Vox Sanguinis (2015) 109, 173180

2015 International Society of Blood Transfusion

ORIGINAL PAPER DOI: 10.1111/vox.12261

A simple genotyping procedure without DNA extraction to

identify rare blood donors
M. Silvy,1,2 J.-C. Br
es,1 A. Grimaldi,2 C. Movia,2 V. Muriel,2 F. Roubinet,1 J. Chiaroni1,2 & P. Bailly1,2
1
Etablissement Francais du Sang, Blood Cell Grand Sud, Montpellier-Marseille, France
2
UMR 7268 ADES  Aix-Marseille Universite -EFS-CNRS, Marseille, France

Background Transfusion-induced alloimmunization has severe clinical conse-

Received: 28 October 2014,
revised 14 January 2015,
accepted 19 January 2015,
published online 8 April 2015
quences including haemolytic transfusion reactions, impaired transfused RBCs
longevity and greater difficulty in finding compatible blood. Molecular analysis
of genomic DNA now permits prediction of blood group phenotypes based on
identification of single nucleotide polymorphisms. Implementation of molecular
technologies in donor centres would be helpful in finding RBC units for special
patient populations, but DNA extraction remains an obstacle to donor genotyping.
Materials and Methods We propose a simple method compatible with high
throughput that allows blood group genotyping using a multiplex commercial kit
without the need for DNA extraction. The principle relies on pre-PCR treatment
of whole blood using heating/cooling procedure in association with a recombi-
nant hotstart polymerase.
Results In a prospective analysis, we yielded 5628 alleles identification and des-
ignated 63 donors with rare blood, that is either negative for a high-frequency
antigen or with a rare combination of common antigens.
Conclusion The procedure was optimized for simplicity of use in genotyping
platform and would allow not only to supply antigen-matched products to recipi-
ents but also to find rare phenotypes. This methodology could also be useful for
establishing a donor repository for human platelet antigens (HPA)-matched plate-
lets since the same issues are involved for patients with neonatal alloimmune
thrombocytopenia or post-transfusion purpura.
Key words: blood group genotyping, multiplex PCR on whole blood, rare blood
donors.

methaemoglobinaemia and severe hyperbilirubinaemia of

Introduction
Proper uses of red blood cell (RBC) transfusions include
treatment of symptomatic anaemia, prophylaxis in life-
threatening anaemia, and restoration of oxygen-carrying
capacity in case of haemorrhage. RBCs are also indicated
for exchange transfusion in sickle-cell disease (SCD),
severe parasitic infection (malaria, babesiosis), severe
Correspondence: Pascal Bailly, EFS Alpes-Mediterranee, Laboratoire
 Aix-Marseille Universite-
dHematologie Moleculaire, UMR 7268 ADES
EFS-CNRS, 207 Boulevard Sainte Marguerite, 13009 Marseille, France
E-mail: pascal.bailly@efs.sante.fr
This study was supported by Etablissement Francais du Sang, Paris,
France (all co-authors).
newborn. One of the main side-effects of transfusion is al-
loimmunization which was shown to be more frequent in
patients with SCD or warm autoimmune haemolytic anae-
mia (AIHA) [1, 2]. Alloimmunization has severe clinical
consequences including triggering of haemolytic transfu-
sion reactions, stimulation of autoantibody production,
impairment of transfused RBCs longevity and greater
difficulty in finding compatible blood [3]. Moreover, the
presence of antibodies is a limitation to haemagglutina-
tion which is the traditional method for testing donor and
patient blood group antigens and irregular antibodies. To
overcome this problem, genotyping tests have been devel-
oped [4, 5] and their benefit in patients care has been
reported in a number of clinical setting [69].

