Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Multiple p21ras effector pathways regulate nuclear factor of activated T cells.

Multiple p21ras effector pathways regulate nuclear factor of activated T cells. The EMBO Journal vol. 15 no. 15 pp.3923-3933, 1996 Multiple p2lras effector pathways regulate nuclear factor of activated T cells the IL-2 gene (Baldari et al., 1992; Rayter et al., 1992). Elisabeth Genot', Steve Cleverley, The major target in the IL-2 gene for p21ras signalling Stefan Henning and Doreen Cantrell is NFAT pathways (Woodrow et al., 1993b). Thus, a Lymphocyte Activation Imperial Cancer Research Laboratory, Fund, constitutively active Ras mutant (p21 -v-Ha-ras), when 44 Lincoln's Inn Fields. London WC2A 3PX. UK in T expressed cells, can synergize with a calcium/ 'Corresponding author calcineurin signalling pathway to activate NFAT (Woodrow et al., 1993a,b). p2lras regulatory effects on The transcription factor, Nuclear Factor of Activated NFAT can be mimicked by phorbol esters that activate T cells (NFAT) is a major target for p2lras and calcium PKC or by expression of constitutively activated mutants signalling pathways in the IL-2 gene and is induced of PKC (Genot et al., 1995). The ability of phorbol esters by p2lras signals acting in synergy with calcium/ to activate p21ras in T cells led to the proposal that PKC calcineurin signals. One p2lras effector pathway effects on NFAT were mediated by p21 ras. However, it is involves the MAP kinase ERK-2, and we have examined now recognized that phorbol ester and p21ras induction its role in NFAT regulation. Expression of dominant of NFAT occur by different, independent mechanisms. In negative MAPKK-1 prevents NFAT induction. Con- particular, p2l.ras but not PKC signals are essential for stitutively active MAPKK-1 fully activates ERK-2 and TCR induction of NFAT (Williams et al., 1995). NFAT the transcription factor Elk-1, but does not substitute proteins are cytosolic in quiescent T cells but they trans- for activated p2lras and synergize with calcium/ locate to the nucleus in TCR-activated cells and combine calcineurin signals to induce NFAT. Expression of with newly induced AP-1 complexes to form a functional dominant negative N17Rac also prevents TCR and transcriptional factor unit (Flanagan et al., 1991). There p2lras activation of NFAT, but without interfering are multiple isoforms of the NFAT cytosolic subunit, with the ERK-2 pathway. The transcriptional activity NFATc, NFATp, NFAT3 and NFAT4, that have different of the NFAT binding site is mediated by a complex patterns of cellular expression but are thought to share comprising member of the NFAT group and AP-1 one property in that their nuclear translocation is controlled family proteins. The induction of AP-1 by p2lras also by calcium/calcineurin-dependent signals (Northrop et al., requires Rac-1 function. Activated Rac-1 could mimic 1994; Rao, 1994; Hoey et al., 1995). The different NFAT activated p2lras to AP-1 induce but not to induce proteins also share the ability to interact with AP-1 to NFAT. Moreover, the combination of activated bind co-operatively to the NFAT/AP- 1 site in composite MAPKK-1 and Rac-1 could not substitute for activated the IL-2 gene (Rao, 1994). The interaction with AP-1 is p21ras and synergize with calcium signals to induce essential for NFAT transcriptional activity, and the role of NFAT. Thus, p2lras regulation of NFAT in T cells p21 ras in NFAT induction has been the explained by requires the activity of multiple effector pathways ability of p21ras signals to stimulate AP-1 in T activity including those regulated by MAPKK-1/ERK-2 and cells (Rayter et al., 1992; Woodrow et al., How- 1993a). Rac-1. the ras effector involved in ever, p21 pathways NFAT NFAT/Rac-l/Ras/MAPKK-1/transduction/ Keiw!ords: regulation have not been defined and are the focus of the TCR present report. T at least two MAP lymphocytes express kinases, ERK- 1 and that are stimulated a ras ERK-2, by p21 in to TCR et pathway response triggering (Izquierdo al., Introduction The MAP kinases are activated a kinase cascade 1993). by involving a MAP kinase kinase that The T cell controls activation (MAPKK) phos- antigen receptor (TCR) and and stimulates the ERK- 1 and -2 kinases the GO-G1 transition of T and the phorylates lymphocytes regulates Cobb and The production of a number of the directly (Marshall, 1994; Goldsmith, 1995). cytokines, including growth of the MAPKK is itself controlled factor interleukin-2 The of IL-2 activity by phosphoryl- (IL-2). regulation produc- ation and, hence, a MAP kinase kinase kinase tion by the TCR requires the co-ordinate action of (MAPKKK) multiple effector ERK-1 and -2. Raf-1 is a that transcription factors that include Nuclear Factor of Activ- regulates p2lras T to be the MAPKKK that lras and ated cells NF-KB and Oct-1 appears couples p2 (NFAT), AP-1, (Schreiber et and The intracellular hence receptors to the MAP kinases Crabtree, 1992; Rao, 1994). signals (Howe al., 1992; et Leevers et factors involve Warne et that regulate these different transcription al., 1993; Izquierdo al., 1994a; al., and -2 are kinase-induced 1994). In neuronal MAPKK-1 tyrosine calcium/calcineurin, protein cells, ERK-1, ras functions et kinase C and critical mediators of al., (PKC) p2lras-mediated signalling pathways p21 (Cowley substrate for the ERKs is the and et 1994). One regulatory (Weiss Littman, 1994; Izquierdo-Pastor al., 1995). protein of a of that can form Inhibition of function in T cells of which is one p21ras by expression Elk-1, family proteins with the activator serum a dominant mutant induction of a negative p21ras prevents ternary complex transcriptional C Oxford Press University E.Genot et al. response factor (SRF) (Marais et al., 1993). Elk-l-SRF and which correlates with -2, the inhibitory effects of complexes are necessary and sufficient mediators of N17Ras on TCR activation of NFAT (Izquierdo et al., c-fos serum response element induction. Thus, p21ras regulation Woodrow et The MAP kinases 1993; al., 1993b). ERK-l of the MAPKK/ERK pathway plays a key role in regulating and -2 are activated a kinase by (MAPKK) that phosphoryl- gene expression (Hill and Treisman, 1995). The ates and stimulates the ERK- 1 and -2 kinases c-fos directly. analysis of the role of the ERKs in the control of gene MAPKK is activated by protein phosphorylation, and expression in fibroblasts has created a paradigm that the mutation of phosphorylation sites at position 217 and/or MAPKKIERK pathway is a major route 221 on whereby p21ras MAPKK- 1 from serine to alanine generates an regulates the transcriptional activity of immediate early inhibitory molecule that can suppress growth factor- genes such as By analogy, there is c-fos. the possibility mediated activation of the ERK kinase cascade (Cowley that the p2lras/MAPKK pathway is the effector et p2lras al., 1994). Accordingly, in initial experiments to explore pathway for NFAT induction. whether regulation of ERKs explains p21ras control of There has been one previous study on the role of NFAT, the effects of the inhibitory mutant of MAPKK MAPKK and the ERKs as p2 Iras effectors in T cells: this 221A on TCR (DN MAPKK) regulation of NFAT were was a study in transgenic mice that showed that analysed. To assess expression the activity of the transcription factor of a dominant negative MAPKK- 1 mutant could mimic NFAT, we employed transient transfection protocols and inhibitory p21 ras mutants and suppress positive selection quantitation of the of a expression CAT reporter gene of thymocytes (Alberola-Ila et al., 1995). However, domin- whose activity is regulated by an enhancer corresponding ant negative Ras mutants can prevent the in vitro to the NFAT binding site. The data in prolifera- Figure 1A show tion of thymocytes, whereas a dominant negative that this NFAT reporter gene is not expressed in unstimu- MAPKK- 1 mutant does not, indicating that not all Ras lated Jurkat cells but is induced in cells stimulated with functions in T cells MAP the UCHT- I are mediated by the kinases antibody that triggers the TCR-CD3 complex. cascade. The Ras-related Rac- 1 has been described GTPase TCR induction of NFAT can be mimicked by the combina- tion of recently as another downstream effector of p21ras (Hall, stimuli that activate PKC and elevate intracellular 1994; Qiu et the of the calcium levels such al., 1995) Accordingly, object as phorbol esters and calciutm iono- present study was to the connections phores. PKC and induce NFAT explore signalling p21ras independently and between the cascade and in similarly independently activate ERK-2 p21ras, MAPKK/ERK Rac-1 (Izquierdo et al., T cells in order to determine the role of these in 1993; Williams et al., 1995). There is proteins thus the possibility p2 lras regulation of NFAT. The data show that that PKC and Ras signals for NFAT induction expression converge of a dominant negative MAPKK-1 mutant can inhibit at the level of the MAP kinases. The data in Figure 1A NFAT induction, but expression of a active show that the phorbol ester constitutively phorbol 12,13 dibutyrate MAPKK- 1 mutant, which could activate ERK-2 in T (Pdbu) has no stimulatory effect on NFAT when used cells, as could not substitute for constitutively active and a single and the calcium ionophore p21ras stimulus, ionomycin with synergize calcium-dependent signals to induce NFAT. has only a weak inductive effect on NFAT. the However, These studies that the ERK of collectively suggest pathway combination calcium ionophore and Pdbu synergizes to is a but not sufficient maximal NFAT necessary p21ras effector pathway give activity. Expression of the inhibitory for NFAT induction in T cells. of a dominant MAPKK- 1 mutant in T cells Expression inhibits TCR induction of Rac- 1 mutant had a effect NFAT and also Pdbu and negative pronounced inhibitory suppresses ionomycin induction on Ras induction of NFAT, which that Rac-1 is of NFAT In four suggests (Figure IA). experiments, the inhibitory an effector for p21ras in NFAT regulation. In effect of the dominant MAPKK mutant fibroblasts, negative on the Rac-l-mediated intracellular signals participate in activa- TCR/NFAT response ranged from 77 to 50%, whereas the tion of Jun kinases and transactivation of c-jun in a inhibitory effects of this mutant on Pdbu plus ionomycin mechanism analogous to p21ras regulation of ERK-2- responses ranged from 30 to 40%. Elk-1 et The (Minden al., 1995). present data show that To monitor ERK activation in cells transfected with expression of active Rac-1 the constitutively can stimulate dominant negative MAPKK- I the mutant, inhibitory AP- 1 transcriptional activity in T mutant was cells, demonstrating that co-expressed in T cells with ERK-2tag, Rac-1 can also stimulate transcription factor in an pathways epitope-tagged ERK-2 construct. Activated ERK-2 is T 1 cells. However, constitutively active Rac- mutants phosphorylated as we have and, shown previously, activ- could not synergize with calcium signals to induce ated isolated from NFAT, ERK-2tag Pdbu- or TCR-activated T indicating that Rac-1 is necessary but not sufficient for cells has reduced electrophoretic mobility in SDS-PAGE all p21 ras-mediated signalling pathways in T cells. Accord- gels as compared with ERK-2tag isolated from quiescent ingly, p21ras effects on NFAT are not mediated by a cells (Izquierdo et al., 1994a,b) (Figure 1B). Accordingly, single linear signalling pathway but must involve the activation multiple of ERK-2tag can be monitored sensitively effectors that include Rac-1 and MAPKK-1. of by analysis its electrophoretic mobility. The data in Figure lB show that co-transfecting the dominant negative MAPKK- 1 mutant with ERK-2tag prevents ERK-2tag Results phosphorylation as induced by either TCR or phorbol ester. The effects of inhibitory and active MAPKK- 1 In further experiments, we examined the consequences mutants on NFAT induction of expressing a constitutively activated 1 MAPKK- mutant Activation of p2lras in T cells can be prevented by on NFAT induction. Mutation of residues 217 and 221 on expression of an inhibitory mutant of p2 Iras, N 17Ras MAPKK- I from serine to glutamic acid generates a (Izquierdo et al., 1993). Expression of N17Ras in T cells constitutively active enzyme, MAPKK-1 217E/221E also suppresses TCR activation of the 1 MAP kinases ERK- The (MAPKK*). expression of this constitutively active 3924 NFAT regulation in T cells A* _\ \ .I( \I * IN\ \1 \1Sk\ C- NI Al CAT _* I HTI _~~~~~~~~~~~~~~~~~~~~ ..... ._ , *. *ER K- 2tag ."p_inim' N. U ERK-2taga i I 1..1^ _ / 0 KJL - 0 Pdbu UCHT-1 _erairitv_. __ MAPKK N1^s DN Fig. 1. The effects of an inhibitory MAPKK mutant on NFAT induction. (A) 107 Jurkat cells were co-transfected by electroporation with the NFAT- CAT reporter construct (15 and the empty vector or the plasmid encoding the dominant negative MAPKK- I mutant (20 ,ug of pEXV3 MAPKK pg) 221A, DN MAPKK) as described in Materials and methods. At 6 h after transfection, transfected cells were either left unstimulated, or stimulated with UCHT-1 (10 .tg/ml), Pdbu (50 ng/ml), ionomycin (0.5 ig/ml) or a combination of Pdbu and ionomycin for 16 h before extracts were made, and CAT reporter activity was assessed. Data show the percentage conversion of chloramphenicol into the acetylated product. Results from one experiment representative of four are presented. (B) Jurkat cells were co-transfected with the pEF-BOS ERK-2tag construct (18 big) together with the empty vector or with the inhibitory MAPKK-1 mutant (20 p,g of pEXV3 MAPKK 221A). At 16 h after cells were left unstimulated transfection, or were stimulated with UCHT-1 (10 jtg/ml) or Pdbu (50 ng/ml) for 10 min. Total cellular extracts were made and run on SDS-PAGE (15% acrylamide-0.075% bisacrylamide). Data show Western blot analysis of ERK-2tag with the 9E10 antibody. (C) Jurkat cells were co-transfected with the NFAT-CAT (15 jg) reporter construct and the empty vector or the constitutively active Ras mutant (15 of pEXV3 v-Ha-ras. Ras*) or pIg constitutively active MAPKK-1 mutant (15 of ,ug pEXV3 MAPKK 217E/221E, MAPKK*). After 6 h. cells transfected with the empty vector were either left unstimulated or were stimulated with Pdbu (50 ng/ml). All sets of transfected cells were then either left unstimulated or were stimulated with ionomycin (0.5 or UCHT- I (10 jg/ml). CAT assay was performed as in (A). Data are shown as the percentage conversion of pg/ml) chloramphenicol into the acetylated product. Results from one experiment representative of five are presented. MAPKK- 1 will activate the MAP kinases ERK- 1 -2 and protein (MBP) as a substrate were carried out on ERK-2tag and has been shown to substitute for activated p21 ras in isolated from MAPKK- 1 immunoprecipitates activated many cellular responses in fibroblasts and neuronal cells or control transfected cells. We have used this system (Cowley et al., 1994). The data in Figure IC are represent- to previously demonstrate that expression of activated ative of five similar experiments and show that NFAT p2lras in T cells is sufficient to stimulate ERK-2 kinase responses were only marginally induced by the activated activity (Izquierdo et TCR induction of ERK-2 al., 1993). MAPKK-1 mutant in combination with ionomycin. No is transient and at 10-15 min In peaks post-stimulation. further induction of the NFAT-CAT construct the experiment presented in it is this reporter Figure 2B, peak could be seen by increasing or decreasing the amount of TCR response that is whereas in the transient shown, MAPKK-1 activated DNA transfected (data not transfection experiments the sustained effects of either shown). In contrast, the constitutively active Ras mutant, v-Ha- activated p21ras or MAPKK-1 mutants on ERK-2 are MAPKK- 1 mutant ras, which has no effect on NFAT alone, synergizes monitored. Expression of activated with in in ERK-2 kinase in strongly ionomycin for NFAT induction. The data induced an -5-fold increase activity 1C thus show that the activated MAPKK- 1 mutant ERK-2 Figure T cells. The activated p21ras induction of activity does not mimic activated for is of the induced the p2lras NFAT induction. 