Springer 2005

Biogerontology (2005) 6: 227232

DOI 10.1007/s10522-005-2617-0

Review article

Bioinformatics and proteomics approaches for aging research

Chaerkady Raghothama1,2, H. C. Harsha1, C. K. Prasad1 & Akhilesh Pandey3,*
1
Institute of Bioinformatics, International Tech Park Ltd., 560066, Bangalore, India; 2Department of Biotechnology, Kasturba Medical College, Manipal, Karnataka, India; 3McKusick-Nathans Institute of Genetic
Medicine and the Department of Biological Chemistry, Johns Hopkins University, Baltimore, MD 21205,
USA; *Author for correspondence (e-mail: pandey@jhmi.edu)
Received 15 February 2005; accepted in revised form 29 June 2005

Key words: bioinformatics, database, genomics, proteomics, systems biology

Abstract
Aging is a natural phenomenon that aects the entire physiology of an organism. Elucidating the molecular
mechanisms underlying this complex process remains a major challenge today. Humans make poor models
for research into aging because of their long life span. Thus, most of the current knowledge is through
studies conducted in lower organisms. Large dierences in life spans make it dicult to extrapolate the
results of experiments carried out in model organisms to humans. Recent advances in genomic and proteomic technologies now permit generation of data pertaining to aging on a large-scale. In addition, several
web-based community resources and databases are available that provide easy access to the available data.
Use of bioinformatics and systems biology type of approaches provide a framework to start dissecting this
complex biological phenomenon. Here, we discuss various genomic, transcriptomic and proteomic
approaches that have the potential to provide a comprehensive mechanistic insight into the aging process.

Introduction
Aging is perhaps one of the most complex biological phenomena. Age invariably affects the
functioning of all organs of an organism. Altered
functioning of even one organ can have a major
impact on the physiology of an entire organism.
One of the major goals of gerontology is to
understand the complex mechanisms involved in
aging at the molecular, cellular and organ levels
that would also facilitate understanding of agerelated diseases. Because of practical limitations
in studying the aging process in humans in vivo,
animal models are frequently used (Gershon and
Gershon 2000, 2002). However, the large dierence between the longevity of most experimental
animal models and humans makes extrapolation
of the results somewhat dicult. Werner syndrome is, to some extent, the closest available
model in humans (Goto 1997) although the

eects are due to a single gene mutation and

may not truly reect the multiple changes that
occur during the normal aging process. Despite
these limitations, research in this area has accelerated with the application of high-throughput
technologies like microarrays and mass spectrometry. Whole transcriptome changes that occur at
cellular and organ levels can be determined from
these studies and individual genes can be investigated further. Equally important are the studies
on reversal of some of the changes observed with
aging through interventions such as hormone
administration (Tollet-Egnell et al. 2001). It is
evident from the high-throughput studies that
several metabolic pathways are aected in aging
and the picture could be even more complex considering that most of these pathways are interconnected. Therefore, a global picture of the
networks in aging will require integration of the
information obtained from individual studies.

228
Bioinformatics deals with the application of computers to store, analyze and interpret complex
biological data. At this juncture, bioinformatics
has great potential, especially in organizing the
fragmented data on aging studies done on dierent animal models and deriving meaningful interpretations. In this article, we discuss the available
resources and bioinformatic approaches for
deciphering the complex networks in human
aging.

factors, mammalian homologs of DAF-16, should

identify additional genes that regulate aging. The
dierences in average life span and susceptibility
to certain diseases among the population have
spurred researchers to use single nucleotide polymorphisms (SNPs) in whole genome association
studies in aging and age-related disorders. This
approach could aid in narrowing down to specic
genes that are more important than others in
determining the life span (Bessenyei et al. 2004).