173
174 M. Silvy et al.
Implementation of molecular technologies in donor
centres would be helpful for finding RBC units for special
patient populations, such as SCD patients with newborn or
patients with multiple alloantibodies. It would also enable
identification of rare donors [4, 10]. A few high-through-
put studies have already been conducted [1012], but the
challenge for the next decade will be to provide blood
banks with a platform allowing beginning-to-end process-
ing, that is from blood sample to results. One of the main
obstacles to donor genotyping is DNA extraction. Indeed,
no secured system allows DNA extraction from thousands
of blood samples per day and multiple sample manipula-
tion increases the risk of cross-contamination. Many
studies have focused on developing simplex PCR amplifi-
cations from whole blood without DNA isolation.
However, PCR inhibitors present in blood samples can lead
to possible false-negative results or reduce sensitivity.
Among the most potent PCR inhibitors reported are hae-
moglobin/heme and IgG fraction [13, 14]. Two main lines
of investigation have been successfully followed, that is
pre-PCR handling of samples (freezing/heating of samples,
microwave treatment) and PCR adjustment [1522].
This study was undertaken to find a simple blood
group genotyping procedure compatible with high-
throughput processing using a multiplex commercial kit
without DNA extraction. The procedure was validated
through identification of 5628 alleles encoding blood
cells antigens.
Materials and methods
Blood samples and DNA template
After informed consent, blood samples were collected from
random donors in EDTA tubes. Samples were collected in
South-East of France and were mainly from Caucasian
descent. Samples stored at 4C were vigorously shaken
before use. For control purposes, genomic DNA was
extracted from 200 ll whole-blood samples using the QIA-
amp Blood DNA Mini kit (Qiagen, Courtaboeuf, France).
Pre-PCR handling procedures
Two pre-PCR handling procedures were used in prelimin-
ary testing. The first procedure consisted of microwave
irradiation of blood samples [17]. Blood samples stored at
-80C were rapidly thawed to 37C before processing. An
aliquot of blood (10 ll) was diluted 10-fold in distilled
water. The mixture was incubated at room temperature for
2 min and then placed in a heat-resistant jar containing
500 ml water and boiled in a microwave oven (700 W) for
7 min. After centrifugation at 13 500g. for 30 s, 10 ll of
the clear supernatant was used directly for PCR.
The second procedure consisted of preheating/cooling
blood in PCR mix [15]. A blood aliquot was introduced
directly in the PCR, and amplification was performed as
on purified DNA with modification of the initial denatur-
ation step. Instead of heating the samples for 5 min at
94C, whole-blood PCRs were heated for 3 min at 94C
and then cooled for 3 min at 55C, and this two-step
incubation was repeated 3 times. Taq DNA polymerase
was added after the three heating/cooling cycles. Since
the addition of Taq DNA polymerase after cycling is
poorly compatible with automation, we also tried addition
before performing the cycles.
Simplex PCR amplification of blood group gene
The two procedures were tested as initial steps for simplex
PCR amplification of two genes encoding blood groups;
that is, the exon 10 of RHD and the exon 6 of SLC14A1
(JK) genes. PCR amplification of exon 10 of the RHD gene
has been published elsewhere [23]. PCR amplification of
exon 6 of the SLC14A1 gene was performed in the pres-
ence of 80 ng/ll BSA, 0
05 unit of Taq DNA polymerase
(Invitrogen, Cergy Pontoise, France) and 200 nM of each
primer (For: 50 -acagagcccatggagctcc-30 and Rev: 50 -tgtaa
gctggtcaattacgct-30 ). PCR conditions were 94C for 5 min
followed by 35 cycles of 94C for 30 s, 58C for 30 s and
72C for 1 min. PCR products were separated on a 2%
(w/v) agarose gel stained with Sight DNA Stain (Eurome-
dex, Souffelweyersheim, France).
Blood group genotyping
Genotyping was performed by the multiplex reverse
sequence-specific oligonucleotide (PCR-RSSO) method
using the LIFECODES RBC kit (Genprobe, Nijlen Belgium)
according to the manufacturers instructions. This multi-
plex assay based on Luminex Multi-Analyte Profiling
technology (xMAP) was designed for the detection of nine
blood group systems through investigation of the follow-
ing alleles: RHCE*C/*c/*E/*e, KEL*01/*02/*03/*04/*06/*07/
*21, JK*01/*02/*02N.06, FY*01/*02/*02M.01/*02N.01,
GYPA*01/*02, GYPB*03/*04/*03N.01/*03N.03 and
DO*01/*02. The test detects also seven additional poly-
morphisms in RHCE gene associated with variant alleles
as well as platelet antigen-1 (HPA-1). Briefly, asymmetric
PCR amplification with biotinylated primers was per-
formed on a Biometra thermocycler. PCR amplicons were
hybridized to complementary allele-specific oligonucleo-
tide probes immobilized on fluorescent-coded microsphere
beads. At the same time, the biotinylated PCR products
were labelled with phycoerythrin-conjugated streptavidin,
and acquisition was performed immediately on a Luminex
100. Automatic genotype determination and data analysis
2015 International Society of Blood Transfusion
Vox Sanguinis (2015) 109, 173180