50% maximal response by TCR, To ensure that the failure of active MAPKK- 1 21 which is consistent with our of the 7E/ previous quantitation 221E mutant to NFAT in T cells did not reflect ERK-2 in T cells activated regulate activity expressing p2lras the of MAPKK-1 to stimulate the et The increase in ERK-2 inability ERKs, experi- (Izquierdo al., 1993). activity ments to monitor ERK-2 in cells the in cells activated MAPKK-1 mutant was to a activity expressing expressing activated activated MAPKK-1 mutant were In initial level which is to the effects of performed. equivalent p2lras the activated MAPKK- 1 mutant was co- on ERK-2 activity experiments, (Figure 2B). function of the transfected activated in T cells with The data in The effector expressed ERK-2tag. Figure T co-trans- 1 mutant on the ERK in Jurkat cells 2A show that isolated from cells MAPKK- endogenous ERK-2tag has a reduced was also examined its to fected with activated MAPKK-1 mutant by assessing ability regulate factor which in to the ERK-2 isolated of the SDS-PAGE phosphorylation transcription Elk-I, mobility similarly et activated T cells. To confirm that this shift is a cellular substrate for ERK-2 from (Marais al., 1993). mobility that can form a correlates with activation of the Elk- 1 is one of a of of ERK-2tag family proteins temary enzyme activator SRF and thus that the activated MAPKK-1 could with the and to demonstrate complex transcriptional in of stimulate ERK-2 kinase basic can a role activity, play key regulation c-fos gene expression assays using myelin 3925 E.Genot et al. )1 1t - ................ =E RK-? ). 1-0111ILVAVJNW :":::.: ... tag OD *ERK -2' tag -- f.hi - - - PdbL t- - --t Y.A K ermnpty KKP APKK wvt nio ERK--2Lj 0UJP _bwk _ Ik ~ 4_-L-exA 1-OP .h -_-- LexA FL;k- 1 '- t III Li -.7 O ( F'dl-,u I-tH r-l Ras" NIAPKK' Fig. 2. The activated MAPKK mutant is able to activate ERK. (A) Jurkat cells were co-transfected with the pEF-BOS construct ERK-2tag (12.5 ,ug) together with the empty or with the activated MAPKK-1 vector, mutant (12.5 of MAPKK or the jg pEXV3 217E/221E) MAPKK-1 wild-type (12.5 ,ug pEXV3 MAPKK wt). At 16 h cells were left post-transfection, unstimulated or were stimulated with Pdbu 50 for 10 ng/ml min. Total cellular extracts were made and run on SDS-PAGE. Data show Western blot of with the 9E10 Jurkat cells analysis ERK-2tag antibody. were (B) co-transfected with the pEF BOS construct with ERK-2tag (15 jig) together 25 of the vector or with the activated Ras of ,ug empty (25 jg pEXV3 v-Ha-ras) or MAPKK-1 (25 jig of pEXV3 MAPKK mutants. At h 217E/221E) 16 cells were left unstimulated or were stimulated post-transfection, with UCHT-1 (10 jg/ml) for 10 min. Cells were the was lysed, ERK-2tag immunoprecipitated with the 9E10 and kinase antibody was activity in assessed a kinase assay using MBP as substrate. into MBP was with Radioactivity incorporated the quantified phosphoimager (Molecular Dynamics). Data show the MBP kinase in as a of the activity ERK-2tag precipitates presented maximal obtained percentage response TCR by stimulation. (C) Jurkat cells were co-transfected with 15 jg of the the fusion N plasmid encoding with 20 protein, pEF Lex-Elk-IC, together ,ug of the vector or with the empty activated Ras (20 of or jig pEXV3 v-Ha-ras) MAPKK-1 (20 of MAPKK mutants. At 16 h jig pEXV3 217E/221E) post- transfection. cells were transfected with the vector and were either left unstimulated or were empty stimulated with or UCHT-1 (10 Pdbu jg/ml) (50 for 30 min. Total ng/ml) cell extracts were made and run on SDS-PAGE. Data show blot of Western LexA-Elk-lC with analysis LexA antibodies. (Hill and Treisman, 1995). It has been shown et The role previously (Qiu al., 1995). of Rac-l in NFAT regulation that activators of the ERK kinase cascade can induce in T cells was explored by monitoring the effects of a phosphorylation of the Elk-I C-terminus and dominant Rac- potentiate negative 1 mutant, N17Rac, and a constitu- Elk-i transcriptional activity (Price et To tively active al., 1995b). Rac-1 mutant, V12Rac (Rac*) (Nobes and monitor this a phosphorylation, fusion protein Hall, 1995). The data in 3A comprising Figure show that expression the C-terminus of Elk-I linked to the LexA of the dominant negative repressor N17Rac molecule inhibits NFAT (LexA-Elk- was co-transfected into T cells with 1C) activ- induction by the TCR and also, although to a lesser extent, ated or MAPKK-1 mutants. The p2iras suppresses the NFAT response phosphorylation induced by phorbol ester of LexA-Elk-1C reduces its electrophoretic in and calcium ionophore. In an that mobility experiment investigated SDS-PAGE gels. the data in 2C show Accordingly, Figure the N 17Rac dose dependency of NFAT induction, it the reduced electrophoretic of LexA-Elk-1 in mobility appeared that Pdbu and ionomycin induction of NFAT Pdbu- or TCR-activated cells with control compared comprised an N17Rac-sensitive and -insensitive com- unstimulated cells. LexA-Elk-1C isolated from T cells ponent (Figure 3B). To examine the selectivity of the either activated expressing p2 Iras or activated MAPKK- effects of the 1 dominant negative mutant of Rac- 1 on TCR mutant has a similar reduced electrophoretic signalling, the effects mobility of N 17Rac- on TCR and Pdbu (Figure 2C). The activated MAPKK and activated p2lras activation of ERK-2 were determined. Accordingly, T had equivalent effects on stimulatory the transcriptional cells were transfected with ERK-2tag either alone or of Elk-i activity (data not shown). together with N17Rac. The data in Figure 3C show that In the summary, data in Figures and 2 show that TCR or Pdbu stimulation or transfection of activated of a expression dominant mutant of negative MAPKK-1 p21ras induces ERK-2 activation while expression of suppresses TCR induction of NFAT. The data demonstrate activated Rac mutant has no effect on ERK-2 activity. also that expression of an activated MAPKK-1 mutant in Co-expressing N17Rac in these cells did not affect p21ras, T cells is sufficient for ERK-2 activation and sufficient TCR or Pdbu activation of ERK-2, showing that the for phosphorylation of the transcription factor Elk-1. inhibitory effects of Ni 7Rac did not result from inhibition However, expression of a constitutively active 1 MAPKK- of ERK-2 (Figure 3C). mutant is not sufficient to substitute for p21ras for NFAT To explore further the role of Rac- 1 in NFAT regulation, induction. Accordingly, to explain p21ras regulation of the effects of an activated Rac-1 mutant (V12Rac) on NFAT in T cells, it is necessary to invoke the existence transcription factor activity were compared with the effects of alternative Ras effector pathways. of the activated p2iras mutant. p2iras can regulate the transcription factor AP-1 activity in a variety of cells The effects of inhibitory and activated mutants of including T cells (Rayter et al., 1992; Woodrow et al., Rac- 1 on NFAT induction 1993b). To compare p2iras and Rac-1 regulation of the The Ras-related GTPase Rac-1 has been identified as a activity of AP- 1, we quantified the expression of an AP- 1- downstream effector of p21ras in fibroblast transformation CAT reporter gene. The AP- 1 reporter gene can be induced 3926 in T cells NFAT regulation i~~~ -- ~ ~ \ FAT CAT LC]ITI '! t *Pd*'bu+ i0000' (tCill .1 n ...... _ _ :\FT- C-x I ---4-.- ._ _ _ . _ _ o r _ lmicro-ram., 10 iiiillioii celk) _ _ V17 Rac concentration J '- _ | __ Il ..,I ~ --: ow &WOO -Mb.- 14'-- .Wmkm- 4movool qww E P,(- : ar: '- ,ti II li-[ 7 _- _ = "dhoLCHF-I. Ras* Rac* db Ras' Rac* Ci Pdbu UCHT-' :I' -= _ : .e~~~~~~~~~~~~~~~~~- em-nv Rac N ` __ 3. The effects of an inhibitory Rac mutant on NFAT induction. (A) Jurkat cells were co-transfected with the NFAT-CAT reporter construct Fig. (12 jig) and 5 jig of the empty vector or the dominant negative Rac mutant-containing plasmid (5 jig of pcDNAl N17Rac). At 6 h post-transfection, cells were either left unstimulated or were stimulated with UCHT-1 (10 ,ug/ml) or a combination of Pdbu (50 ng/ml) and ionomycin (0.5 jig/ml) for 16 h before extracts were made, and CAT reporter activity was assessed. Data show fold induction of NFAT-CAT activity. Results from one experiment representative of five are presented. (B) Jurkat cells were co-transfected with the NFAT-CAT reporter construct (12 jig) and the empty vector or different concentrations of the inhibitory Rac mutant (pcDNAl N17Rac). and the experiment was carried out as in (A). Data are shown as a percentage of the maximum NFAT-CAT induction obtained with UCHT-1 stimulation in the absence of the inhibitory Rac mutant. (C) Jurkat cells were co-transfected with the ERK-2tag construct (18 jig) together with 8 jig of the empty vector, or plasmids encoding active Ras (8 jig of pEF- jg Rac*) mutants, in the absence or presence of 5 jig of the inhibitory Rac mutant BOS v-Ha-ras) or the active Rac (8 of pEFplink V12Rac. (pcDNAl N17Rac). At 16 h post-transfection, cells transfected with the empty vector or with the plasmid encoding the inhibitory mutant were either (10 jg/ml) or Pdbu (50 ng/ml) for 10 min. Total cellular extracts were made and run on SDS- left unstimulated or were stimulated with UCHT-1 of ERK-2tag with the 9E10 antibody. PAGE. The data show Western blot analysis and show that the combination of by Pdbu, TCR triggering with UCHT1 antibodies presented in Figure 5C (Figure 4A). AP-1 1 activated MAPKK could not synergize constitutively active p2lras protein activated Rac- and activated Rac-1 mutant signals to induce NFAT. No further induction activity is also induced by the with calcium demonstrate that TCR construct could be seen when (Figure 4A). Further experiments of the NFAT-CAT reporter and Pdbu induction of AP- 1 activity in T cells are different ratios of the two plasmids (data not transfecting of the dominant negative Rac-1 The data in Figures 1-5 collectively show that prevented by expression shown). mutant N17Rac (Figure 4B). These data show that Rac-1 MAPKK- 1 and Rac- 1 function are each important but not is essential for TCR and Ras regulation of AP-1 sufficient, neither alone nor in combination, for TCR and activity in T cells. Moreover, expression of constitutively activated p21 ras regulation of NFAT. Rac-1 is sufficient to stimulate levels of AP-1 activity NFAT and AP- 1 in T cells with that seen in cells expressing activated p2lras regulation of comparable mutants (Figure 4A). To compare p2iras and requires Rac-1 function p21ras Rac-i in pathways, the MAPKK- 1 regulation of the activity of AP- 1, we quantified To position p2lras signalling in cells Rac- 1 mutants on signals the of the AP-1-CAT reporter gene effect of the dominant negative expression or the active p21ras mutant were either constitutively active p2iras protein generated by constitutively expressing in 4C show that monitored. of N I 7Rac- 1 did not prevent activated MAPKK-1. The data Figure Expression of activated of the constitutively active p21 ras mutant in T AP- 1-CAT is not induced by expression expression activated Rac-1 did not in T cells. cells not Moreover, MAPKK-1 mutant (data shown). nor did AP- 1 a of the active NFAT complex induce activation of ERK-2 (Figure 3C), expression Since is component Pdbu-mediated activation the effects of activated Rac-I of N17Rac or in T cells, we also assessed prevent p2iras- Results in 6A NFAT in T cells. The data in Figure 5A of ERK-2 (Figure 3C). presented Figure on responses of N17Rac did not activated Rac- 1 could not mimic activated show that expression prevent p2iras- show that of which is consistent and with calcium for NFAT mediated phosphorylation Elk-i, p21 ras synergize signals Rac- 1 to be seen when the failure of the dominant negative prevent induction. No NFAT induction could using with the ERK-2 for which activated activation of pathway (Figure the activated Rac in conditions p2 lras-mediated mutant, ERK kinase cascade can induce and under conditions where 3C). Activators of the Ras was effective 5B) (Figure activation the Elk-i C-terminus 1 induce AP- 1. Given that neither transcriptional by (Price activated Rac- could and To monitor Elk-I of activated MAPKK- 1 nor activated Rac- 1 et 1995b; Hill Treisman, 1995) al., expression LexA-Elk-1C was co-transfected ras for we the transcriptional activity, substitute for p21 NFAT, investigated pos- CAT T cells with a LexA reporter a between MAPKK- 1 and Rac- 1. Data into operator-controlled sibility of synergy 3927 E.Genot et al. .E~~ ~ ~ ~ ~ ~ ~ ~ AP'-1 CA *Ail d ,( Stiniuli .. ni c N t co ( il i N17 Rat- nlcentr dtio01 mr]icro1gra m1/l celll tt ITs)1 Fig. 4. The effects of Rac and MAPKK mutants on AP-1 induction. Jurkat cells were co-transfected with the AP-1-CAT construct (A) reporter (2 and 8 of the empty vector or the the active Ras or active Rac gg plasmid encoding constitutively (pEF-BOS v-Ha-ras) (pEFplinkVl2Rac) gg) mutants. After 6 h, transfected cells were either left unstimulated or were stimulated with either UCHT-1 or Pdbu for 16 h (10 .tg/ml) (50 ng/ml) Data a before extracts were made, and CAT was assessed. are as of the maximum AP-1-CAT reporter activity presented percentage response induced by Pdbu. (B) Jurkat cells were co-transfected with the AP-1-CAT construct and the vector or different