Identification of differentially expressed proteins in
senescent human embryonic fibroblasts
Ioannis P. Trougakosa, Aggeliki Saridakib, George Panayotoub,
Efstathios S. Gonosa,*
aLaboratory of Molecular & Cellular Ageing, Institute of Biological Research & Biotechnology,
National Hellenic Research Foundation, 48 Vas. Constantinou Ave., Athens 11635, Greece
bProtein Chemistry Laboratory, Biomedical Sciences Research Center ‘‘Alexander Fleming’’, Athens 16672, Greece
Received 15 July 2005; accepted 24 August 2005
Available online 5 October 2005
Normal human fibroblasts undergo a limited number of divisions in culture, a process known as replicative senescence (RS). Although several
senescence-specific genes have been identified, analysis at the level of protein expression can provide additional insights into the mechanisms that
regulate RS. We have performed a proteomic comparison between young and replicative senescent human embryonic WI-38 fibroblasts and we
have identified 13 proteins, which are differentially expressed in senescent cells. Some of the identified proteins are components of the cellular
cytoskeleton, while others are implicated in key cellular functions including metabolism and energy production, Ca2+signalling, nucleo-
cytoplasmic trafficking and telomerase activity regulation. In summary, our analysis contributes to the list of senescence-associated proteins by
identifying new biomarkers and provides novel information on functional protein networks that are perturbed during replicative senescence of
human fibroblast cultures.
# 2005 Elsevier Ireland Ltd. All rights reserved.
Keywords: 2D PAGE; Ageing; Human fibroblasts; Mass spectrometry; Proteomics; Replicative senescence
Ageing is the outcome of complicated interactions between
genetic factors and the accumulation of a variety of deleterious
stochastic changes overtime (Kirkwood, 2002). Human ageing
can be studied in vitro. Specifically, normal human fibroblasts
undergo a limited number of divisions in culture and
progressively reach a state of irreversible growth arrest, a
process termed as replicative senescence (RS). Replicative
senescence occurs because, owing to the biochemistry of DNA
replication, cells acquire one or more critically short telomere
(Holliday, 1996). Senescent cells have recently been shown to
accumulate with age in human tissues and it, thus, has been
proposed that they contribute to organismal ageing (Campisi,
2000). Moreover, senescent cells acquire phenotypic changes
that may contribute to certain age-related diseases, including
late-life cancer (Campisi, 2005).
Several genes that are linked to senescence have been
successfully identified and cloned on the basis of mRNA
expression changes between young and senescent cells (Gonos
et al., 1998; Lee et al., 1999; Ly et al., 2000). However, these
studies do not consider theweakness of the correlation between
a given level of transcripts and the abundance of the
corresponding proteins. It is well established that analysis at
the level of protein expression can provide additional and
complementary information. The development of proteomic
analysis methods using high-resolution two-dimensional gel
electrophoresis(2DGE) and mass-spectrometry (MS) offers the
advantage of identifying directly changes at the protein level.
Regarding senescence, this methodology has been successfully
applied in identifying new biomarkers during stress-induced
premature senescence (SIPS) in human diploid fibroblasts
(HDFs) (Dierick et al., 2002), as well as in conditionally
immortalized rat embryo fibroblasts (Benvenuti et al., 2002). In
this study,we have performeda proteomic comparison between
Mechanisms of Ageing and Development 127 (2006) 88–92
* Corresponding author. Tel.: +30 210 7273756; fax: +30 210 7273677.
E-mail address: email@example.com (E.S. Gonos).
0047-6374/$ – see front matter # 2005 Elsevier Ireland Ltd. All rights reserved.
young and senescent human embryonic WI-38 fibroblasts and
we have identified 13 differentially expressed proteins.
