Kinase Signal Pathway Ras-Independent Mitogen-Activated Protein Direct Tumor Lysis by NK Cells Uses A

Direct Tumor Lysis by NK Cells Uses a Ras-Independent Mitogen-Activated Protein Kinase Signal Pathway

This information is current as Sheng Wei, Danielle L. Gilvary, Brian C. Corliss, Said Sebti, of September 28, 2021. Jiazhi Sun, David B. Straus, Paul J. Leibson, Joseph A. Trapani, Andrew D. Hamilton, Michael J. Weber and Julie Y. Djeu J Immunol 2000; 165:3811-3819; ;

doi: 10.4049/jimmunol.165.7.3811 Downloaded from http://www.jimmunol.org/content/165/7/3811

References This article cites 55 articles, 27 of which you can access for free at: http://www.jimmunol.org/content/165/7/3811.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 28, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Direct Tumor Lysis by NK Cells Uses a Ras-Independent Mitogen-Activated Protein Kinase Signal Pathway 1

Sheng Wei,*‡ Danielle L. Gilvary,*‡ Brian C. Corliss,* Said Sebti,†‡ Jiazhi Sun,†‡ David B. Straus,§ Paul J. Leibson,¶ Joseph A. Trapani,ʈ Andrew D. Hamilton,** Michael J. Weber,†† and Julie Y. Djeu2*‡

Destruction of tumor cells is a key function of lymphocytes, but the molecular processes driving it are unclear. Analysis of signal molecules indicated that mitogen-activated protein kinase (MAPK)/extracellular regulated kinase 2 critically controlled lytic function in human NK cells. We now have evidence to indicate that target ligation triggers a Ras-independent MAPK pathway that is required for lysis of the ligated tumor cell. Target engagement caused NK cells to rapidly activate MAPK within 5 min, and PD098059 effectively blocked both MAPK activation and tumoricidal function in NK cells. Target engagement also rapidly activated Ras, detected as active Ras-GTP bound to GST-Raf-RBD, a GST fusion protein linked to the Raf protein fragment Downloaded from containing the Ras-GTP binding domain. However, Ras inactivation by pharmacological disruption with the farnesyl transferase inhibitor, FTI-277, had no adverse effect on the ability of NK cells to lyse tumor cells or to express MAPK activation upon target conjugation. Notably, MAPK inactivation with PD098059, but not Ras inactivation with FTI-277, could interfere with perforin and granzyme B polarization within NK cells toward the contacted target cell. Using vaccinia delivery of N17 Ras into NK cells, we demonstrated that IL-2 activated a Ras-dependent MAPK pathway, while target ligation used a Ras-independent MAPK pathway http://www.jimmunol.org/ to trigger lysis in NK cells. The Journal of Immunology, 2000, 165: 3811–3819.

he mitogen-activated protein kinases (MAPK)3 integrate (8–10). Activation of ERK by growth factor receptor triggering multiple intracellular signals within a cell responding to follows a highly conserved set of molecular events, connected to T different external stimuli. Of the MAPK kinase family, upstream signals associated with the receptor. Association of the which includes c-Jun N-terminal kinase/stress-activated protein ki- GRB-2/SOS complex to the receptor recruits Ras to the plasma nase, p38/MAPKAP kinase 2 reactivity kinase (RK), and MAPK/ membrane, resulting in Raf-1 binding to activated GTP-bound extracellular regulated kinase (ERK), the last subgroup is the most Ras. Subsequent activation of Raf-1 leads to its phosphorylation of