concentrations of reporter (2 ,tg) empty the and was carried plasmid encoding the inhibitory Rac mutant (pcDNAI the out as in Data are shown as a of N17Rac), experiment (A). percentage the maximum obtained with Pdbu stimulation in the absence of the Rac mutant. Jurkat cells were co-transfected with the response inhibitory AP-1- (C) CAT construct and 15 of the vector or the active Ras mutant of or reporter (2 gg) gg empty constitutively (15 gg pEXV3 v-Ha-ras) constitutively active MAPKK-1 mutant of MAPKK At 6 h cells transfected with the vector were either (15 tg pEXV3 217E/221E). post-transfection, empty left unstimulated or were stimulated with either UCHT-1 or Pdbu for 16 h before extracts were and (10 .tg/ml) (50 ng/ml) made, CAT reporter activity was assessed. Data are shown as a of the conversion of into the percentage chloramphenicol acetylated product. LexA-Elk- 1 we gene (LexA-OP.tkCAT). C transcription examined the effects of activated Rac- on LexA- is low in unactivated T cells but can be induced Elk-i C The data in activity phosphorylation. Figure 6A show that of a active v-Ha- ofV I by co-expression constitutively p2lras, expression 2Rac- 1 in T cells induced phosphorylation Ras, or by stimulation of T cells with Pdbu of LexA-Elk- IC with that (Figure 6B). comparable induced by activa- Expression of N17Rac did not suppress Pdbu or v-Ha-ras ted p2Iras. Rac-1 function is not required for p2iras induction of LexA-Elk- C. In contrast, in the same induction of Elk-i (Figure 6A and B) which is mediated experiment, Ni7Rac inhibits the AP-1 induced via ERK-2. Nevertheless, these results indicate in T response that, the by constitutively active p21ras and also cells, Rac- has the potential to a kinase (Figure 6C) regulate cascade prevents NFAT induction induced by activated p2 lras plus that can converge with the ERK-2 on the tran- pathway calcium These ionophore (Figure 6D). data thus indicate scription factor Elk- 1. that Rac-I function is for required p21ras induction of AP-1 and NFAT but Rac-1 function is not for necessary Discussion activation p2iras of ERK-2 and Elk-1. Expression of activated Rac-1 did not activate ERK-2 The MAP kinase ERK-2 has been proposed to be a major in T cells (Figure 3C) nor was Rac-I function necessary effector of the p21ras signal transduction pathways in a for TCR- and p2 lras-mediated stimulation of ERK-2 variety of cell systems including the differentiation of (Figure 3C). Nevertheless, it has been described that Rac- 1 PC 12 cells and the transformation of fibroblasts. The plays a role in AP-1 regulation via control of Jun kinases present has study used constitutively active and inhibitory such as JNK/SAPK. Elk-i has also been described to be mutants of a MAP kinase kinase (MAPKK- 1) to examine a substrate for the JNK/SAPK family (Whitmarsh et al., the role of this pathway in TCR signal transduction. 1995; Zinck et al., 1995), and to explore if Rac-I could Activated mutants of MAPKK-1 can stimulate ERK-2 regulate transcription factor phosphorylation in T cells, kinase activity in T cells and can also induce phosphoryl- 3928 NFAT regulation in T cells _._ NIFATCA I u'n- )ItrolI o Ra': m E-k r C A t t 0~ ~~~~~~ 11 t.3 -1 z_r -; tiI i _ II \NEA'T' C.AT' C: ac R{ \ 1\T UY TlO!10 _ ii)1 (TIOMNCllt 0 L(CII-I- -1 :E I r , - ')II --o Stimll - -C DNA 10 mill cell-) coicenritiratioin (miciogratlm ioni Fig. 5. Comparison of the effects of the activated Rac and Ras mutant on NFAT induction. (A) Cells were co-transfected with the NFAT-CAT reporter construct (12 jg) and 8 jig of the empty vector or the plasmid encoding activated Ras (8 jig of pEF-BOS v-Ha-ras) or activated Rac (8 jig of pEFplinkV12Rac) mutants. Transfected cells were either left unstimulated or were stimulated with ionomycin (0.5 or a combination of gg/ml) Pdbu (50 ng/ml) and ionomycin (0.5 ,ug/ml), 6 h post-transfection and for 16 h before extracts were made, and CAT reporter activity was then assessed. Data show the percentage conversion of chloramphenicol into the acetylated product. Results from one experiment representative of four are presented. (B) Jurkat cells were co-transfected with the reporter construct (12 jg) and the empty vector or different concentrations of NFAT-CAT the the activated Ras plasmid encoding (pEF-BOS v-Ha-ras) or activated Rac (pEFplinkV12Rac) mutants, and the experiment was carried out as in (A). Data are shown as a percentage of the conversion of chloramphenicol into the acetylated product. (C) Plasmids encoding the activated Ras (15 jig of pEXV3 v-Ha-ras), the activated MAPKK-1 (15 jg pEXV3 MAPKK 217E/221E) or the activated Rac (15 jig of pEFplinkVl2Rac) mutants, or a combination of the activated MAPKK-1 (15 jg of pEXV3 MAPKK 217E/221E) and activated Rac-1 (15 jig of pEFplinkVl2Rac) mutants were co-transfected with the NFAT-CAT reporter construct (12 jig). Cells transfected with the empty vector were either left unstimulated or were stimulated with Pdbu (50 ng/ml). Transfected cells were then either left unstimulated or stimulated with ionomycin (0.5 jg/mI) or UCHT-1 (10 jg/ml) for 16 h before extracts were made, and CAT reporter activity was assessed. Data show the percentage conversion of chloramphenicol into the acetylated product. Results from one experiment representative of four are presented. ation of the ternary complex factor Elk-1, demonstrating confirms this idea and thus clearly demonstrates that that MAPKK-1 is an effector molecule in the regulation additional p2lras effector pathways must exist in T cells. of the transcription factor Elk-I in T cells as in fibroblasts Recently, the p2iras-related GTPase Rac-1 has been (Hill and Treisman, 1995). The transcription factor NFAT found to be an important mediator of oncogenic trans- is controlled by the TCR by a mechanism involving a formation by Ras (Qiu et al., 1995). Here we describe a 1 in T cells. interaction of and calcium role for Rac- in transcription factor regulation synergistic p2lras signals. did not an MAPKK-1 mutant indi- A dominant mutant of Rac-1 Experiments using inhibitory negative suppress cate that the activation of the MAPKK- 1/ERK- -2 TCR or p21ras activation of the MAP kinase ERK-2 but 1, cascade a role in TCR activation of NFAT because could TCR and induction of NFAT and plays suppress p2lras with the existence of a TCR stimulation of NFAT was prevented by the inhibitory AP-1. These data are consistent in form of MAPKK-1. activated MAPKK- 1 could Rac-l-controlled signalling pathway operating parallel However, mediate iras induction not substitute for 1 ras and with calcium with the ERK-2 pathway to p2 p2 signals synergize with a to stimulate NFAT. A of the effects of of AP- 1 and NFAT. Importantly, experiments signals comparison Rac- 1 mutant showed that Rac-1 was inhibitory p2iras or MAPKK-1 mutants on thymocyte constitutively active that for activation of the in mice concluded a sufficient signal AP-1, although development transgenic recently be mediated could not substitute for the activated functions in T cells cannot active Rac-1 mutant p2iras solely by et for induction. of activated the ERK-2 cascade p2 Iras mutant NFAT Expression signalling (Alberola-lla al., 1995; of the activated MAPKK- in T cells could stimulate and the Swan et The failure 1 MAPKK-1 ERK-2, al., 1995). in NFAT induction Rac- 1 mutant can stimulate kinase to substitute for activated activated parallel signal- p2iras 3929 E.Genot et al. (j .\AI-1 ( ._ .4_ ~.+-i e x l/ U0A4US --*-ii-1 4--E xA lk -'I i'.. ...- O Ras* Rac" PdbLU Ras' Rac" () PdIl- :-. !: ,- \ 1;: ! I'diq \I. empt.y __ N 17 Rac \1.AV C' \T ~~~~. .1 -- 1/ 1 !, |1 >\~ .I~~~~ .,, ~ ~~~~~ I' , -, j~ 1, *. '':}' , \I1 II with the N Fig. 6. p21 ras regulation of NFAT and AP-I Rac-l function. Jurkat cells were co-transfected Lex-Elk-IC fusion requires (A) pEF of protein expression construct together with 8 of the vector or with the encoding activated Ras jg (12 jug) empty plasmid (8 pjg pEF-BOS v-Ha-ras) the Rac mutant or activated Rac (8 ,ig of mutants and in the absence or of the encoding of pEFplinkVl2Rac) presence plasmid inhibitory (8 ig pcDNAl N17Rac). At 16 h cells transfected with the vector were either left unstimulated or were stimulated with post-transfection, empty Pdbu (50 ng/ml) for 30 min. Total cell extracts were made and run on SDS-PAGE. The data show Western blot of LexA-Elk-IC with analysis LexA its antisera. (B), (C) and (D) Jurkat cells were co-transfected with the MLV N Lex-Elk-lC fusion construct and (6 jig) protein expression reporter construct LexA-OP.tkCAT AP-1-CAT or NFAT-CAT constructs with 15 of the vector (3 (B), (2 jg) (C) (12 jg) (D) reporter together ,ug pg) empty Ras mutant of and in the absence or of 2.5 of the or with the plasmid encoding the activated (15 ,ug pEF-BOS v-Ha-ras) presence plasmid Pjg At 6 h cells transfected with the vector or with encoding the inhibitory Rac mutant (pcDNAl N17Rac N17Rac). post-transfection, empty pcDNA1 for 16 h. were carried out as alone were either left unstimulated or were stimulated with Pdbu CAT before. (SOng/ml) assays ling pathways that could and particularly in cell adhesion and phosphorylate transcrip- regulating morphology activate Elk-I in T cells. the (Rodriguez-Viciana et al., Nobes and tionally Nevertheless, 1994; Hall, 1995). combination of activated MAPKK-1 and Rac-I could not Also, Ptdlns 3 kinase can bind directly to activated p21ras, substitute for activated p2lras and synergize with calcium and function as an effector molecule for p2 lras in neuronal to NFAT. signals induce There be at least cells (Rodriguez-Viciana et al., 1994). There is, must, therefore, therefore, one more p21ras effector required for NFAT induction the possibility that Ptdlns 3 kinase could be the link There are several additional (Figure 7). candidate p2lras between Ras and Rac-1. Preliminary experiments show effectors the GTPase Rho et that Ptdlns 3 kinase inhibitors including (Prendergast al., (wortmannin and LY294002) 1995), and also the GTPase Ral et fail to show effect (Hofer al., 1994; any inhibitory on TCR induction of and whose function in NFAT K.Reif and Spaargaren Bischoff, 1994), NFAT (E.Genot, D.A.Cantrell, unpublished regulation remain to be investigated. p21ras effects on the data). The failure of Ptdlns 3 kinase inhibitors to regulate MAPKK-1/ERK-2 cascade in T cells are mediated the NFAT is not with a role for Ptdlns by compatible 3 kinase serine kinase Raf-l (Izquierdo et al., 1994a). It is not as a mediator of p21ras effects on NFAT. Accordingly, clear whether or not there are multiple Raf-1 effector the signalling links between p2lras, Ptdlns 3 kinase and in T cells in Rac- 1 in T pathways but, fibroblasts, Raf-1 and Rac-1 cells may be different from those described in signals synergize to mediate cellular other cell transformation by lineages. However, this question has to be p2lras (Qiu et al., 1995). We currently are exploring investigated further with different and complementary functional interactions between Raf-1 and Rac-1 in NFAT approaches. Concerning the Rac-1 effector pathway for induction. In this context, Jurkat cells expressing constitu- NFAT induction, Rac- 1-regulated protein kinases have tively active Raf-1 show enhanced levels of IL-2 produc- now been identified (Manser et al., 1994; Martin et al., tion but there has not been a direct analysis of the role of 1995), but the role of these protein kinases in Rac-l Raf- in NFAT regulation (Owaki et effector al., 1993). pathways in T cells remains to be discovered. Questioning how p2 lras would regulate Rac- 1, a couple The transcriptional activity of the NFAT binding site of recent reports raise some possibilities. Thus, a report used in the present experiments is mediated by a complex has described phosphatidyl inositol 3' kinase (Ptdlns 3 a comprising member of the NFAT group of DNA binding kinase) as an upstream regulator of Rac-1 in fibroblasts, proteins and AP-l family proteins (Rao, 1994). The AP-l 3930 NFAT regulation in T cells TCR of Fos and Jun proteins which up- the phosphorylation activity. Several parallel MAP regulates their transcription ERK- 1, -2, JNK- 1, kinase signalling pathways involving Fos kinase can all -2 and MPK2/p38 and a Ras-regulated potentially contribute to AP- 1 activation (Deng and Karin, Coso et al., 1995; Hill 1994; Cobb and Goldsmith, 1995; Elk-I which contributes and Treisman, 1995). The TCF via complex formation with to increased c-fos expression factor SRF is a substrate for the ubiquitous transcription -2 (Marais et al., 1993; Whitmarsh the ERKs and JNK-1, \l1APKK-l I is also transcriptionally regulated by et al., 1995). c-jun ,/ i-> Rac- Cilc]iIL'Llr MAP is a direct substrate for JNK- 1 and kinase pathways, JNK- 1-mediated phosphorylation can be activated by R\\K-? ) \Ft\ Recent studies suggest that Rac- 1 (Derijard et al., 1994). \ \ IFt5>. ItsIll role in the activation of the MAP kinases plays a central AI /\ 1;Cn that is exactly analogous to the role of JNK and p38 in ERK pathways (Coso et al., 1995; Minden p2lras the et The present data show that Rac-l signals are al., 1995). APFo-;I II\II necessary and sufficient to induce Elk-I phosphorylation i4 in T cells, which indicates that Rac-1 signals regulate . ;.