Genomic analysis pertaining to aging

Transcriptomic analysis pertaining to aging

The availability of genomic sequences of many

organisms has provided an excellent opportunity
for carrying out bioinformatics analyses. Despite
high sequence conservation and similar gene content, human, chimp and mouse show large differences in their longevity. Efforts are underway to
assess if this can be attributed to differences in
gene regulation. It has already been established
that signaling from DAF-2/insulin receptor to
DAF-16/FOXO regulates longevity, metabolism
and development in Caenorhabditis elegans and
Drosophila melanogaster (Kimura et al. 1997; Lin
et al. 1997; Ogg et al. 1997). This pathway is triggered by insulin and insulin-like growth factor 1
signaling in mammals. The forkhead type transcription factor, DAF-16, functions downstream
of insulin-like signals and binds to a consensus
DNA sequence, TTGTTTAC, which is located
upstream of the target genes (Furuyama et al.
2000). With the goal of identifying the potential
downstream targets of DAF-16, Lee et al. (2003a)
conducted a systematic comparative genomic
study of C. elegans and D. melanogaster genomes.
By searching for its consensus binding-site,
TTGTTTAC, upstream of all sets of orthologous
genes, they determined 17 orthologous gene pairs
and tested their involvement in aging. Extrapolation of such studies to higher mammals should
help unravel the conserved pathways that regulate
aging. To this end, a Comparative Regulatory
Genomics (CORG) database has been developed
that houses conserved non-coding blocks of DNA
sequences upstream of orthologous genes between
human and mouse (Dieterich et al. 2003). A
detailed bioinformatic study and experimental validation of the binding sites of FOXO transcription

DNA microarray studies are an excellent means

of obtaining information regarding transcriptional changes on a global scale. Such studies
have proven to be helpful in deciphering some aspects of aging in different organisms. Grouping
of genes based on their functional annotation has
been made easier by the use of standardized
terms as dened by the Gene Ontology Consortium (Ashburner et al. 2000). Comparison of
involvement of pathways has been simplied by
the development of MAPPFinder, which allows
integration of a freely available pathway visualization tool called GenMAPP with Gene Ontology
annotations (Doniger et al. 2003). DNA microarray studies that have been done on tissue samples from young, normal aged and prematurely
aged humans revealed that genes involved in
DNA and RNA metabolism, cell growth and signal transduction were altered in their expression
(Ly et al. 2000; Kyng et al. 2003). Another study
that focused on gene expression changes associated with age in two areas of the human
prefrontal cortex used a functional class scoring
approach in which every gene in a functional
class was considered irrespective of the magnitude
of change (Pavlidis et al. 2004). This approach
led to identication of genes involved in aging
that would otherwise have been missed. A dierent study identied shared patterns of regulation
across orthologous genes related to aging across
two highly diverged organisms, C. elegans and
D. melanogaster (McCarroll et al. 2004). Similar
comparative functional genomics studies in closely related mammals should take us closer to
understanding mechanisms underlying aging in
humans. Microarray data analysis is still an

229
evolving eld with a paucity of tools that help
integrate information obtained from DNA
microarrays with other functional data such as
proteinprotein interactions and post-translational modications. We anticipate that with
combined development of computational tools
and models, it will be possible to mine DNA
microarray data for inferring regulated networks
as opposed to merely obtaining a list of dierentially expressed genes.

Proteomic analysis pertaining to aging

Proteomics covers both qualitative and quantitative protein proling of a cell, tissue, organ or a
whole organism and the functional characterization of the protein component. Although highthroughput transcriptomic analyses such as DNA
microarrays and RNAi experiments (Lee et al.
2003b) have helped initiate a systematic study of
the aging process, only a limited number of studies to look at proteins in aging have been carried
out thus far (Chen et al. 2003; Cho et al. 2003;
Poon et al. 2004). With developments in mass
spectrometry
and
2D-gel
electrophoresis
(Deshpande et al. 2003), comparative proteomics
shows great promise in analyzing complex protein mixtures. Methods such as isotope-coded
anity tags (ICAT) (Gygi et al. 1999) and stable
isotope labeling by amino acids in cell culture
(SILAC) (Ong et al. 2002) have provided valuable means to study and quantitate global
expression proles in cells and tissues. Highthroughput proteomic strategies using liquid
chromatography tandem mass spectrometry
(LC-MS/MS) have been eciently used for identication of oxidatively induced protein carbonylation in aged mouse brains (Soreghan et al.
2003). Comparative proteomics has been used to
study the eect of aging on mitochondrial
proteins (Chang et al. 2003). Another comparative study using premature aging Hutchinson
Gilford progeria syndrome broblasts revealed
dierential protein expression and glycosylation
in membrane proteins (Robinson et al. 2003).
Proteomic studies on aged samples have also disclosed various non-enzymatic modications such
as glycation and nitration that progress with age
(Bank et al. 1998; Paik et al. 2001; Konova et al.

2004). These studies clearly indicate the value of

additional comprehensive proteomic analyses to
characterize the post-translational modications
and changes in protein abundance that correlated
with aging.

