Diagnostic HDiagnostic Hemoglobinopathies Secon Edition, Emoglobinopathies Secon Edition, August 21, 2015

Diagnostic Hemoglobinopathies
Laboratory Methods and Case Studies
Zia Uddin, PhD

St. John Macomb-Oakland Hospital
Warren, Michigan
Second Edition
August 2015
Editorial Board
Diane M. Maennle, MD
Chairperson
Kenneth F. Tucker, MD
Member
Rita Ellerbrook, PhD
Member
Piero C. Giordano, PhD
Member
Kimberly R. Russell, MT (ASCP), MBA
Member
Contributors and Reviewers

Antonio Amato, MD
Director
Centro Studi Microcitemie Di Roma
A.N.M.I. ONLUS
Via Galla Placidia 28/30
00159 Rome, Rome
Italy
Erol Omer Atalay, MD
Professor, Medical Faculty
Pamukkale University
Kinikli, Denizli
Turkey
Celeste Bento, PhD
Laboratorio de Anemias Congenitas e Hematologia Molecular
Servico de Hematologia, Hospital Pediatrico
Centro Hospitalar e Universitario de Coimbra
Portugal
Aigars Brants, PhD
Scientific Affairs Manager
Sebia, Inc
400-1705 Corporate Drive
NorCross, GA 30093
USA
Thomas E. Burgess, PhD
Technical Director, Quest Diagnostics
Tucker, Georgia
USA
Shahina Daar, MD, PhD
Associate Professor
Department of Hematology
Sultan Qaboos University, Muscat
Sultanate of Oman
II
Angie Duong, MD
Assistant Professor, Hematopathology
Department of Pathology and
Laboratory Medicine
Medical University-South Carolina
Charleston, South Carolina
USA
Technical Director Emeritus
Helena Laboratories, USA
1530 Lindberg Drive
Beaumont, TX 77707
USA
Eitan Fibach, MD
Professor, Department of Hematology
Hadassah-Hebrew University Medical Center
Ein-Kerem, Jerusalem
Israel
Bernard G. Forget, MD
Professor Emeritus of Internal Medicine
Yale School of Medicine
New Haven, CT 06520
USA
Piero C. Giordano, PhD
Hemoglobinopathies Laboratory
Human and Clinical Genetics Department
Leiden University Medical Center
The Netherlands
Dina N. Greene, PhD
Scientific Director, Chemistry
Regional Laboratories, Northern California
The Permanente Medical Group
Berkeley, CA 94710
USA
III
Rosline Hassan, PhD

Professor of Hematology
School of Medical Sciences
University Sains Malaysia, Kelanran
Malaysia
David Hockings, PhD
Formerly with Isolab, USA and
PerkinElmer Corporation, USA
Raleigh-Durham, North Carolina
USA
Prasad Rao Koduri, MD
Division of Hematology-Oncology
Hektoen Institute of Medical Research
Chicago, Illinois 60612
USA
John Lazarchick, M.D.
Professor, Pathology and Laboratory Medicine
Professor, Medicine
Director, Hematopathology
Director, Hematopathology Fellowship Program
Vice Chair, Clinical Pathology
Medical University-South Carolina
Charleston, SC
Elaine Lyon, PhD
Associate Professor of Pathology
University of Utah School of Medicine
Medical Director, Molecular Genetics
ARUP Laboratories, Salt Lake City, UT
USA
Bushra Moiz, PhD
Associate Professor
Department of Pathology and Microbiology
The Agha Khan University Hospital, Karachi
Pakistan
IV
Herbert L. Muncie, MD
Professor, Department of Family Medicine
School of Medicine, Louisiana State University
1542 Tulane Ave
New Orleans, LA 70112
USA
Gul M. Mustafa, PhD
Post-Doctorate Fellow
Department of Pathology
The University of Texas Medical Branch
Galveston, TX 77555
USA
Associate Pathologist
Warren, MI 48093
USA
Jayson Miedema, MD
Post-Doctorate Fellow
Department of Pathology and Laboratory Medicine
University of North Carolina
Chapel Hill, North Carolina
USA
Christopher R. McCudden, PhD
Assistant Professor, Department of Pathology
and Laboratory Medicine, University of Ottawa
Ottawa, Ontario
Canada
Michael A. Nardi, MS
Associate Professor
Department of Pediatrics and Pathology
New York University School of Medicine
New York, NY 100016
USA
John Petersen, PhD

Professor, Department of Pathology
The University of Texas Medical Branch
Galveston, TX 77555
USA
Joseph M. Quashnock, PhD
Laboratory Director
PerkinElmer Genetics, Inc
90 Emerson Lane, Suite 1403
P.O. Box 219
Bridgeville, PA 15017
USA
Semyon A. Risin, MD, PhD
Professor of Pathology & Laboratory Medicine
Director of Laboratory Medicine Restructuring
& Strategic Planning Program
University of Texas Health Science CenterHouston Medical School
6431 Fannin Street, MSB, 2.290
Houston, TX 77030
USA
Maria Cristina Rosatelli, PhD
Professor, Dipartimnto di Scienze Biomediche
e Biotecnologie Universit degli Studi di Cagliari
09121 Cagliari, Sardina
Italy
Donald L Rucknagel, MD, PhD
Professor Emeritus
Department of Human Genetics
University of Michigan, School of Medicine
Ann Arbor, Michigan
USA
Kimberly Russell, MT (ASCP), MBA
Manager & Operations Coordinator
Hematology and Blood Bank
St. John Hospital & Medical Center
and affiliated hospitals of St. John
Providence Health System, Michigan
USA
VI
Luisella Saba, PhD

Professor, Dipartimnto di Scienze Biomediche
e Biotecnologie Universit degli Studi di Cagliari
09121 Cagliari, Sardina
Italy
Dror Sayar, MD, PhD
Department of Pediatrics,
Hematology-Oncology
Tel Hashmer Medical Center
Ramat Gan
Israel
Upendra Srinivas, MD
Department of Hematology
Kokilaben Dhirubhai Ambani Hospital
& Medical Research Institute
Mumbai, Maharashtra
India
Elizabeth Sykes, MD
Clinical Pathologist
William Beaumont Hospital
Royal Oak, Michigan
USA
Ali Taher, MD, PhD
Professor Medicine, Hematology & Oncology
American University of Beirut Medical Center
Beirut
Lebanon
Kenneth F. Tucker, MD
Director, Hematology & Oncology Services
Webber Cancer Center
Warren, Michigan
USA
VII
Zia Uddin, PhD

Consultant, Clinical Chemistry
St John Macomb-Oakland Hospital
Warren, Michigan
USA
Vip Viprakasit, MD, D. Phil
Professor
Department of Paediatrics & Thalassemia Center
Faculty of Medicine Siriraj Hospital, Mahidol University
2 Prannok Road, Bangkoi
Bangkok 10700
Thailand
Dr. Henri Wajcman
Director of Research Emeritus
Editor-in-Chief Hemoglobin
INSERM U955 (Team 11)
Hospital Henri Mondor
94010 Creteil
France
Winfred Wang, MD
Professor of Pediatrics
University of Tennessee College of Medicine
Pediatric Hematologist & Oncologist
St Jude Childrens Research Hospital
Memphis, Tennessee
USA
Andrew N Young, MD, PhD
Department of Pathology & Laboratory Medicine
Emory University School of Medicine
Atlanta, GA 30303
USA
VIII
Financial Disclosure
I neither had nor will have financial relationship
with any of the manufacturers or any other
organization mentioned in the book.
Similarly all the contributors and reviewers
of the book have worked with gratis to further
the cause of education.
This book and its translations into several
languages are provided at no charge.
August 2015
Zia Uddin, PhD
IX
Dedication
This book is dedicated with heartfelt thanks to my
professors responsible for my PhD level education in
Chemistry at the Illinois Institute of Technology, Chicago,
Illinois, and post-doctoral education and training in Clinical
Chemistry at the University of Illinois Medical Center,
Chicago, Illinois.
Illinois Institute of Technology, Chicago, Illinois
Professor Kenneth D. Kopple, PhD
Professor Paul E. Fanta, PhD
Professor Robert Filler, PhD
Professor Sidney I. Miller, PhD
University of Illinois Medical Center, Chicago, Illinois
Professor Newton Ressler, PhD
August 2015
Zia Uddin, PhD
Preface
Higher level education is one of the blessings of God.
Unfortunately, primarily due to economic and logistic reasons
a vast majority of the qualified candidates are denied this
opportunity.
Internet has the potential of mass education at an
infinitesimal cost. This is the 3rd book launched via Internet
by me at no charge.
All the MD/PhD degree holders are most respectfully
requested to utilize the Internet as a means of communication
to launch books at no charge in their areas of expertise.
Love
God
Love
People
Serve
The World
August 2015
Zia Uddin, PhD
XI
Acknowlegement
During the past three years I contacted worldwide >200 family physicians,
clinical chemists, pathologists, hematologists, public health officials and experts in
diagnostic hemoglobinopathy for formatting this book. The contribution of all of these
individuals is heartfelt and very much appreciated.
I am highly indebted to the following persons for their technical support:
Kimberly R. Russell, MT (ASCP), MBA
Jennifer Randazzo, MS (Information Technology)
The following manufacturers and organizations provided technical support,
and facilities for the collection of data for the book:
Helena Laboratories, USA
Sebia, France
PerkinElmer Corporation, USA
Bio-Rad, USA
ARUP Laboratories, USA
Quest Diagnostics, USA
College of American Pathologists, USA
Seven Universities and four Newborn Screening Laboratories, USA
(names are with held as per their request)
Mr. Mathew Garrin, Biomedical Communications and Graphic Arts Department,
Wayne State University, School of Medicine, Detroit has worked on the figures, scans,
and layout of the book. I am very grateful to him for his contribution.
Finally, I would like to thank the following persons for facilitating my work:
Adrian J. Christie, MD, Medical Director of Laboratories
St. John Macomb-Oakland Hospital, Warren, Michigan, USA
Anoop Patel, MD, Assistant Systems Medical Director
St John Providence Health System Laboratories, Warren, Michigan, USA
Mr. Tipton Golias, President & CEO
Helena Laboratories, Beaumont, Texas, USA
August 2015
Zia Uddin, PhD
XII
Table of Contents
Chapter 1
Hemoglobin

1.1
1.2
1.3
1.4
Chapter 2
Hemoglobin Structure
Hemoglobin Function
Hemoglobin Synthesis
Hemoglobin Variants
Red Blood Cell Morphology
10
John Lazarchick, MD
Angie Duong, MD
Chapter 3
Diagnostic Laboratory Methods

3.1
Basic Concepts
44
Jayson Miedema, MD
and Christopher R. McCudden, PhD
3.1.1
3.1.2
3.1.3
3.1.4
3.1.5
3.1.6
3.2
Unstable Hemoglobins
Altered Affinity Hemoglobins
Sickle Solubility Test
Serum Iron, TIBC, Transferrin, Ferritin
Soluble Transferrin Receptor
Hepcidin
Microcytosis
55
Diane Maennle, MD
and Kimberly Russell, MT (ASCP), MBA
3.3
Hereditary Persistence of Fetal Hemoglobin

3.3.1
3.3.2
3.3.3
3.3.4
3.3.5
Introduction
Deletions Associated with the HPFH Phenotype
Non-Deletion Forms of HPFH
HPFH Unlinked to the -Globin Gene Cluster
Conclusion
XIII
62
3.3.6 Hemoglobin F Quantification
3.4
Flow Cytometry Measurements of Cellular Fetal

Hemoglobin, Oxidative Stress and Free Iron in
Hemoglobinopathies 75
Eitan Fibach, MD
3.4.1 Flow Cytometry of Blood Cells
3.4.2 Measurement of Fetal Hemoglobin-Containing
Erythroid Cells
3.4.3 Staining Protocols for F-RBCs and F-Retics (15)
3.4.4 F-Cell Determination for Fetal-Maternal Hemorrhage
(FMH) in Pregnant Patients wit -Thalassemia- A
single Case and General Conclusion (16)
3.4.5 Oxidative Stress
3.4.6 Staining Protocols for ROS and GSH
3.4.7 Intracellular Free Iron
3.4.8 Staining Protocol for LIP
3.5
Solid Phase Electrophoretic Separation
95
Rita Ellerbrook, PhD, and Zia Uddin, PhD

3.5.1
3.5.2
3.5.3
3.5.4
3.5.5
Introduction
Cellulose Acetate Electrophoresis (alkaline pH)
Agarose Gel Electrophoresis (alkaline pH)
Agar Electrophoresis (acid pH)
Interpretation of Hemoglobin Agarose Gel (pH 8.6)
and Agar Gel (pH 6.2) Electrophoresis
3.5.6 Requirements for the Identification of Complex
Hemoglobinopathies
3.6
Capillary Zone Electrophoresis

Zia Uddin, PhD
3.6.1
3.6.2
3.6.3
3.6.4
Introduction
Basic Principle
Application of CZE in Diagnostic
Hemoglobinopathies
Interpretation of CZE Results
XIV
3.7
Isoelectric Focusing
117
107
David Hockings, PhD

3.7.1 Introduction
3.7.2 IEF of Normal Adult Hemoglobin: HbA (Adult),
HbF (Fetal), HbA2
3.7.3 IEF of Normal Newborn Hemoglobins: HbF (Fetal)
and HbA (Adult)
3.7.4 IEF of Beta-Chain Variant Hemoglobins
3.7.5 IEF of Alpha Chain Variant Hemoglobins
3.7.6 IEF of Thalassemias
3.8
High Performance Liquid Chromatography
129
Zia Uddin, PhD

3.8.1 Introduction
3.8.2 Basic Principle
3.8.3 Illustrations
Chapter 4
Globin Chain Analysis

4.1
136
Zia Uddin, PhD

4.1.1 Cellulose Acetate Electrophoresis
(Alkaline and Acid pH)
4.2
Reverse Phase High Performance Liquid

Chromatography
140
Zia Uddin, PhD, and Rita Ellerbrook, PhD
4.3 Globin Chain Gene Mutations: DNA Studies 149

4.3.1 Introduction
4.3.2 Genotyping-PCR Methodology
4.3.3 Mutations
XV
4.4
Electrospray Ionization-Mass Spectrometry

Gul M. Mustafa, PhD and John R. Petersen, PhD
4.5
PCR and Sanger Sequencing
181
Elaine Lyon, PhD

4.5.1 Alpha Globin
4.5.2 Beta Globin
4.5.3 Sequencing
4.5.4 Reporting Sequence variants
4.5.5 DNA Sequence Traces
4.5.6 Conclusion
Chapter 5
Alpha and Beta Thalassemia

Herbert L. Muncie, MD.
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
Epidemiology
Pathophysiology
Alpha Thalassemia
Beta Thalassemia
Diagnosis
Treatment
Complications
Other Treatment Issues
5.8.1
5.8.2
5.8.3
5.8.4
5.8.5
5.8.6
5.8.7
5.8.8
Hypersplenism
Endocrinopathies
Pregnancy
Cardiac
Hypercoagulopathy
Psychosocial
Vitamin Deficiencies
Prognosis
XVI
Chapter 6
Neonatal Screening for
191
166
Hemoglobinopathies
212
Zia Uddin, PhD

6.1
6.2
6.3
6.4
Introduction
Methodologies
Laboratory Reports Format & Interpretation
Examples of Neonatal Screening
6.4.1.
6.4.2
6.4.3
6.4.4
6.5
Chapter 7
Capillary Zone Electrophoresis

Isoelectric focusing
Isoelectric focusing and High Performance
Liquid Chromatography
Isoelectric focusing, High Performance Liquid
Chromatography and DNA studies
Genetic Counseling & Screening
Prenatal Diagnosis of Beta-Thalassemia

and Hemoglobinopathies 236
Maria Cristina Rosatelli, PhD, and Luisella Saba, PhD
Chapter 8
Hemoglobin A1c
266
Zia Uddin, PhD

8.1
8.2
8.3
Introduction
HbA1c Diagnostic Role in Diabetes Mellitus, and
Glycemic Control in Adults
Measurement of HbA1c
8.4
Factors Affecting the Accuracy of Hb A1c Assay
XVII
Case Studies
278
Introduction
Case # 1
Normal Adult
281
Case # 2
Hemoglobin S trait
Case # 3
Hemoglobin S homozygous
Case # 4
Hemoglobin S with hereditary persistence

of fetal hemoglobin (HPFH) 298
Case # 5
Hemoglobin G-Philadelphia trait
Case # 6
Hemoglobin S-G Philadelphia
313
Case # 7
Hemoglobin G-Coushatta trait
321
Case # 8
Hemoglobin C trait
Case # 9
Hemoglobin C homozygous
Case # 10
Hemoglobin C with hereditary persistence

of fetal hemoglobin (HPFH) 340
Case # 11
Hemoglobin S-C disease
Case # 12
Hemoglobin D-Los Angeles (D-Punjab) trait
Case # 13
Hemoglobin S-D disease
286
292
306
327
333
346
353
360
XVIII
Case # 14
Hemoglobin E and Associated Disorders
367
Case # 14 a Hemoglobin E trait
373
Case # 14 b Hemoglobin E homozygous

Case # 14 c Hemoglobin S-E disorders
Case # 15
Hemoglobin S-Korle Bu (G-Accra)
Case # 16
Hemoglobin O-Arab trait
Case # 17
-Thalassemia trait
Case # 18
Hemoglobin S-+- thalassemia
Case # 19
Hemoglobin C-o thalassemia
Case # 20
Hemoglobin Hasharon trait
Case # 21
Hemoglobin Zurich trait
Case # 22
Hemoglobin Lepore trait
Case # 23
Hemoglobin J-Oxford trait
Case # 24
Hemoglobin J-Baltimore trait
Case # 25
Hemoglobin Malmo trait
Case # 26
Hemoglobin Koln trait
Case # 27
Hemoglobin Q-India trait
475
Case # 28
Hemoglobin Dhofar trait
488
XIX
390
396
402
408
415
421
428
434
442
455
466
449
378
384
Chapter 1
Hemoglobin
To attempt a full treatise on hemoglobin in this textbook would be an effort in
futility as the purpose is not to duplicate knowledge already present in the literature.
Rather, this chapter is to provide basic information to the reader which will allow him/her
to properly identify hemoglobin variants in their laboratory. A basic knowledge of the
hemoglobin molecule is absolutely critical to that effort and the sections printed below
are written expressly for that purpose. For a complete treatise on hemoglobin, textbooks
1
such as Disorders of Hemoglobin edited by Steinberg, Forget, Higgs and Nagel should
be consulted.
1.1 Hemoglobin Structure

Composed of 2 distinct globin chains, the complex protein molecule known as
hemoglobin (heme + globin) is arguably THE primary component of the red blood
cell in human beings. In normal adults, the globin chains are either alpha (), beta (),
gamma () or delta (). In addition, during embryonic life in utero, zeta () and epsilon
() chains are present in the first several weeks of life, being rapidly converted to alpha,
beta and gamma chains as development occurs.
Figure 1. Globin chains concentration changes in embryonic, fetal and post-natal

stages of life (Huehns ER, Dance N, Hecht S, Motulsky AG. Human embryonic
hemoglobins. Cold Spring Harbor Symp Quant Biol 1969; 29: 327-331). Adopted
with permission from Blackwell Publishing (Barbara J. Bain, Haemoglobinopathy
Diagnosis, 2nd Edition, 2006).
Each of these globin chains has associated with it a porphyrin molecule
known as heme whose primary function in the red blood cell is the facilitation of
transport of oxygen to the tissues of the human body. The globin portion of the
molecule serves several functions, not the least of which is protection. The internal
pocket of the molecule formed from the convergence of the four globin chains,
provides a hydrophobic environment in which the heme molecules reside. This
2
pocket protects the heme from oxidation and facilitates oxygen transfer to the
tissues of the body. The previously mentioned and chain-containing hemoglobins
have very high oxygen affinities, a factor very important in the early embryonic life of
the fetus.
The hemoglobin molecule can be looked at in four different ways; primary,
secondary, tertiary and quaternary structural views. While outside of the scope of
this volume, each of these structures contributes definitive unique properties to the
various hemoglobin molecules from normal hemoglobins to the very rare and
functionally diverse molecules. The primary structure of all hemoglobins is the order
of amino acids found in the globin chains of the molecule. It is this unique sequence
that is the major differentiator of hemoglobin from each other. The secondary
structure of hemoglobin is the arrangement of these amino acid chains into alpha
2
helices separated by non-helical turns . The tertiary structure is the 3-dimensional

arrangement of these globin chains forming the pocket of hemoglobin that cradles
the iron molecule in its grasp. The quaternary structure is the moving structure of the
molecule that facilitates the oxygenation of the heme molecules in response to the
physiological needs of the human body.
Figure 2. Tertiary structure of a globin chain and the quaternary structure of

hemoglobin molecule (Adopted with permission from Blackwell Publishing, Barbara
J. Bain, Haemoglobinopathy Diagnosis, 2nd Edition, 2006).
The forthcoming sections will elucidate the effects that these structural
considerations have on the hemoglobin molecule and, more specifically, the
abnormal and atypical hemoglobin variants.
1.2
Hemoglobin Function
As mentioned above, the primary function of hemoglobin is to reversibly
transport oxygen to the tissues of the body. In addition, however, this flexible
molecule can also transport carbon dioxide (CO2) and nitrous oxide (NO). The
transport of CO2 is facilitated by reversible carbamoylation (formation of carbamoyl
moiety, i.e., H2NCO-) of the N-terminal amino acids of the globin chains and can,
4
via proton scavenging, keep CO2 in the soluble bicarbonate form . Nitrous oxide is
handled in two different ways by hemoglobin: one as a transporter and the other as
a scavenger. Blood levels of NO are therefore, by definition, a balance between NO
production and NO removal by binding to oxyhemoglobin. Since NO is an extremely
potent vasodilator, hypoxic patients will have lower oxyhemoglobin and therefore
higher amounts of free NO. This free NO can cause significant vasodilation, a
physiological effect that is very desirable in hypoxia.
All hemoglobin molecules, either normal or variant, share the same
functionality in the human body. Therefore, the primary structural differences
mentioned above and in more complete treatises (i.e., amino acid
substitutions/deletions) will be the prime reason for functional differences. It is these
amino acid variances that, along with the secondary, tertiary and quaternary
structural differences, will determine if the variant hemoglobin is either benign or
clinically important.
The bottom line is this whether the hemoglobin is normal or variant in
nature, the prime reason for determining the hemoglobin phenotype of the patient is
to assess the functionality of the hemoglobin. If the variant is normally functioning in
both the heterozygous and homozygous states, the clinical picture is benign. If,
however, the variant has normal properties in the heterozygous state (i.e., trait) but
clinical issues in the homozygous state (i.e., disease), the phenotypic analysis and
subsequent interpretation becomes ultimately important to the patient.
1.3 Hemoglobin Synthesis

The synthesis of hemoglobin, as mentioned before, is under the control of
gene loci on two chromosomes: chromosome 11 (the beta globin or non-alpha
gene) and chromosome 16 (the alpha globin gene). Hemoglobin variants (alpha,
beta, gamma, delta and fusion) are the result of alterations in the nucleotide
sequences of the globin genes and can occur for more than one reason. Mutations
such as point mutations, insertions and deletions can have major, minor or no
influences on hemoglobin function or structure. That being said, the site of the
synthetic variance can in some cases alter the ability of the hemoglobin molecule to
function in a normal manner, i.e., stability, oxygen affinity, solubility or other critical
functions. These alterations truly determine whether the variant hemoglobin is
classified as benign (i.e., no abnormal or pathological effect) or pathological (a
significant physiological effect). The actual nature of the alteration is not of initial
importance to the hemoglobinopathy interpreter. However, once assigned, the
identity of the variant hemoglobin may become of importance when looking at
second generation offspring from the variant carrier, i.e., the pregnant female. For
most hemoglobin variants, the synthetic pathway is of no clinical interest in that the
resulting hemoglobin is benign. It may, however, be of academic interest in that the
identification of the synthetic anomaly can, indeed point to the genetic locus or loci
involved in the alteration, thus giving information to the genetic counselor as to
possible genetic details of the hemoglobinopathy.
As mentioned before, the true reason for identifying the abnormal hemoglobin
or hemoglobins in patients is to identify any associated functional anomalies
6
associated with these hemoglobins. The actual hemoglobin identification in and of

itself is merely of academic interest.
1.4 Hemoglobin Variants

All hemoglobin variants have one thing in common they all involve the
hemoglobin molecule and its functionality. Whether alpha, beta, gamma, delta,
fusion variant, etc., the variant and its effect are judged not on its migration or
concentration but rather on its functionality. The amino acid variation (e.g., glutamic
acid valine at position 6 on the beta chain for hemoglobin S) is the prime effector
of the variants functional alteration(s) and will in most cases be the causative factor
in any abnormal migration that the variant may have versus the normal
hemoglobins (A, F, A2). Most variants therefore will have altered electrophoretic or
chromatographic migrations when compared to the normal variants. Some, such as
hemoglobin Chicago, are not separable by normal electrophoretic techniques and
rely on high performance liquid chromatographic (HPLC) separations to identify its
presence in the blood. As previously mentioned, the presence of variant traits (i.e.,
AS, sickle trait) may or may not be of clinical consequence. Where these traits
really are of importance is in the homozygous state (i.e., SS for hemoglobin S). The
clinical picture dramatically changes with significant physiological changes being
directly associated with the homozygous state. This therefore requires the
interpreter to have several pieces of information specific to the patient at hand
during the interpretation of the hemoglobinopathy. This data includes, but is not
limited to, pregnancy, transfusion history and ethnicity. All of these pieces of
7
information can be critical to the proper identification/interpretation of the

hemoglobin variant in the patients specimen. For example, an elevation of
hemoglobin F in a female patient with a normal hemogram may be evidence of
hereditary persistence of fetal hemoglobin; whereas, if this female is pregnant, the
elevation may be a normal physiological response to the fetal presence in her body.
These data may not be readily available and may require contact with the ordering
healthcare professional to obtain these facts. However obtained, they are
necessary for the proper identification of the hemoglobin variant or variants in the
patients bloodstream and therefore are important in the assignment of a benign or
pathological assessment of the variant hemoglobin.
The variants described in the following chapters all obey the aforementioned
differences, i.e., amino acid substitutions, genetic deletions, sequence modifications,
etc. While not critical, the exact identification of the variant in and of itself is not
normally life-threatening, especially in the heterozygous state, i.e., trait. It is
essential that the variant be properly identified as a mis-identification can lead to
other issues. For example, a mis-interpretation of a hemoglobin G trait (AG) as a
sickle trait (AS), while not in and of itself is clinically an issue, presents real
difficulties for a couple expecting a child. If both partners are AS, there is a 1 in 4
chance that a child born to this couple could be homozygous SS or sickle cell
disease. In the case of an AS mother and an AG father (or vice versa), there is a 1
in 4 chance of a child being born with a phenotype of SG. While on the surface this
may appear as a problem, the SG phenotype is no more of a clinical issue than a
simple AS trait. Without the exact identification of the AG trait, the interpretation and
action taken by attending clinicians may be very different.
References
1. Steinberg, MH, Forget, BG, Higgs, DR and Nagel, RL., Disorders of
Hemoglobin,
Cambridge University Press, 2001.
2. Bain, Barbara J.. in Hemoglobinopathy Diagnosis, 2nd Ed., pg. 4, Blackwell
Publishing, 2006.
3. Bain, Barbara J.. in Hemoglobinopathy Diagnosis, 2nd Ed., pg. 1, Blackwell
Publishing, 2006.
Chapter 2
Red Blood Cell Morphology
9
John Lazarchick, MD
Angie Duong, MD
Knowledge of red blood cell (RBC) morphology is essential for the clinical diagnosis of
hemoglobinopathy. The diameter of RBC, when mature under normal circumstances
is approximately 7-8 microns, and RBC is round, anuclear and biconcave disc-shaped.
A study of RBC morphology includes size, shape, color, inclusions and arrangement. In
this chapter we have presented with pictures of the most commonly encountered RBC
morphologies with legends and few examples of the diseases with abnormal RBC
morphology. In the clinical cases of this book, we have mentioned only the main
features of the peripheral blood smear, therefore a review of this chapter is advised
for a nave reader for the proper diagnosis of hemoglobinopathy.
The following RBC morphology cases are presented in this chapter:
Size:
Macrocyte large
Microcyte small
Normocyte normal
Hemoglobin Content:
Hypochromic low
Normochromic normal
Polychromatic high
Shape and Inclusions:
Anisocytosis
Poikiocytosis
Acanthocyte
Basophilic Stippling
Bite Cell
Blister Cell
Burr Cell (Ecchinocyte)
Heinz Body
Howell-Jolly Body
Pappenheimer Body
Schistocyte
Sickle Cell
Spherocyte
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig.10
Fig.11
Fig.12
Fig.13
Fig.14
Fig 15
Fig.16
Fig.17
Fig.18
Fig.19
10
Stomatocyte
Fig. 20
Target Cell
Fig. 21
Teardrop Cell
Fig. 22
RBC Agglutination
Fig. 23
Rouleaux Formation
Fig. 24
Diseases :
Erythroblastosis Fetalis
Fig. 25
Hemoglobin C Disease
Fig. 26
Hemoglobin C/beta Thalassemia Fig. 27
Hemoglobin S/beta Thalassemia Fig. 28
Hemoglobin SC Disease
Fig. 29
Sickle Cell Disease
Fig. 30
Fetal-maternal Hemorrhage:
Fig. 31
Kleinhauer-Betke Stain
11
Fig. 1 Macrocyte-large
The diameter of RBC >9-14 microns (1.5 to 2 times larger than
normal RBC) and the MCV >100 fL is characteristic of macrocyte.
Macrocytes are mostly oval in shape.
12
Fig. 2 Microcyte-small
RBC, when abnormally smaller (< 5 micron) than normacytic
RBC (7-8 micron) is defined as microcyte (also called
microerythrocyte). The MCV of the microcyte RBC is < 80 fL.
13
Fig. 3 Normocyte-normal
The diameter of RBC, when mature under circumstances is
approximately 7-8 microns, and are round, anuclear, biconcave
disc-shaped with an internal volume of 80-100 fL.
The term normocyte is used when the size of the RBC is normal.
14
Fig. 4 Hypochromasia
Hypochromasia is a descriptive term for red blood cells where the
central pallor is greater than one third the diameter of the red
blood cell (black arrows). This is due to a decrease in the amount
of hemoglobin in the cells. Diseases with prominent
hypochromasia are iron deficiency anemia, anemia of chronic
disease, and sideroblastic anemia. Some cases of
myelodysplastic syndrome can also have hypochromatic red
blood cells. Hypochromasia is reflected in the complete blood
count (CBC) by a decreased mean corpuscular hemo-globin
concentration (MCHC).
Also present are: target cells/codocytes (red arrow),
polychromatic forms (blue arrow), fragmented red blood
cells/schistocytes (green arrows), and tear drop
forms/dacryocytes (yellow arrows). Overall, this smear shows
moderate anisopoikilocytosis.
15
Fig. 5 Normochromic-normal
This descriptive term is applied to a red blood cell with a normal
concentration of hemoglobin. The above figure is a peripheral
blood cell smear of a patient treated for iron deficiency anemia.
Blue arrow shows normochromic-normal RBC. Black arrow shows
hypochromic-microcytic RBC.
16
Fig. 6 Polychromatic-high
This smear demonstrates polychromasia. Numerous
polychromatic forms (black arrows), which are young slightly
larger red blood cells with a purple-tinge due to retained RNA, are
present. Polychromasia is the bone marrows response to anemia,
where the bone marrow releases younger red blood cells.
Sometimes, nucleated red blood cells are also released into the
peripheral blood. Due to their larger size, when many
polychromatic forms are present, the CBC values of mean
corpuscular volume (MCV) as well as RDW (red blood cell
distribution width) will be increased.
In a supravital stains, such as cresyl violet, the retained RNA in
the polychromatic forms precipitate out and these cells are called
reticulocytes. Thus, sometimes the terms polychromatic form is
used interchangeably with reticulocytes.
17
Fig. 7 Anisocytosis
The term anisocytosis refers to size variation seen among red
blood cells. As demonstrated above, there are small red blood
cells as well as large red blood cells, some approaching the size
of a neutrophil (green arrow). Ansiocytosis is a reactive process
where the bone marrow is releasing younger red blood cells,
therefore an increased number of polychromatic forms can also
be seen (black arrow). In the complete blood count (CBC),
anisocytosis is reflected by having an increased red cell
distribution width (RDW).
18
Fig. 8 Poikilocytosis
Poikilocytosis refers to shape variation. In poikilocytosis, the red
blood cells have lost their normal discoid appearance. The
example shown here has a predominance of
ovalocytes/elliptocytes, which are red blood cells that have a
length twice their diameter (a few are indicated by blue arrows).
Also seen are schistocytes (red arrows), which are fragmented
red blood cells. Ovalocytes/Elliptocytes are seen in peripheral
blood smear in some conditions, e.g., thalassemia, iron
deficiency, etc.
Note: When both shape and size variation is seen in the red blood
cells, the term anisopoikilocytosis can be used.
19
Fig. 9 Acanthocyte (Spur Cell)

These are red blood cells with spike-like projections (arrow) of
varying length. They can be seen in both hereditary and acquired
hemolytic anemias including alcoholic liver disease, pyruvate
kinase deficiency, vitamin E deficiency, Huntingtons disease-like
situation and abetalipoproteinemia. In the latter case,
malabsorption of fat, neurologic damage and developmental delay
are noted.
20
Fig. 10 Basophilic Stippling

Red blood cells have multiple fine or coarse small basophilic dotlike inclusions which are due to small clumps of ribonucleic acid
and mitochondria. These inclusions can be seen in a wide variety
of conditions including lead poisoning, hereditary
hemoglobinopathies including unstable hemoglobins,
thalassemias, sideroblastic anemias, megaloblastic anemia and
hereditary pyrimidine 5- nucleotidase deficiency.
21
Fig. 11 Bite Cell

Bite cell (arrow) has a semicircular portion of the membrane
removed. This morphologic abnormality results from splenic
macrophages removing denatured precipitated hemoglobin with
Heinz body formation in these cells. The most common cause of
this finding is glucose-6-phosphate dehydrogenase deficiency.
22
Fig. 12 Blister Cell

Red blood cells with cytoplasmic clearing (large arrows) on one
side and hemoglobin on the other side in a patient with hemolytic
anemia. Multiple polychromatophilic red blood cells
(reticulocytes) are noted (small arrow). In addition, a single cell
with a Howell-Jolly body inclusion (double arrows) is noted,
23
Fig. 13 Burr Cell (Echinocyte)

These are red blood cells with short round membrane projections
with blunt ends (large arrow). Red blood cells with more spikelike projections (small arrow) can also be seen. This finding is
often an artifact of slide preparation but is typically seen in
patients with uremia and pyruvate kinase deficiency.
24
Fig. 14 - Heinz Body

In a RBC when the hemoglobin is denatured (either by a change
of an internal amino acid or glucose-6-phosphatse deficiency,
etc.), the heme portion of hemoglobin molecule is dissociated
from the globin chain. The globin chain after dissociation from the
heme molecule becomes denatured forming a small ball like
structure (black arrow) inside the RBC, and thus called Heinz
body.
25
Fig. 15 Howell-Jolly Body

This red blood cell inclusion (arrow) is round basophilic DNA
remnant usually noted in the outer third of circulating red blood
cells. These inclusions are normally extruded in the bone marrow
during normal erythroid maturation. Howell-Jolly bodies can be
seen in asplenia, conditions associated with hyposplenia including
sickle cell disease and severe hemolytic anemia.
26
Fig. 16 - Pappenheimer Bodies

These are small dark irregular staining granules (large arrow) of
non-heme iron usually noted on the periphery of red blood cells
formed by phagosomes that engulf excess iron. Basophilic
stippling is present in the dysplastic nucleated RBC (small arrow)
These granules stain positive with Prussian blue stain in both the
nucleated RBC and mature red blood cells as shown in the lower
image. They can be found in a variety of conditions including
sideroblastic anemias, thalassemias and myelodyplastic
syndromes.
27
Fig. 17 - Schistocyte (RBC fragments, Helmet Cells)

These are red blood cell fragments typically with two pointed ends
formed when RBCs are sheared by fibrin strands in clotted blood
vessels. Disorders include microangiopathic hemolytic anemia,
disseminated intravascular consumption (DIC), thrombotic
thrombocytopenic purpura (TTP) and hemolytic uremic syndrome
(HUS).
28
Fig. 18 - Sickle Cell

In inherited blood cell disease (change of an amino acid residue
in the globin chain) the shape of the RBC is deformed. The
deformation of RBC resembles (a waxing crescent) a moon
sighted on the first day of lunar month. Since this deformation
looks like a sickle (an implement with a semicircular blade
attached to a short handle, used for cutting grain), therefore this
deformation is called sickle cell.
29
Fig. 19 Spherocytes
This peripheral blood smear is from a patient with autoimmune
hemolytic anemia (AIHA) and is characterized by many
spherocytes (blue arrows) and microspherocytes (black arrows).
Spherocytes are red blood cells that have no central pallor. As the
name implies, microspherocytes are small spherocytes. If the
majority of the cells in a peripheral smear are spherocytes, the
possibility of hereditary spherocytosis arises. Hereditary
spherocytosis is an autosomal dominant disease where one of the
genes that code for red blood cell proteins (such as spectrin and
ankyrin) become mutated.
30
Fig. 20 Stomatocyte
Red blood cells with slit-like central pallor (arrow) caused by a
decrease in surface area to volume ratio associated with a
membrane permeability disorder. Hereditary stomatocytosis is
associated with hemolysis which can be severe. Acquired
stomatocytosis can be seen in acute alcohol intoxication, chronic
liver disease and as drying artifact in peripheral smear
preparation.
31
Fig. 21 - Target Cells

Also known as codocytes, these red blood cells appear to have a
bullseye in the center of the red blood cells central pallor. This
morphologic change is due to a relative excess of cell membrane,
due to decreased cell content or increase in the cells surface
area. Target cells can be seen in liver failure, Hemoglobin C
disease, thalessemias (both alpha and beta), and iron deficiency.
32
Fig. 22 - Tear drop cells

Also known as dacryocytes/dacrocytes (red circles), are distorted
red blood cells where one end of the cell is drawn into a sharp
point. These cells are usually seen in myelophthsic anemias,
which is where the normal marrow space is occupied by nonhematopoietic elements, such as fibrosis or metastatic carcinoma.
It is hypothesized that the shape of the cells is due to the red
blood cells squeezing between fibers or the cells extrinsic to the
marrow.
33
Fig. 23 - RBC Agglutination

Clumping of the red blood cells is due to coating of the RBC
surface with antibodies. Disorders causing the agglutination may
be primary as in cold agglutinin disease or secondary, either
clonal as in lymphoproliferative disorders or polyclonal as seen in
Mycoplasma pneumonia. The upper left insert is from a slide
prepared at room temperature and the upper right insert is a slide
after warming the sample to 370 C with clearing of the
agglutination in a patient with cold agglutinin disease.
34
Fig. 24 Rouleaux Formation

Rouleaux formation is seen in peripheral blood smears in
association with plasma cell neoplasms, most commonly
myeloma. The red cells become stuck together in a stack of
coins formation, due to the excess immunoglobulin proteins
released by malignant plasma cells. Not all cases of plasma cell
neoplasms have rouleaux formation. Rouleaux formation is one of
the causes of an increased erythrocyte sedimentation rate (ESR).
35
Fig. 25 - Erythroblastosis Fetalis

This is an alloimmune hemolytic anemia in the fetus secondary to
placental transfer from mother to fetus during pregnancy of anti
A or B or anti-Rh blood group IgG antibodies. These blood
groups are present on the fetal RBCs but not on the maternal
RBCs which then causes immune hemolysis in the fetal
circulation. As noted on the smear, numerous nucleated RBCs
(large arrow) and polychromatophilic RBCs (small arrow) are
noted.
The case shown above was due to antibodies to Rh D blood
group.
36
Fig. 26 - Hemoglobin C Disease

In this case of homozygous hemoglobin C disease essentially all
of the RBCs are target cells (large arrow). Hemoglobin C crystals
are rod shaped inclusions (Washington Monumentssmall arrow)
in red blood cells in both heterozygous and homozygous
hemoglobin C disease as well as hemoglobin SC disease. Upper
image shows the crystals at a higher magnification.
37
Fig. 27 - Hemoglobin C/beta Thalassemia

Although most patients with this compound heterozygotic state for
hemoglobin C and beta thalassemia are asymptomatic, a mild to
moderate hemolytic anemia can be seen. The red blood cells are
microcytic and hypochromic. Target cells (double arrow) and C
crystals (single arrows) can be seen.
38
Fig. 28 - Hemoglobin S/beta Thalassemia

Hemolytic anemia due to both production of an abnormal
hemoglobin (Hemoglobin S) and decreased synthesis of beta
globin chains (Beta Thalassemia). Individuals have one abnormal
beta chain with substitution of glutamic acid for valine and either
decreased synthesis, beta+, or complete absence of the other
beta chain, beta0. The peripheral smear shows sickle cells,
nucleated red blood cells, polychromasia, microcytosis,
hypochromic, target cells and basophilic stippling. Note the sickle
cell in the insert and the Howell-Jolly body in the other RBC.
39
Fig. 29 - Hemoglobin SC Disease

This is a representative peripheral blood smear from a patient
with hemoglobin SC disease. Hemoglobin SC disease is an
inherited hemoglobinopathy where the two normal genes for
hemoglobin A have been replaced by one hemoglobin S gene and
one hemoglobin C gene. In hemoglobin S, a single nucleotide at
position 6 of the gene is substituted by another nucleotide
(glutamic acid is substituted by valine). A similar phenomenon
occurs in hemoglobin C, where glutamic acid is substituted by
lysine. When both hemoglobin S and hemoglobin C is present,
the genes are codominant and lead to many interesting peripheral
blood findings.
Hemoglobin S produces drepanocytes/sickle cells (black arrows)
which are red blood cells that appear as crescent moon shapes or
continued next page
sickles. Due to the abnormal hemoglobin content, the
deoxygenated red blood cells become stuck in this shape, thus
40
causing vascular occlusions which in turn lead to many

complications such as pain crisis. Sickle cells are seen when
there is no or decreased levels of hemoglobin A (such as
hemoglobin SS disease, hemoglobin SC disease, hemoglobin S
with thalessemia). In sickle cell trait, where there is one normal
hemoglobin A gene and one hemoglobin S gene, sickle cells are
not seen and the patients usually have no clinical symptoms.
Hemoglobin C manifests in peripheral smears as numerous target
cells/codocytes (green arrows). Additionally, in hemoglobin CC
disease and in hemoglobin SC disease, hemoglobin C crystals
(blue circle) can be seen. These crystals are desicated red blood
cells with squared off/blunt edges. In hemoglobin C trait, target
cells are seen but hemoglobin C crystals are not.
41
Fig. 30 - Sickle Cell Disease

Sickle cell disease is a hereditary hemolytic anemia caused by a
single nucleotide substitution (SNP) of valine for glutamic acid in
the beta globin chain of hemoglobin. This results in hemoglobin
polymerizing at low oxygen tension with sickle cell formation
(small arrow). There is marked polychromasia, target cells and
nucleated red blood cells (insetlarge arrow) on the peripheral
smear.
42
Fig. 31- Fetal-maternal Hemorrhage: Kleihauer-Betke Stain

This test relies on the principle that red blood cells containing fetal
hemoglobin (deep red staining RBCs) are less susceptible to acid
elution than adult red blood cells. Its use is a means of
quantitating fetal-maternal hemorrhage in Rh-negative mothers to
determine the dose of Rho (D) immune globulin needed to inhibit
formation of Rh antibodies. It can also be used to detect
hereditary persistence of fetal hemoglobin (HPFH).
Chapter 3
3.1
Basic Concepts
Jayson Miedema, MD, and Christopher R. McCudden, PhD
43
3.1.1 Unstable Hemoglobins

Unstable hemoglobins are characterized by disorders in globin production which
affect the lifespan of the hemoglobin molecule and subsequently the cell leading to
decreased cell stability and increased cell turnover. There are a large number of specific
variants which can result in abnormal hemoglobin production, the most commonly
reported of which is Hb Koln. Many of these abnormal globin chains are a result of
single mutations in the form of deletions (e.g. Hb Gun Hill), insertions (e.g. Hb
Montreal), or substitutions (e.g. Hb Koln) and can result in weakened heme-globin
interactions, subunit interactions, or abnormal folding. These disorders are most
commonly expressed in the heterozygous form, most homozygous situations result in
preterm lethality.
Clinically, these patients often present with symptoms of hemolytic anemia which
can be of varying severity. Symptoms of hemolytic anemia include hyperbilirubinemia,
jaundice, splenomegaly, hyperbilirubinuria or pigmenturia as well as the formation of
Heinz bodies. This pheonotype can present or be exacerbated by infections as well as
certain types of drugs. Specifically sulfonamides, pyridium, and antimalarials are known
to cause exacerbation. Parvovirus can also induce aplastic crisis andHbA2 and HbF
may be increased. The peripheral smear often shows anisocytosis, poikilocytosis,
basophilic stippling, polychromasia and, hypochromasia. Since not all unstable
hemoglobins will give abnormal results on HPLC or electrophoresis and/or these results
can be somewhat non-specific, more definitive testing is often performed.
44
Testing for unstable hemoglobins relies on their decreased stability in heat or

isopropanol alcohol. While normal hemoglobins should be relatively stable in these
conditions, hemoglobins with mutations causing instability tend to be less so and will
precipitate out of solution in these environments. In the context of heat stability testing,
the amount of unstable hemoglobin in a sample is given by the following equation:
(Hb4C-Hb50C)/(Hb4C)x100
Where Hb4C is the hemoglobin concentration at 4 degrees centigrade and Hb50C is
the concentration of hemoglobin at 50 degrees centigrade.
False positives may result from samples greater than 1 week in age as well as from
samples with large amounts of fetal hemoglobin. Additional technical and clinical
information on hemoglobinopathies associated with unstable hemoglobin can be
obtained from:
http://medtextfree.wordpress.com/2011/12/30/chapter-48-hemoglobinopathies
3.1.2 Altered Affinity Hemoglobins

Similar to how certain types of mutations can cause instability of the hemoglobin
molecule, other mutations can cause hemoglobins to have altered affinity for oxygen.
These mutations can be single point mutations, insertions, deletions, elongation,
deletion/insertion mutations and are often named after the city in which they were
discovered (Chesapeake, Capetown, Syracuse, etc.). Both alpha-chain variants, e.g. Hb
Chesapeake, and beta-chain variants, e.g. Hb Olser, Hiroshima, Andrew-Minneapolis,
etc., are known in the literature for altered affinity for oxygen. Many of these are
probably clinically insignificant but when significant most commonly present
45
phenotypically as an increase in oxygen affinity often times resulting clinically in

polycythemia (secondary to the bodies perceived lack of oxygen and subsequent
increase in erythropoietin). Measurement of hemoglobin affinity (p50) is critical to the
diagnosis. Conversely and less frequently described, a decreased affinity for oxygen
can lead to clinical cyanosis.
Testing for altered affinity hemoglobins relies on subsequent changes to the
oxygen dissociation curve and the partial pressure of oxygen at which hemoglobin is
50% saturated, the p50. Because most types of altered affinity hemoglobins cause an
increase in oxygen binding, a left shift in the oxygen dissociation curve results.
Automated systems are available for recording the oxygen dissociation curve and rely
on a Clarke electrode to measure oxygen tension while oxyhemoglobin fraction is
measured by dual wavelength spectrophotometer. Abnormal oxygen dissociation curves
are primarily caused by altered affinity hemoglobins but can also be caused by such
factors
as
pH,
temperature,
pCO2,
and
2,3-diphosphoglycerate
(2,3-DPG).
Measurement of pO2, pCO2, pH and SO2 allows for an estimation of p50 to be

calculated.
3.1.3 Sickle Solubility Testing

Sickle cell anemia is a disease resulting in anemia and painful crises, seen
almost exclusively in African Americans. These crises are caused by inappropriate
aggregation of deformed blood cells in small blood vessels. Widely believed to have
thrived in the gene pool because of its protective effects against malaria, it affects a
large number of people of African descent in its homozygous and clinically significant
46
form. An even greater number of people have sickle cell trait (approximately 8-10% of
African Americans), the heterozygous form, which is largely insignificant from a clinical
standpoint.
Sickle cell testing can be performed in a variety of ways and is currently most
commonly tested via hemoglobin electrophoresis when necessary. However, another
form of testing is known as sickle solubility testing which relies on the property of
increased cell fragility as a result of the glutamic acid to valine substitution at the 6 th
position of the beta globin gene, the most common genetic abnormality of sickle cell
anemia. Sickled red blood cells are soluble when oxygenated but upon deoxygenation
tend toward sickling, polymerization, and precipitation. The addition of sodium
metabisulfite reagent to a sample with hemoglobin S promotes deoxygenating and cell
lyses, creating turbidity in the solution. This turbidity makes it difficult to read a card
through the test tube. A negative test is one in which a card can be read through the
tube, a positive test is one in which the card cannot be read.
Several types of hemoglobins can cause false positives (for example some
types of hemoglobin C) so results should be confirmed by electrophoresis; in other
words, when used, solubility testing should be used as a screening test. The test also
fails to differentiate sickle cell trait (a single copy of the sickle cell gene, heterozygous)
from true sickle cell anemia (both copies are sickle cell, homozygous). Samples with low
hemoglobin concentration (<8%) should be doubled as this low concentration can lead
to false negatives. False positives can occur in the settings of lipemia or samples with
monoclonal proteins (dysproteinemia). Both positive and negative controls should be
47
used as results can be somewhat subjective
3.1.4 Serum Iron, TIBC, Transferrin, and Ferritin

Iron is essential for numerous metabolic functions in the body through its
incorporation into proteins involved in oxygen delivery (hemoglobin, myoglobin) and
electron transport and exchange (cytochromes, catalases). While a detailed description
of iron metabolism is beyond the scope of this compendium (interested readers should
seek the references below), it is worth considering the major mechanisms of iron
homeostasis in the context of erythropoiesis. Iron intake in the diet occurs either as free
iron or as heme. Free iron, in the form of Fe 3+, requires reduction to Fe2+ by enzymes
and transporters to cross the intestinal mucosa; heme iron is absorbed directly by
mucosal cells where it is split from heme intracellularly. Once absorbed by the GI tract,
iron is either stored in association with ferritin or transported into the circulation in the
ferric (Fe3+) form. Because of the toxicity of ferric iron, it is transported in the circulation
bound to transferrin. The main target of transferrin-bound iron is erythroid tissue, which
takes up iron through receptor-mediated endocytosis. As dietary absorption accounts
for <20% of the daily requirement, iron recycling plays an essential role in maintaining
iron stores.
During recycling, senescent red blood cells are phagocytosed by
macrophages in the spleen, liver, and bone marrow. Macrophages store some iron
(bound to ferritin), but most is returned to red cell precursors via transferrin. Unlike
dietary absorption, iron excretion is largely unregulated, where losses occur via
epithelial cell sloughing in the skin and GI tract or through menstrual bleeding in
48
premenopausal women. Accordingly, body stores depend on controlling iron uptake in

the GI tract and recycling.
Disorders of iron homeostasis fall into diseases of excess or deficiency. Iron
deficiency is common, particularly in women, and may result from inadequate intake,
blood loss, and pregnancy; in chronic disease iron deficiency is also common. Iron
excess may occur in hemochromatosis or as a result of repeated transfusions.
Clinically, iron status is assessed by measurement of serum iron, ferritin, transferrin,
and total iron binding capacity (TIBC).
Serum or plasma iron levels can be directly measured using several different
methods. Most commonly, a colorimeteric reaction scheme is used in which iron is
separated from transferrin at low pH (~4) and then reduced to Fe 2+ for dye binding; the
color-complex is detected between 530-600 nm spectrophotometrically. Although iron is
typically increased in cases of iron excess and decreased in cases of deficiency, serum
iron measurement by itself is not particularly useful for diagnosis of iron homeostasis
disorders because of the high intra-individual variation in circulating iron levels.
Total iron binding capacity (TIBC) is another test used to assess iron
homeostasis. TIBC can be measured or calculated. TIBC is measured by adding
excess iron to saturate transferrin (usually transferrin is 30% saturated). Unbound iron
is chelated and removed and then the remaining transferrin-bound iron is measured as
described above yielding the total capacity.
This method can be affected by the
presence
proteins,
of
non-transferrin
iron
binding
particularly
in
cases
of
hemochromatosis and thalassemias. Alternatively, TIBC may be calculated based on

the stoichiometric relationship between transferrin and iron (2 molecules of iron are
49
bound to each molecule of transferrin). TIBC is calculated from measured transferrin

using the following equation: TIBC (g/dL) = 1.43 transferrin (mg/dL). Conversely, the
concentration of transferrin may be calculated from measured TIBC as follows:
Transferrin (mg/dL) = 0.7 TIBC (g/dL). TIBC is increased in iron deficiency and
decreased in chronic anemia of disease and in iron overload (it may be normal or
decreased in thalassemia).
From TIBC and serum iron measurement, it is also possible to calculate the %
transferrin saturation (also known as iron saturation) using a simple formula: %
saturation = serum Fe (g/dL) / TIBC (g/dL) 100. The percent saturation is usually
between 20-50%, supporting an excess capacity for iron binding.
In cases of iron
overload, the % saturation increases dramatically. Saturation is moderately increased

in thalassemia and chronic anemia and in iron deficiency the saturation is decreased.
Ferritin is a large ubiquitous protein and the major iron storing protein in the body.
Ferritin serves to store thousands of iron atoms/molecule in a non-toxic form acting as
an iron reserve.
Ferritin is found in small amounts in the blood, where it can be
measured as an indication of overall iron reserves (1 ng/mL serum iron approximates 10

mg total storage iron). In the blood, ferritin is generally poor in iron content and is
referred to as apoferritin. Circulating ferritin (or apoferritin) is measured using specific
antibodies, commonly by chemiluminescent immunoassay. Serum or plasma ferritin
levels are produced in proportion to dietary iron absorption; serum ferritin is increased
with iron overload and decreased in iron deficiency. Serum ferritin levels change prior
to clinical and morphological manifestations of anemia (e.g. microcytosis) making it a
useful diagnostic marker of iron homeostasis. While considered the most useful of the
50
currently available tests for non-invasively assessing iron stores, ferritin is also an acute
phase reactant and may be normal or even increased when chronic infection or
inflammation occurs in combination with underlying iron deficiency anemia.
In
thalassemias, ferritin is typically elevated reflecting a state of iron overload; in contrast,

ferritin is decreased in iron deficiency making it a useful marker to differentiate causes
of microcytosis.
Transferrin is an iron transporting protein and negative acute phase reactant
produced primarily by the liver. As with ferritin, transferrin is routinely measured by
immunoassay. Most circulating iron is bound to transferrin, binding to Fe 3+ with very
high affinity. Transferrin transports iron absorbed in the GI tract to cells containing
specific receptors, in particular erythroid tissue. Transferrin delivers iron to cells via the
ubiquitously distributed transferrin receptor. Clinically, measurement of transferrin is
useful for hypochromic microcytic anemia workups. Transferrin is increased in iron
deficiency anemias, but normal or decreased in chronic anemia of disease, iron
overload, and thalassemias.
Transferrin is decreased in cases of liver disease,
nephropathy (or other protein loss or malabsorption), and inflammation.
Table 1. Iron Tests in Different Disorders

51
Disorder
Serum
Chronic
Anemia
of Disease
Iron Deficiency
Thalassemia
Hemochromatosi
TIBC
Transferrin
Ferritin
Iron
Saturation
or
or
or
or
or
or
or
s
decreased; within reference interval; increased
3.1.5 Soluble Transferrin Receptor

An additional test that is useful for diagnosis of anemia is the soluble transferrin
receptor (sTfR). The sTfR consists of the N-terminus of the membrane receptor that
can be measured in circulation. Circulating levels reflect the activity of the erythroid
bone marrow, where sTfR levels are decreased in cases of low red cell synthesis (renal
failure and aplastic anemia) and increased in patients with hemoglobinopathies. The
utility of sTfR measurement is that it can differentiate iron deficiency in cases of acute
inflammation because sTfR levels are not affected by inflammatory cytokines.
In
thalassemias, sTfR levels are generally increased in proportion to the severity of the
genotype. Despite the apparent advantages, sTfR testing is not widely used and is not
currently standardized.
3.1.6 Hepcidin
Discovered in 2000, hepcidin is a hormone involved in iron homeostasis.
Hepcidin is produced by the liver and negatively regulates iron balance by inhibiting
52
macrophage recycling and decreasing intestinal absorption. Thus, when iron stores are
replete, hepcidin levels are increased and when iron stores are low, hepcidin is
elevated. Similar to ferritin, hepcidin is an acute phase reactant, making interpretation
of circulating levels in patients with inflammation more challenging.
At the time of
writing, hepcidin testing was not available commercially. The hepcidin in human iron
stores and its diagnostic implications has been recently reviewed (Kroot JJC, Tjalsma
H, Fleming RE, Swinkels DW. Hepcidin in Human Iron Disorders: Diagnostic
Implications: Clin Chem 2011; 57(12): 1650-1669).
Additional Readings
Fairbanks VF, Klee GG. Biochemical aspects of hematology. In Fundamentals of
Clinical Chemistry. Edited by Tietz N. Saunders,1987,789-818.
Guarnone R, Centenara E, Barosi G. Performance characteristics of hemox-analyzer for
assessment of the hemoglobin dissociation curve. Haematologica 1995;80:426-430.
Pincus MR and Abraham NZ. Interpreting laboratory results. In: Henry's Clinical
Diagnosis and Management by Laboratory Methods (Clinical Diagnosis & Management
by Laboratory Methods) Edited by McPherson RA and Pincus MR. 21 st Edition.
Higgins T, Beutler E, Doumas BT. Hemoglobin, Iron and Bilirubin. In Tietz textbook of
clinical chemistry and molecular diagnostics. Edited by Burtis CA, Ashwood ER, Bruns
DE. Elsevier Saunders, 2006,1165-1208.
Marengo-Rowe AJ. Structure-function relations of human hemoglobins. Proc (Bayl Univ
Med Cent) 2006;19:239-245.
Mayomedicallaboratories.com/test-catalog. Accessed April 20, 2011.
Rees DC, Williams TN, Gladwin MT. Sickle-cell disease. The Lancet. 2010;376:20182031.
Steinberg MH. Genetic disorders of hemoglobin oxygen affinity. www.uptodate.com.
Accessed April 28, 2011.
Steinberg MH. Unstable hemoglobin variants. www.uptodate.com. Accessed April 28,
2011.
53
Tietz Textbook of Clinical Chemistry and Molecular Diagnostics. Edited by Burtis CA,
Ashwood ER, and Bruns DE. 5th Edition.
Vichinsky EP. Sickle cell trait. www.uptodate.com. Accessed April 28, 2011.
54
Chapter 3
3.2 Microcytosis
Diane Maennle, MD, and Kimberly Russell, MT (ASCP), MBA
Smaller-than-normal size of Red Blood Cells (RBCs) is defined as microcytosis.
This is quantified by calculating the mean corpuscular volume (MCV) using the following
formula employing the values of hematocrit and RBC count:
MCV = Hematocrit (HCT) X 10 / RBC Count (Million)
In adults, a MCV value of less than 80fL is defined as microcytosis. In pediatric
subjects, the MCV and hemoglobin range distinctly vary with age (Table I).
Table I Age Dependent Mean Hemoglobin and MCV Values 1,2,3,4
Age
Mean Hemoglobin (g/dL)
Mean MCV (fL)
3 to 6 months
11.5
91
6 months to 2 years
12.0
78
2 to 6 years
12.5
81
6 to 12 years
13.5
86
12 to 18 years (female)
14.0
90
12 to 18 years (male)
14.5
88
> 18 years (female)
14.0
90
> 18 years (male)
15.75
90
Iron deficiency anemia, -thalassemia trait, and -thalassemia trait are the most common
causes of microcytosis. However, other clinical conditions are also associated with microcytosis
(Table II).1,3,5,6 In addition to decreased MCV, the patients with -thalassemia trait usually have
55
increased hemoglobin A2. It is pointed out that lower hemoglobin A2 is also observed in patients
with concurrent deficiency of serum iron. Therefore, serum iron deficiency anemia must be ruled
out in order to correctly make the diagnosis of -thalassemia trait in such patients. Conversely,
patients with -thalassemia trait may acquire megaloblastic anemia or liver disease, and may
exhibit a normal range for MCV.7
Table II Diagnostic Reasons of Microcytosis (listed in decending order of frequency)

Children and adolescents Menstruating women
Men and non-menstruating women
Iron deficiency anemia
Thalassemia trait
Thalassemia trait
Anemia of chronic disease
Other hemoglobinopathies
Pregnancy
Unexplained anemia
Lead toxicity
Thalassemia trait
Chronic inflammation
Sideroblastic anemia
Several laboratory tests in addition to the CBC, e.g. serum iron, serum ferritin, total ironbinding capacity (TIBC), transferrin saturation, hemoglobin electrophoresis, and the examination
of the peripheral blood smears (by a pathologist or hematologist), are employed to provide
insight and etiologies of microcytosis (Table III).3,8
56
Table III Laboratory Tests in the Differential Diagnosis of Microcytosis

Suggested diagnosis
Test
Iron deficiency anemia Thalassemia
Serum ferritin
Decreased
Increased
Normal to increased
Normal to increased
RBC
Increased
distribution width
(RDW)
Normal to
increased
Normal
Increased
Serum iron
Decreased
Normal to
increased
Normal to
decreased
Normal to
increased
Total ironbinding capacity
Increased
Normal
Slightly increased
Normal
Transferrin
saturation
Decreased
Normal to
increased
Normal to slightly
decreased
Normal to
increased
Van Vranken3 has recently suggested a protocol for diagnosing the cause of microcytosis
(Figure 1). If the cause remains unclear, the diagnosis of hemoglobinopathy by methods besides
electrophoresis alone is imperative. Note: There is a type-setting error in the presentation of
the protocol suggested by Van Vranken.3 We have corrected this error in the figure 1, and
the journal (American Family Physician) editor was also informed.
57
58
Clinical observations of Kenneth F. Tucker, MD, FACP, a practicing hematologist for the
last forty years:
Ordinary hemoglobin electrophoresis (cellulose acetate or agarose gel
electrophoresis) was only able to detect the more common types of thalassemias. Although
there were several other types, many of them did not have microcytosis. I had a large
number of patients, who had -thalassemia minor and a few with probably -thalassemia,
in which the hemoglobin and hematocrit values were relatively normal. Microcytes may or
may not be present. This diagnosis was suggested by the peripheral smear, and proven by
additional laboratory tests (IFE, globin chain analysis, etc.).
I believe that RDW, which is the average red cell width and reflects standard
deviation of red cell volumes, is a very important test. RDW normal deviation is a bellshaped curve. When this value is 2-3% higher, it represents red cells which have varying
widths. This certainly can be seen in patients who are iron deficient with microcytosis, but
have normal or large cells in addition to megaloblastic or dysplastic marrows, elevated
reticulocytes, vitamin B12 or folic acid deficiency, and other conditions. Despite the
availability of automated cell counters, review of the peripheral film is one of the most
informative and rewarding tests that should be done (by pathologist or hematologist) in any
case in which the cause of anemia is not obvious, e.g., bleeding, pure iron deficiency, pure
vitamin B12 deficiency, etc. It is also emphasized that the RDW test is not sensitive or
specific enough to differentiate iron deficiency and -thalassemia trait. 9
A fairly low to extremely low ferritin is an excellent measure of iron deficiency
anemia. In my practice, regardless of what else is going on, any ferritin level of <10 ng/mL,
means there is iron deficiency. As mentioned above (Table III), elevated ferritin levels are
seen in refractory anemias, all types of chronic inflammatory conditions, etc. Since this test
59
is an acute phase reactant (similar to haptoglobin), it must not be used alone, as the ferritin
level may be normal in these clinical conditions.
Women in the second or third trimester are always anemic. This is similar to patients
who are hypervolemic because of renal or cardiac problems. Red cells in these cases are
not microcytes and when the hypervolemia is corrected, the hemoglobin and hematocrit
rises.
Severe anemia in childhood is usually due to the lack of iron in food, since cows
milk does not contain iron.
A nave reader is advised to also review the Full Color pdf of Complete Blood Count
in Primary Care, Best Practice Journal, June 2008 (www.bpac.org.nz),
especially the section on Hemoglobin and Red Cell Indices (page 15).
References
1. Richardson M. Microcytic anemia [published correction appear in Pediatr Rev. 2007;
28(7): 275, Pediatric Rev. 2009; 30(5): 181, and Pediatr Rev. 2007; 28(4):151]. Pediatr
Rev. 2007; 28(1): 5-14.
2. Beutler E, Waalen J. The definition of anemia: what is the lower limit of normal of the
blood hemoglobin concentration? Blood. 2006; 107(5): 1747-1750.
3. Van Vranken ML. Evaluation of Microcytosis. Am Fam Physician. 2010; 80(9): 1117-1122.
4. RBC indices calculation and laboratory procedure (2006). St. John Health Laboratories,
Warren, MI 48093.
5. Moreno Chulila JA, Romero Colas MS, Gutierrez Martin M. Classification of anemia for
gastroenterologist. World J Gastroenterol. 2009: 15(37):4627-4737.
6. Guralnik JM, Eisenstaedt RS, Ferrucci L, Klein HG, Woodman, RC. Prevalence of anemia
in persons 65 years and older in the United States: evidence for a high rate of
unexplained anemia. Blood. 2004; 104(8): 2263-2268.
7. Bain BJ. Hemoglobinopathy Diagnosis. 2nd ed. Malden, Mass.: Blackwell
Publishing; 2006: 94-106.
8. Hematologic diseases. In: Wallach J. Interpretation of Diagnostic Tests. 8 th ed. Boston,
Mass.: Little Brown and Company; 2006: 385-419.
9. Ntalos G, Chatzinikolaou A, Saouli Z, et al. Discrimination indices as screening
tests for beta-thalassemia trait. Ann Hematol. 2007; 86(7): 487-491.
60
Chapter 3
1
3.3
Hereditary Persistence of Fetal Hemoglobin

3.3.1 Introduction
Hereditary persistence of fetal hemoglobin or HPFH consists of a group of
genetic disorders characterized by the presence of a substantial elevation of fetal
hemoglobin (Hb F) in RBCs of heterozygotes, as well as of homozygotes and
compound heterozygotes for HPFH and other hemoglobinopathies. Increased levels of
Hb F can ameliorate the clinical course of hemoglobinopathies such as thalassemia
and sickle cell anemia. HPFH is usually due to deletions of different sizes involving the
-globin gene cluster, but nondeletion types of disorders have also been identified,
usually due to point mutations in the -globin gene promoters (reviewed in refs. 1-3).
Figure 1 diagrammatically illustrates the spatial organization of the -like globin genes in
the -gene cluster on chromosome 11. However, as discussed later in this chapter,
certain forms of nondeletion HPFH are clearly not linked to the -globin gene cluster.
61
Figure 1. Deletions of the -globin gene cluster associated with fusion proteins and
HPFH. The circle 3 to the -globin gene indicates the 3 -globin gene enhancer. The
filled vertical boxes at the 3 breakpoints of the HPFH-1 and HPFH-6 deletions indicate
the locations of DNA sequences with homology to olfactory receptor genes (adopted
from reference 2). The references for the individual mutations are cited in references 1,
3 and 6.
HPFH is frequently contrasted with thalassemia, which is another genetic
disorder associated with elevated Hb F levels. However, one should probably not
consider the two disorders as being unambiguously separate entities but rather as a
group of disorders with a variety of partially overlapping phenotypes that sometimes
defy classification as one syndrome or the other. The following is a working definition
that is generally applied to the classification of these disorders: thalassemia usually
refers to a group of disorders associated with a mild but definite thalassemia phenotype
of hypochromia and microcytosis together with a modest elevation of Hb F that, in
heterozygotes, is heterogeneously distributed among red cells.
In contrast, HPFH
refers to a group of disorders with substantially higher levels of Hb F, and in which there
is usually no associated phenotype of hypochromia and microcytosis. In addition, the
increased Hb F in heterozygotes with the typical forms of HPFH is distributed in a
relatively uniform (pancellular) fashion among all of the red cells rather than being
distributed in a heterogeneous (heterocellular) fashion among a subpopulation of socalled F cells, as in thalassemia. Homozygotes for both conditions totally lack Hb A
and Hb A2, indicating absence of - and -globin gene expression in cis to the
thalassemia and HPFH determinants.
Although the apparent striking qualitative
difference in cellular distribution of Hb F between HPFH and thalassemia may be

due in great part to the quantitative differences in the amount of Hb F per cell and the
62
sensitivity of the methods used to detect Hb F cytologically, nevertheless it would

appear that the increased amount of Hb F in HPFH is caused by a genetically
determined failure to suppress -globin gene activity postnatally in all erythroid cells,
rather than being due to selective survival of the normally occurring sub-population of F
cells such as occurs in sickle cell anemia, + and o thalassemia.
Nevertheless,
heterocellular forms of HPFH, without a -thalassemia phenotype, have been clearly

defined and characterized. Therefore, in the final analysis, there is definitely some
overlap between these two sets of syndromes at the level of their clinical and
hematological phenotypes, as well as at the level of their molecular basis.
3.3.2 Deletions Associated with the HPFH Phenotype.