-1\F Xl- transcriptional factor kinase pathways in T cells. Future studies will establish whether Rac- 1 effects on Elk- 1 Cvtokine genes phosphorylation in T cells are mediated by JNK-1/SAPK, with or a novel protein kinase pathway. By analogy p38 other models for transcription factor regulation by p21 ras- Fig. 7. Schematic representation of signalling pathways predicted to induce NFAT. Our model for NFAT induction shows calcium! related GTPases, then Rac-1 effects on NFAT could be calcineurin signals regulating NFAT translocation from the cytosol to kinase pathways. Clearly mediated by one of the MAP the nucleus. p2lras /Rac-1 regulation of NFAT involves regulation of 1 in T Rac- does not regulate ERK-2 cells, but it is possible AP-1. MAPKK-1 signalling pathways are also required for NFAT 1 NFAT involves that the role of Rac- in TCR induction of induction and, since MAPKK-1 does not regulate AP-1, we would MAP kinase. It is possible, predict that MAPKK-1 pathways regulate NFAT itself or the NFAT- JNK- 1/SAPK or the p38 MAPKK- AP-1 complex. To explain why the combination of activated in TCR induction of moreover, that the role of Rac-1 1 for activated p21ras for NFAT and activated Rac-1 do not substitute NFAT will be analogous to the role of Rac- 1 in serum have to evoke the existence of at least one more p21ras induction, we activity in fibroblasts, regulation of SRF transcriptional contribution of PKC to NFAT regulation is signalling pathway. The Rho/Rac- 1-controlled signal- because the TCR activates PKC but this is not required which occurs by an unknown complicated for NFAT induction. Phorbol esters can regulate NFAT, but this occurs that does not involve ERK-2, JNK- 1 or p138 ling pathway an pathway that converges on by independent PKC-mediated et (Hill al., 1995). MAPKK-1 and Rac-1. Rac-1 mediate p2lras regulation of AP-1 and may NFAT in T cells but, if the signalling connections between is composed of dimers of members of the Fos and Rac-l in T cells are as complicated as those complex p2lras and Jun family of proteins. The exact nature of the AP-1 in fibroblasts, then it is unlikely that p2iras and Rac-1 it component of NFAT in T cells has not yet been established will function solely in a linear signalling cascade and will in both clearly. Boise et al. have described that a Fra-1-JunB is more likely that Rac-1 and p2lras function Rac- 1 has diverse roles heterodimer is the inducible nuclear component of the and convergent pathways. parallel For in NFAT binding activity, while Northrop et al. have reported in many cellular systems. example, neutrophils, 1 of the NADPH oxidase a dimer of c-Fos (or Fra- 1) and either c-Jun or Rac- is involved in the activation composed In Rac-1 plays a that the nuclear subcomponent may be complex (Xu et al., 1994). fibroblasts, JunD, suggesting of AP- 1-NFAT in of the actin cytoskeleton represented by a heterogeneous population critical role the regulation It and but can also regulate transcriptional (Boise et al., 1993; Northrop et al., 1993). (Nobes Hall, 1995) complexes AP-1 used in the of SRF and controls c-jun transcriptional activity is not clear whether the reporter gene activity measures the of the AP- 1 activation of the c-Jun kinases JNK-l/SAPK (Coso present experiments activity via Minden et In NFAT of distinct AP- 1 members. et Hill et 1995; al., 1995). components of or family al., 1995; al., 1 in T cells is an AP- 1 data are this the activation of JNK- Nevertheless, the present reporter gene context, the in T Rac- 1 function for from co-stimulatory recep- valuable because they show that, cells, integration point signals and can be sufficient for AP- 1 induction tor CD28- and TCR-mediated calcium/calcineurin signals is both necessary It will thus be of interest to determine and that Rac- 1 can mediate Ras effects on AP- 1. Moreover, et 1994). (Su al., whether Rac-1 is an for the activated Rac- 1 mutants were equivalent to the activa- in future studies integration point activation. of AP- 1 induction. This and CD28 for JNK-1/SAPK ted Ras mutants in the context TCR signals has examined 1 1 could thus In the present study p21ras role of Rac- in AP- regulation explain summary, T cell line Jurkat in an to 1 for NFAT induction. to effector in a attempt Rac- Accordingly, pathways requirement contributed for activa- Rac- effector in T characterize the by p21ras consider the 1 pathways cells, signals possible effector involves to focus on the mechanisms for AP- 1 tion of NFAT. One p2iras pathway it is necessary which is shown the data to be both In the activation of AP- 1 results from MAPKK- by present induction. general, 1, sufficient for activation of ERK-2 and the of Fos and Jun family proteins and necessary and increased expression 3931 E.Genot et al. Western blot analysis and MBP kinase assays transcription factor Elk- I in Jurkat cells. Rac- 1 is a Cells cultured in complete medium were harvested at different times sufficient and necessary signal for AP-1 activation and after transfection and washed three times in RPMI alone. SDS was also appears to be involved in p21ras effector pathways added to 1% final concentration. After vortexing, extracts were heated in Jurkat, since p2iras-mediated induction of AP-1 and for 5 min at 100°C. Debris were pelleted by ultracentrifugation for 30 min at 90 000 r.p.m. in the TL100.2 rotor of a Beckman centrifuge. NFAT requires Rac- 1 function. MAPKK- 1- and Rac- 1- Proteins were acetone precipitated from supernatants for 10 min on ice mediated signals may be sufficient for either ERK-2 or and resuspended in reducing Laemmli sample buffer. Disruption of the AP-1 activation but, either alone or in combination, they pellet was completed by sonication. Extracts prepared from equivalent cannot substitute fully for p21ras signals for induction of numbers of living cells were subjected to SDS-PAGE on 10% gels (10% acrylamide-0.27% bisacrylamide), unless otherwise indicated. Proteins NFAT. There must, therefore, be at least one more p21ras were transferred to polyvinyldifluoridine membranes (Immobilon, pathway that is required for NFAT induction. The effector Millipore) by overnight Western blot. Membranes were 'blocked' in T in the present study has proved to Jurkat cell line used phosphate-buffered saline (PBS)-0.05% Tween 20, (PBST) plus 5% system for studies of NFAT regulation. be a valuable model milk for 2 h. One tg/ml of 9E10 antibody or a 1/1000 dilution of LexA likely that there will be multiple Ras antisera was allowed to react with the membrane overnight at 4°C. After It would thus seem extensive washes, membranes were incubated with a dilution of anti- populations that contribute to effectors in normal T cell mouse (1/4000) or anti-rabbit IgG (1/10 000) in PBST for 1 h at room gene induction. This NFAT regulation and hence cytokine temperature. After the washes in PBST, and PBS, membrane-bound of fibroblasts where is analogous to the transformation antibodies were visualized by use of ECL Western blotting detection reagents (Amersham) and Kodak XS 1 films. ERK-2 tag immunoprecipit- Ras effects are mediated by multiple effectors. The Rac- 1 ations with 9E10 antibody and MBP assays were carried out as described effector pathway for NFAT induction is not yet character- previously (Izquierdo et 1993). al., ized. Nevertheless, the NFAT family of transcription factors are important in the regulation of cytokine genes CAT assays and hence important for a successful immune response. CAT assays were performed using the liquid scintillation method of The present data predict that Rac-1 and Rac-1 signalling Sleigh et al. using modifications (Sleigh, 1986). Briefly, 5X 106 cells would be important targets for immune inter- were lysed in 50 ,ul of 0.5% NP-40, 10 mM Tris pH 8, 1 mM EDTA pathways 150 mM NaCI for 10 min on ice. Cellular debris was pelleted and vention. lysates were heated at 68°C for 10 min before use. Assay conditions were 150 mM Tris pH 8, 0.05 tCi [14C]acetyl coenzyme A and 2 mM chloramphenicol. After 18 h at 37C, chloramphenicol was extracted with ethylacetate, and the amount of radioactivity in the acetylated Materials and methods products and non-acetylated substrate of each reaction was determined Reagents by liquid scintillation counting of organic and aqueous phases respect- ively. The amount of chloramphenicol acetylation is calculated as a dibutyrate were from Calbio- lonomycin (Ca2+ salt) and phorbol-12,13 percentage. coenzyme A (at 50-60 mCi/mmol) was chem Corp. (UK). [14C]Acetyl International (Buckinghamshire, UK). Other reagents from Amersham Monoclonal antibodies UCHT-1 and 9E10 reactive were from Sigma. and the c-Myc epitope used a tag for ERK-2 were with human CD3 Acknowledgements from supernatants at I.C.R.F. Rabbit antisera affinity purified hybridoma gift of Richard Treisman and Alberto Gandarillas specific for LexA were a We thank S.Cowley and C.Marshall for the MAPKK-1 constructs, and goat anti-rabbit antibodies are (ICRF). Polyclonal rabbit anti-mouse N.Clipstone and G.Crabtree for the J-TAg cell line, R.Treisman and (Buckinghamshire, UK). for LexA antibodies, M.A.Price for LexA-OP.tkCAT from Amersham International A.Gandarillas and LexA-Elk-IC constructs and for advice, J.Babbage for technical assistance, J.Lunan for help with the manuscript and all members of Plasmid constructs the Lymphocyte Activation Laboratory for helpful comments on the The reporter construct AP- 1-CAT, containing three copies of the sequence manuscript. E.G. is a member of the INSERM organization. S.H. was for the AP-1 site inserted in pBLCAT2. has been described previously supported by a grant from the Deutsche Forschungsgemeinschaft. This et al., 1995). LexA-OP.tkCAT contains two copies of the (Williams work was funded by the Imperial Cancer Research Fund. contains three copies of the NFAT binding LexA reporter, NFAT-CAT upstream of site TAAGGAGGAAAAACTGTTTCATACAGAAGGCG IL-2 driving the reporter gene CAT (Verweij et al., the minimal promoter directing expression of wild-type MAPKK- 1: pEXV3 References 1990). Plasmids MAPKK-1: pEXV3 MAPKK 217E/ MAPKKwt, constitutively active Alberola-lla,J., Forbush,K.A., Krebs,E.G. and Perlmutter,R.M. Seger,R., MAPKK-1: pEXV3 MAPKK 221A and 221E, the dominant negative (1995) 373, 620-623. Natutre, v-Ha-ras or v-Ha-ras, constitutively active p2lras: pEF-BOS pEXV3 and (1992) J. Biol. 267, Baldari,C.T., Macchia,G. Telford,J.L. Chemii.. fusion protein or pEF pEF BOS ERK-2 tag, MLV N-LexA-Elk-IC 4289-4291. N-LexA-Elk-IC fusion protein, have already been described (Medema Boise,L.H., Petryniak,B., Mao,X., June,C.H., Wang,C., Lindsten,C.B., et al., 1991; Cowley et al., 1994; Leevers et al., 1994; Price et al., Bravo,R., Kovary,K., Leiden,J.M. and Thompson,C.B. (1993) Mol. mutants of 1995a). The active and dominant negative Rac-I (Vl2Rac-I Cell. Biol., 113, 1911-1920. and N17Rac-1) have also been described previously (Nobes and Hall, Cobb,M. and Goldsmith,E. (1995) J. Biol. Cheml., 270, 14843-14846. 1995). All plasmids were purified by equilibrium centrifugation in CsCl- Coso,O., Chiariello,M., Yu,J.-C., Teramoto,H., Crespo,P., Xu,N., MIki,T. procedures. ethidium bromide gradients by standard and Gutkind,J. (1995) Cell, 81, 1137-1146. Cowley,S., Paterson,H., Kemp,P. and Marshall,C. (1994) Cell, 77, Cells and transfections 841-852. J-TAg, a subline of the human T acute lymphocytic lymphoma cell line Deng,T. and Karin,M. (1994) 371, 171-175. Natutre, Jurkat, was maintained in RPMI supplemented with 10% heat-inactivated Derijard,B., Hibi,M., Wu,I.-H., Barrett,T., Deng,T., Karin,M. and Su,B., fetal calf serum (FCS), at 37'C in 5% CO, in humidified air. Cells were Davis,R.J. (1994) Cell, 76, 1025-1037. transfected via electroporation (Gene pulser: Bio-Rad, UK) as previously Flanagan,W.M., Corthesy,B., Bram,R.J. and Crabtree,G.R. (1991) Nattlre, ( X described. cells were cells/0.5 ml) in complete Briefly, pulsed 107 352, 803-807. medium at 960 and 320 V. Cells transfected with similar plasmid Genot,E.M., Parker,P.J. and Cantrell,D.A. (1995) J. Biol. Chem?., 270, pF mixtures were pooled and re-aliquoted before stimulation. Transfected 9833-9839. cultures were incubated for 6 h, and stimuli, when required, were applied Hall,A. (1994) Annul. Repi Cell. Biol., 10, 31-54. for an additional 16 h. The amounts of DNA in transfected each Hill,C. and Treisman,R. (1995) Cell, 80, 199-211. were constant adding and Treisman,R. (1995) Cell, 81, 1159-1170. by control plasmids as indicated. Hill,C., Wynne,J. experiment kept 3932 NFAT regulation in T cells Hoey.T.. Sun,Y.-L.. Williamson,K. and Xu,X. (1995) Immunitov, 2. 461-472. HoferEF., Fields.S.. Schneider.C. and (1994) Proc. NatlAcad. Martin,G.S. Sci. USA 91. 11089-11093. Howe.L.R.. Leevers,S.J.. Gomez.N.. Nakielney.S.. Cohen.P. and Marshall,C.J. (1992) Cell, 71, 335-342. Izquierdo.M., Leevers.S.J.. Marshall,C. and CantrelliD.A. (1993) J. EI.). Med.. 178, 1199-1208. Izquierdo,M., Bowden,S. and Cantrell.D. (1994a) J. E.rp Med.. 180. 401-406. Izquierdo,M.. Leevers.S.J.. Williams.D.H.. Marshall.C.J.. Weiss.A. and (1994b) Elu: J. 24. 2462-2468. Cantrell,D.A. Immunaiiiiiol., Izquierdo-Pastor.M.. Reif,K. and CantrellD. (1995) Iinunaiiiiiol. Todayi. 16. 159-164. Paterson,H. and Marshall.C. (1994) 369. 411-414. Leevers,S., Nature, Manser,E., Leung.T.. and (1994) Nature. Salihuddin.H., Zhao,Z.S. Lin,L. 367. 40-46. Marais.R., Wynne.J. and Treisman.R. (1993) Cell. 73. 381-393. Marshall.C.J. (1994) Curr: Opin. Geniet. 4, 82-89. Dev, Martin.G.A.. BollagcA., McCormickEF. and (1995) EMBO J.. 14. Abo,A. 1970-1978. Medema.R.H.. Wubbolts,R. and Bos.J.L. (1991) Mol. Cell. Biol.. 11. 5963-5967. Minden,A.. Lin.A.. Claret,F.-X., Abo.A. and Karin.M. (1995) Cell. 81. 1147-1157. Nobes.C. and HallA. (1995) Cell. 81. 53-62. Northrop,J.P., Ullman,K.S. and Crabtree.G.R. (1993) J. Biol. Clhemli.. 268, 22917-2925. Ho.S.. Northrop,J.. Chen.L.. Thomas.D.. Timmerman.L.. Nolan.G.. Admon.A. and Crabtree.G. (1994) Ncature. 369, 497-502. Owaki,H., Varma,R., Gillis.B.. Bruder.J.T., Davis,L.S. and Rapp,U.R.. Geppert.T.D. 12. (1993) EMBO J., 4367-4373. Prendergast.G.C., Khosravi-Far,R.. Solski,P.A.. Kurzawa.H.. Lebowitz. P.F. and Der,C. (1995) Oncogene. 2289-2296. 10, PriceL.. Nornnan.J.. Ridley.A. and (1995a) Cutr,: Biol.. 5. Koffer,A. 68-73. Price.M.. Rogers.A. and Treisman.R. (I 995b) EMBO J.. 14. 2589-2601. Qiu.R.G.. Chen.J.. McCormick.F. and Symons.M. (1995) Nature. Kirn,D.. 374 457-459. Rao.A. (1994) Iuuululuol. Todav. 15. 274-281. Rayter.S.. Lucas.S.C.. CantreliD. and Downward.J. (I1992) Woodrow.M., EMBO 4549-4556. J., 11, Rodriguez-Viciana.P.. Vanheaesbroeck.B.. Warne,P.H., Dhand,R., Gout,I., Fry.M.J.. Waterfield,M.D. and DownwardJ. Nature. (1994) 370. 527-532. Schreiber.S.L. and Crabtree.G.R. (1992) Iuna711{Z7ol. Todav. 13. 136-142. Sleigh.M.J. (1986) Anial. 156. 251-256. Bioelmeti., and Spaargaren.M. Bischoff.J.R. (1994) Proc. Nalt Acad. Sci. USA. 91. 12609-12613. Su.B.. Estela.J.. Hibi,M., Kallunki,T., Karin,M. and Ben-NeriahY. (1994) Cell. 77. 727-736. Swan.K.A.. Gross,J.A.. Albeerola-lla.J.. Appleby.M.W.. Forbush,K.A.. and Thomas,J.F. Perlmutter.R.M. (1995) EMBO J., 14. 276-285. and Verweij,C.L., Guidos,C. Crabtree.G.R. (1990) J. Biol. Chel., 265. 15788-15795. Warne.P.. and Naitutre. 364. Rodriguez-Viciana.P. Downward.J. (1993) 352-355. Littinan.D.R. 76. WeissA. and (1994) Cell. 263-274. and Davis.R. 269. Whitmarsh.A.. Shore.P.. Sharrocks,A. (1995) Science, 403-407. Cantrell.A.D. and Eli: Williams,H.D., Woodrow.M., Murray,J.E. (1995) Inunuiiinol.. 25. 42-47. J. Clipstone.N.A. and Cantrell.D. J. Med., 178. Woodrow,M., (1993a) E.rp. 15 17-1522. Woodrow.M.. Rayter,S., Downward.J. and CantreIlD.A. (1993b) Ininntol.. 1-9. J. 150. Biol. Xu,X., Barry.D.. Settleman.J.. Schwartz.M. and Bokoch.G. J. (I1994) Clienti.. 23569-23574. and Zinck,R., Cahill.M.. Kracht.M.. Sachsenmaier.C.. Hipskind,R. Nordheim.A. Mol. Cell. Biol.. 4930-4938. (1995) 15, Received December 13. revisedl oni 23, 1996 onz 1995; April http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The EMBO Journal Springer Journals

Multiple p21ras effector pathways regulate nuclear factor of activated T cells.

Genot, E.; Cleverley, S.; Henning, S.; Cantrell, D.
