Apoptosis Protection and Survival Signal by the CD53 Tetraspanin Antigen

Apoptosis protection and survival signal by the CD53 tetraspanin antigen

Mo´ nica Yunta and Pedro A Lazo*

Centro de Investigacioń del Cańcer, Instituto de Biologıá Molecular y Celular del Cańcer, Consejo Superior de Investigaciones Cientı´ficas, Universidad de Salamanca, Campus Miguel de Unamuno, E-37007 Salamanca, Spain

The CD53 antigen is a tetraspanin protein of the EWI-2 (Stipp et al., 2001); they also interact with MHC lymphoid–myeloid lineage, but its implication in biological molecules and other membrane proteins, such as effects is hardly known. Radioresistant tumor cells integrins, particularly those with the b1 subunit express very high levels of this antigen. We have studied (Berditchevski, 2001; Boucheix and Rubinstein, the effect of CD53 antigen ligation on the survival 2001). Tetraspanin proteins can influence several biolo- response of tumor cells to serum deprivation, a well- gical processes, such as cell motility (Higashiyama et al., known stimulator of cell death that may mimic the tumor 1995; Miyake et al., 1995), and homotypic adhesion environment; for this aim IR938F and Jurkat cells, a B- (Masellis-Smith et al., 1990; Lagaudriere-Gesbert et al., and T-cell lymphoma, were used. Ligation of CD53 1997; Lazo et al., 1997; Levy et al., 1998; Yanez-Mo triggers a survival response and reduces the number of et al., 1998; Shibagaki et al., 1999). There isa general cells that enter apoptosis. In CD53- stimulated cells there consensus that tetraspanin molecules modulate interac- is a significant reduction in caspase activation, measured tion with the extracellular environment, and recently the by caspase processing of poly ADP-ribose polymerase, as specific interaction of CD9 with fibronectin has been well as a reduction in the fragmentation of DNA. CD53- characterized (Longhurst et al., 2002), but in general stimulated cells also have an increase in the level of bcl-XL tetraspanin natural ligands are still unknown. Signaling and a reduction of bax protein, two components of the by tetraspanin proteins have generally been attributed to mitochondrial apoptotic pathway, changing their ratio by their interactionswith other membrane proteins,and as 24-fold in the direction of survival. This survival signal such characterized as costimulatory molecules. How- appears to be mediated by activation of the AKT, as ever, tetraspanin proteins have not yet been associated detected by its phosphorylation in Ser473 upon CD53 with the control of specific biological processes, ligation. The CD53 antigen interactions might contribute although some reports show that some antigens to cell survival in poorly vascularized regions of the tumor modulate cell proliferation, cell adhesion and motility. mass. Tetraspanin antigens have been individually studied Oncogene (2003) 22, 1219–1224. doi:10.1038/sj.onc.1206183 in many tumor types. In general, a reduction in antigen levelswascorrelated with poor prognosis(Radford et al., Keywords: CD53; tetraspanin antigen; apoptosis; AKT 1995), such as is the case for CD9 in lung (Higashiyama et al., 1997), breast (Huang et al., 1998) and esophagus carcinomas(Uchida et al., 1999); CD82 in prostate (Dong et al., 1995), pancreatic (Guo et al., 1996) and The tetraspanin proteins constitute a family of breast cancer (Huang et al., 1998); or CD63 in 26 membrane proteinswith four hydrophobic trans- melanoma (Radford et al., 1995). The role of CD9 has membrane domainsand two extracellular loops,which been related to itsmodulation of cell motility, sincethe may account for their extracellular interactions reintroduction of CD9 in the cell actsasa brake and functional specificity. These proteins form com- (Ikeyama et al., 1993). In addition, CD53 deficiency has plexes among themselves, of which the subunits are a clinical phenotype similar to those of inherited defects individual tetraspanin proteins. Some of the tetraspanin of cell adhesion molecules (Mollinedo et al., 1997). proteins are rather ubiquitous, such as CD9 or CD81, Tumor cells express simultaneously several tetraspanin but others have a more restricted pattern of expression antigensthat partially reflect a developmental stage,as asCD37, CD53, peripherin and uroplakin (Berditch- shown in Burkitt lymphomas (Ferrer et al., 1998). In evski, 2001; Boucheix and Rubinstein, 2001; Hemler, other lymphomas, CD53 is overexpressed in germinal 2001). Tetraspanins have been detected both as center and in follicular B-cell lymphomas(Alizadeh free molecules, or interacting with different types of et al., 2000). As a result of the complex tetraspanin receptors, such as EGF-R, HB-EGF (Wang et al., 2002) pattern of gene expression, it is very likely that the or proteinswith immunoglobulin domains,suchas adhesion and migration properties of tumors are conditioned by the alteration in the composition of cell-specific expression patterns (Sugiura and Berditch- *Correspondence: PA Lazo; E-mail: [email protected] evski, 