Classic pancellular HPFH, with absence of -and -globin gene expression from
the affected chromosome, is associated with large deletions in the -globin gene cluster
that remove the -and -globin genes together with variable amounts of their 5 and 3
flanking DNA. At least nine different HPFH deletions of this type have been
characterized that vary in size or length from ~13 kb to ~ 85 kb (1-4), some of which are
illustrated in Fig. 1. The mechanisms by which such deletions cause marked elevation
of Hb F are not well understood, but a number of theories have been proposed.
One theory is based on the model of the proposed mechanism for the marked
elevation of Hb F associated with Hb Kenya. Hb Kenya is a structurally abnormal
hemoglobin that, like Hb Lepore, contains a "hybrid" or fused -like globin chain
resulting from a non-homologous crossing-over event between two globin genes in the
-gene cluster. However, whereas the Lepore crossover occurred between the - and
63
-globin genes, the Kenya gene resulted from crossover between the A- and -globin
genes (Fig. 1). The crossover occurred in the second exons of the A and genes,
between the codons for amino acids 80 to 87, and resulted in deletion of ~24 kb of DNA
between the A to the gene. Unlike Hb Lepore, that is associated with a thalassemic phenotype, Hb Kenya is associated with a phenotype of pancellular G
HPFH: erythrocytes of affected heterozygotes have normal red cell indices and contain
7-23% Hb Kenya as well as approximately 10% Hb F, all of which is of the G type and is
relatively uniformly distributed among the red cells. A proposed explanation for the
HPFH phenotype associated with Hb Kenya is the influence on the G- and Kenya gene
promoters of a well characterized enhancer element located in the 3' flanking DNA of
the -globin gene, shown by the filled circle in Fig. 1, that becomes translocated into
close proximity of the -globin gene promoters by the crossover/deletion event, resulting
in enhanced activity of these promoters.
Among the HPFH deletions, there is a relatively short deletion, called HPFH-5 or
Italian HPFH, that extends from a point ~3 kb 5' to the gene to a point 0.7 kb 3' to the
gene, deleting the gene but not its 3' enhancer. The molecular basis of the HPFH
phenotype associated with this deletion is presumably the influence of the translocated
3' -gene enhancer on the -gene promoters, in a manner analogous to that proposed
for the basis of the HPFH phenotype of the Hb Kenya syndrome. In the case of some of
the other larger HPFH deletions, the DNA preserved at or near the 3 breakpoint of the
deletions has been shown in various types of assays to have enhancer-like activity on
gene expression (2, 5-7). Thus, it has been proposed that the DNA sequences at the
HPFH 3' deletion breakpoints, that become juxtaposed to the genes as a result of the
64
deletion events, may influence -gene expression, in a manner analogous to the

presumed influence of the 3' -gene enhancer on -gene expression in Hb Kenya and
HPFH-5. Mechanisms by which this could occur include the presence of enhancer-like
sequences in the translocated 3' breakpoint DNA or the presence in this DNA of an
active chromatin configuration that could have a spreading and activation effect on the
expression of the neighboring -globin genes.
A second theory for the mechanism of increased -gene expression in deletiontype HPFH is the nature and function of the DNA sequences conserved at the 5
breakpoint of the deletions. The 5 breakpoints of the HPFH deletions, as well as many
of the -thalassemia deletions, are located in the DNA between the and genes,
the so-called -intergene DNA. It has long been proposed that there may exist
negative regulatory or silencer elements in this region of DNA, deletion of which in
HPFH but not in thalassemia, results in markedly impaired postnatal suppression of
-gene activity in all erythroid cells (8). A number of subsequent observations have
been made that support a role for the A-intergene region in the regulation of -gene
expression (reviewed in ref. 9). The Corfu deletion in particular, involving the -gene
and ~6 kb of upstream flanking DNA, is associated in homozygotes with a high HbF
phenotype and removes some interesting structural elements, such as a poly-pyrimidine
region that can serve as a binding site for a multi-protein chromatin remodeling complex
containing the transcription factor Ikaros, and a region of DNA that serves as a promoter
for the synthesis of an intergenic RNA transcript preferentially expressed in adult
erythroid cells (10). This region of DNA also appears to serve as a boundary region
between fetal and adult domains of the -globin gene cluster.
65
The most conclusive evidence for a functional role of the A-intergene DNA in
the regulation of gene expression consists of the observations by Sankaran et al. who
have extensively characterized a negative regulatory transcription factor, called
BCL11A, that down-regulates -gene expression in adult erythroid cells and that binds
to the A-intergene DNA (11-13). BCL11A, originally identified as an important factor in
B-lymphoid cell development, is a component of a multi-protein complex that plays a
negative regulatory role in -gene expression. This complex has been shown to contain
GATA1 as well as all components of the nucleosome and histone deacetylase ( NuRD)repressive complex (14). Additional studies have shown that this complex physically
interacts with another transcription factor called SOX6 that is thought to be a repressor
of embryonic and fetal globin gene expression (15). Chromatin immunoprecipitation
(ChIP) studies have shown that BCL11A binds to a number of regions in the -cluster,
including the upstream locus control region (LCR) and the intergenic region, but does
not bind to the - or -gene promoters (4, 14, 15). Sankaran et al. (4) have
characterized two important deletion mutants with nearly identical distal breakpoints but
different upstream breakpoints around the -gene and its flanking DNA. One mutant
with a more proximal breakpoint has a -thalassemia phenotype, whereas the longer
deletion removing 3.5 kb of additional upstream DNA is associated with a HPFH
phenotype. The authors propose that this 3.5 kb region of DNA contains a silencer
element, deletion of which can cause HPFH. This hypothesis is strengthened by the fact
that the deleted region contains one of the prominent binding sites of BCL11A detected
in their ChIP experiments. These findings provide very strong evidence for a -gene
silencer element in the -gene cluster that associates with a BCL11A-containing
66
repressor complex and that this interaction is an important factor in the suppression of
-gene expression during the perinatal switch from expression of Hb F to Hb A.
3.3.3 Nondeletion Forms of HPFH
In contrast to the deletional types of HPFH syndromes, where both linked G and
A
genes are over expressed, only one or the other gene is usually over expressed in
the best characterized nondeletional types of HPFH associated with high levels of
pancellular Hb F expression. However, less well characterized nondeletion forms of GA
HPFH have been described that are associated with relatively low levels of
heterocellular expression of both genes. Because of the restricted pattern of -globin
gene expression in the G and A forms of nondeletion HPFH, it was assumed that the
mutations in these syndromes must be located near the affected gene and molecular
studies focused initially on the DNA sequence analysis of the over expressed genes in
these disorders.
67
Table 1 adopted from reference 2. The one patient studied was doubly
heterozygous for Hb A and Hb C. About 20% of Hb F (or 8% of the total Hb) was of the
G
G
A
type, and the gene in cis to the -175 mutation carried the -158 C T change.
The references for the individual mutations are cited in references 1 and 3.
The results of these structural analyses revealed a number of different point
mutations in the promoter region of the over expressed gene in individuals with
different types of nondeletion HPFH, as listed in Table 1 (reviewed in refs. 1-3). These
point mutations have clustered primarily in three distinct regions of the 5'-flanking DNA
of the affected genes. The first region is located approximately 200 base pairs from
the "cap site" or site of transcription initiation of the genes (at least five different point
68
mutations involving single nucleotides between residues -195 to -202 from the cap site).
This region of DNA, which had not previously been suspected of playing a role in the
regulation of -gene expression, is very G+C rich and its sequence bears homology to
that of known control elements of other genes that contain the binding site for the
ubiquitous trans-acting protein factor called Sp1. Subsequent studies of the -gene
promoters have demonstrated that the -200 region is also a binding site for Sp1 and for
at least one other ubiquitous DNA binding protein.
The second region containing a mutation associated with nondeletion HPFH is
located at position -175. A point mutation (T->C) at this position of either the G or A
gene is associated with a phenotype of pancellular HPFH with high levels of Hb F (1525%). This region of DNA is noteworthy because it contains an octanucleotide
sequence that is present in the promoter region of a number of genes and is the binding
site of another ubiquitous trans-acting factor called OCT-1. In addition, the octamer
consensus sequence of the -gene promoters is flanked on either side by a consensus
sequence for the hematopoietic-specific transcription factor GATA-1. The point mutation
at position -175 affects the one nucleotide that is present in the partially overlapping
binding sites of both OCT-1 and GATA-1.
The third region affected by a point mutation in nondeletion HPFH is in the area
of a well known regulatory element of globin and other genes: the CCAAT box
sequence. In the genes, the CCAAT box is duplicated and the mutation associated
with the Greek A type of nondeletion HPFH is a G->A substitution at position -117, 2
bases upstream of the distal CCAAT box of the A-globin gene promoter. The base
change disrupts a pentanucleotide sequence, YYTTGA (Y = pyrimidine), that is highly
69
conserved immediately upstream of the CCAAT sequence in all animal fetal and
embryonic genes. At least two other mutations involving the CCAAT box of one or the
other gene have been reported in other cases of HPFH not associated with large
deletions. The CCAAT box region is known to be the binding site of a number of transacting factors, including the ubiquitous factors CCAAT binding protein (CP1) and
CCAAT displacement factor (CDP) as well as the erythroid-specific factor NF-E3.
The unifying model by which these various mutations are thought to affect
hemoglobin switching proposes that these base changes alter the binding of a number
of different trans- acting factors to critical regions of the -gene promoters and thereby
prevent the normal postnatal suppression of -gene expression (reviewed in refs. 1,2).
The mutations could prevent the binding of negative regulatory factors or enhance the
binding of positive regulatory factors. Either mechanism could be operative with one
mutation or the other.
3.3.4 HPFH Unlinked to the -Globin Gene Cluster

A number of studies have identified families in which increased levels of Hb F are
inherited due to a genetic determinant that is unlinked to the -globin gene cluster.
Genome-wide association studies (GWAS), using co-inheritance of single nucleotide
polymorphisms (SNPs) with elevated levels of Hb F, have subsequently demonstrated
the presence of two different quantitative trait loci (QTLs), unlinked to the -globin gene
cluster on chromosome 11, that are associated with inheritance of mildly elevated levels
of Hb F, similar to the phenotype seen in Swiss-type heterocellular HPFH (see section
above on Nondeletion HPFH). These loci are located on chromosome 2 and 6 (16, 17).
70
The locus on chromosome 2 corresponds to the site of the gene encoding BCL11A and
its identification led to the elegant studies of Sankaran and co-workers demonstrating
the role of BCL11A in the regulation of -gene expression. The locus on chromosome 6
is located between the genes encoding HBS1L and MYB. The mechanism by which this
locus causes elevation of Hb F is thus far poorly understood. Finally, mutations in the
gene on chromosome 19 encoding the erythroid-specific transcription factor EKLF1
have been shown to be associated with a form of HPFH (18, 19). The involved
mechanism is probably through the regulation of BCL11A levels, because it has been
demonstrated that EKLF1 binds to the promoter of the BCL11A gene and regulates the
expression of the gene (20).
3.3.5 Conclusion
Significant insights into the normal regulation of expression of the human globin gene cluster have been obtained by a detailed analysis of a group of disorders
called HPFH. On the basis of this information, several important regulatory elements
have been identified for the normal functioning of the individual genes in the cluster
during the developmental switch from fetal to adult hemoglobin gene expression, as well
as for the abnormal persistent expression of the -globin genes in adults with HPFH.
These results provide a more sophisticated understanding of the molecular basis of
these syndromes and point to certain strategies for potential future molecular and
cellular therapies for globin gene disorders.
71
3.3.6 Hemoglobin F Quantification

Hb F can be quantified by several methods, and the most commonly used
procedures in a clinical laboratory are a) radial immunodiffusion, b) Elisa method,
c) HPLC, and d) capillary zone electrophoresis.
References
1. Bollekens JA, Forget BG. Delta beta thalassemia and hereditary persistence of fetal
hemoglobin. Hematol Oncol Clin North Am 1991;5(3):399-422.
2. Forget BG. Molecular basis of hereditary persistence of fetal hemoglobin. Ann N Y
Acad Sci 1998; 850:38-44.
3 Weatherall DJ, Clegg JB. The Thalassaemia Syndromes. 4th ed. Oxford ; Malden,
MA: Blackwell Science; 2001.
4. Sankaran VG, Xu J, Byron R, et al. Functional element necessary for fetal
hemoglobin silencing. N Engl J Med 2011; 365(9):807-14.
5. Feingold, EA, Forget BG. The breakpoint of a large deletion causing hereditary
persistence of fetal hemoglobin occurs within an erythroid DNA domain remote from the
-globin gene cluster. Blood 1989; 74: 21782186.
6. Kosteas T, Palena A, Anagnou NP. Molecular cloning of the breakpoints of the
hereditary persistence of fetal hemoglobin type-6 (HPFH-6) deletion and sequence
analysis of the novel juxtaposed region from the 3' end of the beta-globin gene cluster.
Hum Genet. 1997;100: 441-5.
7. Anagnou NP, Perez-Stable C, Gelinas R, et al. Sequences located 3' to the
breakpoint of the hereditary persistence of fetal hemoglobin-3 deletion exhibit enhancer
activity and can modify the developmental expression of the human fetal A gammaglobin gene in transgenic mice. J. Biol Chem 1995; 270: 10256-63.
8. Huisman TH, Schroeder WA, Efremov GD, et al. The present status of the
heterogeneity of fetal hemoglobin in beta-thalassemia: an attempt to unify some
observations in thalassemia and related conditions. Ann N Y Acad Sci 1974;232(0):10724.
9. Bank A, O'Neill D, Lopez R, et al. Role of intergenic human - -globin sequences in
human hemoglobin switching and reactivation of fetal hemoglobin in adult erythroid
cells. Ann N Y Acad Sci 2005;1054:48-54.
10. Chakalova L, Osborne CS, Dai YF, et al. The Corfu thalassemia deletion disrupts
-globin gene silencing and reveals post-transcriptional regulation of HbF expression.
Blood 2005;105:2154-60.
11. Sankaran VG, Xu J, Orkin SH. Transcriptional silencing of fetal hemoglobin by
BCL11A. Ann N Y Acad Sci. 2010;1202:64-8.
72
12. Sankaran VG, Xu J, Ragoczy T, et al. Developmental and species-divergent globin

switching are driven by BCL11A. Nature 2009;460(7259):1093-7.
13. Sankaran VG, Nathan DG. Reversing the hemoglobin switch. N Engl J Med 2010;
363(23):2258-60.
14. Sankaran VG, Menne TF, Xu J, et al. Human fetal hemoglobin expression is
regulated by the developmental stage-specific repressor BCL11A. Science 2008;
322(5909):1839-42.
15. Xu J, Sankaran VG, Ni M, et al. Transcriptional silencing of -globin by BCL11A
involves long-range interactions and cooperation with SOX6. Genes Dev 2010; 24:78398.
16. Thein SL, Menzel S, Lathrop M, Garner C. Control of fetal hemoglobin: new insights
emerging from genomics and clinical implications. Hum Mol Genet 2009;18(R2):R21623.
17. Galarneau G, Palmer CD, Sankaran VG, Orkin SH, Hirschhorn JN, Lettre G. Fine
mapping at three loci known to affect fetal hemoglobin levels explains additional genetic
variation. Nat Genet 2010;42(12):1049-51.
18. Borg J, Papadopoulos P, Georgitsi M, et al. Haploinsufficiency for the erythroid
transcription factor KLF1 causes hereditary persistence of fetal hemoglobin. Nat Genet
2010;42(9):801-5.
19. Borg J, Patrinos GP, Felice AE, Philipsen S. Erythroid phenotypes associated with
KLF1 mutations. Haematologica 2011; 96:635-8.
20. Zhou D, Liu K, Sun CW, Pawlik KM, Townes TM. KLF1 regulates BCL11A
expression and - to -globin gene switching. Nat Genet 2010; 42:742-4.
73
Chapter 3
3.4 Flow Cytometry Measurements of Cellular Fetal Hemoglobin, Oxidative
Stress and Free Iron in Hemoglobinopathies
Eitan Fibach, MD
3.4.1 Flow Cytometry of Blood Cells
Flow cytometry (FC) is a common methodology in clinical diagnostic and
research laboratories. In hematology, it is mainly used for diagnosis, prognosis,
determining therapy efficacy and follow up of patients with leukemia or lymphoma
(1). It is also used for diagnosis of red blood cell (RBC) abnormalities such as in
Paroxysmal Nocturnal Hemoglobinuria (2) and hereditary spherocytosis (3). In
this review, I will summarize FC methodologies for analysis of RBC (and other
blood cells) from patients with hemoglobinopathies with respect to their fetal
hemoglobin (HbF) and free iron (labile iron pool, LIP) contents and parameters of
the oxidative state.
FC analyzes individual cells in a liquid medium. Most techniques use antibodies
directed against internal (following fixation and premeabilization of the
membrane) or surface antigens. The antibodies are labeled with fluorescence
probes (fluochromes) either directly or indirectly (by a secondary antibody). In
addition to antibodies, other fluorescent compounds can be used. For example,
propidium iodide, which binds stochiometrically to nucleic acids, is commonly
used for determining cell viability and their distribution in the cell cycle (4).
Following staining, the cells are analyzed by a flow cytometer; they are first
hydro-dynamically focused in a narrow sheath of physiological solution before
74
being intercepted by one or more laser beams resulting in light scatter and
fluorescence emission. Depending on the number of laser sources and
fluorescence detectors, several parameters (commonly 6, but up to 18) can be
simultaneously detected on each cell: Forward light scattering and side light
scattering provide correlates with regards to size and granularity of the cells,
respectively, and fluorescence light emission by the fluorochromes correlates
with the expression of different antigens as well as other cellular parameters (see
below).
FC is superior to other techniques in several aspects: (I) Technology is widely available
as mentioned above, most hematology and immunology laboratories use FC for both diagnosis
and research purposes. (II) Only cell-associated fluorescence is measured, excluding soluble or
particulate fluorescence. (III) Each cell is analyzed individually, but since measurement is rapid
(msec), a large number of cells can be analyzed (ranging from 0.1-10 x10 5 cells) within a few
minutes. The results are therefore statistically sound even for very small sub-populations. (IV)
Various sub-populations can be identified and measured simultaneously. (V) The method
produces mean values for each sub-population, and therefore avoids the inaccuracy of
biochemical methods that produce mean value for the whole population. This is of crucial
importance when mixed populations are studied. (VI) The procedure can be automated to permit
high throughput analysis (e.g., for screening of large libraries of compounds for inducers of
HbF). Although the FC data are expressed in arbitrary fluorescence units rather than weight or
molar concentrations, they are useful for comparative purposes.
FC is especially fitting for analysis of blood cells: (I) These cells which can be easily
obtained by blood drawing are present as single cells, thus in contrast to cells of solid tissues,
75
their use does not require harsh procedures for tissue disaggregation (e.g., trypsinization). (II)
They are present as a mixture of various cell types, including numerous subtypes (e.g.,
lymphocytes), with very large (e.g., RBC) to very small (hematopoietic stem cells)
representation. Cells of these sub-types can be identified and "gated" based on differences in
their size (forward light scattering), granularity (side light scattering) and expression of surface
antigens, and can be measured simultaneously. For measurements of various characteristics
(HbF content, oxidative stress parameters and LIP content), the blood sample is stained with
specific probes (as detailed below), and then with fluorescent reagents (usually antibodies)
against surface markers which identify a specific subpopulation. Such markers are glycophorin A
for RBC, CD61 for platelets, CD15 for neutrophils, CD19 for B-lymphocytes and CD3 for Tlymphocytes. CD45 is particularly useful since it is differentially expressed on various nucleated
blood cells (Fig. 1).
76
PMN
RBC
Monocytes
Lymphocytes
CD45
Fig. 1. Flow cytometry of blood cells. A dot plot of blood cells with respect CD45 (FL3-H) and
side light scatter (SSC-H).
3.4.2 Measurement of Fetal Hemoglobin-Containing Erythroid Cells

Fetal hemoglobin (HbF, 22) is the major hemoglobin (Hb) in the prenatal period
that is largely replaced after birth by the adult Hb (HbA, 22) (5). In adults, less than 1%
of the Hb content is HbF which is concentrated in a few RBC, called F-cells (6). High
levels of HbF are frequently seen in hemoglobinopathies (7). Measurement of HbF (as
well as HbA, sickle hemoglobin, HbS, etc.) can assist in diagnosis and in determining
the efficacy of treatment. HbF can be measured by a variety of techniques. Most of the
techniques measure HbF in lysates prepared from RBC. These techniques include
77
spectrofluorometric measurements following treatment with alkaline (to destroy non-fetal

hemoglobins) and staining with benzidine (8), chromatography (ion-exchange HPLC for
hemoglobins and reverse-phase HPLC for globin chains) (9), as well as immunological
techniques, such as Elisa, based on antibodies against HbF (10). However, quantitative
FC measurement of RBC, fluorescently stained with antibodies to HbF (as well as for
the other hemoglobins), has several advantages. For example, in the differential
diagnosis of Hereditary Persistence of Fetal Hemoglobin (11). This condition
encompasses a heterogeneous group of disorders with marked increased levels of HbF.
Based on the cellular distribution of HbF, they are characterized as pan-cellular, where
all RBCs have increased levels of HbF, albeit not always uniformly so; and heterocellular, where nearly all the HbF is confined to a minor, distinct subpopulation of RBCs.
This important distinction is most reliably ascertained by FC.
Epidemiological studies have indicated that high levels of HbF improve the
clinical symptoms of the underlying disease. In sickle cell anemia not only do HbFcontaining cells have a lower concentration of sickle hemoglobin, but HbF inhibits
polymerization of HbS directly, accounting for the lower propensity of such cells to
undergo sickling (12). In -thalassemia, elevated HbF should compensate partially for
the deficiency of -globin chains and reduce the excess of -globin chains. Several
pharmacological agents have been used to stimulate HbF production (13). Hydroxyurea
(HU) is currently the drug of choice (14). When patients are monitored during HU
treatment by measuring HbF in the hemolysate, an increase is usually observed after 23 months (10). HU acts by a still unknown mechanism on the early erythroid precursors
in the bone marrow. It takes several weeks for HbF to accumulate in the peripheral
78
blood to a quantity that allows differences before and after treatment to become
apparent. Measuring differences in F-RBC by FC may be more sensitive, and
measuring F-reticulocytes (retics) may provide early indication of treatment efficacy
(15): Retics have a very short life-span (1-2 days) compared to mature RBC (120 days
in normal subjects) and therefore measuring peripheral blood F-retics more closely
characterizes the current status of HbF production in the bone marrow. Measuring Fretics can indicate the efficacy of the drug and/or the patients compliance several days
after treatment initiation. Such follow up is very important since about 30% of the
patients are non-responders. It is imperative that such patients be identified as early as
possible and the treatment (that is not without potential risks) be discontinued and
replaced by treatment with another drug (e.g., butyroids).
3.4.3 Staining Protocols for F-RBC and F-Retics (15)

Heparinized blood is washed three times in phosphate buffered saline (PBS). For
fixation, 50l of the packed cells are resuspended in 10 ml of PBS containing 4%
formaldehyde for 15-min at room temperature under constant agitation in polypropylene
tubes. For permeabilization, the cells are centrifuged for 3 min at 1,500 g, and 2 ml
methanol-acetone are added to the pellet, mixed and incubated for 1-min at room
temperature. The cells are then washed three times and resuspended in PBS to a final
volume of 0.5 ml (10% suspension).
Anti-HbF monoclonal antibodies (the amount depends on the Manufacturers
instructions or on a pre-performed titration) are added to 5x10 6 cells (5 l of the 10%
suspension) and incubated for 1-hr at 370C, after which the cells are washed in PBS. If
79
the antibodies are fluorochrome-conjugated, the cells are resuspended in PBS and
analyzed directly. In the case of unconjugated antibodies, a secondary antibody
(fluorochrome-conjugated rabbit F(ab)2 anti-mouse IgG) is added for 30-min at room
temperature. For the F-retic count, the blood cells are double labeled with
phycoerythrin-conjugated antibodies to HbF and thiazol orange, a specific nucleic acid
binding green fluorescence dye.
Following staining, the cells are washed and resuspended in PBS and analyzed by FC.
For "acquisition", the threshold is set on forward light scatter to exclude debris and
platelets. Cells are run at about 1000 cells/sec using logarithmic amplification, and data of
104-105 cells are accumulated. RBC are gated based on their forward light scatter and
side light scatter. When the sample is also stained with thiazol orange, RBC are gated based on
their negative staining with thiazol orange, retics - based on their weak staining (because they
contain remnants of RNA) and nucleated cells (including normoblasts) based on their intense
staining; HbF is then specifically determined for each cell population (Fig. 2).
80
Fig. 2. Flow cytometry analysis of F-RBC and F-Retics. Blood cells stained with thiazol-orange
(T.O) and anti-HbF. A. Forward light scatter (FSC) vs. T.O. RBC (negative T.O staining) and
retics (intermediate T.O staining) were gated and their HbF determined (B and C), respectively.
3.4.4 F-Cell Determination for Fetal-Maternal Hemorrhage (FMH) in Pregnant Patients

with -Thalassemia A Single Case and General Conclusion (16)
F-cell analysis is commonly used to detect fetal-maternal hemorrhage (FMH)
where fetal RBC enter the maternal blood circulation due to fetal or maternal trauma or
a placental defect (17). These RBC of fetal origin can be distinguished from the
maternal adult RBC by their fluorescence following staining with an antibody to HbF.
Recently, in order to increase the sensitivity, reproducibility and accuracy of the assay,
81
another marker was introduced carbonic anhydrase (CA) (18). The CA isoenzymes
that are mainly represented by CAI and CAII (19) are fully expressed in the RBC only
after birth (20,21). The "Fetal Cell Count kit" manufactured by IQ Products (Groningen,
the Netherlands), which uses a combination of a murine monoclonal antibody directed
to HbF and a polyclonal antibody to the CAII isoform, has significantly improved this
assay (11,18). Most of the RBC of fetal origin do not express CA but highly express HbF
(CA-HbF++), while RBC in adult blood express CA but do not express HbF (CA +HbF-).
Some adult F-cells which express CA and HbF (CA+HbF+) can be differentiated from
fetal F-cells (CA-HbF++) present in FMH based on the extent of HbF and CA expression.
Until recently, -thalassemia major was lethal. Improvements in treatment, such as the
introduction of blood transfusions and iron chelation, have considerably improved the life
expectancy as well as the quality of the patients life, including the ability of thalassemic women
to give birth. Recently, we were confronted with a case of a possible FMH in a -thalassemic
woman. To establish the usefulness of the CA/HbF procedure, i.e. differentiating between fetal
RBC and the maternal RBC, we screened non-pregnant -thalassemic patients (men and
women). The results demonstrated, in addition to adult non-F RBC (CA +HbF-) and adult F-RBC
(CA+HbF+), two other sub-populations, CA+HbF++ and CA-HbF++. The presence in these patients
of the latter RBC phenotype, which characterizes fetal cells, precludes the use of the CA/HbF
method for the detection of FMH in thalassemia.
3.4.5 Oxidative Stress
The oxidative status of cells is determined by the balance between pro-oxidants
and antioxidants. The reactive oxygen species (ROS) are pro-oxidants which are
generated in most cells mainly during energy production. Although important for various
82
aspects of normal physiology (e.g., signal transduction), ROS interact with and damage
various cell components when they are in excess. To protect against the deleterious
effects of ROS, cells maintain an effective antioxidant system consisting of water- or
lipid-soluble antioxidants and enzymes that remove ROS by metabolic conversion.
When the oxidant/anti-oxidant balance is tilted in favor of the oxidants, oxidative stress
ensues (22). Although oxidative stress is not the primary etiology of
hemoglobinopathies, it mediates several of their pathologies, including hemolysis which
results in chronic anemia. Hemolysis occurs both in the bone marrow, where developing
erythroid precursors undergo enhanced apoptosis (ineffective erythropoiesis) and in the
peripheral blood, where mature RBC undergo lysis in the blood vessels (intra-vascular
hemolysis). Destruction also occurs in reticuloendothelial tissues, such as the spleen,
where mature RBC undergo phagocytosis by resident macrophages (extra-vascular
hemolysis) (22).
Various factors are responsible for oxidative stress in RBC of patients with hemoglobinopathies. In -thalassemia, excess -globin chains form unstable tetramers that
dissociate into monomers and then are oxidized, first to met-Hb and then to
hemichromes which precipitate intracellularly with time (23). Following the release of
heme and iron, there is deposition of the protein moiety on the plasma membrane. The
outcome of this chain of events is enhanced formation of ROS, catalyzed by free iron,
with a variety of deleterious effects on the membrane lipids and proteins, including
oxidation of the membrane protein band 4.1 and a decrease in spectrin/band3 ratio (24).
In -thalassemia, the - and -globins, which are produced in excess, do not precipitate
right away, but form the soluble tetramers 4 (Hb Barts) and later the 4 (HbH), which
83
are less stable than HbA and have an increased susceptibility towards oxidation and
hemichrome formation (23). In sickle cell disease, met-HbS is produced at a higher rate
and is less stable than met-HbA resulting in formation of hemichromes, and release of
heme and iron, with resultant denaturation and precipitation as Heinz bodies (25).
Many approaches have been devised to quantify oxidative stress and its damage
as well as the effects of treatment with anti-oxidants (22). Most of these methods assay
the content of body fluids (mainly blood). FC can be utilized for measurements of
oxidative stress parameters in various blood cells. Although the major target of oxidative
stress in hemoglobinopathies is the RBC, other blood cells are affected as well. Thus,
defects in the abilities of polymorphonuclear cells to adhere to, engulf and lyze bacteria
may result in recurrent infections. Chronic activation of platelets may cause
thromboembolic complications (26,27). In order to study the effects of oxidative stress
on the spectrum of symptoms in hemoglobinopathies, all blood cell lineages should be
studied.
FC of oxidative stress parameters utilizes various probes: ROS can be measured
by staining cells with the non-polar compound, 2-7-dichlorofluorescein diacetate. It
readily diffuses across the membrane and becomes deacetylated by
esterases into a polar derivative that is trapped inside the cells. When it is oxidized by
ROS (mainly peroxides), a green fluorescent product dichlorofluorescin is produced
(28). The intensity of the fluorescence is proportional to the cellular concentration of
ROS. The applicability of the method was validated by the increased fluorescence
measured following treatment with ROS-generating agents such as hydrogen peroxide
and t-butylhydroxyperoxide and with the catalase inhibitor sodium azide, while treatment
84
with ROS scavengers such as N-acetyl cysteine decreased the fluorescence. ROS can
also be measured by dihydrorhodamine 123, which freely enters into cells, and after
oxidation by ROS to rhodamine 123 emits a bright red fluorescence (29).
Reduced glutathione (GSH), the main cellular antioxidant, can be measured
using mercury orange (26), which forms fluorescent adducts with GSH via the
sulphydryl group, producing an S-glutathionyl derivative that emits red-orange
fluorescence (30). The probe reacts more rapidly with non-protein thiols, such as GSH,
compared with thiol-containing proteins, allowing specificity under controlled staining
conditions (31). The validity of this method was confirmed by demonstrating that Nethylmaleimide, which totally blocks thiol groups, decreased the fluorescence in a dosedependent manner. To ascertain that non-protein thiols are being stained, cells were
incubated with diethylmaleate, a specific non-protein thiol-depleting agent. This weak
electrophil of the ,-unsaturated carbonyl group, which reacts with GSH only in the
presence of glutathione transferase, markedly suppressed the mercury orange
fluorescence, suggesting that GSH was the principle thiol which was stained by the dye
(32). Although there is no direct proof that the probe is specific for GSH, the assay
measures predominantly GSH, since it is the main non-protein thiol constituent of the
cellular thiol pool (33).
Other parameters of oxidative stress measured by FC are membrane lipid
peroxidation by staining with fluor-DHPE (26), and externalization of
phosphatidylserine (PS) moieties, a marker of damage to the membrane lipid, by
fluorochrome-conjugated annexin-V (34).
85
3.4.6 Staining Protocols for ROS and GSH

ROS Assay Blood cells are incubated with 2'-7'-dichlorofluorescin diacetate, dissolved
in methanol, at a final concentration of 0.4 mM. After incubation at 37C for 15 min, the cells are
washed and re-suspended in PBS to the original cell concentration.
GSH Assay - Blood cells are washed with PBS and then spun down. The pellet is incubated for
3 min. at room temperature with 40 M (final concentration) of mercury orange. A 100 M stock
solution of mercury orange is made up in acetone and stored at 4C. In both cases, cells are
then washed and resuspended in PBS, and analyzed by FC.
Fig. 3 shows FC measurements of ROS and GSH in normal and thalassemic RBC. The
results indicate that thalassemic RBC have higher ROS but lower GSH contents than
normal RBC.
86
Fig. 3. Flow cytometry of ROS and GSH in normal and thalassemic RBC. Blood cells derived
from a normal donor (A,C) and a thalassemic patient (B,D) were stained for ROS (A,B) and GSH
(C,D) following 1-h pre-incubation with (white) or without (pink) 2 mM H 2O2. Histograms of RBC
are shown.
3.4.7 Intracellular Free Iron

Another contributor to oxidative stress in cells is excess of iron. Iron overload is
generated in thalassemic or sickle RBC as a result of Hb-instability as discussed above.
In addition, iron accumulates in these diseases as a result of increased absorption from
the intestinal mucosa and by a failure to dispose of excess iron acquired by frequent
therapeutic blood transfusions (35). Moreover, iron-containing compounds (Hb or
87
hemin) that are released during hemolysis can add to the iron load and further
aggravate the hemolysis.
Normally, iron is transported in the circulation bound to transferrin and is
transferred into cells through the surface transferrin-receptor (36). Most of the
intracellular iron is firmly bound to various components such as Hb, heme and
cytochrome C; excess is stored in ferritin (37). In iron overload, serum iron which
exceeds the binding capacity of transferrin is present in the form of non-transferrin
bound iron (38). This iron can be taken up through a transferrin-independent pathway,
to form the cellular unbound "labile iron pool" (LIP) (16). The small fraction of LIP was
suggested as a low molecular weight intermediate or transitory pool between
extracellular iron and cellular firmly-bound iron (39). LIP is redox active and it
participates in generation of free radicals by the Fenton and Haber-Weiss reactions and
consequently in cell and tissue damage (40).
Since iron overload plays an important role in the pathology of transfused
patients with -hemoglobinopathies, the patients are commonly treated with iron
chelators. The three chelators currently in clinical use are deferioxamine, deferiprone
and deferasirox (41). Evaluation of iron overload is important for assessing its severity
and for determining the efficacy of iron chelation therapy. The parameters usually tested
are serum ferritin protein level and transferrin iron saturation. However, serum ferritin is
an acute phase reactant that may increase by iron-independent factors such as
infection, inflammation and liver disease (42). In addition, serum ferritin levels often fail
to predict impending cardiac iron overload and ensuing cardio-myopathies (43). The
advent of non-invasive proton relaxation assays (by NMR R2* or T2*) of organs has
88
provided a significant advance in monitoring iron overload, although, similarly to serum

ferritin, substantial changes in these parameters are seen only weeks to months after
the initiation of chelator treatment. In addition, these techniques require expensive
instrumentation that is not always available.FC quantification of the LIP content in
various blood cell types overcomes many of these problems.
3.4.8 Staining Protocol for LIP

Cells are washed twice with saline and incubated at a density of 1x10 6 per ml for 15 min
at 37oC with 0.25 M Calcein Acetoxymethyl Ester (CA-AM). After wash, the cells are treated
with or without Deferiprone (L1, 100 M). Fig. 4 shows the results of LIP measurements in RBC.
LIP is defined as the difference between histograms of cells treated or untreated with L1.
89
Fig. 4. Flow cytometry of labile iron pool (LIP) in RBC. Blood cells were loaded with calcein, then
washed and treated with or without the iron chelator Deferiprone (L1). Distribution fluorescence
(FL1-H) histograms are shown. LIP is defined as the difference between the mean fluorescence
channels of histograms of cells treated or untreated with L1.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Virgo PF, Gibbs GJ. Flow cytometry in clinical pathology. Ann Clin Biochem 2012; 49(Pt
1): 17-28.
Sutherland DR, Keeney M, Illingworth A. Practical guidelines for the high-sensitivity
detection and monitoring of paroxysmal nocturnal hemoglobinuria clones by flow
cytometry. Cytometry B Clin Cytom 2012; 82(4): 195-208.
Kedar PS, Colah RB, Kulkarni S, Ghosh K, Mohanty D. Experience with eosin-5'maleimide as a diagnostic tool for red cell membrane cytoskeleton disorders. Clin Lab
Haematol 2003; 25(6): 373-6.
Krishan A. Rapid flow cytofluorometric analysis of mammalian cell cycle by propidium
iodide staining. J Cell Biol 1975; 66(1): 188-93.
Peterson KR. Hemoglobin switching: new insights. Curr Opin Hematol 2003; 10(2): 1239.
Boyer SH, Belding TK, Margolet L, Noyes AN. Fetal hemoglobin restriction to a few
erythrocytes (F cells) in normal human adults. Science 1975; 188(4186): 361-3.
Bunn H, Forget B. Hemoglobins: Molecular, Genetic and Clinical Aspects. Philadelphia:
WB Saunders Co.; 1986.
Fibach E. Measurement of total and fetal hemoglobin in cultured human erythroid cells by
a novel micromethod. Hemoglobin 1993; 17(1): 41-53.
Huisman TH. Separation of hemoglobins and hemoglobin chains by high-performance
liquid chromatography. J Chromatogr 1987; 418: 277-304.
Epstein N, Epstein M, Boulet A, Fibach E, Rodgers GP. Monoclonal antibody-based
methods for quantitation of hemoglobins: application to evaluating patients with sickle cell
anemia treated with hydroxyurea. Eur J Haematol 1996; 57(1): 17-24.
Leers MP, Pelikan HM, Salemans TH, Giordano PC, Scharnhorst V. Discriminating
fetomaternal hemorrhage from maternal HbF-containing erythrocytes by dual-parameter
flow cytometry. Eur J Obstet Gynecol Reprod Biol 2007; 134(1): 127-9.
Benesch RE, Edalji R, Benesch R, Kwong S. Solubilization of hemoglobin S by other
hemoglobins. Proc Natl Acad Sci U S A 1980; 77(9): 5130-4.
Gambari R, Fibach E. Medicinal chemistry of fetal hemoglobin inducers for treatment of
beta-thalassemia. Curr Med Chem 2007; 14(2): 199-212.
Steinberg MH. Determinants of fetal hemoglobin response to hydroxyurea. Semin
Hematol 1997; 34(3 Suppl 3): 8-14.
Amoyal I, Fibach E. Flow cytometric analysis of fetal hemoglobin in erythroid precursors
of beta-thalassemia. Clin Lab Haematol 2004; 26(3): 187-93.
90
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
Prus E, Fibach E. Heterogeneity of F-cells in -thalassemia. Transfusion 2012, in press.

Sebring ES, Polesky HF. Fetomaternal hemorrhage: incidence, risk factors, time of
occurrence, and clinical effects. Transfusion 1990; 30(4): 344-57.
Porra V, Bernaud J, Gueret P, Bricca P, Rigal D, Follea G, Blanchard D. Identification and
quantification of fetal red blood cells in maternal blood by a dual-color flow cytometric
method: evaluation of the Fetal Cell Count kit. Transfusion 2007; 47(7): 1281-9.
Tashian RE. The carbonic anhydrases: widening perspectives on their evolution,
expression and function. Bioessays 1989; 10(6): 186-92.
Brady HJ, Edwards M, Linch DC, Knott L, Barlow JH, Butterworth PH. Expression of the
human carbonic anhydrase I gene is activated late in fetal erythroid development and
regulated by stage-specific trans-acting factors. Br J Haematol 1990; 76(1): 135-42.
Aliakbar S, Brown PR. Measurement of human erythrocyte CAI and CAII in adult,
newborn, and fetal blood. Clin Biochem 1996; 29(2): 157-64.
Fibach E, Rachmilewitz EA. The role of antioxidants and iron chelators in the treatment of
oxidative stress in thalassemia. Ann N Y Acad Sci 2010; 1202: 10-6.
Rachmilewitz EA. Formation of hemichromes from oxidized hemoglobin subunits. Ann N
Y Acad Sci 1969; 165(1): 171-84.
Advani R, Sorenson S, Shinar E, Lande W, Rachmilewitz E, Schrier SL. Characterization
and comparison of the red blood cell membrane damage in severe human alpha- and
beta-thalassemia. Blood 1992; 79(4): 1058-63.
Winterbourn CC. Oxidative denaturation in congenital hemolytic anemias: the unstable
hemoglobins. Semin Hematol 1990; 27(1): 41-50.
Amer J, Fibach E. Oxidative status of platelets in normal and thalassemic blood. Thromb
Haemost 2004; 92(5): 1052-9.
Amer J, Fibach E. Chronic oxidative stress reduces the respiratory burst response of
neutrophils from beta-thalassaemia patients. Br J Haematol 2005; 129(3): 435-41.
Bass DA, Parce JW, Dechatelet LR, Szejda P, Seeds MC, Thomas M. Flow cytometric
studies of oxidative product formation by neutrophils: a graded response to membrane
stimulation. J Immunol 1983; 130(4): 1910-7.
Rothe G, Oser A, Valet G. Dihydrorhodamine 123: a new flow cytometric indicator for
respiratory burst activity in neutrophil granulocytes. Naturwissenschaften 1988; 75(7):
354-5.
O'Connor JE, Kimler BF, Morgan MC, Tempas KJ. A flow cytometric assay for intracellular
nonprotein thiols using mercury orange. Cytometry 1988; 9(6):529-32.
Hedley DW, Chow S. Evaluation of methods for measuring cellular glutathione content
using flow cytometry. Cytometry 1994; 15(4): 349-58.
Plummer JL, Smith BR, Sies H, Bend JR. Chemical depletion of glutathione in vivo.
Methods Enzymol 1981; 77: 50-9.
Di Simplicio P, Cacace MG, Lusini L, Giannerini F, Giustarini D, Rossi R. Role of protein
-SH groups in redox homeostasis--the erythrocyte as a model system. Arch Biochem
Biophys 1998; 355(2): 145-52.
Freikman I, Amer J, Ringel I, Fibach E. A flow cytometry approach for quantitative
analysis of cellular phosphatidylserine distribution and shedding. Anal Biochem 2009;
393(1): 111-6.
Rund D, Rachmilewitz E. Beta-thalassemia. N Engl J Med 2005; 353(11): 1135-46.
91
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
Richardson D R, Ponka P. The molecular mechanisms of the metabolism and

transport of iron in normal and neoplastic cells. Biochimica et Biophysica Acta
1997; 1331(1): 140.
Konijn AM. Iron metabolism in inflammation. Baillieres Clin Haematol 1994; 7(4): 829-49.
Breuer W, Hershko C, Cabantchik ZI. The importance of non-transferrin bound iron in
disorders of iron metabolism. Transfus Sci 2000; 23(3): 185-92.
Jacobs A. Low molecular weight intracellular iron transport compounds. Blood 1977;
50(3): 433-9.
Cabantchik ZI, Kakhlon O, Epsztejn S, Zanninelli G, Breuer W. Intracellular and
extracellular labile iron pools. Advances in Experimental Medicine and Biology 2003; 509:
5575.
Cappellini MD, Piga A. Current status in iron chelation in hemoglobinopathies. Curr Mol
Med 2008; 8(7): 663-74.
Kalantar-Zadeh K, Kalantar-Zadeh K, Lee GH. The fascinating but deceptive ferritin: to
measure it or not to measure it in chronic kidney disease? Clin J Am Soc Nephrol 2006; 1
Suppl 1: S9-18.
Wood JC. Cardiac iron across different transfusion-dependent diseases. Blood Rev
2008;22 Suppl 2: S14-21.
Davis BH, Olsen S, Bigelow NC, Chen JC. Detection of fetal red cells in fetomaternal
hemorrhage using a fetal hemoglobin monoclonal antibody by flow cytometry. Transfusion
1998; 38(8): 749-56.
Dziegiel MH, Nielsen LK, Berkowicz A. Detecting fetomaternal hemorrhage by flow
cytometry. Curr Opin Hematol 2006; 13(6): 490-5.
Kleihauer E, Braun H, Betke K. [Demonstration of fetal hemoglobin in erythrocytes of a
blood smear]. Klin Wochenschr 1957; 35(12): 637-8.
Navenot JM, Merghoub T, Ducrocq R, Muller JY, Krishnamoorthy R, Blanchard D. New
method for quantitative determination of fetal hemoglobin-containing red blood cells by
flow cytometry: application to sickle-cell disease. Cytometry 1998; 32(3): 186-90.
92
Chapter 3
3.5

Rita Ellerbrook, PhD, and Zia Uddin, PhD
3.5.4 Introduction
Electrophoresis is defined as the movement of charged molecules (e.g. proteins)

under an electrical field, either through a solution (moving boundary electrophoresis) or
through a semi-solid material embedded in a buffer (zone or solid phase
electrophoresis). Historically, the first hemoglobin variant (HbS) identification using
moving boundary electrophoresis was achieved by Professor Linus Pauling 1 in 1949 at
the University of Chicago, Chicago, Illinois. Subsequently the moving boundary
electrophoresis due to experimental difficulties was replaced by solid phase
electrophoretic methods, e.g., cellulose acetate, agarose, and agar, etc.
In view of the convoluted three-dimensional structure of the hemoglobin
molecule, even a single genetic mutation, resulting in the substitution of an amino acid
in the globin chain (e.g. the substitution of the amino acid valine for glutamic acid in the
sixth position of the -chain of hemoglobin molecule) may result in the change of the
secondary/tertiary structure of the hemoglobin molecule/the net charge on the molecule.
This change in the shape/net charge of the hemoglobin molecule is sufficient to modify
its electrophoretic mobility (movement under an electric field), and thus is
advantageously employed for the separation and identification of the hemoglobin
variants. The migration and the identification of hemoglobin variants in solid phase
93
electrophoretic methods are accomplished at alkaline pH (8.6) and acid pH (5.6), and
the commonly used solid phases for this purpose are described here.
3.5.2 Cellulose Acetate Electrophoresis (alkaline pH)

Cellulose upon treatment with acetic anhydride converts into cellulose acetate by
virtue of the acetylation of the hydroxyl groups. The separation characteristic of
cellulose acetate depends on the degree of acetylation reaction and other variables,
e.g., additives used, prewashing procedure utilized by the manufacturer, pore size,
thickness of the membrane, etc. Historically, cellulose acetate electrophoresis (CAE)
was used worldwide in view of the speed of separation, ability to make the membrane
transparent for the quantification of bands by densitometry, ability to store the
transparent membranes for longer periods (plastic backed cellulose acetate plates), no
need for controlled lower temperature for the electrophoresis, low cost, etc. Under the
electrophoretic conditions of pH 8.6, the ionizable groups (e.g. carboxyl group) are
negatively charged thus rendering a negative charge on the hemoglobin molecule. The
relative migration of the hemoglobin towards the anode is dependent on the net
negative charge on the hemoglobin molecule.
CAE laboratory procedure and information about the required hardware and
consumables can be obtained from Helena Laboratories, Beaumont, Texas, USA
(www.helena.com).
94
Fig 1. Computer simulated cellulose acetate electrophoresis of adult

hemoglobins (pH 8.6)
In Figure 1, separation of a few hemoglobin variants by CAE is illustrated. This is a
computer simulation of the separation of hemoglobins. Generally in all electrophoretic
separations, a commercially prepared AFSC control is used to designate the migration
position of the unknown. Hb S, Hb D, Hb Lepore and Hb G migrate in approximately the
same position, therefore further confirmation of the hemoglobin variant is achieved by
additional laboratory tests, e.g., solubility test and citrate agar electrophoresis at pH 5.6
(see below). In case the hemoglobin variant is not identified by these preliminary
laboratory tests, the laboratory employs other procedures, e.g., HPLC, IEF, and DNA
95
studies. The same procedure is also followed about the co-migration of Hb C, Hb E,

and Hb O-Arab upon CAE.
3.5.3 Agarose Gel Electrophoresis (alkaline pH)

Agar is a gelatinous material prepared from certain marine algae, and is a
mixture of agarose and sulfated polysaccharides contaminants called
agaropectin. The highly purified agar (neutral fraction of agar) that is almost free
of agaropectin (ionizable groups like sulfate and carboxylic) is called agarose.
Agarose gel electrophoresis (AGE) at alkaline pH 8.6 is the widely used clinical
laboratory method for the identification of hemoglobin variants. The reason for the
popularity of AGE is due to the lower affinity of agarose for proteins, ability to exhibit
decreased endosmosis, and also the transparency of the film after drying which allows
quantification of the hemoglobin molecule by densitometry. It is emphasized that
hemoglobinopathy is never determined alone by AGE (alkaline pH 8.6), as is the case
with CAE. The resolution of atypical bands or a band co-migrating at the positions of
commonly encountered bands upon AGE (e.g., HbA 2, HbS, etc.) is accomplished by
additional laboratory tests.
Currently the AGE reagents, separation gels, and Peltier cooling device (which
cools the gel during electrophoresis) are supplied by two major manufacturers (Sebia,
France, and Helena Laboratories, USA). Sebias hemoglobin AGE kit (Hydragel) is used
in conjunction with their semi-automated HYDRASYS System. Helena Laboratories,
USA is a pioneer in supplying AGE kits for >35 years. The Helenas QuickGel method
available in manual mode is ideal for smaller volume clinical laboratories, and the same
96
plate form is used in the semi-automated instruments (SPIFE 2000 and SPIFE 3000) for
handling a larger volume of testing. Helenas fully automated instrument (SPIFE 4000)
utilizes a different plate form than QuickGel. Detailed information about AGE procedures
of these two manufacturers can be obtained from their web site (www.sebia.com and
www.helenalaboratories.com).
In Fig 2 we have presented the computer simulation of the electrophoretic
mobilities of the commonly used AFSC control and few hemoglobin bands obtained
from AGE at alkaline pH.
Fig 2. Computer simulation of hemoglobin agarose gel electrophoresis

bands
3.5.4 Agar Electrophoresis (acid pH)

Agar electrophoresis (AE) at acid pH (5.6-6.2) for the identification/confirmation
of hemoglobins has been widely used for > 40 years. Agarose and agaropectin are the
two main components of agar. Both the electrophoresis and electroendosmotic flow
principles are involved in the separation of hemoglobins by AE. Citrate buffer is usually
97
used for the electrophoretic purpose (Beckman-Coulter uses maleate buffer in their
Paragon kit), therefore it is also called citrate agar electrophoresis. Commercially the AE
kits (plates, reagents, consumables, etc.) are also available from Sebia, France
(HYDRAGEL ACID HEMOGLOBIN) and Helena Laboratories, USA (Titan Gel and
QuickGel). In both cases, hemoglobin AFSC control is used to confirm the
electrophoretic mobility of the unknown (i.e. Hb S, Hb C, Hb E, etc.). Quantification of
the bands is not required and the electrophoregrams are evaluated visually. Laboratory
procedures for AE by Sebia and Helena Laboratories can be obtained from their web
sites (www.sebia.com and www.helenalaboratories.com). In Fig 3, we have presented a
computer simulation of an electrophoregram of the AE.
3.5.5 Interpretation of Hemoglobin Agarose Gel (pH 8.6) and Agar

Gel (pH 5.6) Electrophoresis
The commercially available control that consists of a mixture of Hb A, Hb F, Hb S,
and Hb C serves to set the framework upon which the various hemoglobin variant
mobilities are compared. This combination of hemoglobins is run on each
electrophoretic plate and the interpretation is aided by comparing the mobility of the
variant to these hemoglobins in the control material. By assigning the distance from HbA
to HbC an arbitrary distance unit of 10 (under either acid or alkaline conditions), a
relative number may be assigned to any hemoglobin.
Schneider and Barwick2 presented this system of hemoglobin typing and
provided a chart of the relative mobilities of all the hemoglobins fully characterized at
that time. This chart provided preciseness to the characterization not before possible.
Ellerbrook and Matthews3 at Helena Laboratories felt that since the process was a
98
visual one, therefore a quicker way to examine these relative mobilities was to convert
them into a chart as depicted in Appendix I.
It will not be out of place to mention here that for >30 years in the Hemoglobin
Laboratories of Henri Mondor Hospital (Creteil, France), Professor Henri Wajcman and
associates have organized a database of the electrophoretic mobility of > 400 Hb
variants, using a similar format to that proposed by Schneider and Barwick. The
Wajcman group included in their database the results of a) IEF on polyacrylamide gel,
b) electrophoresis on cellulose acetate at alkaline pH, c) citrate agar electrophoresis, d)
electrophoresis of dissociated globin chains in 6M urea at pH 6.0 and 9.0 or in the
presence of Triton X-1004.
An excerpt of eight hemoglobins from the chart developed by Ellerbrook and
Mathews3 is shown in Fig 3 for instructional purposes.
Fig 3. Combined agarose gel (pH 8.6), and citrate agar (pH 5.6) electrophoretic
pattern presentation for instructional purposes.
The area labeled Alkaline on the left side of this figure depicts the mobility of
the named hemoglobins under alkaline conditions. The perpendicular lines
represent the relative mobility of Hb C (-10), Hb S (-5.2), Hb F (-2.6), Hb A (0),
99
Hb J Toronto (4.5) and Hemoglobin I (8.5). In Fig 3 Hb C is seen to have the

least mobility in an alkaline electric field and is depicted squarely on the line.
Each of the control hemoglobins (e.g. AFSC) will also place squarely on the line.
Hemoglobin I and J are extremely rare and their actual presence on the gels as a
control is not necessary because this grid is all about spacing. The gel will have
hemoglobins A, F, S, and C on it to establish spacing. The distance between Hb
S and Hb A is slightly more than the distance from Hb A to Hb J Toronto, and this
distance is slightly less than the distance from Hb J Toronto to Hb I. While
looking at the actual gel the mobility is not a depiction of the leading edge of the
migration but rather the bulk of that hemoglobin band. Denatured hemoglobins
usually run faster than the native form and therefore the leading edge may be a
function of the age of the sample. The sample application in alkaline conditions is
to the left of Hb C, and most hemoglobins at this pH migrate in the same
direction (left side).
The right side of the figure has the similar approach to the mobility under acidic
conditions. The order of migration is different and the direction is reversed. Here the
sample is applied between Hb S and Hb C. Under acidic conditions Hb F is the fastest
moving hemoglobin. The distance from Hb A to Hb C is assigned a new relative
distance of 10. Hb F is assigned the number -4.4, and Hb S is assigned +5.8. The +/sign is relative to the Hb A value of 0 and not due to distance from the application point.
There is no allowance for fast hemoglobins under acidic conditions because there are
none. At a pH of 6.2 or less, fast hemoglobins migrate like Hb A.
100
Looking at the left side of the chart, Hb J Baltimore migrates slower than Hb J
Toronto because the bulk of the hemoglobin has not moved as far as one might expect
Hb J Toronto to move. Acid electrophoresis is of no assistance in this case because
these fast hemoglobins do not migrate differently, and thus all end up lined with Hb A.
The mobility of Hb C Siriraj is not different from Hb C under acidic conditions, but can be
differentiated under alkaline pH. In this case since the alkaline separation would have
been done first the only apparent observation would be the presence of an abnormal
hemoglobin band migrating between Hb F and Hb S. Very few hemoglobins migrate like
hemoglobin S so this second test is very useful in narrowing down the possible identity
of this variant. The chart in Appendix I contains the relative mobilities of 165
hemoglobins. The most common variants were discovered first so this chart should
encompass the relative mobilities of most of the hemoglobins found.
3.5.6 Requirements for the identification of complex hemoglobinopathies

Age, sex, ethnicity, ethnic background of biological parents, blood transfusion
(past three months), CBC with differential, serum iron, TIBC, ferritin, treatment status
(immunotherapy), laboratory results of AGE and AE electrophoresis, capillary zone
electrophoresis, high pressure liquid chromatography, isoelectric focusing, quantitative
results of Hb A2, and Hb F, hemoglobin stability, and globin chain analysis. The
significance of all these parameters shall be obvious from the case studies mentioned
later on in the book.
101
References
1. Pauling L, Itano HA, Singer SH, Wells IC. Sickle cell anemia, a molecular
basis disease. Science 1949; 110: 53.
2. Barwick RC, Schneider RG. The computer-assisted differentiation of hemoglobin
variants, in Human Hemoglobins and Hemoglobinopathies:
A review to1981. Texas Reports on Biology and Medicine 1980-81; 40:
143-156
3. Helena Laboratories, Beaumont, Texas, USA
4. Wajcman H. Electrophoretic Methods for Study of Hemoglobins. Methods in
Molecular medicine, vol 82: Hemoglobin Disorders: Molecular methods and
Protocols, Edited by: Ronald L. Nagel, Humana Press Inc., Totowa, NJ.
102
Appendix 1
Appendix 1 continued next page

103
Appendix 1 continued
104
Chapter 3
3.6 Capillary Zone Electrophoresis
Zia Uddin, PhD
3.6.1 Introduction
During the last five decades separation science has witnessed unparallel growth.
Chromatography and electrophoresis are the main techniques that are routinely used
worldwide for the separation, identification, and quantification of analytes in clinical
laboratories. Capillary zone electrophoresis (CZE) played a significant role in the
completion of the human genome project. Introduction of CZE instruments by BeckmanCoulter, Sebia, and Helena Laboratories not only automated but also increased the
sensitivity, specificity and reproducibility of the clinical laboratory procedures (e.g.,
serum protein electrophoresis, immunotyping, hemoglobin variant identification for both
the adult and newborn). Besides references listed at the end of this section, the
interested reader is advised to also review the online literature on CZE (e.g., Righetti,
PG, and Guttman, A. 2012 Capillary Electrophoresis. eLS.)
3.6.2
Basic Principle
In simple terms CZE is a liquid flow electrophoresis in which buffer has replaced
the solid support medium (e.g., agarose gel), and the separation occurs due to the
interaction of the analyte with the pH of the buffer. For this reason initially CZE was also
called Free Solution Capillary Electrophoresis. In Figure 1, a pictorial illustration of
CZE principle is presented.
105
Figure 1. Capillary Zone Electrophoresis Principle

In CZE two independent phenomena occur, i.e., a) migration of negatively
charged ions toward the positively charged electrode, and b) interaction of the positive
charges from the buffer and the negative charges from the capillary wall leading to
electro-osmotic flow (EOF) from the anode to the cathode. Both of these two processes
(electrophoretic mobility and EOF) can take place at the same time working in opposite
direction thus providing greater resolution.
106
Automated CZE instrument (Figure 2) consists of the following:

a
b
c
d
e
f
g
h
Cathode
Anode
Power supply to generate high voltage (10,000 volts)
Catholyte (buffer solution at the cathode end)
Anolyte (buffer solution at the anode end)
Capillary facilitated with a cooling device
Detector (415 nm for hemoglobins)
Computer for data handling and storage
Figure 2. Capillary Zone Electrophoresis Instrument Components

3.6.3 Application of CZE in Diagnostic Hemoglobinopathies
107
Hemoglobin variants can be separated on CZE as is the case with other proteins.
This method is the most advanced and dedicated alternative to the classic alkaline and
acid electrophoresis and the more sophisticated IEF. Chromatography, the separation
alternative on column, has developed from separation on size, charge and hydrophobic
interaction to the modern dedicated high performance liquid chromatography (HPLC),
as we know today. Both of these dedicated methods (CZE and HPLC) have the
advantage of using minimal amounts of material, of providing a separation in a matter of
minutes, with high reproducibility and sensitivity and above all they are able to measure
virtually all fractions including those present at low levels but essential for the diagnosis
or hemoglobinopathy. In addition these two methods may complement each other up to
a certain extent compensating for specific errors.
3.6.4 Interpretation of CZE Results

The migration time of the hemoglobin variant (since the inception of the injection
of the sample and the moment a specific molecule is detected) is divided into fifteen
(15) zones (Table 1). It is obvious that > 1000 hemoglobin variants cannot be separated
in 15 zones. However, the most common one (e.g., Hb S, C, E, and D) will be putatively
identified by their zone with a specificity >90%. Table 1 shows that there is an overlap of
hemoglobin variants in a particular zone (Z1 Z15). This limitation of CZE is similar to
that experienced with other techniques employed for the identification of hemoglobin
variants, e.g., HPLC (Szuberski J, Oliveira JL, Hoyer JD. A comprehensive analysis of
hemoglobin variants by high performance liquid chromatography. Int J Lab Hematol
2012; 34: 594-604).
108
In Figure 3 we have presented a CZE scan of the most commonly used AFSC
control in the clinical laboratory, which illustrates the position of HbA,
HbF, HbS, and HbC peaks corresponding to their respective zones. Later on in this
book (case studies) we have also presented the CZE scan of the hemoglobin variant for
each case.
One drawback of CZE is the assignment of the migration position of the
hemoglobin bands into Z1-Z15 (Table 1) in cases when the HbA / Hb A2 are absent in
the specimen of interest, e.g. Hb S-C disease. This drawback is due to the shifting of
band positions in the absence of Hb A / Hb A2. This limitation of CZE is avoided by
mixing (1:1 ratio) the specimen devoid of HbA/HbA 2 with a specimen containing Hb A,
and performing the CZE test thus achieving the relevant migration position and zoning
(Z1-Z15).
109
Table 1. Hemoglobin Zones of CZE: Z1 Z15

N
zone
2=
Z(C)
3=
Z(A2)
4=
Z(E)
5=
Z(S)
6=
Z(D)
Hb Variants Library Software release 8.60

Hb Santa Ana free alpha chain, Hb Mizuho (minor peak), Hb delta A'2, Hb alpha
A2, Hb T-Cambodia, "Savaria" Hb A2 variant, "Chad" Hb A2 variant, "Arya" Hb A2
variant, "Hasharon" Hb A2 variant, "Fort de France" Hb A2 variant, "Ottawa" Hb A2
variant, "Shimonoseki" Hb A2 variant, "Stanleyville II" Hb A2 variant, "OIndonesia" Hb A2 variant, "G-Norfolk" Hb A2 variant, "San Antonio" Hb A2 variant,
"Handsworth" Hb A2 variant, "Matsue-Oki" Hb A2 variant, "Memphis" Hb A2
variant, "Q-Iran" Hb A2 variant, "G-Waimanalo" Hb A2 variant, "Russ" Hb A2
variant, "Q-India" Hb A2 variant, "Montgomery" Hb A2 variant, "Watts" Hb A2
variant, "G-Pest" Hb A2 variant, "Winnipeg" Hb A2 variant, "Queens" Hb A2
variant, "Inkster" Hb A2 variant, "Chapel Hill" Hb A2 variant, "Q-Thailand" Hb A2
variant, "Park Ridge" Hb A2 variant
Hb C, Hb F-Hull, Hb F-Texas-I, Hb Constant Spring, Hb C-Harlem (C-Georgetown),
"Boumerdes" Hb A2 variant, "Bassett" Hb A2 variant, "Tarrant" Hb A2 variant,
"Manitoba I" Hb A2 variant, "St. Luke's" Hb A2 variant, "Setif" Hb A2 variant,
"Sunshine Seth" Hb A2 variant, "Titusville" Hb A2 variant, "Swan River" Hb A2
variant, "Manitoba II" Hb A2 variant, "Val de Marne" Hb A2 variant
Hb A2, Hb Chad (E-Keelung), Hb O-Arab, Hb E-Saskatoon, "Dallas" Hb A2 variant*,
"Toulon" Hb A2 variant*, "Bonn" Hb A2 variant*, "Chicago" Hb A2 variant*,
"Fontainebleau" Hb A2 variant*, "Hekinan" Hb A2 variant*, "Mosella" Hb A2
variant*, "Aztec" Hb A2 variant*, "Frankfurt" Hb A2 variant*, "M-Boston" Hb A2
variant*, "Owari" Hb A2 variant*, "Twin Peaks" Hb A2 variant*, "Conakry" Hb A2
variant*, "Gouda" Hb A2 variant*, "Jura" Hb A2 variant, "Nouakchott" Hb A2
variant
Hb E, Hb Seal Rock, Hb Kln (Ube-1), Hb Buenos Aires (minor peak), Hb MSaskatoon (minor peak), Hb A2-Babinga, Hb G-Siriraj, Hb Agenogi, Hb Sabine, Hb
Santa Ana, Hb Savaria, "M-Iwate" Hb A2 variant, "Wayne" Hb A2 variant (peak 1),
Denatured Hb C
Hb S, Hb Arya, Hb Hasharon (Sinai), Hb Dhofar (Yukuhashi), Hb Shimonoseki
(Hikoshima), Hb O-Indonesia (Buginese-X), Hb Ottawa (Siam), Hb Fort de France,
Hb Montgomery, Hb G-Copenhagen, Hb S-Antilles, Hb Handsworth, Hb S-Oman
(peak 2), Hb Hamadan, Hb Russ, Hb Stanleyville II, "Lombard" Hb A2 variant,
"Tatras" Hb A2 variant, "Cemenelum" Hb A2 variant, "Jackson" Hb A2 variant,
"Hopkins-II" Hb A2 variant, "J-Broussais" Hb A2 variant (alpha 2), Denatured Hb OArab
Hb D, Hb Memphis, Hb Leiden, Hb Muravera, Hb D-Bushman, Hb G-Norfolk, Hb SOman (peak 1), Hb Matsue-Oki, Hb Osu Christiansborg, Hb D-Punjab (D-Los
Angeles), Hb G-Waimanalo (Aida), Hb Muskegon, Hb D-Ibadan, Hb Buenos Aires
(minor peak), Hb Q-India, Hb Lepore (Lepore-BW), Hb Q-Iran, Hb Summer Hill, Hb
G-Philadelphia, Hb D-Ouled Rabah, Hb Yaizu, Hb San Antonio, Hb Watts, Hb
Ferrara, Hb Kln (Ube-1), Hb Fort Worth, Hb Korle-Bu (G-Accra), Hb G-Taipei, Hb DIran, Hb St. Luke's, Hb G-Coushatta (G-Saskatoon), Hb Inkster, Hb Winnipeg, Hb
110
7=
Z(F)
9=
Z(A)
10
11
12
13
Zrich, Hb G-Pest, Hb Queens (Ogi), Hb Setif, Hb P-Nilotic, Hb Sunshine Seth, Hb