The EMBO Journal , Volume 15 (15) – Aug 1, 1996
Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/multiple-p21ras-effector-pathways-regulate-nuclear-factor-of-activated-FSLoAzntdF

References (0)

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Springer Journals
Copyright
Copyright © European Molecular Biology Organization 1996
ISSN
0261-4189
eISSN
1460-2075
DOI
10.1002/j.1460-2075.1996.tb00766.x
Publisher site
See Article on Publisher Site

Abstract

The EMBO Journal vol. 15 no. 15 pp.3923-3933, 1996 Multiple p2lras effector pathways regulate nuclear factor of activated T cells the IL-2 gene (Baldari et al., 1992; Rayter et al., 1992). Elisabeth Genot', Steve Cleverley, The major target in the IL-2 gene for p21ras signalling Stefan Henning and Doreen Cantrell is NFAT pathways (Woodrow et al., 1993b). Thus, a Lymphocyte Activation Imperial Cancer Research Laboratory, Fund, constitutively active Ras mutant (p21 -v-Ha-ras), when 44 Lincoln's Inn Fields. London WC2A 3PX. UK in T expressed cells, can synergize with a calcium/ 'Corresponding author calcineurin signalling pathway to activate NFAT (Woodrow et al., 1993a,b). p2lras regulatory effects on The transcription factor, Nuclear Factor of Activated NFAT can be mimicked by phorbol esters that activate T cells (NFAT) is a major target for p2lras and calcium PKC or by expression of constitutively activated mutants signalling pathways in the IL-2 gene and is induced of PKC (Genot et al., 1995). The ability of phorbol esters by p2lras signals acting in synergy with calcium/ to activate p21ras in T cells led to the proposal that PKC calcineurin signals. One p2lras effector pathway effects on NFAT were mediated by p21 ras. However, it is involves the MAP kinase ERK-2, and we have examined now recognized that phorbol ester and p21ras induction its role in NFAT regulation. Expression of dominant of NFAT occur by different, independent mechanisms. In negative MAPKK-1 prevents NFAT induction. Con- particular, p2l.ras but not PKC signals are essential for stitutively active MAPKK-1 fully activates ERK-2 and TCR induction of NFAT (Williams et al., 1995). NFAT the transcription factor Elk-1, but does not substitute proteins are cytosolic in quiescent T cells but they trans- for activated p2lras and synergize with calcium/ locate to the nucleus in TCR-activated cells and combine calcineurin signals to induce NFAT. Expression of with newly induced AP-1 complexes to form a functional dominant negative N17Rac also prevents TCR and transcriptional factor unit (Flanagan et al., 1991). There p2lras activation of NFAT, but without interfering are multiple isoforms of the NFAT cytosolic subunit, with the ERK-2 pathway. The transcriptional activity NFATc, NFATp, NFAT3 and NFAT4, that have different of the NFAT binding site is mediated by a complex patterns of cellular expression but are thought to share comprising member of the NFAT group and AP-1 one property in that their nuclear translocation is controlled family proteins. The induction of AP-1 by p2lras also by calcium/calcineurin-dependent signals (Northrop et al., requires Rac-1 function. Activated Rac-1 could mimic 1994; Rao, 1994; Hoey et al., 1995). The different NFAT activated p2lras to AP-1 induce but not to induce proteins also share the ability to interact with AP-1 to NFAT. Moreover, the combination of activated bind co-operatively to the NFAT/AP- 1 site in composite MAPKK-1 and Rac-1 could not substitute for activated the IL-2 gene (Rao, 1994). The interaction with AP-1 is p21ras and synergize with calcium signals to induce essential for NFAT transcriptional activity, and the role of NFAT. Thus, p2lras regulation of NFAT in T cells p21 ras in NFAT induction has been the explained by requires the activity of multiple effector pathways ability of p21ras signals to stimulate AP-1 in T activity including those regulated by MAPKK-1/ERK-2 and cells (Rayter et al., 1992; Woodrow et al., How- 1993a). Rac-1. the ras effector involved in ever, p21 pathways NFAT NFAT/Rac-l/Ras/MAPKK-1/transduction/ Keiw!ords: regulation have not been defined and are the focus of the TCR present report. T at least two MAP lymphocytes express kinases, ERK- 1 and that are stimulated a ras ERK-2, by p21 in to TCR et pathway response triggering (Izquierdo al., Introduction The MAP kinases are activated a kinase cascade 1993). by involving a MAP kinase kinase that The T cell controls activation (MAPKK) phos- antigen receptor (TCR) and and stimulates the ERK- 1 and -2 kinases the GO-G1 transition of T and the phorylates lymphocytes regulates Cobb and The production of a number of the directly (Marshall, 1994; Goldsmith, 1995). cytokines, including growth of the MAPKK is itself controlled factor interleukin-2 The of IL-2 activity by phosphoryl- (IL-2). regulation produc- ation and, hence, a MAP kinase kinase kinase tion by the TCR requires the co-ordinate action of (MAPKKK) multiple effector ERK-1 and -2. Raf-1 is a that transcription factors that include Nuclear Factor of Activ- regulates p2lras T to be the MAPKKK that lras and ated cells NF-KB and Oct-1 appears couples p2 (NFAT), AP-1, (Schreiber et and The intracellular hence receptors to the MAP kinases Crabtree, 1992; Rao, 1994). signals (Howe al., 1992; et Leevers et factors involve Warne et that regulate these different transcription al., 1993; Izquierdo al., 1994a; al., and -2 are kinase-induced 1994). In neuronal MAPKK-1 tyrosine calcium/calcineurin, protein cells, ERK-1, ras functions et kinase C and critical mediators of al., (PKC) p2lras-mediated signalling pathways p21 (Cowley substrate for the ERKs is the and et 1994). One regulatory (Weiss Littman, 1994; Izquierdo-Pastor al., 1995). protein of a of that can form Inhibition of function in T cells of which is one p21ras by expression Elk-1, family proteins with the activator serum a dominant mutant induction of a negative p21ras prevents ternary complex transcriptional C Oxford Press University E.Genot et al. response factor (SRF) (Marais et al., 1993). Elk-l-SRF and which correlates with -2, the inhibitory effects of complexes are necessary and sufficient mediators of N17Ras on TCR activation of NFAT (Izquierdo et al., c-fos serum response element induction. Thus, p21ras regulation Woodrow et The MAP kinases 1993; al., 1993b). ERK-l of the MAPKK/ERK pathway plays a key role in regulating and -2 are activated a kinase by (MAPKK) that phosphoryl- gene expression (Hill and Treisman, 1995). The ates and stimulates the ERK- 1 and -2 kinases c-fos directly. analysis of the role of the ERKs in the control of gene MAPKK is activated by protein phosphorylation, and expression in fibroblasts has created a paradigm that the mutation of phosphorylation sites at position 217 and/or MAPKKIERK pathway is a major route 221 on whereby p21ras MAPKK- 1 from serine to alanine generates an regulates the transcriptional activity of immediate early inhibitory molecule that can suppress growth factor- genes such as By analogy, there is c-fos. the possibility mediated activation of the ERK kinase cascade (Cowley that the p2lras/MAPKK pathway is the effector et p2lras al., 1994). Accordingly, in initial experiments to explore pathway for NFAT induction. whether regulation of ERKs explains p21ras control of There has been one previous study on the role of NFAT, the effects of the inhibitory mutant of MAPKK MAPKK and the ERKs as p2 Iras effectors in T cells: this 221A on TCR (DN MAPKK) regulation of NFAT were was a study in transgenic mice that showed that analysed. To assess expression the activity of the transcription factor of a dominant negative MAPKK- 1 mutant could mimic NFAT, we employed transient transfection protocols and inhibitory p21 ras mutants and suppress positive selection quantitation of the of a expression CAT reporter gene of thymocytes (Alberola-Ila et al., 1995). However, domin- whose activity is regulated by an enhancer corresponding ant negative Ras mutants can prevent the in vitro to the NFAT binding site. The data in prolifera- Figure 1A show tion of thymocytes, whereas a dominant negative that this NFAT reporter gene is not expressed in unstimu- MAPKK- 1 mutant does not, indicating that not all Ras lated Jurkat cells but is induced in cells stimulated with functions in T cells MAP the UCHT- I are mediated by the kinases antibody that triggers the TCR-CD3 complex. cascade. The Ras-related Rac- 1 has been described GTPase TCR induction of NFAT can be mimicked by the combina- tion of recently as another downstream effector of p21ras (Hall, stimuli that activate PKC and elevate intracellular 1994; Qiu et the of the calcium levels such al., 1995) Accordingly, object as phorbol esters and calciutm iono- present study was to the connections phores. PKC and induce NFAT explore signalling p21ras independently and between the cascade and in similarly independently activate ERK-2 p21ras, MAPKK/ERK Rac-1 (Izquierdo et al., T cells in order to determine the role of these in 1993; Williams et al., 1995). There is proteins thus the possibility p2 lras regulation of NFAT. The data show that that PKC and Ras signals for NFAT induction expression converge of a dominant negative MAPKK-1 mutant can inhibit at the level of the MAP kinases. The data in Figure 1A NFAT induction, but expression of a active show that the phorbol ester constitutively phorbol 12,13 dibutyrate MAPKK- 1 mutant, which could activate ERK-2 in T (Pdbu) has no stimulatory effect on NFAT when used cells, as could not substitute for constitutively active and a single and the calcium ionophore p21ras stimulus, ionomycin with synergize calcium-dependent signals to induce NFAT. has only a weak inductive effect on NFAT. the However, These studies that the ERK of collectively suggest pathway combination calcium ionophore and Pdbu synergizes to is a but not sufficient maximal NFAT necessary p21ras effector pathway give activity. Expression of the inhibitory for NFAT induction in T cells. of a dominant MAPKK- 1 mutant in T cells Expression inhibits TCR induction of Rac- 1 mutant had a effect NFAT and also Pdbu and negative pronounced inhibitory suppresses ionomycin induction on Ras induction of NFAT, which that Rac-1 is of NFAT In four suggests (Figure IA). experiments, the inhibitory an effector for p21ras in NFAT regulation. In effect of the dominant MAPKK mutant fibroblasts, negative on the Rac-l-mediated intracellular signals participate in activa- TCR/NFAT response ranged from 77 to 50%, whereas the tion of Jun kinases and transactivation of c-jun in a inhibitory effects of this mutant on Pdbu plus ionomycin mechanism analogous to p21ras regulation of ERK-2- responses ranged from 30 to 40%. Elk-1 et The (Minden al., 1995). present data show that To monitor ERK activation in cells transfected with expression of active Rac-1 the constitutively can stimulate dominant negative MAPKK- I the mutant, inhibitory AP- 1 transcriptional activity in T mutant was cells, demonstrating that co-expressed in T cells with ERK-2tag, Rac-1 can also stimulate transcription factor in an pathways epitope-tagged ERK-2 construct. Activated ERK-2 is T 1 cells. However, constitutively active Rac- mutants phosphorylated as we have and, shown previously, activ- could not synergize with calcium signals to induce ated isolated from NFAT, ERK-2tag Pdbu- or TCR-activated T indicating that Rac-1 is necessary but not sufficient for cells has reduced electrophoretic mobility in SDS-PAGE all p21 ras-mediated signalling pathways in T cells. Accord- gels as compared with ERK-2tag isolated from quiescent ingly, p21ras effects on NFAT are not mediated by a cells (Izquierdo et al., 1994a,b) (Figure 1B). Accordingly, single linear signalling pathway but must involve the activation multiple of ERK-2tag can be monitored sensitively effectors that include Rac-1 and MAPKK-1. of by analysis its electrophoretic mobility. The data in Figure lB show that co-transfecting the dominant negative MAPKK- 1 mutant with ERK-2tag prevents ERK-2tag Results phosphorylation as induced by either TCR or phorbol ester. The effects of inhibitory and active MAPKK- 1 In further experiments, we examined the consequences mutants on NFAT induction of expressing a constitutively activated 1 MAPKK- mutant Activation of p2lras in T cells can be prevented by on NFAT induction. Mutation of residues 217 and 221 on expression of an inhibitory mutant of p2 Iras, N 17Ras MAPKK- I from serine to glutamic acid generates a (Izquierdo et al., 1993). Expression of N17Ras in T cells constitutively active enzyme, MAPKK-1 217E/221E also suppresses TCR activation of the 1 MAP kinases ERK- The (MAPKK*). expression of this constitutively active 3924 NFAT regulation in T cells A* _\ \ .I( \I * IN\ \1 \1Sk\ C- NI Al CAT _* I HTI _~~~~~~~~~~~~~~~~~~~~ ..... ._ , *. *ER K- 2tag ."p_inim' N. U ERK-2taga i I 1..1^ _ / 0 KJL - 0 Pdbu UCHT-1 _erairitv_. __ MAPKK N1^s DN Fig. 1. The effects of an inhibitory MAPKK mutant on NFAT induction. (A) 107 Jurkat cells were co-transfected by electroporation with the NFAT- CAT reporter construct (15 and the empty vector or the plasmid encoding the dominant negative MAPKK- I mutant (20 ,ug of pEXV3 MAPKK pg) 221A, DN MAPKK) as described in Materials and methods. At 6 h after transfection, transfected cells were either left unstimulated, or stimulated with UCHT-1 (10 .tg/ml), Pdbu (50 ng/ml), ionomycin (0.5 ig/ml) or a combination of Pdbu and ionomycin for 16 h before extracts were made, and CAT reporter activity was assessed. Data show the percentage conversion of chloramphenicol into the acetylated product. Results from one experiment representative of four are presented. (B) Jurkat cells were co-transfected with the pEF-BOS ERK-2tag construct (18 big) together with the empty vector or with the inhibitory MAPKK-1 mutant (20 p,g of pEXV3 MAPKK 221A). At 16 h after cells were left unstimulated transfection, or were stimulated with UCHT-1 (10 jtg/ml) or Pdbu (50 ng/ml) for 10 min. Total cellular extracts were made and run on SDS-PAGE (15% acrylamide-0.075% bisacrylamide). Data show Western blot analysis of ERK-2tag with the 9E10 antibody. (C) Jurkat cells were co-transfected with the NFAT-CAT (15 jg) reporter construct and the empty vector or the constitutively active Ras mutant (15 of pEXV3 v-Ha-ras. Ras*) or pIg constitutively active MAPKK-1 mutant (15 of ,ug pEXV3 MAPKK 217E/221E, MAPKK*). After 6 h. cells transfected with the empty vector were either left unstimulated or were stimulated with Pdbu (50 ng/ml). All sets of transfected cells were then either left unstimulated or were stimulated with ionomycin (0.5 or UCHT- I (10 jg/ml). CAT assay was performed as in (A). Data are shown as the percentage conversion of pg/ml) chloramphenicol into the acetylated product. Results from one experiment representative of five are presented. MAPKK- 1 will activate the MAP kinases ERK- 1 -2 and protein (MBP) as a substrate were carried out on ERK-2tag and has been shown to substitute for activated p21 ras in isolated from MAPKK- 1 immunoprecipitates activated many cellular responses in fibroblasts and neuronal cells or control transfected cells. We have used this system (Cowley et al., 1994). The data in Figure IC are represent- to previously demonstrate that expression of activated ative of five similar experiments and show that NFAT p2lras in T cells is sufficient to stimulate ERK-2 kinase responses were only marginally induced by the activated activity (Izquierdo et TCR induction of ERK-2 al., 1993). MAPKK-1 mutant in combination with ionomycin. No is transient and at 10-15 min In peaks post-stimulation. further induction of the NFAT-CAT construct the experiment presented in it is this reporter Figure 2B, peak could be seen by increasing or decreasing the amount of TCR response that is whereas in the transient shown, MAPKK-1 activated DNA transfected (data not transfection experiments the sustained effects of either shown). In contrast, the constitutively active Ras mutant, v-Ha- activated p21ras or MAPKK-1 mutants on ERK-2 are MAPKK- 1 mutant ras, which has no effect on NFAT alone, synergizes monitored. Expression of activated with in in ERK-2 kinase in strongly ionomycin for NFAT induction. The data induced an -5-fold increase activity 1C thus show that the activated MAPKK- 1 mutant ERK-2 Figure T cells. The activated p21ras induction of activity does not mimic activated for is of the induced the p2lras NFAT induction. 