1999; Testa et al., 1999). Received 31 July 2002; revised 18 October 2002; accepted 23 In thisreport, we focusedour attention on the CD53 October 2002 antigen, a tetraspanin whose expression is mainly Apoptosis protection by CD53 M Yunta and PA Lazo 1220 restricted to the lymphoid–myeloid lineage (Maecker cellswere analysedby flow cytometry to follow their et al., 1997), and which isupregulated in radiation- progression through the cell cycle, or the induction resistant cells (Voehringer et al., 2000). Many cells, when of cell death as a result of the starvation. The result they lose their extracellular matrix interactions or are for IR938F cellsisshownin Figure 2a. After 24 h in placed under poor growth conditions, undergo apopto- the presence of antibody, the number of cells in the sis (Frisch and Ruoslahti, 1997; Frisch and Screaton, apoptosis window is 9.82% (s.d. ¼ 4.53, n ¼ 5), while 2001). In thiswork, we have studied the effect of CD53 in the control nontreated cellsthe apoptotic population antigen ligation as a surrogate mechanism for mimick- represents 25.6% (s.d. ¼ 5.53, n ¼ 5). The control ing itsinteractionswith the extracellular environment, cellsclearly startto die after 24 h, while the cellsin which triggers a protective response against the apop- the presence of anti-CD53 progress through the tosis induced by serum deprivation. cell cycle, with an increase in the S and G2/M First, the presence of the CD53 antigen in rat IR938F populations. At later time points (72 h), the differences cellsand human Jurkat cellswasdetermined by flow are still clearly noticeable despite the severe starvation cytometry (Figure 1). To study the effects of CD53 conditions. ligation on cell survival, we used the cell line IR938F, To demonstrate that the protective effect of CD53 derived from a rat B-cell lymphoma, which expresses ligation isnot restricted to a cell type or antibody, a high levels of CD53 and in which several of its signaling similar experiment was performed using a human cell componentshave already been characterized (Bosca ´ and line, Jurkat cellsderived from a T-cell lymphoma, which Lazo, 1994; Barcia et al., 1996; Lazo et al., 1997). These expresses high levels of the CD53 antigen (Yunta et al., cells, at a density of 5 Â 105 cellsper milliliter, were 2002). After overnight serum deprivation of Jurkat cells, deprived of serum for 24 h before the addition of 10 mg the monoclonal antibody MEM53, specific for human of MRC OX44 specific for the rat CD53 antigen CD53, wasadded to the culture and itseffect was (Angelisova et al., 1990). At different time points, the analysed by flow cytometry. The result is shown in Figure 2b. At 24 h after the addition of MEM53 antibody, the apoptotic population represents 11.9% of the population, while in the control cellsit is21.5%. Therefore, it wasconcluded that ligation of the CD53 antigen inducesa high protection of the apoptosis occurring in different cells as a consequence of serum deprivation. If CD53 ligation promotescell survival, it islikely that it will also reduce the level of activation of the execution mechanism of apoptosis triggered by the serum deprivation response. The execution of the apoptotic process is mediated by the activation of caspases (Igney and Krammer, 2002; Shi, 2002). Activated caspase-3 degrades poly ADP-ribose polymerase (PARP) and generates, in addition to the intact protein of 113 kDa, a shorter proteolytic fragment of 89 kDa. As shown in Figure 3a, at the same time points, more noticeably at 12 and 24 h, the band of 85 kDa isclearly detected in the untreated control cells, but not in the MRC OX44-treated cells. A similar effect was observed in Jurkat cells in the presence of MEM53 (not shown). Also, the ligation of CD53 reducesthe fragmentation of total DNA in response to starvation, as shown for IR938F cells(Figure 3b). Theseobservationsconfirmthat the ligation of the CD53 antibody isprotecting the cells from the activation of apoptosis as a response to serum deprivation. It islikely that if CD53 ligation inducesapoptosis protection, moleculesthat regulate apoptosiswill modify their levelsin the direction of a protective effect. Figure 1 Flow cytometry analysis of rat IR938F cells and human The bcl-XL and bax proteinshave oppositeeffectson Jurkat cells showing the expression of the CD53 antigen on the cell apoptosis; while bcl-XL hasa protective effect (Boise surface. In the case of IR938F cells, the specific antibody used was et al., 1993), bax isa promoter of cell death (Miyashita MRC OX44. In the case of Jurkat cells, the CD53-specific antibody et al., 1994). Therefore, the levelsof both proteinswere was MEM53. The specific signal is the darker profile. The lighter trace is the isotype control antibody used for background signal. monitored in starved IR938F cells in the presence of The analysis was performed in a FACScalibur flow cytometer from MRC OX44 or in controls. As shown in Figure 4, Becton-Dickinson following the ligation of the CD53 antigen there isan