extensively investigated, particularly in events associated with MAPK/ERK kinase 1 (MEK-1), which, in turn, becomes activated as by guest on September 28, 2021 growth factor receptors and G protein-coupled receptors (1, 2). a dual kinase, with the ability to phosphorylate the TEY motif ERKs are differentiated from the other MAPKs by their conserved in ERK. TEY motif, which becomes dually phosphorylated at the Tyr and ERK has largely been considered a growth- and differentiation- Thr sites upon activation. They play a critical role in gene expres- related signal molecule, involved in transcriptional control of crit- sion related to cell growth and differentiation. They have the abil- ical genes essential for cell proliferation/maturation, but it is likely ity to activate nuclear transcription factors, including Elk-1, c-Jun, that it plays a crucial role in other cellular events. For example, c-Myc, NF-IL-6, and TAL1 (3–7), as well as cytoplasmic sub- ERK has recently been reported to activate superoxide release (11) rsk strates, such as phospholipase A2, pp90 , and RNA polymerase and modulate cell-cell or cell-matrix interactions (12, 13). Cell secretion may also be driven by ERK (14). We therefore undertook the task of investigating whether ERK was important in lympho- *Immunology Program and †Drug Discovery Program, and ‡H. Lee Moffitt Cancer Center, Department of Biochemistry and Molecular Biology, University of South cyte function against tumor cells. Target ligation can direct two Florida College of Medicine, Tampa, FL 33612; §Department of Pathology, Uni- separate and distinct functions in lymphocytes, dictated by the in- versity of Chicago, Chicago, IL 60637; ¶Department of Immunology, Mayo Clinic, Rochester, MN 55905; ʈCellular Cytotoxicity Laboratory, Austin Research Institute, duction and effector phases. In the induction phase, the event trig- Heidelberg, Victoria, Australia; **Department of Chemistry, Yale University, New gered by TCR coupling to its Ag on quiescent T cells leads to †† Haven, CT 06511; and Department of Microbiology, University of Virginia, Char- signals that up-regulate genes associated with cytokine induction lottesville, VA 22908 and cell proliferation, resulting in clonal expansion and functional Received for publication May 30, 2000. Accepted for publication July 14, 2000. activation (15). Upon re-encounter of the Ag, activated T cells The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance express lytic function in the effector phase. The signal pathways with 18 U.S.C. Section 1734 solely to indicate this fact. coupled to the TCR during the initial activation phase must differ 1 This work was supported by a grant from the U.S. Public Health Service (CA83146) from those during the effector phase, because different functions and by American Heart Association Grant AHA 970175. prevail. Research to date has focused on the induction phase, and 2 Address correspondence and reprint requests to Dr. Julie Y. Djeu, Immunology it is established that the TCR is coupled to nonreceptor tyrosine Program, H. Lee Moffitt Cancer Center, University of South Florida, 12902 Magnolia Drive, Tampa, FL 33612. E-mail address: djeu@moffitt.usf.edu kinases, Lck and Zap70 (15). These kinases, in turn, trigger downstream events via calcium- and Ras-dependent pathways. Zap70 3 Abbreviations used in this paper: MAPK, mitogen-activated protein kinase; FTI, farnesyl transferase inhibitor; ERK, extracellular regulated kinase 2; LGL, large gran- can phosphorylate a membrane-bound adaptor protein, linker for ular lymphocyte; PLC, phospholipase C; RBD, Ras-GTP binding domain; PI 3-ki- activation of T cells (LAT), which subsequently binds phospho- nase, phosphoinositide 3-kinase; MEK-1, MAPK/ERK kinase 1; TRITC, tetrameth- ␥ ␥ ylrhodamine isothiocyanate; RK, MAPKAP kinase 2 reactivity kinase; GRB2, growth lipase C 1 (PLC 1) and growth factor receptor binding protein 2 factor receptor binding protein 2. (GRB2) (16, 17). Activation of PLC␥1 leads to inositol phosphate

