Ann. N.Y. Acad. Sci. ISSN 0077-8923

ANNALS OF THE NEW YORK ACADEMY OF SCIENCESIssue: Aging, Cancer, and Age-Related Diseases

Evaluating the role of p38 MAP kinase in growth of Wernersyndrome fibroblasts

Terence Davis,1 Marcus A. Bachler,1 Fiona S. Wyllie,1 Mark C. Bagley,2 and David Kipling1

1Department of Pathology, Cardiff University School of Medicine, Heath Park, Cardiff, United Kingdom. 2School of Chemistry,Main Building, Cardiff University, Park Place, Cardiff, United Kingdom

Address for correspondence: Dr. Terence Davis, Department of Pathology, School of Medicine, Henry Wellcome Building,Heath Park, Cardiff, CF14 4XN, Wales, UK. davist2@cardiff.ac.uk

The accelerated aging of Werner syndrome (WS) fibroblasts can be prevented by treatment with the p38 kinaseinhibitor SB203580. If accelerated cellular senescence underlies the premature ageing features seen in this humanaging model, then p38 inhibitors may have therapeutic potential in WS. However, SB203580 can inhibit in vitro

several kinases that are involved in control of cellular growth, in particular, c-Raf1, CK1, and RIPK2. Thus, a betterunderstanding of the role of this inhibitor in WS cells is required. Here we use a combination of kinase inhibitors andsmall intefering RNA-induced gene knockdown to show that it is inhibition of the stress-induced p38 MAP kinasethat is the most plausible explanation for the effects of SB203580 on the growth of WS cells. As the development ofhighly selective p38 inhibitors with low toxicity is a major effort of the pharmaceuticals sector, these studies helppave the way for possible therapeutics for WS.

Keywords: ageing; c-Raf1; casein kinase 1; Rip kinase 2; stress signalling; SB203580; replicative senescence

Werner syndrome (WS) is a premature aging dis-order widely used as a model of normal human ag-ing.1 Individuals with WS display a premature onsetof many age-related phenotypes, including juvenilebilateral cataracts, tight skin and skin atrophy, pre-mature hair greying and hair loss, and early onsetof age-related inflammatory diseases such as type IIdiabetes mellitus, osteoporosis, arteriosclerosis, andatherosclerosis. Not all tissues show premature ag-ing, however, as there is no immune dysfunction oraccelerated central nervous system aging.

Human cells have a limited cell division capac-ity before entering the replicative senescence that ispostulated to contribute to normal aging,1 and WSfibroblasts have a much-reduced life span.2 How-ever, not all WS cell types show this accelerated ag-ing, with T cells having a normal replicative lifespan.3 As WS individuals manifest accelerated ag-ing in the skin but not in the immune system, ithas been postulated that the mechanism of WS mayinvolve accelerated cell aging.1

The premature senescence of WS cells may be astress response, as the stress-activated p38 MAPK

pathway is activated in young WS cells. Treatmentwith the p38 inhibitor SB203580 essentially revertsthe WS cellular phenotype by increasing the growthrate and the cellular life span of primary WS cellsto within the ranges seen for normal fibroblasts andrescuing their senescence-like morphology, thus im-plicating a role for both p38 and stress signalingin WS.4 As the premature replicative senescence isthought to play a role in the premature aging in WSand as p38 activation plays a role in inflammatorydiseases,5 these data suggest that p38 inhibitors mayhave an application as possible therapeutics for thispremature aging syndrome.

However, although the most plausible target forthe actions of SB203580 is p38, other kinases thatplay a role in cellular growth may be inhibited bySB203580, especially at the concentrations used inthe previous study, including c-Raf1, casein kinase1 (CK1), and receptor-interacting serine-threoninekinase 2 (RIPK2).6 In this work we used a combi-nation of genetic and kinase inhibitor approachesto explore the possibility that SB203580 exerts itsgrowth-promoting effects by inhibition of these

doi: 10.1111/j.1749-6632.2010.05195.xAnn. N.Y. Acad. Sci. 1197 (2010) 45–48 c© 2010 New York Academy of Sciences. 45

p38 MAP kinase and Werner syndrome Davis et al.

kinases. Two cell types were used in these experi-ments, both immortalized by the ectopic expressionof human telomerase.7 The WS cells are adult der-mal fibroblast strain AG03141 from the Coriell CellBank. 4 The HCA2 cells are normal human diploidneonatal foreskin fibroblasts kindly supplied by J.Smith (Houston, TX, USA).8 Growth conditionswere as previously described.4 The small interferingRNA (siRNA) transfection protocol has been pre-viously described;9 the siRNA used was a validatedsiRNA for RIPK2 (Ambion AM51325).