Titusville, "Le Lamentin" Hb A2 variant, "J-Meerut" Hb A2 variant, "J-Rajappen" Hb
A2 variant, "J-Anatolia" Hb A2 variant, "J-Oxford" Hb A2 variant, "Ube 2" Hb A2
variant, "J-Broussais" Hb A2 variant (alpha 1), "J-Toronto" Hb A2 variant, "Mexico"
Hb A2 variant, "J-Tongariki" Hb A2 variant, "Neuilly-sur-Marne" Hb A2 variant, "JParis-I" Hb A2 variant (alpha 2), "J-Habana" Hb A2 variant, "J-Paris-I" Hb A2 variant
(alpha 1), "Wayne" Hb A2 variant (peak 2), Denatured Hb E
Hb F, Hb Willamette, Hb Alabama, Hb Chapel Hill, Hb Park Ridge, Hb Porto Alegre,
Hb Q-Thailand (G-Taichung), Hb Sabine, Hb Bassett, Hb Rampa, Hb G-San Jos, Hb
Barcelona, Hb Geldrop Santa Anna, Hb Richmond, Hb Boumerdes, Hb Swan River,
Hb Burke, Hb Tarrant, Hb Presbyterian, Hb Manitoba II, Hb Manitoba I, Hb Port
Phillip, "J-Rovigo" Hb A2 variant, Denatured Hb S, Denatured Hb D-Punjab
Hb Lansing, Hb Hinsdale, Hb Ypsilanti (Ypsi - peak 1), Hb Alberta, Hb Val de Marne
(Footscray), Hb Kempsey, Hb Shelby (Hb Leslie), Hb Atlanta, Hb Ypsilanti (Ypsi peak 2), Hb Rainier, Hb Athens-GA (Waco), Hb Debrousse
Hb A, Hb Olympia, Hb Gorwihl (Hinchingbrooke), Hb Phnom Penh*, Hb Silver
Springs*, Hb La Coruna*, Hb Bougardirey-Mali*, Hb Dallas*, Hb Toulon*, Hb
Austin*, Hb Bonn*, Hb Buenos Aires (Bryn Mawr, major peak)*, Hb Chicago*, Hb
Okayama*, Hb Fontainebleau*, Hb Raleigh*, Hb Hekinan*, Hb Mosella*, Hb Aztec*,
Hb Little Rock*, Hb Frankfurt*, Hb Bethesda*, Hb M-Boston (M-Osaka)*, Hb
Brisbane (Great Lakes)*, Hb Mizuho*, Hb Grange Blanche*, Hb San Diego*, Hb MSaskatoon (main peak)*, Hb Malm*, Hb Minneapolis Laos*, Hb Owari*, Hb Rhode
Island (Southwark)*, Hb Twin Peaks*, Hb Wood*, Hb Conakry*, Hb Coimbra
(Ingelheim)*, Hb Linkping (Meilahti)*, Hb Templeuve*, Hb Alzette*, Hb Ty Gard*,
Hb Gouda*, Hb Syracuse*, Hb Fort Dodge, Hb Camperdown, Hb Jura
Hb Nouakchott, Hb Wayne (peak 1), Hb M-Iwate (M-Kankakee), Hb Camden
(Tokuchi), Hb Hope
Hb Vaasa, Hb Providence (X-Asn peak), Hb Tacoma, Hb Corsica, Hb K-Woolwich, Hb
Lombard, Hb Andrew Minneapolis, Hb Fannin Lubbock, Hb Kaohsiung (New York),
Hb Osler (Fort Gordon), Hb Himeji, Hb Jackson, Hb Tatras, "I (I-Texas)" Hb A2
variant
Hb Bart's, Hb Cemenelum, Hb Wayne (peak 2), Hb Hopkins-II, Hb J-Calabria (JBari), Hb J-Tongariki, Hb Providence (X-Asp peak), Hb J-Meerut (J-Birmingham), Hb JBroussais (Tagawa-I - alpha 2), Hb J-Rajappen, Hb Grady (Dakar), Hb Le Lamentin,
Hb J-Anatolia, Hb Hikari, Hb J-Broussais (Tagawa-I - alpha 1), Hb J-Chicago, Hb JToronto, Hb J-Oxford (I-Interlaken), Hb Ube-2, Hb J-Meinung (J-Bangkok), Hb Neuillysur-Marne, Hb Mexico (J-Paris-II), Hb J-Paris-I (J-Aljezur - alpha 1), Hb J-Habana, Hb JBaltimore (N-New Haven), Hb J-Paris-I (J-Aljezur - alpha 2), Hb K-Ibadan
Hb N-Baltimore (Hopkins-I), Hb J-Rovigo, Hb J-Norfolk (Kagoshima), Hb J-Kaohsiung
(J-Honolulu)
14
Hb N-Seattle
15
Hb I (I-Texas)
111
Figure 3. CZE scan of AFSC control
112
References
1
Borbely N, Phelan L, Szydlo R, Bain B. Capillary zone electrophoresis for

haemoglobinopathy diagnosis. J Clin Path 2013; 66: 29-39.
2. Greene DN, Pyle AL, Chang JS, Hoke C, Lorey T. Comparison of Sebia
Capillarys Flex Capillary electrophoresis with the BioRad Variant II high pressure
liquid chromatography in the evaluation of hemoglobinopathies. Clinica Chimica
Acta 2012; 413: 1232-1238
3. Keren DF, Shalhoub R, Gulbranson R, Hedstrom D. Expression of Hemoglobin
Variant Migration by Capillary Electrophoresis Relative to Hemoglobin A2
Improves Precision. Am J Clin Path 2012; 137: 660-664
4. Sae-ung N, Sriyorakun H, Fucharoen G, Yamsri S, Sanchaisuriya K, Fucharoen
S. Phenotypic expression of hemoglobin A2, E and F in various hemoglobin E
related disorders. Blood Cells, Molecules, and Diseases 2012; 48: 11-15.
5. Sangkitporn S, Sangkitporn SK, Tanjatham S, Suwannakan B, Rithapirom S,
Yodtup C, Yowang A, Duangruang S. Multicenter Validation of Fully Automated
Capillary Electrophoresis Method for Diagnosis of Thalassemias and
Hemoglobinopathies in Thailand. Southeast Asian J Trop Med Public Health
2011; 6(5):1224-1232.[PubMed]
6. Fucharoen G, Srivorakun H, Singsanan S, Fucharoen S. Presumptive diagnosis of
Hemoglobinopathies in Southeast Asia using a capillary electrophoresis system.
Int. Jnl. Lab. Hem. 2011; 33: 424-433.
7. Wajcman H, Moradkhani K. Abnormal haemoglobins: detection & characterization.
Indian J Med Res 2011;134 (4): 538-546
8. Liao C, Zhou J-Y, Xie X-M, Li J, Li DZ. Detection of Hb Constant Spring by a Capillary
Electrophoresis Method. Hemoglobin 2010; 34(2): 175-178.
9. Cotton F, Nalaviolle X, Vertongen F, Gulbis B. Evaluation of an Automated Capillary
Electrophoresis System in the Screening for Hemoglobinopathies. Clin Lab 2009;
55: 217-221.
10. Van Delft P, Lenters E, Bakker-Verweij M, De Korte M, Baylan U, Harteveld CL,
Giordano PC. Evaluating five dedicated automatic devices for haemoglobinopathy
dianostics in multi-ethinic populations. Int Jnl Lab Hem 2009; 31: 484-495
11. Winichagoon P, Svasti S, Munkongdee T, Chaiya W, Boonmongkol P, Chantrakul N,
Fucharoen S. Rapid diagnosis of thalassemias and other hemoglobinopathies by
capillary electrophoresis system. Translational Research 2008, 152 (4): 178-184
12. Keren DF, Hedstrom D, Gulbransom R, Ou C-N, Bak R. Comparison of Sebia Capillarys
Capillary Electrophoresis With the Primus High-Pressure Liquid Chromatography in the
Evaluation of Hemoglobinopathies. Am J Clin Pathol 2008, 130: 824-831.
13. Wang J, Zhou S, Huang W, Kiu Y, Cheng C, Lu Xin, Cheng J. CE-based analysis of
hemoglobin and its applications in clinical analysis. Electrophoresis 2006; 27: 31083124.
14. Louhabi A, Philippe M, Lali S, Wallenmacq, Maisin D. Evaluation of a new Sebia kit for
analysis of hemoglobin fractions and variants on the Capillarys system. Clin Chem Lab
Med 2006; 44(3): 340-345.
15. Chang P-L, Kuo I-T, Chiu T-C, Chang H-T. Fast and sensitive diagnosis of thalassemia
by capillary electrophoresis. Anal Bioanal Chem 2004; 379: 404-410.
113
16. Jenkins M, Ratnaike S. Capillary Electrophoresis of Hemoglobin. Clin Chem Lab

Med 2003; 41(6): 747-754.
17. Gulbis B, Fontaine B, Vertongen F, Cotton F. The place of capillary
electrophoresis techniques in screening for hemoglobinopathies.
Ann Clin Biochem 2003; 40: 659-662
18. Gerritsma J, Sinnige D, Drieze C, Sittrop B, Houtsma P, Ulshorst-Jansen NH,
Huisman W. Quantitative and qualitative analysis of hemoglobin variants using
capillary zone electrophoresis. Ann Clin Biochem 2000; 37 (3): 380-389.
19. Castagnola M, Messana I, Cassiano L, Rabino R, Rossetti DV, Giardina B. The use of
capillary electrophoresis for the determination of hemoglobin variants. Electrophoresis
1995; 16(1): 1492-1498.
20.
http://www72.homepage.villanova.edu/frederick.vogt/ppt/2007/Capillary_Electrop
horesis.ppt
21.
http://chemwiki.ucdavis.edu/Analytical_Chemistry/Instrumental_Analysis/Capillar
y_Electrophoresis?highlight=capillary+zone+electrophoresis
114
Chapter 3
3.7
Isoelectric Focusing
David R. Hocking, PhD
3.7.1 Introduction
Isoelectric focusing (IEF), also known as electrofocusing and isoelectricfocusing
electrophoresis, is a separation method that resolves complex mixtures of proteins by
their isoelectric points (pI). IEF is a type of electrophoresis that forms a pH gradient
during the run. The technique is capable of extremely high resolution. The formation of
a pH gradient is accomplished by blending a mixture of small molecular weight carrier
ampholytes into a support matrix, or gel, usually of purified high-grade agarose. An
anolyte solution (i.e. acetic acid) and a catholyte solution (i.e. ethanolamine) are
saturated onto paper electrode wicks then are placed directly on opposite ends on the
surface of the agarose gel. Proteins (i.e. hemoglobins) that are to be separated are
placed near the cathode wicks using a clear plastic with rectangular wells cut out. The
protein solution (i.e. hemoglobin hemolysate) is then pipetted in the defined wells and
allowed to diffuse into the gel. An electric current is then passed through the medium.
The proteins move through the changing pH gradient until it reaches a point in which the
pH of that molecules pI is reached. At this point the protein no longer has an electric
charge and becomes neutral, or isoelectric (due to the protonation or de-protonation of
the associated functional amino and carboxyl groups) and as such will not proceed any
further within the gel. The proteins become focused into sharp stationary bands with
each protein positioned at a point within the newly formed pH gradient corresponding to
its pI.
115
Note: All the IEF figures in this compendium were obtained after agarose gel
electrophoresis on the Wallac Resolve Hemoglobin System (Perkin Elmer), and
the scans were procured using the Wallac WS-1010 IsoScan Imaging System
(Perkin Elmer).
3.7.2 IEF of Normal Adult Hemoglobins: HbA (Adult), HbF (Fetal), HbA 2
Normal adult hemoglobins are comprised of , , and globin chains paired as
~96% HbA (2 2), ~3% HbA2 (2 2) and <2% HbF (2 2) tetramers (Figure 1). One can
usually find the glycated form of HbA, or HbA 1c , anodal to it as shown in the Figure 2.
Figure 1.
116
Aging bands, HbA3, are also anodal to HbA and are the result of posttranslational modifications such as acetylation and glutathione
attachment.
It should be noted that beta-chain variants such as HbS, HbE, HbD, etc., will also
Figure 2.
display glycated forms anodal to the variant (HbS 1c, HbE1c, HbD1c). This observation is
critical to note should the patient exhibit symptoms of diabetes where their blood
glucose values are documented to be high. The percentage can be upwards of 10-20%
in cases of uncontrolled blood glucose levels.
3.7.3 IEF of Normal Newborn Hemoglobins: HbF (Fetal) and HbA (Adult)
Normal newborn hemoglobins are comprised of ~ 60-85% HbF ( 2 2) and 1540% of HbA (2 2). It is very rare to see HbA2. About 10% of HbF is partially acetylated
HbFac, which results in higher oxygen affinity, an important property needed for
newborns. Aging hemoglobin bands, or HbF3 are always anodal to HbFac, and are the
result of glutathione (an antioxidant, preventing damage to important cellular
components caused by reactive oxygen species) attachment.
Representative patterns of newborn hemoglobin are shown in Figure 3. Fresh
cord blood is shown in channel 3a. A sample that was collected and stored using a filter
paper is shown in channel 3b. Note the increased levels of HbF 3 in the stored blood
collected on filter paper.
117
Figure 3. IEF of normal newborn hemoglobins: HbF (fetal) and HbA (adult)
3.7.4 IEF of Beta-Chain Variant Hemoglobins

It is customary to report hemoglobins in the order of greatest concentrations.
Heterozygotes are usually expressed as HbAX where X represents the beta-chain
variant i.e. Sickle Cell (S), C, D, E or name of the variant. Examples are HbAS, HbAE or
HbAD-Punjab. A few examples are shown (note percentages).
118
Figure 4. Hemoglobin A-S Trait

Figure 5. Hemoglobin D-Punjab Trait
Figure 6. Hemoglobin A-E Trait
Figure 7. Hemoglobin A-C Trait

These
examples show the beta-chain mutation along with the Relative Charge Value (RCV)
Figure 9. Hemoglobin S/+-thal
Figure 8. Hemoglobin A-O Trait
change as a result of the substitution of

the normal amino acid found in HbA. In each case the HbA is in greater concentration
than the beta variant (HbX). This pattern is true for all positive RCV values.
119
Background on Hb O-Arab: This rare hemoglobin variant emerged about 2,000

years ago on a singular haplotype, characteristic of the Greek Pomaks. Its frequency
increased as a consequence of high genetic drift within this population, and it was
dispersed throughout the Mediterranean basin and Middle East with minor variations of
its haplotypic pattern. (Haematologica. 2005 Feb;90(2):255-7. HbO-Arab mutation
originated in the Pomak population of Greek Thrace. Papadopoulos V, Dermitzakis E,
Konstantinidou D, Petridis D, Xanthopoulidis G, Loukopoulos D).
The example in Figure 9 shows adult hemoglobin that has more HbS than HbA.
The patient has reduced HbA, an increase in HbA 2 and shows >than 3% of HbF. These
findings indicate that the patient has a beta thalassemia (reduced amount of beta-globin
chains from one parent) along with sickle cell hemoglobin (HbS) from the other parent.
Note the aging bands from sample storage.
3.7.5
IEF of Alpha-Chain Variant Hemoglobins

An individual inherits two sets of alpha globin genes, 1 and 2, from each
parent. If one of the alpha genes has a mutation, then one out of the four, or ~25% of
the hemoglobin, will be the variant, not the typical 50% from a beta-chain variant. The
affected alpha globin chain will form dimers with the non-alpha globin chains. HbGPhiladelphia is a common alpha-chain variant that is shown below (Figure 10).
Note that the percentage of HbG-Philadelphia (Figure 10) relative to HbA is less
than what is seen in the beta variant HbD or HbD-Punjab. This is a clue to suspecting
an alpha variant. Additionally you should also observe that there should be another
120
band cathodal to HbA2. This is due to the variant alpha globin chain combining with the
delta chain.
121
The example in Figure 11 is a rare combination of the beta HbS variant and the
alpha HbG-Philadelphia variant. Note the presence of four prominent bands: HbA, HbGPhiladelphia, HbS and the hybrid, HbSG-Philadelphia, the tetramer formed by the
dimers of -GPhiladelphia and S. Also note the HbA 2 variant that resulted from the G
and dimers. It will be seen cathodal (negative electrode) to the hybrid.
3.7.6 IEF of Thalassemias
A
Figure 10. Hemoglobin G-Philadelphia Trait
typical +-
thalassemia is shown in Figure 12. Note that the percentage of HbA is reduced (95%)
and the amount of circulating HbA2 is increased (>3.5%). Beta thalassemias occur in
persons of Mediterranean origin, and to a lesser extent, Chinese, other Asians, and
African Americans. +-thalassemia is also known as Thalassemia Minor and occurs if
you receive the defective beta-globin gene from only one parent. Persons with this form
of the disorder are carriers of the disease, Cooleys anemia or beta thalassemia major
(0), if their other partner also passes their defective gene to the baby.
Figure 11. Hemoglobin ASG-Philadelphia
122
The pattern in Figure 13 is typical of those individuals presenting with a severe form of
Sickle Cell disease. In this example, the patient inherited the HbS from one parent and
is missing the beta globin gene from the other parent. The patient, though missing a
beat globin gene, has compensated for the missing beta-globin gene with the
persistence of making HbF from the gamma-globin gene.
Figure 12. Hemoglobin +-thal
Conclusion
IEF can be an important tool in assiting the laboratorian in the dection and
interpretation of hemoglobin variants. The technique offers improved resolution over
traditional electrophoretic methods and is useful for both adult and newborn patients. By
careful observation, one can determine if the variant is either a or variant or
combination. One can also correctly interpret -thalassemias.
References:
1. David R. Hocking. The Separation and Identification of Hemoglobin Variant by
Isoelectric Focusing Electrophoresis (May 2004), Catalog # HC-60, Perkin
123
Elmer Life and Analytical Sciences, Wallac Oy, P.O. Box 10, FIN-20101
Turku, Finland. Tel. 358-2-2678111 Fax. 358-2-2678357
Web site: www.perkinelmer.com
2. Additional information about the IEF procedures and instruments can be
solicited from:
a) Petra Furu, Ph.D., Global Business Manager, Specialty Diagnostics,
Perkin Elmer , Mustionkatu 6 / 20750 Turku / Finland.
e-mail: petra.furu@perkinelmer.com Tel. 358 2 267 8497
b) William R. Fisher, Technical Support Specialist, Specialty Diagnostics,
Perkin Elmer, 520 South Main Street, Akron, OH 44311, USA
e-mail: william.fisher@perkinelmer.com Tel. 330-564-4883
124
Chapter 3
3.8
High Performance Liquid Chromatography

Zia Uddin, PhD
3.8.1. Introduction
In 1973 I had the privilege of attending a short course on High Performance
Liquid Chromatography (HPLC), sponsored by the American Chemical Society at
Virginia Polytechnic Institute, Blacksburgh, Virginia, USA. The teachers of this course
were Drs. Lloyd R. Snyder and Joseph J. Kirkland. These two scientists are responsible
for several advancements in HPLC, and their most significant contribution in
collaboration with Dr. John W. Dolan is their latest book (Snyder LR, Kirkland JJ, Dolan
JW, Introduction to Modern Liquid Chromatography, 3 rd Edition, John Wiley & Sons,
Hoboken, NJ 20010). Persons interested in HPLC shall find this book very helpful in
understanding the theory and practice of HPLC, and the components of HPLC (solvent
system, pump, injection port, column, stationary phase, detector, computer, etc.).
Additional literature about HPLC can be accessed from the following Internet sites:
http://www.lcresources.com
http://lchromatography.com/hplc find/index.html
http://tech.groups.yaho.com/group/chrom-L/links
http://userpages.umbc.edu/~dfrey1/Freylink
http://www.chromatographyonline.com
Note: The name High Pressure Liquid Chromatography was initially used, however now
the word Pressure is replaced by Performance. In this book we shall therefore use
High Performance Liquid Chromatography nomenclature.
125
3.8.2. Basic Principle

Liquid chromatography (LC) consists of a liquid mobile phase
and a stationary phase and the separation is accomplished by the
distribution of the solutes between these phases. Manual LC procedure is slow
and needs additional steps for the identification of the compound of interest. In
HPLC the separation process is expedited by forcing the mobile phase under
high pressure through the column, and almost all the steps of the operation are
automated (Figure 1).
Figure 1. Key components of a HPLC system, a) computer, b) detector, c) column, d)

injection port, e) pump, f) solvent reservoir. (adopted from Snyder LR, Kirkland JJ,
Dolan, JW, Introduction to Modern Liquid Chromatography, 3 rd Edition. John Wiley &
Sons, Hoboken, NJ 20010).
126
The identification of a compound of interest in HPLC is ascertained by

matching its retention times (time required to separate a compound after the
injection step) with a standard or control. Several kinds of detectors are
employed in HPLC for detection purposes, e.g., spectrophotometric, flurometric,
electrochemical, etc. Another development in the identification of a compound
after HPLC is coupling it with mass spectrometry (Chapter 3.4). This technique is
very useful when the retention time of the compound is not previously known.
The identification is achieved by the m/z value of the ion associated with the
compound of interest, e.g., globin chain of a hemoglobin variant (Chapter 3.4).
3.8.3 Illustrations
a) Quantification of Hb A2, and Hb A1c :
One of the widely used procedure employing HPLC is the
quantification of Hb A1c and Hb A2 ( Figure 2).
127
Figure 2. Peak at 0.81 (Hb A1c) and at 3.1 (Hb A2). Adopted from the Technical Manual
of D-10, Bio-Rad, Hercules, CA.
b) Hb OIndonesia in India: a rare observation
The father is heterozygous for Hb OIndonesia and the mother is normal, however the
1
daughter has an HPLC pattern similar to her father (Figure 3). Although the normal
hemoglobin fractions (Hb F, Hb A, Hb A2)as well as the common variants (Hb S and Hb
C) all have distinct retention times,there are less prevalent variants with similar or
128
identical retention times. In these cases additional laboratory procedures must be

utilized for a conclusive diagnosis.
Figure 3. a) HPLC of daughter, b) HPLC of father, c) HPLC of mother
129
c) Apparent hemoglobinopathies caused by blood transfusions

Any spurious peak in HPLC can cause confusion and lead to
unnecessary additional testing. It is advised that in order to reduce
unwarranted commotion, the patients medical record should be examined
for recent blood transfusions. Figure 4 illustrates an example of a patient
with Hb SS disease on hyper-transfusion regimen who received a unit of
blood from a donor heterozygote for Hb O-Arab as demonstrated by a
small, but prominent peak eluting after Hb S.
Figure 4. HPLC of a Hb SS patient transfused with one unit of Hb

A-O Arab blood.
130
Cited references:
1.
Chopra A, Fisher C, Khunger JM, Pati H. Hemoglobin O Indonesia in

India: a rare observation. Ann Hematol 2011; 90: 353-354
2.
Kozarski TB, Howantiz PJ, Howantiz JH, Lilic N, Chauhan YS.
Blood transfusions leading to apparent Hemoglobin C, S, and
O-Arab Hemoglobinopathies. Arch Pathol Lab Med 2006; 130: 1830-33.
Additional references:
3.
Szuberski J., Oliveira JL, Hoyer JD. A comprehensive analysis of hemoglobin
variants by high performance liquid chromatography (HPLC). International
Journal of Hematology 2012; 34: 594-604.
4.
Ondei LS, Zamaro PJA, Mangonaro PH, Valencio CR, BoniniDomingos CR. HPLC determination of hemoglobins to establish
reference values with the aid of statistics and informatics. Genetics
and Molecular Research 2007; 6(2): 453-460.
5.
Mais DD, Boxer LA, Gulbranson RD, Keren DF. Hemoglobin Ypsilanti.
A High-Oxygen-Affinity Hemoglobin Demonstrated by Two Automated
High-Pressure Liquid Chromatography Systems. Am J Clin Path 2007; 128:
850-853.
6.
Joutovsky A, Hadzi-Nesic J, Nardi MA. HPLC Retention Time as a Diagnostic
Tool for Hemoglobin Variants and Hemoglobinopathies: A Study of 60 000
Samples in a Clinical Diagnostic Laboratory. Clin Chem 2004; 50: 1736-47.
7.
Ou C-N, Rognerud CL. Diagnosis of hemoglobinopathies: electrophoresis vs.
HPLC. Clin Chim Acta 2001; 313: 187-94.
8.
Fucharoen S, Winichagoon P, Wisedpanichkij R, et al. Prenatal and postnatal
diagnosis of thalassemias and hemoglobinopathies by HPLC. Clin Chem
1998; 44: 740-748.
9.
Riou J, Godart C, Hurtrel D, Mathis M, Bimet C, et al. Cation-exchange HPLC
evaluated for presumptive identification of hemoglobin variants.
Clin Chem 1997; 43: 34-39.
10.
Huisman THJ. Review: Separation of Hemoglobins and Hemoglobin Chains
By High-Pressure Liquid Chromatography. J. Chromatogr 1987; 418: 277-304
11.
Colah RB, Surve R, Sawant P, DSouza E, Italia K, Phanasgaonker S,
Nadkarni AH, Gorakshaker AC. HPLC studies in hemoglobinopathies.
Indian J Pediatr 2007; 74(7): 657-62
12. Sachdey R, Dam AR, Tyagi G. Detection of Hb variants and
hemoglobinopathies in Indian population using HPLC: Report of 2600
cases. Indian J Pathol Microbiol 2010; 53: 57-62.
13. Rao S, Kar R, Gupta SK, Chopra A, Saxena R. Spectrum of
haemoglobinopathies diagnosed by cation exchange-HPLC and
modulating effects of nutritional deficiency anemias from north India.
Indian J Med Res 2010; 132: 513-519.
Chapter 4
131
4.1 Solid Phase Electrophoretic Separation

Zia Uddin, PhD
In the early stages of the development of the diagnostic hemoglobinopathies,

polyacrylamide gel electrophoresis (in urea, acid and non-ionic detergent Triton-X-100)
and cellulose acetate electrophoresis (alkaline and acid pH) were utilized for globin
chain analysis. These techniques provided information about the globin chains that
contained the substitution. However, due to scientific limitations (selection of known
variants as controls with mobility similar to that of the unknown), these techniques were
abandoned in favor of other methods as described in this chapter. Recently capillary
zone electrophoresis was also used for the separation of globin chains. For historical
reasons we have briefly presented the basic features of cellulose acetate
electrophoresis of globin chains.
4.1.1 Cellulose Acetate Electrophoresis (Alkaline and Acid pH)

First the heme groups and the globin chains are dissociated from the hemoglobin
molecule using 2-mercaptoethanol and urea. Electrophoresis at alkaline pH is
performed using the tris-ethylenediaminetetraacetic acid buffer at pH 8.8-9.5. In Figure
1 the relative mobilities of globin chains at alkaline pH are presented.
There is not much difference in the mobilities of globin chains between the alkaline
(8.8 9.5) and acidic (6.0-6.2) pH, and in both cases the alpha chains migrate towards
the cathode and the beta chains towards the anode. In Figure 2 the relative mobilities of
globin chains at acidic pH are presented.
132
Fig 1.
Relative Mobilities of Globin Chains (Cellulose Acetate Electrophoresis

at pH 8.8-9.5). Adopted from Laboratory Methods for Detecting
Hemoglobinopathies, Division of Host Factors, Center for Infectious
Diseases, Center for Disease Control, Atlanta, GA (September 1984)
133
Fig 2.
Relative Mobilities of Globin Chains (Cellulose Acetate Electrophoresis

at pH 6.0-6.2). Adopted from Laboratory Methods for Detecting
Hemoglobinopathies, Division of Host Factors, Center for Infectious
Diseases, Center for Disease Control, Atlanta, GA (September 1984)
References
134
1. Ueda S, Schneider RG. Rapid identification of polypeptide chains of

hemoglobin by cellulose acetate electrophoresis of hemolysates. Blood
1969; 34: 230.
2. Schneider RG. Differentiation of electrophoretically similar hemoglobinssuch as S,D,G and P; or A2, C,E, and O- by electrophoresis of the globin
chains. Clin Chem 1974; 20(9): 1111-1115.
3. Shihabi ZK, Hinsdale ME. Simplified hemoglobin chain detection by
capillary electrophoresis. Electrophoresis 2005; 26: 581-585
135
Chapter 4
4.2 Reverse-Phase High Performance Liquid Chromatography
Zia Uddin, PhD and Rita Ellerbrook, PhD
Conventional charge based separation techniques (electrophoresis, ion-exchange liquid
chromatography, and isoelectric focusing) are sometimes ineffective in the separation of
hemoglobins, when the amino acid substitution does not cause a net charge differential.
Several hemoglobin variants migrate upon electrophoresis and elute upon ion-exchange liquid
chromatography in the positions of hemoglobin A, S, D, A2 or C. Further clarification is
necessary for newborn screening or in cases of unexplained clinical disorders. Additional
testing is required to resolve this matter, e.g., DNA studies, reverse-phase chromatography
(RPC), liquid chromatography-mass spectrometry (LC-mass) primarily employing the
electrospray ionization (ESI) technique, and Sanger sequencing.
There are three main chromatographic techniques for the separation of peptides and
proteins, e.g., a) size exclusion, b) ion-exchange, and c) hydrophobic interactions. For a
detailed study of the theory and practice of the liquid chromatography of peptides and proteins
in general and reverse phase high pressure liquid chromatography (RP-HPLC) in particular
(Chapter 13, Section 13.4), the interested reader is advised to review the 3 rd edition of
Introduction to Modern Liquid Chromatography, by Lloyd L. Snyder, Joseph J. Kirkland and
John W. Dolan (A John Wiley and Sons, Inc. Publication, 2010). Howard and Martin 1 first
introduced RPC in 1950, and since then, several improvements in the methodology and
advancements in its application in the separation of peptides and proteins were achieved. The
136
recent literature on RPC can be accessed via the Internet (http://www.lcresources.com) and
the specialized journals in the field.
The separation of globin chains by RP-HPLC is based on the hydrophobicity of the
globin chains, which is defined as a tendency of not combining with water or incapable of
dissolving in water. The RP-HPLC consists of a non-polar column in combination with a polar
mixture of water plus an organic solvent as a mobile phase. In this section, we shall
demonstrate the usefulness of RP-HPLC in the separation of globin chains leading to the
identification of hemoglobin variants. Experimental details of RP-HPLC of globin chains
(hemoglobin specimen preparation, selection of column, solvent system, high pressure liquid
chromatography instrumentation, temperature, retention times, detection system, etc.) were
provided by the work of three research groups in this field in Italy, France and USA 2-7.
A few RP-HPLC chromatograms (Fig 1-5) are shown to illustrate the application of this
technique in the separation of globin chains. These chromatograms are either replicated
exactly as cited in the literature (abscissa depicting actual retention times in minutes), or for
comparison, as a normalized scale for the retention times. In the normalized scale, the
retention time for the normal chain is 10, and for the normal chain is 20 (Fig 4). The elution
window is of 0.5 units width. It is emphasized that the retention times of RP-HPLC might vary
depending upon the experimental conditions, but the overall shape of the chromatogram is
highly reproducible.
137
Normal Cord Blood: Fetal blood obtained at 18-20 weeks of gestation age, shows the
preponderance of chains (Fig 1).
Fig 1. RP-HPLC chromatogram of a normal cord blood (Leone L, Monteleone M,

Gabutti V, Amione C. Reversed-Phase High Performance Liquid Chromatography of Human
Hemoglobin Chains. J Chromatogr. 1985, 321: 407-419) 2.
Normal Adult Blood: The first peak (Fig 2) at 10 minutes is the heme molecule
followed by two major peaks, a) chain of Hb A (31-35 minutes), and b) chain
of Hb A (43-48 minutes).
138
Fig 2. RP-HPLC chromatogram of a normal adult blood (Kutlar F, Kutlar A, Huisman THJ.
Separation of Normal and Abnormal Hemoglobin Chains by Reverse-Phase High Performance
Chromatography. J. Chromatogr 1986, 357: 147-153) 3.
Adult hemoglobin A-S trait : In hemoglobin S the variation in the chain is due to the
substitution of glutamic acid by valine [6(A3)]. This is shown in Fig 3 by the separation of A
and S chains.
139
Fig. 3. RP-HPLC chromatogram of an adult Hb A-S trait (Leone L, Monteleleone M, Gabutti V,

Amione C. Reversed-Phase High Performance Liquid Chromatography of Human Hemoglobin
Chains. J. Chromatogr 1985, 321: 407-419)2.
Hemoglobin S interacts with Hb D-Punjab [121(GH4) GluGln] causing sickle cell
disease. Hemoglonin S also interacts with Hb Korle-Bu [73(E17) AspAsn], but in the
opposite direction, i.e., inhibiting sickling. Both of these hemoglobin variants
(Hb D-Punjab and Hb Korle-Bu) are frequently found in the population sector dominated by Hb
S. The separation of Hb D-Punjab and Hb Korle-Bu is difficult from cellulose acetate
140
electrophoresis and isoelectric focusing, however both the D-Punjab and Korle-Bu chains can be
easily separated by RP-HPLC7.
Several electrophoretic separation techniques did not distinguish 4 Hb Camperdown
[104(G6) ArgSer] from Hb Sherwood Forest [104(G6) ArgThr]. In this example there is
no change on the charge of the two hemoglobin variants as only the hydrogen atom on serine
is replaced by a methyl group of threonine. The substitution of serine by threonine on the same
position of the chain changes the hydrophobicity (presumably by altering the
secondary/tertiary structure of the globin chain), thus resulting in their separation by RP-HPLC
(Fig 4).
Fig. 4. Normalized scale of retention times of globin chains on RP-HPLC, a) retention times of
A (10), (20), G (28) and A (35), b) Hb Campertown (14.1-14.5), and c) Hb Sherwood
Forest (16.1-16.5). Adopted from: Wajcman H, Riou J, Yapo AP. Globin Chain Analysis by
Reversed Phase High Performance Liquid Chromatography: Recent Developments.
Hemoglobin 2002, 26: 272-2844.
141
Another interesting illustration8 of the usefulness of RP-HPLC was the resolution of

hemoglobinopathy during newborn screening, provided by Hb Sinai-Greenspring [34(16)
ValIle, GTC>ATC]. IEF showed an abnormal band (slightly anodal to HbA), and HPLC (Fig
5a) was also inconclusive except the broadening of the band due to a hemoglobin variant. RPHPLC did indicate a distinct band due to a variant Hb between the and chains (Fig 5b).
Substitution of amino acid valine at position 34 of the -globin chain by isoleucine changed the
hydrophobicity of the protein molecule and thus allowed the separation of two chains by RPHPLC (Fig 5b).
142
Fig 5. Cation exchange HPLC chromatogram (a) of infant with Hb-Sinai-Greenspring,

and RP-HPLC chromatogram (b). Adopted from: Dainer E, Wenk RE, Luddy R, Elam D,
Holley L, Kutlar A, Kutlar F. Two new hemoglobin variants: Hb Sinai-Greenspring [34
(16) ValIle, GTC>ATC] and Hb Sinai-Bel Air [53 (D4) AlaAsp, GCT>GAT].
Hemoglobin 2008; 32(6): 588-5918
143
Henri Wajcman and associates published the retention times on RP-HPLC of over 200
abnormal globin chains which were also made available on the web 7. Additional
chromatographic and electrophoretic information about hemoglobin variants can be obtained
from the database9-11.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Howard GA, Martin JP. The separation of the C12-C18 Fatty Acids by ReversedPhase Partition Chromatography. Biochem J 1950; 46: 532-538.
Leone L, Monteleone M, Gabutti V, Amione C. Reversed-Phase High
Performance Liquid Chromatography of Human Globin Chains. J Chromatogr
1985; 321: 407-419.
Kutlar F, Kutlar A, Huisman THJ. Separation of Normal and Abnormal Hemoglobin
Chains by Reversed-Phase High-Performance Liquid Chromatography. J
Chromatogr 1986; 357:147-153.
Wajcman H, Riou J, Yapo AP. Globin Chain Analysis by Reversed Phase High
Performance Liquid Chromatography: Recent Developments. Hemoglobin 2002; 26:
271-284.
Yapo PA, Datte JY, Yapo A, Wajcman H. Separation of Adult Chains of Abnormal
Hemoglobin: Identification by Reversed-Phase High-Performance Liquid
Chromatography. J Clin Lab Anal 2004;18: 65-69.
Zanella-Cleon I, Becchi M, Lecan P, Giordano PC, Wajcman H, Francina A.
Detection of a Thalassemic -Chain Variant (Hemoglobin Groene Hart) by
Reversed-Phase Liquid Chromatography. Clin Chem 2008; 54:1053-1059.
Wajcman H, Riou J. Globin chain analysis: An important tool in phenotype
study of hemoglobin disorders. Clinical Biochemistry 2009; 42:1802-1806.
Dainer E, Wenk RE, Luddy R, Elam D, Holley L, Kutlar A, Kutlar F. Two new
hemoglobin variants: Hb Sinai-Greenspring [34 (16) ValIle, GTC>ATC] and Hb
Sinai-Bel Air [53 (D4) AlaAsp, GCT>GAT]. Hemoglobin 2008; 32(6): 588-591
Hardison RC, Chui DHK, Giardine B, et al. HbVar: a relational database of
human hemoglobin variants and thalassemia mutations at the globin gene server.
Hum Mutat 2002; 19: 225-33 (http://globin.bx.psu.edu/hbvar/smenu.html).
Giardine B, van Baal S, Kaimakis P, et al. HbVar database of human
hemoglobin variants and thalassemia mutations: 2007 update. Hum Mutat
2007; 28(2): 206.
Patrinos GP, Giardine B, Riemer C, et al. Improvements in the Hbvar
database of human hemoglobin variants and thalassemia mutations for
population and sequence variation studies. Nucleic Acids Res 2004 Jan 1;
32: D537-41(Database issue).
144
Chapter 4
4.3 Globin Chain Gene Mutations: DNA Studies
2
4.3.1 Introduction
Hemoglobin A is the designation for the normal hemoglobin that exists after birth.
Hemoglobin A is a tetramer with two alpha chains and two beta chains (22).
Hemoglobin A2 is a minor component of the hemoglobin found in red cells after birth
and consists of two alpha chains and two delta chains (22). Hemoglobin A2 generally
comprises less than 3% of the total red cell hemoglobin. Hemoglobin F is the
predominant hemoglobin during fetal development. The molecule is a tetramer of two
alpha chains and two gamma chains (22). Hemoglobinopathies result from amino
acid changes in the alpha or beta globin chains. Most of the mutations are single amino
acid substitutions caused by a single base change, however, other amino acid
mutations can be found due to various base alterations such as:
1.
More than one amino acid change e.g. the alpha chain mutation of Hb J
Singapore with Asn>Asp and Ala>Gly, the beta chain mutation of Hb
Poissy with Gly>Arg and Ala>Pro.
2.
Elongation of the chain due to frameshifts or insertions such as Hb

Constant Spring or Hb Doha.
3.
Shortened chains due to deletions such as Hb Leiden.
4.
Hybrids such as the Lepore globin gene that is a crossover of beta and
delta globin genes that produces hemoglobin made up of two normal
145
alpha chains and two Hb Lepore chains.

Additionally, though much rarer, there are also changes in the gamma chains
(Hb F) and delta chains (Hb A2). Over 1,000 hemoglobin mutations have been
described. For a detailed list of the mutations, the reader is directed to the Globin Gene
Server of Pennsylvania State University at:
http://globin.cse.psu.edu/html/ and Department of Microbiology of the
University of Massachusetts at:
http://www.umass.edu/microbio/chime/hemoglob/index.htm.
Mutations that cause diminished production of the globin molecules are termed
Thalassemia. Equal numbers of alpha and beta chains are necessary for normal
hemoglobin synthesis.
4.3.2 Genotyping - PCR Methodology

Determining the genotype requires DNA from the subject and the synthesis of a
primer and probe for the known mutation. The subjects DNA, a primer, a reporting
probe, DNA bases, and DNA polymerase enzyme are incubated a number of times to
amplify the mutation sufficiently to be detected with a labeled probe. However, the
procedure has limitations; the first is that the mutation must be known so that a unique
primer and probe can be made, secondly, a sufficient amount of sample DNA must be
present to make a sufficient quantity of PCR product (amplicon) which is then detected
and reported by the probe.
Methods that have been employed over the years for identifying single mutations are:
146
1 Restriction Fragment Length Polymorphism (RFLP) detection in which

specific restriction enzyme digested DNA is separated by electrophoresis 1.
2.
Binding of a labeled Allele-Specific Oligonucleotide (ASO) probe to

amplified DNA2.
3.
Allele-Specific PCR (ASP), PCR Amplification of Specific Alleles (PASA), or

Amplification Refractory Mutation System (ARMS), in which the presence or
absence of a normal or mutant sequence is determined by whether the PCR
products generated by specific primers can be detected through a reporting
system such as electrophoresis, or a fluorescent, chemical, colorimetric, or
electric signal. The signals may be read directly by the human eye
(electrophoresis) or detected by instrumentation in which case they may also be
quantitated3.
Some additional methods for multi-mutation detection by PCR assays include:

1
Allele-Specific Primer Extension (ASPE) assays that detect the

incorporation of a labeled nucleotide that binds at a single nucleotide
polymorphism (SNP) and is linked to an oligonucleotide that is bound next
to the SNP site3.
2.
Binding labeled multiplex ASPE products to mutation specific beads that

can generate identifying signals in solution when separated by laser flow
cytometry as is done by the Luminex4.
3.
Oligonucleotide Ligation Assay (OLA) based on the binding and ligation of

an allele-specific probe to a common downstream sequence reporter
147
probe, which generates a specific fluorescent signal from the completed

ligation products separated on electrophoresis 5.
4.
Hybridizing PCR amplification products to electrode-bound allele-specific

probes (printed circuit board, microarray, chip-based) to generate electric
signals6.
5.
Fluorescence Resonance Energy Transfer (FRET) fluorescent signals

generated by Cleavase treated PCR products 7.
PCR amplification products are produced by incubating extracted DNA from the
specimen with DNA primers, the substrate nucleic acid bases of adenosine, thymidine,
guanosine, and cytidine, DNA polymerase, and a DNA detection probe. The mixture is
repeatedly heated to ~ 95 C and cooled; each heating and cooling cycle doubles the
amount of PCR product produced; most PCR assays use 25-40 cycles.
Rapid cycle PCR is based upon the low heat capacity of air and the ability to
ramp through temperatures at a far greater rate than instruments using thermocyclers
that rely upon heating and cooling an aluminum block.
Instruments such as the
LightCycler from Roche also incorporate the improvement of using glass capillary
tubes to serve as both the reaction vessel and optical cuvette. Detection is by the
Fluorescence Resonance Energy Transfer (FRET) method described below, however,
the time required to complete 25-40 cycles is on the order of 30-40 minutes as opposed
to 3-4 hours for aluminum block thermocyclers.
Detection of FRET probes is performed by measuring hybrid stability as modified
by the introduction of base pair mismatch/es. Mismatch destabilization is measured by
148
observing the melting temperature of the FRET probe as monitored by fluorescence

output.
Fluorescence is generated using two fluorophores.
The first or Light-Up
fluorophore is excited at an appropriate wavelength. The emission of the Light-Up

fluorophore is in turn used to excite the detection fluorophore.
emission of the detection fluorophore is monitored.
The subsequent
In order for resonance energy
transfer to occur between the Light-Up and detection fluorophores and produce a
florescent signal, the two fluorophores must be in close proximity. Proximity is achieved
by conjugating the fluorophores to oligonucleotides such that when the oligonucleotides
are hybridized to their target in an amplicon, the fluorophores are held in proximity. The
mixture is then heated and a melting curve is generated by the slow thermal denaturing
of the probe-template hybrid. Melting curves are generated by monitoring the loss of
fluorescence over the course of denaturation. Melting peaks are generated by plotting
the inverse derivative of fluorescence verses temperature (-dF/dT) - the bigger the
mismatch between the amplicon and the probe, the lower the melting temperature.
Because most hemoglobinopathies are single amino acid mutations such as
base substitution or base pair insertion or deletion, the ASP method is the commonly
used technology. In this procedure, allele-specific primers for sequences are designed
to bind to and amplify a small region surrounding the site of the known mutation. A
probe of oligonucleotides, which matches the normal or abnormal sequence, binds to
the PCR products. The probes incorporate a label (fluorophore) that produces a signal
to show that binding has taken place and a specific sequence has been detected.
149
4.3.3 Mutations
Hemoglobin beta is the name of the hemoglobin gene and is abbreviated HBB.
Sickle cell anemia is the most common mutation and primarily affects African-Americans
with a frequency of 1:400.
The defect causes red cells to distort and block small
capillaries. The -globin gene is located on the small (p) arm of chromosome 11 in the
region of 15.5 (HBB; MIM # 141900; 11p15.5). The mutation is the replacement of an
adenosine with a thymidine in the DNA that causes the substitution of valine for glutamic
acid at position 6 in the beta-globin chain. The codon sequence is shown below, GAG,
in the sixth position below, codes for glutamic acid; the replacement of adenosine (A)
with thymidine (T) produces GTG that codes for valine.
1
3
GTG GAC CTG ACT
Val
Asp
Leu
Thr
GTG GAC CTG ACT
6
9
CCT GAG GAG AAG TCT
Glu
Pro
Glu Lys Ser
Val
CCT GTG GAG AAG TCT
- - - (wildtype)
- - - (Hb S)
Hemoglobin C is a mutation in the same codon which replaces the first

guanosine with adenosine (GAG becomes AAG) causing the glutamic acid to be
replaced with lysine.
1
3
GTG GAC CTG ACT
Val
Asp
Leu
Thr
GTG GAC CTG ACT
6
9
CCT GAG GAG AAG TCT
Glu
Pro
Glu Lys Ser
Lys
CCT AAG GAG AAG TCT
150
- - - (wildtype)
- - - (Hb C)
Similar single point mutations cause other variants of hemoglobin. Hemoglobin E

results when glutamic acid is replaced with the amino acid lysine at position 26 in globin (Glu>Lys) due to the same GAG>AAG mutation that causes hemoglobin C at
codon 6. It is the second most common hemoglobin variant. When the hemoglobin E
mutation is present with hemoglobin S, Hb SE disease, the person may have more
severe signs and symptoms associated with sickle cell anemia, such as episodes of
pain, anemia, and abnormal spleen function.
Hemoglobin D-Punjab also known as Hb D-Los Angeles, Hb D-Chicago, Hb DNorth Carolina, Hb D-Portugal, and Hb Oak Ridge is an abnormality due to the
replacement of glutamic acid with glutamine on the hemoglobin beta chain. The
mutation is GAA>CAA
at codon 121 (121 Glu>Gln). Hb D is primarily found in the
Indus River Valley (Punjab) region of Pakistan and Northwestern India but is
widespread, and has been observed in persons from China, England, Holland,
Australia, Greece, Serbia, Bosnia Herzegovina, Macedonia, Montenegro, and Turkey.
It is the fourth most frequently occurring hemoglobin variant.
Heterozygotes for Hb D are normal. Homozygosity for Hb D is associated with
normal hemoglobin levels, decreased osmotic fragility, and some target cells.
Compound heterozygotes for Hb D and -Thalassemia have mild anemia and
microcytosis.
Hb D has been found in combination with Hb S, Hb C, Hb E, -
thalassemia, and in the homozygous state. Hemoglobin D has been shown to interact
with the sickle hemoglobin gene S. Individuals who are compound heterozygotes for
Hb S and Hb D-Los Angeles (SD) have moderately severe hemolytic anemia and
151
occasional pain episodes. Populations that have a high frequency of sickle hemoglobin
(SD) disease are those of Asian and Latin American descent.
Hemoglobin O-Arab is an abnormality due to the amino acid substitution of lysine
for glutamic acid at the 121st position in the beta globin gene. The genetic mutation is a
GAA>AAA at this codon (121 Glu>Lys). The mutation is also known as Hb Egypt and
Hb O-Thrace. The mutation is found mainly in African-Americans, Gypsies, in Pomaks
(a population group in the Balkan countries) and in Arabian, Egyptian, and Black
families of the US and western hemisphere.
-------------------------------------------------------------------------- See the DNA codon table for degeneracy (redundant) codons.
Hemoglobin O-Arab is important when found with sickle syndromes. Compound
heterozygotes for Hb S and Hb O-Arab have hemoglobin concentrations in the range of

7-8 g/dL with reticulocytosis, jaundice, splenomegaly, episodes of pain, and many other
complications seen in Hb SS disease.
manifestations.
microcytosis.
Heterozygote carriers have no clinical
Homozygous individuals usually present with mild anemia and

Compound heterozygotes for Hb O-Arab and -thalassemia have
manifestations similar to thalassemia intermedia.

Thalassemias are named by the chain that is deficient. In -Thalassemia, there
is an insufficient amount of the beta subunit due to mutations such as -29A>G, -88C>T,
and IVS1+6T>C. The excess alpha subunits precipitate and eventually damage the red
blood cells.
In severe -thalassemia, the -globin subunits begin to associate into
tetramers due to the reduced concentration of alpha chain. The tetramers of -globin
do not transport oxygen.
No comparable tetramers of -globin subunits form with
severe -thalassemia.
152
Below are several melting curves representing the various signals obtained
during an analysis. In allele specific binding assays, it is preferred that the primer and
detection probe are the sequences for the mutation and not the wildtype (normal)
sequence. Because the mismatch in base sequences causes the melting temperature
to be lower, the use of the wild type sequence as the detection probe will indeed
demonstrate a lower melting temperature when a mismatch is present, however it will
not be known as to which base/s mismatch (mutation) was present. The use of the
mutation as the template will always result in the specific mutation producing the highest
melting temperature.
The Hemoglobin S templates were used in the analysis of wildtype (normal)
hemoglobin in Figure 1 and shows a melting point of 55.5 oC.
Figure 1. Hemoglobin A (WT*) bound to Hemoglobin S probe, melting point is 55.5 C.

* WT - Wild Type, the commonly occurring type - no mutation.
153
Figure 2 shows a melting curve for a carrier, both hemoglobin sequences were
detected.
Hemoglobin S has the higher melting temperature of 62.5 C while the
wildtype melts at 55.5 C. A homozygous sickle disease individual would show only one
melting point at 62.5 C.
Figure 2. Hemoglobin A (WT) and Hemoglobin S (Mutant) bound to Hemoglobin S

probe. Melting temperatures: WT - 55.5 C and Mutant S - 62.5 C.
As pointed out earlier, Hemoglobin S and Hemoglobin C differ from the wild type
by only one base in the same codon number 6 of the HBB gene. The Hb C mismatch
causes an even lower melting temperature than Hb S or the wildtype. Figure 3 shows
two melting points indicating a Hemoglobin C carrier with Hb C melting at 49.8 C and
the wildtype again at 55.5 C. This detection of two mutations is an example of a
multiplexed assay. This type of multiplexing is only useful when the bases involved in
154
the mutation are very close, e.g. 3 bases, otherwise the energy transfer would not be
very efficient and no fluorescent signal would be detected.
Figure
3.
Hemoglobin C (Mutant) bound to Hemoglobin S probe.
Melting temperatures: Mutant C - 49.8 C and WT - 55.5 C
155
Figure 4 is an example of a non-preferred base sequence for the Hemoglobin E

mutation in which the wildtype probe is used to detect the mismatch at codon 26 of
GAG to AAG. Here the wildtype melts at 70.3 C and the Hb E mutation melts at 65.2
C.
Figure 4. Hemoglobin E (mutant) bound to Hemoglobin A (WT) probe.

Melting temperatures: Mutant E - 65.2 C and WT - 70.3 C
156
Figure 5 illustrates an analytical run in which a normal, no Hemoglobin D

present patient is in red, a carrier control specimen is in blue, and the green line is the
no DNA control which must not give a signal.
Figure 5. Hemoglobin A (WT) and Hemoglobin D (Mutant) bound to Hemoglobin D

Probe. Melting temperatures: WT - 50.95 C and Mutant D - 63.68C.
157
Figure 6 is an example of a melting curve for Hemoglobin O. Note that among

the assays shown, there is no correlation of melting temperatures for the wildtype or
mutations. This is because each primer and probe set is different for each specific
mutation.
Figure 6. Hemoglobin A (WT) and Hemoglobin O (Mutant) bound to Hemoglobin D

Probe. Melting temperatures: WT - 51.52 C and Mutant D - 61.91C .
158
Degeneracy Table
Amino Acid
DNA codons
Alanine
GCT, GCC, GCA, GCG
Arginine
CGT, CGC, CGA, CGG, AGA, AGG
Asparagine
AAT, AAC
Aspartic acid
GAT, GAC
Cysteine
TGT, TGC
Glutamic acid
GAA, GAG
Glutamine
CAA, CAG
Glycine
GGT, GGC, GGA, GGG
Histidine
CAT, CAC
Isoleucine
ATT, ATC, ATA
Leucine
CTT, CTC, CTA, CTG, TTA, TTG
Lysine
AAA, AAG
Methionine
ATG
Phenylalanine
TTT, TTC
Proline
CCT, CCC, CCA, CCG
Serine
TCT, TCC, TCA, TCG, AGT, AGC
Threonine
ACT, ACC, ACA, ACG
Tryptophan
TGG
Tyrosine
TAT, TAC
Valine
GTT, GTC, GTA, GTG
Stop codons
TAA, TAG, TGA
159
DNA Codon Table
First
T
TTT
TTC
TTA
TTG
C
Phe
Leu
CTT
C
CTC
CTA
Leu
TCT
TCC
TCA
TAT
TAC
Ser
TAA
TCG
TAG
CCT
CAT
CCC
CCA
CAC
Pro
CAA
CTG
CCG
CAG
ATT
ACT
AAT
ATC
Ile
ATA
ATG
GTC
GTA
GTG
ACC
ACA
Met
GTT
G
Val
AAC
Thr
AAA
ACG
AAG
GCT
GAT
GCC
GCA
GCG
Ala
GAC
GAA
GAG
G
Tyr
stop
His
Gln
Asn
Lys
Asp
Glu
TGT
TGC
Third
Cys
T
C
TGA
stop
TGG
Trp
CGT
CGC
CGA
T
Arg
CGG
AGT
AGC
AGA
AGG
GGA
GGG
A
G
Ser
Arg
GGT
GGC
T
C
A
G
T
Gly
C
A
G
References
2
3
4
1 3Enzymatic amplification of beta-globin genomic sequences and restriction site

analysis for diagnosis of sickle cell anemia. R. K. Saiki, S. Scharf, F. Faloona, K.
B. Mullis, H. A. Erlich, and N. Arnheim, Science (1985) 230:13501354.
4Specific Enzymatic Amplification of DNA In Vitro: The Ploymerase Chain Reaction. K.
Mullis, F. Faloona, S. Scharf, R. Saiki, G. Horn, H. Erlich, Cold Spring Harbor Symposia
on Quantitative Biology (1986) LI:263-273.
5PCR Second Edition - The Basics. M. McPherson and S. Moller, Taylor & Francis
Pub., New York (2006).
Luminex Corporation, 612212 Technology Boulevard, Austin, TX 78727.
160
5
6
Automated DNA diagnostics using an ELISA-based oligonucleotide ligation assay. D. A.

Nickerson, R. Kaiser, S. Lappin, J. Stewartt, L. Hood, and U. Landegrent, Proc. Nat.
Acad. Sci. USA (1990) 87:8923-8927.
7Design of electrochemical biosensor systems for the detection of specific DNA
sequences in PCR-amplified nucleic acids related to the catechol-O-methyltransferase
Val108/158Met polymorphism based on intrinsic guanine signal. D. Ozkan-Ariksoysal,
B. Tezcanli,B. Kosova, and M.Ozsoz, Anal Chem. (2008) 80(3):588-596.
New Cleavase Fragment Length Polymorphism Method Improves the Mutation
Detection Assay, M. C. Oldenburg and M. Siebert, BioTechniques (2000) 28:351-357.
General References
2
3
4
5
6
7
8
9
C8ompound Heterozygosity Hb S/Hb Hope ( 136Gly>Asp): a Pitfall in the Newborn

Screening for Sickle Cell Disease. R. Ducrocq, A. Bevier, A. Leneveu, M. MaierRedelsperger, J. Bardakdian-Michau, C. Badens, and J. Elion, Journal of Med
Screening (1998) 5:27-30.
Rapid -globin Genotyping by Multiplexing Probe Melting Temperature and Color. M.
Herrmann, S. Dobrowolski, and C. Wittwer, Clinical Chemistry (2000) 46:425-428.
Identification of Hb D-Punjab gene: application of DNA amplification in the study of
abnormal hemoglobins. Y. T. Zeng., S. Z. Huang, Z. R. Ren, and H. J. Li, Am J Hum
Genet. (1989) 44(6):886-9.
The inherited diseases of hemoglobin are an emerging global health burden. D. J.
Weatherall. Blood (2010) 115:4331.
Percentages of abnormal hemoglobins in adults with a heterozygosity for an alphachain and/or a beta-chain variant. T. H. Huisman. Am J Hematol (1983) 14:393.
http://www.ncbi.nlm.nih.gov/books/NBK1426/
Beta-Thalassemia,
GeneReviews
[Internet], A. Cao and R. Galanello. (2000 updated 2010).
Construction of a Genetic Linkage Map in Man Using Restriction Fragment Length
Polymorphisms. D. Botstein, R. L. White, M. Skolnick, and R. W. Davis, Am J Hum
Genet (1980) 32:314-331.
Specific Enzymatic Amplification of DNA In Vitro: The Ploymerase Chain Reaction. K.
Mullis, F. Faloona, S. Scharf, R. Saiki, G. Horn, H. Erlich, Cold Spring Harbor Symposia
on Quantitative Biology (1986) LI:263-273.
High-throughput SNP genotyping. S. Jenkins and N. Gibson. Comparative and
Functional Genomics (2002) 3(1):57-66.
161
Chapter 4
Global Chain Analysis
4.4 Electrospray Ionization-Mass Spectrometry
Gul M. Mustafa, PhD and John R Petersen, PhD
Mass spectrometry (MS) is an analytical technique that identifies the chemical
composition of a sample on the basis of the mass-to-charge ratio (m/z) of charged ions.
The technique has both qualitative (structure) and quantitative (molecular mass or
concentration) uses. Another way of thinking about mass spectrometry is that it can be
considered as the worlds most accurate scale. Mass spectrometers can be divided
into three fundamental parts, namely the ionization source, the analyzer, and the
detector (Figure 1). The molecules of interest are first introduced into the ionization
source of the mass spectrometer, where they are ionized to acquire positive or negative
charges. This is done because ions are far easier to manipulate as compared to
molecules that do not have a charge. The ions then travel through the mass analyzer
and arrive at different parts of the detector according to their mass to charge (m/z) ratio.
After the ions make contact with the detector, useable signals are generated and
recorded via a computer. The computer displays the signals graphically as a mass
spectrum showing the relative abundance of the signals according to their m/z ratio. The
analyzer and detector of the mass spectrometer, and often the ionization source too, are
maintained under high vacuum to allow the ions to travel from one end of the instrument
to the other without colliding with air molecules which decreases the signal. The entire
operation and often the sample introduction process are under complete data system
162
control on modern mass spectrometers.
163
Figure: 1
The method of sample introduction to the ionization source often depends on the
ionization method being used, as well as the type and complexity of the sample. Many
ionization methods are available and each has its own advantages and disadvantages.
The ionization method used depends on the nature and type of sample under
investigation and the mass spectrometer available. Figure 2 shows various ionization
methods of ionization such as Atmospheric Pressure Chemical Ionization (APCI),
Atmospheric Pressure Photo-Ionization (APPI), Electron Impact (EI), and Electrospray
Ionization (ESI). The ionization methods used for the majority of biochemical analyses
are Electrospray Ionization (ESI) and Matrix Assisted Laser Desorption Ionization
(MALDI)
Figure: 2
164
Mass spectrometry using ESI is called electrospray ionization mass spectrometry

(ESI-MS) or, less commonly, electrospray mass spectrometry (ES-MS). Electrospray
ionization mass spectrometry was pioneered by John Bennet Fenn, who shared the
Nobel Prize in Chemistry with Koichi Tanaka in 2002 for his work on the subject (1). One
of the original instruments used by Dr. Fenn is on display at the Chemical Heritage
Foundation in Philadelphia, Pennsylvania. This technique of ionization is especially
useful in producing ions from macromolecules because it overcomes the propensity of
these molecules to fragment when ionized and as such is considered a soft ionization
technique. When analyzing biological molecules of large molecular mass, ESI-MS is
very useful because it does not cause fragmentation of the macromolecules into smaller
charged particles; rather it creates small droplets containing the macromolecule being
ionized and solvent allowing analysis of the molecular weight of the intact
macromolecule. Solvent can then be removed causing the formation of even smaller
droplets, creating protonation of the macromolecules. These protonated and desolvated
molecular ions are then passed through the mass analyzer to the detector, and the
mass of the sample is determined (Figure: 3). This method can be performed on solid or
liquid samples, and allows analysis of nonvolatile or thermally unstable molecules which
means that ionization of proteins, peptides, olgiopeptides, and some inorganic
molecules can be easily performed. The spectrum is shown with the mass-to-charge
(m/z) ratio on the x-axis, and the relative intensity (%) of each peak shown on the yaxis. The quantitative analysis is done by considering the mass to charge ratios of the
various peaks in the spectrum. Calculations to determine the unknown mass, (Mr) from
the spectral data are performed using; p = m/z.
165
The ionization mechanism first involves the liquid containing the analyte(s) of
interest to be dispersed by electrospray into a fine aerosol. Because the ion formation
involves extensive solvent evaporation, the typical solvents for electrospray ionization
are prepared by mixing water with volatile organic compounds (e.g. methanol,
acetonitrile). To decrease the initial droplet size, compounds that increase the
conductivity (e.g. acetic acid) are customarily added to the solution. Large-flow
electrosprays can benefit from additional nebulization by an inert gas such as nitrogen.
Figure: 3
There are some clear advantages and disadvantages of using electrospray

ionization mass spectrometry as an analytical method. It is one of the softest ionization
methods available; thus it can not only analyze molecules that have high molecular
166
masses but also has the ability to analyze biological samples that are defined by noncovalent interactions. Since the m/z ratio range of a quadrupole instrument is fairly
small, the mass of the sample can be determined with a high amount of
accuracy. Sensitivity of the instrument is also impressive making it useful in both
quantitative and qualitative measurements. The major disadvantage of ESI-MS is that in
the analysis of mixtures the results are unreliable. In addition to the difficulty in handling
mixtures the multiple charges that are attached to the molecular ions can make for
confusing spectral data. The apparatus is also very difficult to clean and has a tendency
to become contaminated with residues from previous experiments.
In recent years, electrospray ionization (ESI) mass spectrometry has become an
increasingly important method in proteomics not only to analyze peptides but also to
study proteins and protein complexes of increasing size and complexity in structural
biology. The analysis of proteins and protein complexes by mass spectrometry
(macromolecular mass spectrometry) has become possible because of the
development of the relatively gentle ionization procedure related to ESI, which retains
non-covalent interactions. The mass-to-charge (m/z) ratios of these proteins can well be
over 10,000 daltons, and therefore, time-of-flight (TOF) analyzers with orthogonal
injection are the most commonly used analyzers in the field of macromolecules. The
m/z analysis of larger proteins and protein complexes is not a routine technique, since a
careful optimization of the operating conditions is always required. Despite the
theoretically unlimited mass range of TOF analyzers, most instruments have detection
problems when the m/z values exceed 4,000 daltons. It has been shown that a pressure
increase in the first and second vacuum chamber of the mass spectrometer is an
167
absolute requirement for the analysis of large proteins (2-6). The increased pressure
leads to collisional cooling and focusing of large ions in the ion guides and, therefore,
improved transmission through the ion guides and the TOF (5). In ESI-MS, the ion
signal is proportional to analyte concentration and largely independent of flow rate and
injection volume used for sample introduction. The signal is linear from the limit of
detection (usually pmol/L) to around 10 mol/L of analyte concentration. For quantitative
measurements, it is important to incorporate an internal standard in the procedure to
compensate for losses during sample preparation and variable detection sensitivity of
the MS system. The internal standard should have a structure similar to that of the
analyte and the ideal practice is to synthesize an internal standard by incorporating
stable isotopes on the molecules of interest. When an ideal internal standard is not
available, molecules with similar structure can also be used. Another critical issue in
quantitative ESI-MS is suppression of ionization due to matrix interference. A biological
sample can give significantly lower ionization signals compared to pure standard
solutions with similar analyte concentrations. This phenomenon is the result of high
concentrations of non-volatile materials, such as salts and lipids, present in the spray
with the analyte. To overcome the matrix interference, extensive sample purification
processes are required. However, these elaborate procedures are time-consuming and
can cause poor recovery. A recent development is to use short Liquid Chromatography
(LC) columns (or guard columns) and apply a fast High Pressure Liquid
Chromatography (HPLC) purification (e.g. for 25 minutes) prior to MS analysis. The
HPLC serves to separate the non-volatile compounds from the analyte. For HPLC
systems with column-switching capability, the analyte in the biological sample can be
168
purified and concentrated on separate columns before MS analysis. Unlike many other
techniques which measure one analyte at a time, these techniques can measure
multiple analytes (>40) at one time. In recent years the scope of testing using these
techniques has expanded from toxicological purposes to newborn screening to
hormones, proteins, and enzymes.
In recent years a change in the way MS is being used in clinical laboratories has
occurred. In the past MS was commonly used in conjunction with gas chromatograph
(GC). Today it is not uncommon to see MS being coupled to LC in the routine clinical
laboratories. Once considered too expensive and cumbersome to use except in forensic
and reference settings, such systems are now used routinely to generate data for
patient care. Although mass spectrometry has long been recognized as an important
and powerful analytical tool, there were a number of challenges that had to be
overcome to be used in the clinical setting for more than a few special applications. GCMS was introduced into the clinical laboratory more than two decades prior to LC-MS.
With the advent of relatively small, inexpensive, and user-friendly LC/MS and LC
tandem MS (LC/MS-MS) systems along with advances in column chemistries the door
has been opened to many analyses not possible with GC/MS (7). Although the initial
capital investment for LC ESI-MS equipment is substantial compared to other routine
clinical laboratory analyzers, its operational costs are low. The cost-effectiveness of this
technique comes from the fact that it can measure multiple analytes at the same time.
This technology can be expected to exert an important influence in how analytes, both
large and small, are detected and quantified in the clinical laboratory service.
169
Since the first report on the successful measurement of large bio-molecules by

ESI-MS, there has been a revolution in the identification of protein molecules in
biochemical research. MS also found its way into the analysis of hemoglobin (Hb)
analysis. In 1981, Wada et al. pioneered the analysis of tryptic peptides of Hb by MS.
The development of the soft ionization techniques (ESI and MALDI) has made it
possible to use MS to study intact globin chains. In 1990, the 1st application of ESI-MS
involving intact Hb chains was reported by Falick et al (8). Since then ESI and MALDITOF MS has become more common in routine Hb variant analysis.
Electrospray ionization is efficient in generating cluster ions for structural
elucidation of macromolecules. This has fostered a new and improved approach (vs.
electrophoresis) for identification and quantification of hemoglobin variants. The use of
MS techniques has led to the discovery of more than 60 new mutations and even the
intact Hb tetramer can be analyzed using a nano ESI-MS technique. Furthermore,
MALDI-TOF MS is a highly sensitive method that enables the analysis of Hb chains
from a single red blood cell. Final identification of a variant is achieved either by
molecular biology techniques or by protein sequence analysis, in which MS now also
occupies a key position. In variants with mutation sites close to the termini of the chain
were identified by ESI-MS/MS of the intact Hb chain. With the understanding of
glycohemoglobin (GHb) structure, an IFCC reference method for glycohemoglobin
assay has been established using ESI-MS. It represented a significant advancement for
the standardization of HbA1c in diabetic monitoring. ESI-MS has also become the
preferred technique for a rapid systematic approach to definitive characterization of Hb
variants. In addition, hemoglobin (Hb) variants need to be identified for the investigation
170
of hemolytic anemia, methemoglobinemia, sickle cell disease and thalassemia.