2. Materials and methods
2.1. Cell culture
Human diploid WI-38 fibroblasts were obtained from the European Collec-
tion of Cell Cultures and were maintained in Dulbecco’s modified Eagle’s
medium (Gibco Life Technologies, Inc.), supplemented with 10% (v/v) foetal
calf serum, 2 mM glutamine, 100 U/ml penicillin, 100 mg/ml streptomycin and
1% non-essential amino acids. Proliferating WI-38 fibroblasts were sub-cul-
tured at a split ratio 1:2 when theywere confluent until theyreached senescence
at about 45 cell population doublings (CPD) (Petropoulou et al., 2001). In all
experimental procedures, cells were fed approximately 16 h prior to the assay.
2.2. Sample preparation, 2DGE
Triplicates of WI-38 cell cultures at CPD21 or CPD45 were rinsed with ice-
coldPBSandthenlysedin 2%CHAPS, 15 mg/mlDTTand50 mMTrispH7.4.
Treatment with benzonase at 37 8C for 1 h removed interfering nucleic acids
from cell extracts and proteins were further purified by acetone precipitation at
?20 8C for 3 h. The protein pellet was finally re-suspended in 350 ml re-
hydration buffer (7 M urea, 2 M thiourea, 2% CHAPS, 10 mg/ml DTT, 2% IPG
buffer, 0.01% bromophenol blue) and after the removal of any insoluble
material, samples were loaded on immobilised pH gradient strips (pH 3–10
non-linear, Pharmacia Biotech). Isoelectric focusing on a MultiPhor II appa-
ratus (Pharmacia Biotech) and preparation of the IPG strips for the second
dimension were carried as per manufacturer’s instructions. After the isoelectric
focusing, the IPG strips were embedded on 12% polyacrylamide gel slabs for
the second dimension run according to the discontinuous buffer system of
2.3. In-gel tryptic digestion
Protein spots were visualised after staining of the gels with silver nitrate.
Protein patterns in distinct gels were manually compared. We deliberately
avoided the use of automated, software-driven procedures (e.g. the Melanie
software), because in our experience they increase the number of artefacts and
are useful only in high-throughput studies where manual inspection of a large
protein expression or those not repeated in all three independent experiments
and,thusweselectedonlyforthosewitha highlevel ofconfidence. Anexample
is given in Fig. 1A. Spots which were senescence-specific, were detected and
their expression levels (individual spot intensities) were compared from
scanned gel images by using the software Image Quant (Molecular Dynamics).
Spots were then excised from gels and cut into small pieces. The gel
particles were destained with acetonitrile and proteins were reduced with
10 mM DTT at 56 8C for 1 h and alkylated with 10 mg/ml iodoacetamide
for 45 min in the dark before being subjected to digestion with 5 ng/ml trypsin
overnight. The resulting peptides were eluted from gel particles following
successivewashes with 50% acetonitrile/5% formic acid and then concentrated
by centrifugal lyophilisation that removed the organic solvent. The volume of
the sample was adjusted to 10 ml with 0.1% formic acid and injected in the
reversed-phase nano-high performance liquid chromatography (nano-HPLC)
2.4. Nano-HPLC separation, nano-spray ion-trap mass spectrometer
Nano-HPLC was performed for the separation of the tryptic peptides using
the ultimate HPLC system (LC Packings) and a PepMap reversed phase C18
column (75 mm ? 15 cm, LC Packings). The injected samples were eluted at
150 nl/min with a 5–80% (v/v) acetonitrile/water gradient containing 0.1%
formic acid over 35 min. The separated tryptic fragments were visualised by
detection of the absorbance at 214 nm and were introduced online into a LCQ
Deca NSI-MS equipped with a nano-electrospray source (ThermoFinnigan).
During the first scanning event, the spectrum representing the mass to charge
ratio (m/z) of all the ions detected was collected on a real-time basis. The mass
of the most abundant ion was then calculated based on the data obtained by the
secondzoomscanandoncethe parentionwasidentified, themassspectrometer
was set up to obtain the collision-induced dissociation MS–MS spectrum.