were performed using the GENPROBE RBC MATCHIT!-RBC soft-
ware (Genprobe) according to the manufacturers instruc-
tions. Genotyping was performed on either whole blood (3
or 5 ll) or control DNA samples (100 ng).
Results
Efficacy of simplex PCR amplification after
pretreatment
To assess the suitability of the two procedures for PCR
amplification, we performed simplex PCR amplification of
exon 10 of the RHD and exon 6 of the SLC14A1 genes
(Fig. 1). Amplification failed in all samples following
microwave irradiation. After pretreatment using the heat-
ing/cooling cycle procedure on 3 ll of whole blood,
amplification of RHD showed a weak specific band at
382 bp in 2 out of three samples investigated. Amplifica-
tion of SLC14A1 was more efficient with a fairly specific
band at 267 bp. No obvious effect on PCR efficiencies
was observed by increasing blood volume to 5 ll.
Feasibility of blood group genotyping without
DNA extraction
Based on the results, blood group genotyping was per-
formed on 3 ll of whole blood using the LIFECODES RBC
kit (Genprobe) after the heating/cooling cycle procedure.
We also tested the procedure modified by adding Taq DNA
polymerase, that is either Taq polymerases (Genprobe) or
KAPA2G Robust DNA hotstart polymerase (Clinisciences,
Nanterre, France) prior to cycling. Four samples with various
genotypes were investigated, and results were compared
with those obtained on control DNA samples (Table 1;
the results of sample #4 are available in supplementary data
Pre-PCR handling Microwave irradiation
Blood dilution 1/10 1/50
Volume 10 l 10 l
Sample 1 2 3 1 2 3 Ctrl
JK
RHD
Fig. 1 Simplex PCR amplication following the 2 procedures. The blood of three RhD-positive donors (numbered 13) was tested for PCR amplication
of SLC14A1 exon 6 and RHD exon 10. PCR products were separated on a 2% (w/v) agarose gel. DNA sample was used as positive control for amplica-
tion (Ctrl).
2015 International Society of Blood Transfusion
Vox Sanguinis (2015) 109, 173180
Blood group genotyping without DNA extraction 175
Figs S1 and S2). When Taq DNA polymerase was added
after heating/cooling cycles, raw MFI (mean fluorescence
intensity) ranged from 13% to 257% of the control values
with a correct allele assignment for all SNPs. When Genp-
robes Taq DNA polymerase was added prior to cycling,
MFI was from 3% to 114% of control values and no
adjusted data were generated by the MATCHIT Software for 23
probes since raw data were below the threshold. Almost
40% (23/59) of the probes with MFI below 40% of control
were assigned as failed or undetermined by the MATCHIT
Software. Thus, to minimize failure rate, we search for
conditions allowing MFI higher than 50% of control.
When KAPA2G Robust DNA hotstart polymerase was
added before the heating/cooling cycles, MFI was from
26% to 340% of control values and allele assignment was
successful. Higher MFI (59402%) was obtained by
increasing the volume of blood used to 5 ll. To clarify
the importance of the three heating/cooling cycles, we
tested the efficiency of amplification using KAPA2G
Robust DNA hotstart polymerase on blood samples (3 ll
or 5 ll) omitting these cycles. Raw MFI was ranging from
20% to 269% and from 43% to 422% using 3 and 5 ll of
blood, respectively (Table 1, Fig. S2).
Since no adjusted data were generated by the MATCHIT
Software for some probes with MFI below 40% of control,
we choose the condition giving MFI higher than 50% of
control for all SNP in all samples, that is addition of
KAPA2G Robust DNA hotstart polymerase before the
heating/cooling cycles using 5 ll of blood for pursuing
the study.
Intra- and interassay variations
Blood and DNA samples (n = 2) were analysed five times in
a single experiment and in four independent experiments
Heating/cooling cycles
none none
3 l 5 l
Size
1 2 3 1 2 3 Ctrl (bp)
267
382
176 M. Silvy et al.