Occasionally, these variants are detected incidentally because they interfere with the
measurement of GHb.
Identification and quantification of hemoglobin variants

Globin chain analysis is an important tool in phenotype study of hemoglobin
disorders. The majority of hemoglobin variants result from changes in the amino acid
sequence of either the or non- globin chains of hemoglobin with the majority of these
changes due to a single point mutation in the globin gene. Substitution, insertion,
deletion or the combination of deletion with insertion of a different amino acid than those
normally present, results in changes to the amino acid sequence.
Worldwide, an estimated 150 million people carry Hb variants (9) and
hemoglobinopathies are the most common inherited disorders, constituting a significant
healthcare problem (10). Hemoglobin (Hb) variants lead to inherited disorders with
variable clinical manifestations. Therefore, reliable detection and identification methods
are essential. Among more than 900 hemoglobin (Hb) variants currently described in the
HbVar database of the globin Gene Server, variants with elongated chains are very rare
(11,12). In this database, Hb variants leading to a charge difference are significantly
over represented compared with neutral Hb variants. This result is surprising, because
only 5 of the 20 amino acids contain either a basic (Lys, Arg, His) or an acidic (Asp, Glu)
side chain, whereas the other 15 amino acid side chains are uncharged. Thirty-six of
171
141 amino acids in the -chain and 38 of 146 residues in the -chain are charged
residues and the rest are neutral so they cannot be detected by these traditional
analytical techniques, such as ion-exchange HPLC and isoelectric focusing (IEF) on
polyacrylamide gel, as these techniques depend on the presence of charge differences
induced by the mutation. Also in the past, definitive characterization of Hb variants
involved tedious and time-consuming analytical procedures requiring days and even
months for completion. Recently, a strategy for rapid definitive characterization of Hb
variants to identify a single mutated; inserted or deleted amino acid residue was
reported using ESI-MS. In case of Hb San Martin [b6(A3)GluVal;b85(F1) PheLeu],
the second mutation leads to an unstable protein causing chronic hemolytic anemia in
the heterozygous carrier (13). Molecular diagnosis, achieved by DNA analyses, shows
the presence of two mutations, but protein or familial studies was required to prove that
the two mutations are carried by the same allele and not interacting in trans. The
identification by MS methods of a new Hb variant: Hb S-Clichy [b6(A3)GluVal;b8(A5)
LysThr], which presents a double mutation located on the same bT-1 tryptic peptide.
This new variant adds the amino acid substitution of Hb Rio Grande[b8(A5) LysThr]
(14) on the same b-globin chain, to that of Hb S. Difficulties encountered in structural
determinations are caused by the presence of two abnormalities in the same
polypeptide chain. Variants with two amino acid substitutions on the same globin chain
as in Hb S-Clichy, demonstrated the importance of including MS studies.
The procedure comprises the following steps:
I.
Molecular weight profiling of intact and globin chains by direct ESI-MS on a

500-fold dilution of the whole blood sample. The cluster ion spectrum is then
172
deconvoluted to a true molecular weight scale using computer software that is

usually supplied with the MS analyzer system. This step can detect Hb variants
with molecular weight difference of more than 6 Da when compared with the wild
type globin chains (15).
II.
Overnight trypsin digestion for investigation of the amino acid substitution on the
Hb variants. ESI-MS on the tryptic digest can identify the specific peptide
harboring the substituted amino acid.
III.
ESI-MS/MS of the target peptide can provide the amino acid sequence of the
peptide and thus the position of the substituted amino acid.
These performances can be applied at different steps of the globin variant
analysis process: either as a screening method or as an additional technique to confirm

the results from classical analytical methods. ESI-MS can also identify 95% of the Hb
variants in over 250 samples with a turn-around-time of not more than 2 days for each
sample, making it a powerful tool for Hb analysis.
It must be considered that the 3-dimensional structure of the globins is
determined principally by the residues that form the interhelical and helix-heme
packings (16), and substitutions in these sites may lead to conformational changes in
the proteins. The substitution effect also depends on the 3-dimensional position, viz.
internal or external. For example, the variant Hb Sun Prairie (130Ala3Pro) is silent in
IEF, whereas Hb Fontainebleu (21Ala3Pro) is detectable (12). The substituted amino
acid is internal in Hb Sun Prairie and external in Hb Fontainebleu. As a very simple
model, the calculation of the isoelectric points(pI)-shifts does not consider
conformational changes that might alter the mobility. Therefore, mutations leading to a
173
distinct conformational change can diverge from the predicted behavior. Furthermore,
the model cannot predict reliably unstable variants. Nevertheless, pI calculations and
the evaluation of the method-specific detect ability allow the prediction of the number of
the currently undetected, silent variants. So it is now recommended that other methods
that are not based on electrophoretic or chromatographic mobility should be applied in
Hb variant analysis. In this regard ESI and MALDI-TOF MS are the suggested methods
that enable the detection of variants when the mass difference between the abnormal
and the wild-type globin chains exceeds 6 Da. This limitation in MS determination is due
to the complexity between the normal and mutated globin chains which can be
overcome by using high resolution instruments (FT-ICR, Orbitrap) or by special
precautions on low resolution instruments. For these low mass differences between
normal and variant globin chains, MS analysis of digested peptides is required.
As calculated by various studies, MS method is able to detect 92% of the
undetected variants. Among MS techniques for studying Hb variants, ESI-MS is the
most frequently used and can be associated with peptide sequencing using tandem MS,
but it often gives multiple charged fragment ions. On the other hand, MALDI-TOF MS
gives single-charge peptide ions and has been used for identification of some single
mutation Hb variants. Indeed, with additional MS analysis of lysate samples 3 new
variants, Hb Zurich-Hottingen, Hb Zurich-Langstrasse and Hb Riccarton were detected
by using ESI-MS. Neither variant had a clinical impact. These neutral variants are
exclusively found by MS and are chromatographically silent. Also in an Hb Malay
sample, only the MS analysis revealed the variant chain, as opposed to cationexchange HPLC which identified it as a thalassaemia. Recapitulating, 4 out of 2105
174
samples (0.2%) or 1% of the abnormal samples would be missed without the use of MS
analysis. In ESI-MS, the sample preparation is very simple and requires only the dilution
of the lysate sample. Two important drawbacks of the MS methods are worth
mentioning. First, its insufficient resolution prevents the detection of Hb mutations with
small mass differences of the globin chains. The precision of normal low-resolution
mass measurements is insufficient to distinguish the wild-type chain from several chain
variants, such as Hb C, D, or E. Owing to the isotopic pattern, even high-resolution MS
did not separate globin chains that differed only in 1 or 2 Da from the normal chains
(17). Two intact globin chains are not observed as separate entities in MS unless their
masses differ from one another by more than 6 Da. Second, MS as described here is
only a qualitative technique, and in particular, minor Hb fractions such as HbA1C or
HbA2, which are important for diagnosis of diabetes mellitus or thalassemia,
respectively, cannot be quantified. However high resolution MS enables detection of
variants with low mass difference (<2 Da). Also different signals in the spectrum shows
isotopic pattern.
References
1. Fenn, JB. Electrospray Ionization Mass Spectrometry: How It All Began. J.
Biomol. Techl. 13:101118; 2002.
2. Sanglier S, Leize E, Van Dorsselaer A, Zal F. Comparative ESI-MS study of
approximately 2.2 M Da native hemocyanins from deep-sea and shore crabs:
from protein oligomeric state to biotope. J. Am. Soc. Mass Spectrom.14:419-429;
2003.
3. Schmidt A, Bahr U, Karas M. Influence of pressure in the first pumping stage on
analyte desolvation and fragmentation in nano-ESI MS. Anal. Chem.73:60406046; 2001.
4. Tahallah N, Pinkse M, Maier CS, Heck A. The effect of the source pressure on
the abundance of ions of noncovalent protein assemblies in an electrospray
175
ionization orthogonal time-of-flight instrument. J. Rapid Commun. Mass

Spectrom. 15:596-601; 2001.
5. Chernushevich IV, Thomson BA. Collisional cooling of large ions in electrospray
mass spectrometry. Anal. Chem. 76:1754-1760; 2004.
6. Krutchinsky AN, Chernushevich IV, Spicer VL, Ens W, Standing KG. A Collisional
damping interface for an electrospray ionization time-of-flight mass spectrometry.
J. Am. Soc. Mass Spectrom. 9: 569-579; 1998.
7. Plumb RS and Balogh MP. The changing face of LC-MS: from experts tousers.
Current Trends in Mass Spectrometry November. 1419; 2008.
8. Falick AM, Shackleton CH, Green BN, Witkowska HE. Tandem mass
spectrometry in the clinicalanalysis of variant hemoglobins. Rapid Commun Mass
Spectrom. 4:396400; 1990.
9. Shimizu A, Nakanishi T, Miyazaki A. Detection and characterization of variant and
modified structures of proteins in blood and tissues by mass spectrometry. Mass
Spectrom Rev. 25: 686712; 2006.
10. Daniel YA, Turner C, Haynes RM, Hunt BJ, Dalton RN. Rapid and specific
detection of clinically significant haemoglobinopathies using electrospray mass
spectrometry-mass spectrometry. Br J Haematol. 130:63543; 2005.
11. Hardison RC, Chui DH, Giardine B, Riemer C, Patrinos GP, Anagnou N. HbVar: a
relationaldatabase of human hemoglobin variants and thalassemia mutations at
the globin gene server. Hum Mutat. 19:22533; 2002.
12. HbVar database. http://globin.cse.psu.edu (accessed July 2011).
13. Feliu-Torres A, Eandi-Eberle S, Calvo K, et al. Hemoglobin San Martin: A new
unstable variant associated with Hemoglobin S in an Argentinean boy.
Proceedings of the 49th American Society of Hematology Meeting, Atlanta, GA,
December 811, Blood. 110:3806; 2007.
14. Moo-Penn WF, Johnson, MH, Mc Guffey, JE, Jue, DL, Therrell, BL, Jr.
Hemoglobin Rio Grande [b8(A5) LYSTHR]: A new variant found in a MexicanAmerican family. Hemoglobin. 7(1):9195; 1983.
15. Wild BJ, Green BN, Stephens AD. The potential of electrospray ionization mass
spectrometry for the diagnosis of hemoglobin variants found in newborn
screening. Blood Cells Mol Dis. 33:308-317; 2004.
16. Lesk AM, Chothia C. How different amino acid sequences determine similar
protein structures: the structure and evolutionary dynamics of the globins. J Mol
Biol. 136:22570; 1980.
17. Peter K, Marlis S, Karin Z, Oliver S, Markus S, Bernd R, Silke SD, Leopold U,
Thomas K, Claus WH, Hannes F, and Heinz. T. Mass Spectrometry: A Tool for
Enhanced Detection of Hemoglobin Variants. Clinical Chemistry. 54(1): 6976;
2008.
176
Chapter 4
4.5 PCR and Sanger Sequencing
Elaine Lyon, PhD
Molecular methods are commonly employed to identify hemoglobin variants.
Polymerase chain reaction (PCR) exponentially amplifies regions of DNA allowing direct
genotyping or targeted mutation analysis. Detecting common alpha globin deletions is
accomplished by amplifying over deletion breakpoints or using quantitative methods to detect
copy number changes. Sanger sequencing is considered the gold standard for mutation
detection, and can confirm abnormal hemoglobinopathy and thalassemia variants. Molecular
analysis confirms a diagnosis, detects carrier status, and predicts disease prenatally in highrisk pregnancies. This section will describe general methods, applications and challenges in
PCR and Sanger sequencing for alpha and beta globin molecular analysis.
4.5.1 Alpha Globin

Two alpha globin genes (HBA1 and HBA2) are present on each chromosome 16,
resulting in a normal copy number of four genes (represented by /). One or both alpha
globin genes may be deleted on a single chromosome, with the severity of disease
corresponding to the overall number of deleted alpha globin genes. If two alpha globin genes
are deleted (alpha-thalassemia trait), it is important to determine whether both genes are
deleted on one chromosome (--/), or if each chromosome contains a single gene deletion (/-). If both parents carry a chromosome with two deletions (--/), their offspring are at risk
for Hb Bart hydropsfetalis syndrome (--/--). Common deletions include 3.7kB and 4.2kB
177
deletion which eliminate a single gene, while a 20.5kB deletion and the SEA, MED, FIL and
THAI gene rearrangements delete portions of or completely HBA1 and HBA2. In one assay
commonly used in clinical laboratories, PCR primers are designed to flank the breakpoints,
amplifying a product only when the deletion is present. By multiplexing primers, any of the
common deletions can be detected in a single reaction (1). Amplification products are
visualized by gel electrophoresis (figure 1). Other methods to identify deletions include
quantitative PCR analyses such as multiplexed-ligation probe amplification (MLPA) or high
resolution (exonic level) microarray. These methods are capable of detecting known and
previously unknown alpha globin deletions and alpha globin triplications (2,3). Given the
frequency of alpha globin deletions, a molecular work-up for alpha thalassemia often begins
with a test for deletions.
Figure 1. Gel electrophoresis for common alpha globin deletions. Each patient is tested with
control primers for HBA2 and LIS genes. In a separate reaction, each patient is tested for a
common deletion with a multiplex of deletion-specific primers. M: size marker, C: control
primers (HBA2 and LIS1), D: deletion primers. Patient 1: no common deletions, patient 2; 3.7
kB heterozygous, patient 3; 3.7kb/SEA compound heterozygous. Note that in patient 2, the
control HBA2 band is not present, as this patient has a deletion of this region on both
chromsomes.
178
The alpha globin genes also harbor sequence variants, and full gene sequencing is also
available, although alpha globin sequencing poses challenges. Sequencing is performed on
both genes (HBA1 and HBA2) that are highly homologous, and primers are designed to be
specific for each gene. Sequencing the entire coding region (3 exons for each gene),
intron/exon boundaries, proximal promoter regions, 5 and 3untranslated regions, and
polyadenylation signals provides a comprehensive sequencing test. Sequencing should be
combined with deletion analysis because deletions are not detected by sequencing, and an
apparent homozygous sequence variant may be one copy of the variant with a deletion on the
opposite chromosome. Samples homozygous for the 3.7kB deletion may not be able to be
amplified for sequencing, resulting in a failed test. However, the common 3.7kB deletion has
a single functional gene, but mutations have also been described in that fusion gene. (4,5) To
be able to identify a mutation in a chromosome with the 3.7kB deletion, primers must be
designed that will amplify over the deletion breakpoints.
4.5.2 Beta Globin

Molecular genotyping assays targeting common beta globin mutations, (e.g. HbS and
Hb C) are available to confirm the mutations for sickle cell or S/C disease. But due to the
number of mutations that have been described, full gene sequencing identifies any sequence
variant and complements other types of globin analysis for hemoglobinopathies and
thalassemias. The mutation spectrum for beta globin is well characterized, and includes
coding region mutations, splice-site mutations, regulatory mutations, and deletions. Therefore
a comprehensive assay consists of sequencing the three exons of the HBB gene, the
intron/exon boundaries, the proximal promoter region, the 5 and 3 untranslated regions
179
(UTR), and known pathogenic deep intronic mutations (e.g. IVS-II-654, IVS-II-705 and IVS-II745). Large beta globin deletions or delta-beta fusion genes (e.g. Lepore,) will not be detected
by a sequencing assay designed only for HBB, and require a different analysis. Similar to
methods to detect alpha globin deletions, primers can be designed to amplify over the
breakpoints. One example of this is the 619 base pair (bp) deletion found in Indian and other
Asian populations (6). Recently, novel beta globin deletions have been detected by other
methods, such as exonic-level microarrays or MLPA (2,7). Specific large deletions in the beta
globin gene cluster is one of two molecular mechanisms that can result in HPFH. The other
mechanism is point mutations in the promoters of the gamma globin genes (HBG1 and HBG2).
4.5.3 Sequencing
Sequencing assays begin with PCR for the regions to be interrogated. Primers are
designed to avoid known variants at their 3 end which would prevent polymerase extension,
resulting in a drop-out of that allele (8). PCR products are treated with ExoSAP (exonuclease
1 from shrimp alkaline phosphatase) to remove excess primers.
PCR primers may be tagged with a M13 sequence which allows sequencing of all amplicons
from the same M13 primers. Alternatively, a second set of sequencing primers internal to the
PCR primers may be used. The sequencing reaction utilizes fluorescently labeled dideoxynucleotides (ddATP, ddTTP, ddCTP and ddGTP, collectively refered to as ddNTPs) which
lack a 3 hydroxyl group on the sugar residue and prevent the newly synthesized product from
extending to the next base when incorporated into the product. Sanger sequencing is
performed as a linear rather than exponential amplification, with separate sequence results
180
each from the 5 and 3 directions (bi-directionally). After the sequencing reactions, the
products are again processed to remove excess primers. Sephadex columns are often used
to bind the sequencing products, which are eluted as purified products.
Sequencing products are separated by capillary electrophoresis using a polymer with a singlebase resolution. The last base of each fragment is the ddNTPs with a fluorescent dye which is
incorporated into the product. The sequence is visualized as an electropherogram and aligned
to a reference sequence. The difference between the reference and the patient sequence are
examined to determine the type of mutation present. To accurately identify the mutation, it
should be identified in both the 5 and 3 direction.
4.5.4 Reporting Sequence Variants

In standard nomenclature, nucleotides are numbered from the ATG of the start codon.
The protein position is predicted and standard nomenclature is from the methionine of the
translated product, but common or traditional nomenclature (also known as legacy
nomenclature) may be from the mature protein (gene reviews for beta and alpha). For alpha
and beta globin, the traditional nomenclature differs by one amino acid than the standard
nomenclature, representing the cleavage of methionine. Reports should clarify if they are
using the standard or the common (i.e. amino acids numbered from the mature protein).
A variant is described as to whether it is structural (hemoglobinopathy) variant or
quantitative (thalassemia variant). For example, a beta thalassemia mutation is
classified as a B(0) [the absence of beta chains] or B(+) [reduced amount of beta
chains] mutation. One copy of a thalassemia mutation is consistent with beta
thalassemia minor, while two copies on opposite chromosomes are consistent with beta
181
thalassemia major. On occasion, two mutations may occur on the same chromosome.
Sequence analysis cant determine the phase of two mutations (whether on the same or
opposite chromosomes). HPLC or parental studies may be necessary to evaluate
phase. Over ten complex variants with the Glu6Val variant are listed in the globin gene
server (6). One example is Hb S-Oman, with Glu6Val and Glu121Lys variants on the
same chromosome. The standard nomenclature for the nucleotide changes is HBB:c.
[20A>T;364G>A] (6). These two variants are also seen alone as in Hb S and Hb 0Arab. The combination of these two mutations on opposite chromosomes is consistent
with severe sickle cell disease, whereas if they are on the same chromosome, the
individual is a carrier of an HBB mutation.
4.5.5 DNA Sequence Traces:

Several sequencing electropherograms are presented to illustrate the application. The
first shows the common alpha globin variant Constant Spring (Hb CS). This alpha +
thalassemia variant changes the normal termination codon to an elongated 3 end of the
protein. The second shows a sequence with two beta globin mutations. Information from
family or other laboratory studies could determine if these two variants are on the same or
different chromosomes.
182
Figure 2.Alpha globin sequencing. An apparent homozygous Hb CS variant (c.427T>C,

Term>Gln) is detected (arrows). The sequence from the patient (Forward and Reverse) are
compared against a reference sequence. Differences between the reference and patient
sequences are shown in the middle panes.
183
Figure 3.Beta globin sequencing. A compound heterozygous genotype is detected. The first
mutation is Hb S (c.20A>T, Glu6Val), while the second affects a splice site resulting in a
beta(0) thalassemia mutation. (c.92+1G>A). The yellow arrow in the electropherogram
indicates the exonic region. The sequence from the patient (Forward and Reverse) are
compared against a reference sequence. Differences between the patient and reference
sequences are shown in the middle panes.
4.5.6 Conclusion
Molecular analysis confirms the familial variant in individuals who are carriers of or
affected with globin gene variants. In prenatal analysis, molecular studies are often the most
direct method to predict the status of a fetus. If molecular testing is used prenatally, the
parents should first be tested to identify the familial mutations. In addition, amniotic fluid,
amniocyte or chorionic villi cell cultures should be tested for contamination from the mother. If
the samples show maternal cell contamination, the results may not accurately reflect the fetus
genotype and a second sample should be obtained.
184
The alpha and beta loci have complex structures that lead to a variety of molecular
anomalies, such as sequence variants, and large gene rearrangements resulting in deletions
or duplications.
Because many mutations in HBA1, HBA2 and HBB are well understood, the
interpretations are typically straight forward. However, because these loci have complex
structures that lead to a variety of molecular anomalies, molecular results should be combined
with clinical, family and other laboratory findings.
185
References
1 Tan AS, Quah TC, Low PS, Chong SS. A rapid and reliable 7-deletion
multiplex polymerase chain reaction assay for -thalassemia. Blood. 2001;
98(1):250251.
2 Phylipsen M, Chaibunruang A, Vogelaar IP, Balak JR, Schaap RA, Ariyurek Y,
Fucharoen S, den Dunnen JT, Giordano PC, Bakker E, Harteveld CL. Finetiling array CGH to improve diagnostics for - and -thalassemia
rearrangements. Hum Mutat. 2012 Jan; 33(1):272-80.
3 Galanello R, Cao A. Alpha-Thalassemia. 2005 Nov 1 [Updated 2011 Jun 7].
In: Pagon RA, Bird TD, Dolan CR, et al., editors. GeneReviews [Internet].
Seattle (WA): University of Washington, Seattle; 1993-. Available from:
http://www.ncbi.nlm.nih.gov/books/NBK1435/accessed 10-04-12
4 Zhao P, Buller-Burckle AM, Peng M, Anderson A, Han ZJ, Gallivan MV.
Secondary mutation (c.94_95delAG) in a -3.7 allele associated with Hb H
disease in two unrelated African American individuals homozygous for the
-(3.7) deletion (-3.7/-3.7T). Hemoglobin. 2012; 36(1):103-7.
5 Brennan SO, Chan T, Duncan J. Novel 2 gene deletion (c.349_359 del
GAGTTCACCCC) identified in association with the -3.7 deletion.
Hemoglobin. 2012; 36(1):93-7.
6 Hardison RC, Chui DHK, Giardine B, et al. HbVar: a relational database of
human hemoglobin variants and thalassemia mutations at the globin gene
server. Hum Mutat 2002; 19: 225-33 http://globin.cse.psu.edu/accessed 1004-2012
7 Mikula M, Buller-Burckle A, Gallivan M, Sun W, Franklin CR, Strom CM.The
importance of globin deletion analysis in the evaluation of patients with
thalassemia.Int J Lab Hematol. 2011 Jun;33(3):310-7
8 Pont-Kingdon G, Gedge F, Wooderchak-Donahue W, Schrijver I, Weck KE,
Kant JA, Oglesbee D, Bayrak-Toydemir P, Lyon E; Biochemical and Molecular
Genetic Resource Committee of the College of American Pathologists.
Design and analytical validation of clinical DNA sequencing assays.Arch
Pathol Lab Med. 2012 Jan;136(1):41-6.
186
Chapter 5
Alpha and Beta Thalassemia
Herbert L. Muncie, Jr., MD
Alpha () and beta () thalassemia are hematological disorders that are the
result of a decreased or absent synthesis of a globin chain. 1 These genetic alterations
may have been the result of selective pressure from Plasmodium falciparum malaria
from which thalassemia carriers are relatively protected from invasion. 2, 3 The altered
globin chain synthesis can be asymptomatic or can cause severe hemolytic anemia and
even death.
5.1
Epidemiology
The thalassemias are prevalent in the tropical and subtropical regions of the
world and affect men and women equally. Alpha thalassemia is found more often in
persons of African or Southeast Asian descent and -thalassemia occurs more often in
persons of Mediterranean, African or Southeast Asian descent. Thalassemia trait can be
found in 5 to 30 percent of these populations. 4 An estimated 1.5% of the global
population are -thalassemia carriers but only approximately 200,000 people are alive
with -thalassemia major.5, 6
5.2
Pathophysiology
Hemoglobin has an iron-containing heme ring and four globin chains: normally
two alpha and two nonalpha. The composition of these four globin chains determines
the hemoglobin type. The predominant in utero hemoglobin, fetal hemoglobin (Hb F),
187
has two and two gamma () chains (2 / 2). Adult hemoglobin A (Hb A) has two and
two chains (2/2) and hemoglobin A2 (HbA2) has two and two delta () chains
(2/2).
The transition from -globin synthesis (Hb F) to -globin synthesis (Hb A) begins
before birth. Therefore, at birth approximately 20% to 30% of hemoglobin is Hb A and
the remainder is HbF.7 This transition continues and is usually completed from 6 to 24
months of age. At that time normal children will have mostly Hb A (>96%), small
amounts of Hb A2 (2.0 3.4%) and very small amounts of Hb F (< 1%). 8
5.3
Alpha Thalassemia
Alpha thalassemia occurs when there is reduced or absent -globin chain
synthesis with subsequent excess -globin chains.9, 10 Two genes on chromosome 16

control -globin synthesis (/). Most defects are due to deletions of one or more of
these genes. Since two genes on each chromosome 16 control the production of globin chains, there are four possible phenotypical presentations. With a single gene
deletion (-/) the result is -thalassemia silent carrier state which is asymptomatic
with normal hematological indices. With two gene deletions (-/-; --/) the result is thalassemia trait (minor) which frequently causes microcytosis without anemia. If three
genes are deleted (--/-) there will be significant amounts of hemoglobin H (Hb H)
consisting of four -globin chains (4). The result of significant amounts of Hb H is thalassemia intermedia (Hb H disease), which causes hemolytic anemia, microcytosis
and splenomegaly. While most cases of Hb H disease are deletional, non-deletional
forms do occur and are often more symptomatic. Hemoglobin Constant Spring is an 188
globin chain variant that is longer than normal and produced in only small quantities. It
therefore behaves in a similar manner to an alpha gene deletion.
11
When Hb Constant
Spring is inherited with a 2 alpha gene deletion, the condition may be referred to as Hb
H / Constant Spring. Finally if all four genes are deleted (--/--) the result will be
significant amounts of hemoglobin Barts (Hb Barts) with four gamma chains ( 4). With
increased Hb Barts and total absence of Hb F, -thalassemia major results leading to
hydrops fetalis, which is incompatible with life.
5.4
12
Beta Thalassemia
Beta thalassemia occurs when there is reduced or absent -globin chain
synthesis with subsequent excess -globin chains. 3, 6 Most often a mutation is the
genetic defect, with more than 200 reported; a deletion is quite rare. One gene on each
chromosome 11 controls the production of -globin chains (,), therefore, there are two
phenotypical presentations. If a child inherits one normal gene from one parent (/)
and a defective gene from the other parent (-/), the result is -thalassemia trait (minor)
which causes an asymptomatic mild microcytic anemia. If both genes are defective, the
result depends on the degree they are deficient in -globin chain production. If -globin
chain production is severely reduced, the person will have -thalassemia major (Cooley
anemia). Most individuals with -thalassemia are asymptomatic at birth because of the
presence of significant amounts of Hb F. As the -globin chain synthesis decreases,
infants may experience symptoms starting at six months of age. If the -globin chain
synthesis is only partially reduced, the person will have -thalassemia intermedia with
189
less severe symptoms and survival beyond 20 years of age without life-long blood
transfusions.
5.5
Diagnosis
Except for or thalassemia major, the diagnosis of thalassemia is usually
made incidentally when a patient is found to have microcytosis with or without anemia.
The most common etiologies for microcytic anemia are iron deficiency, thalassemia,
lead toxicity and sideroblastic anemia. The patients medical history, mean corpuscular
volume (MCV), red cell count and the red cell distribution width (RDW) can help exclude
many of these etiologies (Table 1). The MCV in -thalassemia trait is usually lower than
in -thalassemia trait. In Hb H disease the MCV will be as low as 64 fl. 13 Mentzer index
(MCV/red blood cell count) was proposed (which is not true in children) to predict the
likelihood of thalassemia. If the ratio is > 13, iron deficiency is the likely etiology
whereas thalassemia is associated with a value < 13. An exact ratio of 13 would be
uncertain.14
The RDW can be helpful in distinguishing thalassemia from iron deficiency and
sideroblastic anemia. With iron deficiency or sideroblastic anemia the RDW is almost
always elevated while it is elevated in approximately 50% of thalassemia trait patients. 15
Therefore, with a microcytic anemia, if the RDW is normal the diagnosis is usually
thalassemia trait. However, if the RDW is elevated additional tests to evaluate for iron
deficiency and sideroblastic anemia will be needed. 16
A serum ferritin level is the best single test to rule out iron deficiency in the absence of
inflammation.17 Serum iron, total iron binding capacity and transferrin may not be
190
needed in distinguishing iron deficiency from thalassemia. A peripheral smear or bone

marrow aspirate can rule out sideroblastic anemia. If lead toxicity is suspected, a serum
lead level will be needed. And finally a hemoglobin electrophoresis/ HPLC can evaluate
for hemoglobinopathies and may confirm the diagnosis of thalassemia.
In the past the diagnosis of -thalassemia in adults was by exclusion. If a patient
had a microcytic (MCV < 80 fl) hypochromic (MCH < 27 pg) anemia, normal iron studies
and a normal hemoglobin electrophoresis and Hb A2 it was assumed the patient had thalassemia trait (minor). Now high-performance liquid chromatography (HPLC) can
often provide an accurate diagnosis in neonates. In infants, if increased amounts of Hb
H or Hb Barts are found in cord blood or neonatal blood, the diagnosis of -thalassemia
is confirmed. Infants who are silent carriers may have a slightly increased amount of Hb
Barts (1 2%) at birth while infants with -thalassemia trait have a moderately
increased amount (5 6%).10
In adults with -thalassemia trait (minor) the HPLC or hemoglobin
electrophoresis will show reduced Hb A levels (<96%), elevated Hb A2 levels (>3.5%)
and often elevated Hb F levels (1.0 4.0%).3, 4 However, a normal amount of Hb A2
does not exclude thalassemia in some patients. Patients with Iron deficiency often have
lower Hb A2 levels and the Hb A2 quantification may need to be repeated after iron
supplement therapy.18 Genetic coinheritances may reduce Hb A2 production making it
difficult to diagnosis -thalassemia. If the Hb A2 level is below normal (<2.5%) but with a
normal Hb F level and microcytosis, the patient has -thalassemia intermedia, i.e. Hb H
disease (Table 2)19. Reviews on measuring and interpreting Hb A2 and Hb F levels are
available.8, 20
191
Beta thalassemia major is diagnosed during the infants first year of life. The
infant usually displays growth retardation, pallor, irritability and later jaundice with
abdominal swelling. Children who develop these symptoms after their first birthday will
be diagnosed with -thalassemia intermedia.
5.6
Treatment
Patients with or thalassemia trait (minor) require no treatment or regular long-
term follow-up. While these patients have a microcytic anemia they are not iron deficient
and should not be given iron supplements. If iron deficiency did develop, iron
supplements would be appropriate.4, 21
Beta thalassemia major requires treatment with blood transfusions starting as
early as six months of age. Transfusions correct the anemia, suppress erythropoiesis
and inhibit intestinal iron absorption. Transfusions are initiated either when the
hemoglobin is < 7 g/dl for more than 2 weeks (without another etiology) or if other
factors such as facial changes, poor growth, bony expansion or splenomegaly occur.
Without blood transfusions these patients would not survive into adulthood. They will
need periodic (every 2 - 4 weeks) transfusions (lifelong) to maintain their hemoglobin
higher than 9.5 g/dL.4, 22 The post-transfusion hemoglobin goal is 13 14 g/dl. Beta
thalassemia intermedia patients require transfusions only when their reduced
hemoglobin interferes with their quality of life or it effects their growth and development.
Transfusions will occasionally be needed for Hb H disease depending on the severity of
the condition.
192
Patients who receive frequent blood transfusions or who have increased

intestinal iron absorption will eventually develop iron overload since their bodies cannot
actively eliminate excess iron. To treat the iron overload, -thalassemia major patients
will require iron chelation therapy starting either around age 5, when the serum ferritin
exceeds 1000 ng/ml, or after they have had 10-20 transfusions. 23 A liver biopsy is the
gold standard for iron overload diagnosis.24 Beta thalassemia intermedia patients will
begin chelation when the serum ferritin is > 300 mcg/L. 3 Deferoxamine either
subcutaneously or intravenously has been the chelation treatment of choice although
long-term compliance is difficult.25 An oral alternative is deferasirox (Exjade).26 Iron
chelation therapy is relatively benign although it is time consuming and expensive
(Table 3).
The only curative therapy for patients with -thalassemia major is a bone marrow
transplant. In low-risk patients with no hepatomegaly, no portal fibrosis on liver biopsy
and not receiving regular chelation therapy, hematopoietic stem cell transplantation
usually has excellent results.4
5.7
Complications
Patients with or thalassemia trait (minor) have no complications. Patients with
Hb H disease, -thalassemia major or -thalassemia intermedia have hemolysis, growth

retardation and skeletal abnormalities as a consequence of the over stimulation of the
bone marrow and ineffective erythropoesis.21, 27 Infants with significant amounts of Hb
Barts usually die in utero or shortly after birth due to autoimmune hydrops fetalis.
193
Because of the need for multiple blood transfusions in -thalassemia major or in some
cases of Hb H disease and the increased intestinal iron absorption with -thalassemia
intermedia, patients develop iron overload which damages visceral organs (liver, spleen,
endocrine organs) and the heart which is the primary cause of early death. 28
Splenomegaly invariably develops in symptomatic thalassemia and can worsen the
anemia. The risk of hepatocellular carcinoma is increased due to iron overload hepatic
damage, longer survival and viral infection with hepatitis B and/or C. 29 Gallstones are
more prevalent with -thalassemia intermedia than with -thalassemia major.
Beta thalassemia major and intermedia cause a hypercoaguable state. 30 This effect
increases the risk of thromboembolic events especially after splenectomy.31
Osteoporosis was found in 51% of patients over age 12 with -thalassemia major.32
5.8
Other Treatment Issues
5.8.1
Hypersplenism
Splenectomy is required for patients whose splenomegaly causes a marked
increase in their need for blood transfusions, i.e. the annual red cell requirement
exceeds 180 200 ml/kg.6 Because of the importance of the spleen in clearing bacteria
and preventing sepsis, the surgery is not done until the patient is at least 4 years old.
One month prior to the surgery the child should be given the pneumococcal
polysaccharide vaccine. They should also receive the pneumococcal conjugate vaccine
series if they had not received it during infancy. For the first two years after the surgery
patients should take penicillin 250 mg twice a day. For children the antibiotic prophylaxis
194
continues until age 16.33 Gallbladder removal should be considered if gallstones are
present34 at the time of splenectomy.
5.8.2 Endocrinopathies
While growth retardation can occur with thalassemia, growth hormone therapy
has limited effectiveness and is not recommended. If hypogonadism develops,
hormonal therapy is effective.35 Bone mineral density has been increased with
alendronate, pamidronate and zolendronate; however, studies evaluating fracture
reduction are needed.36
5.8.3 Pregnancy
Couples from high risk ethnic groups should be encouraged to seek
preconception genetic counseling.37 Individuals with a low MCV (<80fl) and MCH (<27
pg) could be assessed with hemoglobin electrophoresis / HPLC. 12 An efficient way to
identify mutations is to study their parents hematology and screen them for single
mutations.38
For couples, if both partners have -thalassemia trait, their child will have a one
in four chance of having -thalassemia major and a two in four chance of -thalassemia
trait.(Figure 1) With four genes controlling the expression of -globin chains, the
inheritance pattern is more complex. If two genes are defective, the phenotype is
influenced by whether the defective genes are on the same chromosome (cis) or
different chromosomes (trans), e.g. if one parent is an -thalassemia silent carrier (-,
) and one parent has -thalassemia trait [(cis),(--,)], they have a one in four chance
195
their child will have Hb H disease. Whereas, if the -thalassemia trait parents defect is
trans (-, -), their children have no risk of Hb H disease (Figure 2).
Once pregnancy occurs, patients should be counseled regarding prenatal
diagnostic testing options. An amniocentesis at approximately 15 weeks gestation or a
chorionic villus sample (CVS) obtained at 10 11 weeks gestation can detect point
mutations or deletions utilizing polymerase chain reaction (PCR) testing. Other
diagnostic options include DNA analysis of fetal cells obtained by amniocentesis and in
the future analysis of maternal blood fetal cells. 39, 40 If Hb Barts is detected, the mother
has an increased risk of pre-eclampsia and postpartum hemorrhage.
For couples using in vitro fertilization, preimplantation genetic testing is available. 41
5.8.4 Cardiac
When iron overload occurs, cardiac infiltration and death are significant
concerns. Serum ferritin levels have been used to predict cardiac complications with
improved survival if levels are kept below 2,500 ng per ml (2500 mcg per L). 42 Ferritin
levels are unreliable when significant liver disease develops. 43
5.8.5 Hypercoagulopathy
While the risk of thromboembolic events in patients with -thalassemia major or
intermedia is increased, no trials have evaluated prevention of these complications with
anticoagulants. A consensus recommendation for patients with a thrombosis history is
prophylactic treatment with low molecular weight heparin especially before surgery and
during pregnancy. Because estrogen containing contraceptives may increase the risk of
196
thrombosis, an alternative form of contraception should be recommended for these

women during their reproductive years.
5.8.6
Psychosocial
The impact on a patient and their family of a chronic disease such as -
thalassemia major that requires lifelong treatment is significant. Providing education

regarding the inheritance patterns, the prenatal diagnostics options and the need for
psychological support may help patients better manage their disease. However, based
on the available evidence no specific therapy or combination of therapies can be
recommended.44
5.8.7 Vitamin Deficiencies

With the increase in erythropoesis some patients may develop folic acid
deficiency. For these patients a 1 mg folic acid supplement daily is recommended. 34
However, patients receiving frequent transfusions rarely have this problem. While
oxidative stress may contribute to the complications, the use of antioxidants has not
improved the anemia nor reduced the morbidity or mortality of thalassemia. 34 If a
transfusion dependent patient has proven vitamin C deficiency, supplements are
recommended.
5.8.8 Prognosis
197
Beta thalassemia major patients live an average of 17 years and usually die by
age 30. With regular blood transfusions and compliance with chelation therapy, their life
can extend beyond age 40.6 Their deaths are commonly caused by cardiac
complications of iron overload.28 Thalassemia trait patients have a normal life
expectancy.
Sources of Additional Information:

Cooley's Anemia Foundation http://www.cooleysanemia.org or
http://www.thalassemia.org
Thalassemia International Federation www.thalassaemia.org.cy
198
Table 1 Hematologic Indices for Iron deficiency and thalassemias

Indices
MCV (abnormal <80
Iron Deficiency
-thalassemia
-thalassemia
-thalassemia
Low
Low
trait
Low
major
Low
High
Normal
Normal,
Normal,
occasionally
occasionally high
fl in adults; < 70 fl age

6 months 6 years; <
75 fl age 7 12 years)
RDW (Adult normal 11.5 14.5%)
Low
Normal
high
Normal
Normal
> 13
< 13
< 13
< 13
count)
Hb electrophoresis
Normal (may
Adults : normal
Reduced HbA ,
Reduced HbA,
(Adult normals
have reduced
Newborns:
increased
increased HbA2,
HbA - > 96%
HbA2 before
cord blood may
HbA2, and
and increased
HbA2 2.5 -3.5%
iron therapy is
have HbH or
increased HbF
HbF
given)
Hb Bart
Ferritin (adult normal

male 20 250
ng/ml; female 10
120 ng/ml)
Mentzer Index
Children (MCV/RBC
HbF - < 1%)
199
Table 2 Hb A2 levels with iron deficiency and thalassemia 9

Diagnosis
HbA2 level
2.5 3.5 %
1.6 3.2 %
2.0 3.2 %
1 2.4 %
> 4.0 %
> 4.0 %
Normal
Iron deficiency
-thalassemia silent carrier or trait (minor)
HbH disease
-thalassemia trait (minor)
-thalassemia major
200
Table 3 Chelation Therapy Treatment Options

Therapeutic
Route of
Dosage
agent
Desferoxamine
administration
Subcutaneous
Adults - 30 50 mg/kg
Deferasirox
Deferiprone
infusion over 8 - 12
Children - 20 40
hours
Oral
Oral
mg/kg
20 - 30mg/kg/day
75 100 mg/kg/day
(only available
in the US
through FDA
Treatment Use
Program)
201
Frequency of therapy
5 7 days/week
Once a day
3 times/day
Figure 1 Beta thalassemia trait genetics
X
Mother
(-,)
-thalassemia trait
Children:
(-,-)
( ,)
-thalassemia major
Father
(-,)
-thalassemia trait
(-,)
-thalassemia trait
(-,)
-thalassemia trait
Normal
or intermedia
Note: Shaded symbols indicate an abnormal -globin gene on chromosome 11.
202
203
204
References
1. Muncie HL Jr, Campbell J. Alpha and beta thalassemia. Am Fam Physician 2009;
80 (4): 339-344.
2. Mantikou E, Arkesteijn SG, Beckhoven JM, Kerkhoffs JL, Harteveld CL, Giordano PC.
A brief review of newborn screening methods for hemoglobinopathies and preliminary
results selecting beta thalassemia carriers at birth by quantitative estimation of the
HbA fraction. Clinical Biochemistry 2009; 42(18): 1780-1785.
3. Cao A, Galanello R. Beta-thalassemia. Genetics in Medicine 2010; 12(2): 61-76.
4. Rund D, Rachmilewitz E. Beta-thalassemia. N Engl J Med. 2005; 353(11): 1135-1144.
5. Thalassemia International Federation. Thalassemia. Available at:
http://www.thalassaemia.org.cy/index.html. Accessed 04/10, 2011.
6. Galanello R, Origa R. Beta-thalassemia. Orphanet J Rare Dis 2010; 5:11.
7. Richardson M. Microcytic anemia. Pediatr Rev 2007; 28 (1): 5-14.
8. Mosca A, Paleari R, Ivaldi G, Galanello R, Giordano PC. The role of haemoglobin
A(2) testing in the diagnosis of thalassaemias and related haemoglobinopathies. J
Clin Pathol 2009; 62:13-17.
9. Harteveld CL, Higgs DR. Alpha-thalassaemia. Orphanet J Rare Dis 2010; 5:13.
10. Galanello R, Cao A. Alpha-thalassemia. Genetics in Medicine 2011; 13 (2): 83-88.
11. Chen FE, Ooi C, Ha SY, et al. Genetic and clinical features of hemoglobin H disease
in Chinese patients. N Engl J Med. 2000; 343(8): 544-550.
12. Leung TN, Lau TK, Chung TKH. Thalassaemia screening in pregnany. Curr Opin
Obstet Gynecol 2005; 17 (2): 129-134.
13. Origa R, Sollaino MC, Giagu N, et al. Clinical and molecular analysis of
haemoglobin H disease in Sardinia: Haematological, obstetric and cardiac aspects
in patients with different genotypes. Br J Haematol 2007; 136(2): 326-332.
14. Mentzer WC,Jr. Differentiation of iron deficiency from thalassaemia trait. Lancet
1973; 1(7808): 882.
15. Flynn MM, Reppun TS, Bhagavan NV. Limitations of red blood cell distribution width
(RDW) in evaluation of microcytosis. Am J Clin Pathol 1986; 85(4): 445-449.
16. Marsh WL Jr, Bishop JW, Darcy TP. Evaluation of red cell volume distribution width
(RDW). Hematol Pathol 1987; 1(2): 117-123.
17. Guyatt GH, Oxman AD, Ali M, Willan A, McIlroy W, Patterson C. Laboratory
diagnosis of iron-deficiency anemia: An overview. Journal of General Internal
Medicine 1992; 7(2) : 145-153.
18. Kattamis C, Lagos P, Metaxotou-Mavromati A, Matsaniotis N. Serum iron and
unsaturated iron-binding capacity in the -thalassaemia trait: their relation to the
levels of haemoglobins A, A 2 , and F. J Med Genet 1972; 9(2): 154-159.
19. Van Delft P, Lenters E, Bakker-Verweij M, et al. Evaluating five dedicated automatic
devices for haemoglobinopathy diagnostics in multi-ethnic populations. Int J Lab
Hematol 2009; 31(5): 484-495.
20. Mosca A, Paleari R, Leone D, Ivaldi G. The relevance of hemoglobin F
measurement in the diagnosis of thalassemias and related hemoglobinopathies.
Clin Biochem 2009; 42(18): 1797-1801.
21. Olivieri NF. The beta-thalassemias. N Engl J Med 1999; 341(2): 99-109.
205
22. Cazzola M, Borgna-Pignatti C, Locatelli F, Ponchio L, Beguin Y, De Stefano P. A

moderate transfusion regimen may reduce iron loading in beta-thalassemia major
without producing excessive expansion of erythropoiesis. Transfusion 1997; 37(2):
135-140.
23. Roberts DJ, Brunskill SJ, Doree C, Williams S, Howard J, Hyde CJ. Oral deferiprone
for iron chelation in people with thalassaemia. Cochrane Database of Systematic
Reviews 2007: 3: CD004839.
24. Angelucci E, Brittenham GM, McLaren CE, et al. Hepatic iron concentration and
total body iron stores in thalassemia major. N Engl J Med 2000; 343(5): 327-331.
25. Delea TE, Edelsberg J, Sofrygin O, et al. Consequences and costs of
noncompliance with iron chelation therapy in patients with transfusion-dependent
thalassemia: a literature review. Transfusion 2007; 47(10): 1919-1929.
26. Deferasirox (exjade): A new iron chelator. Drugs Ther. Med Lett 2006; 48(1233): 3536.
27. Parano E, Pavone V, Di Gregorio F, Pavone P, Trifiletti RR. Extraordinary intrathecal
bone reaction in beta-thalassaemia intermedia. Lancet 1999; 354(9182): 922.
28. Modell B, Khan M, Darlison M. Survival in beta-thalassaemia major in the UK: Data
from the UK Thalassaemia Register. Lancet 2000; 355(9220): 2051-2052.
29. Borgna-Pignatti C, Vergine G, Lombardo T, et al. Hepatocellular carcinoma in the
thalassaemia syndromes. Br J Haematol 2004; 124(1): 114-117.
30. Eldor A, Rachmilewitz EA. The hypercoagulable state in thalassemia. Blood 2002;
99(1): 36-43.
31. Tso SC, Chan TK, Todd D. Venous thrombosis in haemoglobin H disease after
splenectomy. Aust N Z J Med 1982; 126): 635-638.
32. Jensen CE, Tuck SM, Agnew JE, et al. High prevalence of low bone mass in
thalassaemia major. Br J Haematol 1998; 103(4): 911-915.
33. Davies JM. Barnes R. Milligan D. British Committee for Standards in Haematology.
Working Party of the Haematology/Oncology Task Force. Update of guidelines for
the prevention and treatment of infection in patients with an absent or dysfunctional
spleen. Clin Med 2002; 2(5): 440-443.
34. Borgna-Pignatti C. Modern treatment of thalassaemia intermedia. Br J Haematol
2007; 138(3): 291-304.
35. De Sanctis V. Growth and puberty and its management in thalassaemia. Horm Res.
2002; 58 Suppl 1: 72-79.
36. Gaudio A, Morabito N, Xourafa A, et al. Bisphosphonates in the treatment of
thalassemia-associated osteoporosis. J Endocrinol Invest 2008; 31 (2): 181-184.
37. ACOG Practice Bulletin No. 78: hemoglobinopathies in pregnancy. ACOG
Committee on Obstetrics. Obstetrics & Gynecology 2007; 109(1): 229-237.
38. Old JM. Screening and genetic diagnosis of haemoglobin disorders. Blood Rev.
2003;17(1): 43-53.
39. Li Y, Di Naro E, Vitucci A, Zimmermann B, Holzgreve W, Hahn S. Detection of
paternally inherited fetal point mutations for beta-thalassemia using sizefractionated cell-free DNA in maternal plasma. JAMA 2005; 293(7): 843-849.
40. Papasavva T, Kalakoutis G, Kalikas I, et al. Noninvasive prenatal diagnostic assay
for the detection of beta-thalassemia. Ann N Y Acad Sci 2006; 1075: 148-153.
206
41. Braude P, Pickering S, Flinter F, Ogilvie CM. Preimplantation genetic diagnosis.

[erratum appears in nat rev genet. 2003 feb;4(2):157.]. Nature Reviews Genetics.
2002;3:941-953.
42. Hoffbrand AV, Cohen A, Hershko C. Role of deferiprone in chelation therapy for
transfusional iron overload. Blood 2003;102(1):17-24.
43. Brittenham GM, Cohen AR, McLaren CE, et al. Hepatic iron stores and plasma
ferritin concentration in patients with sickle cell anemia and thalassemia major. Am
J Hematol 1993; 42(1): 81-85.
44. Anie KA, Massaglia P. Psychological therapies for thalassaemia. Cochrane
Database of Systematic Reviews 2001; 3: CD002890.
207
Chapter 6
Neonatal Screening for Hemoglobinopathies
Zia Uddin, PhD
With the technical support of Patrick Hopkins, Joseph Quashnock,
Aigars Brants, Christine Moore, DAndra Morin, Rachel Lee, Mahin Azimi,
and Bonifacio Dy
6.1 Introduction
A gratifying achievement of my professional career was the organization of an
interdisciplinary conference on Perinatal Care and Neonatal Screening in 1978 at
South Macomb Hospital (now a part of St. John Providence Health System), Warren,
Michigan.
CLINICAL CHEMISTRY, VOL 24, No. 7, 1978

Seven specialists eminent in the respective fields recently presented
their research work to a group of obstetricians, gynecologists,
pediatricians, pathologists, and clinical chemists from Michigan and
Windsor (Canada) at a recent interdisciplinary conference in Warren,
Michigan, sponsored by Detroit-Macomb Hospitals Association.
Speakers and their topics were: Dr. Joseph Bieniarz, Amniocentesis in
Perspective: Diagnostic Value of Ultrasonography and Protocol for
Monitoring High Risk Pregnancy. Dr. Keith H. Marks, Elective Delivery
of the Term Fetus: An Obstetrical Hazard. Dr. John. L. Kitzmiller,
Management and Outcome of Pregnancy in Diabetes Mellitus.
Dr. Norman F. Gant, Supine Hypertension Test and the Clinical
Management of Pregnancy-Induced Hypertension. Dr. Thomas P.
Foley, Neonatal Screening for Congenital Hypothyroidism and
Clinical Treatment. Dr. Samuel Meites, Clinical Chemistry Laboratory
in Neonatology. Mrs. Ann Bennett, Public Health and Neonatal
Screening.
For additional information write Zia Uddin, Ph.D., Perinatal Care &
Neonatal Screening Conference, South Macomb Hospital, 11800
Twelve Mile Road, Warren, Mich. 48093.
208
After this conference I started neonatal T4 screening for the four major hospitals
in South Eastern Michigan. Subsequently, with the vision of the then Governor of
Michigan (Honorable Mr. William Milliken), a law was passed for the establishment of a
state of the art laboratory in Lansing, Michigan for the screening of inborn errors of
metabolism and hemoglobinopathies. By statute, each state in the USA performs
newborn screening (NBS), however, the number of tests/neonate and the
methodologies utilized vary from state to state. In 2006, Honorable Senators Edward M.
Kennedy, Barack H. Obama, and Hillary R. Clinton proposed a uniform standard and
protocol of NBS in USA. An integral part of this proposal was to extend this program to
resource-poor countries under the auspices of the United States Agency for
International Development. Unfortunately, due to political events in USA and the death
of Senator Edward M. Kennedy, nothing materialized in this direction.
NBS in America always includes analysis for hemoglobinopathies as described
by the Health Resources and Services Administration (HRSA) Maternal and Child
1
Health Program of the U. S. Department of Health and Human Services .

During the last twenty years, I had the opportunity to introduce NBS in a few
developing countries (Kuwait, Iraq, Bahrain, Egypt, and Saudi Arabia) with the
cooperation of Quest Diagnostics, USA and Laboratory Corporation of America, USA.
PerkinElmer Genetics is the most popular private laboratory in USA that provides NBS
services worldwide. Besides the popularity of PerkinElmer Genetics, several countries
in Europe and Asia have instituted liquid chromatography-mass spectrometry (Applied
209
Bioscience), high performance liquid chromatography (Bio-Rad and Trinity Biotech), and
isoelectric focusing instruments (PerkinElmers Resolve, and Helenas SPIFE) for NBS.
6.2 Methodologies
Isoelectric focusing (IEF) and high performance liquid chromatography (HPLC) are
the two most commonly used screening methods for hemoglobinopathies in the
neonate. Recently Sebia (Evry, France) has introduced the capillary zone
electrophoresis (CZE) method (Neonate Hb Fast System) for the newborn screening of
hemoglobinopathies. In order to resolve abnormal results of NBS,the blood of the
biological parents (and sometimes of the siblings) are analyzed by IEF, HPLC, agarose
gel electrophoresis (pH 8.6 and 6.2) and capillary zone electrophoresis to ascertain
genetic inheritance of the abnormality in the neonate. Finally, the diagnosis is confirmed
by means of DNA mutation studies,but for Hb S the final diagnosis can be confirmed by
a Sickle test, and complete blood count (CBC) with manual differential. Further
confirmation, if desired, for Hb S in a newborn can be achieved by testing the blood of
the parents. Electrospray ionization-mass spectrometry (Chapter 3.4), PCR and Sanger
sequencing (Chapter 3.5) are also used to confirm DNA mutation. A table of screening
2
methods by individual states in America can be obtained via internet .
The principle of the assay by IEF (Chapter 2.7) and HPLC (Chapter 2.8) for the
screening of hemoglobinopathies in the neonate and the adult are similar. However
certain adjustments are required for the neonate specimen (dried blood spot on filter
paper) handling and processing. The Resolve kit and instrumentation of the
210
PerkinElmer and SPIFE 2000 or 3000 instrument of the Helena Laboratories are the
most commonly used methods for the screening of hemoglobinopathies by IEF. HPLC is
most commonly performed employing ion-exchange chromatography by the NBS
instruments manufactured by BioRad, USA., or Trinity Biotechs Ultra Resolution
System.
The principle of the CZE for the screening of hemoglobinopathies (Chapter 2.6)
in the adults and neonates is identical, except that modifications in the automated
instrument are necessary for the handling of a dried blood sample on filter paper from
the neonate (Figure 1). Sebia is the main supplier of capillary zone electrophoresis
instrument and reagents for the NBS of hemoglobinopathies.
Figure 1. Automated hemolysis, sample dilution and analysis instrument (Sebia)

The hemoglobin variants in the neonate by the Sebia Capillarys Neonat
211
Fast System separated into windows or zones (N 1-N13) as illustrated in section

5.4. This method has been evaluated
3-5
and found satisfactory, however as with
IEF and HPLC methodologies for NBS, results have to be considered provisional and
confirmatory procedures are always required because many rare hemoglobin variants
migrate or elute on the same position of the common one and because different
homozygous or hemizygous genotypes look identical with these methods.
6.3 Laboratory Reports Format and Interpretation

All the NBS laboratories in America have an elaborate management system for the
procurement, handling and processing of dried blood on filter paper (Guthrie card).
Additional facilities are provided for confirmatory testing and the counseling of the
parents. Advisory consultative services and treatment modalities are also provided by
the medical staff of the NBS facilities in America. The details of all these services are
available online as well as in a printed version.
Normal patterns and abbreviations:
A normal newborn typically displays about 70% Hb F and 20% Hb A and perhaps
traces of Hb A2. The abbreviation used to indicate normal and abnormal patterns by the
laboratories are:
FA
Fetal hemoglobin is greater than adult hemoglobin. This is

observed in a newborn < 3 weeks of age.
212
AF
Adult hemoglobin is greater than fetal hemoglobin. This is usually

observed in a newborn > 3 weeks of age unless transfused within the last
eight weeks and anemia is not suspected.
Abnormal patterns and abbreviations:

Fa
Lower Hb A levels then expected for the gestational age usually

indicate a -thalassemia carrier that could have been born from a couple
at risk (50% chance). Reporting these carriers will allow
the couple to consider prospective primary prevention.
BFA? The presence of Hb Barts in a further normal pattern will indicate

+
-thalassemia (2-5% heterozygous), (5-10% + homozygous or
0 heterozygous), (>10% could indicate Hb H disease). Hb Barts
in absence of Hb F indicates hydrops foetalis.
FF
Absence of Hb A may indicate a delayed appearance of Hb A

(early prematurity), hereditary persistence of fetal hemoglobin
(HPFH), or -thalassemia major.
FAS
Hb F + Hb A and Hb S indicates heterozygous Hb AS trait (sickle

cell trait).
FSS
Patterns with only Hb F and Hb S may indicate homozygous Hb S

or Hb S--thalassemia (both resulting in sickle cell disease).
FAC
Hb F, Hb A and Hb C indicates heterozygous Hb AC trait.
FCC Only Hb F and Hb C indicates either homozygous Hb C or

Hb C--thalassemia.
FSC
Hb F + Hb S and Hb C indicates compound heterozygous Hb S and

Hb C (sickle cell disease).
FAE
Hb F, Hb A, and Hb E indicates heterozygous Hb AE trait.
FEE
Hb F and Hb E only may indicate mild Hb E homozygosity or

severe Hb E--thalassemia.
FSE
Hb F, Hb S and Hb E indicates Hb S/E compound heterozygosity

(sickle cell disease).
FAD
Hb F, Hb A and Hb DPunjab indicates heterozygous Hb AD trait

(an asymptomatic condition).
213
FDD Only hb F and hb D indicates Hb D homozygous or Hb D-thalassemia (both mild conditions).

FDS
Hb F, Hb S and Hb D indicates compound heterozygous Hb SD

(sickle cell disease).
FAO Hb F, Hb A, and Hb OArab indicates heterozygous Hb A-OArab trait

(an asymptomatic condition).
FSS-Barts Homozygous Hb S or Hb S--thalassemia (severe sickle cell
disease condition).
Earlier, laboratories used the symbol X to designate a hemoglobin variant
which could not be identified by the NBS laboratory and further testing was suggested.
This practice of using X for a hemoglobin variant was abandoned by some
laboratories and now the symbol V is also used for this purpose.
It is emphasized here that some abnormal hemoglobin designated by V is
often reported in newborn screening, because the screening laboratories do not have all
the diagnostic methods available. In these situations the physician is advised to have
definitive diagnostic testing done at a specialized laboratory, e.g.
Georgia Health Sciences University Sickle Cell Center, Augusta, Georgia
(http://www.georgiahealth.edu/centers/sicklecell).
The interpretation of NBS in a premature neonate is subject to a possibility of false
6
positive results , therefore the blood is retested when the adjusted gestational age is 40
weeks and two months after transfusion if executed.
6.4 Examples of Neonatal Screening
214
In this section, selected cases are presented to illustrate the laboratory data
obtained from NBS from commonly used methods.
6.4.1 Capillary Zone Electrophoresis

Capillary zone electrophoresis (CZE) scans of the most commonly expected
hemoglobin variants are presented in Figures 2-9. These scans were obtained after
analyzing the dried blood spot on the Sebia Capillarys Neonat Fast System. The blood
sample was collected by capillary puncture between 2-5 days after birth from neonates
of gestational age > 38 weeks. We have also provided the percentage of major
hemoglobin bands.
215
216
217
218
219
6.4.2 Isoelectric Focusing

Isoelectric focusing is the most widely used NBS method for hemoglobinopathies
in America. Here again, confirmatory testing is desired for accurate diagnosis of any
abnormal hemoglobin. One method of confirmation, if feasible, is to include the testing
of the biological parents. In Figure 10, we have presented the IEF results (Hb SC
disease) of a newborn, along with that of the father and mother of the newborn. The
father has Hb AS trait, and the mother has Hb AC trait, therefore there is a 25% chance
of the genetic inheritance of Hb SC disease in the newborn.
Figure 10. IEF results of newborn (Hb SC disease), father (Hb AS trait), and mother (Hb
AC trait).
220
6.4.3 Isoelectric Focusing and High Performance Liquid Chromatography

Generally speaking, it is a common practice in some laboratories in USA to
further evaluate the IEF results for abnormal cases by HPLC and vice versa. In Figures
11-17, we have presented the scans for both of these methods; for the normal and a
few abnormal variants in a newborn. However, absolute certainty is never achieved by
these two methods and DNA sequencing is the method of choice to confirm the variant
and eventually the halotype to define the prognosis, tailor the best treatment and also to
allow primary prevention in case of another pregnancy (Section 5.4.4).
The IEF figures provided in this section were obtained using the RESOLVE
neonatal hemoglobin test kit and testing equipment (PerkinElmer). In all the IEF Figures
(11-17), we have presented at the top the IEF of the traditional laboratory control
AFSC. Details about this procedure can be obtained from:
http://www.perkinelmer.com/CMSResources/Images/44-72976FLY_Hemoglob_12449784.pdf
The high performance liquid chromatography scans provided in this section were
obtained using the Trinity Biotechs Ultra Resolution System. Details about this
procedure and instrumentation can be obtained from:
http://www.trinitybiotech.com/HbA1c_HB/Instruments/Pages/Ultra2Variant.aspx
221
Isoelectric
focusing
High performance liquid chromatography
Figure 11 Normal: FA
IEF of normal phenotype displays three prominent bands, Hb F, Hb A, and acetylated
Hb F. Hb F is the prominent band in newborns. Hb A (the middle band) in the IEF
pattern often appeared weaker in premature babies compared to full term babies. In all
the HPLC separations the prominent Hb peaks (i.e. with significant concentration)
eluted at specific retention times.
Isoelectric
focusing
Isoelectric
focusing
Figure 13 Hb AE trait FAE
222
Figure
12 Hb
AS trait
FAS
Isoelectric
focusing
Figure 14 Hb AC trait FAC

Isoelectric
focusing
Figure 15 Hb SC disease FSC

Isoelectric
focusing
Figure 16 Hb S disease FS
223
Isoelectric
focusing
Figure 17 Hb Barts FABarts

Note: A reviewer of this chapter pointed out the possibility that the fastest band on IEF is
Hb H and the second is Hb Barts, as Hb H affected babies also have fast bands on IEF.
Another reviewer suggested that Hb H affected babies have three fast bands on IEF,
with a Barts result on HPLC exceeds 25%, and usually greater than 30%. This case
displayed < 10% Barts from HPLC and displayed the typical two-banded Barts that is
observed on one and two gene deletion alpha thalassemia carriers.
224
6.4.4. Isoelectric focusing, High Performance Liquid Chromatography and DNA

Studies
In Figure 18, we present the typical IEF result of a full-term newborn with only Hb
F and Hb S, and no Hb A. This pattern suggests in order of probability the following
diagnostic options:
a) Hb S homozygous (both genes code for Hb S, genotype associated with
severe SCD).
b) Hemizygous Hb S / -thalassemia (one gene codes for Hb S and the other is
not active, genotype associated with severe SCD).
c) Hemizygous Hb S / deletional HPFH (one gene codes for Hb S and the other
is deleted, associated with mild SCD conditions).
d) Double heterozygous Hb S / Hb Lepore (a combination associated with SCD)
e) Double heterozygous Hb S-like / -thalassemia (a combination not associated
with SCD).
f) Double heterozygous Hb S / Hb S-like (the last migrating like Hb S but causing
no SCD).
g) Homozygous for the same Hb S-like variant heterozygous for two Hb S-like
variants
This means that even the simple SCD newborn pattern comes with different
diagnostic options that have to be sorted out at the DNA level.
225
Figure 18. IEF of newborn
Another example of a complex interpretation is shown in Figure 19. The HPLC

pattern of the newborn shows 54% of Hb F with three additional and significant bands at
a retention time known for a) Hb A at 0.87 minutes (1.3%), b) Hb E/A 2 at 1.04 minutes
(8.3%) and c) Hb S at 1.2 minutes (6.3%). It is emphasized that a newborn cannot be
assigned Hb A2 and Hb E was not detected by IEF. Therefore, the band in HPLC at 1.04
minutes is due to a hemoglobin variant to be defined at the molecular level.
226
Figure 19. HPLC of newborn
DNA sequencing revealed Hb S heterozygous mutation at codon 6 and a second

point mutation at codon 43 of the -globin gene leading to a GluAla amino acid
substitution known as Hb G-Galveston.Therefore the newborn was diagnosed at the
molecular level as compound heterozygous Hb S / Hb G-Galveston, a combination
which is not associated with SCD. It is noteworthy to mention that like Hb G-Galveston
227
elutes on HPLC at the position of the common Hb E, many other variants elute at the
position of Hb S, Hb D, or Hb C and therefore molecular confirmation is always needed.
References (Section 6.1 6.4)

1.
2.
3.
4.
5.
6.
Lin K, Barton M. Screening for Hemoglobinopathies in Newborns.