and intracellular calcium mobilization, while formation of the containing 10% FCS with 2 mM L-glutamine, 100 U/ml penicillin, and 100 GRB2-SOS complex recruits Ras to the membrane, turning on the ␮g/ml streptomycin (30). Raf/MEK/MAPK pathway (18, 19). MAPK is known to activate Isolation of large granular lymphocytes (LGL) from peripheral several important transcription factors required for IL-2 and other blood cytokine gene induction (20). The involvement of Vav and 76-kDa SH2 domain leukocyte protein in T cell signal pathways has also NK cells were derived from LGL isolated from PBMC of normal volunteers as previously described (4). After plastic adherence for1hat37°C been recently described, and the ability of 76-kDa SH2 domain and passage through nylon wool, the nonadherent mononuclear cells were leukocyte protein to bind Grb2 could provide another link to the placed on a four-step discontinuous Percoll gradient. The cells recovered Ras-dependent MAPK pathway (21, 22). from 42.5–45.0% Percoll usually contained 75–90% LGL, as assessed by Studies to date have thus documented that the conserved Ras/ Giemsa-stained morphology, and were used to test NK function. Raf/MEK/MAPK pathway is invoked by ligands binding to their Pharmacological inhibition of MAPK and Ras Ag receptors. This ﬁnding is expected, given that Ag recognition ϫ 6 leads to lymphocyte activation, resulting in gene expression related For MAPK inhibition, NK92 cells or fresh LGL (2.5 10 cells/ml) were incubated for2hat37°C with medium or serial dilutions of PD098059 to clonal expansion, a noted example of which is IL-2. However, (New England Biolabs, Beverly MA) or an equal amount of DMSO to once activated, T cells must express a separate and distinct func- dilute the highest concentration of inhibitor (31). Ras inhibition with a tion, i.e., the ability to lyse relevant target cells. In contrast to farnesyl transferase inhibitor, FTI-277, also dissolved in DMSO, was per- well-documented molecular signals that are coupled to cytokine formed under the same conditions, except the incubation period was ex- and proliferative responses in lymphocytes, information on mo- tended to 24 h (35, 36).

lecular events coupled to lytic effector function is scarce. NK cells Fixation of target cells Downloaded from use the same machinery as cytolytic T cells (23), and the avail- Raji tumor cells were washed with PBS once and incubated with 1% para- ability of lytic NK cell lines provides a convenient, useful tool in formaldehyde (methanol-free) in PBS, pH 7.4, on ice for 30 min. Then the deciphering these mechanisms. A recent interest in signal path- cells were washed four times with PBS to remove all paraformaldehyde. ways associated with lytic function has led to the identiﬁcation of Syk70, Vav, Rac, and Pyk2 as critical components (24–29). We Preparation of GST-Raf-RBD