Table 1 Mean uorescence intensity (MFI) range for genotyping on blood in the different conditions tested (expressed in % of control DNA)

Geneprobes Taq KAPA2G Robust DNA hotstart

Heating/cooling cycles + + + + - -
Taq added After Before Before Before
Sample Blood volume (ll) 3 3 3 5 3 5

#1 MFI (% range) 95257 23114 26339 88402 86269 80422

Probes with MFI <50% (n) 0 16 2 0 0 0
#2 MFI (% range) 54191 892 38340 86231 42179 68288
Probes with MFI <50% (n) 0 17 1 0 1 0
#3 MFI (% range) 13171 389 36186 59211 20164 45206
Probes with MFI <50% (n) 14 17 9 0 10 3
#4 MFI (% range) 1461 751 53207 73270 36107 43189
Probes with MFI <50% (n) 10 23 0 0 5 1

Values in bold are the min and max in each condition.

for intra- and interassay variations, respectively. Coeffi-
cients of variation (CV) were calculated both on raw MFI
and adjusted values. In intra-assays, CVMFI were ranging
from 27% to 31% on blood and from 42% to 59% on DNA.
Adjusted values showed a CV of 11% on both blood
and DNA. As expected, interassay experiments presented
a higher CV both on MFI (50% and 34% on blood and
DNA, respectively) and adjusted values (25% and 17%,
respectively).
Prospective genotyping of blood samples without
DNA extraction
Prospective blood group genotyping without DNA extrac-
tion was performed on 5 ll of whole blood with the addi-
tion of KAPA2G Robust DNA hotstart polymerase prior to
heating/cooling cycling. A total of 209 donor blood sam-
ples were investigated for 28 alleles, and genotype results
were compared to those obtained on control DNA samples
(Table 2). Eight samples (3
8%) failed when genotyping
was performed on blood, while analysis gave correct
results on DNA. Detection of KEL*03/*04 genotype failed
in one DNA sample (low MFI). Among the 5628 alleles
investigated, we considered as non-reliable typing with a
probe value <0
05 to the cut-off. Thus, 45 samples (27
blood samples and 18 DNA) were tested twice and in all
cases the second investigation showed both a high MFI
and a correct deviation between probe values and cut-off.
It should be noted that among the samples tested twice,
first determinations were incorrect for (i) MNS*03/*04 in
five blood samples, (ii) KEL*03/*04 in two blood samples
and (iii) RHCE*C/*c in one blood and one DNA sample.
It is noteworthy that some samples present a rare pre-
dicted phenotype due to homozygosity for a rare allele or
rare antigen combination (Table 3). The genotyping kit
used in our study also investigates eight SNPs of RHCE
gene associated with a variety of alleles mainly found in
people from African ancestry. Two samples were found
heterozygous for 733C>G transversion, two for the
122A>G transition and one for the 1025C>T transition.
Altogether, no discordance was noted between blood
and DNA samples as long as deviation between probe
value and cut-off is 0
05.
Discussion
In transfusion medicine, SNP typing is a powerful adjunct
to serologic testing and is the best method for typing
transfused patients [79]. Recent study has shown that
molecular matching of patients and RBC units reduced al-
loimmunization risks [6]. Currently, commercial kits for
blood group genotyping are available, but DNA extrac-
tion hampers automation of PCR for large-scale applica-
tions and was shown to represent a significant proportion
of genotyping cost [10]. To circumvent the DNA purifica-
tion step for genotyping of donors, we attempted to per-
form blood group genotyping directly on blood samples.
Since genotyping was to be performed using a commer-
cial multiplex kit, PCR conditions could not be changed
and our only option was to focus on sample pretreatment
that had to be compatible with automation, that is a sim-
ple workflow with limited treatment steps.
The two pre-PCR sample handling procedures described
here were microwave irradiation of blood [17] and heat-
ing/cooling cycling [15]. Since first validation by simplex
PCR amplification of exon 10 of the RHD and exon 6 of
the SLC14A1 genes failed after microwave irradiation,
that option was dropped and investigation was carried
on with heating/cooling cycling. Amplification with
successful blood group genotyping was achieved on 3 ll