Oncogene Apoptosis protection by CD53 M Yunta and PA Lazo 1221

Figure 2 Protection from apoptosis by ligation of the CD53 antigen in rat IR938F cells (a) and human Jurkat cells( b). The cellswere analysed by flow cytometry at specific time points after antibody addition, OX44 or MEM53. The reduction in the apoptosis window (M1) is clearly detectable at 24 h, as indicated by the arrow in both cell types. M2 represents the G0/C1 cell cycle phase, M3 represents the S phase, and M4 represents the G2/M cell cycle population. The DNA contents in different cell populations were determined by ethidium bromide staining in a FACScalibur flow cytometer (Becton-Dickinson). The rat B-cell immunocytoma IR938F was grown in DMEM supplemented with 10% fetal calf serum. The human Jurkat cell line, from a T-cell lymphoma, was grown in DMEM supplemented with 10% fetal calf serum. The cells were grown at 371C with 5% CO2 and 98% humidity

increase in the level of bcl-XL that reachesa maximum by OX44 were sensitive to inhibitors of PI3kinase (Lazo of a sixfold increase (Figure 4a), as well as a fourfold et al., 1997). Therefore, it isvery likely that the survival decrease in the level of proapoptotic bax (Figure 4b). signal generated by CD53 ligation might implicate Overall, these variations in protein levels represent a the AKT pathway. To test this possibility, the phos- 24-fold change in the bcl-XL /bax ratio in favor of a phorylation of AKT in Ser473 as part of the response to protective reaction in the presence of MRC OX44. CD53 ligation wasdetermined. Rat IR938F cellswere These apoptosis protective changes are already notice- treated with the MRC OX44 antibody and the able at 9–12 h poststimulation. All these observations phosphorylation of AKT was determined at different are consistent with the generation of a signal in CD53 time points. As shown in Figure 5, there is a peak of that triggers a protective response against the apoptosis AKT phosphorylation in Ser473 10 min after addition of induced by serum deprivation that implicates the the MEM53 antibody, suggesting that this might be a mitochondrial pathway (Igney and Krammer, 2002; component of the survival signal initiated by CD53 Shi, 2002). antigen ligation. A similar effect was also induced in One of the main cellular survival pathways is Jurkat cells. mediated by the activation of the PI3kinase-AKT The signaling pathways implicated in apoptosis pathway (Gottlieb et al., 2002). AKT isactivated by protection are complex. In IR938F cellsand in Jurkat phosphorylation in Thr308 and Ser473 (Scheid cells, in addition to the phosphorylation of AKT and Woodgett, 2001). Several studies have previously detected in thisreport, other pathwaysimplicated shown that phosphoinositol kinases are activated as in proliferation and apoptosis are also activated. part of the response to CD53 ligation (Bosca´ and Lazo, The signaling pathways activated by CD53 antigen 1994; Zhang et al., 2001), and the effectsinduced ligation to initiate a survival response are likely to be