50% maximal response by TCR, To ensure that the failure of active MAPKK- 1 21 which is consistent with our of the 7E/ previous quantitation 221E mutant to NFAT in T cells did not reflect ERK-2 in T cells activated regulate activity expressing p2lras the of MAPKK-1 to stimulate the et The increase in ERK-2 inability ERKs, experi- (Izquierdo al., 1993). activity ments to monitor ERK-2 in cells the in cells activated MAPKK-1 mutant was to a activity expressing expressing activated activated MAPKK-1 mutant were In initial level which is to the effects of performed. equivalent p2lras the activated MAPKK- 1 mutant was co- on ERK-2 activity experiments, (Figure 2B). function of the transfected activated in T cells with The data in The effector expressed ERK-2tag. Figure T co-trans- 1 mutant on the ERK in Jurkat cells 2A show that isolated from cells MAPKK- endogenous ERK-2tag has a reduced was also examined its to fected with activated MAPKK-1 mutant by assessing ability regulate factor which in to the ERK-2 isolated of the SDS-PAGE phosphorylation transcription Elk-I, mobility similarly et activated T cells. To confirm that this shift is a cellular substrate for ERK-2 from (Marais al., 1993). mobility that can form a correlates with activation of the Elk- 1 is one of a of of ERK-2tag family proteins temary enzyme activator SRF and thus that the activated MAPKK-1 could with the and to demonstrate complex transcriptional in of stimulate ERK-2 kinase basic can a role activity, play key regulation c-fos gene expression assays using myelin 3925 E.Genot et al. )1 1t - ................ =E RK-? ). 1-0111ILVAVJNW :":::.: ... tag OD *ERK -2' tag -- f.hi - - - PdbL t- - --t Y.A K ermnpty KKP APKK wvt nio ERK--2Lj 0UJP _bwk _ Ik ~ 4_-L-exA 1-OP .h -_-- LexA FL;k- 1 '- t III Li -.7 O ( F'dl-,u I-tH r-l Ras" NIAPKK' Fig. 2. The activated MAPKK mutant is able to activate ERK. (A) Jurkat cells were co-transfected with the pEF-BOS construct ERK-2tag (12.5 ,ug) together with the empty or with the activated MAPKK-1 vector, mutant (12.5 of MAPKK or the jg pEXV3 217E/221E) MAPKK-1 wild-type (12.5 ,ug pEXV3 MAPKK wt). At 16 h cells were left post-transfection, unstimulated or were stimulated with Pdbu 50 for 10 ng/ml min. Total cellular extracts were made and run on SDS-PAGE. Data show Western blot of with the 9E10 Jurkat cells analysis ERK-2tag antibody. were (B) co-transfected with the pEF BOS construct with ERK-2tag (15 jig) together 25 of the vector or with the activated Ras of ,ug empty (25 jg pEXV3 v-Ha-ras) or MAPKK-1 (25 jig of pEXV3 MAPKK mutants. At h 217E/221E) 16 cells were left unstimulated or were stimulated post-transfection, with UCHT-1 (10 jg/ml) for 10 min. Cells were the was lysed, ERK-2tag immunoprecipitated with the 9E10 and kinase antibody was activity in assessed a kinase assay using MBP as substrate. into MBP was with Radioactivity incorporated the quantified phosphoimager (Molecular Dynamics). Data show the MBP kinase in as a of the activity ERK-2tag precipitates presented maximal obtained percentage response TCR by stimulation. (C) Jurkat cells were co-transfected with 15 jg of the the fusion N plasmid encoding with 20 protein, pEF Lex-Elk-IC, together ,ug of the vector or with the empty activated Ras (20 of or jig pEXV3 v-Ha-ras) MAPKK-1 (20 of MAPKK mutants. At 16 h jig pEXV3 217E/221E) post- transfection. cells were transfected with the vector and were either left unstimulated or were empty stimulated with or UCHT-1 (10 Pdbu jg/ml) (50 for 30 min. Total ng/ml) cell extracts were made and run on SDS-PAGE. Data show blot of Western LexA-Elk-lC with analysis LexA antibodies. (Hill and Treisman, 1995). It has been shown et The role previously (Qiu al., 1995). of Rac-l in NFAT regulation that activators of the ERK kinase cascade can induce in T cells was explored by monitoring the effects of a phosphorylation of the Elk-I C-terminus and dominant Rac- potentiate negative 1 mutant, N17Rac, and a constitu- Elk-i transcriptional activity (Price et To tively active al., 1995b). Rac-1 mutant, V12Rac (Rac*) (Nobes and monitor this a phosphorylation, fusion protein Hall, 1995). The data in 3A comprising Figure show that expression the C-terminus of Elk-I linked to the LexA of the dominant negative repressor N17Rac molecule inhibits NFAT (LexA-Elk- was co-transfected into T cells with 1C) activ- induction by the TCR and also, although to a lesser extent, ated or MAPKK-1 mutants. The p2iras suppresses the NFAT response phosphorylation induced by phorbol ester of LexA-Elk-1C reduces its electrophoretic in and calcium ionophore. In an that mobility experiment investigated SDS-PAGE gels. the data in 2C show Accordingly, Figure the N 17Rac dose dependency of NFAT induction, it the reduced electrophoretic of LexA-Elk-1 in mobility appeared that Pdbu and ionomycin induction of NFAT Pdbu- or TCR-activated cells with control compared comprised an N17Rac-sensitive and -insensitive com- unstimulated cells. LexA-Elk-1C isolated from T cells ponent (Figure 3B). To examine the selectivity of the either activated expressing p2 Iras or activated MAPKK- effects of the 1 dominant negative mutant of Rac- 1 on TCR mutant has a similar reduced electrophoretic signalling, the effects mobility of N 17Rac- on TCR and Pdbu (Figure 2C). The activated MAPKK and activated p2lras activation of ERK-2 were determined. Accordingly, T had equivalent effects on stimulatory the transcriptional cells were transfected with ERK-2tag either alone or of Elk-i activity (data not shown). together with N17Rac. The data in Figure 3C show that In the summary, data in Figures and 2 show that TCR or Pdbu stimulation or transfection of activated of a expression dominant mutant of negative MAPKK-1 p21ras induces ERK-2 activation while expression of suppresses TCR induction of NFAT. The data demonstrate activated Rac mutant has no effect on ERK-2 activity. also that expression of an activated MAPKK-1 mutant in Co-expressing N17Rac in these cells did not affect p21ras, T cells is sufficient for ERK-2 activation and sufficient TCR or Pdbu activation of ERK-2, showing that the for phosphorylation of the transcription factor Elk-1. inhibitory effects of Ni 7Rac did not result from inhibition However, expression of a constitutively active 1 MAPKK- of ERK-2 (Figure 3C). mutant is not sufficient to substitute for p21ras for NFAT To explore further the role of Rac- 1 in NFAT regulation, induction. Accordingly, to explain p21ras regulation of the effects of an activated Rac-1 mutant (V12Rac) on NFAT in T cells, it is necessary to invoke the existence transcription factor activity were compared with the effects of alternative Ras effector pathways. of the activated p2iras mutant. p2iras can regulate the transcription factor AP-1 activity in a variety of cells The effects of inhibitory and activated mutants of including T cells (Rayter et al., 1992; Woodrow et al., Rac- 1 on NFAT induction 1993b). To compare p2iras and Rac-1 regulation of the The Ras-related GTPase Rac-1 has been identified as a activity of AP- 1, we quantified the expression of an AP- 1- downstream effector of p21ras in fibroblast transformation CAT reporter gene. The AP- 1 reporter gene can be induced 3926 in T cells NFAT regulation i~~~ -- ~ ~ \ FAT CAT LC]ITI '! t *Pd*'bu+ i0000' (tCill .1 n ...... _ _ :\FT- C-x I ---4-.- ._ _ _ . _ _ o r _ lmicro-ram., 10 iiiillioii celk) _ _ V17 Rac concentration J '- _ | __ Il ..,I ~ --: ow &WOO -Mb.- 14'-- .Wmkm- 4movool qww E P,(- : ar: '- ,ti II li-[ 7 _- _ = "dhoLCHF-I. Ras* Rac* db Ras' Rac* Ci Pdbu UCHT-' :I' -= _ : .e~~~~~~~~~~~~~~~~~- em-nv Rac N ` __ 3. The effects of an inhibitory Rac mutant on NFAT induction. (A) Jurkat cells were co-transfected with the NFAT-CAT reporter construct Fig. (12 jig) and 5 jig of the empty vector or the dominant negative Rac mutant-containing plasmid (5 jig of pcDNAl N17Rac). At 6 h post-transfection, cells were either left unstimulated or were stimulated with UCHT-1 (10 ,ug/ml) or a combination of Pdbu (50 ng/ml) and ionomycin (0.5 jig/ml) for 16 h before extracts were made, and CAT reporter activity was assessed. Data show fold induction of NFAT-CAT activity. Results from one experiment representative of five are presented. (B) Jurkat cells were co-transfected with the NFAT-CAT reporter construct (12 jig) and the empty vector or different concentrations of the inhibitory Rac mutant (pcDNAl N17Rac). and the experiment was carried out as in (A). Data are shown as a percentage of the maximum NFAT-CAT induction obtained with UCHT-1 stimulation in the absence of the inhibitory Rac mutant. (C) Jurkat cells were co-transfected with the ERK-2tag construct (18 jig) together with 8 jig of the empty vector, or plasmids encoding active Ras (8 jig of pEF- jg Rac*) mutants, in the absence or presence of 5 jig of the inhibitory Rac mutant BOS v-Ha-ras) or the active Rac (8 of pEFplink V12Rac. (pcDNAl N17Rac). At 16 h post-transfection, cells transfected with the empty vector or with the plasmid encoding the inhibitory mutant were either (10 jg/ml) or Pdbu (50 ng/ml) for 10 min. Total cellular extracts were made and run on SDS- left unstimulated or were stimulated with UCHT-1 of ERK-2tag with the 9E10 antibody. PAGE. The data show Western blot analysis and show that the combination of by Pdbu, TCR triggering with UCHT1 antibodies presented in Figure 5C (Figure 4A). AP-1 1 activated MAPKK could not synergize constitutively active p2lras protein activated Rac- and activated Rac-1 mutant signals to induce NFAT. No further induction activity is also induced by the with calcium demonstrate that TCR construct could be seen when (Figure 4A). Further experiments of the NFAT-CAT reporter and Pdbu induction of AP- 1 activity in T cells are different ratios of the two plasmids (data not transfecting of the dominant negative Rac-1 The data in Figures 1-5 collectively show that prevented by expression shown). mutant N17Rac (Figure 4B). These data show that Rac-1 MAPKK- 1 and Rac- 1 function are each important but not is essential for TCR and Ras regulation of AP-1 sufficient, neither alone nor in combination, for TCR and activity in T cells. Moreover, expression of constitutively activated p21 ras regulation of NFAT. Rac-1 is sufficient to stimulate levels of AP-1 activity NFAT and AP- 1 in T cells with that seen in cells expressing activated p2lras regulation of comparable mutants (Figure 4A). To compare p2iras and requires Rac-1 function p21ras Rac-i in pathways, the MAPKK- 1 regulation of the activity of AP- 1, we quantified To position p2lras signalling in cells Rac- 1 mutants on signals the of the AP-1-CAT reporter gene effect of the dominant negative expression or the active p21ras mutant were either constitutively active p2iras protein generated by constitutively expressing in 4C show that monitored. of N I 7Rac- 1 did not prevent activated MAPKK-1. The data Figure Expression of activated of the constitutively active p21 ras mutant in T AP- 1-CAT is not induced by expression expression activated Rac-1 did not in T cells. cells not Moreover, MAPKK-1 mutant (data shown). nor did AP- 1 a of the active NFAT complex induce activation of ERK-2 (Figure 3C), expression Since is component Pdbu-mediated activation the effects of activated Rac-I of N17Rac or in T cells, we also assessed prevent p2iras- Results in 6A NFAT in T cells. The data in Figure 5A of ERK-2 (Figure 3C). presented Figure on responses of N17Rac did not activated Rac- 1 could not mimic activated show that expression prevent p2iras- show that of which is consistent and with calcium for NFAT mediated phosphorylation Elk-i, p21 ras synergize signals Rac- 1 to be seen when the failure of the dominant negative prevent induction. No NFAT induction could using with the ERK-2 for which activated activation of pathway (Figure the activated Rac in conditions p2 lras-mediated mutant, ERK kinase cascade can induce and under conditions where 3C). Activators of the Ras was effective 5B) (Figure activation the Elk-i C-terminus 1 induce AP- 1. Given that neither transcriptional by (Price activated Rac- could and To monitor Elk-I of activated MAPKK- 1 nor activated Rac- 1 et 1995b; Hill Treisman, 1995) al., expression LexA-Elk-1C was co-transfected ras for we the transcriptional activity, substitute for p21 NFAT, investigated pos- CAT T cells with a LexA reporter a between MAPKK- 1 and Rac- 1. Data into operator-controlled sibility of synergy 3927 E.Genot et al. .E~~ ~ ~ ~ ~ ~ ~ ~ AP'-1 CA *Ail d ,( Stiniuli .. ni c N t co ( il i N17 Rat- nlcentr dtio01 mr]icro1gra m1/l celll tt ITs)1 Fig. 4. The effects of Rac and MAPKK mutants on AP-1 induction. Jurkat cells were co-transfected with the AP-1-CAT construct (A) reporter (2 and 8 of the empty vector or the the active Ras or active Rac gg plasmid encoding constitutively (pEF-BOS v-Ha-ras) (pEFplinkVl2Rac) gg) mutants. After 6 h, transfected cells were either left unstimulated or were stimulated with either UCHT-1 or Pdbu for 16 h (10 .tg/ml) (50 ng/ml) Data a before extracts were made, and CAT was assessed. are