have pinpointed MAPK/ERK2 as another key signal molecule in The GST fusion protein linked to the Raf fragment that contains the Ras- http://www.jimmunol.org/ the lytic pathway (30). We first demonstrated that target engage- GTP binding domain, GST-Raf-RBD, was constructed using pGEX-KT. ment rapidly activated MAPK phosphorylation and kinase func- GST-Raf-RBD was recovered from Escherichia coli by sonication and lysis with 1% Triton X-100. After centrifugation of the lysate at 15,000 tion (within 5 min) in human NK cells. To link ERK2 activation to rpm for 15 min at 4°C, the supernatant was collected and incubated for 15 NK biological function, we recently showed that PD098059, a spe- min at 4°C with glutathione agarose on a rotator. The agarose beads with cific inhibitor of MEK1 that lies directly upstream of MAPK (31), the bound GST-Raf-RBD were then collected and washed three times could potently suppress NK lysis of live Raji tumor cells in a 5-h before use. 51 Cr release assay. More importantly, immunostaining with fluo- Detection of Ras activation by GST-Raf-RBD rescence-conjugated anti-perforin or granzyme B indicated that PD098059 blocked the polarization of perforin and granzyme B Active Ras was measured by its ability to bind GST-Raf-RBD. Control by guest on September 28, 2021 NK92 cells or NK92 cells treated with DMSO or FTI-277 for 24 h at 37°C that occurs in NK cells upon target ligation. Transient transfection were rested in IL-2-free medium for4hat37°C to reduce background with dominant negative ERK2, but not wild-type ERK2, into NK phosphorylation. To initiate target ligation, the NK92 cells, suspended at cells effectively interfered with their lytic capacity. Thus, these 1 ϫ 107/ml, were mixed with an equal number of paraformaldehyde-fixed results documented a pivotal role of MAPK in NK effector func- Raji target cells and rapidly pelleted at 1000 rpm in a microcentrifuge at tion and demonstrated that ERK2 may regulate the events that 4°C, followed by incubation for 0–30 min at 37°C. For IL-2 activation, NK92 cells were incubated with 100 u/ml of IL-2 for 0–30 min at 37°C. mobilize perforin and granzyme B to the point of contact with the The cell pellet was solubilized at 4°C for 30 min in lysis buffer containing engaged target cell. Two other laboratories have also recently doc- 1% Nonidet P-40, 10 mM Tris, 140 mM NaCl, 0.1 mM PMSF, 10 mM umented the importance of MAPK in NK cells, either in direct or iodoacetamide, 50 mM NaF, 1 mM EDTA, 0.4 mM sodium orthovanadate, ␮ ␮ ␮ indirect CD16-mediated lytic function (32, 33). 10 g/ml leupeptin, 10 g/pepstatin, and 10 g/ml aprotinin. Cell lysates were centrifuged at 15,000 rpm in a microcentrifuge for 10 min at 4°C to In a number of models of cell growth and differentiation, the remove nuclei and cell debris. To detect activated Ras, the cell lysates were Ras-dependent MAPK pathway is inevitably involved (1, 2, 20). incubated for2hat4°Conarotator with 15 ␮l of the GST-Raf-RBD/ Whether this same pathway is called upon during the lytic process glutathione agarose bead preparation. The beads containing the bound Ras- in NK cells is unknown. We thus set out to examine whether target GTP were collected by centrifugation and washed three times with washing cells triggered a Ras-dependent MAPK pathway in NK cells and buffer (0.1% Nonidet P-40, 10 mM Tris, 140 mM NaCl, 0.1 mM PMSF, 10 mM iodoacetamide, 50 mM sodium fluoride, 1 mM EDTA, and 0.4 mM compared