Database resources for aging studies

Databases are immensely valuable as they can
pool information from related experiments and
offer a common platform for in silico research. To
date, there are two major web resources that house
a wealth of information regarding the biology of
aging. SAGE KE (Science of Aging Knowledge
Environment: http://sageke.sciencemag.org) is an
online resource that provides a historical perspective and covers the recent developments in aging
research (Davenport et al. 2001). It also contains
a database, AGEID (Kaeberlein et al. 2002),
which provides information on gene/intervention
pertaining to age-associated disorders such as
Alzheimers and Parkinsons disease. Another useful resource that has recently been developed is
HAGR (Human Ageing Genomic Resources:
http://genomics.senescence.info), a collection of
online resources featuring two databases: GenAge
and AnAge (de Magalhaes et al. 2005).
GenAge (de Magalhaes and Toussaint 2004) is a
human aging database that catalogues over 200
human genes associated with human progeroid
syndromes and derived human homologues of
genes known to be involved in longevity of model
organisms while AnAge is an integrative database
of the aging phenotype in several species. Dierent
databases when integrated together using bioinformatic tools can yield novel information. For
example, a recent study (de Magalhaes and
Toussaint 2004) eectively utilized proteinprotein
interaction data from Human Protein Reference
Database (HPRD) (Peri et al. 2003), a database
that we have developed previously, with the genes
associated with aging from GenAge to create
gene networks inuencing aging. By grouping
genes that are known to aect aging in model
organisms into functional clusters, the authors
constructed gene networks inuencing aging by
focusing on specic pathways. As most of the
genes that have been studied in the context of
extension of life span in mice appear to aect the

230
GH/IGF-1 pathways, an interaction map was
constructed by obtaining the interactors of these
proteins from Human Protein Reference Database
(Peri et al. 2003) and literature searches. The same
has been done with the genes involved in DNA
repair and metabolism, as a number of genes
involved in accelerated aging phenotype in mammals appear to be related to DNA metabolism.
The interactors with highest connectivity in these
pathways have been hypothesized to play a role in
regulating aging. With improvements in the protein network analysis and availability of larger
datasets, it should be possible to probe the mechanisms involved in this complex machinery in greater detail. Such analyses would also be useful in
bringing out the new potential partners in the
aging process, which are otherwise overlooked in
the independent experiments. To identify agingrelated genes, another study (Ferrarini et al. 2004)
extrapolated the data from General Repository for
Interaction Datasets (GRID) (Breitkreutz et al.
2003) which contains interaction datasets of
S. cerevisiae, C. elegans and D. melanogaster along
with the genes listed in the SAGE KE resource
(Strauss and LaMarco 2002). A model was proposed which allows novel aging-related genes to be
identied based on their connectivity in a network.
Aging genes were observed to have the highest
connections in the network (hubs). With data on
proteinprotein interactions accumulating steadily, this model can be useful in identifying novel
candidate genes in the future.

Future directions: A systems biology approach

One of the promising approaches to decipher and
understand complex biological processes is a
systems biology approach (Hood 2003). This
focuses on studying the interrelationships of all
the elements that comprise a system rather than
studying each element individually. The growing
realization that eukaryotes have conserved pathways, which regulate life span, has prompted
researchers to search for genes that govern these
pathways. Many genes have also been studied in
the closest model organism, the laboratory mouse
that increases longevity. Development of various
statistical methods to correlate the data obtained
from one organism to another has allowed us to

extrapolate results obtained from such studies

(McCarroll et al. 2004). With the large amount
of data already available on aging from other
organisms, the time is ripe for an integrated approach by combining biochemical, physiological
and molecular studies that would provide fundamental insights into the aging process.

Conclusions
Progressing from the era of single gene knock
out studies, biology has now taken a big leap by
the development of high-throughput genomic
and proteomic techniques. These advancements
have led to generation of enormous amount of
data that needs to be analyzed systematically.
Experts from many different disciplines such as
mathematics and computational biology are
required to work together to develop approaches
and tools that will make it possible to analyze
and interpret the complex data. With global
approaches coupled to careful hypothesis-driven
approaches, we are getting closer to solving the
mysteries of aging.

Acknowledgement
Akhilesh Pandey is supported by Beckman Young
Investigator Award, Sidney Kimmel Scholar
Award and a Grant (U54 RR020839) from the
National Institutes of Health. He serves as Chief
Scientic Advisor to the Institute of Bioinformatics. The terms of this arrangement are being managed by the Johns Hopkins University in
accordance with its conict of interest policies.