as of the maximum AP-1-CAT reporter activity presented percentage response induced by Pdbu. (B) Jurkat cells were co-transfected with the AP-1-CAT construct and the vector or different concentrations of reporter (2 ,tg) empty the and was carried plasmid encoding the inhibitory Rac mutant (pcDNAI the out as in Data are shown as a of N17Rac), experiment (A). percentage the maximum obtained with Pdbu stimulation in the absence of the Rac mutant. Jurkat cells were co-transfected with the response inhibitory AP-1- (C) CAT construct and 15 of the vector or the active Ras mutant of or reporter (2 gg) gg empty constitutively (15 gg pEXV3 v-Ha-ras) constitutively active MAPKK-1 mutant of MAPKK At 6 h cells transfected with the vector were either (15 tg pEXV3 217E/221E). post-transfection, empty left unstimulated or were stimulated with either UCHT-1 or Pdbu for 16 h before extracts were and (10 .tg/ml) (50 ng/ml) made, CAT reporter activity was assessed. Data are shown as a of the conversion of into the percentage chloramphenicol acetylated product. LexA-Elk- 1 we gene (LexA-OP.tkCAT). C transcription examined the effects of activated Rac- on LexA- is low in unactivated T cells but can be induced Elk-i C The data in activity phosphorylation. Figure 6A show that of a active v-Ha- ofV I by co-expression constitutively p2lras, expression 2Rac- 1 in T cells induced phosphorylation Ras, or by stimulation of T cells with Pdbu of LexA-Elk- IC with that (Figure 6B). comparable induced by activa- Expression of N17Rac did not suppress Pdbu or v-Ha-ras ted p2Iras. Rac-1 function is not required for p2iras induction of LexA-Elk- C. In contrast, in the same induction of Elk-i (Figure 6A and B) which is mediated experiment, Ni7Rac inhibits the AP-1 induced via ERK-2. Nevertheless, these results indicate in T response that, the by constitutively active p21ras and also cells, Rac- has the potential to a kinase (Figure 6C) regulate cascade prevents NFAT induction induced by activated p2 lras plus that can converge with the ERK-2 on the tran- pathway calcium These ionophore (Figure 6D). data thus indicate scription factor Elk- 1. that Rac-I function is for required p21ras induction of AP-1 and NFAT but Rac-1 function is not for necessary Discussion activation p2iras of ERK-2 and Elk-1. Expression of activated Rac-1 did not activate ERK-2 The MAP kinase ERK-2 has been proposed to be a major in T cells (Figure 3C) nor was Rac-I function necessary effector of the p21ras signal transduction pathways in a for TCR- and p2 lras-mediated stimulation of ERK-2 variety of cell systems including the differentiation of (Figure 3C). Nevertheless, it has been described that Rac- 1 PC 12 cells and the transformation of fibroblasts. The plays a role in AP-1 regulation via control of Jun kinases present has study used constitutively active and inhibitory such as JNK/SAPK. Elk-i has also been described to be mutants of a MAP kinase kinase (MAPKK- 1) to examine a substrate for the JNK/SAPK family (Whitmarsh et al., the role of this pathway in TCR signal transduction. 1995; Zinck et al., 1995), and to explore if Rac-I could Activated mutants of MAPKK-1 can stimulate ERK-2 regulate transcription factor phosphorylation in T cells, kinase activity in T cells and can also induce phosphoryl- 3928 NFAT regulation in T cells _._ NIFATCA I u'n- )ItrolI o Ra': m E-k r C A t t 0~ ~~~~~~ 11 t.3 -1 z_r -; tiI i _ II \NEA'T' C.AT' C: ac R{ \ 1\T UY TlO!10 _ ii)1 (TIOMNCllt 0 L(CII-I- -1 :E I r , - ')II --o Stimll - -C DNA 10 mill cell-) coicenritiratioin (miciogratlm ioni Fig. 5. Comparison of the effects of the activated Rac and Ras mutant on NFAT induction. (A) Cells were co-transfected with the NFAT-CAT reporter construct (12 jg) and 8 jig of the empty vector or the plasmid encoding activated Ras (8 jig of pEF-BOS v-Ha-ras) or activated Rac (8 jig of pEFplinkV12Rac) mutants. Transfected cells were either left unstimulated or were stimulated with ionomycin (0.5 or a combination of gg/ml) Pdbu (50 ng/ml) and ionomycin (0.5 ,ug/ml), 6 h post-transfection and for 16 h before extracts were made, and CAT reporter activity was then assessed. Data show the percentage conversion of chloramphenicol into the acetylated product. Results from one experiment representative of four are presented. (B) Jurkat cells were co-transfected with the reporter construct (12 jg) and the empty vector or different concentrations of NFAT-CAT the the activated Ras plasmid encoding (pEF-BOS v-Ha-ras) or activated Rac (pEFplinkV12Rac) mutants, and the experiment was carried out as in (A). Data are shown as a percentage of the conversion of chloramphenicol into the acetylated product. (C) Plasmids encoding the activated Ras (15 jig of pEXV3 v-Ha-ras), the activated MAPKK-1 (15 jg pEXV3 MAPKK 217E/221E) or the activated Rac (15 jig of pEFplinkVl2Rac) mutants, or a combination of the activated MAPKK-1 (15 jg of pEXV3 MAPKK 217E/221E) and activated Rac-1 (15 jig of pEFplinkVl2Rac) mutants were co-transfected with the NFAT-CAT reporter construct (12 jig). Cells transfected with the empty vector were either left unstimulated or were stimulated with Pdbu (50 ng/ml). Transfected cells were then either left unstimulated or stimulated with ionomycin (0.5 jg/mI) or UCHT-1 (10 jg/ml) for 16 h before extracts were made, and CAT reporter activity was assessed. Data show the percentage conversion of chloramphenicol into the acetylated product. Results from one experiment representative of four are presented. ation of the ternary complex factor Elk-1, demonstrating confirms this idea and thus clearly demonstrates that that MAPKK-1 is an effector molecule in the regulation additional p2lras effector pathways must exist in T cells. of the transcription factor Elk-I in T cells as in fibroblasts Recently, the p2iras-related GTPase Rac-1 has been (Hill and Treisman, 1995). The transcription factor NFAT found to be an important mediator of oncogenic trans- is controlled by the TCR by a mechanism involving a formation by Ras (Qiu et al., 1995). Here we describe a 1 in T cells. interaction of and calcium role for Rac- in transcription factor regulation synergistic p2lras signals. did not an MAPKK-1 mutant indi- A dominant mutant of Rac-1 Experiments using inhibitory negative suppress cate that the activation of the MAPKK- 1/ERK- -2 TCR or p21ras activation of the MAP kinase ERK-2 but 1, cascade a role in TCR activation of NFAT because could TCR and induction of NFAT and plays suppress p2lras with the existence of a TCR stimulation of NFAT was prevented by the inhibitory AP-1. These data are consistent in form of MAPKK-1. activated MAPKK- 1 could Rac-l-controlled signalling pathway operating parallel However, mediate iras induction not substitute for 1 ras and with calcium with the ERK-2 pathway to p2 p2 signals synergize with a to stimulate NFAT. A of the effects of of AP- 1 and NFAT. Importantly, experiments signals comparison Rac- 1 mutant showed that Rac-1 was inhibitory p2iras or MAPKK-1 mutants on thymocyte constitutively active that for activation of the in mice concluded a sufficient signal AP-1, although development transgenic recently be mediated could not substitute for the activated functions in T cells cannot active Rac-1 mutant p2iras solely by et for induction. of activated the ERK-2 cascade p2 Iras mutant NFAT Expression signalling (Alberola-lla al., 1995; of the activated MAPKK- in T cells could stimulate and the Swan et The failure 1 MAPKK-1 ERK-2, al., 1995). in NFAT induction Rac- 1 mutant can stimulate kinase to substitute for activated activated parallel signal- p2iras 3929 E.Genot et al. (j .\AI-1 ( ._ .4_ ~.+-i e x l/ U0A4US --*-ii-1 4--E xA lk -'I i'.. ...- O Ras* Rac" PdbLU Ras' Rac" () PdIl- :-. !: ,- \ 1;: ! I'diq \I. empt.y __ N 17 Rac \1.AV C' \T ~~~~. .1 -- 1/ 1 !, |1 >\~ .I~~~~ .,, ~ ~~~~~ I' , -, j~ 1, *. '':}' , \I1 II with the N Fig. 6. p21 ras regulation of NFAT and AP-I Rac-l function. Jurkat cells were co-transfected Lex-Elk-IC fusion requires (A) pEF of protein expression construct together with 8 of the vector or with the encoding activated Ras jg (12 jug) empty plasmid (8 pjg pEF-BOS v-Ha-ras) the Rac mutant or activated Rac (8 ,ig of mutants and in the absence or of the encoding of pEFplinkVl2Rac) presence plasmid inhibitory (8 ig pcDNAl N17Rac). At 16 h cells transfected with the vector were either left unstimulated or were stimulated with post-transfection, empty Pdbu (50 ng/ml) for 30 min. Total cell extracts were made and run on SDS-PAGE. The data show Western blot of LexA-Elk-IC with analysis LexA its antisera. (B), (C) and (D) Jurkat cells were co-transfected with the MLV N Lex-Elk-lC fusion construct and (6 jig) protein expression reporter construct LexA-OP.tkCAT AP-1-CAT or NFAT-CAT constructs with 15 of the vector (3 (B), (2 jg) (C) (12 jg) (D) reporter together ,ug pg) empty Ras mutant of and in the absence or of 2.5 of the or with the plasmid encoding the activated (15 ,ug pEF-BOS v-Ha-ras) presence plasmid Pjg At 6 h cells transfected with the vector or with encoding the inhibitory Rac mutant (pcDNAl N17Rac N17Rac). post-transfection, empty pcDNA1 for 16 h. were carried out as alone were either left unstimulated or were stimulated with Pdbu CAT before. (SOng/ml) assays ling pathways that could and particularly in cell adhesion and phosphorylate transcrip- regulating morphology activate Elk-I in T cells. the (Rodriguez-Viciana et al., Nobes and tionally Nevertheless, 1994; Hall, 1995). combination of activated MAPKK-1 and Rac-I could not Also, Ptdlns 3 kinase can bind directly to activated p21ras, substitute for activated p2lras and synergize with calcium and function as an effector molecule for p2 lras in neuronal to NFAT. signals induce There be at least cells (Rodriguez-Viciana et al., 1994). There is, must, therefore, therefore, one more p21ras effector required for NFAT induction the possibility that Ptdlns 3 kinase could be the link There are several additional (Figure 7). candidate p2lras between Ras and Rac-1. Preliminary experiments show effectors the GTPase Rho et that Ptdlns 3 kinase inhibitors including (Prendergast al., (wortmannin and LY294002) 1995), and also the GTPase Ral et fail to show effect (Hofer al., 1994; any inhibitory on TCR induction of and whose function in NFAT K.Reif and Spaargaren Bischoff, 1994), NFAT (E.Genot, D.A.Cantrell, unpublished regulation remain to be investigated. p21ras effects on the data). The failure of Ptdlns 3 kinase inhibitors to regulate MAPKK-1/ERK-2 cascade in T cells are mediated the NFAT is not with a role for Ptdlns by compatible 3 kinase serine kinase Raf-l (Izquierdo et al., 1994a). It is not as a mediator of p21ras effects on NFAT. Accordingly, clear whether or not there are multiple Raf-1 effector the signalling links between p2lras, Ptdlns 3 kinase and in T cells in Rac- 1 in T pathways but, fibroblasts, Raf-1 and Rac-1 cells may be different from those described in signals synergize to mediate cellular other cell transformation by lineages. However, this question has to be p2lras (Qiu et al., 1995). We currently are exploring investigated further with different and complementary functional interactions between Raf-1 and Rac-1 in NFAT approaches. Concerning the Rac-1 effector pathway for induction. In this context, Jurkat cells expressing constitu- NFAT induction, Rac- 1-regulated protein kinases have tively active Raf-1 show enhanced levels of IL-2 produc- now been identified (Manser et al., 1994; Martin et al., tion but there has not been a direct analysis of the role of 1995), but the role of these protein kinases in Rac-l Raf- in NFAT regulation (Owaki et effector al., 1993). pathways in T cells remains to be discovered. Questioning how p2 lras would regulate Rac- 1, a couple The transcriptional activity of the NFAT binding site of recent reports raise some possibilities. Thus, a report used in the present experiments is mediated by a complex has described phosphatidyl inositol 3' kinase (Ptdlns 3 a comprising member of the NFAT group of DNA binding kinase) as an upstream regulator of Rac-1 in fibroblasts, proteins and AP-l family proteins (Rao, 1994). The AP-l 3930 NFAT regulation in T cells TCR of Fos and Jun proteins which up- the phosphorylation activity. Several parallel MAP regulates their transcription ERK- 1, -2, JNK- 1, kinase signalling pathways involving Fos kinase can all -2 and MPK2/p38 and a Ras-regulated potentially contribute to AP- 1 activation (Deng and Karin, Coso et al., 1995; Hill 1994; Cobb and Goldsmith, 1995; Elk-I which contributes and Treisman, 1995). The TCF via complex formation with to increased c-fos expression factor SRF is a substrate for the ubiquitous transcription -2 (Marais et al., 1993; Whitmarsh the ERKs and JNK-1, \l1APKK-l I is also transcriptionally regulated by et al., 1995). c-jun ,/ i-> Rac- Cilc]iIL'Llr MAP is a direct substrate for JNK- 1 and kinase pathways, JNK- 1-mediated phosphorylation can be activated by R\\K-? ) \Ft\ Recent studies suggest that Rac- 1 (Derijard et al., 1994). \ \ IFt5>. ItsIll role in the activation of the MAP kinases plays a central AI /\ 1;Cn that is exactly analogous to the role of JNK and p38 in ERK pathways (Coso et al., 1995; Minden p2lras the et The present data show that Rac-l signals are al., 1995). APFo-;I II\II necessary and sufficient to induce Elk-I phosphorylation i4 in T cells, which indicates that Rac-1 signals regulate . ;.