Reaffirmation Update for the U.S. Preventive Services Task Force. Evidence
Synthesis No. 52. Rockville, MD: Agency for Healthcare Research and
Quality, August 2007. AHRQ Publication No. 07-05104-EF-1. Available at
http://www.ahrq.gov/clinic/serfiles.htm#sicklecell
http://nnsis.uthscsa.edu/xreports.aspx?xreportID=47&formid=104&fclr=1
Giordano PG. Newborn screening for hemoglobinopathies using
capillary electrophoresis. Methods Mol Biol 2013; 919: 131-45.
Renom G, Mereau C, Maboudov P, Perini JM. Potential of the Sebia
Capillarys neonat fast automated system for neonatal screening of
sickle cell disease. Clin Chem Lab Med 2009; 47(11): 1423-32.
Mantikou E, Harteveld CL, Giordano PC. Newborn screening for
hemoglobinopathies using capillary electrophoresis technology:
Testing the Capillarys Neonate Fast Hb device. Clin Biochem 2010;
43: 1345-1350.
Hustace T, Fleisher JM, Varela AMS, Podda A, Alvarez O. Increased
Prevalence of False Positive Hemoglobinopathy Newborn Screening in
Premature Infants. Pediatric Blood Cancer 2011; 57: 1039-1043.
References related to neonatal screening experience for hemoglobinopathies:
Bouva MJ, Mohrmann K, Brinkman Henri BJM, Kemper-Proper EA, Elvers

B, Loeber JG, Verheul Francesco EAM, Giordano PC. Implementing
Neonatal screening for haemoglobinopathies in the Netherlands. J Med
Screen 2010; 17: 58-65
Michlitsch J, Azimi M, Hoppe C, Walters MC, Lubin B, Lorey F, Vichinsky

E. Newborn Screening for Hemoglobinopathies in California. Pediatr Blood
Cancer 2009; 52: 486-490.
Kafando E, Nacoulma E, Quattara Y, Ayeroue J, Cotton F, Sawadogo,

Gulbis B. Neonatal haemoglobinopathy screening in Burkina Faso. J Clin
Pathol 2009; 62: 39-41.
Streetly A, Latinovic R, Hall K, Henthorn J. Implementation of universal

228
newborn bloodspot screening for sickle cell disease and other clinically
significant haemoglobinopathies in England: screening results for 2005-7. J Clin
Pathol 2009; 62: 26-30.
Gulbis B, Cotton F, Ferster A, Ketelslegers O, Dresse MF, Ronge-Collard

E, Minon JM, Le PQ, Vertongen F. Neonatal haemoglobinopathy screening in
Belgium. J Clin Pathol 2009; 62: 49-52.
Bardakdjian-Michau J, Bahuau M, Hurtrel D, Godart C, Riou J, Mathis M,

Goossens M. Neonatal screening for sickle cell disease in France. J Clin
Pathol 2009; 62: 31-33.
Adorno EV, Couto FD, de Moura Neto JP, Menezes JF, Rego M, dos Reis
MG, Goncalves MS. Hemoglobinopathies in newborns from Salvador, Bahia,
Northeast Brazil. Cad. Saude Publica, Ruio de Janeiro 2005; 21(1): 292-298.
6.5 Genetic Counseling & Screening:

After a careful review of the literature on the worldwide prevalence of
thalassemia and hemoglobinopathies, it is my estimate that by 2050 more than
500 million individuals will be affected by these genetic disorders. During the past two
decades, attempts have been made to provide premarital and prenatal genetic
counseling and screening in both the endemic and non-endemic (in view of migration)
countries, however achieving a thalassemia-and hemoglobinopathy free generation
seems unlikely to me. Although treatment modalities for sickle cell anemia have been
investigated since 1967, including the latest promising treatment with antidepressants in
these individuals by increasing the concentration of Hb F, permanent cure is illusive.
Impediments for the worldwide implementation of a prevention and control
program are: a) financial resources, b) technical personnel, c) religious and social
considerations, d) education of the entire population about the benefits of this program,
and e) poor and resource-lacking population problem.
229
Indeed it is very promising that various religious organizations (Muslims and

Jews) have authorized screening for genetic diseases after taking into consideration the
halachic concerns.
Country and state specific genetic counseling and screening programs (Thailand,
Cyprus, etc.) are steps in the right direction, and let us hope that these initiatives
blossom into an elaborate undertaking.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Jewish Womens Health. http://www.jewishwomenshealth.org/article.php?

article=32
Strauss BS. Genetic Counseling for Thalassemia in the Islamic Republic of
Iran. Perspectives in Biology and Medicine 2009; 52(3): 364-376
Larijani B, Anaraki FZ. Islamic principles and decision making in bioethics.
Nature Genetics 2008; 40(2): 123.
Norton ME. Genetic screening and counseling. Current Opinion in Obstetrics
and Gynecology 2009, 20: 157-163.
Zlotogora J. Population programs for the detection of couples at risk for
severe monogenic genetic diseases. Hum Genet 2009; 126: 247-253.
Al-Ama JY. Attitudes towards mandatory national premarital screening for
hereditary hemolytic disorders. Health Policy 2010; 97: 32-37.
Theodoridou S, Alemayehou M, Prappas N, Karakasidou O, Aletra V, Plata E,
Tsaftaridis P, Karababa P, Boussiou M, Sinopoulou K, Hatzi A, Voskaridou E,
Loutradi A, Manitsa A. Carrier Screening and Prenatal Diagnosis of
Hemoglobinopathies. A Study of Indigenous and Immigrant Couples in
Northern Greece, over the last 5 years. Hemoglobin 2008; 32(5): 434-439.
Koren A, Zalman L, Palmor H, Zamir RB, Levin C, Openheim A, Daniel
Spiegel E, Shalev S, Filon D. Sickle Cell Anemia in Northern Israel: Screening
and Prevention. IMAJ 2008; 11: 229-234.
Yamsri S, Sanchaisuriya K, Fucharoen G, Sae-ung N, Ratanasiri T,
Fucharoen S. Prevention of severe thalassemia in northeast Thailand: 16
years of experience at a single university center. Prenat Diagn 2010; 30: 540546.
Tarazi I, Al-Najjar E, Lulu N, Sirdah M. Obligatory premarital tests for
thalassemia in the Gaza Strip: evaluation and recommendations. Int Jnl Lab
Hem 2007; 29: 111-118.
230
11.
12.
13.
14.
Al-Allawi NA, Al-Dousky AA. Frequency of haemoglobinopathies at premarital

health screening in Dohuk, Iraq: implications for a regional prevention
programme. Eastern Mediterranean Health Journal 2010; 16(4): 381-385.
Karimi M, Jamalian N, Yarmohammadi H, Askarnejad A, Afrasiabi A, Hashemi
A. Premarital screening for -thalassemia in Southern Iran: opinions for
improving the programme. Journal of Medical Screening 2007; 14(2): 62-66.
Al-Sulaiman A, Suliman A, Al-Mishari M, Al-Sawadi A, Owaidah TM.
Knowledge And Attitude Toward The Hemoglobinopathies Premarital
Screening Program in Saudi Arabia: Population Based Survey. Hemoglobin
2008; 32(6): 531-538.
El-Tayeb E-N H, Yaqoob M, Abdur-Rahim K, Gustavson K-H. Prevalence of Thalassemia and Sickle Cell Traits in Premarital Screening in Al-Qassim,
Saudi Arabia. Genetic Counseling 2008; 19(2): 211-218.
Chapter 7
231
Prenatal Diagnosis of -Thalassemias and Hemoglobinopathies

Maria Christina Rosatelli, PhD and Luisella Saba, PhD
Abstract
Prenatal diagnosis of -thalassemia was accomplished for the first time in the
1970s by globin chain synthesis analysis on fetal blood obtained by placental aspiration
at 18-22 weeks gestation. Since then, the molecular definition of the -globin gene
pathology, the development of procedures of DNA analysis, and the introduction of
chorionic villous sampling have dramatically improved prenatal diagnosis of this disease
and of related disorders. Much information is now available about the molecular
mechanisms of the diseases and the molecular testing is widespread. A prenatal
diagnosis has to provide an accurate, safe and early result, an efficient screening of the
population and a rapid molecular characterization of the couple at risk, are necessary
prerequisites. In the last decades earlier and less invasive approaches for prenatal
diagnosis were developed. An overview of the most promising procedure will be done.
Moreover, in order to reduce the choice of interrupting the pregnancy in case of affected
fetus, Preimplantation or Preconceptional Genetic Diagnosis (PGD) has been setting up
for several diseases including thalassemias.
Rosatelli MC, Saba, L. Prenatal Diagnosis of Beta-Thalassemia and

Hemoglobinopathies. Mediterr J Hematol Infec. Dis. 2009; 1(1): e200911
This can also be accessed from http://www.mjhid.org/article/view/5079.
We acknowledge all those concerned with this publication.
Introduction
232
-thalassemias and hemoglobinopathies are among the most common

autosomal recessive diseases with a high frequency in the population of the
Mediterranean area, the Middle East, the Indian subcontinent, the Far East, Tropical
Africa and the Caribbean [1]. However, in the last decades, the steady migratory flows
have rendered these pathologies much more widespread, thus representing a general
public health problem. In the '70s the set-up of globin chain synthesis analysis for the
detection of a little amount of -chains in fetal blood during the 18 th-22nd week of
gestation [2] has allowed the development of screening programs of the general
population, based on the identification of the couple at risk, and, in addition, the offer of
prenatal diagnosis testing. At that time the thalassemic patients had limited lifespan and
prenatal diagnosis represented the only option for the control of the disease. Such
programs first started in Sardinia, Continental Italy, Cyprus, and Greece [3,4,5,6].
Prenatal diagnosis on fetal blood, even if it represented for couples at-risk an
opportunity to generate healthy sons, was not easily accepted. The late gestational age
in which fetal diagnosis was carried out, the risk of misdiagnosis due to a not clear cutoff between some heterozygotes and affected fetuses, the high risk of miscarriage due
to the sampling procedures, made indeed the procedure difficult to accept from the
couples.
The continuous advances in the knowledge of the molecular pathology of the
disease, the discovery of restriction fragment length polymorphisms (RFLP) linked to
the -like globin gene, the development of methodologies for mutation detection and the
application of the villocentesis for the recovery of nucleated fetal cells, allowed a fast
improvement both in feasibility and acceptability of prenatal diagnosis. For a short
233
period, in the eighties, the diagnosis of thalassemia was obtained either indirectly by
linkage analysis using RFLP at the -globin cluster [7] or directly by oligonucleotide
hybridization on electro-phoretically separated DNA fragments [8] or by enzymatic
digestion of mutated sites. A major impulse has been given by the PCR technology that
allowed the development of a number of procedures, for easier mutation detection, as
well as the development of both PGD and non-invasive prenatal diagnosis procedures.
Nowadays thalassemias are detected directly by the analysis of amplified DNA from
fetal trophoblast or, more rarely, from amniotic fluid cells.
In this review we will delineate current procedures for prenatal and
preimplantation diagnosis of thalassemias as well as the most promising approaches for
non-invasive prenatal diagnosis.
Prenatal Diagnosis
Detection Methods:
Detection of molecular defect in both parents is a prerequisite for prenatal
diagnosis of the disease. The majority of defects affecting the -globin gene are point
mutations that occur in critical areas for its function, or single/few base addition/deletion
that change the frame in which triplets are translated into protein. Very rarely thalassemia results from gross rearrangement in the -globin gene cluster. In spite of
the marked molecular heterogeneity, a limited number of molecular defects are
prevalent in every at risk population. This may be very useful in practice, because a
panel of most frequent mutations to be searched for can be designed according the
carrier's ethnic origin [9]. Known mutation detection is caried out by a number of PCR234
based techniques. (ARMS, Amplification Refractory Mutation System) and the reverse
oligonucleotide hybridization with specific oligonucleotide probes (RDB, Reverse
Oligonucleotide-probe analysis).
Primer-specific Amplification:
The method is based on the principle that a primer carrying a mismatch in its 3'
region cannot anneal on its template. With this method, the target DNA fragment is
amplified in two separate PCR reactions using a common primer and either of the two
following primers: one complimentary to the mutation to be detected (-thalassemia
primer) and one complementary, at the same position, to the normal DNA (normal
primer). Normal DNA is amplified only by the normal primer while the DNA from
homozygotes only by the -thalassemia primer and DNA from heterozygotes by both
primers. A different sized fragment of the -globin gene is simultaneously co-amplified
as an internal control of the PCR reaction [10]. The method is very simple as it
requires, for each mutation to be searched, only two PCR reactions followed by agarose
gel electrophoresis. A further improvement of the methodology can be obtained by
multiplexing the primers for more than one mutation. In good hands the method is very
safe and particularity useful in fetal DNA analysis to search for mutations previously
detected in the parents.
Reverse Oligonucleotide Hybridization:

235
When the spectrum of mutations to be searched is complex, ARMS is not the

most appropriate method. In this case RDB results can be more informative and
efficient. The method uses membrane-bound allele-specific oligonucleotide probes that
hybridize to the complementary sequence of the PCR product prepared using patient
DNA as starting template [11]. In this format, multiple pairs of normal and mutant allelespecific oligonucleotides can be placed on a small strip of membrane. Hybridization
with PCR-amplified -globin gene is able to detect, in a single procedure, any of the
mutations screened. Up to 20-30 mutations have indeed been screened in one single
step and several commercial kits are available to detect the most common beta
thalassemia mutations in Mediterranean population.
Other Known Mutation-detection Procedures:

Several other methods have been developed to search for known mutations, i.e.
oligonucleotide ligation assay [12], restriction enzyme digestion of PCR products [13];
however some of them have been abandoned in routine diagnostics as they are less
informative or more complex.
In recent years a real time PCR assay has been successfully applied to both
carriers screening and prenatal diagnosis [14]. This is a one-step method that is based
on the use of fluorescent hybridization probes followed by a melting curve analysis.
This method, which allows the simultaneous multiple mutation detection, has been
successfully applied also to the detection of maternal contamination. In spite of these
advantages its use is still limited as it needs a dedicated apparatus as well as an
accurate population-based design of detection probes.
236
Technically, we can realistically predict further simplification and full automation

of the procedures for the detection of the -thalassemia mutations is commercially
available, which are not completely automated and quite expensive. Among them, the
oligonucleotide microchip-based assays have been proposed many times for the largescale detection of mutations in genetic diseases, including -thalassemia [15]. Given
the alternative features of high throughput and automation, the DNA chip has the
potential to become a valuable method in future applications of mutation detection in
medicine. At the moment, the technology developed several years ago is not yet
transferred in the clinical practice, due to the higher costs and to the lower analytical
sensitivity and specificity.
Unknown Mutations Detection:

When carriers escape to the above mutation detection approaches, further
investigations need to be carried out by alternative methods which uncover the
presence of unknown mutations by scanning the whole gene. Denaturing gradient gel
electrophoresis (DGGE) [16,17,18], Denaturing High Pressure Liquid Chromatography
(dHPLC) and Single Strand Conformation Polymorphism (SSCP) [19] are the most
widely used in the last years, followed by direct sequencing analysis [20] which
characterizes the undefined mutation found by these methods. Nowadays, considering
the small size of the -globin gene (1,8kb), the simplified technologies available and the
reduced costs of analysis, direct sequencing, based on cycle sequencing with
fluorescent dye terminators and automated capillary DNA sequencing technology,
237
seems to be the faster and most useful approach to detect unknown thalassemia
mutations.
If a mutation is not detected by sequence analysis, we search for the presence of
small deletions by polyacrylamide gel electrophoresis of amplicons designed for the
most frequent small deletional defects of the -globin gene (gap PCR). Furthermore,
the presence of larger deletions of the cluster may be identified by Southern-blotting or
more recently by Multiple Ligation-Dependent Probe Amplification (MLPA) for which a
commercial kit is available (SALSA MLPA KIT P102 HBB-MRC Holland).
In a very limited number of cases, direct sequencing from position -600 to 60 bp
downstream from the -globin gene and methods for deletion detection, failed to detect
the disease causing defect. In these cases, the molecular defect may reside either in
the locus control region of the -globin gene cluster, or in one of the genes, outside the
-globin gene region, encoding for regulatory proteins acting in trans on the function of
the -globin gene. Very recently it has been proved that the -thalassemia-like
phenotype could be caused by the coinheritance of a -globin gene defect and a
duplication of the -globin gene cluster, which results in an excess of chain. In these
selected cases, the characterization of these -globin gene rearrangements (SALSA
MLPA KIT P140-B2 HBA-MRC Holland) can be routinely carried out with success by
MLPA analysis.
Genetic Counseling of the Couple at Risk:

238
Both members of the couple at risk are counseled in a non-directive way. The
nature of the disease, the implications of being carriers and reproductive choices are
analyzed, specifically those concerning birth control, including prenatal or
preimplantation diagnosis and the possibility, in case of affected fetus HLA compatible to
not interrupt pregnancy. As for fetal testing, detailed information is offered regarding the
risk of fetal mortality, the risk of misdiagnosis, and the mortality and morbidity of an
abortion in case of affected fetus.
Fetal DNA Sampling:

Fetal DNA for analysis can be obtained from either amniocytes or chorionic villi.
At present the most widely used procedure is chorionic villi sampling, because of the
clear advantage of being carried out during the first trimester of pregnancy, generally at
the 10th-12th week of gestation [21, 22, and 23]. The risk of fetal mortality associated
with both methods is in the order of 1-2%. Chorionic villi may be obtained
transcervically or transabdominally, the last being most widely used, mainly because it
has a low infection rate and a lower incidence of amniotic fluid leakage. Moreover it is a
simple procedure, largely preferred by pregnant women, which can be carried out also
in late gestational age. Samples obtained by villocentesis need to be accurately
dissected under inverted microscope in order to remove maternal decidua, that
represent the major cause of diagnostic error in prenatal diagnosis of monogenic
diseases.
Fetal DNA Analysis:

239
Fetal DNA is analyzed using the same methods described above for detection of
known mutations during carrier molecular screening. To limit the possibility of
misdiagnosis, we analyze chorionic villous DNA with two different procedures: i.e. RDB
hybridization and primer-specific amplification, using distinct couple primers.
Misdiagnosis may occur for several reasons: failure to amplify one copy of the target
DNA fragment, mispaternity, maternal contamination, and sample exchange.
Misdiagnosis for failure of DNA amplification is obviously limited by the double approach
described above. To avoid misdiagnosis due to maternal contamination as well as
mispaternity and/or sample exchange, a fetal DNA microsatellite analysis is usually
performed to verify the presence of one allele from each parent [9]. In our hands, by the
above mentioned PCR-based procedures, no misdiagnoses have occurred in more than
5000 cases. Figure 1 shows the overall results of the Sardinia prenatal diagnosis
program since the beginning of 1976 up to the end of the past year.
240
Currently, prenatal diagnosis is a widely applied and well-accepted procedure.

Among the patients screened we have found an acceptability of 99.3% for early prenatal
diagnosis by CVS. This data, if compared with previously utilized procedures such as
fetal blood sampling, with an acceptability of 93.2%, and 96.4% by amniocentesis,
demonstrates how the acceptance of the procedure depends on its precocity [22].
The screening program in the Mediterranean countries has proven to be very successful
in reducing the number of thalassemia patients. In Sardinia, thalassemia major was
present in 1 in 250 births, and has declined to 1 in 4000 births (Figure 1). Other
countries in which such thalassemia programs have been introduced also show similar
trends.
241
Preimplantation and Preconceptional Genetic Diagnosis:

The progress in assisted reproduction and molecular genetics techniques,
particularly the advent of PCR that has made possible to analyze the genotype of a
single cell, has paved the way for preimplantation genetic diagnosis (PGD) [24,25].
This technique was introduced as an option for avoiding the decision to terminate an
established pregnancy diagnosed as affected by conventional approaches. The term
preimplantation genetic diagnosis describes those procedures which involve the
removal of one or more nuclei from oocytes (polar bodies) or embryos (blastomeres of
trophectoderm cells) to test for mutation in the target gene or aneuploidy before transfer.
PGD requires that couples at risk undergo in vitro fertilization (IVF) even if not
infertile and for this reason a multidisciplinary approach including an appropriate genetic
counseling and the referral to both a fertility clinic and to a highly specialized molecular
genetics laboratory is mandatory.
Counseling for couples considering PGD must include additional information
regarding at least the risk associated with IVF procedures and with embryo biopsy, the
technical limitations of DNA analysis, including the risk of failure of the procedure as
well as that of misdiagnosis, and the need of subsequent prenatal diagnosis to confirm
the result. Beyond that, the possibility that no embryos may be transferred and the
dispositions of the embryos not transferred have also to be seriously considered.
242
Cell Biopsy:
Preimplantation may be carried out by either cleavage-stage biopsy of 1-2
blastomeres, from an eight-cell embryo three days after in vitro fertilization carried out
by ICSI (Intracytoplasmatic Sperm Injection), or by the biopsy of polar bodies.
For cleavage-stage biopsy the embryo is grown in vitro until it reaches a six-eight cell
stage which usually occurs on the third day after insemination. Polar bodies diagnosis,
pioneered by Verlinsky and his group in 2006 is based on the analysis of the first polar
body of unfertilized eggs [27], and may lead to distinguish between unfertilized eggs
that carry the defective gene and those without the defect. The successive sampling
and analysis of the second polar body that is extruded from the oocyte after fertilization
and completion of the second meiotic division, is carried out in order to avoid
misdiagnosis due to the high rate of recombination that happens during the first meiosis.
By fertilizing in vitro only the eggs without the defect and replacing them in the mother, a
successful pregnancy with a normal fetus can be obtained. Recently a preconceptional
genetic diagnosis based on the analysis of only the first polar body has been proposed
for countries in which the use of PGD and manipulation of embryos is prohibited [28].
This approach although permitting to avoid the manipulation, cryopreservation and/or
discard of sovranumerary and/or affected embryos, shows several problems: the need
to obtain more than 10-12 oocytes, the increased risk of diagnostic error and the
increased risk of the technical difficulties. Blastocyst biopsy, even if it has the advantage
to provide a higher number of cells, is at present more rarely used because of the
difficulties of the embryos to reach this stage in IVF programs. The cleavage-stage
243
biopsy of blastomeres from an eight-cell embryo is the most frequently used PGD
procedure all over the world.
Detection Methods:
Methods for mutation detection in OGD are always based on multiple steps of
PCR. Mutations are detected in PCR products by various methods that combine speed,
analytical sensitivity and specificity. In particular, a first round of multiplex PCR is
performed to amplify both the -globin gene region including the mutation and one or
more polymorphic loci. Secondly, two separated nested PCR reactions are performed
to amplify the two or more selected genomic regions. Finally, the polymorphic alleles
are directly detected by capillary electrophoresis of the amplified fragment, while the
presence of -globin gene mutations are identified by the subsequent mini sequencing
reaction [29]. This approach is expressly designed to detect the presence of the globin gene mutations and to monitor, in the same sample, the presence of
contamination as well as the eventual allele drop-out that represent the most frequent
causes of error in PGD.
Quality Control:
For both techniques a prenatal diagnosis by villocentesis is recommended in
order to avoid diagnostic errors. Successful pregnancies following the transfer of
human embryos in which the -globin gene defect has been excluded, occur only in 2025% of cases and the birth rate of a child is even lower. Due to the low birth rate most
women have to undergo PGD several times in order to give birth to a healthy child [30].
244
Transfer of no more than 1-2 embryos is strongly recommended in order to avoid

multiple pregnancies [31]. Elective Single Embryo Transfer (eSET) is in fact a wellestablished procedure which has demonstrated to ensure a better prognosis of IVF
patients [32}.
PD or PGD?
Among clinical geneticists there has been much discussion about the main goal
of PD. Some have argued that the main aim is to avoid the birth of an affected child.
Others have emphasized the reproductive confidence and the purpose of informing the
couples at risk about the status of the fetus. Several studies indicate that if there is no
PD option, a large proportion (up to 50%) of the couples at high risk of an affected child
refrain from pregnancy despite their wish to reproduce. When PD is possible many
more at-risk couples dare to embark on a pregnancy.
Most experts consider PGD as an additional option for couples at risk and not as
a replacement for conventional prenatal diagnosis. PGD is still considered a highly
specialized experimental procedure with limited results, mainly dedicated to couples
against abortion for ethical and religious reasons and to a small proportion of couples
who have experienced repeated abortion, that ask for referral for this procedure.
At present its use in routine monitoring of pregnancies at risk is precluded by the
technical demand for these procedures, the difficulty in organizing the service, and the
high costs.
245
Simplification of preimplantation and preconception genetic diagnosis, together

with an increase in the pregnancy rate may lead to a more extensive use of the
procedure in the future.
Non-Invasive Prenatal Diagnosis (NIPD):

Analysis of Fetal Cells in Maternal Blood:
In the sections below the most significant studies, which have been carried out in
this field of research, are briefly summarized. The mot relevant results have been
grouped in three different sections, according to the different cell type in which they
have been acquired. A separate section is dedicated to NIPD of -thalassemia.
Trophoblasts:
The first evidence that fetal cells circulate in maternal peripheral blood dates
back to 1893 when George Schmorl observed the presence of placentally derived
trophoblasts in the lungs of 17 autopsied women affected by severe eclampsia [33].
In 1959 Douglas [34] established that migration of trophoblasts is a normal process
during pregnancy and twenty-five years later, Covone et al [35] demonstrated that these
cells could be detected in healthy pregnant women as early as six weeks gestation.
They also found that an increased concentration of trophoblast cells were frequently
presenting in women affected by preeclampsia. Further studies have established that
trophoblasts are entrapped in the maternal lungs and rapidly removed from the
pulmonary circulation [36].
246
Tropoblast-specific cell-surface antigens have not yet been characterized and

several experimental evidences have shown that the H315, initially described as the
specific antigen for trophoblasts, is indeed absorbed in maternal leucocytes [37].
These are some of the reasons why, in recent years, trophoblasts are no longer
considered as the best target cells for non-invasive prenatal diagnosis. Nevertheless,
this line of research has not yet been completely abandoned as the characterization of
trophoblast-specific antigens is one of the objectives of the SAFE (Special Non-Invasive
Advances in Fetal and Neonatal Evaluation) Network (for more information please visit
www.safenoe.org).
Lymphocyte:
Fetal lymphocytes are the second cell type which has been extensively studied
as a possible source of fetal DNA. In 1969 Walknowska et al [38] detected for the first
time 46, XY karyotype cells in maternal peripheral blood of women bearing male
fetuses. Ten years later Herzenberg and colleagues described the use of FACS
(Fluorescent Activated Cell Sorting) as a method for the enrichment of fetal lymphocyte
expressing the HLA-A2 paternal antigen [39]. Detection of Y chromosome was then
obtained in the enriched cells deposited directly onto microscope slides, thus confirming
their fetal origin.
Unfortunately other groups have failed to replicate these results with success,
even when cytogenetic analysis was carried out in fetal cells that were flow sorted on
the basis of several HLA differences and by using monoclonal antibodies.
In the same years further studies demonstrated that lymphocytes were not
removed from maternal circulation after delivery. One of the earliest studies provided
247
the first evidences that fetal lymphocytes persist in maternal circulation one year after
delivery was published in 1974 by Bianchi et al [40]. Several years later Bianchi et al
described the presence of fetal progenitor cells 27 years after delivery [41]. For these
reasons also lymphocytes, as trophoblasts, became an unattractive candidate for noninvasive prenatal diagnosis.
Erythroblasts:
One of the main advantages to study fetal erythroid cells is that they are
nucleated, terminally differentiated short-lived cells and for this reason they do not
persist in maternal circulation for a long time after delivery. Furthermore, first primitive
erythroblasts appear in the embryonic bloodstream around the four-five week gestations
so they can be detected early during gestation.
Nevertheless, their isolation from maternal peripheral blood is still problematic
because of their rarity and the lack of a fetal specific antibody.
In 1990 Bianchi [41] first described a method for fetal nucleated erythroid cells CD71
transferrin receptor, highly expressed in erythroid cells. Two years later Ganshirt-Ahlert
et al [43] obtained similar results by using a new detection system called MACS
(Magnetic Cell Sorting) which is based on the use of antibodies labeled with magnetic
beads.
Since then, both systems have been extensively improved and used, by several
groups, following different approaches which can consist in the positive selection of
CD71 and/or glycophorin-A fetal cells and/or in the negative depletion of CD45 maternal
cells. Usually, in both cases, a previous density (Ficoll or Histopaque) gradient
centrifugation step is carried out to remove non-nucleated maternal erythrocytes. A
248
schematic workflow resuming one of the strategies used for isolating fetal NRBCs from
maternal peripheral blood is represented in Figure 2. Finally both MACS and FACS
sorted cells are labeled with fluorescent antibodies which recognize embryonic (, ) or
fetal () hemoglobin chains and are eventually subjected to FISH analysis for
chromosome Y detection. An example of positive labeling with the antibody for gammaglobin conjugated with FITC is shown in Figure 3. Molecular characterization can
eventually be carried out in positive fluorescent cells isolated by laser microdissection.
Even with the high progress made in the last twenty years in this field, the methods for
erythroblasts enrichment are still limited as they mostly result in the recovery of fetal
samples with low yield (FACS) and scarce purity (MACS), being variably contaminated
by maternal cells.
For these reasons in recent years several studies have been addressed to the
proteomic field with the attempt to characterize novel fetal erythroblast cell-specific
surface markers. For example, bi-dimensional electrophoresis coupled with mass
spectrometry has allowed the identification of 2 proteins, differentially expressed in
sickle erythrocytes in comparison to healthy erythrocytes, and the detection of proteins
up-or downregulated in fetal erythroid cells in comparison to their adult counterparts.
Some of these results have been published as a full-patent application and the data
concerning the new antibodies developed against these new targets expect to be
validated in large samples of maternal blood [44].
In addition, further developments in
fetal cell recovery are expected to be obtained through the application of micro-fluidic
rare-cell capture technologies [45] which are being developed to detect not only fetal but
also cancer as well as other rare cells in biologic fluids.
249
250
Analysis of Fetal Cells in Maternal Blood and Non Invasive Prenatal Diagnosis
(NIPD) of -Thalassemia:
Despite the difficulties encountered to find the best target cell and the best
method for their enrichment and isolation, several attempts have been made in the last
twenty years, to transfer the results of these researches into clinical practice.
Unfortunately the lack of reproducibility of experiments hardly makes the isolation of
fetal cells from maternal blood as a first choice method of NIPD of monogenic disorders.
Below the most significant results obtained in NIPD of -thalassemia are briefly
summarized. The first example of non invasive prenatal diagnosis of
hemoglobinopathies was described in 1990 by Camaschella et al [46]. The genetic test
was carried out in three selected couples where the mother was a carrier of thalassemia and the father of the Hb Lepore-Boston trait. The absence/presence of the
paternal trait was successfully detected in PCR amplified samples DNA extracted from
T-cell samples were obtained by incubating Ficoll-separated cells of the mother with the
CD 3-specific MoAb Leu 4 and then separating the positive cells with goat-anti-mouse
immunoglobulin G (1gG)-coated immunomagnetic beads.
In those years most of the studies were addressed to couples carrying different
mutations and only aimed to the exclusion of the paternal allele in the enriched fetal
cells, as most of the times they were contaminated from maternal cells.
In subsequent years, even if the fetal cells enrichment and selection methods
have been greatly improved, other IP diagnosis have been carried out but with
fluctuating results. Below are described three significant examples of NIPD realized,
with different levels of success, by using single or pooled erythroid cells.
251
In 1996 the group of Y.W.Kan47 reported the successful identification of two fetal
genotypes by using fetal nucleated erythroid cells selected by MACS, anti- globin
immunostaining and then isolated by microscopy and cell scraping. The
presence/absence of sickle cell and beta thalassemia mutations of both parents were
finally detected by Reverse Dot Blot in PCR amplified samples constituted by pools of
fetal dissected cells.
A few years later the group of Di Naro [48] replicated these results using a
slightly different procedure for erythroblast enrichment which was carried out by Percoll
and Gastrografin multiple gradient centrifugation. Mutation detection was then obtained
by automated sequencing of single cells amplified by PCR. According to authors, even
if the risk of allele drop out is higher when amplifying single cells, however the possibility
to study several individual, instead of pooled, cells guarantees an accurate diagnosis of
the fetal DNA.
More recently the group of Kolialexi [49] has hardly tried to replicate these
results. In this study, NIPD was performed through magnetic cell sorting (MACS) and
microdissection of single NRBCs with a laser micromanipulation system. Single-cell
genotyping was achieved by nested real-time PCR for genotyping -globin gene
mutations; a multiplexed minifingerprinting was used to confirm the origin of the isolated
cells and to exclude their possible contamination. A total of 224 cells were isolated but
only half of them were successfully amplified. In the majority (n=80) of these cells
minifingerprinting was not informative because of allele dropout or homozygosity. In the
rest of the samples, 22 cells resulted to be of fetal origin, 26 maternal while 80 were non
informative.
252
Analysis of Fetal DNA in Maternal Plasma and Non Invasive Prenatal Diagnosis
(NIPD) of Thalassemia:
The existence of cell-free nucleic acids within the human plasma was firstly
reported in 1948 by Mendel and Metais [50] which described their presence both in
normal subjects and in individuals affected by various diseases. Some decades later
other studies have confirmed the presence of circulating DNA as well as of RNA in
several pathological conditions (pancreatitis, inflammatory diseases, cancer, diabetes,
etc) [51].
In 1997 Lo et al discovered for the first time that a fetus may release cell-free
fetal DNA (cffDNA) into maternal plasma, thus providing an alternative to fetal cells for
noninvasive prenatal diagnosis [52].
In recent years more information has been acquired about the concentration, the
origin and the characteristics of the cell-free fetal DNA and several procedures have
been developed in order to use it in prenatal diagnosis.
The cell-free DNA is constantly present in peripheral blood of non pregnant
women and its concentration increases during pregnancy. The cell-free fetal DNA
represents the 3-5% of the DNA present in maternal plasma from which, after delivery, it
is rapidly cleared.
Recent studies carried out by microfluidic digital PCR have revealed that cffDNA
can be present at even higher concentrations which can reach up to 10-20% of total
DNA in maternal plasma [53]. Nevertheless, because of the high background of
maternal DNA, an enrichment step is needed to obtain highly purified fetal DNA
samples suitable for non invasive prenatal diagnosis.
253
Size-fractionation agarose gel electro-phoresis is one of the methods developed

for fetal DNA enrichment and consists in the isolation of short-length DNA fragments
(<300 bp of length) which is the medium length of the cffDNA. This method coupled
with the peptide-nucleic-acid clamp (PNA) PCR, which selectively suppresses the
amplification of maternal alleles, and with the Allele-specific Real-Time PCR for
mutation detection, has been used with success by Li et al [54] to detect mutations of
paternal origin in fetuses at risk for -thalassemia.
More recently [55] the same group has described a new procedure, still based on
size fractionation method, but coupled with MALDI-TOF mass-spectrometry, a mediumthroughput platform which detects with high sensitivity the presence of known and
unknown point mutations. In this case no suppression of maternal allele was caried out
and the molecular diagnosis was addressed to the exclusion of the paternal mutated
allele. The analysis by MALDI-TOF preceded by size fractionation has given improved
results, in comparison to the absence of enrichment, in the detection of the codon 39 thalassemia paternal allele. Nevertheless, for eventual future diagnostic application the
protocol needs to be validated in larger samples, even if the high cost of the
instrumentation required makes this platform difficult to apply in routine diagnostics and
the size fractionation is considered an enrichment method more susceptible to maternal
contamination.
The use of peptide-nucleic-acid clamping to suppress the amplification of normal
maternal alleles was first described by Cremonesi in 2004 [56]. Peptide nucleic acid is
artificially synthesized polymers similar to nucleic acids and able to hybridize DNA
sequences. The PNA/DNA hybrids are more stable than equivalent DNA/DNA hybrids
254
but less stable in the presence of single-pair mismatches. In that paper their ability to
clamp wild type -globin sequences was proved in artificial mixture of DNA samples
enriched with increased amounts of wild type alleles, by using a microchip platform to
detect the -thalassemia mutations.
Four years later [56] the efficacy of PNA was evaluated with success in 41 non
invasive prenatal diagnosis of -thalassemia and in combination with three different
techniques (microelectronic chip, pyro sequencing and direct sequencing) to detect fetal
DNA mutations in maternal plasma.
Despite its successful application, this strategy, as the other above described
technologies, is still restricted to couples which carry different mutated alleles and
aimed to the detection of mutated paternal alleles.
Another method recently described for NIPD of -thalassemia is called APEX
namely Arrayed Primer Extension. This is a mutation detection system which is based
on the combined use of the microchip technology and the single nucleotide base
extension method. This system has been recently described by the group of
Papasavva [57] and used to characterize the presence of the paternal -thalassemia
mutations and associated -globin gene SNPs, in cffDNA isolated from maternal
plasma. The possibility to study the polymorphisms associated to the mutated alleles
represent a feature of great value since it would give the possibility to extend NPID to
couples which carry the same mutated allele. Prerequisite for its application is to find
informative SNPs associated with parental mutations which can help to discriminate the
paternal mutated allele and to characterize the haplotype inherited from the fetus. The
authors of the paper described the correct application of this methodology in the NIPD
255
of six out of seven couples at risk for -thalassemia, carried out in the Cypriot
population.
Future Perspective:
As previously reported, one of the major problems which still limits the application
of the described protocols in clinical practice is the impossibility to obtain highly purified
fetal, cellular as well as cffDNA, samples which could allow the detection of parental
alleles, even when they are identical. Few clinical applications of NIPD are actually
restricted to the detection of the Y chromosome, for fetal sex determination, or the
Rhesus D gene, in Rhesus D negative women, or, in general, of genetic loci which are
absent in the maternal genome.
In recent years a great improvement has been obtained in the field of the
technologies which can explore the presence of sequence variations even in single
molecules of DNA. The concept of "Digital PCR" was firstly introduced in 1992 by
Sykes [58] who described a method to determine the number of starting DNA templates
by doing Poisson statistical analysis of PCR results obtained in limiting dilutions. The
more recent development of the emulsion PCR (emPCR) have further enhanced the
possibility to study single molecules of DNA by using a small volume of reactions, wateroil emulsions and microfluidic as well as high-throughput platforms (for a review of both
methods and application to NIPD please see Zimmermann et al [59].
Recent applications of these technologies in the field of NIPD, and in particular in
the diagnosis of aneuploidies and monogenic disorders, have shown that these
methodologies may find useful application in the near future, even if several drawbacks
256
need to be solved and wider validation studies should be carried out before transferring
their use in routine diagnostics.
References
1. Wetherall DJ, Clegg JB. The Thalassemia Syndromes. Wiley-Blackwell; 2008.
2. Kan YW, Golbus MS, Klein P, Dozy AM. Successful application of prenatal diagnosis
in a pregnancy at risk for homozygous -thalassemia. New Engl J of Med. 1975;
292:1096-9. [PubMed]
3. Cao A. Results of programs for antenatal detection of thalassemia in reducing the
incidence of the disorder. Bool Rev. 1987; 1:169-176. [PubMed]
4. Cao A, Rosatelli MC, Galanello R. Control of -thalassemia by carrier screening,
genetic counseling and prenatal diagnosis: the Sardinian experience Variation in
the human genome. Chirchester, England: Wiley; 1996. pp 137-155. Ciba
Foundation Symposium 197.
5. Loukopulos D. Current states of thalassemia and sickle cell syndromes in Greece.
Semin Hematol. 1996; 33: 76-86. [PubMed]
6. Angastiniotis M, Kyriakidou S, Hadjiminas M. The Cyprus Thalassemia Control
Program, White Plains. Vol.23. New York: March of Dimes Birth Defects
Foundation; 1988. pp.417-432. Original Article Series. [PubMed]
7. Kan YW, Lee KY, Furbetta M, Angius A, Cao A. Polymorphism of DNA sequence in
New Engl J of Med. 1980; 302: 185-8. [PubMed]
8. Pirastu M, Kan YW, Cao A, Conner BJ, Teplitz RL, Wallace RB. Prenatal diagnosis of
-thalassemia. Detection of a single nucleotide mutation in DNA. New Engl J of
Med. 1983; 309: 284-7. [PubMed]
9. Rosatelli MC, Tuveri T, Scalas MT, Leoni GB, Sardu R, Faa V, Meloni A, Pischedda
MA, Demurtas M, Monni G, et al. Molecular screening and fetal diagnosis of thalassemia in the Italian population. Hum Genet. 1992; 83: 590-2. [PubMed]
10. Newton CR, Graham A, Hepteinstall Le, Powell SJ, Summers C, Kalsheker N, Smith
JC, Markham AF. Analysis of any point mutation in DNA. The amplification
refractory mutation system (ARMS) Nucleic Acid Research. 1989; 17:2503-16.
[PMC free article] [PubMed}
11. Saiki RK, Walsh PS, Levenson CH, Erlich HA. Genetic analysis of amplified DNA
with immobilized sequence-specific oligonucleotide probes. Proc. Natl. Acad. Sci
(USA) 1989; 86:6230-4. [PMC free article] [PubMed]
12. Nickerson DA, Kaiser R, Lappin S, Stewart J, Hood L, Landegren U. Automated
DNA diagnostic using an Elisa-based oligonucleotide ligation Assay. Proc. Natl.
Acad. Sci (USA) 1990; 87: 8923-7. [PMC free article] [PubMed]
13. Pirastu M, Ristaldi MS, Cao A. Prenatal diagnosis of -thalassemia based on
restriction endonuclease analysis of amplified fetal DNA. J Med Genet. 1989; 26:
363-7. [PMC free article] [PubMed]
257
14. Traeger-Synodinos J, Vrettou C, Kanavakis E. Rapid detection of fetal Mendelian

disorders: thalassemia and sickle cell syndromes. Methods Mol Biol. 2008; 444:
133-45. [PubMed]
15. Foglieni B, Cremonesi L, Travi M, Ravani A, Giambona A, Rosatelli MC, Perra C,
Fortina P, Ferrari M. Beta-thalassemia microelectronic chip: a fast and accurate
method for mutation detection. Clin Chem. 2004 Jan; 50(1): 73-9. [PubMed]
16. Myers RM, Fisher SG, Lerman LS, Maniatis T. Nearly all single base substitution in
DNA fragments joined to a GS-clamp can be detected by denaturing gradient gel
electrophoresis. Nucleic Acids Research. 1985; 13: 3131-5. [PMC free article]
[PubMed]
17. Cai SP, Kan YW. Identification of the multiple -thalassemia mutations by
denaturing gradient gel electrophoresis. J Clin Invest. 190; 85: 550-3. [PMC free
article] [PubMed]
18. Rosatelli MC, Dozy A, Faa V, Meloni A, Sardu R, Saba L, Kan YW, Cao A. Molecular
characterization of -thalassemia in the Sardinian population. Am J Hum Genet.
1992; 50: 422-6. [PMC free article] [PubMed]
19. Orita M, Iwahana H, Kanazawa H, et al. Detection of polymorphisms of human DNA
by gel electrophoresis as single-strand conformation polymorphism. Proc. Natl.
Acad. Sci (USA) 1989; 86: 2766-70. [PMC free article] [PubMed]
20. Sanger F, Micklen S, Coulson AR. DNA sequencing with chain terminating
inhibitors. Proc. Natl. Acad. Sci (USA) 1977; 74: 5463-7. [PMC free article]
[PubMed]
21. Hogge WA, Schonberg SA, Golbus MS. Chorionic villous sampling: experience of
the first 1000 cases. Am J Obstet Gynaecol. 1986; 154: 1249-52. [PubMed]
22. Cao A, Cossu P, Monni G, Rosatelli C. Chorionic villous sampling and acceptance
rate of prenatal diagnosis. Pren Diagno. 1987; 7: 531-3. [PubMed]
23. Brambati B, Lanzani A, Oldrini A. Transabdominal chorionic villous sampling,
clinical experience of 1159 cases. Pren Diagn. 1988; 8: 609-13. [PubMed]
24. Monk M, Holding C. Amplification of a -hemoglobin sequence in individual human
oocytes and polar bodies. Lancet. 1990; 325: 985-8. [PubMed]
25. Handyside AH, Kontogianni E, Hardy K, Winston R. Pregnancies from biopsied
human preimplantation embryos sexed by Y-specific DNA amplification. Nature.
1990; 344: 768-70. [PubMed]
26. Verlinsky Y, Ginsberg N, Lifchez A, Valle J, Moise J, Strom CM. Analysis of the first
polar body: preconception genetic diagnosis. Hum Reprod. 1990 Oct; 5: 826-9.
[PubMed]
27. Strom CM, Rechitsky S, Wolf G, Cieslak J, Kuliev A, Verlinsky Y. Preimplantation
diagnosis of autosomal dominant retinitis pigmentosum using two simultaneous
single cell assays for a point mutation in the rhodopsin gene. Mol Hum Reprod.
1998 Apr; 4: 351-5. [PubMed]
28. Fiorentino F, Biricik A, Nuccitelli A, De Palma R, Kahraman S, Sertyel S, Karadayi H,
Cottone G, Baldi M, Caserta D, Moscarini M. Rapid protocol for pre-conception
genetic diagnosis of single gene mutations by first polar body analysis: a possible
solution for the Italian patients. Prenat Diagn. 2008 Jan; 28: 62-4. [PubMed]
258
29. Monni G, Cau G, Usai V, Perra G, Lai R, Ibba G, Faa V, Incani F, Rosatelli MC.
Preimplantation genetic diagnosis for beta-thalassemia: the Sardinian experience.
Prenat Diagn. 2004 Dec 15; 24(12): 949-54. [PubMed]
30. Goossens V, Harton G, Moutou C, traeger-Synodinos J, Van Rij M, Harper JC.
ESHRE PGD Consortium data collection IX: cycles from January to December
2006 with pregnancy follow-up to October 2007. Hum reprod. 2009 Aug; 24: 1786810. [PubMed]
31. Cutting R, Morroll D, Roberts SA, Pickering S, Rutherford A, BFS and ACE Elective
single embryo transfer: guidelines for practice British Fertility Society and
Association of Clinical Embryologists. Hum Fertil (Camb) 2008 Sep; 11: 131-46.
[PubMed]
32. Schmorl G. Pathologisch-anatomische Untersuchungen ueber Publer eklampsie.
Vogel; Leipzig: 1893.
33. Douglas GW, Thomas L, Carr M, Cullen NM, Morris R. Trophoblasts in the
circulating blood during pregnancy. Am J Obstet Gynecol. 1959 Nov; 78: 960-73.
[PubMed]
34. Covone AE, Johnson PM, Mutton D, Adinolfi M. Trophoblast cells in peripheral
blood of pregnant women. Lancet. 1984; ii: 841-843. [PubMed]
35. Attwood HD, Park WW. Embolism to the lungs by trophoblasts. J. Obstet.
Gynaecol. Br. Commonw. 1960; 68: 611-617. [PubMed]
36. Bertero Mt, Camaschella C, Serra a, Bergui L, Caligaris-Cappio F. Circulating,
trophoblast cells in pregnancy have maternal genetic markers. Prenat Diagn. 1988;
8: 585-90. [PubMed]
37. Walknoska J, Conte FA, Grumbach MM. Practical and theoretical implication of
fetal/maternal lymphocyte transfer. Lancet. 1969; i: 1119-1122. [PubMed]
38. Herzenberg LA, Bianchi DW, Schroder J, Cann HM, Iverson GM. Fetal cells in the
blood of pregnant women: detection and enrichment by fluorescence activated cell
sorting. Proc. Natl. Acad. Sci USA. 1979; 76: 1453-1455. [PMC free article]
[PubMed]
39. Schroder J, Tilikainen A, de la Chapell A. Fetal leucocytes in maternal circulation
after delivery. Transplantation. 1974; 17: 346-360. [PubMed]
40. Bianchi DW, Zickwolf GK, Weil GJ, Sylvester S, DeMaria MA. Male fetal progenitor
cells persist in maternal blood for as long as 27 years postpartum. Proc. Natl. Acad.
Sci. USA. 1996; 93: 705-708. [PMC free article] [PubMed]
41. Bianchi DW, Flint AF, Pizzimenti M, Knoll JH, Latt SA. Isolation of fetal DNA from
nucleated erythrocytes in maternal blood. Proc. Natl. Acad. Sci. USA. 1990; 87:
3279-3283. [PMC free article] [PubMed]
42. Ganshirt-Ahlert D, Burschyk M, Garritsen HSP, Helmer L, Miny P, Horst J, Schneider
HP, Holzgreve W. Magnetic cell sorting and the transferrin receptor as potential
means of prenatal diagnosis from maternal blood. Am. J. Obstet. Gynecol. 1992;
166: 1350-1355. [PubMed]
43. Avent ND, Plummer ZE, Madgett TE, Maddocks DG, Soothill PW. Post-genomics
studies and their application to non-invasive prenatal diagnosis. Semin Fetal
Neonatal Med. 2008; 13:91-8. Review. [PubMed]
44. Huang R, Barber TA, Schmidt MA, Tompkins RG, Toner M, Bianchi DW, Kapur R,
Flejter WL. A microfluidics approach for the isolation of nucleated red blood cells
259
(NRBCs) from the peripheral blood of pregnant women. Prenat Diagn. 2008 Oct;
28: 892-9. [PubMed]
45. Camaschella C, Alfarano A, Gottardi E, Travi M, Primignani P, Caligaris Cappio F,
Saglio G. Prenatal diagnosis of fetal hemoglobin Lepore-Boston disease on
maternal peripheral blood. Blood. 1990; 75: 2102-2106. [PubMed]
46. Cheung MC, Goldberg JD, Kan YW. Prenatal diagnosis of sickle cell anemia and
thalassemia by analysis of fetal cells in maternal blood. Nat Genet. 1996 Nov; 14:
264-8. [PubMed]
47. Di Naro E, Ghezzi F, Vitucci A, Tannoia N, Campanale D, D'Addario V, Holzgreve W,
Hahn S. Prenatal diagnosis of beta-thalassemia using fetal erythroblasts enriched
from maternal blood by a novel gradient. Mol Hum Reprod. 2000 Jun; 6: 571-4.
[PubMed]
48. Kolialexi A, Vrettou C, Traeger-Synodinos J, Burgemeister R, Papantoniou N,
Kanavakis E, Antsaklis A, Mavrou A. Noninvasive prenatal diagnosis of betathalassemia using individual fetal erythroblasts isolated from maternal blood after
enrichment. Prenat Diagn. 2007 Dec; 27: 1228-32. [PubMed]
49. Mendel P, Metais P. Les acides nucleiques du plasma sanguine chez l'homme. C.
R. Acad. Sci. Paris. 1948; 142: 241-243.
50. Gahan PB, Swaminathan R. Circulating nucleic acids in plasma and serum. Recent
developments. Ann N Y Acad. Sci. 2008 Aug; 1137: 1-6. [PubMed]
51. Lo YM, Corbetta N, Chamberlain PF, Rai V, Sargent IL, Redman CW, Wainscoat JS.
Presence of fetal DNA in maternal plasma and serum. Lancet. 1997 Aug 16; 350:
485-487. [PubMed]
52. Microfluidics digital PCR reveals a higher than expected fraction of fetal DNA in
maternal plasma. Lun FM, Chiu RW, Allen Chan KC, Yeung Leung T, Kin Lau T,
Dennis Lo YM. Clin Chem. 2008 Oct; 54: 1664-72. [PubMed]
53. Detection of paternally inherited fetal point mutations for beta-thalassemia using
size-fractionated cell-free DNA in maternal plasma. Li Y, Di Naro E, Vittuci A,
Zimmerman B, Holzgreve W, Hahn S. JAMA. 2005 Feb 16; 293: 843-9. [PubMed]
54. Li Y, Di Naro E, Vitucci A, Grill S, Zhong XY, Holzgreve W, Hahn S. Size
Fractionation of Cell-Free DNA in Maternal Plasma Improves the Detection of a
Paternally Inherited beta-Thalassemia Point Mutation by MALDI-TOF Mass
Spectrometry. Fetal Diagn Ther. 2009 Jun 5; 2: 246-249. [PubMed]
55. Cremonesi L, Galbiati S, Foglieni B, Smid M, Gambini D, Ferrari A, Viora E,
Campogrande M, Pagliano M, Travi M, Piga A, Restagno G, Ferrari M. Feasibility
study for a microchip-based approach for noninvasive prenatal diagnosis of genetic
diseases. Ann N Y Acad. Sci. 2004 Jun; 1022: 105-12. [PubMed]
56. Galbiati S, Foglieni B, Travi M, Curcio C, Restagno G, Sbaiz L, Smid M, Pasi F,
Ferrari A, Ferrari M, Cremonesi L. Peptide-nucleic acid-mediated enriched
polymerase chain reaction as a key point for noninvasive prenatal diagnosis of
beta-thalassemia. Haematologica. 2008 Apr; 93: 610-4. [PubMed]
57. Papasavva T, Kalikas I, Kyrri A, Kleanthous M. Arrayed primer extension for the
noninvasive prenatal diagnosis of beta-thalassemia based on detection of single
nucleotide polymorphisms. Ann N Y Acad. Sci. 2008 Aug; 1137: 302-8. [PubMed]
260
58. Sykes PJ, Neoh SH, Brisco MJ, Hughes E, Condon J, Morley AA. () Quantitation of
targets for PCR by use of limiting dilution. Biotechniques. 1992; 13: 444-449.
[PubMed]
59. Zimmermann BG, Grill S, Holzgreve W, Zhong XY, Jackson LG, Hahn S. Digital
PCR: a powerful new tool for noninvasive prenatal diagnosis? Prenat Diagn. 2008
Dec; 28: 1087-93. [PubMed]
261
Chapter 8
Hemoglobin A1c
Zia Uddin, PhD
8.1 Introduction
The term glycated hemoglobin refers to the non-enzymatic, irreversible, covalent
bonding of glucose at one or both N-terminal valine residues of the hemoglobin -chain,
and the N-terminus of the -chains, or the -amino groups of lysine residues (Figure 1).
Figure 1. Non-enzymatic glycation of hemoglobin
The term normal hemoglobin phenotype beyond the neonatal period involves a
major fraction due to Hb A (22), and a minor fraction of Hb A2 (22). Occasionally a
very minor fraction of Hb F (22) is also detected. Further chromatographic analysis of
262
Hb A showed that it contains a number of minor hemoglobins, e.g., Hb A1a1, Hb A1a2, Hb

A1b and Hb A1c.These four minor fractions of Hb A were collectively referred as 1) Hb
A1, 2) fast hemoglobins, 3) glycosylated hemoglobins, 4) glycated hemoglobins , or
5) glycohemoglobins. To provide more consistency in nomenclature the Joint
Commission on Biochemical Nomenclature of the International Union of Pure and
Applied Chemistry has recommended the term glycated hemoglobin instead of the
above mentioned five names used in the literature.
The Hb A1a1 and Hb A1a2 fractions that are covalently bonded to Glucose-6phosphate account for 10% of the glycated hemoglobin. In Hb A1b the N-terminus of
the -chain is covalently bonded to pyruvic acid instead of glucose molecule, and this
also accounts for 10% of the glycated hemoglobin.
Hb A1c is a specific species of glycated hemoglobin resulting from covalent
1
bonding of glucose to the N-terminal valine of the hemoglobin -chain. Hb A1c

accounts for 80% of the glycated hemoglobin, and more importantly it is the only
portion of the glycated hemoglobin that is elevated in diabetes. Since hemoglobin
remains in the red blood cell during its entire life span ( 120 days), the constantly
changing glucose level in the cell will directly effect the formation of Hb A1c. Therefore,
the measurement of Hb A1c is directly proportional to the time averaged glucose levels.
263
The fast moving forms of Hb A were separated in the late 1950s and recognized as
being associated with diabetes in the late 1960s
. Since Hb A1a1, Hb A1a2 and Hb A1b
are not elevated in diabetes, the clinical focus has been solely on Hb A1c. For clinical
testing purposes the term Hb A1c analysis is referred to as A1c or the A1c test with the
word hemoglobin omitted as a matter of convenience.
8.2 Hb A1c Diagnostic Role in Diabetes Mellitus, and Glycemic Control in Adults
After three decades of investigation and evaluation of numerous proposals by
various scientific and clinical organizations, Hb A1c found its status as a diagnostic test
4
for diabetes mellitus. One of the four criteria for the diagnosis of diabetes mellitus are:
Hb A1c > or = 6.5% (48 mmol/mL)
Fasting plasma glucose > or = 126 mg/dL (7.0 mmol/L)
2-h plasma glucose > or = 200 mg/dL (11.1 mmol/L) during an Oral
Glucose Tolerance Test
Symptoms of hyperglycemia and casual plasma glucose > or = 200 mg/dL

(11.1 mmol/L)
Several investigators have recently suggested the combined use of fasting glucose and
Hb A1c for the diagnosis of diabetes mellitus.
5-7
Beyond diagnosis, modification of medical
treatment for diabetics is now being performed based on the laboratory test results of Hb A 1c.
One objective would be to get glucose levels to as close to normal as possible with minimal or
264
no hypoglycemia. American Diabetic Association (ADA) has suggested the lowering of Hb A 1c

< 7% for non-pregnant adults for reducing microvascular, and neuropathic complications of the
8
disease (type I and II). Recently a follow up study of the ACCORD (Action to Control
Cardiovascular Risk in Diabetes) stipulated that the best target of Hb A1c in middle aged or
older patients with cardiovascular risk factors is between 7.0 and 7.9%.
Hb A1c is widely used to judge the treatment of diabetes and adjustment of the
medication dose when necessary. In chronic glycemia the blood glucose is monitored more
frequently (once a day or more). Since Hb A1c is measured less frequently and in percent, and
is a complicated process to explain to the patient, it is convenient for the physician to relate
the result to glucose concentration (in mg/dL or mmol/L) over the preceding 5-12 weeks. This
derived glucose concentration from Hb A1c value is called Estimated Average Value (eAG).
Patients monitor their blood glucose and their physician can relate that performance to the
eAG. This way the patients can see the effect of their behavior over time on the test outcome.
The only way this feat could be accomplished, if the result for Hb A1c be the same no matter
where the result was run. This simple feat required the cooperation of many government
agencies and all Hb A1c laboratory testing manufacturing facilities and was brought about by
the determination of the Diabetes Control and Complication Trial (DCCT) Research Group and
the American Diabetes Association.
The mathematical relationship then developed between HbA1c and eAG is based on
the following linear regression equation.
10
265
eAG (mg/dL)
= (28.7 x Hb A1c %) - 46.7
eAG (mmol/L) = (1.59 x Hb A1c %) - 2.59

Table 1 provides the National Glycohemoglobin Standardization Program (NGSP)
Values
11
of Hb A1c % and its corresponding eAG.
Table 1.
NGSP (Hb A1c%)
eAG(mg/dL)
eAG (mmol/L)
97
5.4
126
7.0
154
8.6
183
10.2
212
11.8
10
240
13.4
11
269
14.9
12
298
16.5
8.3 Measurement of Hb A1c
266
Currently over 100 different methods are available for quantification of Hb A1c.
Most available Hb A1c methods are certified by the NGSP
12
and are based on one of
the following techniques:
Immunoassay
Boronate affinity binding/HPLC
Ion-exchange HPLC
Capillary zone electrophoresis
Enzymatic
Measurement of Hb A1c was recently reviewed (December 12, 2012) by David B.

13
Sacks.
(also available on line)
http://care.diabetesjournals.org/content/35/12/2674.full)). It is encouraging to note that

most of the commercial diagnostic manufacturers for Hb A1c test kits are now attempting
to provide an acceptable Hb A1c for eAG calculation.
8.4
Factors Affecting the Accuracy of Hb A1c Assay

In spite of the efficacy of Hb A1c in the diagnosis and the management of
diabetes (type I and II), several factors influence the accuracy of its laboratory results,
e.g., a) hemolytic disease or other conditions with reduced red blood cell survival, b)
recent blood loss, c) iron deficiency anemia, d) patients with renal failure, and e)
hemoglobin variants. All these interferences cannot be easily delineated by the
laboratory personnel and the physician. Due to the diluvial of methods, reagents, and
267
instruments for the assay of Hb A1c , it is impossible for the laboratory to be aware of
the methods limitation with respect to the presumptive interference by >1000
hemoglobin variants reported so far in the literature (http://globin.cse.psu.edu). In the
case of the most common hemoglobinoathies (AS, AE, AC, AD), Hb A1c can be
accurately measured if the correct method is used. The affect of these hemoglobin
variants (AS, AE, AC, AD) and elevated Hb F in HPFH (not pathological) on the results
of Hb A1c by the most often used methods is presented in Table 2.
268
Table 2. Hb A1c methods: Effects of Hemoglobin Variants (Hb C, Hb S, Hb E

and Hb D traits) and Elevated Fetal Hemoglobin (Hb F). Updated
March 2013. (with the permission of http://www.ngsp.org/interf.asp).
The methods are listed in an alphabetic order of manufacturers name.
The criteria used to determine whether or not a method shows
interference that is clinically significant (indicated by Yes) is >
+ or 7% at 6 and/or 9% Hb A1c. In the absence of data for a specific
method (designated by @), it can generally be assumed that
immunoassay methods do not have clinically significant interference
from Hb E and Hb D because the E and D substitutions are distant
from the N-terminus of the hemoglobin -chain. In the absence of
data for a specific method (designated by $), it can generally be
assumed that both immunoassay and boronate affinity methods show
interference from Hb F levels above 10-15%.
In situations where Hb A1c cannot be reliably measured, an alternative is the

assay of serum frustosamine. Fructosamine is the generic name for plasma protein
ketoamines and is also known as glycated serum protein (GSP). Frustosamine provides
evaluation of glucose status over a short period of time (2-3 weeks rather than months).
269
Several studies have shown a correlation of Hb A1c with fructosamine and was thus
recommended in patients with hemoglobinopathies.
14
References
1
2
3
4
5
6
8
9
10.
11.
Sacks DB. Carbohydrates. In Burtis CA, Ashwood ER, eds. Tietz

Fundamentals of Clinical Chemistry. 5th Ed. St. Louis: W.B. Saunders
2011; 452-457.
Allen DW, Schroeder WA, Balog J. Observations on the chromatographic
Heterogeneity of Normal Adult and Fetal Human Hemoglobin: A Study of
the Effects of Crystallization and Chromatography on the Heterogeneity
and Isoleucine Content. Am J Chem Soc 1958; 80: 1628-34.
Rahbar S. An abnormal hemoglobin in red cells of diabetics. Clin Chim
Acta 1968; 22: 296-98
Sacks DB, Arnold M, Bakris GL, Bruns DE, Horvath AR, Kirkman MS,
Lernmark A, Metzger BE, Nathan DM. Guidelines and Recommendations
for Laboratory Analyzers in the Diagnosis and Management of Diabetes
Mellitus. Clin Chem 2011; 57: 793-798.
Inzucchi SE. Diagnosis of Diabetes. N Engl J Med 2012; 367: 542-50.
Hu Y, Kiu W, et al. Combined use of fasting plasma glucose and glycated
hemoglobin A1c in the screening of diabetes and impaired glucose
tolerance. Acta Diabetol 2010; 47: 231-36.
Heianza Y, Hara S, Arase Y, et al. Hb A1c 5.7-6.4% and impaired fasting
plasma glucose for diagnosis of prediabetes and risk of progression to
diabetes in Japan (TOPICS 3): a longitudinal study. Lancet 2011; 378:
147-55.
American Diabetic Association Clinical Practice Recommendations:
Executive Summary: Standard Methods of Care in Diabetes-2010.
Diabetes Care 2010: 33, suppl. 1: S4-5.
Gerstein HC, Miller ME, Genuth S, et al. ACCORD Study Group.
Long term effects of intensive glucose lowering on cardiovascular
outcomes. N Engl J Med 2011; 364 (9): 818-828.
Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ,
and for the A1c-Derived Average Glucose (ADAG) Study Group.
Translating the A1c Assay into Estimated Average Glucose
Values. Diabetes Care 2008; 31: 1473-1478.
Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ,
FOR THE A1c-Derived Average Glucose (ADAG) Study Group.
Translating the A1c Assay into Estimated Average Glucose Values.
270
12.
13.
14.
Diabetes Care 2008; 31: 1-6.