this pathway to that stimulated by IL-2. We demonstrate sodium orthovanadate). Samples were then boiled for 5 min in loading here that target ligation and IL-2 activated separate signal events to buffer and separated by 10% SDS-PAGE followed by Western blotting lead to MAPK activation, and target cells primarily triggered a with anti-pan- Ras (Transduction Laboratories, Lexington, KY). The pro- Ras-independent MAPK-dependent event in NK cells that drives teins were detected by the enhanced chemiluminescence detection system (ECL, Amersham, Arlington Heights, IL). lytic function. Detection of active MAPK Materials and Methods Cell lysates prepared from IL-2-rested NK92 cells that had been exposed Cells to fixed Raji tumor cells or IL-2 for 0–30 min at 37°C were resolved by gel electrophoresis. Active MAPK was detected by Western blotting with anti- Human IL-2-dependent NK92 cells, provided by Dr. H.-G. Klingeman active MAPK, which specifically recognizes the phosphorylated TEY mo- (Terry Fox Laboratory, Vancouver, Canada) (34), were maintained in tif in this kinase (New England Biolabs). Equal loading was assessed by ␣-MEM containing 100 U/ml of human rIL-2 (Chiron, Emeryville, CA). reblotting with anti-pan- ERK (Transduction Laboratories). This medium was also supplemented with 20% FCS, 2 mM glutamine, 100 U/ml penicillin, 100 ␮g/ml streptomycin, 1% nonessential amino acids, 1% Vaccinia viral delivery of N17Ras sodium pyruvate, and 5 ϫ 10Ϫ5 M 2-ME. All tissue culture reagents were purchased from Life Technologies (Grand Island, NY). The NK target Recombinant vaccinia viruses encoding wild-type H-Ras and N17-Ras cells, Raji lymphoma and K562 erythroleukemic tumor cells (American were constructed using the vector, pSC11, in recombination with the WR Type Culture Collection, Manassas, VA), were cultured in RPMI 1640 strain of vaccinia. CD56 expressing vaccinia was used as a control. In some The Journal of Immunology 3813 cases, GB-expressing vaccinia, provided by Dr. Andrew M. Scharenberg (Harvard Medical School, Boston, MA), was substituted as a control. For infection, NK92 cells were incubated with various vaccinia constructs for 1 h at 37°C in serum-free medium at a multiplicity of infection of 10. Cells were then further incubated in serum-containing medium for5hat37°C. Cytotoxicity assay A 51Cr release assay was performed as previously described, using Raji tumor cells as targets for NK92 effector cells and K562 tumor cells for fresh LGL (30). Briefly, target tumor cells were labeled with 200 ␮Ci of sodium [51Cr]chromate (Amersham, Arlington Heights, IL) in 0.2 ml of medium at 37°C in a 5% CO2 atmosphere for 1 h. The cells were then washed three times and added to effector cells at 5 ϫ 103 cells/well in 96-well round-bottom microplates, resulting in E:T cell ratios ranging from 50:1 to 2.5:1 in a final volume of 0.2 ml in each well. After 5-h incubation at 37°C, 100 ␮l of culture supernatants were harvested and counted in a gamma counter. The percent specific 51Cr release was determined by the equation ((experimental cpm Ϫ spontaneous cpm)/total cpm incorpo- rated) ϫ 100. All determinations were performed in triplicate, and the SEM FIGURE 1. Effect of PD098059 on NK92 lysis of tumor cells. NK92 of all assays was calculated and was typically around 5% of the mean cells, pretreated for 1 h at 37°C with 12.5–100 ␮M PD098059 or the or less. amount of DMSO used to dilute the highest concentration of PD098059,