Oncogene Apoptosis protection by CD53 M Yunta and PA Lazo 1222

Figure 4 Effect of CD53 ligation by MRC OX44 in the level of bcl-XL and bax in IR938F cells. The time after antibody addition is indicated. Each gel lane wasloaded with 100 mg of whole cell extract. The blot wasdeveloped with a specificantibody for either bcl-XL or bax proteins. The gel loading was determined by the detection of b-actin that wasusedfor the normalization of bandsin the quantification of the digitized blots. The specific antibodies for Figure 3 CD53 ligation reduces PARP processing by caspase bcl-XL (clone 2H12) and for bax (clone 6A7) were from BD- activation (a) and total DNA fragmentation (b) in IR938F cells. Pharmingen (San Diego, CA, USA). The blot wasdeveloped with Assay of proteolytic processing of PARP by caspase activation. an ECL chemiluminescence kit from Amersham IR928F cells, after serum deprivation for 12 h, were incubated in the presence of a control or of the MRC OX44 antibody, specific for the rat CD53 antigen. The complete PARP protein has 113 kDa, and the caspase processed PARP fragment has 89 kDa. The PARP protein wasdetected with a monoclonal antibody from Enzyme Systems Products (Livermore, CA, USA) at a 1 : 5000 dilution for Western blots. The antibody against the PCNA antigen was obtained from Transduction Laboratories and used at a 1 : 5000 dilution for western blots. The control for loading was b- actin, which wasdetected with the AC-15 antibody from Sigma (St Louis, MO, USA) and used at a 1 : 5000 dilution. The products were analysed by SDS–PAGE under denaturing conditions and transferred to Immobilon-P membranes (Millipore, Bedford, MA, USA). The membraneswere blocked with 5% skimmedmilk in PBS, and then incubated with the specific antibody, followed by a rabbit anti-mouse IgG with peroxidase and developed with an ECL chemiluminescence kit from Amersham. The films were digitized at high resolution in a UMAX scanner. Quantification was performed using either the PCBas program from Fujibas phos- phorimager (Fuji, Japan) or a Personal Imager from Bio-Rad (Hercules, CA, USA). The total DNA (10 mg) wasanalysedin a 1% agarose gel in TAE with 1 kb size marker (Invitrogen, Carlsbad, CA, USA) Figure 5 Phosphorylation of AKT as a response to ligation of the CD53 antigen in IR938F with MRC OX44 antibody (top) and Jurkat cellswith MEM53 (bottom) . The time after antibody addition isindicated. Ligation of CD53 antigen in IR938F cells. very complex, and not just mediated by a unique The upper Western blot was developed with an antiphospho-AKT pathway. It isknown that CD53 ligation activates (Ser473) polyclonal antibody from Cell Signaling (Beverley, MA, USA), and the total AKT protein wasdetected with a polyclonal phospholipase C and protein kinase C (Bosca´ and Lazo, anti-AKT antibody from Santa Cruz (Santa Cruz, CA, USA). The 1994; Barcia et al., 1996; Lazo et al., 1997); the latter gel wasloaded with 100 mg of whole cell extract even physically interacts with tetraspanin antigens after itstranslocation to the membrane (Zhang et al., 2001). Also, there is a transient activation of the N-terminal Jun kinase (JNK) after CD53 ligation (Yunta et al., 2002). All these pathways, protein kinase C (Musashi adhesion, both are likely to be a consequence of the et al., 2000; Herr and Debatin, 2001), JNK (Herr and simultaneous activation of several signaling pathways Debatin, 2001) and AKT (Kennedy et al., 1997) have initiated in CD53. been associated with apoptosis protection. In IR938F In thisreport, it isshownthat ligation of the rat cells, the ligation of CD53 also induced homotypic CD53 antigen clearly protectsfrom the apoptotic adhesion (Lazo et al., 1997); a similar effect on adhesion response that is initiated as a result of serum depriva- isinduced in human Jurkat cells,which ismore tion. noticeable after 24 h (Figure 6). Asa result of the timing There is some evidence where the presence of CD53 implicated in the two effects, apoptosis protection and antigen has been correlated with cell survival. First, the