2015 International Society of Blood Transfusion

Vox Sanguinis (2015) 109, 173180
 Blood group genotyping without DNA extraction 177

Table 2 Occurrence of blood group alleles in prospective study

Occurrence (%) [24]

Blood group system Allele/polymorphism Associated phenotype Number of positive Caucasian Black

MNS GYPA*01 M+ 157 78 74

GYPA*02 N+ 152 72 75
GYPB*03 S+ 106 55 31
GYPB*04 s+ 173 89 93
GYPB*03N.01 Uvar(NY) 0 nd <1
GYPB*03N.03 Uvar(P2) 0 nd <1
RH (RHCE) RHCE*C C+ 142 29 22
RHCE*c c+ 161 80 98
RHCE*E E+ 48 29 22
RHCE*e e+ 197 98 98
Kell (KEL) KEL*01 K+ 19 9 2
KEL*02 k+ 201 99
8 100
KEL*03 Kpa+ 5 2 <0
01
KEL*04 Kpb+ 201 100 100
KEL*21 Kpc+ 0 <0
01 <0
01
KEL*06 Jsa+ 0 <0
01 20
KEL*07 Jsb+ 201 100 99
Duffy (FY) FY*01 Fya+ 132 66 10
FY*02 Fyb+ 170 83 23
FY*02M.01 Fyb weak 11 <2 nd
FY*02N.01 Fyb null 3 nd 80
Kidd (JK) JK*01 Jka+ 157 77 92
JK*02 Jkb+ 143 74 49
JK*02N.06 Jk null 0 nda nd
Dombrock (DO) DO*01 Doa+ 139 67 55
DO*02 Dob+ 171 82 89
HPA-1 HPA*1a HPA-1a+ 197 97
9 99
0
HPA*1b HPA-1b+ 55 28
8 18
8
Polymorphisms in RHCE 122G na 2 na na
733G na 2 na na
1025T na 1 na na

nd, not determined; na, not applicable.

a
This allele was found exclusively in Finns [25].

whole-blood samples when Taq DNA polymerase was
added after the heating/cooling cycles. However, to make
automation easier, we tried performing amplification
after the addition of two different DNA polymerases
prior to cycling. Using Genprobes Taq DNA polymerase,
genotyping succeeded on DNA but failed on blood
suggesting that the enzyme was inhibited by blood
compounds. Using KAPA2G Robust DNA hotstart poly-
merase, a recombinant Taq DNA polymerase selected for
its improved tolerance to common PCR inhibitors,
genotyping was successful on 5 ll of blood samples.
Intra- and interassay tests showed that the procedure was
as robust on blood samples as on DNA.
The accuracy of genotyping after pre-PCR treatment
using recombinant polymerase added prior to cycling
was evaluated in a cohort of 209 blood samples. No
result was obtained in eight samples probably because
of the presence of a blood compound, eliminated during
DNA purification, which might hamper genotyping on
blood. Results showed 100% concordance for the 5628
alleles investigated. However, 30/5628 and 21/5628
alleles for blood samples and DNAs respectively had to
be tested twice to make sure of genotyping results.
These weak failure rates are likely accountable for man-
ual pipetting and might be lowered by automation. Con-
sistent with the low frequency of GYPB*03N.01,
GYPB*03N.03, KEL*06, KEL*21 and JK*02N.06 alleles,
none of these alleles were encountered in our cohort.
Previous data show that GYPB*03N.01 and
GYPB*03N.03 alleles (encoding Uvar(NY) and Uvar(P2)