The c-Raf1 MAPK kinase kinase functions in thegrowth proliferative pathway in response to growthfactors. SB203580 inhibits c-Raf1 (50% inhibitoryconcentration [IC50] = 2 �M) and, as SB203580prevents accelerated aging of WS cells at 10 �M,4

this may be due to c-Raf1 inhibition. GW5074 is aspecific c-Raf1 inhibitor that shows complete in vivoinhibition at 1.0 �M and does not inhibit p38.10

Treatment with GW5074 up to 5.0 �M has no effecton the growth rate or cellular morphology of WScells (Fig. 1A and data not shown), suggesting thatinhibition of c-Raf1 is not responsible for the effectson WS cells that are seen with SB203580. However,it has been shown that despite c-Raf1 inhibition theMEK1/2–ERK1/2 pathway remains active and thereis no change in ERK1/2 phosphorylation.10,11 In-deed, there is no reduction of MEK1/2 or ERK1/2phosphorylation in GW5074-treated WS or HCA2cells (Fig. 1B). A possible interpretation is that inhi-bition of c-Raf1 upregulates a compensatory path-way that activates MEK1/2, the nature of which isnot fully understood.11 The observation that the in-hibition of c-Raf1 by SB203580 (and other c-Raf1inhibitors, such as ZM336372) is also counterbal-anced in this way10 provides support for this mech-anism in WS cells. Alternatively, it is possible that aconstitutive activation of MEK1/2 is seen in WS dueto the known genetic instability in this condition,although this is speculation. The conclusion is thateven if SB203580 treatment inhibited c-Raf1 in vivo,this would have little effect on the growth or mor-phology of WS cells, as c-Raf1 acts via MEK1/2 inproliferation, seemingly ruling out the involvementof c-Raf1.

CK1 has several isoforms involved in the trans-duction of the Wnt signaling pathway leading to cel-lular proliferation via activation of �-catenin. CK1�primes �-catenin degradation, and so inhibitingCK1� may activate cell growth.12 Conversely, CK1�

and CK1ε activate �-catenin signaling, so their in-hibition would inhibit cellular growth.13 IC261 is aCK1 inhibitor with an in vivo IC50 of 1.5 �M forCK1� and CK1ε and >16 �M for CK1�.14 At levelsaround and above the IC50 for CK1�, IC261 treat-ment results in cessation of cellular growth (Fig. 1C),in association with a change in morphology andapoptosis (Fig. 1D), in agreement with the obser-vation that IC261 at 1.0 �M induces apoptosis introphoblast cells that is due to CK1� and CK1ε in-hibition.15 As SB203580 has only been shown toinhibit CK1 isoforms in vitro,6 these data suggestthat SB203580 does not inhibit CK1 in vivo at thelevels used.

The kinase RIPK2 is upstream of p38 and is in-hibited by SB20580, with a lower IC50 than p38.6,16

Inhibition of RIPK2 can prevent the activation ofp38, an observation that could explain the down-regulation of p38 in WS cells by SB203580.4 As thereare no specific inhibitors for this kinase, an siRNAapproach was taken. Using a RIPK2 siRNA, success-ful knockdown of RIPK2 was seen in WS cells thatpersisted for at least 72 h (Fig. 1E). No knockdownwas seen in controls. Repeated siRNA transfectionsat 3-day intervals for 12 days were growth inhibitoryfor RIPK2, whereas the mock- and SCR-transfectedcells grew normally (Fig. 1F). SB203580 treatmentincreased the growth rate for control cells but hadno effect on RIPK2-transfected cells. Similar effectsare seen with HCA2 cells (data not shown). Thedata suggest that RIPK2 inhibition may not be in-volved in the positive growth effect of SB203580 onWS cells, as SB203580 is a known in vivo RIPK2 in-hibitor.16 This also suggests that the downregulationof p38 activity in WS cells treated with SB203580 isdue to an alternate mechanism, perhaps by prevent-ing p38 autophopshorylation.17 It must be notedthat knockdown of RIPK2 is not equivalent to in-hibition of the RIPK2 kinase activity, and a fullerunderstanding of possible roles for RIPK2 may re-quire specific inhibitors for the kinase activity ofRIPK2.

The above data strongly support the hypothe-sis that it is the inhibition of p38 and not othergrowth-related kinases that is responsible for therescue of the WS cellular phenotype by SB203580.As the development of novel p38 inhibitors withincreased specificity and efficacy and low toxicityis a major goal of the pharmaceuticals sector,5 andas a fully reflective mouse model of WS has been

46 Ann. N.Y. Acad. Sci. 1197 (2010) 45–48 c© 2010 New York Academy of Sciences.

Davis et al. p38 MAP kinase and Werner syndrome

Figure 1. (A, C) Growth rates of WS cells treated with either GW5074 or IC261, plotted as the percentage of thegrowth rate of untreated cells. The growth rate was defined as PD = log2(N f /N i), where Nf is the final cell count andNi is the initial cell count. ∗, complete growth inhibition. (B) Immunoblot of WS and HCA2 cells treated with 5 �MGW5074 and probed for ERK1/2, phospho-(T202/Y204)ERK1/2 (pERK1/2), MEK1/2, and phospho(S218/S222)-MEK1/2 (pMEK1/2). (D) Bright-field view of WS cells treated with IC261 at 1 �M (arrows indicate apoptotic cells).Bar = 100 �m. (E) Immunoblot of siRNA-treated WS cells probed for RIPK2 and actin (SCR, scrambled siRNA).(F) Growth curves of siRNA-treated WS cells: squares are control cells, circles are SCR-treated cells, triangles areRIPK2 siRNA-treated cells, and closed symbols are cells treated with SB203580. Arrows indicate times when cells weretransfected with siRNA.