List of NGSP Certified Methods-Hb A1c (updated 11/2012).
http:/www.ngsp.org
Sacks DB, Measurement of Hemoglobin A1c. A new twist on the
path to harmony. Diabetes Care 2012; 35 (12): 2674-2680.
http://labtestsonline.org/understanding/analytes/fructosamine/tab/test.
Additional references (not quoted above) concerning hemoglobin variant

interference in the assay of Hb A1c.
i
ii
iii
iv
vi
vii
viii
ix
Sofronescu A-G, Williams LM, Andrews DM, Zhu Y. Unexpected

Hemoglobin A1c Results. Clin Chem 2011; 57:2, 153-157
Selvin E, Steffes MW, Ballantyne CM, Hoogeveen RC, Coresh J, Brancati
FL. Racial Differences in Glycemic Markers: A cross-sectional Analysis of
Community-Based Data. Ann Inter Med 2011; 154: 303-309
Bergman A-C, Beshara S, Byman I, Karim R, Landin B. A New -Chain
Variant: Hb Stockholm [7(A4)GluAsp] Causes Falsely Low A1c.
Hemoglobin 2009; 33(2): 137-142
Williams JP, Jackson H, Green BN. Hb Belleville [10(a&)AlaThr] Affects
the Determination of HbA1c by Routine Cation Exchange High
Performance Liquid Chromatography. Hemoglonin 2009; 33(1): 45-50.
+
Zhu Y, Williams LM. Falsely elevated hemoglobin A1c due to S-beta thalassemia interference in Bio-Rad Variant II Turbo HbA 1c assay. Clin
Chem Acta 2009; 409(1-2): 18-20.
Thevarajah M, Nadzimah MN, Chew YY. Interference of hemoglobin A1c
(HbA1c) detection using ion-exchange high performance liquid
chromatography (HPLC) method by clinically silent hemoglobin variant in
University Malaya Medical Center (UMMC)- A case report. Clin Biochem
2009; 42: 430-434.
Mongia SK, Little RR, Rohlfing CL, Hanson S, Roberts RF, Owen WE,
DCosta MA, Reyes CA, Luzzi VI, Roberts WL. Effects of Hemoglobin C
and S on the Results of 14 Commercial Glycated Hemoglobin Assays. Am
J Clin Pathol 2008; 130: 136-140.
Barakat O, Krishnan STM, Dhatariya K. Falsely low HbA 1c value due to a
rare variant of hemoglobin J-Baltimore. Primary Care Diabetes 2008; 2:
155-157.
Little RR, Rohlfing CL, Hanson S, Connolly S, Higgins T, Weykamp CW,
DCosta M, Luzzi V, Owen WE, Roberts WL. Effects of Hemoglobin (Hb) E
and HbD Traits on Measurements of Glycated Hb (HbA 1c) by 23 Methods.
Clin Chem 2008; 54:8, 1277-1282.
271
x
xi
Lee S-T, Weykamp CW, Lee Y-W, Kim J-W, Ki C-S. Effects of 7
Hemoglobin Variants on the Measurement of Glycohemoglobin by 14
Analytical Methods. Clin Chem 2007; 53(12): 2202-2205.
Roberts WL. Hemoglobin Constant Spring can interfere with Glycated
Hemoglobin Measurements by boronate Affinity Chromatography. Clin
Chem 2007; 53(1): 142-43.
272
Case Studies
Introduction
The following case (# 1-28) studies include laboratory data representing results
from five different hemoglobin separation methods commonly used in the clinical
laboratory. Due to the large number of variants possible the mandate is that
more than one separation method be used in identification. The question is which
two methods would provide discriminative information. The results of the lab
tests for each case are presented in a tabular form to assist in these choices.
The alkaline electrophoresis images are of Helena SPIFE Alkaline
Electrophoretic results but identical separation results would have also been
obtained using alkaline cellulose acetate, Helena Biosciences SAS alkaline
hemoglobin gels, Helena Quick Gels or Sebia Hydrasys alkaline hemoglobin
gels.
The acid electrophoretic images are of Helena SPIFE or QUICK Gel Acid
electrophoresis. For Acid electrophoretic separation, two classes of media have
been used with differing separation results. Historically, acid hemoglobin
separation was done on agar using citric acid buffer. Helena SPIFE and Quick
Gels are of this type. Agarose purified from agar has more recently been used by
Beckman, Sebia and Helena BioSciences. The purified nature of the agarose
makes these products easier to produce but historically they lacked easily
available documentation of the differences in mobilities compared to the
273
historically used agar. These differences have been documented in the table
associated with the attached case studies. All acid agarose data were adopted
from (Variant Hemoglobins. a Guide to Identification. 1 st edition, by Barbara J.
Bain , Barbara J. Wild , Adrian D. Stephens, Lorraine A. Phelan . Published
2010 by Wiley-Blackwell Publishing Ltd).
All Capillary Zone Electrophoresis (CZE) data were generated using the Sebia
Capillarys System. This CZE system separates hemoglobins into 15 zones and
provides a list of possible variants that migrate in that zone. The operator then
selects the hemoglobin variant they expect that peak to represent. The peaks in
the CZE reports in the case studies have been labeled in such a fashion but a
different assignment could have been made by the operator had they had
information warranting the choice. Details of other vendor results would require
contact with the vendor but the goal again would be to maintain equality as close
as possible and the assumption would be that the order of separation would not
be different.
All isoelectric focusing images are actual or simulated from actual data obtained
with the Helena Isoelectric Focusing Gels either on the SPIFE or the REP
systems. The Perkin Elmer Resolve (formerly IsoLab) isoelectric focusing
systems would obtain the same results, because the pH range of the ampholytes
are the same. These agarose gels contain acrylamide to sharpen the bands.
Again the end user is the final discriminator. In this case, proper selection of the
274
controls determines the degree of discrimination possible instead of the number

of Zones available. If there is Hb S control and the variant migrates anodal to Hb
S, the variant might be Hb D or Hb G but you may not reliably report which, even
though they are both anodal to Hb S. You only know it is not Hb S. If the control
is Hb D or Hb G then you may report based on the migration compared to that
control.
The High Performance Liquid Chromatography (HPLC) separations were all
obtained using BioRad Variant information from several sources.
This data for cases 1-28 is the cooperative effort of many institutions.
Hemoglobin screening is done on neonates as well as adults. Sometimes data
from these rarer hemoglobin variants may include Hb F at low levels that is
ignored in the discussion because its presence is to be expected due to the
patients age. In this regard, some discrepancy in the data may appear. The
presence of an alpha chain variant on a newborn can be complicated by this
temporary presence of gamma chains. The gamma chains compete for the
variant alpha chains as well as the normal resulting in two gamma alpha
possibilities. In neonates the Hb A2 is barely visible because delta chain
production is just beginning. If sufficient delta chains are expressed they also
would show a competition for alpha bands resulting in Hb A2 and a smaller
alpha variant band. These complications will be discussed in the cases in which
they are encountered.
Case # 1
Normal Adult
71 years old male, recent medical examination showed no abnormality.

275
Laboratory Data:
Hemoglobin
Hematocrit
RBC
MCV
MCH
MCHC
RDW
Platelet
13.5
39.6
4.7
84.3
28.8
34.1
13.5
200
13.5 -18.5 g/dL

38.0 - 54.0 %
3
4.6 - 6.2 Mil/mm
80 - 100 fL
27 - 34 pg
31 - 36%
11.5 - 14.5%
3
150 - 400 Th/mm
Hb A
Hb A2
Hb F
98.0
1.8
0.2
94.3 - 98.5%
1.5 - 3.7%
0.0 - 2.0%
Peripheral Blood Smear:
No abnormality was detected.
Agarose Gel Electrophoresis (pH 8.6)
Case # 1 Normal Adult
Citrate Agar Electrophoresis (pH 6.2)
276
277
Interpretation & Discussion:

Agarose gel electrophoresis at alkaline pH 8.6 showed a major band
278
(98%) in the position of Hb A, and a very faint band of Hb A2 (1.8%) in the

position of Hb C. Another very faint band detected cathodal to Hb A2 is due to
the enzyme carbonic anhydrase. This carbonic anhydrase band is mostly
detected in fresh specimens of blood.
Acid electrophoresis at pH 6.2 does not separate Hb A from Hb A2 ,
therefore only one major band is shown in the position of Hb A. A smudged band
cathodal to Hb A includes Hb F, and modified forms of Hb A such as HbA1c.
Isoelectric focusing showed a major band in the position of Hb A, and a
very faint band in the position of Hb A2. A smudged minor band anodal to Hb A
represents modified Hb A such as acetylated Hb F, denatured Hb A, and Hb A1c.
From capillary zone electrophoresis a major peak of Hb A was detected in

window Z9, and a minor peak due to Hb A2 was present in window Z3. No other
peaks were observed.
High performance liquid chromatography showed a major peak at a
retention time of 2.42 (peak value) ascribed to Hb A and a minor peak due to Hb
A2 at a retention time of 3.64 (peak value). There are 2-3 minor peaks before Hb A
and after Hb F, and these peaks represent Hb A1c fraction besides fractions of
other hemoglobins.
The term normal hemoglobin phenotype beyond the neonatal period
involves a major band due to Hb A (22), and a minor band of Hb A2 (22).
279
Occasionally a very faint band of Hb F (22) is also detected. By definition the

concentration of both the Hb A2 and Hb F must be in the normal range for that
method regardless of the methodology used.
Reference
Bain BJ. Hemoglobin and the genetics of hemoglobin synthesis: In:
Haemoglobinopathy Diagnosis, Blackwell Publishing, second edition, 2006, pp 12-22.
280
Case # 2 Hemoglobin S trait

A 20 year old female African-American pre-nursing student in a local community college
was screened for hemoglobinopathy by her family physician. Her physical examination
and chemistry profile were normal.
Laboratory Data:
Hemoglobin
Hematocrit
RBC
MCV
MCH
MCHC
RDW
Platelet
Hb A
Hb S
Hb A2
Hb F
13.1
39.6
4.4
82.1
27.3
32.1
12.6
267
59.2
38.4
1.8
0.6
12.0 16.0 g/dL

35.0 - 48.0 %
3
4.0 5.5 Mil/mm
79-98 fL
26-34 pg
31-36%
11.5-14.5%
3
150-400 Th/mm
94.3-98.5%
1.5 -3.7%
0.0-2.0%
Peripheral Blood Smear: No abnormality was present.

Solubility test for Hb S was positive.
281
282
Interpretation & Discussion

283
Summary of Results
Method
Hb A
area
Hb S
area
Alk Agarose
Major
band
(Hb A)
Major
band
(Hb S)
Acid
Agar/Agaros
e
Major
band
(Hb A+
Hb A2)
Major
peak
(Hb A)
Zone 9
Major
band
(Hb S)
IEF
Major
band
(Hb A)
Major
band
(Hb S)
HPLC
Major
peak
(Hb A)
RT=2.3
4
Major
peak
(Hb S)
RT=4.2
6
CZE
Major
peak
(Hb S)
Zone 5
Hb
A2/C
area
Minor
band
(Hb A2)
Minor
peak
(Hb A2)
Zone 3
Minor
band
(Hb A2)
Minor
peak
(Hb A2)
RT=3.6
5
Since the solubility test was positive and the aberrant band fell between
35 40%, a diagnosis of Hb S trait was made. Concentrations of Hb S other than
35 40% require consideration of the effect of a transfusion, the possibility of
iron deficiency, a concurrent Hb S--thalassemia (Hb S < 33%), a Hb S-thalassemia (Hb S >49%) or the possibility that the fraction may not be Hb S at
all. Mutation at the 6th amino acid position of the chain [6 (A3) GluVal)
causes the substitution of glutamic acid by valine that results in the formation of
Hb S.
284
Since one negative charge is reduced by this mutation, Hb S migrates

slower than Hb A in alkaline and acid electrophoretic procedures. There are other
Hb variants that migrate in the position of Hb S in alkaline electrophoresis, but
not in acid. Use of Acid electrophoresis eliminates all the other common
hemoglobin variants that migrate in the Hb S alkaline area or by CZE, IEF or
HPLC. Other identification methods do exist.
Individuals with Hb S should be advised that it is almost a benign and
2
innocuous condition . However, there are exceptions and in some individuals:

hematuria and aseptic necrosis of bone has been reported. If the hematuria
persists for a long time and is profuse, then the possibility of bladder cancer by
cystoscopy and bladder cancer markers must be evaluated.
3
Recently, a new sickling hemoglobin (Hb S-San Martin) was reported

from an Argentinean family. Besides the usual -globin chain mutation
associated with sickle cell [6(A3)GluVal, (GAGGTG)], an additional
mutation on the same -globin chain [105 (G7) LeuPro (CTCCCC) ] was
confirmed by the DNA studies. The electrophoretic mobility of Hb S-San Martin
at both the alkaline pH (8.6) and acid pH (6.2) was identical with the Hb S. This is a
rare occurrence and only ten (10) hemoglobin variants out of >1000 variants
discovered so far have double mutation on the same -globin chain besides the
sickle cell mutation.
References
285
1.
2.
3.
Bain BJ. Sickle cell haemoglobin and its interactions with other variant
haemoglobins and with thalassaemias. In: Bain BJ, Ed. Haemoglobinopathy
Diagnosis, 2nd edition, Blackwell Publishing; 2006:141-149.
Steinberg MH. Sickle cell trait. In: Steinberg MH, Forget BG, Higgs DR,
Nagel RC, eds. Disorders of Hemoglobin: Genetics, Pathophysiology and Clinical
Management. Cambridge, England: Cambridge University Press; 2001: 811-830.
Feliu-Torres A, Eberle SE, Bragos IM, Sciuccati G, Ojeda MJ, Calvo KL,
Voss ME, Pratti AF, Milani AC, Bonduel M, Diaz L, Noguera NI. Hb S-San Martin:
A new sickling hemoglobin with two amino acid substitutions
[6(A3)GluVal;105(G7)LeuPro]. Hemoglobin 2010; 34(5): 500-504.
Case # 3 Hemoglobin S homozygous

286
A 21 years old African American female came to the emergency department of a

hospital complaining abdominal and joint pain.
Laboratory Data:
Hemoglobin
7.6
RBC
2.6
MCV
80.0
RDW
21.0
Platelet
201
Hb A
4.2
Hb S
90.0
Hb A2
2.8
Hb F
3.0
(Hemoglobin fractions from HPLC)
12.0 16.0 g/dL

3
4.0 - 5.5 Mil/mm
78 - 98 fL
11.5 -14.5%
3
150 - 400 Th/mm
94.3 - 98.5%
1.5 - 3.7%
0.0 - 2.0%
2+ poly morphic, 1+ target cells, few HowellJolly bodies, sickle cells

Sickle cell solubility test for Hb S: Positive.

287
288
Note: The original CZE on this specimen showed no presence of Hb A, therefore the analysis was
repeated after mixing the specimen 1:1 with a normal blood. This is the standard practice in cases
whenever Hb A is not detected.

289
Summary of Results
Method
Hb A
area
Alk
Agarose
Acid Agar
/ Agarose
CZE
Hb S
area
Major
band
(Hb S)
Major
band
(Hb S)
Major
peak
(Hb S)
Zone 5
IEF
Major
band
(Hb S)
HPLC
Major
peak
(Hb S)
RT=4.38
Hb A2/C
area
Minor
band
(Hb A2)
Minor
peak
(Hb A2)
Zone 3
Minor
band
(Hb A2)
Minor
peak
(Hb A2)
RT=3.6
Minor
peak
(Hb F)
RT=1.04
Agarose gel electrophoresis (alkaline pH 8.6) and citrate agar

electrophoresis (acid pH 6.2) showed only one major band in the position of Hb
S. In view of the positive sickle cell solubility test a diagnosis of homozygous Hb
S disease was apparent.
It must be emphasized that due to co-migrating hemoglobin variants a
confirmatory discriminatory test must be run. The selection of these confirmatory
tests must be done with an eye on the results, for instance CZE would not be a
good confirmatory test for the identification of Hb S vs Hb Dhofar following
alkaline electrophoresis, because the migration is not different. Acid
290
electrophoresis will suffice as confirmation for either of them. Other readily

available tests that can be of use are HPLC and IEF.
If hemoglobinopathy testing is performed within three months of a blood
transfusion, the separation pattern will indicate the presence of Hb A from the
transfused blood and thus complicate the interpretation of the results. Therefore,
it is advised that all the laboratories obtain the blood transfusion history before
interpreting hemoglobin results.
Sometimes it is impossible to know the patient transfusion history,
especially if the patient arrived in the emergency department of the hospital.
About eight years ago, a very unusual case was observed by me in our hospital.
The Hb S diseased patient without insurance and facing sickle cell crisis went to
the emergency department of a large hospital in Detroit. The patient was
transfused with two units of blood and then discharged. He felt a little better after
blood transfusion, but two days later he went to the emergency department of
another large hospital in Detroit and received a second transfusion. Two days
later, this patient was examined in the emergency department of our hospital. In
our laboratory, the hemoglobin assays indicated Hb A (60%), Hb S (34%), Hb A2
(2.5%), and Hb F (3.5%). These results are suggestive of Hb S trait without
knowing the blood transfusion history of the patient. Therefore, in order to make
a correct diagnosis of a hemoglobin variant, it is prudent to know the recent blood
transfusion record.
References
1.
Kutlar A. Sickle Cell Disease: A Multigenic Perspective of a Single Gene

291
2.
3.
4.
5.
Disorder. Hemoglobin 2007; 31 (2): 209-224.

Steinberg MH. Genetic Etiologies for Phenotypic Diversity in Sickle Cell
Anemia. The Scientific World Journal 2009; 9: 46-67.
Bain BJ. Sickle cell anemia, In: Bain BJ, Ed. Hemoglobinopathy
Diagnosis, 2nd edition, Blackwell Publishing; 2006: 150-164.
Beutler E. The sickle cell diseases and related disorders. In: Beutler E,
Lichtman MA, Coller BS, Kipps TJ, Seligsohn U, eds. Williams
Hematology, 6th ed. New York, NY: McGraw-Hill; 2000: 581-606.
Nagel RC, Platt VS. General pathophysiology of sickle cell anemia. In:
Steinberg MH, Forget BG, Higgs DR, Nagle RL, eds. Disorders of
Hemoglobin: Genetics, Pathophysiology and Clinical Management.
Cambridge, England: Cambridge University Press; 2001: 494-526.
Case # 4
Hemoglobin S with hereditary persistence of

fetal hemoglobin (HPFH)
African-American adult male, apparently healthy and without any previously known major
clinical condition visited his family physician for his annual check-up.
292
Laboratory Data:
Hemoglobin
13.4
RBC
4.78
MCV
80.9
MCH
28.0
Hb A
5.7
Hb S
56.0
Hb A2
3.3
Hb F
35.0
Peripheral Blood Smear: No abnormality was noticed.
13.5 - 18.5 g/dL

3
4.6 - 6.2 Mil/mm
80 - 100 fL
27 - 34 pg
94.3 - 98.5
1.5 - 3.7%
0.0 - 2.0%
Case # 4 Hemoglobin S with hereditary persistence of fetal hemoglobin (HPFH)
293
294

295
Summary of Results
Metho
d
Alk
Agaros
e
Hb F
area
Major
band
(Hb F)
Acid
Agar /
Agarose
Major
band
(Hb F)
Major
peak
(Hb F)
Zone 7
Major
band
(Hb S)
Major
peak
(Hb S )
Zone 5
IEF
Major
band
(Hb F)
Major
band
(Hb S)
HPLC
Major
peak
(Hb F)
RT=1.16
Major
peak
(Hb S)
RT=4.38
CZE
Hb A
area
Hb S
area
Major
band
(Hb S)
Hb A2/C
area
Faint
band
(Hb A2)
Minor
peak
(Hb A2)
Zone 3
Faint
band
(Hb A2)
Minor
peak
(Hb A2)
RT=3.6
These laboratory results must have been somewhat of a surprise for an

asymptomatic patient. The few commonly encountered hemoglobins which
migrate in the position of Hb S on alkaline agarose gel electrophoresis are Hb D,
Hb G and Lepore all of which are ruled out by the results of acid agar gel
electrophoresis. In addition the positive Hb S solubility test assures this patient
has only Hb S and High Persistence of Fetal Hemoglobin. Analysis of this
specimen by CZE requires a modification of the procedure because there is no
hemoglobin A present for the software to use as a home base for comparison to
other hemoglobin mobilities. The CZE analysis would be repeated after mixing in
1:1 ratio with a normal blood specimen.
296
From all the five laboratory methods, three abnormalities are evident:
a
b
c
Absence of Hb A
Hb S (56%)
Hb F (35%)
The percentage of Hb S and Hb F suggests the following diagnostic possibilities:

i
ii
iii
Homozygous Hb S disease with failure to suppress Hb F production

Hb S-- thalassemia, with failure to suppress Hb F production
Hb S- HPFH due to a deletional mutation in the non S gene
Hb S HPFH patients are the result of a point mutation on one beta gene
forming Hb S and a deletion of the delta and beta area on the other gene
permitting the production of Hb F to continue. The Hb F expression will be 25
35%. This is a pancellar condition so every erythrocyte will contain Hb F as well
as Hb S and the damage caused by Hb S is not seen. Generally speaking
patients with Hb S-HPFH are clinically well, with a benign clinical course, little
evidence of hemolysis and without severe anemia. It is prudent to make a
clinical diagnosis based on all available resources. In this case other laboratory
data showed a positive sickle solubility test, a normal CBC, serum iron and
ferritin, and no other abnormalities except for some sickling. Consultation with the
physician indicated the patient was clinically well and certainly had not been
treated with hydroxyurea. This patient is presumed to be Hb S HPFH.
Approximately 1% of Homozygous S patients present with 5% or less Hb
F and these patients clinically do better than those without Hb F. Therefore much
has been done to increase the production of Hb F in homozygous S patients in
general. Degree of success of hydroxyurea treatment has been very variable for
297
unknown but at least to some extent genetic reasons. A patient with

homozygous Hb S disease may present with Hb F levels (15% - 30%) following
treatment with Hydroxyurea. Among the symptom ameliorating effects of
hydroxyurea is the apparent interference in the suppression of Hb F manufacture
and the production of nitric oxide. Since Hb F is higher in oxygen affinity than Hb
S and deoxyhemoglobin S polymerizes, its presence protects the cells from
sickling and other but not all symptoms of Sickle Cell Disease. The problem with
this type of fetal persistence is that it is not pancellular. Not all erythrocytes
contain Hb F even though the Hb F is elevated. Those cells without the Hb
F are not protected. That said as a result of several Clincal Trials including BABY
HUG several agencies have recommended use of hydroxyl urea for treatment of
Sickle Cell Disease [McGann PT, Ware RE. Hydroxyurea for sickle cell anemia:
What have we learned and what questions still remain? Curr Opin Hematol 2011;
18(3): 158-165].
In the unlikely circumstance that it was not known if the patient had been
treated with hydroxyl urea there is the possibility that he might have been a
homozygous patient who at the moment his blood was drawn was not very
symptomatic but his Hb F had been chemically altered. The two conditions could
be separated by doing a Kleihuer Betke acid elution test or flow cytometry
(monoclonal antibody agains - chains) for the study of pancellular vs
heterocellular distribution of the Hb F. An essentially homogeneous distribution
establishes the Hb S-HPFH diagnosis.
Hb S--thalassemia is also highly unlikely because of the clinical picture.
298
Patients with Hb S- -thalassemia even in the presence of Hb F would have a

thalassemic clinical picture. Hydroxyurea has been used for treatment of betathal patients with some success so the possibility exists that it might be helpful in
a case of Hb S - -thalassemia. Far less data exists on this treatment even
though it is known that the presence of Hb F lessens the clinical picture.
A few cases of clinical aberrations, e.g. minor joint or abdominal pains,
asceptic necrosis of bone, palpable spleen were reported in persons with Hb SHPFH (Fairbanks VF. Hemoglobinopathies and Thalassemias. New York, NY:
Brian C. Decker; 1980:136). Recently Whyte et al (see below reference # 5)
reported massive splenic infarction in an adolescent with Hb S-HPFH. Therefore
the condition is not benign.
References
1. Murray N, Serjeant BE, Serjeant GR. Sickle cell-hereditary persistence of
fetal hemoglobin and its differentiation from other sickle cell syndromes. Br J Haemotol
1988; 6: 89-92. (available online since March 2008).
2. Hoyer JD, Connie SP, Fairbanks VF, Hanson CA, Katzmann JA. Flow cytometric
measurement of hemoglobin F in RBCs: Diagnostic usefulness in the distinction of
hereditary persistence of fetal hemoglobin (HPFH) and hemoglobin S-HPFH from other
conditions with elevated levels of hemoglobin F. Am J Clin Pathol 2002; 117: 857-863.
3. Akinsheye I, Al-Sultan A, Solovieff N, Ngo D, Baldwin CT, Sebastiani P, Chui DH,
Steinberg MH. Fetal hemoglobin in sickle cell anemia.Blood 2011; 118: 19-27.
4. Ngo D, Aygun B, Akinsheye I, Hanjins JS, Bhan I, Luo HY, Steinberg MH, Chui DH.
Fetal haemoglobin levels and haematological characteristics of compound
hegterozygotes for haemoglobin S and deletional hereditary persistence of fetal
hemoglobin. Br J Haematol 2012; 156(2): 259-64.
5. Whyte D, Forget BG, Chui DH, Luo HY, Pashankar F. Massive splenic infarction in
an adolescent with hemoglobin S-HPFH. Pediatr Blood Cancer 2013; 60(7): 49-51.
6. Chapter 2.3 of this book: Bernard G. Forget, MD. Hereditary Persistence of Fetal
Hemoglobin
299
7. Bain BJ. Hereditary persistence of fetal haemoglobin and other inherited causes of
an increased proportion of haemoglobin F. In: Haemoglobinopathy Diagnosis, Blackwell
Publishing, second edition, 2006, pp119-127.
Case # 5 Hemoglonin G-Philadelphia trait

Adult African-American male with no abnormalities.
Laboratory Data:
Hemoglobin
RBC
14.5
5.06
13.5 -18.5 g/dL

3
4.6 - 6.2 Mil/mm
300
MCV
84.0
80 - 100 fL
MCH
28.7
27 - 34 pg
Hb A
77.1
94.3 - 98.5%
Hb A2
0.9
1.5 - 3.7%
Hb F
0.0
0.0 - 2.0%
Hb G
22%
Peripheral Blood Smear: No abnormality was detected.
Sickle cell solubility test: Negative.
Case # 5 Hemoglobin G-Philadelphia trait
301

302

Summary of Results
303
Method
Hb A
area
Hb S
area
Alk Agarose
Major
band
(Hb A)
Major
band
(Hb G)
Acid
Agar/Agaro
se
Major
band
(Hb A+
Hb A2 +
Hb G+
Hb G2)
Major
peak
(Hb A)
Zone 9
Major
peak
(Hb G)
Zone 6
CZE
IEF
HPLC
Major
band
(Hb A)
Major
band
anodal
to Hb S
(Hb G)
Major
peak
(Hb A)
RT=2.4
5
Major
peak
(Hb G)
RT=4.0
4
Hb
A2/C
area
Minor
band
(Hb A2)
Minor
band
close to
carbonic
anhydras
e
Minor
peak
(Hb A2)
Zone 3
Minor
band
(Hb A2)
Minor
peak
(Hb G2)
Zone 1
Minor
peak
(Hb A2)
RT=3.6
Minor
peak
RT=4.54.6
Minor band
(Hb G2)
as far
cathodal to
A2 as G is
anodal to it.
* Note: HPLC retention time (RT) varies with the type of the instrument used and
several other factors, e.g. temperature etc.
Electrophoretic migration bands at less than 1% may be difficult to detect

on alkaline electrophoresis. If the presence of a minor band is expected the
sample amount may be increased. Quantification of the results must not be done
on over applied samples because you will have exceeded the quantitative
linearity of the system.
304
The sickle cell solubility test was negative, ruling out the possibility of Hb
S. The separation table shows the presence of a non Hb S band migrating in the
Hb S area for all methods except Acid Electrophoresis. Common variants found
anodal to Hb S on alkaline electrophoresis that migrate in Hb A position on acid
electrophoretic conditions are Hb D, Hb G-Philadelphia and Lepore. Of these
options only Hb G- Philadelphia is an -chain variant. If -chain variant is
expressed in a large enough percentage to compete with normal -chains for
combination with chains a new small modified delta band is created. In
individuals with Hb G-Philadelphia [68(E17)AsnLys], a combination of Hb GPhiladelphia -chains with normal -chains leads to the formation of about 1% of
a molecule Hb G2 (2G2). Hb G2 has no clinical significance, but plays an
important role in the distinction between Hb D and Hb G-Philadelphia. Since Hb
G-Philadelphia is entirely innocuous, globin chain electrophoresis and DNA
studies are usually not necessary.
There is a temptation to analyze the available hemoglobin variants by
percentage since Hemoglobin Lepore runs less than 15 % and Hb D runs about
40 while Hb G-Philadelphia trait runs 20-25% in the heterozygote. Differentiation
between Hb D and Hb G-Philadelphia on the basis of the percentage of the
variant is not advised because the percentages of either would be effected by a
concurrent -thalassemia -2 trait or homozygous -thalassemia-2 (see below).
The single alpha gene deletion resulting in -thalassemia-2 trait is found in 1/3 of
African Americans therefore this silent mutation could be likely found in
305
association with Hb G-Philadelphia in this ethnic population.

Of the four alpha genes located on chromosome 16 (two on each
chromosome), alpha gene mutations lead to the following possibilities (adopted
with the permission of College of American pathologists: Hoyer JD and Kroft SH,
eds. Color Atlas of Hemoglobin Disorders. College of American Pathologists,
Northfield, IL, 2003; 67).
1
Hb G trait with no thalassemia, Hb G 20 -25% no hematologic effect
Hb G trait, One gene deleted (-thalassemia-2 trait), Hb G 25-35% usually

no hematologic effect
Hb G-trait; Two genes deleted (homozygous -thallasemia-2), Hb G 3545%; microcytosis.
Homozygous Hb G; Two genes deleted (homozygous - thalassemia-2), Hb

G 95%, microcytosis.
References
1.
2.
3.
4.
5.
6.
Keren DF. Clinical Evaluation of Hemoglobinopathies: Part II. Structural

Changes, Ward Medical Laboratory, Archived Issues 2003; 3: 1-11.
Available online (http://www.wardlab.com/14-3.html).
Hoyer JD and Kroft SH, eds. Color Atlas of Hemoglobin Disorders.
College of American Pathologists, Northfield, IL, 2003; 65.
Bain BJ. Hemoglobin G-Philadelphia trait: In: Haemoglobinopathy
Diagnosis, Blackwell Publishing, second edition, 2006, pp 212.
Milner PF, Huisman TH. Studies of the proportion and synthesis of
haemoglobin G-Philadelphia in red cells of heterozygotes, a homozygote,
and a heterozygote for both haemoglobin G and alpha thalassemia. Br J
Haematol 1976; 34: 207-220. (Available online from July 2008).
Baine BM, Rucknagel DL, Dublin DA Jr, Adams JG III. Trimodality in the
proportion of hemoglobin G-Philadelphia in heterozygotes; evidence of
heterogeneity in the number of human alpha chain locations. Proc Natl
Acad Sci. 1976; 73: 3633-36.
Reider RF, Woodbury DH, Rucknagel DL. The interaction of 306
7.
thalassemia and hemoglobin G-Philadelphia. Br. J Haematol. 1976; 32:

159-65.
Khalil MSM, Timbs A, Hendrson S, Schuh A, Hussein MRA, Old J.
Haemoglobin (Hb) G-Philadelphia, Hb Stanleyville-II, Hb G-Norfolk, Hb
Matsue-Oki and Hb Mizushi can form a panel of -chain variants that
overlap in their phenotype: the novel use of StyI to screen for Hb GPhiladelphia. Intl Jnl Lab Hem 2011; 33: 318-325.
Case # 6 Hemoglobin S-G Philadelphia

Adult African American female who was asymptomatic.
Laboratory Data:
307
Hemoglobin
11.7
RBC
4.29
MCV
81.6
MCH
27.4
Hb A
54.0
Hb S
19.9
Hb G
17.8
Hb A2
1.1
Hb F
0.2
Hb S-G Hybrid
7.0
Peripheral Blood Smear: No abnormality.
Sickle cell solubility test for hemoglobin S: Positive.
Unstable hemoglobin (isopropanol) Test: Negative.
No record of blood transfusion during the past six months.
308
12.0 -16.0 g/dL

3
4.0 5.5 Mil/mm
79 - 98 fL
27- 34 pg
94.3 98.5%
1.5-3.7%
0.0-2.0%

309

310
Summary of Results
Method
Alk
Agarose
Acid Agar
/ Agarose
CZE
Hb A
area
Major
band
Hb A
Major
band
(Hb A +
G)
Major
peak
(Hb A)
Zone 9
IEF
HPLC
*Note:
Medium
peak
(Hb A)
RT=2.35
Hb S
area
Major
band
Hb S +
G
Major
band
(Hb S)
Major
peak
Zone 6
Major
peak
poorly
separated
from Zone
6 in Zone
5
Major Hb
G band
Anodal
to Hb S
Major
band
(Hb S)
Minor
peak
(Hb A2)
RT=3.58
Hb A2/C
area
Minor
band
(Hb A2)
Minor
Hb A2
peak
Zone 3
Medium
peak
(Hb G)
RT=4.0
Minor
band
cathodal to
A2
Major Hb
GS
hybrid
peak
Zone 2
Minor
Hb G
A2
peak
Zone 1
Medium
band (Hb
S-G
hybrid +
A2 )
Minor
Hb G2
Band
Medium
peak
(Hb S)
RT=4.24
Medium
peak
(Hb S-G
hybrid)
RT=4.8
HPLC retention time (RT) varies with the type of the instrument used and
Agarose gel electrophoresis (pH 8.6) showed three major bands in the
familiar positions of Hb A ( 50%), Hb S ( 38%), and Hb A2/C (>10%) and a
barely visible minor band slightly cathodal to the carbonic anhydrase position. It
should be emphasized here that Hb A, Hb S, and Hb C cannot all be
manufactured in any single person because there are only 2 beta genes and
these hemoglobins represent three different beta compositions. Either this patient
311
had a transfusion or one of the hemoglobin variants is not a beta chain variant.
The transfusion could be to a patient with Hb S and C or a transfusion using
blood from an Hb A-S heterozygous donor or an Hb A-C heterozygous donor to a
patient who was a heterozygote of the other type. These unlikely scenarios were
all ruled out as the patient received no blood transfusion.
Beta-thalassemia in conjunction with Hb A-S trait can result in an elevated
Hb A2 which migrates with or near Hb C by most of these methods. In S-thalassemia the Hb A2 is rarely higher than 10% so a >10% band is unlikely
Hb A2.
Secondly in S--thalassemia patients Hb F concentration is often
increased especially if the patient is thalassemic to the point that the Hb A2 is
very elevated but in this patient the Hb F was normal ( 0.2%).
The identity of the small barely visible minor band is the key to the
identification. The most common alpha chain variant is Hb G-Philadelphia which
would present in the Hb S area at 30 to 35%. This alpha chain variant then
competes with the unmodified alpha chains to combine with the beta and delta
chains available. Since the sickle solubility test was positive we know the band in
the position of Hb S is indeed at least partly due to the S beta gene combined
with normal alpha chains. This Hb S beta gene when combined with a modified
Hb G-Philadelphia alpha gene creates a new double hemoglobin variant
combination, Hb S-G Philadelphia hybrid which unfortunately migrates with Hb A2
on alkaline, acid or IEF electrophoresis. This explains the elevated Hb A2.
312
If
half of the alpha chains are modified they would be competing also with the
unmodified alpha chains for delta chains. The unmodified alpha chain delta
combination is Hb A2 seen normally and the modified alpha variant delta
combination is new hemoglobin, Hb G2 which migrates close to the carbonic
anhydrase. The number of different hemoglobin molecules created by a Hb S GPhiladelphia double mutation is 6. The Hb S-G hybdrid migrates with A2 on
acid, alkaline and IEF electrophoresis.
IEF, CZE and HPLC data support the presence of a heterozygous Hb GPhiladelphia [68(E17)AsnLys] and Hb S in that two distinct, approximately
equal bands or peaks were seen in the position of Hb S and Hb G. IEF indicated
that the Hb G band is closer to the Hb A band (more anodal) than the Hb S. Two
additional bands in the position of Hb A2 and Hb G2 were also detected from IEF
although the low intensity of the Hb G2 band made it difficult to see.
CZE showed six distinct peaks in the following zones with alleged
hemoglobins indicated in parenthesis:
i
ii
iii
iv
v
vi
Zone 9 (Hb A)
Zone 6 (Hb G-Philadelphia)
Zone 5 (Hb S)
Zone 3 (Hb A2)
Zone 2 (Hb S/G hybrid)
Zone 1 (Hb G2)
HPLC showed the following major peaks:

a
b
Hb F ( 0.2%)
Hb A (54%; RT = 2.35)
313
c
d
e
f
Hb A2 (1.1%, RT= 3.58)

Hb G (17.8%, RT= 4.0)
Hb S (19.9%, RT = 4.24)
Hb S/G hybrid (7%, RT=4.8)
All the data affirm the presence of a double heterozygous presentation of

an abnormal chain (Hb S) and an abnormal chain (Hb G-Philadelphia) in
A
conjunction with normal and chains ( and ) found in Hb A. The abnormal

chains end up competing with their normal counterparts creating all the possible
combinations listed below.
A
A
2 2)
G
A
Hb G ( 2 2)
A
HB A2 ( 2 2)
Hb A (
S
2 2)
G
S
S/G ( 2 2)
G
G2 ( 2 2)
Hb S (
Hb
Hb
Had this patient been a newborn the situation would further have been
complicated by the addition of 2 new gamma chain containing forms of HbF.
Hb S-G Philadelphia double heterozygous hemoglobinopathies are
essentially healthy and without anemia.
References
1
Kirk CM, Papadea CN, Lazarchik J. Laboratory Recognition of a Rare

Hemoglobinopathy. Hemoglobin SS and SGPhiladelphia Associated with Thalassemia -2. Arch Pathol Lab Med 1999; 123: 963-966.
Gu LH, Wilson JB, Molchanova TP, McKie KM, Huisman THJ. Three Sickle
Cell Anemia Patients each with a Different Chain Variant. Diagnostic
Complications. Hemoglobin 1993; 17(4): 295-301.
Kutlar F, Kutlar A, Nuguid E, Prachal J, Huisman. Usefulness of HPLC
Methodology for the Characterization of Combinations of the Common Chain variants Hb S, C, and O-Arab, and the Chain variant in GPhiladelphia. Hemoglobin 1993; 17(1):, 55-66.
314
4
5
LeCrone CN, Jones JA, Detter JC. Hemoglobin G Trait and S Trait in the
Same Patient. Hemotology 1983; 49(3): 165-167.
Lawrence C, Hirsch RE, Fataliev NA, Patel S, Fabry ME, Nagel RL. Molecular
interactions between Hb alpha-G Philadelphia, Hb C, Hb S: phenotypic
implications for SC -G Philadelphia disease. Blood 1997; 90: 2819-2825.
Case # 7 Hemoglobin G-Coushatta trait

A 24 year old male resident of Cheyenne River Indian Reservation, South Dakota, USA. No
physical abnormality. Blood sent to a reference laboratory for hemoglobin electrophoresis.
315
Laboratory Data:
Hemoglobin
RBC
MCV
RDW
Platelet
12.7
4.49
81
13.2
243
Hb A
Hb A2
Hb F
Hb variant
56.0
2
1
41.0%
13.5-18.5 g/dL
3
4.6-6.2 Mil/mm
80-100 fL
11.5-14.5%
3
150-400 Th/mm
94.3-98.5%
1.5-3.7%
0.0-2.0%

Peripheral Blood Smear: No abnormality noticed.
Sickle cell solubility test for Hb S: Negative
Unstable hemoglobin (isopropanol) test: Negative.
316

317

318
Summary of Results
Method
Hb A
area
Alk
Agarose
Major
band
(Hb A)
Acid
Agar /
Agarose
Major
band
(Hb A+
Hb A2 +
Hb G)
Major
peak
(Hb A)
Zone 9
CZE
IEF
Major
band
(Hb A)
HPLC
Minor
peak
(Hb F)
RT=1.0
5
Hb S
area
Major
band
Major
peak
( Hb G )
Zone 6
Major
Hb G
band
anodal to S
Major
peak
(Hb A)
RT=2.5
Hb
A2/C
area
Minor
band
(Hb A2)
Minor
peak
(Hb A2)
Zone 3
Minor
band
(Hb A2)
No G2
band
was
detected
Major
peak
(Hb G +
Hb A2)
RT=3.6
*Note: HPLC retention time (RT) varies with the type of the instrument used and several
other factors, e.g. temperature etc.
Agarose gel electrophoresis (pH 8.6) indicated major bands in the position
of Hb A and at the position of Hb S. Besides a minor band at the position of Hb
A2 and carbonic anhydrase band no other band was detected. Citrate agar
electrophoresis (pH 6.2) showed one major band at the position of Hb A, and a
faint band was also detected in the position of Hb F. CZE showed major peaks
in Zone 9 (Hb A), and Zone 6 (Hb variant) and a minor peak in Zone 3 (Hb A2).
IEF indicated that the second major band was in the position of Hb G, or
319
possibly Hb D but not Hb S, however a Hb G 2 (2 2) band was not detected.

HPLC showed two major peaks at the position of Hb A and Hb A2 rather than
one toward the center of the pattern as seen with all the alkaline electrophoretic
separations (pH 8.6). The tentative identification of the Hb variant (41%
concentration from alkaline agarose gel electrophoresis at pH 8.6) was achieved
by eliminating commonly encountered hemoglobin variants (e.g. Hb S, Hb GPhiladelphia, Hb Lepore, Hb Hasharon, etc) on the basis of the laboratory
results.
Hb S was also ruled out by a normal sickle cell solubility test. The most
commonly noticed Hb G variant (Hb G Philadelphia) is noticed mostly in African
Americans. The presence of this -chain variant was ruled out because the minor
G
Hb G2 (2 2) band was not detected by IEF or by agarose gel electrophoresis

(pH 8.6) and because alpha chain variants are found in a lower percentage than
-chain variants. Hb Hasharon and Hb Lepore are also ruled out on the basis of
low concentration. Furthermore Hb Lepore produces a thalassemic picture
including microcytosis, and that was not exhibited in this case. Hemoglobin
variant of 41% is extremely high for Hb Hasharon and Hb Lepore. Hemoglobins
D-Los Angeles and Hb G- trait are closely migrating variants with no clinical
manifestation. Generally speaking they are found in different ethnic groups.
The safest interpretation for this case is that this patient has Hb G trait (-chain
variant ) known as Hb G-Coushatta[ 22 (4) GluAla (GAAGCA)] because of the
320
(American Indian) ethnicity. It is emphasized that Hb G-Coushatta is not limited to

American Indian tribes, and this hemoglobinopathy also know as Hb G-Saskatoon, Hb
G-Taegu, or Hb G- Hsin Chu, has been reported in Chinese, Korean, Japanese, Thai,
Turkish, and Algerian nationals and is harmless.
Homozygous Hb G-Coushatta is very rare and exhibits microcytosis.
Recently a compound heterozygote for Hb E and Hb G-Coushatta was reported
in a Thai family by amplification refractory mutation system-polymerase chain
reaction (ARMS-PCR). It may not be worth the cost to further solidify the identity
of the hemoglobin variant in a situation like this where the variant is functioning
normally.
References
1
2
3
4
5.
6
Worrawut C, Viprakasit V. Further identification of Hb G-Coushatta

[22(4) GluAla (GAAGCA)] in Thailand by the polymerase chain
reaction-single-strand conformation polymorphism technique and by
amplification refractory mutation system-polymerase chain reaction.
Hemoglobin 2007; 31(1): 93-99.
Ohba Y, Miyaji T, Hirosaki T, Matsuoka M, Koresawa M, Iuchi I. Occurrence of
Hemoglobin G Coushatta in Japan. Hemoglobin 1978; 2(5): 437-441.
Wong SC, Tesanovic M, Poon M-C. Detection of two abnormal hemoglobins,
Hb Manitoba and Hb G-Coushatta, during analysis of glycohemoglobin (A 1c)
by high performance liquid chromatography. Clin Chem 1991; 38(8): 14561459.
Li J, Wilson D, Plonczynski M, Harrell A, Cook CB, Scheer WD, Zeng Y-T,
Coleman MB, Steinberg MH. Genetic studies suggest a multicentric origin for
Hb G-Coushatta [22(4)GluAla]. Hemoglobin 1999; 23(1): 57-67.
Boissel JP, Wajcman H, Labie D, Dahmane M, Benabadji M.
[Hemoglobin G-Coushatta (beta 22(4) glu leads to ala) in Algeria: an
homozygous case]. Nouv Rev Fr Hematol 1979; 21:225-230.
Dincol G, Dincol K, Erdem S. Hb G-Coushatta or alpha 2 beta 22 (4)
GluAla in a Turkish male. Hemoglobin 1989; 13: 75-77.
Case # 8 Hemoglobin C trait

A 28 year old African American male. No physical abnormalities. Participated regularly in
basketball and never complained about fatigue.
321
Laboratory Data:
Hemoglobin
RBC
MCV
RDW
Platelet
14.8
4.91
77
15.1
248
13.5-18.5 g/dL
3
4.6-6.2 Mil/mm
80-100 fL
11.5-14.5%
3
150-400 Th/mm
Hb A
58.0
Hb A2
2
Hb F
1
Hb variant
39.0%
94.3-98.5%
1.5-3.7%
0.0-2.0%
Peripheral Blood Smear: 1 microcytosis and numerous target cells.

Sickle cell solubility for Hb S: Negative
Unstable hemoglobin (isopropanol) test: Negative
322
323

Summary of Results
324
Method
Alk
Agarose
Acid Agar
/Agarose
CZE
IEF
HPLC
Hb A
area
Major
band
(Hb A)
Major
band
(Hb A+
Hb A2)
Major
peak
(Hb A)
Zone 9
Hb S
area
Hb A2/C
area
Major
band
Major
band
Minor
peak
(Hb A2)
Zone 3
Major
peak
(Hb C)
Zone 2
Major
band
(Hb A)
Minor
band
(Hb A2)
Major
peak
(Hb A)
RT=2.45
Minor
peak
(Hb A2)
RT=3.6
Major
band
cathodal
to A2
(Hb C)
Major
Peak
(Hb C)
RT=5.l0
Agarose gel electrophoresis (pH 8.6) exhibited a major band in the

position of Hb A, and another intense band (40%) in the position of Hb C/Hb E/
Hb O-Arab/ Hb A2. The intense band is not due to Hb A2 only in view of
the fact that the concentration of Hb A2 is never > 10%. Citrate agar
electrophoresis (pH 6.2) indicated two bands. One band was in the position of Hb
A and another band in the position of Hb C. Hb E, Hb O-Arab, and Hb C-Harlem
are ruled out on the basis of citrate agar electrophoresis (pH 6.2), as none of
these migrate in the position of Hb C by this method. Combination of alkaline and
325
acid pH electrophoresis suggested that the Hb variant is most likely Hb C. IEF

also indicated two major bands in the position of Hb A and Hb C. CZE also
indicated two major peaks in Zone 9 (Hb A) and Zone 2 (Hb C). HPLC results
were concordant with above stated observations from IEF and CZE, i.e. one
major peak eluted in the position of Hb A (retention time 2.45 minutes) and the
second major peak eluted in the C-window (retention time 5.10 minutes).
+
The peripheral blood smear examination (1 microcytosis and target cells),

negative for sickle cell solubility and hemoglobin instability tests, and the five laboratory
tests led towards the assignment of the Hb variant as Hb C. In order to be Hb C
trait the percentage of Hb C should be less than Hb A, therefore the diagnosis of
Hb C trait was made.
Hb C is a -chain variant [6 (A3) GluLys], caused by the substitution of
glutamic acid by lysine in the sixth position. Hb C trait is prevalent in 2-3% in
African Americans, and rarely found in other ethnic groups. Clinically the Hb C
trait phenotype is insignificant.
References
1
Bain BJ. Hemoglobin C trait: In: Haemoglobinopathy Diagnosis, Blackwell

Publishing, 2nd edition, 2006, pp 192-195.
326
2
3
4
5
Wajcman H, Moradkhani K. Abnormal haemoglobins: detection &

characterization. Indian J Med Res 2011; 134: 538-546
Joutovsky A, Nardi M. Hemoglobin C and Hemoglobin O-Arab variants can be
diagnosed using the Bio-Rad Variant II High Performance Liquid
Chromatography System without further confirmatory tests. Arch Pathol Lab
Med 2004; 128: 435-439.
Joutovsky A, Hadzi-Nesic J, Nardi MA. HPLC retention time as a diagnostic
tool for hemoglobin variants and hemoglobinopathies: A study of 60 000
samples in a clinical diagnostic laboratory. Clin Chem 2004; 50: 1736-1747.
Keren DF, Hedstrom D, Gulbranson R, Ou Ching-Nan, Richard B. Comparison of
Sebia Capillary Electrophoresis with the Primus High-Pressure Liquid
Chromatography in the evaluation of hemoglobinopathies. Am J Clin Pathol 2008;
130: 824-831
Case # 9
Hemoglobin C homozygous
African-American male (22 years old) with no physical complaints.
Laboratory Data:
327
Hemoglobin
12.1
RBC
4.3
MCV
73
RDW
13.3
Platelet
248
Hb A
Not detected
Hb A2
2.5
Hb F
1.6
Hb variant
95.9%
13.5 - 18.5 g/dL

3
4.6 - 6.2 Mil/mm
80 -100 fL
11.5 - 14.5%
3
150 - 400 Th/mm
94.3 - 98.5%
1.5 - 3.7%
0.0 - 2.0%
Peripheral Blood Smear: Target cells, spherocytes, and poikilocytosis.

Sickle cell Hb S solubility test: Negative
Case # 9 Hemoglobin C homozygous
328

329

330
Summary of Results
Method
Hb A
area
Hb S
area
Alk
Agarose
Acid Agar
/Agarose
CZE
IEF
HPLC
Hb A2/C
area
Major
band
Major
band
Minor
peak
(Hb A2)
Zone 3
Minor
band
Minor
peak
(Hb A2)
RT=3.6
Major
peak
(Hb C)
Zone 2
Major
band
cathodal
to A2
(Hb C)
Major
peak
(Hb C)
RT=5.06
Agarose gel electrophoresis (pH 8.6) showed only one intense and major band in
the position of Hb C/E/O, and Hb A was not detected. Citrate agar electrophoresis (pH
6.2) also showed one intense band in the position of Hb C, therefore at the very outset
the presence of Hb E and O were ruled out. It appeared that the solitary band in the Hb C
position is most likely due to the substitution of amino acid lysine with glutamic acid at
the sixth position of -chain [ 6(A3) GluLys]. Hb C has prevalence of 0.017% among
the African-Americans in the United States, but it has been also reported in persons of
Hispanic and Sicilian ancestry.
331
Other laboratory tests (CZE, HPLC, and IEF) also indicated the prominent Hb C
band or peak, however contrary to alkaline and acid electrophoresis (see above) minor
bands or peaks due to Hb F ( 1.6%) and Hb A2 ( 2.5%) were also detected. Absence
of Hb A by all the five methods in this person suggested either homozygous Hb C or Hb
+
C/0-thalassemia and ruled out Hb C/ -thalassemia (Case # 19).

The clear distinction between homozygous Hb C and Hb C/ 0-thalassemia
(double heterozygous state for both Hb C and 0-thalassemia) is problematic,
because the clinical features are similar in both cases. Careful evaluation of
peripheral blood smear, CBC, anemia status, quantitative values of Hb F and Hb
A2, and evaluation of hemoglobinopathy in the biological parents are helpful for
the exactness of the diagnosis.
Fairhurst and Casella reported a diagnosis of homozygous Hb C disease
in a Ghanian child [N Engl J Med 2004; 350(26): e24], with hemoglobin (9.0
g/dL),HCT (24.3), MCV (53.8), RDW (28.8), and an uncorrected reticulocyte
count of 1.6%. The peripheral blood smear (Figure 1) indicated characteristic
features of homozygous Hb C: target cells (arrows), microspherocytes
(arrowheads), rod-shaped cells containing hemoglobin C crystals (asterisk),
anisocytosis, and poikilocytosis. Schwab and Abelson [N Engl J Med 2004;
351(15): 1577] questioned the diagnosis of homozygous Hb C on the basis of
extremely low MCV and the clinical status of the child, and suggested the
diagnosis of Hb C/0-thalassemia.
332
Figure 1. Peripheral blood smear of the Ghanaian child

(adopted with the permission of the N Engl J Med)
The following characteristics are helpful in the differential diagnosis

between the two possibilities:
Test
Hb A2
Hb C/0-thalassemia
Homozygous Hb C
3.2 3.9%
Elevated in most cases
Hb F
0.8 1.9%
3 10% (generally > 5%)
MCV
68 - 76
55 - 70
On the basis of Hb A2 ( 2.5%), Hb F ( 1.6), MCV (73), mild anemia, a

333
tentative diagnosis of homozygous Hb C is reasonable, however for confirmation,

additional tests in the biological parents are mandatory.
Persons with homozygous Hb C rarely have clinical symptoms and live a
normal life. Symptoms that may develop in these persons include:
Reduced red blood cell counts during infection or illness

jaundice
Increased risk for gallstones
Enlarged spleen
Episodes of pain
Increased risk for infection
Hemoglobin C is known to protect individuals against clinical Plasmodium

falciparum malaria.
References
1
2
3
4
5
6.
7.
8.
Bunn HF, Forget BG, Hemoglobin: molecular, genetic and clinical aspects. 1 st
edition, Philadelphia, PA: WB Saunders Co; 1986: 421-425.
Nagel RL, Steinberg MH. Hb S/C disease and Hb C disorders. In: Steinberg
MH, Forget BG, Higgs DR, Nagle RL, eds. Disorders of Hemoglobin:
Genetics, Pathophysiology and Clinical Management. Cambridge, England:
Cambridge University Press; 2001: 756-785.
Fairhurst RM, Casella JF. Homozygous hemoglobin C disease. N Engl J Med
2004; 350: e24 (Web only).
(Available at www.nejm.org/cgi/content/full/350/26/e24).
Schwab JG, Abelson HT. Hemoglobin C. N Engl J Med 2004; 351(15): 1577.
Weatherall DJ, Clegg JB. The thalassemia syndrome, 4th edition, Oxford,
England: Blackwell Science, 2001: 415-419.
Modiano D, Luoni G, Sirima BS, et al. Hemoglobin C protects against
clinical Plasmodium falciparum malaria. Nature 2001; 414 (6861): 305-8.
[Medline].
Rihet P, Flori L, Tall F. Hemoglobin C is associated with reduced
Plasmodium falciparum parasitemia and low risk of mild malaria. Hum
Mol Genet 2004; 13(1): 1-6.
Hoyer JD, Kroft SH. Color Atlas of Hemoglobin Disorders. College of
American Pathology 2003. Case # 8 (pp 45), Case # 15 (pp 75), Case
# 29 (pp 135), Case # 30 (pp 139).
Case # 10
Hemoglobin C with hereditary persistence

of fetal hemoglobin (HPFH)
334
A 23 years old white female presented to the Emergency Department of the hospital
(2011) complaining of pelvic pain. She was found to have a ruptured right hemorrhagic
ovarian cyst which was suspected on CT and ultrasound and then confirmed by
laparoscopy. No blood transfusion was executed.
Laboratory Data:
Hemoglobin
11.9
12.0 -16.0 g/dL
3
RBC
4.8
4.0 - 5.5 Mil/mm
MCV
74
79 - 98 fL
RDW
20.8
11.5 -14.5%
Hb A
Not detected
94.3 - 98.5%
Hb A2
2.2
1.5 - 3.7%
Hb F
29.4
0.0 - 2.0%
Hb variant
68.4%
Peripheral Blood Smear: Abundant target cells
Sickle cell solubility test for hemoglobin S: Negative
Flow cytometry (monoclonal antibody for Hb F) showed a homogeneous distribution of
Hb F.
Case # 10 Hemoglobin C with hereditary persistence of fetal hemoglobin (HPFH)
335

336

337
Note: HPLC and hemoglobin electrophoresis tests were performed at three independent
laboratories, and all the results were concordant.
Method
Alk
Agarose
Acid Agar
/Agarose
CZE
IEF
HPLC
Hb A
area
Hb S
area
Major
band
in Hb F
area
Major
Band in
Hb F
area
Major
peak Hb
F
Zone 7
Hb A2/C
area
Major
band
Major
band
Very
minor
peak
(Hb A2)
Zone 3
Minor
band
(Hb A2)
Major
band
in
Hb F
area
Major
peak
(Hb F)
RT=1.1
5
Very
minor
peak
(Hb A2)
RT=3.6
Major
peak
(Hb C)
Zone 2
Major
band
cathodal
to A2
(Hb C)
Major
peak
(Hb C)
RT=5.14
*Note: HPLC retention time (RT) varies with the type of the instrument and several other
factors, e.g. temperature etc.
Agarose gel electrophoresis (pH 8.6) indicated the absence of Hb A and

the presence of two major bands. One major band was detected in the position
of Hb F ( 29%) and another major band ( 68%) was detected in the position of
Hb C/E/O. Hb E and O were ruled out on the basis of citrate agar electrophoresis
(pH 6.2), as only two major bands were detected in the position of Hb C and
Hb F. IEF, CZE, and HPLC also confirmed the presence of only two major
338
hemoglobins (Hb C and Hb F) in this patient.

This suggested two possibilities, a) heterozygosity for Hb C or b)
heterozygosity for a deletional form of hereditary persistence of fetal hemoglobin
(HPFH). The presence of Hb C > 50% also suggested the presence of HPFH.
Hb C with hereditary persistence of fetal hemoglobin is the diagnosis of
this patient. Generally speaking homozygous Hb C disease (Case # 9) is rare
and is associated with abundant target cells, microcytosis, reticulocytosis, and
minimal hemolytic disease. Contrary to this, Hb C with HPFH is clinically similar
to Hb C trait (Case # 8).
References
1
Bain BJ. Hereditary persistence of fetal hemoglobin and other inherited causes
of an increased proportion of hemoglobin F: In: Hemoglobinopathy Diagnosis,
Blackwell Publishing, 2nd edition, 2006, pp 119-127.
2 Bollekens JA, Forget BG. thalassemia and hereditary persistence of fetal
hemoglobin. Hematol Oncol Clin North Am. 1991; 5: 399-422.
3 Hoyer JD, Penz CS, Fairbanks VF, et al. Flow cytometric measurement of
hemoglobin F in RBCs: diagnostic usefulness in the distinction of hereditary
persistence of fetal hemoglobin (HPFH) and hemoglobin S-HPFH from other
conditions with elevated levels of hemoglobin F. Am J Clin Pathol 2002; 117: 857863.
4 Weatherall DJ, Legg JB. Hereditary persistence of fetal hemoglobin. In: The
thalassemia Syndromes. 4th ed. Oxford: Blackwell Science, 2001: 450-484.
5 Wood WB. Hereditary persistence of fetal hemoglobin and thalassemia. In:
Steinberg MH, Forget BG, Higgs DR, Nagel RL. Disorders of Hemoglobin:
339
Genetics, Pathophysiology, and Clinical Management, Cambridge, England:

6 Pissard S, Mrad A, Beuzard Y, Romeo PH. A new type of hereditary persistence
of fetal hemoglobin (HPFH): HPFH Tunisia beta + (+C-200) G gamma. Br J
Haematol 1996; 95(1): 67-72.
7 Martin AW, Lippmann SB, Keeling MM, Lynch JA, Martinez M. Hemoglobin C in
association with hereditary persistence of fetal hemoglobin. Postgrad Med 1987;
81(8): 133-37.
Case # 11
Hemoglobin S-C disease
340
A 22 year old African American male, who was working at the Chrysler Stamping plant,
complained of headache and difficulty in breathing. His supervisor suspected carbon monoxide
poisoning and sent him to the Emergency Department.
Laboratory Data:
Hemoglobin
10.8
RBC
3.6
MCV
90.1
MCH
30.0
Hb A2
2.4
Hb F
1.8
Hb Variant-1
49.0%
Hb Variant-2
46.8%
13.5 - 18.5 g/dL

3
4.6 - 6.2 Mil/mm
80 - 100 fL
27 - 34 pg
1.5 - 3.7%
0.0 - 2.0%
Peripheral Blood Smear: Target cells present. Rare spherocyte seen.

Slight anisocytosis and polychromasia.
Case # 11 Hemoglobin S-C disease

341

342

Summary of Results
343
Metho
d
Hb A
area
Hb S
area
Alk
Agaros
e
Major
band
Acid
Agar/
Agarose
Major
band
Major
peak
Hb S
(Zone 5)
Major
band
(Hb S)
CZE
IEF
HPLC
Major
peak
(Hb S)
RT=4.37
Hb
A2/C
area
Major
band
Major
band
Minor
peak
(Hb A2)
Zone 3
Very
minor
band
(Hb A2)
Very
minor
peak
(Hb A2)
RT=3.6
Major
peak
(Hb C)
Zone 2
Major band
(Hb C)
slightly
cathodal to
A2
Major
peak
(Hb C)
RT=5.12
Agarose gel electrophoresis (pH 8.6) indicated the absence of Hb A, but

two intense bands were detected in the position of Hb S and Hb C/E/O/A 2.
Citrate agar electrophoresis (pH 6.2) also showed the absence of a band in the
usual position of Hb A, here again two bands were detected in the position of Hb
S and Hb C. The citrate agar electrophoresis (pH 6.2) ruled out the possibility of
Hb S-E since Hb E migrates with Hb A in this system. The possibility of Hb
S-O Arab was also ruled out, however with less certainty since Hb O-Arab
migrates between Hb S and Hb A in this system. Since two separate major
344
bands in the position of Hb S and Hb C were detected, Hb C-Harlem (also called

Hb C-Georgetown) was also ruled out, because this variant migrates with Hb S
upon citrate agar electrophoresis (pH 6.2).
IEF confirmed the above results (absence of Hb A, and two major bands
in the position of Hb S and Hb C). CZE has the advantage of fewer variants with
mobility similar to Hb S, HbC, and Hb E. Since Hb A was absent in the patient,
no zones were detected upon CZE. Therefore, the patients blood specimen was
mixed (1:1) with a normal blood specimen, and two major peaks in the patient
were present in Zone 2 (Hb C), and Zone 5 (Hb S). Similarly HPLC showed two
major peaks (besides very minor peaks for Hb F and Hb A2) in the S window
(RT= 4.37 minutes) and C window (RT= 5.12 minutes).
All the above stated tests support the diagnosis of Hb S-C disease in this
patient.
Hb S-C disease is observed in approximately 0.13% of African Americans,
which is approximately half of the homozygous Hb S disease. Most clinical
manifestations of homozygous Hb S disease are also seen in Hb S-C disease,
but in a somewhat milder form.
A characteristic of Hb S-C disease (first pointed out by Professor Virgil F.
Fairbanks, MD, Mayo Clinic, Rochester, MN) is that the concentration
of Hb S is always slightly greater than Hb C. In addition, the cellular dehydration
that occurs as a consequence of the presence of Hb C promotes the distortion of
345
the shape of the red blood cells (Professor James D. Hoyer, MD, Mayo
Clinic, Rochester, MN).
Hemoglobin C-Harlem (also called Hb C-Georgetown) is a rare double chain mutation hemoglobin (
6(A3) GluVal
73(E73) AspAsn
) and patients
heterozygous for only Hb C-Harlem are asymptomatic. Compound heterozygous

state (e.g. Hb S-C-Harlem) exhibits sickling, and also clinical severity.
The diagnosis of Hb S-C disease and homozygous Hb S disease is
usually straight forward in the appropriate clinical context (e.g. African American
patient).The diagnosis of Hb S-O Arab disease, Hb S-C-Harlem disease requires
the evaluation of a large number of laboratory tests in conjunction with the clinical
status of the patient. Special attention is required if the patient has been recently
transfused.
References:
1
2.
3
Lionett F, Hammoudi N, Stojanovic KS, Avellino V, Grateau G, Girot R,

Haymann J-P. Hemoglobin SC disease complications: a clinical study of 179
cases. Haematologica 2012; 97(8): 1136-1141.
OKeefe EK, Rhodes MM, Woodworth A. A patient with a Previous
Diagnosis of Hemoglobin S/C Disease with an unusually Severe Disease
Course. Clin Chem 2008; 55(6): 1228-1231.
Bain BJ. Sickle cell/hemoglobin C disease: In: Hemoglobinopathy
Diagnosis, Blackwell Publishing, 2 nd edition, 2006, pp 164-170.
Joutovsky A, Nardi M. Hemoglobin C and Hemoglobin O-Arab variants
can be diagnosed using the Bio-Rad Variant II High-Performance
Liquid Chromatography System without further confirmatory tests.
Arch Pathol Lab Med 2004; 128: 435-439.
Nagel RL, Fabry ME, Steinberg MH. The paradox of hemoglobin SC
disease. Blood Reviews 2003; 17: 167-178.
346
6
7
8
9
10
Powars DR. Hiti A, Ramicone E, Johnson C, Chan L. Outcome in

Hemoglobin SC disease: A four-decade observational study of clinical,
hematologic, and genetic factors. Am J Hematol 2002; 70: 206-215.
Koduri PR, Agbemadzo B, Nathan S. Hemoglobin S-C disease
revisited: Clinical study of 106 adults. Am J Hematol 2001; 68: 298300.
Nagel RL, Steinberg MH, Hb S/C disease and Hb C disorders. In:
Steinberg MH, Forget BG, Higgs DR, Nagle RL. Disorders of
Hemoglobin: Genetics, Pathophysiology and Clinical Management.
Cambridge, England: Cambridge University Press; 2001; 756-785.
Bunn HF, Forget BG. Hemoglobin: Molecular, Genetic and Clinical
Aspects. 1st ed. Philadelphia, PA: WB Saunders Co; 1986; 533-536.
Bunn HF, Noguchi CT, Hofrichter J, Schechter GP, Schechter AN,
Eaton WA. Molecular and cellular pathogenesis of hemoglobin S/C
disease. Proc Natl Acad Sci USA. 1982; 79: 7527-7531.
Case # 12 Hemoglobin D-Los Angeles (D-Punjab) trait

First year resident (male, 26 years old) in the Department of Surgery. Originally from
India (State of Punjab). Healthy and physically robust.
Laboratory Data:
347
Hemoglobin
14.7
RBC
4.9
MCV
82
MCH
30.2
Platelet
239
Hb A
58.0
Hb A2
1.5
Hb F
0.3
Hb Variant
40.2
13.5 - 18.5 g/dL

3
4.6 - 6.2 Mil/mm
80 - 100 fL
27 - 34 pg
3
150 - 400 Th/mm
94.3 - 98.5%
1.5 - 3.7%
0.0 - 2.0%
Peripheral Blood Smear: No abnormality

348
349

Summary of Results
350
Method
Hb A
area
Hb S
area
Alk
Agarose
Major
band
(Hb A)
Major
band
Major
peak
(Hb A)
Zone 9
Major
band
(Hb A)
Major
peak
(Hb A)
RT=2.42
Major
band
Acid Agar
/Agarose
CZE
IEF
HPLC
Major
peak
(Hb D)
Zone 6
Major Hb D
band
slightly
anodic to S
Hb
A2/C
area
Minor
band
Minor
peak
(Hb A2)
Zone 3
Minor
band
(Hb A2)
Minor
peak
(Hb A2)
RT=3.6
Major
peak
(Hb D)
RT=3.99
*Note: HPLC retention time (RT) varies with the type of the instrument and several other
factors, e.g. temperature etc.
Agarose gel electrophoresis (pH 8.6) showed two major bands in

approximately equal intensity at the positions of Hb A and Hb S. Citrate agar
electrophoresis (pH 6.2) showed only one major band ( 100%) and barely
visible staining in the Hb F position. Several hemoglobin variants migrate in the
position of Hb S upon agarose gel electrophoresis (pH 8.6), and among them,
the most frequently noticed are Hb G, Hb D, and very rarely Hb Korle-Bu (GAcra).
Hb S was easily ruled out on the basis of the negative sickle solubility test,
Hb G (-chain variant) was ruled out on the basis of the absence of Hb G 2 band
351
(2G2) and the observation that the percentage of the abnormal variant
approaches 50%. -chain variants percentages do not run this high without other
genetic complications (see Case# 5).
The differentiation between Hb D-Los Angeles and other kinds of
heterozygous D hemoglobins (which are also -chain variants) or heterozygous
Hb Korle-Bu (G-Accra) on the basis of electrophoretic tests (alkaline, acid, IEF,
CZE) was not possible with certainty due to identical mobilities. HPLC
differentiated Hb D-Los Angeles from Hb Korle-Bu. We have summarized the
HPLC retention times from three separate studies for Hb D-Los Angeles and Hb
Korle-Bu:
Nardi et al*
Nardi-2013**
Hoyer et al
Hb Korle-Bu
3.92 + 0.050
3.9+ 0.034
3.88+ 0.08
Hb D- Los Angeles
4.18+ 0.007
4.11+ 0.078
4.08+ 0.08
* Bio-Rad Variant II (Clin Chem 2004; 50: 1736-1747)

** Bio-Rad Variant II (personal communication)
Bio-Rad Variant Classic (Intl J Lab Hematol 2012; 34: 594-604)
It is the observation of Professor Michael A. Nardi (personal communication) that
Hb Korle-Bu rarely separates from Hb A2 (due to the closeness of their retention times),
while Hb D-Los Angeles always separates from Hb A2.
In view of the laboratory tests, the diagnosis of Hb D-Los Angeles trait

was most likely. Since the patient had a clinically silent and harmless
condition, it was not advised to perform globin chain analysis and DNA studies.
352
Hb D-Los Angeles results from a substitution of glutamic acid by glutamine

on position 121 of the -chain [121(GH4)GluGln.GAA>CAA] and is a
harmless condition. Hb D-Los Angeles has been found double heterozygotes for
other variants (e.g., Hb S, Hb C, Hb E). Hb D-Los Angeles in combination with
Hb S causes a severe sickling disorder (Case # 13).
Homozygous Hb D-Los Angeles patients exhibit normal hematologic
indices (e.g. hemoglobin, RBC), and no evidence of hemolysis. However,
o
patients with Homozygous Hb D-Los Angeles and -Thalassemia do have a

mild anemia and mild hemolysis.
References
1
2
3
4
5
6
7
Pandey S, Mishra RM, Pandey S, Saxena R. Homozygous hemoglobin D with

alpha thalassemia: case report. Open Journal of Hematology 2011; 2: 1-4.
Basmanj MT, Karimpoor M, Amirian A, Jafrinejad M, Katouzian L, Valei A,
Bayat F, Kordafshari A, Zeinali S. Co-inheritance of Hemoglobin D and thalassemia Traits in Three Iranian Families: Clinical Relevance.
Archives of Iranian Medicine 2011;14(1): 61-63.
Srinivas U, Pati HP, Saxena R. Hemoglobin D-Punjab syndromes in India: a
single center experience on cation-exchange high performance liquid
chromatography. Hematology 2010; 15 (3): 178-181.
Yavarian M, Karimi M, Paran F, Neven C, Harteveld CL, Giordano PC. Multi
Centric Origin of Hemoglobin D-Punjab [121(GH4)GluGLN, GAA>CAA].
Hemoglobin 2005; 29 (4): 307-310.
Atalay EO, Koyuncu H, Turgut B, Atalay A, Yildiz S, Bahadir A, Koseler A.
High incidence of Hb D-Los Angeles [121(GH4)GluGln] in Denizli
Province, Aegean Region of Turkey. Hemoglobin 2005; 29(4): 307-310.
Owaidah TM, Al-Saleh MM, Al-Hellani AM. Hemoglobin D/-thallasemia and
-thalassemia major in a Saudi family. Saudi Med J 2005; 26(4): 674-677.
Thornburg CD, Zimmerman SA, Schultz WH, Ware RE. An infant with
Homozygous D-Iran. Journal of Pediatric Hematology/Oncology 2001; 23(1):
67-68.
353
8
9
10
11
12
13
14
El-Kalla S, Mathews AR. Hb D-Punjab in the United Arab Emirates.