References
Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry
JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris
MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese
JC, Richardson JE, Ringwald M, Rubin GM and Sherlock G
(2000) Gene ontology: tool for the unication of biology. The
Gene Ontology Consortium. Nat Genet 25: 2529
Bank RA, Bayliss MT, Lafeber FP, Maroudas A and
Tekoppele JM (1998) Ageing and zonal variation in posttranslational modication of collagen in normal human
articular cartilage. The age-related increase in non-enzymatic

231
glycation aects biomechanical properties of cartilage. Biochem J 330: 345351
Bessenyei B, Marka M, Urban L, Zeher M and Semsei I (2004)
Single nucleotide polymorphisms: aging and diseases.
Biogerontology 5: 291303
Breitkreutz BJ, Stark C and Tyers M (2003) The GRID: The
General Repository for Interaction Datasets. Genome Biol 4:
R23
Chang J, Van Remmen H, Cornell J, Richardson A and Ward
WF (2003) Comparative proteomics: characterization of a
two-dimensional gel electrophoresis system to study the eect
of aging on mitochondrial proteins. Mech Ageing Dev 124:
3341
Chen W, Ji J, Xu X, He S and Ru B (2003) Proteomic
comparison between human young and old brains by twodimensional gel electrophoresis and identication of proteins.
Int J Dev Neurosci 21: 209216
Cho YM, Bae SH, Choi BK, Cho SY, Song CW, Yoo JK and
Paik YK (2003) Dierential expression of the liver proteome
in senescence accelerated mice. Proteomics 3: 18831894
Davenport RJ, Strauss E and LaMarco K (2001) A SAGE KE
Primer. Sci Aging Knowledge Environ 2001: vp1
de Magalhaes JP and Toussaint O (2004) GenAge: a genomic
and proteomic network map of human ageing. FEBS Lett
571: 243247
de Magalhaes JP, Costa J and Toussaint O (2005) HAGR: the
Human Ageing Genomic Resources. Nucleic Acids Res
33(Database Issue), D537543
Deshpande KS, Harsha HC and Pandey A (2003) Microarrays
and mass spectrometry the future of proteomics. Curr
Genomics 4: 605613
Dieterich C, Wang H, Rateitschak K, Luz H and Vingron M
(2003) CORG: a database for COmparative Regulatory
Genomics. Nucleic Acids Res 31: 5557
Doniger SW, Salomonis N, Dahlquist KD, Vranizan K, Lawlor
SC and Conklin BR (2003) MAPPFinder: using Gene
Ontology and GenMAPP to create a global gene-expression
prole from microarray data. Genome Biol 4: R7
Ferrarini L, Bertelli L, Feala J, McCulloch AD and Paternostro
G (2004) A more ecient search strategy for aging genes
based on connectivity. Bioinformatics 21: 338348
Furuyama T, Nakazawa T, Nakano I and Mori N (2000)
Identication of the dierential distribution patterns of
mRNAs and consensus binding sequences for mouse
DAF-16 homologues. Biochem J 349: 629634
Gershon H and Gershon D (2000) The budding yeast,
Saccharomyces cerevisiae, as a model for aging research: a
critical review. Mech Ageing Dev 120: 122
Gershon H and Gershon D (2002) Caenorhabditis elegans a
paradigm for aging research: advantages and limitations.
Mech Ageing Dev 123: 261274
Goto M (1997) Hierarchical deterioration of body systems in
Werners syndrome: implications for normal ageing. Mech
Ageing Dev 98: 239254
Gygi SP, Rist B, Gerber SA, Turecek F, Gelb MH and
Aebersold R (1999) Quantitative analysis of complex protein
mixtures using isotope-coded anity tags. Nat Biotechnol 17:
994999