-1\F Xl- transcriptional factor kinase pathways in T cells. Future studies will establish whether Rac- 1 effects on Elk- 1 Cvtokine genes phosphorylation in T cells are mediated by JNK-1/SAPK, with or a novel protein kinase pathway. By analogy p38 other models for transcription factor regulation by p21 ras- Fig. 7. Schematic representation of signalling pathways predicted to induce NFAT. Our model for NFAT induction shows calcium! related GTPases, then Rac-1 effects on NFAT could be calcineurin signals regulating NFAT translocation from the cytosol to kinase pathways. Clearly mediated by one of the MAP the nucleus. p2lras /Rac-1 regulation of NFAT involves regulation of 1 in T Rac- does not regulate ERK-2 cells, but it is possible AP-1. MAPKK-1 signalling pathways are also required for NFAT 1 NFAT involves that the role of Rac- in TCR induction of induction and, since MAPKK-1 does not regulate AP-1, we would MAP kinase. It is possible, predict that MAPKK-1 pathways regulate NFAT itself or the NFAT- JNK- 1/SAPK or the p38 MAPKK- AP-1 complex. To explain why the combination of activated in TCR induction of moreover, that the role of Rac-1 1 for activated p21ras for NFAT and activated Rac-1 do not substitute NFAT will be analogous to the role of Rac- 1 in serum have to evoke the existence of at least one more p21ras induction, we activity in fibroblasts, regulation of SRF transcriptional contribution of PKC to NFAT regulation is signalling pathway. The Rho/Rac- 1-controlled signal- because the TCR activates PKC but this is not required which occurs by an unknown complicated for NFAT induction. Phorbol esters can regulate NFAT, but this occurs that does not involve ERK-2, JNK- 1 or p138 ling pathway an pathway that converges on by independent PKC-mediated et (Hill al., 1995). MAPKK-1 and Rac-1. Rac-1 mediate p2lras regulation of AP-1 and may NFAT in T cells but, if the signalling connections between is composed of dimers of members of the Fos and Rac-l in T cells are as complicated as those complex p2lras and Jun family of proteins. The exact nature of the AP-1 in fibroblasts, then it is unlikely that p2iras and Rac-1 it component of NFAT in T cells has not yet been established will function solely in a linear signalling cascade and will in both clearly. Boise et al. have described that a Fra-1-JunB is more likely that Rac-1 and p2lras function Rac- 1 has diverse roles heterodimer is the inducible nuclear component of the and convergent pathways. parallel For in NFAT binding activity, while Northrop et al. have reported in many cellular systems. example, neutrophils, 1 of the NADPH oxidase a dimer of c-Fos (or Fra- 1) and either c-Jun or Rac- is involved in the activation composed In Rac-1 plays a that the nuclear subcomponent may be complex (Xu et al., 1994). fibroblasts, JunD, suggesting of AP- 1-NFAT in of the actin cytoskeleton represented by a heterogeneous population critical role the regulation It and but can also regulate transcriptional (Boise et al., 1993; Northrop et al., 1993). (Nobes Hall, 1995) complexes AP-1 used in the of SRF and controls c-jun transcriptional activity is not clear whether the reporter gene activity measures the of the AP- 1 activation of the c-Jun kinases JNK-l/SAPK (Coso present experiments activity via Minden et In NFAT of distinct AP- 1 members. et Hill et 1995; al., 1995). components of or family al., 1995; al., 1 in T cells is an AP- 1 data are this the activation of JNK- Nevertheless, the present reporter gene context, the in T Rac- 1 function for from co-stimulatory recep- valuable because they show that, cells, integration point signals and can be sufficient for AP- 1 induction tor CD28- and TCR-mediated calcium/calcineurin signals is both necessary It will thus be of interest to determine and that Rac- 1 can mediate Ras effects on AP- 1. Moreover, et 1994). (Su al., whether Rac-1 is an for the activated Rac- 1 mutants were equivalent to the activa- in future studies integration point activation. of AP- 1 induction. This and CD28 for JNK-1/SAPK ted Ras mutants in the context TCR signals has examined 1 1 could thus In the present study p21ras role of Rac- in AP- regulation explain summary, T cell line Jurkat in an to 1 for NFAT induction. to effector in a attempt Rac- Accordingly, pathways requirement contributed for activa- Rac- effector in T characterize the by p21ras consider the 1 pathways cells, signals possible effector involves to focus on the mechanisms for AP- 1 tion of NFAT. One p2iras pathway it is necessary which is shown the data to be both In the activation of AP- 1 results from MAPKK- by present induction. general, 1, sufficient for activation of ERK-2 and the of Fos and Jun family proteins and necessary and increased expression 3931 E.Genot et al. Western blot analysis and MBP kinase assays transcription factor Elk- I in Jurkat cells. Rac- 1 is a Cells cultured in complete medium were harvested at different times sufficient and necessary signal for AP-1 activation and after transfection and washed three times in RPMI alone. SDS was also appears to be involved in p21ras effector pathways added to 1% final concentration. After vortexing, extracts were heated in Jurkat, since p2iras-mediated induction of AP-1 and for 5 min at 100°C. Debris were pelleted by ultracentrifugation for 30 min at 90 000 r.p.m. in the TL100.2 rotor of a Beckman centrifuge. NFAT requires Rac- 1 function. MAPKK- 1- and Rac- 1- Proteins were acetone precipitated from supernatants for 10 min on ice mediated signals may be sufficient for either ERK-2 or and resuspended in reducing Laemmli sample buffer. Disruption of the AP-1 activation but, either alone or in combination, they pellet was completed by sonication. Extracts prepared from equivalent cannot substitute fully for p21ras signals for induction of numbers of living cells were subjected to SDS-PAGE on 10% gels (10% acrylamide-0.27% bisacrylamide), unless otherwise indicated. Proteins NFAT. There must, therefore, be at least one more p21ras were transferred to polyvinyldifluoridine membranes (Immobilon, pathway that is required for NFAT induction. The effector Millipore) by overnight Western blot. Membranes were 'blocked' in T in the present study has proved to Jurkat cell line used phosphate-buffered saline (PBS)-0.05% Tween 20, (PBST) plus 5% system for studies of NFAT regulation. be a valuable model milk for 2 h. One tg/ml of 9E10 antibody or a 1/1000 dilution of LexA likely that there will be multiple Ras antisera was allowed to react with the membrane overnight at 4°C. After It would thus seem extensive washes, membranes were incubated with a dilution of anti- populations that contribute to effectors in normal T cell mouse (1/4000) or anti-rabbit IgG (1/10 000) in PBST for 1 h at room gene induction. This NFAT regulation and hence cytokine temperature. After the washes in PBST, and PBS, membrane-bound of fibroblasts where is analogous to the transformation antibodies were visualized by use of ECL Western blotting detection reagents (Amersham) and Kodak XS 1 films. ERK-2 tag immunoprecipit- Ras effects are mediated by multiple effectors. The Rac- 1 ations with 9E10 antibody and MBP assays were carried out as described effector pathway for NFAT induction is not yet character- previously (Izquierdo et 1993). al., ized. Nevertheless, the NFAT family of transcription factors are important in the regulation of cytokine genes CAT assays and hence important for a successful immune response. CAT assays were performed using the liquid scintillation method of The present data predict that Rac-1 and Rac-1 signalling Sleigh et al. using modifications (Sleigh, 1986). Briefly, 5X 106 cells would be important targets for immune inter- were lysed in 50 ,ul of 0.5% NP-40, 10 mM Tris pH 8, 1 mM EDTA pathways 150 mM NaCI for 10 min on ice. Cellular debris was pelleted and vention. lysates were heated at 68°C for 10 min before use. Assay conditions were 150 mM Tris pH 8, 0.05 tCi [14C]acetyl coenzyme A and 2 mM chloramphenicol. After 18 h at 37C, chloramphenicol was extracted with ethylacetate, and the amount of radioactivity in the acetylated Materials and methods products and non-acetylated substrate of each reaction was determined Reagents by liquid scintillation counting of organic and aqueous phases respect- ively. The amount of chloramphenicol acetylation is calculated as a dibutyrate were from Calbio- lonomycin (Ca2+ salt) and phorbol-12,13 percentage. coenzyme A (at 50-60 mCi/mmol) was chem Corp. (UK). [14C]Acetyl International (Buckinghamshire, UK). Other reagents from Amersham Monoclonal antibodies UCHT-1 and 9E10 reactive were from Sigma. and the c-Myc epitope used a tag for ERK-2 were with human CD3 Acknowledgements from supernatants at I.C.R.F. Rabbit antisera affinity purified hybridoma gift of Richard Treisman and Alberto Gandarillas specific for LexA were a We thank S.Cowley and C.Marshall for the MAPKK-1 constructs, and goat anti-rabbit antibodies are (ICRF). Polyclonal rabbit anti-mouse N.Clipstone and G.Crabtree for the J-TAg cell line, R.Treisman and (Buckinghamshire, UK). for LexA antibodies, M.A.Price for LexA-OP.tkCAT from Amersham International A.Gandarillas and LexA-Elk-IC constructs and for advice, J.Babbage for technical assistance, J.Lunan for help with the manuscript and all members of Plasmid constructs the Lymphocyte Activation Laboratory for helpful comments on the The reporter construct AP- 1-CAT, containing three copies of the sequence manuscript. E.G. is a member of the INSERM organization. S.H. was for the AP-1 site inserted in pBLCAT2. has been described previously supported by a grant from the Deutsche Forschungsgemeinschaft. This et al., 1995). LexA-OP.tkCAT contains two copies of the (Williams work was funded by the Imperial Cancer Research Fund. contains three copies of the NFAT binding LexA reporter, NFAT-CAT upstream of site TAAGGAGGAAAAACTGTTTCATACAGAAGGCG IL-2 driving the reporter gene CAT (Verweij et al., the minimal promoter directing expression of wild-type MAPKK- 1: pEXV3 References 1990). Plasmids MAPKK-1: pEXV3 MAPKK 217E/ MAPKKwt, constitutively active Alberola-lla,J., Forbush,K.A., Krebs,E.G. and Perlmutter,R.M. Seger,R., MAPKK-1: pEXV3 MAPKK 221A and 221E, the dominant negative (1995) 373, 620-623. Natutre, v-Ha-ras or v-Ha-ras, constitutively active p2lras: pEF-BOS pEXV3 and (1992) J. Biol. 267, Baldari,C.T., Macchia,G. Telford,J.L. Chemii.. fusion protein or pEF pEF BOS ERK-2 tag, MLV N-LexA-Elk-IC 4289-4291. N-LexA-Elk-IC fusion protein, have already been described (Medema Boise,L.H., Petryniak,B., Mao,X., June,C.H., Wang,C., Lindsten,C.B., et al., 1991; Cowley et al., 1994; Leevers et al., 1994; Price et al., Bravo,R., Kovary,K., Leiden,J.M. and Thompson,C.B. (1993) Mol. mutants of 1995a). The active and dominant negative Rac-I (Vl2Rac-I Cell. Biol., 113, 1911-1920. and N17Rac-1) have also been described previously (Nobes and Hall, Cobb,M. and Goldsmith,E. (1995) J. Biol. Cheml., 270, 14843-14846. 1995). All plasmids were purified by equilibrium centrifugation in CsCl- Coso,O., Chiariello,M., Yu,J.-C., Teramoto,H., Crespo,P., Xu,N., MIki,T. procedures. ethidium bromide gradients by standard and Gutkind,J. (1995) Cell, 81, 1137-1146. Cowley,S., Paterson,H., Kemp,P. and Marshall,C. (1994) Cell, 77, Cells and transfections 841-852. J-TAg, a subline of the human T acute lymphocytic lymphoma cell line Deng,T. and Karin,M. (1994) 371, 171-175. Natutre, Jurkat, was maintained in RPMI supplemented with 10% heat-inactivated Derijard,B., Hibi,M., Wu,I.-H., Barrett,T., Deng,T., Karin,M. and Su,B., fetal calf serum (FCS), at 37'C in 5% CO, in humidified air. Cells were Davis,R.J. (1994) Cell, 76, 1025-1037. transfected via electroporation (Gene pulser: Bio-Rad, UK) as previously Flanagan,W.M., Corthesy,B., Bram,R.J. and Crabtree,G.R. (1991) Nattlre, ( X described. cells were cells/0.5 ml) in complete Briefly, pulsed 107 352, 803-807. medium at 960 and 320 V. Cells transfected with similar plasmid Genot,E.M., Parker,P.J. and Cantrell,D.A. (1995) J. Biol. Chem?., 270, pF mixtures were pooled and re-aliquoted before stimulation. Transfected 9833-9839. cultures were incubated for 6 h, and stimuli, when required, were applied Hall,A. (1994) Annul. Repi Cell. Biol., 10, 31-54. for an additional 16 h. The amounts of DNA in transfected each Hill,C. and Treisman,R. (1995) Cell, 80, 199-211. were constant adding and Treisman,R. (1995) Cell, 81, 1159-1170. by control plasmids as indicated. Hill,C., Wynne,J. experiment kept 3932 NFAT regulation in T cells Hoey.T.. Sun,Y.-L.. Williamson,K. and Xu,X. (1995) Immunitov, 2. 461-472. HoferEF., Fields.S.. Schneider.C. and (1994) Proc. NatlAcad. Martin,G.S. Sci. USA 91. 11089-11093. Howe.L.R.. Leevers,S.J.. Gomez.N.. Nakielney.S.. Cohen.P. and Marshall,C.J. (1992) Cell, 71, 335-342. Izquierdo.M., Leevers.S.J.. Marshall,C. and CantrelliD.A. (1993) J. EI.). Med.. 178, 1199-1208. Izquierdo,M., Bowden,S. and Cantrell.D. (1994a) J. E.rp Med.. 180. 401-406. Izquierdo,M.. Leevers.S.J.. Williams.D.H.. Marshall.C.J.. Weiss.A. and (1994b) Elu: J. 24. 2462-2468. Cantrell,D.A. Immunaiiiiiol., Izquierdo-Pastor.M.. Reif,K. and CantrellD. (1995) Iinunaiiiiiol. Todayi. 16. 159-164. Paterson,H. and Marshall.C. (1994) 369. 411-414. Leevers,S., Nature, Manser,E., Leung.T.. and (1994) Nature. Salihuddin.H., Zhao,Z.S. Lin,L. 367. 40-46. Marais.R., Wynne.J. and Treisman.R. (1993) Cell. 73. 381-393. Marshall.C.J. (1994) Curr: Opin. Geniet. 4, 82-89. Dev, Martin.G.A.. BollagcA., McCormickEF. and (1995) EMBO J.. 14. Abo,A. 1970-1978. Medema.R.H.. Wubbolts,R. and Bos.J.L. (1991) Mol. Cell. Biol.. 11. 5963-5967. Minden,A.. Lin.A.. Claret,F.-X., Abo.A. and Karin.M. (1995) Cell. 81. 1147-1157. Nobes.C. and HallA. (1995) Cell. 81. 53-62. Northrop,J.P., Ullman,K.S. and Crabtree.G.R. (1993) J. Biol. Clhemli.. 268, 22917-2925. Ho.S.. Northrop,J.. Chen.L.. Thomas.D.. Timmerman.L.. Nolan.G.. Admon.A. and Crabtree.G. (1994) Ncature. 369, 497-502. Owaki,H., Varma,R., Gillis.B.. Bruder.J.T., Davis,L.S. and Rapp,U.R.. Geppert.T.D. 12. (1993) EMBO J., 4367-4373. Prendergast.G.C., Khosravi-Far,R.. Solski,P.A.. Kurzawa.H.. Lebowitz. P.F. and Der,C. (1995) Oncogene. 2289-2296. 10, PriceL.. Nornnan.J.. Ridley.A. and (1995a) Cutr,: Biol.. 5. Koffer,A. 68-73. Price.M.. Rogers.A. and Treisman.R. (I 995b) EMBO J.. 14. 2589-2601. Qiu.R.G.. Chen.J.. McCormick.F. and Symons.M. (1995) Nature. Kirn,D.. 374 457-459. Rao.A. (1994) Iuuululuol. Todav. 15. 274-281. Rayter.S.. Lucas.S.C.. CantreliD. and Downward.J. (I1992) Woodrow.M., EMBO 4549-4556. J., 11, Rodriguez-Viciana.P.. Vanheaesbroeck.B.. Warne,P.H., Dhand,R., Gout,I., Fry.M.J.. Waterfield,M.D. and DownwardJ. Nature. (1994) 370. 527-532. Schreiber.S.L. and Crabtree.G.R. (1992) Iuna711{Z7ol. Todav. 13. 136-142. Sleigh.M.J. (1986) Anial. 156. 251-256. Bioelmeti., and Spaargaren.M. Bischoff.J.R. (1994) Proc. Nalt Acad. Sci. USA. 91. 12609-12613. Su.B.. Estela.J.. Hibi,M., Kallunki,T., Karin,M. and Ben-NeriahY. (1994) Cell. 77. 727-736. Swan.K.A.. Gross,J.A.. Albeerola-lla.J.. Appleby.M.W.. Forbush,K.A.. and Thomas,J.F. Perlmutter.R.M. (1995) EMBO J., 14. 276-285. and Verweij,C.L., Guidos,C. Crabtree.G.R. (1990) J. Biol. Chel., 265. 15788-15795. Warne.P.. and Naitutre. 364. Rodriguez-Viciana.P. Downward.J. (1993) 352-355. Littinan.D.R. 76. WeissA. and (1994) Cell. 263-274. and Davis.R. 269. Whitmarsh.A.. Shore.P.. Sharrocks,A. (1995) Science, 403-407. Cantrell.A.D. and Eli: Williams,H.D., Woodrow.M., Murray,J.E. (1995) Inunuiiinol.. 25. 42-47. J. Clipstone.N.A. and Cantrell.D. J. Med., 178. Woodrow,M., (1993a) E.rp. 15 17-1522. Woodrow.M.. Rayter,S., Downward.J. and CantreIlD.A. (1993b) Ininntol.. 1-9. J. 150. Biol. Xu,X., Barry.D.. Settleman.J.. Schwartz.M. and Bokoch.G. J. (I1994) Clienti.. 23569-23574. and Zinck,R., Cahill.M.. Kracht.M.. Sachsenmaier.C.. Hipskind,R. Nordheim.A. Mol. Cell. Biol.. 4930-4938. (1995) 15, Received December 13. revisedl oni 23, 1996 onz 1995; April

Journal

The EMBO Journal – Springer Journals

Published: Aug 1, 1996

There are no references for this article.