Hemoglobin 1997; 21(4): 369-375.
Zago MA, Costa FF. Hb D-Los Angeles in Brazil: Simple Heterozygotes and
Associations with -Thalassemia and with Hb S. Hemoglobin 1988; 12(4):
399-403.
Harano T, Harano K, Ueda S, Nakaya K. Hb D-Los Angeles [121 GluGln]
in Japan. Hemoglobin 1987; 11(2): 177-180.
Li HJ, Liu DX, Li L, Liu ZG, Lo SL, Zhao J, Han XP, Yu WZ. A Note About The
Incidence And Origin of Hb D-Punjab in Xinjiang, Peoples Republic of China.
Hemoglobin 1986; 10(6): 667-671.
Husquinet H, Parent MT, Galacteros F. Hemoglobin D-Los Angeles [121
(GH4)GluGln] in the Province of Liege, Belgium. Hemoglobin 1986; 10(6):
587-592.
Baiget M, del Rio E, Gimferrer E. Hemoglobin D-Punjab (121 GluGln) in a
Spanish Family. Hemoglobin 1982; 6(2):193-198.
Ramot B, Rotem J, Rahbar S, Jacobs AS, Udem L, Ranney HM. Hemoglobin
D-Punjab in a Bulgarian Jewish Family. Israel J. Med. Sci. 1969; 5(5):10661070.
Case # 13
Hemoglobin S-D disease
17 years old male patient. No other information provided due to the privacy requested
by the patient. No record of blood transfusion during the past three months.
Laboratory Data:
Hemoglobin
10.2
13.5 - 18.5 g/dL

354
RBC
MCV
MCH
Platelet
3.2
90.7
30.9
229
Hb A (mostly Hb A1c)
6.0
4.6 - 6.2 Mil/mm

80 - 100 fL
27 - 34 pg
3
150 - 400 Th/mm
Hb A2
2.7
Hb F
2.5
Hb Variant-1
49.3
Hb Variant-2
39.5
1.5 - 3.7%
0.0 - 2.0%
Peripheral Blood Smear: Moderate sickle cells. Target cells and polychromasia.
Hemoglobin instability (isopropanol) test: Negative.
Case # 13 Hemoglobin S-D disease
355
356

357
Summary of Results
Method
Alk
Agarose
Acid Agar
/Agarose
CZE
Hb A
area
Hb S
area
Major
band
(Hb D)
Major
band
Major
band
(Hb S)
Major*
peak
(Hb D)
Zone 6
IEF
HPLC
Minor
peak
(Hb F)
RT=1.0
5
Major
peak
(Hb D)
RT=4.0
Major*
peak
(Hb S)
Zone 5
Major
band
(Hb S)
Major
peak
(Hb S)
RT=4.3
Hb
A2/C
area
Minor
band
Minor
peak
(Hb A2)
Zone 3
Major
band
anodal to
S position
(Hb D)
Minor
band
(Hb A2)
Minor
peak
(Hb A2)
RT=3.6
* Overlap of the two peaks (Zone 5-6) due to approximately equal and higher
concentration of Hb S and Hb D-Los Angeles.
Note: HPLC retention time (RT) varies with the type of instrument used and several
Agarose gel electrophoresis (pH 8.6) showed a major and very intense
band in the position of Hb S. Another band of faint intensity was detected in
the Hb F position. A faint band in the position of Hb C/E/O/A 2 was also noticed.
No band was detected in the position of Hb A. Citrate agar electrophoresis (pH
6.2) presented with two major bands in approximately equal intensity in the
358
position of Hb A and Hb S. A faint band was also detected in the position of

Hb F. Since the sickle cell solubility test was positive, therefore the band in the
Hb S position upon citrate agar electrophoresis (pH 6.2) suggested the presence
of a -chain variant (Hb S). The migration of a band of equal intensity in the
position of Hb A upon citrate agar electrophoresis (pH 6.2) suggested the
presence another hemoglobin variant (since Hb A was absent upon alkaline
agarose gel electrophoresis). Several hemoglobin variants (e.g. Hb G, Hb D-Los
Angeles, Hb Korle-Bu, etc) exhibit this kind of migration pattern, therefore
assignment of this hemoglobin variant was deferred.
IEF confirmed the presence of Hb S, however another band in between
the customary position of Hb G and Hb S was also prominent. The presence of
Hb G from IEF was ruled out positively as no band in the position of Hb G 2
(2G2) was detected. Since Hb D-Los Angeles and Hb Korle-Bu have similar
mobilities upon IEF, therefore a distinction could not be made between these two
possibilities. HPLC was helpful in differentiating between the Hb D-Los Angeles
and Hb Korle-Bu variants, as Hb D-Los Angeles has a longer retention time (4.0
minutes) as compared to Hb Korle-Bu (3.75 minutes). Hb S eluted at retention
time of 4.3 minutes, thus the two major bands in this case were separated nicely
upon HPLC.
In a heterozygous situation upon CZE, Hb S migrates in Zone 5 and Hb DLos Angeles in Zone 6. In this case since the concentration of the two variants is
intense ( 40-49% from HPLC), thus clearly separated peaks were not detected
359
but the scan positively showed two overlapping peaks in the position of Zones 56. Distinct peaks for Hb F and Hb A2 from CZE were noticed in Zone 7 and and
Zone 3 respectively.
The specimens of father and mother of this person were not available for
additional studies. Furthermore globin chain and DNA studies were also not done
on the blood of this person. On the basis of the available laboratory data a
tentative diagnosis of a double heterozygosity of Hb S [6 (A3) GluVal] and
Hb D-Los Angeles [121(GH4)GluGln.GAA>CAA] was advised to the
physician.
Hb D-Los Angeles in both the heterozygous (Case # 12) and homozygous
state is clinically silent and harmless. However patients with homozygous Hb Do
Los Angeles and -thalassemia do have mild anemia and also exhibit mild
hemolysis. Hb D-Los Angeles is not itself a sickling hemoglobin, but compound
heterozygosity (Hb S + Hb D-Los Angeles) produces a severe sickle cell anemia
because Hb D-Los Angeles enhances Hb S polymerization by forming an
additional contact stabilizing the Hb S polymer.
References
1
2
Adekile A, Mullah-Ali A, Akar NA. Does Elevated Hemoglobin F Modulate the

Phenotype of Hb SD-Los Angeles?. Acta Haematol 2010; 123: 135-139.
Isoa EM. Current Trends in the Management of Sickle Cell Disease: An
Overview. Benin J Postgraduate Med 2009; 11:50-73.
360
3
4
5
6
7
8
9
10
11
12
Mukherjee MB, Surve RR, Gangakhedkar RR, Mohanty D, Colah RB.

Hemoglobin sickle D Punjab-a case report. Indian J Hum Genetics 2005;
11(3): 154-155.
Jiskoot PMC, Halsey C, Rivers R, Bain BJ, Wilkins BS. Unusual splenic
sinusoidal iron overlaod in sickle cell/haemoglobin D-Punjab disease. J Clin
Pathol 2004; 57: 539-540.
Athanasiou-Metaxa M, Economou M, Tstra I, Pratsidou P, Tsantali C. CoInheritance of Hemoglobin D-Punjab and Hemoglobin S: Case Report. J Ped
Hematology/Oncology 2002; 24(5): 421.
Perea FJ, Casas-Castaneda M, Villalobos-Arambula AR, Barajas H, Alverez
F, Camacho A, Hermosillo RM, Ibrarra B. Hb D-Los Angeles Associated with
Hb S or -Thalassemia in Four Mexican Mestizo Families. Hemoglobin 1999;
23(3): 231-237.
Dash S. Haemoglobin S-D Disease in a Bahraini Child. Bahrain Med Bulletin
1995; 17(4): 154-56.
Samperi P, Dibenedetto SP, Cataldo AD, Mancuso GR, Schiliro G. Unusual
Sickle Cell Disease observed for the First Time in Italy. Haematologica 1990;
75: 464-66.
McCurdy PR, Lorkin PA, Casey R, Lehmann H, Uddin DE, Dickson LG.
Hemoglobin S-G (S-D) Syndrome. The American J of Med 1974; 57: 665-670.
Barton LL, Stark AR, Zarkowsky HS,.Hemoglobin S-D disease in a Negro
Child. The Journal of Pediatrics 1973; 82(1): 164-165.
Ozsoylu S. Haemoglobin S-D Disease in a Turkish Family. Scand. J Haematol
1969; 6: 10-14.
Cawein MJ, Lappat EJ, Brangle RW, Farley CH. Hemoglobin S-D Disease.
Annals of Internal Medicine 1966; 64(1): 62-70.
Case # 14 Hemoglobin E and Associated Disorders

The contents of this section are presented from Hoyer JD, Kroft SH, eds. Color
Atlas of Hemoglobin Disorders: A Compendium Based on Proficiency Testing.
Northfield, IL: College of American Pathologists; 2003 (Reproduced with Permission).
361
In addition to Hb E, several other disorders of hemoglobin are prevalent in the

Southeast Asian population. Therefore, Hb E may be encountered in conjunction with
another abnormality. A description of the various Hb E-associated disorders is provided
below.
1.
Hb E trait. A harmless condition characterized by mild microcytosis and

often by erythrocytosis. No icterus, no splenomegaly, no anemia. MCV
about 75 fL (adult). Electrophoresis: Hb E 30-35%, Hb A 65-70%, Hb F
<2%.
2.
Homozygous Hb E. A harmless condition characterized only by mild

microcytosis and erythrocytosis. No icterus, no splenomegaly, no anemia
(hemoglobin concentration >11 g/dL in females, >14 g/dL in males). MCV
about 67 fL (adults). Electrophoresis: Hb E about 99%, the rest Hb F.
3.
Hb E trait/-thalassemia. This combination results in microcytosis, but

usually no other adverse effects (no anemia, no splenomegaly, no icterus).
Serum ferritin assay is required to differentiate this condition from Hb E
trait/iron deficiency. Electrophoresis (1 gene deletion): Hb E 25-30%;
remainder Hb A; Hb F normal. Electrophoresis (2 gene deletion): Hb E
20-25%; remainder Hb A; Hb F normal. Since Hb E and Hb A2 co-migrate
in all electrophoresis media and co-elute from chromatography columns, a
common laboratory error is to ascribe the electrophoresis findings to thalassemia trait. However, in the latter, Hb A2 is always <10%.
4.
Hb E trait/Hb H disease. In this disorder, Hb E trait is inherited in

conjunction with a three locus gene deletion. This is a moderately
362
severe thalassemic disorder with features identical to Hb H disease.

However, electrophoresis does not reveal Hb H. Instead, Hb E represents
about 10-15% of hemoglobin; most of the remainder is Hb A. This
paradox is due to reduced total synthesis of globin chains. As a result,
not enough surplus chains are present to form tetramers (Hb H).
Instead, Hb Bart's is present. (Thus, this condition has also been called
"Hb A + E + Bart's Disease").
5.
Homozygous Hb E/Hb H disease. This disorder has the same features as

Hb H disease. However, electrophoresis reveals mostly Hb E (about 95%)
and a small proportion of Hb F. It is believed that in this condition, the
tetramers co-migrate with Hb E in all electrophoresis media.

6.
Hb E trait/-thalassemia/Hb Constant Spring. Features are the same as 4

and 5 above, except for faint additional hemoglobin bands (as many as
five) between the positions of Hb E and the site of application. These
additional faint bands represent Hb Constant Spring.
7.
Hb E trait/iron deficiency. A benign condition characterized by

microcytosis, often erythrocytosis, and anemia. The anemia is due to iron
deficiency and thus may be minimal to severe. There is no icterus and no
splenomegaly. Electrophoresis shows the same pattern as Hb E trait/thalassemia. The combination should be suspected in an anemic patient
with an "Hb A2" concentration of 10-20%. The diagnosis is confirmed by a
serum ferritin assay. Following treatment, repeat electrophoresis will show
363
Hb E representing 30-35% of total (unless the patient also has Hb E

trait/-thalassemia).
8.
Hb E/-thalassemia. This is a serious thalassemic disorder due to

compound heterozygosity for both Hb E trait and -thalassemia trait.
Characteristics are severe anemia, icterus, marked splenomegaly, and
microcytosis. Affected children suffer all the problems of -thalassemia
major. Most require frequent transfusions and should also receive iron
chelation therapy. This is the most common severe thalassemia of
Southeast Asians. Neurologic manifestations are often reported that are
due to brain or spinal cord compression by extramedullary hematopoietic
tumors, which may cause paraplegia. The tumors respond to
radiotherapy. Electrophoresis: Hb E is 40-90% total; the rest is Hb F.
(Note: Because these patients usually require transfusion, Hb A may be
present from donor blood). It should be pointed out that it is not necessary
to document elevated Hb A2 levels to establish a diagnosis of Hb E/thalassemia. The diagnosis is easily established on the basis of an Hb E
concentration >40% with the remainder representing Hb F (usually 3060%) and an absence of Hb A.
9.
Hb E/-thalassemia, post-splenectomy. Same condition as # 8 (see

above), but often confusing in laboratories. Splenectomy is a common
treatment in Hb E/-thalassemia and is reputed to be beneficial for those
with severe anemia. The post-splenectomy blood picture is characterized
by marked normoblastemia and a positive solubility test for sickling
364
hemoglobin. The latter is due to the large number of normoblast nuclei

causing strong persistent turbidity. Pulmonary artery occlusion is a
common complication in splenectomized patients with Hb E/thalassemia. Prophylactic therapy with daily doses of aspirin or
dipyridamole is indicated for all patients with this disorder who have been
splenectomized.
Note: It will not be out of place to mention here that another disorder Hb S-E
heterozygous has been also diagnosed in persons of Southeast Asian
origin (Case # 14 C).
References
1 Fucharoen S, Weatherall DJ. The Hemoglobin E Thalassemias. Cold Spring
Harb Perspect Med 2012; 2: a011734.
2 Sae-ung N, Srivorakun H, Fucharoen G, Yamsri S, Sanchaisuriya K,
Fucharoen S. Phenotypic expression of hemoglobins A2 , E and F in various
hemoglobin E related disorders. Blood Cells, Molecules, and Diseases 2012;
48: 11-16.
3 Tatu T, Kasinrerk W. A novel test tube method of screening for hemoglobin E.
Int. Lab. Hem 2012; 34: 59-64.
4 Moiz B, Hashmi MR, Nasir A, Rashid A, Moatter T. Hemoglobin E syndromes
in Pakistani population. NMC Blood Disorders 2012; 12: 1-6.
5 Khan MR, Aziz MA, Shah MSU, Imam H. Hemoglobin E trait- in Rajshahi,
Bangladesh. Bangladesh Med ResCounc Bull 2012; 38: 72-73.
6 Tamminga RYJ, Doombos ME. Muskiet FAJ, Koetse HA. Rhabdomyolysis in a
child with Hb SE. Pediatric Hematology-Oncology 2012; 29(3): 267-269.
7 Edison ES, Shaji RV, Chandy M, Srivasta A. Interaction of Hemoglobin E with
Other Abnormal Hmoglobins. Acta Haematol 2011; 126: 246-248.
8 Tay SH, Teng GG, Poon M, Lee VKM, Lim AYN. A Case of Hemoglobin SE
Presenting with Sickle Cell Crisis: Case Report and Histological Correlation.
Annl Acad Med 2011; 40 (12): 552-553.
9 Colah R, Gorakshakar A, Nadkarni A. Global burden, distribution and
prevention of -thalassemias and hemoglobin E disorders. Expert Review of
Hematology 2010; 3: 103-117.
10 Patel J, Patel A, Patel J, Kaur A, Patel V. Prevalence of Haemoglobinopathies
in Gujrat, India: A Cross-Sectional Study. The Internet Journal of Hematology
2009; 5 (1): DOI: 10.5589/1764.
365
11 Intorasoot S, Thongpung R, Tragoolpua K, Chottayaporn M. Hemoglobin E

Detection Using PCR with Confronting Two-Pair Primers. J Med Assoc Thai
2008; 91: 1677-1680.
12 Masiello D., Heeney MM, Adewoye AH, Eung SH, Luo Hong-Yuan, Dteinberg
MH, Chui D HK. Hemoglobin S-E Disease- A Concise Review. Am H Hematol
2007; 82: 643-649.
13 Jetsrisuparb A, Sanchaisuriya K, Fucharoen G, Fucharoen S, Wiangnon S,
Jetsrisuparb C, Sirijirachai J, Chansoong K. Development of Severe Anemia
During Fever Episodes in Patients with Hemoglobin E trait and Hemoglobin H
Disease Combinations. J Pediatr Hematol Oncol 2006; 28 (4): 249-253.
14 Bain BJ. Hemoglobin E. Other significant hemoglobinopathies. In:
Hemoglobinopathy Diagnosis. 2nd Ed, 2006, pg 201-209, Blackwell
Publishing, London.
15 Edison ES, Shaji RV, Srivastava A, Chandy M. Compound Heterozygosity for
Hb E and Hb Lepore-Hillandia in India: First report and potential diagnostic
pitfalls. Hemoglobin 2005; 29(3): 221-224.
16 Andino L, Risin SA. Pathologic Quiz case. A 24-Year-Old Woman With
Abnormal Hemoglobin and Thrombocytopenia. Arch Pathol Lab Med 2005;
129: 257-258.
17 Mishra P, Pati HP, Chatterjee T, Dixit A, Choudhary DR, Srinivas MV,
Mahapatra M, Choudhary VP. HB SE Disease: a clinico-hematological profile.
Ann Hematol 2005; 84: 667-670.
18 Sirichotiyakul S. Tongprasert F, Tonsong T. Screening for hemoglobin E trait in
pregnant women. Intl J Gyn & Obstet 2004; 86: 390-392.
19 Fucharoen S, Sanchaisuriya K, Fucharoen G, Panyasai S, Devenish R, Luv
L. Interaction of hemoglobin E and several forms of -thalassemia in
Cambodian families. Haematologica 2003; 88: 1092-1098.
20 Piplani S. Hemoglobin E Disorders in the North East India. JAPI 2000; 48(11):
1082-1084.
21 Fucharoen S. Hemoglobin E disorders. In: Steinberg MH, Forget BG, Higgs
DR, Nagel RL, eds. Disorders of Hemoglobin: Genetics, Pathophysiology and
Clinical Management. Cambridge, England: Cambridge University Press;
2001: 1139-1154.
22 Gupta R, Jarvis M, Yardumian A. Compound Heterozygosity for hemoglobin S
and hemoglobin E. Br J Haematol 2000, 108: 463.
23 Joseph VJ, Sunny AO, Pandit N, Yeshwanth M. Double Heterozygosity for
Hemoglobin S and E. Indian Pediatrics 1992; 29: 895-897.
24 Fairbanks VF, Gilchrist GS, Brimhali B, Jereb JA, Goldston EC. Hemoglobin E
Trait Rexamined: A Case of Microcytosis and Erythrocytosis. Blood 1979;
53(1): 109-115.
366
Case # 14 a
Hemoglobin E trait
A 56 year old male of Southeast Asian origin, migrated to America in 1972 with his
parents. Physical examination showed no abnormalities.
Laboratory Data:
Hemoglobin
RBC
MCV
RDW
Platelet
14.8
5.7
74
13.5
240
Hb A
Hb A2 (CZE)
Hb variant
64.0
2.0
34.0
13.5-18.5 g/dL
3
4.6-6.2 Mil/mm
80-100 fL
11.5-14.5%
3
150-400 Th/mm
94.3-98.5%
1.5-3.7%
367
Peripheral Blood Smear: Slight microcytosis, and occasional target cells

Hemoglobin instability (isopropanol) test: Positive
368
369

Summary of Results
370
Method
Hb A
area
Alk Agarose
Major
band
(Hb A)
Acid
Agar/Agaros
e
Major
band
(Hb A+
Hb E+
Hb A2)
Major
peak
(Hb A)
Zone 9
CZE
IEF
Hb S
area
Major
peak
(Hb E)
Zone 4
Major
band
(Hb E)
Slightly
anodal to
Hb A2
Major
band
(Hb A)
HPLC
Very
faint
peak
(Hb F)
RT=1.0
7
Hb
A2/C
area
Major
band
(Hb E +
Hb A2)
Major
peak
(Hb A)
RT=2.4
2
Minor
peak
(Hb A2)
Zone 3
Very
minor
band
(Hb A2)
Major
peak
(Hb E +
Hb A2)
RT=3.6
3
Alkaline agarose electrophoresis (pH 8.6) showed two major bands in the
position of Hb A (65%), and Hb C/E/O/A 2 (35%). Citrate agar electrophoresis
showed only one band in the position of Hb A. This kind of electrophoretic
migration pattern (pH 8.6 and 6.2) ruled out the possibility of Hb C
and Hb O, and suggested the possibility of Hb E, as Hb A2 is never > 10%. IEF
also showed a major band in the position of Hb A and another band slightly
371
anodal to Hb A2 suggesting the presence of Hb E variant. Hb E and Hb A2 coeluted upon HPLC, therefore their quantification was not feasible. However,
CZE presented three distinct peaks in the zones for Hb A, Hb E and Hb A2 and
also provided quantification of the peaks.
Hb E is a -chain variant (226Glu-Lys) and is the second most prevalent

hemoglobin variant in the world after Hb S. It is prevalent in sixteen Southeast
Asian countries, however it is also encountered in Europe and North America.
A diagnosis of Hb E trait was made in view of the electrophoretic results and the
following characteristics:
Microcytosis, hypochromia, target cells, irregularly contracted cells, basophilic

stippling or any combination of these features of the peripheral blood film.
Negative sickle cell solubility test.
Positive isopropanol test for unstable hemoglobin.
Per se harmless condition and not associated with anemia.
Hb A2 and Hb F in the normal range.
Hb E in the range of 30-39%. Hb E is a -chain variant, however the chain is
synthesized in Hb E trait at a reduced rate in comparison with . In view of this

E
slower ribosomal synthesis of chain, a mild thalassemic blood picture is also

witnessed.
Case # 14 b
Hemoglobin E homozygous
A 25 year old female from Laos-Northern Vietnam region, asymptomatic; physical examination
showed no abnormalities. No icterus, no splenomegaly.
372
Laboratory Data:
Hemoglobin
RBC
MCV
MCH
Platelet
14.1
5.7
68
24.9
223
12.0-16.0 g/dL
3
4.0-5.5 Mil/mm
80-100 fL
26-34 pg
3
150-400 Th/mm
Hb A
Hb A2 (CZE)
Hb F (HPLC)
Hb variant
Not Detected
0.5
0.8
98.7%
94.3-98.5%
1.5-3.7%
0.0-2.0%
Significant microcytosis, hypochromia, target cells

and occasional basophilic stippling of erythrocytes.
373

374

375
Summary of Results
Metho
d
Hb A
area
Hb S
area
Hb
A2/C
area
Alk
Agaros
e
Acid
Agar /
Agarose
Major
band
(Hb E +
Hb A2)
Major
band
(Hb E +
Hb A2)
CZE
Major
peak
(Hb E)
Zone 4
Very
minor
peak
(Hb A2)
Zone 3
IEF
Major
band
(Hb E)
slightly
anodal to
Hb A2
HPLC*
Major
peak
(Hb E +
Hb A2)
RT=3.65
In addition to Hb F peak at RT= 1.05 minutes, there are two additional minor
peaks at RT= 1.75 minutes and RT= 2.1 minutes. The peak at RT=2.1 minutes
(A0 window) must not be construed as Hb A. Similar peaks were detected upon
HPLC (Bio-Rad Variant II), and were alleged to post-translationally modified
Hb E (Other Significant Hemoglobinopathies. In: Hemoglobinopathy Diagnosis,
Bain, Barbara J., 2nd edition, pg 206, Blackwell Publishing, 2006).
376
Alkaline agarose electrophoresis (pH 8.6) showed only one band in the
position of Hb C/E/O/Hb A2, therefore Hb A was not present. Citrate agar
electrophoresis (pH 6.2) indicated only one major band in the position of Hb A,
thus the possibility of Hb C and Hb O was ruled out. IEF also indicated the
absence of Hb A and only one major band slightly anodal to Hb A2 was detected.
Absence of Hb A was also shown by HPLC, and only one major peak eluted at
RT = 3.65. CZE clarified the ambiguity about the solitary band /peak in
electrophoretic methods and HPLC, and two peaks were detected at Zone 4
(major peak presumably of Hb E) and Zone 3 (minor peak of Hb A2).
A diagnosis of Hb E homozygous was made in view of the electrophoretic,
and HPLC results and the following characteristics:
Absence of anemia and hemolysis. Spleen was not enlarged.

Negative sickle cell solubility test.
Isopropanol test positive for unstable hemoglobin.
Increased red cell count, normal hemoglobin, decreased MCV and MCH.
Significant microcytosis, hypochromia, and variable number of target cells.
Harmless condition.
Hb A2 and Hb F in the normal range.

Hb E concentration >95 %.
Homozygous Hb E is a clinically benign condition. Unfortunately, it is
prevalent in the population areas (e.g. Cambodia and Northeastern India) that
also have the higher frequency of -thalassemia minor. Therefore genetic
counseling is advised to prevent the occurrence of severe thalassemic
o
(Hb E/ thalassemia) disorders in their children.
377
On the basis of hematological studies alone, homozygous Hb E may be

confused with iron deficiency and -thalassemia. The following characteristic
features can distinguish between Hb E/-thalassemia and homozygous Hb E.
In Hb E/-thalassemia the concentration of Hb E varies between 40-70%, and

the Hb F concentration is found in the range of 30-60%.
In homozygous Hb E the Hb F concentration is in the normal range, and Hb E

concentration is >95%.
Case # 14 c
Hemoglobin S-E disorder
A 29 year old female nurse (Southeast Asian descent) complained of knee joint pain and
weakness of the lower extremities. Hemoglobin electrophoresis was ordered after lower MCV
378
was found from CBC. Clinical and laboratory data of her parents were not available to the
physician.
Laboratory Data:
Hemoglobin
RBC
MCV
MCH
Platelet
12.2
4.6
72
26
212
12.0-16.0 g/dL
3
4.0-5.5 Mil/mm
80-100 fL
26-34 pg
3
150-400 Th/mm
Hb A
Hb A2 (CZE)
Hb F (HPLC)
Hb S
Hb E
Not Detected
0.5
0.5
55%
44%
94.3-98.5%
1.5-3.7%
0.0-2.0%
Peripheral Blood Smear: Hypochromia, microcytosis, irregular contracted cells,

occasional target cells, polychromatic cells.
Sickle cell solubility test for Hb S: Positive
Case # 14 c Hemoglobin S-E heterozygous disorder
379

380

381
Summary of Results
Metho
d
Hb A
area
Alk
Agaros
e
Acid
Agar/
Agaros
e
Major
band
(Hb E)
Hb S
area
Hb
A2/C
area
Major
band
(Hb S)
Major
band
(Hb E)
Major
band
(Hb S)
CZE
Major
peak
(Hb S)
Zone 5
Major
peak
(Hb E)
Zone 4
Minor
peak
(Hb A2)
Zone 3
IEF
Major
band
(Hb S)
Major
band
(Hb E)
slightly
anodal to
Hb A2
HPLC*
Major
peak
(Hb S)
RT=4.48
Major
peak
(Hb E +
Hb A2)
RT=3.65
*Note: HPLC retention time (RT) varies with the type of the instrument used and
Alkaline agarose gel electrophoresis (pH 8.6) showed two major bands in
the position of Hb S (55%) and Hb C/E/O/A 2 (45%). Citrate agar
electrophoresis (pH 6.2) confirmed the presence of a major band due to Hb S and
presence of another band at the position of Hb A ruled out the possibility of Hb C
382
and Hb O. IEF indicated a major band in the position of S and another major
band in the position of Hb A2. These three electrophoretic methods suggested
the presence of double heterozygous Hb S and Hb E. HPLC was not helpful as
both Hb E and Hb A2 co-eluted at RT = 3.65. CZE provided a clear separation of
the three hemoglobin entities, i.e. Hb S (Zone 5), Hb E (Zone 4) and Hb A2 (Zone
3), therefore a presumptive diagnosis of Hb S-E double heterozygous disorder
was made.
Hb S [6 (A3) GluVal) and Hb E [26 (8) GluLys] are the two most
prevalent hemoglobin variants in the world. However, due to their existence in
different ethnic groups and continents, compound heterozygosity for Hb S and
Hb E is extremely rare. As of 2011 only thirty (30) cases were reported, therefore
hematological parameters are too scant to provide a module for diagnostic
purposes.
Majority of Hb S-E subjects have mild or absent anemia, microcytic
indices, and some target cells. Contrary to some earlier reports, a severe sickle
cell crisis was recently reported in a 66-year-old Bangladeshi woman in
Singapore (Ann Acad Med 2011; 40: 552-553).
A recent review of Hb S-E double heterozygosity by Masiello et al (Am J
Hematol 2007; 82: 643-649) mentioned that patients aged 18 and younger are
usually well. Sickling-related complications, including potentially life threatening
acute chest syndrome was developed in a majority of cases. Generally these
383
patients have Hb S concentration in the range of 60-65%, which is also similar to

+
the patients with Hb S- -thalassemia, and therefore hematological features and

+
clinical course of these patients appeared to parallel those of Hb S- thalassemia.

Coincidental complication per se not related to hemoglobinopathy was
also reported, e.g. idiopathic thrombocytopenic purpura in a 28-year-old woman
with Hb S-E heterozygosity (Arch Pathol Lab Med 2005, 129: 257-58).
Case # 15 Hemoglobin S-Korle Bu (G-Accra)
384
A 23 year old female administrative assistant of Ghanaian decent at the Embassy of

Ghana in Washington, DC was found to have abnormal hemoglobinopathy. Physical
examination was unremarkable.
Laboratory Data:
Hemoglobin
RBC
MCV
MCH
MCHC
RDW
Platelet
Hb A
Hb S (HPLC)
Hb Korle-Bu
Hb A2 (CZE)
Hb F
13.1
4.4
82.1
27.3
32.1
12.6
267
Not Detected
53.4 %
45%
1.6
Not Detected

Hemoglobin instability (isopropanol) test: Negative

385
12.0 16.0 g/dL

3
4.0 5.5 Mil/mm
79-98 fL
26-34 pg
31-36%
11.5-14.5%
3
150-400 Th/mm
94.3-98.5%
1.5 -3.7%
0.0-2.0%
386
Case #15 Hemoglobin SKorle Bu (G-Accra)

Summary of Results
Method
Hb A
Hb S
387
Hb
area
area
Broad
Major
band
starting
anodic to
S
(Hb
Korle-Bu
+ Hb S)
Alk Agarose
A2/C
area
Very
faint
band
(Hb A2)
Acid Agar
Major
band
slightly
anodic
to Hb A
position
(Hb
KorleBu)
Major
band
(Hb S)
Acid
Agarose
Major
band in Hb
A position
(Hb KorleBu)
Major
band
(Hb S)
Major
peak
Zone 6
(Hb
KorleBu)
Major
peak
Zone 5
(Hb S)
Minor
peak
Zone 3
(Hb A2)
Major
band
(Hb S)
Very
faint
band
(Hb A2)
Major
peak
RT=4.48
(Hb S)
Major
peak
(Hb A2 +
Hb
Korle
Bu)
RT=3.75
CZE
Small
degradatio
n peaks
Zone 7
Major
band
slightly
anodal
to Hb S
band
(HbKorleBu)
IEF
HPLC
Alkaline agarose electrophoresis (pH 8.6) showed a major band in the

388
area of Hb S and a very faint band in the Hb A2 area, but Hb A was not
detected. Citrate agar electrophoresis (pH 6.2) showed two variants in the
position of Hb S and very slightly anodic to the Hb A band position.
IEF also indicated the presence of two variants, one in the position of Hb S and
another band anodal to Hb S in the migration position of Hb D-Los Angeles/GPhiladelphia and few other variants. The presence of Hb S as one of the variants
was also confirmed by HPLC, CZE and the positive solubility test. It is obvious
that the three electrophoretic methods (i.e. alkaline agarose, acid agar, and IEF)
could not identify the second major band with certainty. Hb G-Philadelphia was
ruled out due to the absence of G2 variant (G22) in both the alkaline agarose
electrophoresis and IEF (Case # 5). In situations like this the patients history and
clinical status may indicate the likelihood of the hemoglobin variant. Both of these
hemoglobin variants (Hb D-Los Angeles and Hb Korle-Bu) are found in the
population sector dominated by Hb S. HPLC is helpful in the separation of Hb DLos Angeles from Hb Korle-Bu, since Hb D-Los Angeles has a longer elution time
(Case # 12) as compared to Hb Korle-Bu. The D-Los Angeles chain can be easily
separated from Korle-Bu chain by reverse phase chromatography, but all these
additional tests are not necessary. Hb S interacts with Hb D-Los Angeles causing
sickle cell disease. Hb S also interacts with Hb Korle-Bu, but in opposite
direction, i.e. inhibiting sickling. This patient does not have an abnormal blood
picture so the second variant in this case is most likely Hb Korle-Bu (G-Accra).
Korle-Bu means valley of the Korle lagoon, and this hemoglobin was
named after its discovery at Korle-Bu Hospital, Accra, Ghana. Initially it was
389
called Hb G, and since several other hemoglobins with mobility similar to Hb G

were discovered, its name was changed to Hb Accra. The reason for changing its
name from Hb Accra to Hb Korle-Bu is not known to me. This mutation has been
reported from the Ivory Coast, Costa Rica, and Mexico, so it is not highly
prevalent but is widely spread.
Hb Korle-Bu is the result of mutation GAT AAT at codon 73 of the
chain [(E73)AspAsn]. Both the heterozygote and homozygote of Hb Korle-Bu
are clinically normal.
References
1
2
3
4
AKL PS, Kutlar F, Patel N, Salisbury CL, Lane P, Young AN. Compound
Heterozygosity for Hemoglobin S [6(A3)Glu6Val] and Hemoglobin Korle-Bu
[73(E17)ASP73Asn]. Lab Hematol 2009; 15: 19-23.
Chico A, Padros A, Novials A. The Korle-Bu Hb Variant in Caucasian Women
With Type I Diabetes. A pitfall in the assessment of diabetes control. Diabetes
Care 2004; 27(9): 2280-2281.
Vukelja SJ. Hemoglobin Korle-Bu (G-Accra) in Combination with Hemoglobin
C. Am J Hematol 1993; 42(4): 412.
Nagel RL, Lin MJ, Witkowska HE, Fabry ME, Bestak M, Hirsch RE.
Compound Heterozygosity for Hemoglobin and Korle-Bu: Moderate Microcytic
Hemolytic Anemia and Acceleration of Crystal Formation. Blood 1993; 82(6):
1907-1912.
Konotey-Ahulu FID, Gallo E, Lehman H, Ringelhann B. Hemoglobin Korle-Bu
(73 aspartic acidasparagine). J Med Genet 1968; 5: 107-111.
Case # 16
Hemoglobin O-Arab trait
A 23 year old male student of Northern African descent at Michigan State University,
East Lansing, Michigan, USA.
Laboratory Data:
390
Hemoglobin
14.8
RBC
4.9
MCV
86
MCH
29.3
MCHC
32.6
RDW
12.5
Platelet
273
Hb A
56.5%
Hb O-Arab
41%
Hb A2
2%
Hb F
0.5%
12.0 16.0 g/dL

3
4.0 5.5 Mil/mm
79-98 fL
26-34 pg
31-36%
11.5-14.5%
3
150-400 Th/mm
94.3-98.5
1.5 -3.7
0.0-2.0

Case # 16 Hemoglobin O-Arab trait
391
392

Summary of Results
393
Metho
d
Hb A
area
Alk
Agaros
e
Major
band
(Hb A)
Acid
Agar
Actually
on Hb S
side of
Hb A
Acid
Agaros
e
CZE
Major
band
Major
band
(Hb A)
Small
peak
(Hb F)
Zone 7
IEF
Hb S
area
(Hb OArab)
May
appear
as a
broader
Hb A
band
Major
band
(Hb OArab)
Actually
migrates
cathodal
to Hb S)
Major
band
Actually
on Hb C
side of
Hb S
(Hb OArab)
Major
peak
(Hb A)
Zone 9
Small
peak
(Hb F)
RT=1.08
Actually on
Hb S side
of Hb C
Minor
peak
(Hb OArab)
Zone 3
Major
band
Major
band
HPLC
Hb
A2/C
area
Major
band
(Hb A)
(Hb OArab)
Major
peak
(Hb A)
RT=2.38
Minor
peak
(Hb A2)
RT=3.64
Major
peak
(Hb OArab)
RT=4.3
Note: Separation under acidic conditions has traditionally been done with agar instead of
agarose because all the early descriptions of hemoglobin variants contained data collected in
this manner. As the separation quality of agar has deteriorated many vendors have chosen to
switch to use of the purified agarose in order to maintain a more constant product. Hemoglobin
O-Arab migrates close to hemoglobin A historically and with well selected current agars. In the
heterozygous state one broad band is seen starting with Hb A but extending on toward Hb S
somewhat. When agarose is substituted Hb O-Arab migrates close but on the Hb C side of the
Hb S location. The user is advised to take note of the separation media used for acid
electrophoresis in the interpretation of the results.
Since the concentration of the band that migrated at or near Hb C/E/O/A 2

position on alkaline electrophoresis was significantly > 10%, Hb A2
394
was ruled out because Hb A2 virtually never has such an increase. Citrate agar
electrophoresis (pH 6.2) eliminates the possibility of Hb C or Hb C-Harlem.

Incidentally the migration of Hb C-Harlem, Hb O-Arab and Hb E is virtually
identical upon IEF, therefore it was not helpful in the differentiation of these
variants. HPLC and CZE show characteristic elution times (RT) and migration
mobilities respectively for Hb O-Arab. In view of the characteristic laboratory tests
and normal peripheral blood smear a tentative diagnosis of Hb O-Arab trait was
made.
Hemoglobin O-Arab was first discovered (in association with Hb S) in an
Arabic-speaking Israeli village (Giser-A-Zarke), and thus got its name as Hb OArab. It is the same village Sayar reported the homozygous Hb O-Arab
(reference 2). It is emphasized here that Hb O-Arab is not prevalent in Israel or
the Jewish population. However, three homozygous Hb O-Arab cases from
progeny of parents who originally came from South Sudan were recently reported
[Sayar D. Clinical and Hematological Features of Homozygous Hemoglobin OArab (Beta 121 GluLys). Pediatr Blood Cancer 2013; 60: 506-507]. Hb O-Arab
has been found in Northern Africa (Tunisia), African-American, Saudi Arabia,
Bulgaria and the Mediterranean littoral. Hemoglobin O-Arab is a -chain variant
(121 GluLys), and exhibits no evidence of hemolysis and anemia in the
heterozygous state. Persons with Hb O-Arab trait are clinically well. Homozygous
Hb O-Arab exhibits a mild anemia. Hb O-Arab interacts with Hb S (double
heterozygous) and produces a disorder similar to homozygous Hb S disease with
all of its characteristic features. Hb O-Arab also interacts with -thalassemia, and
395
these individuals exhibit moderately severe hemolytic anemia and splenomegaly.
References
1
2
3
4
5
7
8
Bain BJ. Hemoglobin O-Arab, other significant hemoglobins. In:

Hemoglobinopathy Diagnosis, 2066, 2nd edition, Blackwell Publishing, pg 21315.
Sayar D. Clinical and Hematological Features of Homozygous Hemoglobin-O
Arab [Beta 121 GluLys]. Pediatr Blood Cancer 2013; 60: 506-507.
Zimmerman SA, OBranski EE, Rosse WF, Ware RE. Hemoglobin S/O Arab :
Thirteen New Cases and Review of the Literature. Am J Hematol 1999; 60:
279-284.
Sangore A, Sanogo I, Meite M, Ambofo Y, Abe Sopie V, Segbena A, Tolo A.
Hemoglobin O Arab Disease in Ivory Coast and West Africa. Medicine
Tropicale 1992; 52(2): 163-167.
Altay C, Gurgey A, Huisman Titus TJ. Homozygosity For Hemoglobin O-Arab
121 GluLys
(22
) Hb O-Arab Disease. The Turkish Journal of Pediatrics 1986;
28: 67-72.
Rachmilewitz EA, Tamari H, Liff F, Ueda Y, Nagel RL. The interaction of
+
hemoglobin O Arab with Hb S and thalassemia among Israeli Arabs. Hum
Genet 1985; 70: 119-125.
Ballas SK, Atwater J, Burka ER. Hemoglonin S-O Arab--Thalassemia: Globin
Biosynthesis and Clinical Picture. Hemoglobin 1977; 1(7): 651-662.
Milner PF, Miller C, Grey R, Seakins M, DeJong WW, Went LN. Hemoglobin
O Arab in four negro families and its interaction with hemoglobin S and
hemoglobin G. N Eng J Med 1970; 283(26): 1417-1425.
Case # 17 -Thalassemia trait

A 27 year old Caucasian female.
Laboratory Data:
396
Hemoglobin
12.5
RBC
5.13
MCV
63.8
MCH
20.5
RDW
12.1
Platelet
267
Serum Iron
110
Ferritin
75
Hb A
93.5
Hb A2
6.0
Hb F
0.5
12.0 16.0 g/dL

3
4.0 5.5 Mil/mm
79-98 fL
26-34 pg
11.5-14.5%
3
150-400 Th/mm
30-160 ug/dL
8-120 ng/mL
1.5 -3.7%
0.0-2.0%
Peripheral Blood Smear: Hypochromasia, microcytosis, target cells,

basophilic stippling
Sickle solubility test for hemoglobin S: Negative
Unstable hemoglobin test (isopropanol): Negative
397
398
Case #17 -Thalassemia trait

Summary of Results
399
Metho
d
Hb A
area
Alk
Agaros
e
Major
band
(Hb A)
Acid
Agar/
Agaros
e
Major
band
(Hb A +
Hb A2)
Major
peak
(Hb A)
Zone 9
CZE
Small
peak
(Hb F)
Zone 7
IEF
Hb S
area
Minor
peak
(Hb A2)
Zone 3
Minor
band
(Hb A2)
Minor
peak
(Hb A2)
RT=3.64
Major
band
(Hb A)
HPLC
Small
peak
(Hb F)
RT=1.05
Hb
A2/C
area
Minor
band
(Hb A2)
Major
peak
(Hb A)
RT=2.42
Note: HPLC retention time (RT) varies with the type of instrument used and
Alkaline agarose electrophoresis (pH 8.6) showed no abnormality except
that the staining of the band at the Hb A2 position was relatively denser than the
normal adult. No abnormal band was detected from citrate agar
electrophoresis (pH 6.2). Two bands in the migration position of Hb A (major
band) and Hb A2 (minor band but more intense than a normal adult) were
indicated by IEF. CZE and HPLC results were concordant suggesting
an increased concentration of Hb A2 and no other abnormal peaks.
Hemoglobinopathies can be classified as a manufacture of a modified
400
globin chain or a failure or decrease in the ability to manufacture a particular

globin chain. This latter set of conditions is referred to as a thalassemia. A
decreased ability to manufacture beta chains is called -thalassemia and results
in small erythrocytes (microcytosis) and a decreased amount of hemoglobin per
erythrocytes and thinness of the cell (hypochromasia). Due to insufficient beta
chains there is a surplus of alpha chains which bind to the red blood cell
membranes causing damage and an occasional clump of alpha chains is the
center of the Target Cells. The delta chains compete with the beta chains
present with the delta chains getting a larger proportion in this beta chain
deprived environment and this accounts for an elevated Hb A2.
The hematological and morphological parameters along with elevated
Hb A2 suggested the diagnosis of -thalassemia trait. In the presence of serum
iron deficiency Hb A2 can be falsely lower, therefore quantification should be
done again after the correction of the iron deficiency. -thalassemia trait is
clinically a benign condition most often found in persons of the Mediterranean,
Chinese, African American and other Asian ethnic groups. However problems
arise when a thalassemic gene is inherited from both parents, e.g. causing
Cooleys anemia (thalassemia major). Incidentally this disease was first
discovered in an Italian population by Dr. Denton Cooley in Detroit, Michigan,
USA.
A thorough review of the articles mentioned in the references is strongly
advised to make a correct diagnosis of various kinds of thalassemias (minor,
intermedia, and major) and also its interactions with several other hemoglobin
401
variants. Molecular characterization is necessary for genetic counseling when

both parents are carriers of -thalassemia minor or other hemoglobinopathies.
References
1
2
3
4
5
6
7
8
Galanello R, Origa R. Review: Beta-thalassemia. Orphanet J Rare Diseases

2010; 5(11): 1-15. http://www.ojrd.com/content/5/1/11
Cao A, Galanello R. Review: Beta-thalassemia. Genetics in Medicine
2010;12(2): 61-76.
Colah R, Gorakshakar A, Nadkarni A. Global burden, distribution and
prevention of -thalassemias and Hemoglobin F disorders. Expert Review of
Hematology 2010; 3(1): 103-116.
El Rassi F, Cappellini D, Inati A, Taher A. Beta-thalassemia Intermedia: An
Overview. Pediatric Annals 2008; 37(5): 302-328.
Bain BJ. -thalassemia trait. The , , and -thalassemia and related
conditions. In: Hemoglobinopathy Diagnosis, 2 nd Edition, Blackwell Publishing;
2006: 95-105.
Oliveri N, Weatherall DJ. Clinical Aspects of thalassemia. In: Steinberg MH,
Forget BG, Higgs DR, Nagel RL, eds. Disorders of Hemoglobin: Genetics,
Pathophysiology, and Clinical Management. Cambridge, England:
Cambriadge University Press; 2001: 277-341.
Weatherall DJ, Clegg JB. The thalassemias. In: The Thalassemia
Syndromes, 4th ed. Oxford, England: Blackwell Science, Ltd; 2001: 287-356.
Qari MH, Wali Y, Albagshi MH, Aishahrani M, Alzahrani A, Alhijji IA, Almomen
A, Aljefri A, Al-Saeed HH, Abdullah S, Al-Rustamani A, Mahour K, Mousa SA.
Regional consensus opinion for the management of beta thalassemia major
in the Arab Gulf Area. Orphanet J Rare Diseases 2013; 8: 143. Available from
http://www.ojrd.com/content/8/1/143
Case # 18 Hemoglobin S-+- thalassemia

A 37 year old African American female.
Laboratory Data:
Hemoglobin
RBC
10.3
4.28
12.0 16.0 g/dL

3
4.0 5.5 Mil/mm
402
MCV
74.8
MCH
23.9
RDW
16.2
Platelet
203
Hb A
22.1
Hb A2
6.4
Hb F
3.4
Hb S
68.1
79-98 fL
26-34 pg
11.5-14.5%
3
150-400 Th/mm
94.3-98.5%
1.5 -3.7%
0.0-2.0%
Peripheral Blood Smear: microcytosis, rare target cells,

moderate poikilocytosis
No record of blood transfusion during the past seven months.
Case # 18 Hemoglobin S- -thalassemia
403
Case # 18 Hemoglobin S- -thalassemia

404
Case # 18 Hemoglobin S-+-thalassemia

Summary of Results
405
Metho
d
Hb A
area
Hb S
area
Hb
A2/C
area
Minor
band
(Hb A2)
Alk
Agaros
e
Very
weak
band
(Hb F)
Small
band
(Hb A)
Major
band
(Hb S)
Acid
Agar/
Agaros
e
Weak
band
(Hb F)
Small
band
(Hb A+
Hb A2)
Major
band
(Hb S)
CZE
Small
peak
(Hb F)
Zone 7
Small
peak
(Hb A)
Zone 9
Major
peak
(Hb S)
Zone 5
Minor
peak
(Hb A2)
Zone 3
IEF
Very
small
band
(Hb F)
Very
small
band
(Hb A)
Major
band
(Hb S)
Minor
band
(Hb A2)
HPLC
Small
peak
(Hb F)
RT=1.05
Small
peak
(Hb A)
RT=2.40
Major
peak
(Hb S)
RT=4.32
Minor
peak
(Hb A2)
RT=3.6
Agarose gel electrophoresis (pH 8.6) showed one major band in the
position of Hb S, minor band (less in intensity than Hb S band) in the migration
position of Hb A, and another band in the position of Hb C/E/O/A 2 with intensity
greater than a normal adult. Citrate agar electrophoresis (pH 6.2) indicated
increased Hb F than a normal adult, and a major band in the position of Hb S,
major band in the position of Hb A but less in intensity than Hb S. IEF, CZE, and
HPLC also provided concordant results and the evidence for the following four
406
bands:
Hb S
Hb A
Hb A2
Hb F
Major band-positive sickle cell test

Major band-concentration less than Hb S
Minor band-concentration greater than a normal adult
Minor band concentration greater than normal adult
Quantitatively increased Hb A2 (6.4% by HPLC) suggested the presence of thalassemia in conjunction with Hb S. Two types of Hb S--thalassemia are
found in African Americans:
+
Type 1 with Hb A concentration 5-15%
Type 2 with Hb A concentration 20-30%
Hb S- -thalassemia:
Hb S- -thalassemia
+
This case represents Hb S- -thalassemia Type 2. It is emphasized here that

precaution is warranted in the interpretation at any time Hb A is less than Hb S.
This kind of situation can be encountered in homozygous sickle cell disease with
recent blood transfusion.
Hb S--thalassemia in African Americans is present in two clinically
significant conditions. If the Hb A is completely absent then it is termed Hb S0
-thalassemia, and is clinically similar to homozygous sickle cell disease.

If Hb A is present (Type 1 or Type 2) the person will have a milder
clinical course, and elevation of Hb F is also characteristic feature of Hb S+
-thalassemia.
Appendix:
0
I looked for a case of Hb S- -thalassemia for > three years but no luck. Just
yesterday a 23 year female came to the Emergency Department with a severe sickle
407
cell crisis. I immediately contacted the attending physician and my associates involved
in this book to include the data of this patient for the benefit of readers for understanding
the distinction between the two types of Hb S--thalassemias.
CBC
Hemoglobin
Hematocrit
RBC
MCV
MCH
RDW
Platelet
Reticulocyte Count
5.3
15.1
2.13
70.7
24.8
24.3
407
13.5
12.0 16.0 g/dL

35.0 - 48.0%
3
4.0 - 5.5 Mil/mm
79 - 98 fL
26 - 34 pg
11.5 - 14,5 %
3
150 - 400 Th/mm
0.5 - 1.5%
Alkaline agarose (pH 8.6), Citrate agar (pH 6.2), IEF, HPLC and CZE indicated the
absence of Hb A. The concentration of hemoglobin fractions from CZE were:
Hb A
Hb A2
Not Detected
4.5
Hb F
Hb S
8.6
86.9
On the basis of CBC and laboratory results a diagnosis of Hb S- -thalassemia

is most likely.
References
1.
2.
3.
4.
Bain BJ. Sickle cell/ thalassemia, Sickle cell hemoglobin and its interactions
with other variant hemoglobins and with thalassemia. In: Hemoglobinopathy
Diagnosis, 2006, 2nd edition. Blackwell Publishing,
England, pg 170-173.
Steinberg MH. Compound heterozygous and other sickle
hemoglobinopathies. In: Steinberg MH, Forget BG, Higgs DR, Nagle RL.
Disorders of Hemoglobin: Genetics, Pathophsiology and Clinical Management.
Cambridge , England: Cambridge University Press; 2001: 786-792.
Sunna EI, Gharaibeh NS, Knapp DD, Bashir NA. Prevalence of
Hemoglobins S and -Thalassemia in Northern Jordan. J Obstet Gynecol
Res 1996; 22(1): 17-20.
Gonzalez-Redondo JM, Kutlar A, Kutlar F, McKie VC, Mckie KM, Baysai
E, Huisman THJ. Molecular Characterization of Hb S (C) -Thalassemia
408
5.
in American Blacks. Am J Hematol 1991; 38: 9-14.

Serjeant GR, Ashcroft Mt, Serjeant BE, Milner PF. The clinical features of
sickle cell/-thalassemia in Jamaica. Br J Hematol 1973; 24: 19-30.
Case # 19 Hemoglobin C-0-thalassemia

A 17 year old female student from Turkey (most likely of Eti Turkish descent) visiting a
prestigious high school in Michigan to brush up her English. She was asymptomatic.
She declined to participate in athletic activities because she felt fatigue upon physical activity.
Low hemoglobin and MCV triggered hemoglobin electrophoresis by the physician.
409
Laboratory Data:
Hemoglobin
10.3
12.0 16.0 g/dL
3
RBC
5.4
4.0 5.5 Mil/mm
MCV
66.5
79-98 fL
MCH
20.9
26-34 pg
Hb A (all methods)
Not Detected
94.3-98.5%
Hb A2
5.9
1.5 -3.7%
Hb F
1.4
0.0-2.0%
Hb C
87.0
Hb A1c + other fractions
5.7
Peripheral Blood Smear: Hypochomosia, microcytosis, target cells,
anisocytosis and poikilocytosis
Case # 19 Hemoglobin C- -thalassemia
410
Case # 19 Hemoglobin C- -thalassemia

411
Case # 19 Hemoglobin C-o-thalassemia

Summary of Results
412
Method
Alk
Agarose
Hb A
area
Hb S
area
Faint
band
(Hb F)
Hb A2/C
area
Major
band
(Hb C +
Hb A2)
Acid Agar
/Agarose
Faint
band
(Hb F)
Major
band
(Hb C +
Hb A2)
CZE
Minor
peak
(Hb F)
Zone 7
IEF
HPLC
Minor
peak
(Hb F)
RT=1.0
5
Minor
peak
(Hb A2)
Zone 3
Major
peak
(Hb C)
Zone 2
Minor
band
(Hb A2)
Major
Hb C
band
cathodal
to A2
Minor
peak
(Hb A2)
RT=3.6
Major
peak
(Hb C)
RT=5.09
Note: Hb A was not detected by any of the six methods.
Alkaline agarose electrophoresis (pH 8.6) showed a single band in the

position of Hb C/E/O/Hb A2, thus indicating the absence of Hb A. Citrate agar
electrophoresis (pH 6.2) also indicated the absence of Hb A, and a major band
(87%) was indicated in the position of Hb C.
IEF showed a major band in the position of Hb C (cathodal to
Hb A2), minor band in the position of Hb A2 and a faint band in the position of Hb
413
F. CZE also indicated three peaks:

Hb C
Hb F
Hb A2
major peak in Zone 2

minor peak in Zone 7
minor peak in Zone 3 but obscured by
the larger peak in Zone 2 (Hb C)
HPLC separated the hemoglobins into three peaks, i.e. Hb C, Hb A2 and Hb F

and also provided quantitative results.
Increased Hb A2 (5.9%), absence of Hb A, microcytosis, target cells and a

major Hb C peak (87%) from HPLC suggested the presence of compound
heterozygosity for Hb C and -thalassemia. It is emphasized here that a
0
distinction between homozygous Hb C and Hb C- -thalassemia is not feasible

from alkaline agarose gel electrophoresis (pH 8.6) alone due to lack of the
correct quantitative value of Hb A2 because of the overlap of Hb C and Hb
+
A2 bands. Absence of Hb A as in this case rules out the possibility of Hb C- thalassemia.

Due to similarity in clinical features it is sometimes not possible to
0
differentiate with certainty between homozygous Hb C and Hb C- -thalassemia.

0
Similar clinical features of homozygous Hb C and Hb C- -thalassemia are:

Mild to moderate chronic hemolytic anemia, with hemoglobin levels in the range
of 8-12 g/dL and splenomegaly. The blood film shows large number of target
cells, folded cells, scattered spherocytes, hypochromia, microcytosis and
polychromasia.
0
The two parameters that lead to the putative diagnosis of Hb C- -
414
thalassemia in this case are MCV (55-70) and increased Hb A2.

However, the Hb A2 fraction could be overestimated and HbC/beta
syndromes are usually characterized by some hemolysis and

spelenomegaly while this patient is asymptomatic. Microcytosis could be
caused by alpha thalassemia and the genotype should be confirmed by
direct sequencing of the beta genes.
References
1
3
4
5
Kumar S, Rana M, Handoo A, Saxena R, Verma IC, Bhargava M, Sood SK.

Case report of Hb C/0-thalassemia from India. Int Jnl Lab Hem 2007; 29:
381-385.
Nagel RL, Steinberg MH, Hb S/C disease and Hb C disorders. In: Steinberg
MH, Forget BG, Higgs DR, Nagel RL. Disorders of Hemoglobin: Genetics,
Pathohysiology and Clinical Management.
Cambridge, England: Cambridge University Press; 2001; 756-785.
Fattoum S, Guemira F, Abdennebi M, Ben Abdeladhim A. [Hbc/betathalassemia association. Eleven cases observed in Tunisia]. Ann Pediatr
(Paris) 1993; 40(1): 45-8.
Maberrry MC, Mason RA, Cunningham G, Pritchard JA. Pregnancy
Complicated by Hemoglobin CC and C--Thalassemia Disease. Obstet
Gynecol 1990; 76: 324-327.
Ozsoylu S, Sipahioglu H, Altay F. Hemoglobin C-beta (O) thalassemia. Isr J
Med Sci 1989; 25: 410-412.
Case # 20
Hemoglobin Hasharon trait
A 55 year old male computer programmer with age onset diabetes mellitus, was screened for
hemoglobinopathy since one of his family member was anemic, and another having a
hemoglobin variant. His parents (Ashkenazi Jews) migrated from Poland to Detroit, Michigan.
415
Laboratory Data:
Hemoglobin
RBC
MCV
MCH
RDW
Platelet
Hb A
Hb A2 (CZE)
Hb F (CZE)
Hb variant (CZE)
14.8
5.1
88.0
28.3
12.3
203
77.3
1.6
0.8
20.3
12.0 18.0 g/dL

3
4.6 6.2 Mil/mm
80 - 100 fL
27 - 34 pg
11.5 -14.5%
3
15 - 400 Th/mm
94.3 - 98.5%
1.5 - 3.7%
0.0 - 2.0%
Peripheral Blood Smear: No abnormality was detected.

Hemoglobin instability (isopropanol) test was positive
but heat stability test was negative.
Case # 20 Hemoglobin Hasharon trait
416

417
Interpretation and Discussion

Summary of Results
418
Metho
d
Hb A
area
Hb S
area
Alk
Agaros
e
Major
band
Major band
slightly
toward C
Acid
Agar
Major
band
(Hb A)
(Hb A)
Acid
Agaros
e
CZE
Major
band
(Hb A)
Small
peak
Major
peak
(Hb F)
(Hb A)
Zone 7
Zone 9
(Hb
Hasharon)
Major band
slightly
toward A
(Hb
Hasharon)
Major band
directly in S
position
(Hb
Hasharon)
Major
peak
(Hb
Minor
peak
Zone 3
Hasharon)
(Hb A2)
Zone 5
IEF
Major
band
(Hb A)
Hb
A2/C
area
Minor
band
(Hb A2)
Major
band
slightly
cathodal
Minor
band
(Hb A2)
Very
small
peak
(Hb F)
RT=1.05
Major
peak
(Hb A)
RT=2.40
Major
peak
eluted
between S
and C
(Hb
Hasharon)
RT=4.74
(Hb-Hash
aron-A2)
A very
faint band
cathodal to
Hb C
Hb A2
of Hb S.
(Hb
Hasharon)
HPLC
Very
minor
peak
Zone 1
(2 2)
Minor
peak
(Hb A2)
RT=3.58
Alkaline agarose electrophoresis (pH 8.6) showed a major band in the

Hb A region, and another major band of lesser intensity cathodal to Hb S. A very
419
faint band was present in Hb A2 position. Citrate agar electrophoresis (pH 6.2)
also revealed two major bands with intensities equivocal to that described for
agarose gel electrophoresis (pH 8.6) in the respective Hb A and Hb S positions.
The sickle cell test was negative, ruling out the presence of Hb S.
IEF showed four bands: two intense bands and two faint bands. One
intense band in the position of Hb A, and a second band slightly cathodal to
Hb S. Additionally, there was a very faint band migrating in the Hb A2 position
and a second faint band in the delta chain variant position (cathodal to Hb C).
Hb A2 variants are due to the presence of an abnormal -chain as seen in Hb GPhiladelphia trait (Case # 5) or due to the presence of an abnormal delta chain.
Since this specimen also has an abnormal Hb A, the Hb A2 variant is likely due
to an alpha mutation.
CZE showed a major peak in the Hb A zone (Zone 9), and a lesser
intense peak than Hb A in Zone 5. Two minor peaks in the position of Hb F
(Zone 7) and Hb A2 (Zone 3) were also detected as well as a very small peak
in Zone 1.
HPLC showed a major peak for Hb A and two minor peaks for
Hb A2 and Hb F. Another major peak was detected between the Hb S and Hb
C window.
A narrative report was communicated to the attending physician with a
request for consultation with him to identify the exact hemoglobinopathy. The
420
physician communicated that the patient is an orthodox Ashkenazic Jew of Polish

origin.
Consistently typical migration patterns by the four electrophoretic
methods, elution retention times upon HPLC and the Ashkenazic Jewish ethnic
origin of the patient suggested the possibility of Hb Hasharon trait. Hb Hasharon
was first discovered in Hasharon Hospital, Israel in an Ashkanezic Jew, whose
father was from Poland and mother from Romania. It is -chain variant caused
by a mutation on condom 47 that results in the substitution of aspartic acid by
histidine (47 AspHis).
The presence of Hb Hasharon is typically found in Ashkanezic Jews (who
have also migrated to several countries after World War II), and Italians from
Ferrara district of Italy. Hb Hasharon has not been recognized in Sephardic
Jews. No consistent clinical and hematological abnormalities are associated with
Hb Hasharon. It is innocuous hemoglobinopathy, however some patients have
indicated drug-Induced (sulfonamide, dapsone) hemolytic anemia.
The percentage of Hb Hasharon varies between Ashkenezic Jews and the
subjects of the Ferrara district of Italy. The Hb Hasharon concentration in Italians
of Ferrara district origin is usually in the range of 30-35%. Contrary to this the
Ashkanezic Jews have the Hb Hasharon concentration in the range of 15-20%.
The DNA studies have determined that this difference is because of an
underlying -thalassemia (-thalassemia-2) in Italians of Ferrara area. Thus,
these individuals have both an alpha chain mutation and an alpha deletion. The
421
Ashkanezic Jews have no evidence of the presence of -thalassemia-2 trait.
References
1
2
3
4
5
Unstable hemoglobin variants, Martin H. Steinberg, MD, www.uptodate.com

2013 UpTodate.
http://www.uptodate.com/contents/unstable-hemoglobin-variants
Zur B, Ludwig M, Stoffel-Wagner B. Hemoglobin Hasharon and Hemoglobin
NYU in subjects of German origin. Biochemia Medica 2011; 21: 321-25.
Chinelato-Fernandes AR, Mendiburu CF, Bonini-Domingos CR. Utilization of
different methodologies for the characterization of Hb Hasharon
heterozygotes. Genet Mol Res 2006; 5: 1-6.
Eliakim R, Rachmilewitz EA. Hemoglobinopathise in Israel. Hemoglobin 1983;
7: 479-85.
Mavilio F, Marinucci A, Fontanarosa PP, Tentori L, Cappellozza G.
Hemoglobin Hasharon [2 47(CD5) AspHis2] linked to -Thalassemia in
Northern Italian carriers. Acta Haemat. 1980; 63: 305-311.
del Senno L, Bernardi F, Marchetti G, et al. Organization of globin genes
47
and mRNA translation in subjects carrying hemoglobin Hasharon ( Asp
replaced by His) from the Ferrara region (Northern Italy). Eur J Biochem
1980; 111(1): 125-130.
Alberti R, Mariuzzi GM, Marinucci M, Bruni E, Tentori L. Hemoglobin
Hasharon in a north Italian community. J Med Genet 1975; 12: 294-98.
Case # 21
Hemoglobin Zurich trait
A 18 year old white female student from Grand Rapids, Michigan. Parents migrated from
Europe, but no information available about their ethnicity and country of origin.
Her physical examination was normal.
Laboratory Data:
422
Hemoglobin
RBC
MCV
MCH
RDW
Platelet
Hb A
Hb A2 (HPLC)
Hb F (HPLC)
Hb variant
11.6
4.4
102.0
28.4
12.3
228
66.0
1.6
0.8
31.6%
12.0 16.0 g/dL

3
4.0 5.5 Mil/mm
79 - 98 fL
26 - 34 pg
11.5 -14.5%
3
15 - 400 Th/mm
94.3 - 98.5%
1.5 - 3.7%
0.0 - 2.0%
Peripheral Blood Smear: Macrocytic red blood cells. Serum iron and ferritin were
normal; very mild anemia with slight reticulocytosis.
No congenital deficiency of glucose-6-phosphate dehydrogenase.
Case # 21 Hemoglobin Zurich trait
423

424

Summary of Results
425
Metho
d
Hb A
area
Hb S
area
Alk
Agaros
e
Acid
Agar
Major
band
Major
band
(Hb A)
Hb
A2/C
area
Minor
band
(Hb A2)
Major
band
(Hb A +
Hb A2+ Hb
variant)
CZE
IEF
HPLC
Very
minor
peak
(Hb F)
RT=1.06
Major
peak
Zone 9
(Hb A)
Major
band
Major
peak
Zone 6
Minor
peak
Zone 3
Major
band
Minor
band
(Hb A)
Slightly
cathodal to
Hb S
(Hb A2)
(Hb A2)
Major
peak
(Hb A)
RT=2.34
Major
peak
RT=3.55
*The hemoglobin variant eluted with Hb A2, and Hb A2 was in the normal range from
CZE.
Note: HPLC retention time (RT) varies with the type of instrument used and several
Please note that we do not have data of acid agarose electrophoresis separation
at this time.
Agarose gel electrophoresis (pH 8.6) showed two major bands; one at
the migration position of Hb A, and one at the Hb S position. In addition,
a very weak band was noticed in the position of Hb C/E/O/Hb A2. Citrate
426
agar electrophoreis (pH 6.2) showed only one band in the position of Hb A and
a very weak, smudged band in the position of Hb F.
In view of the negative sickle cell test and the migration patterns of the
alkaline and acid electrophoresis, the presence of Hb S was ruled out. Similarly,
the possibility of Hb G-Philadelphia and Hb D- Los Angeles was eliminated on
the basis of the positive (isopropanol) instability test, the absence of G 2 band of
Hb G-Philadelphia and a lower percentage of the variant as compared to Hb DLos Angeles. Hemoglobin electrophoresis on this patient was performed about
seven years ago in our laboratory. At that time, neither the HPLC nor the CZE
testing instruments were available in our laboratory. After consultation with the
attending physician, the specimen was sent to a reference laboratory for globin
chain analysis by reverse phase HPLC and DNA studies. Note: The IEF, CZE
and HPLC scans inserted here in this case are adopted from other sources for
educational purposes.
The globin chain analysis and DNA studies provided the correct
identification of the hemoglobin variant (31.6% from alkaline agarose
electrophoresis) as Hb Zurich. Hb Zurich is an unstable hemoglobin and found
only in the heterozygous state. It is caused by the substitution of histidine by
arginine at the 63rd position of the -chain [22
63
(E7) HisArg]. Hb Zurich
was initially found in Europeans of Swiss descent, but later this variant
was reported in Japanese, and Brazilian citizens of non-Swiss ancestry.
427
Physicians should be advised of possible induced or exacerbated

hemolysis in subjects with Hb Zurich by exposure to oxidant drugs, e.g.
sulfonamides, sulfones, phenacitin-like analgesics, and most of the local
anesthetics.
References
1.
2.
3.
4.
5.
6.
7.
8.
Unstable hemoglobin variants. Steinberg MH. www.uptodate.com 2013

UpTodate. August 2013.
http://www.uptodate.com/contents/unstable-hemoglobin-variant
Aguinaga MdP, Wright CJ, Roa PD, Terrel F, Turner EA, Houston M.
Molecular Diagnosis and Characterization of Hb Zurich [63(E7)HisArg]
Carriers in a Kentucky Family. Hemoglobin 1998; 22 (5 & 6): 509-515.
Harano T, Harano K, Nagasaka I, Yamasaki S. Hb Zurich [63(E7)HisArg]
Found in a Japanese Woman. Hemoglobin 1996; 20 (4): 429-434.
Miranda SRP, Kimura EM, Saad STO, Costa FF. Identification of Hb Zurich
[2263(E7)HisArg] by DNA Analysis in a Brazilian Family. Hemoglobin
1994; 18 (4 & 5): 337-341.
Zinkham WH, Winslow RM. Unstable Hemoglobins: Influence of Environment
on Phenotypic Expression of a Genetic Disorder. Medicine 1989; 68(5): 309320.
Zinkham WH, Houtchens RA, Caughey WS. Relation between variations in
the phenotypic expression of an unstable hemoglobin disorder (hemoglobin
Zurich) and carboxyhemoglobin levels. Am J Med 1983; 74: 23-29.
Murata K, Yamamoto S, Hirano Y, Omine Mitsuhiro O, Tsuchiya J, Ohba Y,
Miyaji T. First Japanese Family with the Unstable Hemoglobin Zurich
[63(E7)HisArg]. Jap J Med 1982; 21 (1): 40-45.
Dickerman JD, Holtzman NA, Zinkman WH. Hemoglobin Zurich. A Third
Family Presenting with Hemolytic Reactions to Sulfonamides. Am J Med
1973; 55: 638-642.
Case # 22
Hemoglobin Lepore trait
A 41 year old male employee of General Motors, Detroit, Michigan. Parents migrated
from Italy.
Laboratory Data:
Hemoglobin
14.3
13.5 -18.5 g/dL

428
RBC
MCV
MCH
RDW
Platelet
5.72
69
22.4
13.2
243
4.6 - 6.2 Mil/mm

80 -100 fL
27 34 pg
11.5 -14.5%
3
150 - 400 Th/mm
Hb A
Hb A2
Hb F
Hb variant (CZE)
80.7
2.1
5.4
11.8
94.3 - 98.5%
1.5 - 3.7%
0.0 - 2.0%
Peripheral Blood Smear: Microcytosis, hypochromasia, target cells, poikilocytosis.