The data were collected using the Xcalibur software (Finnigan) and were
subsequently used to search protein databases with the TurboSequest software.
The following default score values were used as cut-off parameters during the
search:Xcorr> 1.0,DCn> 0.1,Sp> 500,Rsp< 10andapeptidemasstolerance
of 1.0. Following this initial filtering, all candidate peptides were further
evaluated by manual inspection of the spectra and the Sequest parameters.
The ions produced were checked for each peptide, with an acceptance limit of
>60% coverage. The Xcorr values were also evaluated depending on the length
of the peptide, since longer peptides produce greater numbers. Only peptides
ending with a lysine or arginine residue were included. Finally, for each
identified protein the predicted or published molecular weight was checked
with that observed on the gels. An example of such analysis is given in Fig. 1B.
3. Results and discussion
We visualised approximately 1500 protein spots after gel
staining with silver nitrate (not shown), of which 13 were
senescence-specific. As shown in Table 1, eight proteins (a-
enolase, b-actin, annexins I and VI, creatine kinase B chain,
glutathione transferase omega-1, tubulin b-1 chain and
vimentin) were consistently found up-regulated in senescent
fibroblasts, whereas five proteins (RAN specific GTPase-
activating protein, type II keratin subunit, telomerase binding
protein p23, L-lactase dehydrogenase A chain and the ATP-
dependent RNA helicase p47) were found to be down-regulated
in all samples preparations. Few of these proteins have been
linked previously to ageing and/or cellular senescence in
various tissues and animal species. For instance, a-enolase, b-
actin, annexin I and creatine kinase B chain were also isolated
by Dierick et al. (2002) after a proteomic analysis of HDFs
The observed differential up-regulation of the cytoskeletal
proteins during RS, is in accordance with the changes in the
cytoskeleton structure and cell morphology which accompany
the senescence phenotype. More specifically, b-actin is known
to play a role in the reorganization of the cytoskeleton in
conditions generating stress fibres such as SIPS (Chen et al.,
2000). Tubulins contribute to the formation of microtubules,
which are known to be modified during cellular ageing (Raes,
1991). Vimentin has been directly linked to the senescence
phenotype since its over-expression has been shown to induce a
2001). Finally, considering that the primary function of type I
and type II keratins is to impact mechanical strength to cells
(Kirfelet al., 2003),the observed down-regulation of the type II
keratin subunit may be related to the fragility of the senescent
Three proteins involved in energy production and/or
metabolism were identified as senescence-specific. In parti-
cular, a-enolase, a highly conserved cytoplasmic glycolytic
enzyme, has been linked to ageing of the heart of Fisher 344/
Brown Norway F1 rats (Kanski et al., 2005). Interestingly, a-
enolase is down-regulated during non-small cell lung cancer
I.P. Trougakos et al./Mechanisms of Ageing and Development 127 (2006) 88–92 89
I.P. Trougakos et al./Mechanisms of Ageing and Development 127 (2006) 88–9290
Fig.1. (A)Detailsofrepresentative 2DgelsofCPD21andCPD45cells,showinga characteristic changeinproteinexpression(arrowed).Asisevidentbycomparing
are non-lineargels.(B)Anexampleofb0andy0fragmentsidentifiedbyMS/MSanalysisofa TelomeraseBindingProteinpeptide,theb0andy0fragmentsarepeptide-
in the right panel.