2015 International Society of Blood Transfusion

Vox Sanguinis (2015) 109, 173180
178 M. Silvy et al.
Table 3 Rare alleles and low-frequency antigens combination found [10,
27]
Negative for high-frequency antigens
e-
Fy(a-,b-)
Dob-
HPA-1a-
Rare for combination of common antigens
C- E- Fya- Jkb-
C- E- Fyb- Jka-
C- E- Fyb- Jkb-
E- c- Fya- Jka-
E- c- Fya- Jkb-
E- c- Fyb- Jka-
E- c- Fyb- Jkb-
E- c- Fya- s-
E- c- Fyb- s-
E- c- Jka- s-
E- c- Jkb- s-
Total
phenotypes, respectively) are absent in Caucasian as
opposed to frequency of 0
1% and 0
7% in African
Americans, respectively [4]. The frequency of KEL*06
(encoding Jsb-positive phenotype) is <0
01% in Cauca-
sians and 20% and Blacks. KEL*21 (encoding Kpc+ phe-
notype) is rare in all populations (<0
01%) [24], and
JK*02N.06 has been found exclusively in Finns (0
03%)
[25]. In all, 23 out of 28 alleles and 3 out of 8 probes
in RHCE were validated in our donors.
Our very simple and accurate procedure could be
easily used in an automated high-throughput genotyp-
ing platform for donor typing. Indeed, high-throughput
genotyping could be useful in a number of transfu-
sional settings. The most obvious situation would
involve finding blood for patients at high risk of allo-
antibody productions due to frequent chronic blood
transfusion for medical conditions such as sickle-cell
disease, AIHA and aplastic anaemia. Other applications
for red cell genotyping would include identifying red
cells lacking blood group antigens reactive with any
pre-existing clinically significant antibodies in patients
with complex serological problems [26], increasing the
inventory of antigen-negative blood, matching RBC
components to the recipient blood type and finding of
rare donor products [10, 27].
This study has especially interesting implications with
regard to finding rare donors. Typically, there are two
types of rare donors. The first type is a person whose
RBCs lack certain high-prevalence antigens such as
k-negative phenotype that occurs in 2/1000 [24, 28] of
Caucasians or Jsb-negative phenotype that is almost
absent in Caucasians but occurs in 1/149 of Afro-Carib-
bean [29]. In 2007, only 8 donors lacking Jsb-negative
N=
phenotype were indexed in the French National Refer-
ence Laboratory for Blood Groups (CNRGSParis) [28].
4 The second type of rare donor has RBCs displaying
1 several common antigens forming a unique combination
30 such as E-, c-, Fya- and Jkb-negative combination
4 [10, 27, 28]. This prospective analysis of blood donor
allowed identification of 62 samples with a rare
1
phenotype. We also identified two samples with the
1
122A>G transition in RHCE which encode the RH8
1
6
(Cw) antigen. This antigen expressed in 2% Caucasians
5 is implicated in mild-to-severe haemolytic transfusion
1 reaction [24].
1 Another avenue for further assessment of the geno-
3 typing procedure described herein is the identification
1 of compatible human platelet antigens (HPA) that
1 involves issues similar to those related to RBC identifi-
2 cation. Establishing a blood donor repository for HPA
62 matched platelets (PLTs) would allow adequate PLT
support for patients with neonatal alloimmune thrombo-
cytopenia or post-transfusion purpura requiring anti-
gen-matched PLT transfusion [30]. Development of
genotyping platforms would provide a tool to increase
the antigen-negative inventory at a fraction of the cost
associated with serology [31, 32]. Based on the proof-
of-principle reported here, our procedure optimized for
simplicity and ease could be used in an automated
high-throughput genotyping platform for donor typing
to supply antigen-matched products and find rare phe-
notypes. Extension of the CE marking to blood group
SNP detection without need for DNA extraction can be
considered by manufacturers.
Acknowledgements
We wish to thank the staff of haematology laboratories of
EFS Alpes-Mediterranee and EFS Pyrenees-Mediterranee
for providing the blood donor samples used in this study.
We thank Andrew Corsini for proofreading the manuscript.
Authorship contributions
MS, JCB and PB designed the research, FR and JC pro-
vided donors samples, AG, CM and VM performed the
analysis, and MS and PB analysed the data and wrote the
manuscript.
Disclosure and conict of interest
All authors have read the journals policy on disclosure
of potential conflicts of interest and declare that they
2015 International Society of Blood Transfusion
Vox Sanguinis (2015) 109, 173180