Patient was not transfused during the past four months.
Case # 22 Hemoglobin Lepore trait
429
430

Summary of Results
431
Method
Hb A
area
Hb S
area
Alk Agarose
Major
band
(Hb A)
Major
band
(Hb
Lepore)
Acid
Agar/Agaros
e
Major
band
(Hb A+
Hb
Lepore
+ Hb
A2)
CZE
Small
peak
(Hb F)
Zone 7
IEF
HPLC
Small
peak
(Hb F)
RT=1.0
5
Hb
A2/C
area
Minor
band
(Hb A2)
Major
peak
(Hb A)
Zone 9
Major
peak
(Hb
Lepore)
Zone 6
Minor
peak
(Hb A2)
Zone 3
Major
band
(Hb A)
Major
band in Hb
G position
(Hb
Lepore)
Minor
band
(Hb A2)
Major
peak
Hb A2 &
Hb Lepore
(RT=3.5)
Major
peak
(Hb A)
RT=2.4
5
.
Since the sickle cell test was negative and also no band was observed in
the area of Hb S upon acid electrophoresis, the presence of Hb S was ruled out.
The possibility of other commonly encountered variants (e.g. Hb D, Hb G, Hb
Russ) that exhibit alkaline and acid electrophoresis pattern similar to this case
432
were ruled out. Hb D has a concentration of approximately 40% in the

heterozygous state, and this variant quantified at 11.8% by CZE. Hb G was ruled
out due to the absence of the -chain variant ( G2 2) band/peak by
electrophoretic methods (alkaline agarose, IEF and CZE). This variant was
associated with microcytosis, while Hb Russ is not.
With IEF the hemoglobin variant migrated exactly in the position of Hb G,
even though the presence of Hb G was ruled out on the basis of the absence of
the delta chain Hb G band (see above). Similarly differential diagnosis of the
hemoglobin variant with CZE was not helpful due to the overlap of several
hemoglobins in zone 6.
HPLC showed increased intensity of the Hb A2 peak (RT=3.5), which was
inconsistent with other electrophoretic methods. This suggested that the
hemoglobin variant eluted with Hb A2. Another thing the HPLC ruled out was the
presence of the other common variants exhibiting electrophoretic patterns similar
to this case (Hb D, Hb G, Hb Russ), because none of these variants elute with
Hb A2.
In view of the thalassemic peripheral blood picture, low concentration of
the variant (11.8%), and the separation data a diagnosis of Hb Lepore trait was
made.
433
Hb Lepore is hybrid (fused globin chains) hemoglobin consisting of two

-globin chains and two - hybrid chains. In the - hybrid chain the N-terminal
position of the -chain joins at the C-terminal of the -chain. There are three
genetically controlled - chains fusions (hybrid formation) that are
characterized by their fusion points of the amino acids in the chains. This
characteristic fusion of and chains leads to the following variants of Hb
Lepore:
i)
Hb Lepore-Boston
[(1-87) (115-146)]
In this case of Hb Lepore, the hybrid - chain consists of the

first 87 amino acids of the -chain and the last 32 amino acids
of the -chain. This variant is also called Hb Lepore-Washington.
ii)
Hb Lepore-Baltimore
[(1-50) (86-146)]
iii)
Hb Lepore-Hollandia
[(1-22) (50-146)]
Among these three variants, Hb Lepore-Boston is found with some

frequency in people of Mediterranean descent.
Lepore- Boston, (Lepore-Washington) migrates in the same position as
Hb S in alkaline conditions. However Lepore Hollandia and Lepore Baltimore
migrate slightly slower than Hb S in alkaline conditions (Bain, BJ, Wild BJ,
Stephens AD and Phelan L. Variant Hemoglobins: A Guide to identification;
2010: Wiley-Blackwell Publishing).
To the best of our knowledge both CZE and HPLC do not differentiate
among these variants.
434
All the Lepore traits have the same clinical symptoms. Hb Lepore trait is a
stable hemoglobin but exhibits features of thalassemia minor due to the
decreased production of -chains. -thalassemia intermedia or major type
disorder is exhibited by homozygous Hb Lepore. Hb Lepore interacts with Hb S
to give Hb S/Hb Lepore, however very few cases (<18) were reported in the
literature. Hb S/Hb Lepore patients exhibited mild microcytic anemia, and
clinically were either asymptomatic or with complications generally associated
with Hb S disease. A case of Hb E interaction with Hb Lepore-Hollandia was
described in the literature but without any significant clinical condition except
microcytic anemia without the need for transfusion.
References
1.
2.
3.
4.
5.
6.
McKeown SM, Carmichael H, Markowitz RB, Kutlar A, Holley L, Kutlar F.

Rare Occurrence of Hb Lepore-Baltimore in African Americans: Molecular
Characteristics and Variations of Hb Lepores. Ann Hematol 2009; 88: 545548.
Pasanga J, George E, Nagaratnam M. Haemoglobin Lepore in a Malay
family: a case report. Malasian J Pathol 2005; 27(1): 33-37.
Shaji RV, Edison ES, Krishamoorthy R, Chandy M, Srivatava A. Hb
Lepore in the Indian Population. Hemoglobin 2003; 27: 7-14.
Viprakasit V, Pung-Amritt P, Suwanthon L, Clark K, Tanphachitr VS.
Complex interactions of [delta] [beta] hybrid haemoglobin (Hb Lepore Hollandia)
Hb E([beta]26G>A) and [alpha]+ thalassemia in a Thai Family. Eu J Haematol 2002;
68-107-12.
Goncalves I, Henriques A, Raimundo A, Picanco I, Reis A, Correia Jr E,
Santos E, Nogueria P, Osorio-Almedia L. Fetal Hemoglobin Elevation in
Hb Lepore Heterozygotes and its correlation with Globin Cluster Linked
Determinants. Am J Hematol 2002; 69: 95-102.
Forget BG. Molecular mechanism of beta-thalassemia. In: Steinberg MH,
Forget BG, Higgs DR, Nagel RC, eds. Disorders of Hemoglobin: Genetics,
435
7.
8.
9.
10.
11.
Pathophysiology and Clinical Management. Cambridge, England:

Ropero P, Gonzalez FA, Sanchez J, Anguta E, Asenjo S, Arco AD, Murga
MJ, Ramos R, Fernandez C, Villegas A. Identification of the Hb Lepore
phenotype by HPLC. Haematologica 1999; 84: 1081-1084.
Romana M, Diara JP, Merghoub T, Leclard L, Saint-Martin C, Berchel C,
Merault G. Hemoglobin Sickle-Lepore: An Unusual Case of Sickle Cell
Disease. Acta Haematologica 1997; 98: 170-71.
Fairbanks VF, McCormick DJ, Kubik KS, Rezuke WN, Black D, Ochaney
MS. Hb S/Hb Lepore with Mild Sickling Symptoms: A Hemoglobin Variant with
Mostly Delta-Chain Sequences Ameliorates Sickle-Cell Disease. Am J Hematol
1997; 54: 164-165.
Waye JS, Eng B, Patterson M, Chui DHK, Chang LS, Coglonis B,
Poon AO, Oliveri NF. Hb E/Hb LeporeHollandia in a Family From
Bangladesh. Am J Hematol 1994; 47: 262-265.
Hemoglobin Sickle-Lepore: Report: Report of Two Siblings and Review of
the Literature. Am J Hematol 1993; 44: 192-195.
Case # 23
Hemoglobin J-Oxford trait
A 55 year old male farmer from Saginaw, Michigan. His mother belonged to a French
settlement in Newfoundland, Canada. Ancestors from father side immigrated from
Norway. No abnormality was found from an annual medical examination except a slight
elevation of serum cholesterol.
436
Laboratory Data
Hemoglobin
RBC
MCV
MCH
Platelet
14.8
5.1
90.7
29.9
279
13.5 - 18.5 g/dL

3
4.6 - 6.2 Mil/mm
80 - 100 fL
27 - 34 pg
3
150 - 400 Th/mm
Hb A2
Hb F
Hb A
Hb Variant
2.2
0.8
72.0
25.0 %
1.5 - 3.7%
0.0 - 2.0%
94.3 98.5%

Case # 23 Hemoglobin J-Oxford trait
437

438

439
Summary of Results
Metho
Hb A
d
area
Alk
Agaros
e
Major
band as
much
anodal to
Hb A as
Hb S is
cathodal
to Hb A
(Hb J)
Acid
Agar/
Agaros
e
Hb S
area
Major
band
(Hb A)
Hb
A2/C
area
Very
faint
band
(Hb A2)
See note
below
Major
band
(Hb A +
Hb A2+
Hb J)
CZE
Major
peak
(Hb J)
Zone 12
Major
peak
(Hb A)
Zone 9
IEF
Major
band
anodal
to Hb A
(Hb J)
Very
small
peak
(Hb F)
RT=1.05
Major
band
(Hb A)
HPLC
Major
peak
(Hb A)
RT=2.44
Barely
visible
peak
(Hb "JOxfordA2")
Zone 6
Very
minor
peak
(Hb A2)
Zone 3
Very
minor
band
(Hb A2)
Major
peak
(Hb J)
RT=1.62
Barely
visible
peak
(Hb A2)
RT=3.64
Note: A faint band due to Hb "A2-J" was detected in the position of Hb S,

when a concentrated hemolysate was used for heavier application
in alkaline agarose electrophoresis (pH 8.6).
In the later stages of our investigation, when Hb J was considered in this

440
case, the alkaline agarose gel electrophoresis (pH 8.6) was repeated with a
heavier application, and a faint Hb "A2-J" band was detected in the area of Hb S.
The reason for this additional test was to rule out a beta chain variant form of Hb
J. Because the intensity of the fast band is far less than that of the Hb A band
this variant is most likely an -chain variant.
During my discussion with the attending physician the following points
were brought to his attention:
i
We suspect an Hb J (-chain variant) trait, probably Hb J-Oxford

trait [15(A13) GlyAsp].
ii
There are >50 Hb J variants that are known in the literature. In

addition to that there are > 24 Hb variants which are not designated
as Hb J variant but exhibit electrophoretic mobilities akin to Hb J.
Most of these Hb J variants are entirely without any clinical or
hematological manifestations.
Note: As of today 57 hemoglobins are designated Hb J by
electrophoretic mobility and they are roughly divided equally
between and chain variants. Six of these are unstable
and one has increased oxygen affinity.
iii)
There are three Hb J variants reported in the literature as

associated with clinical disorders:
a)
Hb J-Altgeld (unstable hemoglobin hemolytic anemia)
b)
Hb J-Cape Town (erythrocytosis due to high oxygen affinity)

441
Hb J-Buda (erythrocytosis resulting from interaction with

Hb G-Pest in persons doubly heterozygous for Hb J-Buda
and Hb G-Pest)
iv)
The exact identification of the Hb J trait variant in the patient is not

necessary since the patient is clinically normal. Further testing to
designate the type of Hb J (DNA sequencing, LC-Mass) may be
deferred indefinitely due to exorbitantly associated cost.
At the time this patient was analyzed, both the CZE and HPLC testing
facilities were not available in our laboratory. For instructional purposes, we have
illustrated the CZE and HPLC scans of another established Hb J-Oxford trait
patient. Both the CZE (major peak in zone 12) and HPLC (minor peak at RT=
1.62) provided concurrent evidence for Hb J trait (-chain variant).
Hb J was first discovered in 1956 (Thorup OA, Itano HH, Wheyby M,
Leavll BS. Hemoglobin J. Science 1956; 123: 889-90), and the 57 variants found
so far are roughly divided equally between -chain and -chain variants. Hb J
variants are rarely found, but have been reported from the USA, northern
European countries, China and Japan. The only Hb J variant which is
encountered with any notable frequency, is Hb J-Baltimore (Case # 24).
Hb H, J, I, N, K, Camden and Hope are designated "fast hemoglobins" in

442
view of their faster mobility on agarose gel electrophoresis (pH 8.6). Additional
cases of other "fast hemoglobins" will be included in the 2nd edition of the book.
References
1.
2.
3.
Caruso D, Crestani M, Riva LD, Mitro N, Giavarini F, Mozzi R, Franzini C.

Mass spectrometry and DNA sequencing are complementary techniques for
characterizing hemoglobin variants: the example of hemoglobin J-Oxford.
Haematologica 2004; 89(5): 608-609.
Joutousky A, Hadzi-Nesic J, Nardi MA. HPLC Retention Time as a Diagnostic
Tool for Hemoglobin Variants and Hemoglobinopathies: A study of 60 000
Samples in a Clinical Diagnostic Laboratory. Clin Chem 2004; 50(10): 17361747.
Harano K, Harano T, Shibata S, Mori H, Ueda S, Imai K, Ohba Y, Irimajiri
K. Hb J-Oxford [15(A13) Gly---Asp] in Japan. Hemoglobin 1984; 8(2):
197-198.
Case # 24
Hemoglobin J-Baltimore trait
A 33 year old male, residing in Windsor, Canada, whose ancestors migrated from
Europe. While donating blood, his hemoglobin was found to be low, therefore his family
physician ordered hemoglobin electrophoresis.
443
Laboratory Data
Hemoglobin
RBC
MCV
MCH
Platelet
12.8
4.9
86.0
28.3
232
13.5 - 18.5 g/dL

3
4.6 - 6.2 Mil/mm
80 - 100 fL
27 - 34 pg
3
150 - 400 Th/mm
Hb A2
Hb F
Hb A
Hb Variant (HPLC)
2.3
3.8
55.2
38.7 %
1.5 - 3.7%
0.0 - 2.0%
94.3 98.5%

Case # 24 Hemoglobin J-Baltimore trait
444

445

446
Summary of Results
Metho
d
Alk
Agaros
e
Hb A
area
Major
band as
much
anodal to
Hb A as
Hb S is
cathodal
to Hb A
(Hb J)
Acid
Agar/
Agaros
e
Major
band
(Hb A)
Hb S
area
Note:
Both
major
bands of
equal
intensity
Hb
A2/C
area
Very
faint
band
(Hb A2)
Major
band
(Hb A +
Hb A2+
Hb J)
CZE
Major
peak
(Hb J)
Zone 12
Major
peak
(Hb A)
Zone 9
IEF
Major
band
anodal
to Hb A
(Hb J)
Major
band
(Hb A)
HPLC
Small
peak
(Hb F)
RT=1.05
Major
peak
(Hb A)
RT=2.4
Note:
Minor
peak
(Hb A2)
Zone 3
Minor
band
(Hb A2)
Note:
Both
major
bands of
equal
intensity
Major
peak
(Hb J)
RT=1.8
Very
small
peak
(Hb A2)
RT=3.6
Since Hb J-Baltimore is a -chain variant, therefore the major bands

(Hb A and Hb J-Baltimore) are approximately in equal concentration.
Agarose gel electrophoresis (pH 8.6) showed two major bands. One band
was in the position of Hb A. Another major band was approximately as much
447
anodal to Hb A as Hb S is cathodal to Hb A. Visually, the intensity of both of

these two bands was similar. No other band was detected besides a minor band
for Hb A2. There are several fast moving hemoglobin variants. Among them are
Hb H and Hb I, which migrate much faster towards anode than Hb J (see Case #
23). Similarly, Hb N also migrates slightly faster than Hb J. There are several chain variants that exhibit similar migration patterns as this patient. These chain variants are usually present in a 1:3 ratio relative to Hb A. Since the
intensity of the two major bands on alkaline agarose electrophoresis was similar,
it was suggestive of a -chain variant in this case.
Citrate agar electrophoresis (pH 6.2) showed only one major band in the
position of Hb A. IEF also showed two major bands, i.e. one in the position of Hb
A and another more anodal to it in the position of Hb J. The attending physician
was consulted about the clinical condition of the patient. Since no abnormality
was noted by the physician except a slightly lower hemoglobin, a report was
submitted advising the presence of a harmless Hb J- trait without any
hematological or clinical consequences..
For instructional purposes, we have included the CZE and HPLC scans of
Hb J-Baltimore (-chain variant), which is the most prevalent -chain variant
among the category of Hb J variants (both and chains).
CZE showed a major peak (approximately 40%) in zone 12 (Hb JBaltimore), and HPLC also showed a major peak at a retention time of 1.8
448
minutes (Hb J-Baltimore). We suspect that this case was most likely a
representative of Hb J-Baltimore. Confirmation by DNA studies, globin chain
analysis, and LC-Mass spectrometry would be necessary for definitive diagnosis.
However, for financial reasons, all these additional tests are not required in view
of the benign status of the hemoglobinopathy in the patient.
Hb J-Baltimore (also called J-Trinidad, J-Ireland, J-Georgia) is a -chain
variant [ 16(A13) GlyAsp] and is encountered rarely in Afro Americans, and
very rarely in Europeans.
References
1.
2.
3.
4.
Arribalzaga K, Ricard MP, Carreno DL, Sanchez J, Gonzalez A, Ropero P,

+
Villegas A. Hb J-Baltimore [16(A13)GlyAsp] Associated with Thalassemia in a Spanish Family. Hemoglobin 1996; 20(1): 79-84.
Landin B, Jeppsson J-O. Rare -Chain Hemoglobin Variants Found in
Swedish Patients During Hb A1c Analysis. Hemoglobin 1993; 17(4): 303-318.
Vandenesch F, Baklouti F, Francina A, Vianey-Liaud C, Bertrand A, Le
Devehat C, Delaunay J. Hemoglobin J-Baltimore [16(A13)GlyAsp]:
Interference with the assay of Hb A1c. Clin Chim Acta 1987; 168(2): 121-28.
Musumeci S, Schiliro G, Fisher A, Musco A, Marinucci M, Mavilio F,
Fontanarosa PP, Tentori L. Hb J-Baltimore [16(A13)GlyAsp] in Association
with -Thalassemia in a Sicilian Family. Hemoglobin 1979; 3(6): 459-464.
Case # 25 Hemoglobin Malmo trait

An 18 year old high school student from Warren, Michigan was hurt during football
practice and brought to the Emergency Department of the hospital. He had a ruddy face
and complained of pain in the lower extremities.
Laboratory Data:
449
Hemoglobin
RBC
MCV
MCH
Platelets
Hb A (HPLC)
Hb A2
Hb F
Hb Variant (HPLC)
20.1
6.8
88.0
29.7
270.0
56.0
2.0
4.0
38%
13.5 -18.5 g/dL

3
4.6 - 6.2 Mil/mm
80 - 100 fL
27 - 34 pg
3
150 400 Th/mm
94.3 - 98.5%
1.5 - 3.7%
0.0 - 2.0%
Peripheral Blood Smear: Crowding of erythrocytes.

450
451

Note: Acid agarose data was not available for Hb Malmo.
452
Method
Alk
Agarose
Acid Agar
Minor
Hb F
band
detected
CZE
IEF
Faint
band
in
Hb F
area
HPLC
Major
band
slightly
anodal
to Hb A
band
Minor
Major
peak
peak
(Hb F)
(Hb
RT=1.03 Malmo)
RT=1.66
Hb A
area
Major
band
(Hb A +
Hb
Variant)
Major
band
(Hb A +
Hb
Variant)
Major
peak
(Hb A +
Hb
Malmo)
Zone 9
Major
band in
Hb A
position
Hb S
area
Hb A2/C
area
Very
faint
band
(Hb A2)
Minor
peak
(Hb A2)
Zone 3
Faint
band in
Hb A2
area
Major
peak
(Hb A)
RT=2.44
Minor
peak
(Hb A2)
RT=3.6
*Note:
HPLC retention times varies with the type of the instrument used and several
other factors, e.g. temperature, etc.
CZE was not available in our laboratory when we encountered this case,
therefore the CZE scan provided here is of another Hb Malmo patient. However
453
the CZE scan of a Hb Malmo trait was not helpful in the identification of the
variant since both the Hb A and Hb Malmo migrated together in zone 9.
IEF did separate Hb Malmo from Hb A in a pattern identical with a proven
case of Malmo hemoglobinopathy reported in the literature. Hb Malmo migrated
slightly anodal to Hb A, and the migration mobility was much less than some fast
hemoglobins (e.g. Hb J-Oxford, Hb J-Baltimore, Hb N, Hb I, etc).
HPLC provided two clearly separated major peaks and three small peaks
of Hb F, Hb A1c and Hb A2. One major peak was due to Hb A (RT= 2.43
minutes), and another with a faster elution time (RT=1.66), was due to the
hemoglobin variant of this case. In summary, Hb Malmo elutes with Hb A or
before, depending on the chromatographic system used. In our system it eluted
before Hb A.
We were aware that occasionally erythrocytosis has been found to be
associated with high-oxygen affinity hemoglobins, but we did not have the
capability to determine a hemoglobin-oxygen dissociation curve and its p50 ( the
point on the curve where the hemoglobin molecule is half-saturated with oxygen).
Normally, the hemoglobin-oxygen dissociation curve is sigmoid-shaped.

High affinity hemoglobins, e.g. Hb Malmo, show a markedly leftward shifted
curve (p50 of about 13 torr compared to normal values of 26-30 torr) resulting in
454
a hyperbolic shape. The oxygen delivery to the tissues is impaired whenever the
oxygen affinity is high (low p50). Erythropoietin production is stimulated, which in
turn increases the red cell mass, resulting in erythrocytosis.
After consultation with the attending physician, a narrative report was
submitted stating that a hemoglobin variant is present and in view of marked
erythrocytosis a possibility of a high affinity hemoglobin cannot be ruled out.
Fortunately, the parents of the patient agreed to provide their blood for
analysis. The mother was found to have a normal CBC and hemoglobin pattern.
The father, who had immigrated from Sweden to USA belonged to a family with
known erythrocytosis. Some years ago, when he complained of fatigue,
headaches and lethargy a diagnosis of Hb Malmo was made in Sweden. In order
to relieve his symptoms, phelebotomy was performed. The electrophoretic
(alkaline, acid, IEF) results and HPLC curve were identical for both the father and
son.
In view of the ancestral background and the laboratory results on both the
patient and the parents, a putative diagnosis of Hb Malmo was made. The
attending physician and the family were advised that the hemoglobin disorder
455
was essentially benign. However, the patient should refrain from smoking and
should be followed periodically for any signs of fatigue, headaches or lightheadedness.
More than 100 high oxygen affinity hemoglobin variants are reported in
the literature. Hb Malmo is a member of this class and is the result of the
substitution of glutamine for a histidine amino acid at the 97 th amino acid of the
chain [97(FG4)HisGln]. This mutation is in the area of the peptide chain that
moves during the oxygenation deoxygenation process. The substitution inhibits
movement in such a manner that deoxygenation becomes more difficult and
deters transfer to the tissue to the point that the patient would become anemic if
the body did not compensate by making excess erythrocytes. The amino acids
from position 94 through 103 constitute a nonhelical section of the beta chain and
mutations effecting ionicity of those positions effects the spacing between the
alpha and beta chains near the point of oxygenation. This area of the globin
chain is called the FG segment or FG corner and a list of these variants is
found in Table 1 (courtesy of Hoyer & Kraft, College of American Pathologists).
The fit between the alpha and beta chains is critical because the gap becomes
narrower when oxygen is attached to the ferrous iron and expands as oxygen is
released. A second region of the beta chain (amino acids 143 through 146 on
456
the Carboxy end of the molecule) also effects this spatial control. Several
hemoglobin variants have been identified as possessing mutations in this area
and thus assisting in understanding the synchronous action involved in the
oxygenation / deoxygenation process. Table 2 is a list of these variants
(also courtesy of Hoyer & Kraft, College of American Pathologists).
In all the cases on this list except for Heathrow and Brigham the
mutation effects the shape of the globin chains such that the electrophoretic
mobility is altered so the mutations are not silent (personal communication, Rita
Ellerbrook, PhD, Helena Laboratories, USA).
Hb Malmo only exists in the heterozygous state. The homozygous state
has not been reported and is thus most probably incompatible with life.
Reference
1
2
3
4
5
Bain BJ. High-affinity Hemoglobins. In: Hemoglobinopathies Diagnosis,

Blackwell Publishing, Oxford, United Kingdom. 2006, 224-226.
Steinberg MH. Genetic disorders of hemoglobin oxygen affinity.
www.uotodate.com 2013 UpToDate
Fernandez FAG, Villegas A, Ropero P, Carreno MD, Anguita E, Polo M,
Pascual A, Henandez A. Hemoglobinopathies with high oxygen affinity.
Experience of Erythropathology Cooperative Spanish Group. Ann Hematol
2009; 88: 235-238.
Giordano PC, Harteveld, Brand A, Willems LNA, Kluin-Nelemans HC, Plug

RJ, Batelaan DN, Bernini LF. Hb Malmo[-97(FG-4) HisGln] leading to
polycythema in a Dutch family. Ann Hematol 1996; 73: 183-188.
Landin B, Berglund S, Wallman K. Two Different Mutations in Codon 97 of the
-Globin Gene Cause Hb Malmo in Sweden. Am J Hematol 1996; 51: 32-36.
457
6
7
Girino M, Riccardi A, Mosca A, Paleari R, Bonomo P. Double Heterozygosity

for Hemoglobin Malmo [97 (FG4) HisGln] and -Thalassemia Traits.
Haematologica 1989; 74: 187-90.
Boyer SH, Charache S, Fairbanks VF, Maldonado JE, Noyes A, Gayle EE.
Hemoglobin Malmo -97 (FG-4) HistidineGlutamine: A Cause of
Polycythemia. J Clin Invest 1972; 51: 666-676.
Table 1. Chain Variants in FG Corner and G Helix

Position
Helical #
Substitution
Name
Effect
94
FG1
AspHis
Barcelona
polycythemia
458
AspAsn
Bunbury
normal
95
FG2
LysAsn
LysGlu
Detroit
normal
N-Baltimore normal
97
FG4
HisGln
HisLeu
Malmo
Wood
98
FG5
ValMet
Koln
polycythemia
polycythemia
ValGly
ValAla
hemolysis, O2
Affinity
Nottingham hemolysis
Djelfa
(?)
99
G1
AspAsn
AspHis
AspAla
AspTyr
AspGly
AspVal
Kempsey
Yakima
Radcliffe
Ypsilanti
Hotel Dieu
Chemilly
polycythemia
polycythemia
polycythemia
polycythemia
polycythemia
polycythemia
100
G2
ProLeu
Brigham
polycythemia
101
G3
GluGly
GluGln
GluAsp
GluLys
Alberta
Rush
Potomac
British
Columbia
polycythemia
hemolysis
polycythemia
polycythemia
102
G4
AsnLys
AsnThr
AsnSer
AsnTyr
Richmond
Kansas
Beth Israel
Saint Mande
normal
cyanosis
cyanosis
cyanosis
103
G5
PheLeu
Heathrow
polycythemia
Table 2. Chain Variants Near the C-Terminus

Position
Helical #
Substitution
Name
459
Effect
143
144
H21
HC1
HisArg
HisGln
HisPro
HisAsp
Abbruzzo
Little Rock
Syracuse
Rancho
Mirage
polycythemia
polycythemia
polycythemia
(?)
HisTyr
Old
Dominion
normal CBC
O2 affinity
LysAsn
Andrewpolycythemia
Minneapolis
LysGlu
Mito
polycythemia
145
HC2
TyrHis
TyrCys
TyrAsn
TyrStop
Bethesda
Rainier
Osler
McKees
Rocks
polycythemia
polycythemia
polycythemia
polycythemia
146
HC3
HisAsp
HisPro
HisArg
Hiroshima
York
Cochin-Port
Royal
Cowtown
Kodaira
polycythemia
polycythemia
(?)
HisLeu
HisGln
polycythemia
polycythemia
The contents of these tables are presented from Hoyer JD, Kroft SH, eds. Color Atlas of
Hemoglobin Disorders: A Compendium Based on Proficiency Testing. Northfield, IL:
College of American Pathologists: 2003 (Reproduced with Permission)
Case # 26
Hemoglobin Koln trait
A 18 year old female student. No ancestral information was available. Physical examination
revealed scleral icterus and spleen palpable 4 cm below left costal margin.
Laboratory Data:
Hemoglobin
RBC
MCV
RDW
10.7
3.9
106
13.6
12.0 -16.0 g/dL

3
4.0 - 5.5 Mil/mm
79 - 98 fL
11.5 -14.5%
460
Platelet
235
150 400 Th/mm
WBC
Reticulocyte
Serum Iron
Total Bilirubin
Indirect Bilirubin
Hb A (HPLC)
Hb A2 (HPLC)
7.1
2.9
39
2.8
2.4
74.8%
2.2
4.0 11.0 Th/mm

0.7 - 1.8 %
30 160 ug/dL
0.0 1.5 mg/dL
0.0 0.4 mg/dL
1.5 - 3.7%
Peripheral Blood Smear: Mild macrocytic anemia with slight hypochromasia,

polychromasia, and occasional target cells.
Case # 26 Hemoglobin Koln trait
461
462

Summary of Results
463
Metho
d
Hb A
area
Hb S
area
Alk
Agaros
e
Major
band
Hb A
Broad
smudge
from Hb S
through
Hb C
Acid
Agar /
Agaros
e
CZE
Major
band
(All
Hbs)
Major
peak
(Hb A +
Zone 9
Minor
peak
(Hb
Koln)
Zone 6
Major
band
(Hb A)
Broad
smudge of
Hb Koln
from Hb
Hb Kln)
IEF
A-Hb A2
HPLC
Major
peak
(Hb A)
RT=2.39
Two
barely
visible
minor
peaks (Hb
Koln)
RT=4.54.8
Hb
A2/C
area
No discrete
band
detected
Minor
peak
(Hb
Koln)
Zone 4
Minor
peak
(Hb A2
+ Hb
Koln)
Zone 3
Barely
visible
minor
band
Hb A2
Minor
peak
Hb A2
RT=3.6
Minor
peak
(Hb
Koln)
RT=4.95
# Due to the instability of Hb Koln, several minor bands were noticed in HPLC. A similar
phenomenon was observed in other electrophoretic methods except in the acid
agar/agarose electrophoresis, where only one major band was detected.
Alkaline agarose gel electrophoresis (pH 8.6) showed a major band in the
position of Hb A and a smudged band migrating on both the anodal and cathodal
sides of Hb S extending further towards the area of Hb C. Citrate agar
electrophoresis (pH 6.2) showed only one band in the position of Hb A. IEF
showed an intense band in the position of Hb A, and a broad smear between Hb
464
A and Hb A2 (similar to a fresh painting with a brush). CZE showed a major peak
in zone 9 (Hb A area) and three minor peaks in zones 3, 4 and 6. HPLC showed
a major peak at a retention time (RT) of 2.39 minutes (Hb A), and minor peaks
from a RT of 3.5 5.0 minutes.
The laboratory tests indicated that either the hemoglobin variant did
not separate from Hb A, or it formed discrete peaks, or it formed blurred bands.
Since the hemoglobin instability test was positive, the attending physician was
advised of the possibility of an unstable hemoglobin variant.
In August 2013, Professor Steinberg reviewed (reference # 1) the current
clinical and hematological characteristics of unstable hemoglobins. According to
this recent review, approximately 140 of the1028 known mutations of hemoglobin
were found to be unstable. To date, it is not feasible to identify an unstable
hemoglobin variant (either alpha-, beta-, gamma, and delta-globin chains
abnormalities) by the commonly used laboratory methods.
In most worldwide laboratories, Hb E, Hb H, Hb Hasharon and Hb
Koln are the most frequently reported unstable hemoglobin variants. Three of
these variants (Hb E, Hb H, and Hb Hasharon) were excluded in our patient on
the basis of their electrophoretic mobilities and retention times on HPLC. Hb E
(Case # 14a) migrates in the position of Hb A2/E/O/C. Hb H is a fast migrating
hemoglobin variant (more anodal to Hb A) on alkaline agarose gel
electrophoresis (pH 8.6). Hb Hasharon (Case # 20) migrates in the Hb S area on
465
both the alkaline and acid electrophoresis.

Hemoglobin Koln, a prevalent unstable hemoglobin, has a rather
atypical electrophoretic migration pattern, which is helpful in its identification
in conjunction with the associated clinical and hematological manifestations of
the patient.
In Hb Koln, a valine amino acid at the 98th position of the -chain
(98ValMet) is substituted by the amino acid methionine. Since both the valine
and methionine amino acids are neutral amino acids, there would be no net
change in the charge. However Hb Koln does separate from Hb A on agarose
gel electrophoresis (pH 8.6). This anomaly is explained by the modification at the
site of -chain contact with the heme molecule, which causes a quaternary
structure change of the hemoglobin molecule. As a consequence of this,
especially during alkaline electrophoresis, Hb Koln loses heme groups from the
abnormal -chain and thus loses negative charge. The hemoglobin migrating in
the Hb S position on alkaline electrophoresis is essentially des-heme hemoglobin
Koln.
The diagnosis of Hb Koln was substantiated by the following observations:
i)
ii)
iii)
iv)
v)
vi)
vii)
Unstable hemoglobin
Negative sickle cell solubility test for Hb S
Minimal or no anemia
Splenomegaly and regenerative erythrocyte changes even in the
absence of anemia
Hypochromasia and macrocytosis are usually evident
Increased oxygen affinity
Atypical migration pattern on alkaline agarose gel electrophoresis
(pH 8.6), IEF, CZE and multiple peaks on HPLC.
466
In view of the above considerations, the physician was advised of the

possibility of Hb Koln. Due to the autosomal dominant mode of familial
transmission, the parents of the patient should also be analyzed for the
confirmation of Hb Koln.
Generally, patients with Hb Koln are asymptomatic unless complications
develop, e.g. blow to the left side of the abdomen by contact sports, exacerbation
of hemolysis during infection (e.g. upper respiratory tract infection), or after use
of some medications, e.g. sulfonamides. Hb Koln exists only in the heterozygous
state; life is not compatible with homozygous Hb Koln.
Hb Koln and Hb Hasharon (Case # 20) are two of the many unstable
hemoglobins that have been identified. Unstable hemoglobins may be suspected
in patients presenting with the symptoms of congenital non-sherocytic anemia
with splenomegaly and pigmented gallstones, or with hemolytic anemia in which
the red cells contain Heinz bodies and are sensitive to oxidant drugs as
sulfonamides, or with mild anemia in which the reticulocyte count is elevated
compared to the amount of hemoglobin, or with a peripheral blood smear
containing target cells, basophilic stippling, a few anisocytes and
hypochromic red cells. If the condition was present in a newborn, the cause
might be a gamma chain variant whose effect would be eliminated as beta chain
production increased. Likewise, an asymptomatic newborn, in which clinical
condition began to develop at 4 to 6 months of age might be a beta chain variant.
467
A suspicious peripheral blood smear would lead to a Heinz body test and
hemoglobin instability (isopropanol) test, but the later two tests should not be run
in the presence of Hb F levels >5%.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Steinberg MH. Unstable hemoglobin variants. www.uptodate.com

UpToDate. Literature review current through August 2013.
Chang YH, Hur M, Lee DS, Park SS, Kim BK, Park S, Ohba Y, Hattori
Y, Cho HI. The first case of Hb Koln [98(FG5)ValMet] in Korea.
Hemoglobin 1999; 23(3): 287-289.
Chang J-G, Yang T-Y, Perng L-I, Wang J-C, Tsan K-W. Hb Koln
[98(FG5)ValMet] : The first case found in a Chinese family. Hemoglobin
1998; 22 (5&6): 535-536.
Landin B, Frostad B, Brune M, Ljung R. Haemoglobin Koln as de
novo mutations in Sweden: Diagnosis by PCR and specific enzymatic
cleavage. Eur J Haematol 1994; 52:156-61.
Indrak K, Brabec V, Wilson JB, Webber BB, Huisman THJ. Hb Koln or
2298(FG5)VALMet in a Czechoslovakian family. Hemoglobin
1991; 15(1& 2): 133-135.
Ohba Y. Unstable hemoglobins. Hemoglobin 1990; 14: 353-388.
Bird AR, Karabus CD, Hartley PS, Lehman H. Haemoglobin Koln in
Cape Town. A case reprt. S Afr Med J 1987; 72: 154-156.
Gurgey A, Altay C. Hemoglobin Koln [ 98(FG5) ValMet] in a
Turkish child. The Turkish Journal of Pediatrics 1982; 24: 271-73.
Ricco G, Ravazzolo R, Rege-Cambrin G, Capaldi A, Trento M, Leechi
M, Sartori ML, Furlani C, Rietto GB, Rabino-Massa E. Koln
haemoglobinopathy in Italy. Pan. Med 1981; 23:227-233.
Stirling M. Koln Haemoglobinopathy in a Second Scotish Family. Scott
Med J 1980; 25: 121-125.
Egan EL, Fairbanks VF. Postsplenectomy Erythrocytosis in
Hemoglobin Koln Disease. N Eng J Med 1973; 288: 929-931.
Hallen J, Charlesworth D, Lehmann H. Haemoglobin Koln in a Jewish
Family. Acta Med. Scand. 1972; 191: 177-180.
Luan Eng L-I, Lopez CG, Eapen JS, Eravelly J, Wiltshire BG,
Lehmann H. Unstable Haemoglobin Koln Disease in Members of a
Malay Family. J Med Genetics 1972; 9: 340-43.
468
Case # 27 Hemoglobin Q-India trait

27 year old male from India. No abnormality detected. Physical examination was
unremarkable.
Laboratory Data
Hemoglobin
RBC
14.8
5.4
13.5 - 18.5 g/dL

3
4.6 - 6.2 Mil/mm
469
MCV
MCH
Platelet
86.0
28.3
232
Hb A2
Hb F
Hb A (HPLC)
Hb Variant (HPLC)
1.1
0.8
78.8
19.3 %
80 - 100 fL
27 - 34 pg
3
150 - 400 Th/mm
1.5 - 3.7%
0.0 - 2.0%
94.3 - 98.5%

Case # 27 Hemoglobin-Q India trait
470
Case # 27 Hemoglobin-Q India trait

471
Case # 27 Hemoglobin Q-India trait

Summary of Results
Metho
Hb A
Hb S
472
Hb
area
area
Alk
Agaros
e
Major
band
(Hb A)
Major
band
slightly
towards
Hb A side
of Hb S
(Hb QIndia)
Acid
Agar
Major
band
between
Hb A and
Hb S (Hb
A + Hb
Q-India)
Hb A and
Hb QIndia
combine
as a
broader
band
Acid
Agaros
e
Major
band
(Hb A +
Hb QIndia)
Exactly
in the
position
of Hb A
CZE
Major
peak
(Hb A)
Zone 9
Major
peak
(QIndia)
Zone 6
IEF
Major
band
(Hb A)
Major
band
(Hb QIndia)
HPLC
Major
peak
(Hb A)
RT=2.39
Major
peak
(Hb QIndia)
RT=4.7
A2/C
area
Faint
band
(Hb A2)
Slightly
anodal
towards
Hb S
Minor
peak
(Hb A2)
Zone 3
Slightly
anodal
towards
Hb A
Minor
peak
(Hb QIndia +
Hb A2 )
Zone 1
Faint
band
(Hb A2)
Faint
peak
(Hb A2)
RT=3.6
The diagnosis of Hb Q-India trait is cumbersome on alkaline

electrophoresis and may be mistaken for Hb S/D/ Lepore because the band in
473
the Hb S area is close to Hb S. Sickling or sickle solubility tests should be

employed to rule out Hb S. These and other variants are characterized through
additional laboratory tests (HPLC and IEF). Molecular analysis is required for
establishing a definitive diagnosis. This can be achieved through ARMS-PCR,
RFLP-PCR and mass spectrometry. However, first two of these are fraught with
technical pitfalls and should be done with inclusion of appropriate controls to
acquire correct results. LC/ESI-MS offers a rapid and unambiguous
characterization of individual chains. DNA sequence is currently the most
accurate way of identifying Hb Q-India in a blood sample.
Review of Hemoglobin Q
Bushra Moiz, PhD
Introduction
Hb Q was first reported in 1958 in a Chinese patient (1). Since then a number of
cases have been described in Asians. It is a rare hemoglobinopathy resulting from a
single point mutation (GACCAC) of -1 globin gene present on chromosome 16. The
resulting hemoglobin is modified structurally at polypeptide chain replacing aspartic
acid by histidine. Depending on the implicated codon, three variants have been
described namely Hb Q-India ( 64 AspHis), Hb Q-Thailand ( 74 AspHis), and Hb
Q-Iran ( 75 AspHis). Using computerized models for protein structure, it was
observed that there is no difference between the predicted secondary structures of
normal -globin and that of Hb Q-India (2). In contrast, Hb Q-Iran carries an extra helix
while Hb Q-Thailand carries two extra helices. The predicted results of tertiary structure
474
also support these findings (2). Since the residue and hence charge changes involve
the surface of the hemoglobin tetramer, the properties of the hemoglobin molecule are
not affected (3).
Hb Q-Thailand [74(EF3) AspHis] is often found in Thailand, China, and other
Southeast Asian countries (4). It has several synonyms including Hb Mahidol, QChinese, G-Taichung, Kurashiki, and Asabara (5). The alpha-Q-Thailand gene is
strongly linked to gene deletion and has important implications in the identification and
diagnosis of hemoglobinopathies and thalassemias. Subjects with Hb Q-Thailand
invariably show microcytosis as the variant is invariably linked to (-
4.2
). However,
individuals who are doubly mutated for Hb Q-Thailand and thalassemia may be more
severely anemic (6,7). More complex interaction of Hb Q-Thailand with Hb E, Constant
Spring and hereditary persistence of fetal hemoglobin has been described in the
literature (8-10).
Hb Q-Iran was first described in 1970 by Lorkin et al (11) and later by Rahimi in
an Iranian individual (12).
Hb Q-India was first reported in 1972 by Sukumaran in a Sindhi family (13). Later
reports were published by Dash (14), Abraham (3) and Desai (15); their observations
came from Sindhi and Punjabi families. Hb Q-India usually occurs in the heterozygous
state (
Q-India
/ and /), however double heterozygotes with both (-
and /) and -thalassemia (
Q-India
Q-India
/ and / ) were reported (3, 14). A novel
Hb D-Punjab / Hb Q-India was recently reported in an Indian diabetic (16). No report of

a homozygous state has ever been published.
475
Clinical manifestation
The presence of Hb Q does not impart any functional deficit since the
hemoglobin is not altered structurally at its tertiary level (8). Hb Q is a stable
molecule and has normal oxygen affinity. Therefore, Hb Q is clinically silent
in a heterozygous individual. In contrast, subjects who are compound
heterozygotes with other hemoglobinopathies exhibit a thalassemic phenotype.
For example, co-inheritance of Hb Q-India and -thalassemia results in a mild
anemia (17). Similarly, Hb Q-H disease caused by the co-inheritance of Hb Qo
Thailand and -thalassemia [--/- ] presents with a chronic, hemolytic anemia

with associated jaundice and splenomegaly (18). The severity of anemia may
warrant blood transfusions and or splenectomy (18). Subjects with a single copy
of Hb Q-Iran do not show any distinctive clinical manifestation (19). Interestingly,
a report from Turkey described a subject with homozygous Hb Q-Iran who was
clinically asymptomatic (20).
Diagnostic laboratory tests and interpretation

CBC and peripheral blood smear
An individual heterozygote with Hb Q-Thailand usually shows a
slight microcytosis. A thalassemic blood picture similar to Hb H disease is
observed in Hb Q-H disease (18). The peripheral blood smear in this
disorder showed anisocytosis, poikilocytosis, nucleated red cells and
target cells (6). Intracellular crystals were also observed in red cells in
brilliant cresyl blue preparations (6) similar to Hb H disease.
476
Hb Q-India usually demonstrates normochromic normocytic red cell

indices with normal or near normal hemoglobin (3). However, a few cases
with mild anemia and microcytosis have been reported in the literature
(10). Hb Q-India with co-inherited -thalassemic trait usually shows
hypochromic microcytic red cell indices with mild anemia (14,17).
Hb Q-Iran shows normal red cell indices with normal hemoglobin
irrespective of zygosity (12,20).
Alkaline agarose gel, citrate agar and acid agarose electrophoresis
Hb Q migrates in the position of Hb S/D/Lepore on agarose gel
o
and cellulose acetate electrophoresis (pH 8.6). Presence of chains

lead to the appearance of accessory bands corresponding to abnormal
hemoglobin and hence double bands of Hb A2 can be observed (14).
Thus, Hb Q can easily be misinterpreted as Hb S if confirmatory testing
such as sickling or sickle solubility test are not performed. On citrate
agar (pH 6.2), it migrates between Hb A and Hb S. Upon electrophoretic
migration in agarose medium at pH 6.2, Hb Q migrates exactly in the
position of Hb A.
HPLC
.The separation of hemoglobin variants depends on their retention
times (21). Hb Q-India, Iran and Thailand exhibited retention times similar
to that of other -chain variants in the range of 4.76 to 4.78 minutes (3). In
477
a heterozygote, Hb Q-India and Iran usually eluted as 17-19% of the total

hemoglobin. In contrast, Hb Q-Thailand represented 30-35% of the total
hemoglobin in the heterozygote state, because of the gene deletion
accompanying it (5). Hb Q % may be further decreased with
concomitant iron deficiency or with co-inheritance of -thalassemia
trait (14,17).
IEF
Hb Q-India focused in the position of Hb S. However, both the Hb
Q-Iran and Hb Q-Thailand migrated slightly anodal to Hb S, i.e. towards
Hb A.
Mass spectrometry
Liquid chromatography electrospray ionization mass spectrometry
(LC/ESI-MS) has the advantage of providing molecular information in
individual polypeptide chains and of the hemoglobin molecule (23).
LC/ESI-MS has been described in recent years to evaluate unknown Hb
variants including Hb Q-India (24). It detects the presence of a mutant chain differing in mass from a normal -chain by 22 DA. The later is
assigned to a mutation of an aspartic acid residue to a histidine residue
thus identifying Hb Q. The site can be identified by tandem-mass analysis
of a tryptic digested fragment encompassing residues V62-K90 of
hemoglobin -chain. Sequencing these fragments can establish the
diagnosis of Hb Q-India.
478
ARMS-PCR
This technique can be used for the successful detection of various
hemoglobin variants including Hb Q-India (3). This technique is based on
the amplification of allele specific primers because of 3-terminal matches
and mismatches. The methodology is simple, rapid and inexpensive;
however, it is non-specific since either sub-optimal amplification or
deteriorating primers can lead to false positive results (25).
RFLP-PCR
Recently, a restriction enzyme digestion assay was employed for
the diagnosis of Hb Q-India (22). Restriction enzyme EaeI was utilized in
RFLP-PCR since Hb Q-India abolishes the recognition site of this enzyme.
It can be used as a simple, robust and alternative method to ARMS-PCR
for DNA diagnosis of Hb Q-India. However, any other rare variant that
abolishes the same EaeI restriction site would also be detected. Hence,
RFLP-PCR can be used as an adjuvant test after HPLC and or IEF for
primary diagnosis of Hb Q-India.
Gene sequencing
This is the most definitive technique for identifying a hemoglobin
variant. Recently, Bhat described DNA sequencing in a patient with Hb QIndia (26). This methodology of sequence electrophoretogram clearly
demonstrates the specific location of the mutation of Hb Q-India.
479
Not only did it show that the codon of GAC encoding aspartic acid was
mutated to the codon CAC encoding for histidine, but it also depicted the
zygosity of the patient (26).
References
1
2
3
4
5
6
7
9
10
11
12
13
Vella F, Wells RH, Ager JA, Lehmann H. A haemoglobinopathy involving

haemoglobin H and a new (Q) haemoglobin. Br Med J 1958; 1: 752-755.
Yadav AK. Comparative analysis of protein structure of common Hb Q
variants. Indian J Pathol Microbiol 2010; 53: 696-698.
Abraham R, Thomas M, Britt R, Fischer C, Old J. Hb Q-India: an uncommon
variant diagnosed in three Punjabi patients with diabetes is identified by a
novel DNA analysis test. J Clin Pathol 2003; 56: 296-299.
Higgs DR, Hunt DM, Drysdale HC, Clegg JB, Pressley L, Weatherall DJ. The
genetic basis of Hb Q-H disease. Br J Haematol 1980; 46: 387-400.
Hoyer JD, Kroft HS, editors. Color Atlas of Hemoglobin Disorders. A
Compendium Based on Proficiency Testing, 159 pp, Northfield, Illinois,
College of American Pathologists, 2003.
Lieinjo LE, Pillay RP, Thuraisingham V, Further Cases of Hb Q-H disease (Hb
Q-alpha thalassemia). Blood 1966, 28: 830-839.
Beris P, Huber P, Miescher PA, Wilson JB, Kutlar A, Chen SS, Huisman TH.
Hb Q-Thailand Hb H disease in a Chinese living in Geneva, Switzerland:
Characterization of the variant and identification of the two alpha-thalassemic
chromosomes. Am J Hematol 1987; 24: 395-400.
Sanchaisuriya K, Chunpanich S, Fucharoen S, Fucharoen G, Sanchaisuriya
P, Changtrakun Y. Association of Hb Q-Thailand with homozygous Hb E and
heterozygous Hb Constant Spring in pregnancy. Eur J Haematol 2005; 74:
221-227.
Li D, Liao C, Li J, Xie X, Zhong H. Association of Hb Q-Thailand with
heterozygous Hb E in a Chinese patient. Hemoglobin 2008; 32: 319-321.
Zheng W, Liu Y, Chen D, Rong K, Ge Y, Gong C, Chen H. Complex
interaction of Hb Q-Thailand and Hb E with alpha (0)-thalassemia and
hereditary persistence of fetal hemoglobin in a Chinese family. Ann Hematol
2010; 89: 883-888.
Lorkin PA, Charlesworth D, Lehmann H, Rahbar S, Tuchinda S, Eng Li. Two
haemoglobins Q, alpha-74 (EF3) and alpha-75(EF4) aspartic acid to histidine.
Br J Haematol 1970; 19: 117-125.
Rahimi Z, Aktamipour R, Vaisi-Raygani A, Nagel RL, Muniz A. An Iranian child
with Hb Q-Iran [alpha75(EF4)AspHis]/-alpha3.7 kb/IVSII.1 GA]: first
report. J Pediatr Hematol Oncol 2007; 29: 649-651.
Sukumaran PK, Merchant SM, Desai MP, Wiltshire BG, Lehmann H.
Haemoglobin Q India [alpha64(E13) aspartic acid histidine] associated with
480
14
15
16
17
18
19
20
21
22
23
24
25
26
beta-thalassemia observed in three Sindhi families. J Med Genet 1972; 9:

436-442.
Dash S, Huisman TH. Hemoglobin Q-India [64(E13) AspHis] and beta
thalassemia: a case report from Punjab (North India). Eur J Haematol 1988;
40: 281.
Desai DV, Dhanani H, Kapoor AK, Yeluri SV. Hb Q-India in a Sindhi family: an
uncommon hemoglobin variant. Lab Hematol 2004; 10: 212-214.
Higgins T, Schnabl K, Savoy M, Rowe P, Flamini M, Bananda S. A novel
double heterozygous Hb D-Pubjab / Hb Q-India hemoglobinopathy. Clin
Biochem 2012; 45: 264-266.
Moiz B, Moatter T, Hashmi MR, Hashmi N, Kauser T, Nasir A, Khurshid M.
Identification of Hemoglobin Q India (alpha 1-64 AspHis) through ARMSPCR. First report from Pakistan. Ann Hematol 2008; 87: 385-389.
Leung KF, ma ES, Chan AY, Chan LC. Clinical phenotype of haemoglobin QH disease. J Clin Pathol 2004; 57: 81-82.
Rahimi Z, Rezei M, Nagel RL, Muniz A. Molecular and hematological analysis
of hemoglobin Q-Iran and hemoglobin Setif in Iranian families. Arch Iran Med
2008; 11: 382-386.
Ozdag H YI, Akar N. First observation of homozygote Hb Q-Iran [alpha
75(EF4) AspHis]. Turk J Hematol 2008; 25: 48-50.
Joutovsky A, Hadzi-Nesic, Nardi MA. HPLC retention time as a diagnostic tool
for hemoglobin variants and hemoglobinopathies: a study of 60 000 samples
in a clinical diagnostic laboratory. Clin Chem 2004; 50: 1736-1747.
Khalil MS, Henderson S, Schuh A, Hussein MR, Old J. The first use of Eael
restriction enzyme in DNA diagnosis of Hb Q-India. Intl J Lab Hematol 2011;
33: 492-497.
Wild BJ, Green BN,, Cooper EK, Lalloz MR, Erten S, Stephens AD, Layton
DM. Rapid identification of hemoglobin variants by electrospray ionization
mass spectrometry. Blood Cells Mol Dis 2001; 27: 691-704.
Mandal AK, Bisht S, Bhat VS, Krishnaswamy PR, Balaram P. Electrospray
mass spectrometric characterization of hemoglobin Q (Hb Q-India) and a
double mutant hemoglobin S/D in clinical sampes. Clin Biochem 2008; 41: 7581.
Old JM, Khan SN, Verma I, Fucharoen S, Kleanthous M, Ioannou P, Kotea N,
Fisher C, Riazuddin S, Saxena R, Winichagoon P, Kyriacou K, Al-Qudbaili F,
Khan B. A multi-center study in order to further define the molecular basis of
beta-thalassemia in Thailand, Pakistan, Sri Lanka, Mauritius, Syria, and India,
and to develop a simple molecular diagnostic strategy by amplification
refractory mutation system-polymerase chain reaction. Hemoglobin 2001; 25:
397-407.
Bhat VS, Dewan KK, Krishnaswamy PR, Mandal AK, Balaram P.
Characterization of a hemoglobin variant: Hb Q-India / IVS 1-1 [G>T]-beta
thalassemia. Indian J Clin Biochem 2010; 25: 99-104.
481
Case # 28
Hemoglobin Dhofar trait
A 24 year old male, belonging to Qara tribes from the Dhofar region of the Sultanate of
Oman. The patient had not been transfused during the past six months.
Laboratory Data:
Hemoglobin
RBC
MCV
MCH
RDW
Platelet
Hb A
Hb A2 (HPLC)
13.0
5.1
68
22.4
13.2
243
81.5
4.1
13.5 -18.5 g/dL

3
4.6 - 6.2 Mil/mm
80 -100 fL
27 34 pg
11.5 -14.5%
3
150 - 400 Th/mm
94.3 - 98.5%
1.5 - 3.7%
482
Hb F
Hb Dhofar (HPLC)
0.8
13.6
0.0 - 2.0%
Peripheral Blood Smear: Microcytosis, hypochromasia, target cells.

Note: In HPLC, Hb A2 is slightly under estimated due to overlap with Hb Dhofar peak,
but in heterozygotes , Hb A2 is found to be raised if quantified by capillary zone
electrophoresis or elution after cellulose acetate electrophoresis.
Case # 28 Hemoglobin Dhofar trait
483
484

485
Summary of Results
Metho
d
Hb A
area
Hb S
area
Alk
Agaros
e
Major
band
(Hb A)
Medium
size
band
(Hb
Dhofar)
Acid
Agar
Major band
( Hb A +
Hb Dhofar)
Hb A and
Hb Dhofar
combine as
a broader
band
Acid
Agaros
e
Major
band (Hb
A + Hb
Dhofar)
Major
band
broadened
by Hb
Dhofar
CZE
Major
peak
(Hb A)
Zone 9
Medium
size peak
(Hb
Dhofar)
Zone 5
Minor
peak (Hb
A2) Zone
3
IEF
Major
band
(Hb A)
Medium
size
band
(Hb
Dhofar)
Minor
band
(Hb A2)
Major
peak
(Hb A)
RT=2.35
Medium
size peak
(Hb
Dhofar)
RT=4.04
Minor
peak (Hb
A2)
RT=3.6
HPLC
*Note:
Hb
A2/C
area
Minor
band
(Hb A2)
HPLC retention time (RT) varies with the type of the instrument used and
Alkaline agarose gel electrophoresis (pH 8.6) showed a major band in the
486
position of Hb A and a medium size band in the area of Hb S/G/D/Lepore/Korle

Bu, and a few other variants. Citrate agar electrophoresis (pH 6.2) showed one
major band in the area of Hb A and no other major band was detected. Hb S
was ruled out due to the negative sickle cell solubility test and absence of a band
in the Hb S area on acid electrophoresis. Hb Korle Bu was also ruled out due to
the absence of a band in the Hb G region on IEF. Hb G-Philadelphia was ruled out due
to the absence of a G2 band on alkaline agarose electrophoresis and IEF. Both Hb D
and Hb Lepore migrate in the area of Hb G on IEF, therefore the possibility of these two
variants was also ruled out; Hb Dhofar migrates in the area of Hb S on IEF.
It is emphasized here that the identification of Hb Dhofar by alkaline and
acid electrophoretic methods alone is not prudent. Secondly, the range of Hb
Dhofar (26 59% in homozygous and compound heterozygous and 8.8 21.5.%
in heterozygous) can be a confounding factor in its differentiation with Hb D-Los
Angeles trait (Case # 12).
HPLC was informative since Hb Dhofar eluted at a retention time slightly
longer than Hb A2 and not in the Hb S window. Here again, the retention time
(4.04 minutes) of Hb Dhofar was in the D window, thus not providing conclusive
evidence for its differentiation.
CZE scan of the patient indicated that Hb Dhofar peak migrated in

zone 5 (Hb S zone), thus other possibilities (migration peak assigned in zone 6
487
for Hb D-Los Angeles, Hb G-Philadelphia, Hb Lepore, etc) were ruled out.
Hb Dhofar [
29 (GGC-GGT) Gly-Gly
58 (CCT-CGT) ProArg
] exists
predominantly in the Sultanate of Oman and with a thalassemic phenotype.
References
1.
2.
3.
4.
5.
6.
Daar S, Gravell D, Hussein HM, Pathare AV, Wali Y, Krishnamoorthy R.

Haematological and clinical features of -thalassemia associated with Hb
Dhofar. Eur J Haematol 2008; 80: 67-70.
Williamson D, Brown KP, Langdown JV, Baglin TP. Haemoglobin Dhofar is
+
linked to the codon 29 C-T(IVSI nt-3) splice mutation which causes beta
thalassemia. Br J Haematol 1995; 90: 229-31.
Marengo-Rowe AJ, Lorkin PA, Gallo E, Lehmann H. Haemoglobin Dhofar- a
new variant from Southern Arabia. Biochim Biophys Acta 1968; 168: 58-63.
Haemoglobin Dhofar-58 (ProArg) heterozygote. In: Variant Haemoglobins:
A Guide to Identification. Bain BJ, Wild BJ, Stephens AD, Phelan L. pp 188.
Wiley-Blackwell, United Kingdom, 2010.
Tony S, Daar S, Zachariah N, Wali Y. Prepubertal Hypertransfusion in
Thalassemia Intermedia: Sustained Positive Effects on Growth, Splenic
Function and Endocrine Parameters. Oman Med J 2012; 27(6). Available
from http://wwwomjournal.org/fultext_pdf.aspx?DetailsID=321&type=fultext
Qari MH, Wali Y, Albagshi MH, Aishahrani M, Alzahrani A, Alhijji IA, Almomen
A, Aljefri A, Al-Saeed HH, Abdullah S, Al-Rustamani A, Mahour K, Mousa SA.
Regional consensus opinion for the management of beta thalassemia major
in the Arab Gulf Area. Orphanet J Rare Diseases 2013; 8: 143. Available from
http://www.ojrd.com/content/8/1/143
488

Diagnostic HDiagnostic Hemoglobinopathies Secon Edition, Emoglobinopathies Secon Edition, August 21, 2015

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Diagnostic HDiagnostic Hemoglobinopathies Secon Edition, Emoglobinopathies Secon Edition, August 21, 2015

Uploaded by

Copyright:

Available Formats

Diagnostic Hemoglobinopathies

Laboratory Methods and Case Studies

Zia Uddin, PhD

Rita Ellerbrook, PhD

Piero C. Giordano, PhD

Kimberly R. Russell, MT (ASCP), MBA

Contributors and Reviewers

Rosline Hassan, PhD

John Petersen, PhD

Luisella Saba, PhD

Zia Uddin, PhD

Zia Uddin, PhD

Zia Uddin, PhD

Zia Uddin, PhD

Zia Uddin, PhD

Thomas E. Burgess, PhD

Red Blood Cell Morphology

Diagnostic Laboratory Methods

Hereditary Persistence of Fetal Hemoglobin

3.3.6 Hemoglobin F Quantification

Flow Cytometry Measurements of Cellular Fetal

Solid Phase Electrophoretic Separation

Rita Ellerbrook, PhD, and Zia Uddin, PhD

Capillary Zone Electrophoresis

David Hockings, PhD

High Performance Liquid Chromatography

Zia Uddin, PhD

Globin Chain Analysis

Solid Phase Electrophoretic Separation

Zia Uddin, PhD

Reverse Phase High Performance Liquid

Zia Uddin, PhD, and Rita Ellerbrook, PhD

4.3 Globin Chain Gene Mutations: DNA Studies 149

Electrospray Ionization-Mass Spectrometry

PCR and Sanger Sequencing

Elaine Lyon, PhD

Alpha and Beta Thalassemia

Neonatal Screening for

Zia Uddin, PhD

Capillary Zone Electrophoresis

Genetic Counseling & Screening

Prenatal Diagnosis of Beta-Thalassemia

Zia Uddin, PhD

Factors Affecting the Accuracy of Hb A1c Assay

Hemoglobin S with hereditary persistence

Hemoglobin G-Philadelphia trait

Hemoglobin S-G Philadelphia

Hemoglobin G-Coushatta trait

Hemoglobin C with hereditary persistence

Hemoglobin S-C disease

Hemoglobin D-Los Angeles (D-Punjab) trait

Hemoglobin S-D disease

Hemoglobin E and Associated Disorders

Case # 14 a Hemoglobin E trait

Case # 14 b Hemoglobin E homozygous

Hemoglobin S-Korle Bu (G-Accra)

Hemoglobin O-Arab trait

Hemoglobin S-+- thalassemia

Hemoglobin C-o thalassemia

Hemoglobin Hasharon trait

Hemoglobin Zurich trait

Hemoglobin Lepore trait

Hemoglobin J-Oxford trait