Oncogene Apoptosis protection by CD53 M Yunta and PA Lazo 1223 presence of CD53 is implicated in thymic selection; CD53-expressing cells are those that survive and mature (Puls et al., 2002). Second, the rescue of monocytes induced by GM-CSF, mediated by the transcription factor PU.1, isaccompanied by an increasein the expression of CD53 (Henkel et al., 2002). Interestingly, PU.1 isone of the main regulatory factorsof the CD53 gene (Herna´ ndez-Torres et al., 2001). Finally, there are some highly specialized members of the tetraspanin family, such as the RDS/peripherin protein in the retina (Travis et al., 1991), where mutationsresult in the apoptotic degeneration of retinal cells and subsequent blindness (van Soest et al., 1999). Also, CD53 reﬂects a phenotype of resistance to radiation, as has been demonstrated by the overexpression of CD53 (Voehrin- ger et al., 2000). In conclusion, the observations made in this report suggest that CD53 antigen stimulation might, under some circumstances, have a protective role against programmed cell death that might partly explain a role for CD53 when the antigen is overexpressed in tumors.

Acknowledgements Figure 6 Homotypic adhesion of Jurkat cells. Morphology of Jurkat cellsat different timesafter antibody addition to illustrate This work was supported by grants from Ministerio de Ciencia the large cellular aggregatesinduced by the speciﬁcantibody. The y Tecnologı´ a (SAF2000/0169), Junta de Castilla y Leo´ n (CSI1/ antibody wasadded to the culture medium at a concentration of 01) and Fondo de Investigacio´ n Sanitaria (FIS-PI02-0585) to 10 mg/ml PAL.

References

Alizadeh AA, Eisen MB, Davis RE, Ma C, Lossos IS, Guo X, Friess H, Graber HU, Kashigawi M, Zimmerman A, Rosenwald A, Boldrick JC, Sabet H, Tran T, Yu X, Powell Kore M and Buchler MW. (1996). Cancer Res., 56, 4876– JI, Yang L, Marti GE, Moore T, Hudson Jr J, Lu L, Lewis 4880. DB, Tibshirani R, Sherlock G, Chan WC, Greiner TC, Hemler ME. (2001). J. Cell Biol., 155, 1103–1108. Weisenburger DD, Armitage JO, Warnke R, Staudt LM. Henkel GW, McKercher SR and Maki RA. (2002). Int. (2000). Nature, 403, 503–511. Immunol., 14, 723–732. Angelisova P, Vlcek C, Stefanova I, Lipoldova M and Horejsi Herna´ ndez-TorresJ, Yunta M and Lazo PA. (2001). J. Biol. V. (1990). Immunogenetics, 32, 281–285. Chem., 276, 35405–35413. Barcia R, Garcı´ a-VargasS, BoscaL and Lazo PA. (1996). Herr I and Debatin KM. (2001). Blood, 98, 2603–2614. Cell. Immunol., 169, 107–112. Higashiyama M, Doi O, Kodama K, Yokouchi H, Adachi M, Berditchevski F. (2001). J. Cell Sci., 114, 4143–4151. Huang C, Taki T, Kasugai T, Ishiguro S, Nakamori S and Boise LH, Gonzalez-Garcia M, Postema CE, Ding L, Lindsten Miyake M. (1997). Int. J. Cancer, 205–211. T, Turka LA, Mao X, Nunez G and Thompson CB. (1993). Higashiyama M, Taki T, Ieki Y, Adachi M, Huang C, Koh T, Cell, 74, 597–608. Kodama K, Doi O and Miyake M. (1995). Cancer Res., 55, Bosca´ L and Lazo PA. (1994). J. Exp. Med., 179, 1119 6040–6044. 1126. Huang CI, Kohno N, Ogawa E, Adachi M, Taki T and Boucheix C and Rubinstein E. (2001). Cell. Mol. Life Sci., 58, Miyake M. (1998). Am. J. Pathol., 153, 973–983. 1189–1205. Igney FH and Krammer PH. (2002). Nature Rev. Cancer, 2, Dong J, Lamb PW, Rinker-Schaeffer CW, Vukanovic J, 277–288. Ichikawa T, Isaacs JT and Barrett JC. (1995). Science, 268, Ikeyama S, Koyama M, Yamaoko M, Sasada R and Miyake 884–886. M. (1993). J. Exp. Med., 177, 1231–1237. Ferrer M, Yunta M and Lazo PA. (1998). Clin. Exp. Immunol., Kennedy SG, Wagner AJ, Conzen SD, Jordan J, Bellacosa A, 113, 346–352. Tsichlis PN and Hay N. (1997). Genes Dev., 11, Frisch SM and Ruoslahti E. (1997). Curr. Opin. Cell Biol., 9, 701–713. 701–706. Lagaudriere-Gesbert C, La Naour F, Lebel-Binay S, Billard Frisch SM and Screaton RA. (2001). Curr. Opin. Cell Biol., 13, M, Lemichez E, Boquet P, Boucheix C, Conjeaud H and 555–562. Rubinstein E. (1997). Cell. Immunol., 182, 105–112. Gottlieb TM, Leal JF, Seger R, Taya Y and Oren M. (2002). Lazo PA, CuevasL, Gutierrez del Arroyo A and Orue E. Oncogene, 21, 1299–1303. (1997). Cell. Immunol., 178, 132–140.