have no conflict of interests in the subject matter of their
article. All authors declare that they have no financial or
References
1 Flickinger C: In search of red blood
cells for alloimmunized patients with
sickle cell disease. Immunohematology
2006; 22:136142
2 Branch DR, Petz LD: Detecting alloan-
tibodies in patients with autoantibod-
ies. Transfusion 1999; 39:610
3 Aygun B, Padmanabhan S, Paley C,
et al.: Clinical significance of RBC al-
loantibodies and autoantibodies in
sickle cell patients who received trans-
fusions. Transfusion 2002; 42:3743
4 Hashmi G, Shariff T, Zhang Y, et al.:
Determination of 24 minor red blood
cell antigens for more than 2000 blood
donors by high-throughput DNA analy-
sis. Transfusion 2007; 47:736747
5 Di Cristofaro J, Silvy M, Chiaroni J,
et al.: Single PCR multiplex SNaPshot
reaction for detection of eleven blood
group nucleotide polymorphisms: opti-
mization, validation, and one year of
routine clinical use. J Mol Diagn 2010;
12:453460
6 Ribeiro KR, Guarnieri MH, da Costa DC,
et al.: DNA array analysis for red blood
cell antigens facilitates the transfusion
support with antigen-matched blood in
patients with sickle cell disease. Vox
Sang 2009; 97:147152
7 El Kenz H, Efira A, Le PQ, et al.:
Transfusion support of autoimmune
hemolytic anemia: how could the
blood group genotyping help? Transl
Res 2014; 163:3642
8 Guelsin GA, Sell AM, Castilho L,
et al.: Benefits of blood group geno-
typing in multi-transfused patients
from the south of Brazil. J Clin Lab
Anal 2010; 24:311316
9 Bakanay SM, Ozturk A, Ileri T, et al.:
Blood group genotyping in multi-
transfused patients. Transfus Apher
Sci 2013; 48:257261
10 Perreault J, Lavoie J, Painchaud P,
et al.: Set-up and routine use of a
database of 10,555 genotyped blood
donors to facilitate the screening of
compatible blood components for allo-
immunized patients. Vox Sang 2009;
97:6168
2015 International Society of Blood Transfusion
Vox Sanguinis (2015) 109, 173180
Blood group genotyping without DNA extraction 179
personal relationship with organizations that could poten-
tially be perceived as influencing the described research.
11 Le Goff GC, Br
es JC, Rigal D, et al.:
Robust, high-throughput solution for
blood group genotyping. Anal Chem
2010; 82:61856192
12 Hopp K, Weber K, Bellissimo D, et al.:
High-throughput red blood cell anti-
gen genotyping using a nanofluidic
real-time polymerase chain reaction
platform. Transfusion 2010; 50:4046
13 Al-Soud WA, J onsson LJ, R^adstr
om P:
Identification and characterization of
immunoglobulin G in blood as a major
inhibitor of diagnostic PCR. J Clin
Microbiol 2000; 38:345350
14 Al-Soud WA, R adstr
om P: Purification
and characterization of PCR-inhibitory
components in blood cells. J Clin
Microbiol 2001; 39:485493
15 Mercier B, Gaucher C, Feugeas O,
et al.: Direct PCR from whole blood,
without DNA extraction. Nucleic Acids
Res 1990; 18:5908
16 Ohhara M, Kurosu Y, Esumi M: Direct
PCR of whole blood and hair shafts by
microwave treatment. Biotechniques
1994; 17:726728