recently reported,18 these studies make future proof-of-principle studies possible as to whether p38 in-hibitors will be effective as therapeutic agents inpremature aging diseases. What makes this discov-ery of wider interest is the fact that p38 is also stimu-lated in a wide range of inflammatory conditions,19

including atherosclerosis, type II diabetes, and os-teoporosis, all with high rates of incidence in WSindividuals. As such it is tempting to classify WSas an example of the class of diseases known as“inflamm-aging.20 Moreover, as inflammatory dis-

eases are a major cause of morbidity during nor-mal human aging20 it is possible that p38 inhibitorsmay form the basis of antiaging therapeutics, notonly for individuals with WS but also for normalindividuals.

Acknowledgments

This work was supported by grants from theEPSRC/BBSRC SPARC and the BBSRC SCIBS(BB/D5241401) initiatives.

Ann. N.Y. Acad. Sci. 1197 (2010) 45–48 c© 2010 New York Academy of Sciences. 47

p38 MAP kinase and Werner syndrome Davis et al.

Conflicts of interest

The authors declare no conflicts of interest.

References

1. Kipling, D. et al. 2004. What can progeroid syndromes

tell us about human aging? Science 305: 1426–1431.

2. Tollefsbol, T. & H. Cohen. 1984. Werner’s syndrome: an

underdiagnosed disorder resembling premature aging.

Age 75–88.

3. James, S.E. et al. 2000. Werner’s syndrome T lymphocytes

display a normal in vitro life-span. Mech. Ageing Dev.

121: 139–149.

4. Davis, T. et al. 2005. Prevention of accelerated cell aging

in Werner syndrome using a p38 mitogen-activated pro-

tein kinase inhibitor. J. Gerontol. A Biol. Sci. Med. Sci. 60:

1386–1393.

5. Force, T. et al. 2004. Inhibitors of protein kinase signaling

pathways: emerging therapies for cardiovascular disease.

Circulation 109: 1196–1205.

6. Godl, K. & H. Daub. 2004. Proteomic analysis of ki-

nase inhibitor selectivity and function. Cell Cycle 3: 393–

395.

7. Davis, T. et al. 2006. Prevention of accelerated cell aging

in the Werner syndrome. Ann. N.Y. Acad. Sci. 1067: 243–

247.

8. Gire, V. & D. Wynford-Thomas. 1998. Reinitiation of

DNA synthesis and cell division in senescent human fi-

broblasts by microinjection of p53 antibodies. Mol. Cell.

Biol. 18: 1611–1621.

9. Davis, T. et al. 2005. Replicative senescence in sheep

fibroblasts is a p53 dependent process. Exp. Gerontol. 40:

17–26.

10. Chin, P.C. et al. 2004. The c-Raf inhibitor GW5074 pro-

vides neuroprotection in vitro and in an animal model

of neurodegeneration through a MEK-ERK and Akt-

independent mechanism. J. Neurochem. 90: 595–608.

11. Hall-Jackson, C.A. et al. 1999. Effect of SB 203580 on

the activity of c-Raf in vitro and in vivo. Oncogene 18:

2047–2054.

12. Liu, C. et al. 2002. Control of beta-catenin phosphoryla-

tion/degradation by a dual-kinase mechanism. Cell 108:

837–847.

13. Peters, J.M. et al. 1999. Casein kinase I transduces Wnt

signals. Nature 401: 345–350.

14. Li, G., H. Yin & J. Kuret. 2004. Casein kinase 1 delta phos-

phorylates tau and disrupts its binding to microtubules.

J. Biol. Chem. 279: 15938–15945.

15. Stoter, M. et al. 2005. Inhibition of casein kinase I delta

alters mitotic spindle formation and induces apoptosis

in trophoblast cells. Oncogene 24: 7964–7975.

16. Argast, G.M., N. Fausto & J.S. Campbell. 2005. Inhibition

of RIP2/RIck/CARDIAK activity by pyridinyl imidazole

inhibitors of p38 MAPK. Mol. Cell. Biochem. 268: 129–

140.

17. Ge, B. et al. 2002. MAPKK-independent activation of

p38� mediated by TAB1-dependent autophosphoryla-

tion of p38�. Science 295: 1291–1294.

18. Chang, S. et al. 2004. Essential role of limiting telomeres

in the pathogenesis of Werner syndrome. Nat. Genet. 36:

877–882.

19. Davis, T. & D. Kipling. 2006. Werner syndrome as an

example of inflamm-aging: possible therapeutic oppor-

tunities for a progeroid syndrome? Rejuvenation Res. 9:

402–407.

20. Franceschi, C. 2007. Inflammaging as a major character-

istic of old people: can it be prevented or cured? Nutr.

Rev. 65: S173–176.

48 Ann. N.Y. Acad. Sci. 1197 (2010) 45–48 c© 2010 New York Academy of Sciences.