Oncogene Apoptosis protection by CD53 M Yunta and PA Lazo 1224 Levy S, Todd SC and Maecker HT. (1998). Annu. Rev. Shibagaki N, Hanada K, Yamashita H, Shimada S and Immunol., 16, 89–109. Hamada H. (1999). Eur. J. Immunol., 29, 4081–4091. Longhurst CM, Jacobs JD, White MM, Crossno Jr JT, Stipp CS, Kolesnikova TV and Hemler ME. (2001). J. Biol. Fitzgerald DA, Bao J, Fitzgerald TJ, Raghow R and Chem., 276, 40545–40554. JenningsLK. (2002). J. Biol. Chem, 277, 32445–32452. Sugiura T and Berditchevski F. (1999). J. Cell Biol., 146, 1375– Maecker HT, Todd SC and Levy S. (1997). FASEB J., 11, 1389. 428–442. Testa JE, Brooks PC, Lin JM and Quigley JP. (1999). Cancer Masellis-Smith A, Jensen GS, Seehafer JG, Slupsky JR and Res., 59, 3812–3820. Shaw ARE. (1990). J. Immunol., 144, 1607–1613. TravisGH, Sutcliffe JG and Bok D. (1991). Neuron, 6, 61–70. Miyake M, Nakano K, Ieki Y, Adachi M, Huang C, Itoi S, Uchida S, Shimada Y, Watanabe G, Li ZG, Hong T, Miyake Koh T and Taki T. (1995). Cancer Res., 55, 4127–4131. M and Imamura M. (1999). Br. J. Cancer, 79, 1168–1173. Miyashita T, Krajewski S, Krajewski M, Wang HG, Lin HK, van Soest S, Westerveld A, de Jong PT, Bleeker-Wagemakers Lieberman DA, Hoffman B and Reed JC. (1994). Oncogene, EM and Bergen AA. (1999). Surv. Ophthalmol., 43, 321–334. 9, 1799–1805. Voehringer DW, Hirschberg DL, Xiao J, Lu Q, Roederer M, Mollinedo F, Fonta´ n G, Barasoaı´ n I and Lazo PA. (1997). Lock CB, Herzenberg LA and Steinman L. (2000). Proc. Clin. Diag. Lab. Immunol., 4, 229–231. Natl. Acad. Sci. USA, 97, 2680–2685. Musashi M, Ota S and Shiroshita N. (2000). Int. J. Hematol., Wang YD, De Vos J, Jourdan M, Couderc G, Lu ZY, Rossi 72, 12–19. JF and Klein B. (2002). Oncogene, 21, 2584–2592. PulsKL, HogquistKA, Reilly N and Wright MD. (2002). Int Yanez-Mo M, Alfranca A, CabanasC, Marazuela M, Tejedor Immunol, 14, 249–258. R, Ursa MA, Ashman LK, de Landazuri MO and Sanchez- Radford KJ, Mallesh J and Hersey P. (1995). Int. J. Cancer, Madrid F. (1998). J. Cell Biol., 141, 791–804. 62, 631–635. Yunta M, Oliva JL, Barcia R, Horejsi V, Angelisova P and Scheid MP and Woodgett JR. (2001). Nat. Rev. Mol. Cell. Lazo PA. (2002). Eur J. Biochem. 269, 1012–1021. Biol., 2, 760–768. Zhang XA, Bontrager AL and Hemler ME. (2001). J. Biol. Shi Y. (2002). Mol. Cell, 9, 459–470. Chem. 276, 25005–25003.

Oncogene