Proteins identified in the proteomic analysis of replicative senescent WI-38 human fibroblasts
ESI spot identificationAccession number
Up-regulated in senescence
Creatine kinase B chain
Glutathione transferase omega 1
Tubulin b-1 chain
Down-regulated in senescence
Type II keratin subunit
Telomerase-binding protein p23
L-Lactate dehydrogenase A chain
Probable ATP-dependent RNA helicase p47
aEstimate of the fold-difference in protein spot intensity; +++, >5-fold; ++, >2–5-fold; +, 2-fold.
and it has been suggested that it may function as a tumour
suppressor (Chang et al., 2003). Creatine kinase isoenzymes
play a central role in energy transduction in tissues with large,
fluctuating energy demands, such as skeletal muscle, heart,
brain and spermatozoa. Creatine kinase was found to be up-
regulated after a proteome analysis in the brains of the aged
senescence-accelerated-prone-mouse-strain 8 (SAMP8) (Poon
et al., 2004). In support to our data, both a-enolase and creatine
kinase B chain enzymes were found by Dierick et al. (2002) to
be up-regulated in HDFs undergoing RS. L-Lactate dehydro-
genase A chain (LDHA) is involved in anaerobic glycolysis. In
recent studies, it has been found to be deregulated in cancer
(Ishikawa et al., 2004), while its expression in human
fibroblasts is suppressed by the catalytic subunit of telomerase
(hTERT) and is induced by the Ataxia telangiectasia mutated
(ATM) gene (Baross et al., 2004).
The function of glutathione-transferase omega-1 (GSTO1)
is not well understood but recent data have associated GSTO1
with the Alzheimer and Parkinson diseases (Li et al., 2003)
which are both related to advance ageing. ATP-dependent RNA
helicase p47 protein encodes for a nuclear member of the
DEAD protein family of ATP-dependent RNA helicases with
unknown function (Peelman et al., 1995). The DEAD protein
family has more than 40 members, including the eukaryotic
translation initiation factor-4A (eIF-4A), the human nuclear
vasa (Peelman et al., 1995).
Annexins are a family of calcium and membrane-binding
proteins that have been involved in diverse cellular functions
including vesicle trafficking, cell division, apoptosis, growth
regulation and calcium signalling (Hayes and Moss, 2004).
a proteome analysis ofinvitro cultured fibroblasts from healthy
subjects of different ages (Boraldi et al., 2003), has an anti-
inflammatory effect and exerts profound inhibitory effects on
both neutrophil and monocyte migration. It was also recently
recognised and ingested by the phagocytes (Parente and Solito,
2004). This latter function is of particular interest, if one
considers the necessity for removal of the senescent cells that
gradually accumulate in a living organism during ageing
(Krtolica and Campisi, 2002). Notably, annexin I was
previously found to be down-regulated in senescent HDFs
(Dierick et al., 2002). This discrepancy can be attributed to the
complexity of the systems being analysed (i.e. whole
proteomes), differences between independent clones of the
same cell line as well as methodological variations.
Two other RS biomarkers, which are both involved in key
cellular functions, are the RAN-specific GTPase-activating
protein (RanGAP) and the telomerase-binding protein p23.
RanGAP binds to the GTPase Ran and, along with RCC1 and
RanBP1, regulate the action of this protein, which is a well
known regulator of protein transport across the nuclear
envelope (Steggerda and Paschal, 2002). In agreement with
our data which suggest a reduced nucleo-cytoplasmic traffick-
ing in senescent cells, other factors of the nucleo-cytoplasmic
during ageing in human dermal fibroblasts obtained from
young, mature and old donors, both at the mRNA and protein
level (Ly et al., 2000; Pujol et al., 2002). Considering that this
down-regulation was accompanied by a reduction in protein
import in fibroblasts derived from old donors (Pujol et al.,
2002), it seems that diminished nucleo-cytoplasmic trafficking
is a major characteristic of the RS phenotype. p23 is a
molecular chaperone that along with other five subunits
compose the telomerase complex. Blockade of the interaction
between p23 and hTERT inhibits the assembly of active
telomerase both in vitro and in vivo (Holt et al., 1999), while
antisense treatment of the p23 mRNA results in decreased or
abolished telomerase activity (Chang et al., 2002). Since
telomerase activation represents a key event towards cellular
immortalisation and eventual transformation(Rangarajan et al.,
2004), p23 down-regulation in senescent cells may represent an
additional blockade to the pro-oncogenic event of telomerase
In conclusion, our analysis has identified novel biomarkers
of RS, providing information on functional protein networks
that are perturbed during RS of human fibroblast cultures.
the HellenicGeneralSecretariat ofResearchandTechnology to
E.S.G. and to G.P. and by a European Union full-cost grant
(‘‘Functionage’’: QLK6-CT-2001-00310) to E.S.G.