17 Jadaon MM, Dashti AA, Lewis HL,
et al.: Whole-blood polymerase chain
reaction and restriction fragment
length polymorphism: a simplified
method by microwave irradiation. Med
Princ Pract 2009; 18:280283
18 Li H, Xu H, Zhao C, et al.: A PCR
amplification method without DNA
extraction. Electrophoresis 2011;
32:394397
19 Nishimura N, Nakayama T, Tonoike H,
et al.: Direct polymerase chain reac-
tion from whole blood without DNA
isolation. Ann Clin Biochem 2000;
37:674680
20 Nishimura N, Nakayama T, Tonoike H,
et al.: Various applications of direct
PCR using blood samples. Clin Lab
2002; 48:377384
21 Yang YG, Kim JY, Song YH, et al.: A
novel buffer system, AnyDirect, can
improve polymerase chain reaction
from whole blood without DNA isola-
tion. Clin Chim Acta 2007; 380:112
117
22 Klintschar M, Neuhuber F: Evaluation
of an alkaline lysis method for the
extraction of DNA from whole blood
and forensic stains for STR analysis. J
Forensic Sci 2000; 45:669673
23 Silvy M, Chapel-Fernandes S, Beley S,
et al.: Molecular characterization of a
new D- - haplotype in a Comorian
man. Vox Sang 2012; 103:352355
24 Reid ME, Lomas-Francis C, Olsson ML.
The Blood Group Antigens. Fact Book.
3rd ed. London: Academic Press. 2012.
25 Sidoux-Walter F, Lucien N, Nissinen R,
et al.: Molecular heterogeneity of the
Jk(null) phenotype: expression analysis
of the Jk(S291P) mutation found in
Finns. Blood 2000; 96:15661573
26 Anstee DJ: Red cell genotyping and
the future of pretransfusion testing.
Blood 2009; 114:248256
27 Revelli N, Villa MA, Paccapelo C,
et al.: The lombardy rare donor pro-
gramme. Blood Transfus 2013; 12:
S249S255
28 Peyrard T, Pham BN, Le Pennec PY,
et al.: The rare blood groups: a public
health challenge. Transfus Clin Biol
2008; 15:109119
29 Silvy M, Di Cristofaro J, Beley S,
et al.: Identification of RHCE and KEL
alleles in large cohorts of Afro-Carib-
bean and Comorian donors by multi-
plex SNaPshot and fragment assays: a
transfusion support for sickle cell dis-
ease patients. Br J Haematol 2011;
154:260270
30 Shehata N, Denomme GA, Hannach B,
et al.: Mass-scale high-throughput
multiplex polymerase chain reaction
for human platelet antigen single-
nucleotide polymorphisms screening of
apheresis platelet donors. Transfusion
2011; 51:20282033
31 Denomme GA, Johnson ST, Pietz BC:
Mass-scale red cell genotyping of
blood donors. Transfus Apher Sci
2011; 44:9399
32 Veldhuisen B, van der Schoot CE, de
Haas M: Blood group genotyping:
from patient to high-throughput donor
screening. Vox Sang 2009; 97:198206
180 M. Silvy et al.

Supporting Information
Additional Supporting Information may be found in the online version of this article:
Fig. S1 Evaluation of Genprobes Taq DNA polymerase for multiplex genotyping with pre-PCR treatment using the heat-
ing/cooling procedure.
Fig. S2 Evaluation of KAPA2G Robust DNA hotstart polymerase for multiplex genotyping with pre-PCR treatment
using the heating/cooling procedure.

2015 International Society of Blood Transfusion

Vox Sanguinis (2015) 109, 173180