Hood L (2003) Systems biology: integrating technology, biology, and computation. Mech Ageing Dev 124: 916
Kaeberlein M, Jegalian B and McVey M (2002) AGEID: a
database of aging genes and interventions. Mech Ageing Dev
123: 11151119
Kimura KD, Tissenbaum HA, Liu Y and Ruvkun G (1997)
daf-2, an insulin receptor-like gene that regulates longevity
and diapause in Caenorhabditis elegans. Science 277: 942946
Konova E, Baydano S, Atanasova M and Velkova A (2004)
Age-related changes in the glycation of human aortic elastin.
Exp Gerontol 39: 249254
Kyng KJ, May A, Kolvraa S and Bohr VA (2003) Gene
expression proling in Werner syndrome closely resembles
that of normal aging. Proc Natl Acad Sci USA 100: 12259
12264
Lee SS, Kennedy S, Tolonen AC and Ruvkun G (2003a)
DAF-16 target genes that control C. elegans life-span and
metabolism. Science 300: 644647
Lee SS, Lee RY, Fraser AG, Kamath RS, Ahringer J and
Ruvkun G (2003b) A systematic RNAi screen identies a
critical role for mitochondria in C. elegans longevity. Nat
Genet 33: 4048
Lin K, Dorman JB, Rodan A and Kenyon C (1997) daf-16: an
HNF-3/forkhead family member that can function to double
the life-span of Caenorhabditis elegans. Science 278: 1319
1322
Ly DH, Lockhart DJ, Lerner RA and Schultz PG (2000)
Mitotic misregulation and human aging. Science 287: 2486
2492
McCarroll SA, Murphy CT, Zou S, Pletcher SD, Chin CS, Jan
YN, Kenyon C, Bargmann CI and Li H (2004) Comparing
genomic expression patterns across species identies shared
transcriptional prole in aging. Nat Genet 36: 197204
Ong SE, Blagoev B, Kratchmarova I, Kristensen DB, Steen H,
Pandey A and Mann M (2002) Stable isotope labeling by
amino acids in cell culture, SILAC, as a simple and accurate
approach to expression proteomics. Mol Cell Proteomics 1:
376386
Ogg S, Paradis S, Gottlieb S, Patterson GI, Lee L, Tissenbaum
HA and Ruvkun G (1997) The Fork head transcription
factor DAF-16 transduces insulin-like metabolic and longevity signals in C. elegans. Nature 389: 994999
Paik DC, Dillon J, Galicia E and Tilson MD (2001) The nitrite/
collagen reaction: non-enzymatic nitration as a model system
for age-related damage. Connect Tissue Res 42: 111122
Pavlidis P, Qin J, Arango V, Mann JJ and Sibille E (2004)
Using the gene ontology for microarray data mining: a
comparison of methods and application to age eects in
human prefrontal cortex. Neurochem Res 29: 12131222
Peri S, Navarro JD, Amanchy R, Kristiansen TZ, Jonnalagadda
CK, Surendranath V, Niranjan V, Muthusamy B, Gandhi TK,
Gronborg M, Ibarrola N, Deshpande N, Shanker K,
Shivashankar HN, Rashmi BP, Ramya MA, Zhao Z, Chandrika KN, Padma N, Harsha HC, Yatish AJ, Kavitha MP,
Menezes M, Choudhury DR, Suresh S, Ghosh N, Saravana R,
Chandran S, Krishna S, Joy M, Anand SK, Madavan V, Joseph
A, Wong GW, Schiemann WP, Constantinescu SN, Huang L,
Khosravi-Far R, Steen H, Tewari M, Ghaari S, Blobe GC,

232
Dang CV, Garcia JG, Pevsner J, Jensen ON, Roepstor P,
Deshpande KS, Chinnaiyan AM, Hamosh A, Chakravarti A
and Pandey A (2003) Development of human protein reference
database as an initial platform for approaching systems biology
in humans. Genome Res 13: 23632371
Poon HF, Castegna A, Farr SA, Thongboonkerd V, Lynn BC,
Banks WA, Morley JE, Klein JB and Buttereld DA (2004)
Quantitative proteomics analysis of specic protein expression and oxidative modication in aged senescenceaccelerated-prone 8 mice brain. Neuroscience 126: 915926
Robinson LJ, Karlsson NG, Weiss AS and Packer NH (2003)
Proteomic analysis of the genetic premature aging disease
Hutchinson Gilford progeria syndrome reveals dierential

protein expression and glycosylation. J Proteome Res 2: 556

557
Soreghan BA, Yang F, Thomas SN, Hsu J and Yang AJ (2003)
High-throughput proteomic-based identication of oxidatively induced protein carbonylation in mouse brain. Pharm
Res 20: 17131720
Strauss E and LaMarco K (2002) Science of aging knowledge
environment: one-stop shopping for researchers in the eld of
aging. Exp Gerontol 37: 12971301
Tollet-Egnell P, Flores-Morales A, Stahlberg N, Malek RL,
Lee N and Norstedt G (2001) Gene expression prole of the
aging process in rat liver: normalizing eects of growth
hormone replacement. Mol Endocrinol 15: 308318