Baross, A., Schertzer, M., Zuyderduyn, S.D., Jones, S.J., Marra, M.A., Lans-
dorp, P.M., 2004. Effect of TERT and ATM on gene expression profiles in
human fibroblasts. Genes Chromosomes Cancer 39, 298–310.
Benvenuti, S., Cramer, R., Bruce, J., Waterfield, M.D., Jat, P.S., 2002. Identi-
fication of novel candidates for replicative senescence by functional pro-
teomics. Oncogene 21, 4403–4413.
Boraldi, F., Bini, L., Liberatori, S., Armini, A., Pallini, V., Tiozzo, R., Pasquali-
Ronchetti, I., Quaglino, D., 2003. Proteome analysis of dermal fibroblasts
cultured in vitro from human healthy subjects of different ages. Proteomics
Campisi, J., 2000. Cancer, aging and cellular senescence. In Vivo 14, 183–188.
Campisi, J., 2005. Senescent cells, tumor suppression and organismal aging:
good citizens, bad neighbors. Cell 120, 513–522.
Chang, J.T., Chen, Y.L., Yang, H.T., Chen, C.Y., Cheng, A.J., 2002. Differential
regulation of telomerase activity by six telomerase subunits. Eur. J. Bio-
chem. 269, 3442–3450.
Chang, Y.S., Wu, W., Walsh, G., Hong, W.K., Mao, L., 2003. Enolase-alpha is
frequently down-regulated in non-small cell lung cancer and predicts
aggressive biological behavior. Clin. Cancer Res. 9, 3641–3644.
Chen, Q.M., Tu, V.C., Catania, J., Burton, M., Toussaint, O., Dilley, T., 2000.
Involvement of Rb family proteins, focal adhesion proteins and protein
Sci. 113, 4087–4097.
Dierick, J.F., Kalume, D.E., Wenders, F., Salmon, M., Dieu, M., Raes, M.,
Roepstorff, P., Toussaint, O., 2002. Identification of 30 protein species
involved in replicative senescence and stress-induced premature senes-
cence. FEBS Lett. 531, 499–504.
Gonos, E.S., Derventzi, A., Kveiborg, M., Agiostratidou, G., Kassem, M.,
Clark, B.F.C., Jat, P.S., Rattan, S.I.S., 1998. Cloning and identification of
genes that associate with mammalian senescence. Exp. Cell Res. 240, 66–
I.P. Trougakos et al./Mechanisms of Ageing and Development 127 (2006) 88–92 91
Hayes, M.J., Moss, S.E., 2004. Annexins and disease. Biochem. Biophys. Res.
Commun. 322, 1166–1170.
Holliday, R., 1996. Endless quest. Bioessays 18, 3–5.
Holt, S.E., Aisner, D.L., Baur, J., Tesmer, V.M., Dy, M., Ouellette, M., Trager,
J.B., Morin, G.B., Toft, D.O., Shay, J.W., Wright, W.E., White, M.A., 1999.
Functional requirement of p23 and Hsp90 in telomerase complexes. Genes
Dev. 13, 817–826.
Ishikawa, J., Taniguchi, T., Higashi, H., Miura, K., Suzuki, K., Takeshita, A.,
Maekawa, M., 2004. High lactate dehydrogenase isoenzyme 1 in a patient
LDHA gene. Clin. Chem. 50, 1826–1828.
Kanski, J., Behring, A., Pelling, J., Schoneich, C., 2005. Proteomic identifica-
tion of 3-nitrotyrosine-containing rat cardiac proteins: effects of biological
aging. Am. J. Physiol. Heart Circ. Physiol. 288, H371–H381.
Kirfel, J., Magin, T.M., Reichelt, J., 2003. Keratins: a structural scaffold with
emerging functions. Cell Mol. Life Sci. 60, 56–71.
Kirkwood, T.B., 2002. Evolution of ageing. Mech. Ageing Dev. 123, 737–745.
Krtolica, A., Campisi, J., 2002. Cancer and aging: a model for the cancer
promotingeffects ofthe agingstroma.Int. J. Biochem.Cell. Biol.34,1401–
Li, Y.J., Oliveira, S.A., Xu, P., Martin, E.R., Stenger, J.E., Scherzer, C.R.,
Hauser, M.A., Scott, W.K., Small, G.W., Nance, M.A., Watts, R.L., Hubble,
J.P., Koller, W.C., Pahwa, R., Stern, M.B., Hiner, B.C., Jankovic, J., Goetz,
C.G., Mastaglia, F., Middleton, L.T., Roses, A.D., Saunders, A.M., Schme-
chel, D.E., Gullans, S.R., Haines, J.L., Gilbert, J.R., Vance, J.M., Pericak-
Vance, M.A., Hulette, C., Welsh-Bohmer, K.A., 2003. Glutathione S-
transferase omega-1 modifies age-at-onset of Alzheimer disease and Par-
kinson disease. Hum. Mol. Genet. 12, 3259–3267.
Lee, C.K., Klopp, R.G., Weindruch, R., Prolla, T.A., 1999. Gene expression
profileof agingand itsretardationby caloricrestriction.Science 285,1390–
Ly, D.H., Lockhart, D.J., Lerner, R.A., Schultz, P.G., 2000. Mitotic misregula-
tion and human aging. Science 287, 2486–2492.
Nishio, K., Inoue, A., Qiao, S., Kondo, H., Mimura, A., 2001. Senescence and
cytoskeleton: overproduction of vimentin induces senescent-like morphol-
ogy in human fibroblasts. Histochem. Cell. Biol. 116, 321–327.
Parente, L., Solito, E., 2004. Annexin 1: more than an anti-phospholipase
protein. Inflamm. Res. 53, 125–132.
Peelman, L.J., Chardon, P., Nunes, M., Renard, C., Geffrotin, C., Vaiman, M.,
Van Zeveren, A., Coppieters, W., Van de Weghe, A., Bouquet, Y., Choy,
W.W., Strominger, J.L., Spies, T., 1995. The BAT1 gene in the MHC
encodes an evolutionarily conserved putative nuclear RNA helicase of the
DEAD family. Genomics 26, 210–218.
Petropoulou, C., Trougakos, I.P., Kolettas, E., Toussaint, O., Gonos, E.S., 2001.
Clusterin/apolipoprotein J is a novel biomarker of cellular senescence, that
Lett. 509, 287–297.
Poon, H.F., Castegna, A., Farr, S.A., Thongboonkerd, V., Lynn, B.C., Banks,
W.A., Morley, J.E., Klein, J.B., Butterfield, D.A., 2004. Quantitative
proteomics analysis of specific protein expression and oxidative modifica-
tion in aged senescence-accelerated-prone 8 mice brain. Neuroscience 126,
Pujol, G., Soderqvist, H., Radu, A., 2002. Age-associated reduction of nuclear
Raes, M., 1991. Involvement of microtubules in modifications associated with
cellular aging. Mutat. Res. 256, 149–168.
type-specific requirements for cellular transformation. Cancer Cell 6, 171–
Steggerda, S.M., Paschal, B.M., 2002. Regulation of nuclear import and export
by the GTPase Ran. Int. Rev. Cytol. 217, 41–91.
I.P. Trougakos et al./Mechanisms of Ageing and Development 127 (2006) 88–92 92