Immunostaining were tested in triplicate wells at the indicated E:T cell ratios for lysis of Downloaded from 51Cr-labeled Raji tumor cells. The SEM of each percent cytotoxicity value NK92 cells, untreated or pretreated with 50 ␮M PD098059 for1hat37°C was Ͻ5% of the mean and is not shown. or 15 ␮M FTI-277 for 24 h at 37°C, were added to Raji cells at a 1:1 ratio in a total volume of 100 ␮l. The cells were spun rapidly at 1000 rpm for 1 min in a cold microcentrifuge, and then incubated for 0–10 min at 37°C. DMSO-treated NK92 cells were included as a control. The cells were then tration used to dissolve the highest concentration of PD098059, centrifuged onto a microscope slide and fixed at Ϫ20°C with methanol/ acetone (3/1) for 20 min (30). The slides were air-dried and rehydrated for had little adverse effect on NK lysis. http://www.jimmunol.org/ 2 h in several changes of PBS. All procedures were performed at room We next examined whether contact with tumor target cells could temperature. Polyclonal rabbit anti-human IgM (Sigma, St. Louis, MO) trigger MAPK activation in NK92 cells. For detection of active was used to differentiate Raji B lymphocytic cells from NK92 cells. Mono- MAPK, we first rested NK92 cells in IL-2-free medium for4hto clonal anti-human perforin (Endogen/T Cell Sciences, Woburn, MA) and anti-granzyme B (37) were used to detect these lytic components from reduce background phosphorylation in the IL-2-dependent cells. NK92 cells. Anti-IgM together with anti-perforin (or anti-granzyme B), The rested NK92 effector cells were then added to fixed Raji tumor each diluted 1/200 with 0.1% Nonidet P-40 in 1% BSA in PBS, were cells at a 1:1 ratio, pelleted at 1000 rpm for 30 s at 4°C, and applied to the slide for 1 h. After several washes with PBS for 2 h, the incubated for 0–30 min at 37°C. Paraformaldehyde fixation of slides were incubated for 25 min with goat anti-rabbit IgG tetramethylrho- target cells ensured that all detectable active MAPK originated damine isothiocyanate (TRITC)-labeled Ab (Sigma) diluted 1/80 or goat anti-mouse Ig FITC-labeled Ab (Sigma), diluted 1/100 in 0.1% Nonidet from NK92 cells in the cell mixture. Western blot analysis of the by guest on September 28, 2021 P-40 in PBS containing 1% BSA. The slides were then washed several lysates prepared from these cells with anti-active MAPK indicated times with PBS and covered with coverslips in a mounting medium of that a transient, but rapid, MAPK activation took place at 5 min in antifade/DAPI. Immunofluorescence was observed with a Leitz Orthoplan NK cells after target ligation (Fig. 2A). PD098059, but not DMSO, 2 microscope (Rockleigh, NJ), and images were captured by a CCD camera with the Smart Capture Program (Vysis, Downers Grove, IL). On each pretreatment for 1 h was able to inhibit this event in NK cells (Fig. slide 100 NK92/Raji conjugates were evaluated for perforin or granzyme B 2B). Analysis of total MAPK by stripping of the original gel and mobilization. reblotting with anti-pan-ERK demonstrated that MAPK was Several controls were performed, i.e., NK92 cells alone or Raji tumor equally present in all lanes. cells alone stained only with FITC-labeled goat anti-mouse Ig or with TRITC-labeled goat anti-rabbit IgG, to check for nonspecific binding of the Ras activation by target ligation in NK cells secondary Abs. Nonspecific binding was not detected, and the results were omitted from the figures for clarity. We next undertook the task of examining whether Ras was involved. The first attempt was to define whether Ras was activated Results by target ligation in NK92 cells. Inactive Ras binds GDP, which is MAPK activation by target ligation in NK cells exchanged for GTP upon activation, acquiring the ability to bind Our earlier work had employed the NK tumor cell line, YT, to Raf. To pursue Ras activation, we thus employed a GST fusion establish the importance of MAPK (30). To carry out this work protein that is linked to the Raf protein segment that contains the further, we have turned to a widely used IL-2-dependent NK cell activated GST-Raf-RBD (40). Active Ras could then be captured line, NK92 (34, 38, 39). This move was taken because NK92 cells by binding to GST-Raf-RBD and detected with anti-pan-Ras. retain all the biological properties of normal NK cells and possess NK92 cells, IL-2-starved for 4 h, were incubated with fixed Raji a markedly higher capacity to lyse Raji tumor cells than YT cells, tumor cells for 0–30 min (Fig. 3). As a positive control, a separate ranging from about 30% at an E:T cell ratio of 5:1, while YT group of NK92 cells was stimulated with IL-2, which is docu- effector cells only killed around 30% at 50:1. We thus needed to mented to activate Ras. Cell lysates were then prepared and incu- re-establish the baseline that MAPK critically controls lytic func- bated for 2 h with GST-Raf-RBD precoupled to glutathione- tion in NK92 cells. NK92 cells were pretreated with 12.5–100 ␮M Sepharose beads. Western blot analysis of the proteins bound to PD098059, a MAPK pathway specific inhibitor, for1hat37°C GST-Raf-RBD demonstrated that Ras was quickly activated before incubation with 51Cr-labeled Raji tumor cells for5hat within 2–5 min of target ligation in NK92 cells; similarly, IL-2 37°C to test for lysis. PD098059 pretreatment effectively inhibited treatment activated Ras, which peaked at 5–15 min and was still NK92 lysis of Raji tumor cells in a dose-dependent manner, as detectable at 30 min. Stripping and reblotting with anti-active shown in a representative experiment of four that were performed MAPK showed that MAPK activation followed the same rapid (Fig. 1). Marked inhibition was observed with 25 ␮M, with com- kinetics with target ligation and markedly longer kinetics with IL-2 plete inhibition attained at 100 ␮M. DMSO, tested at the concen- (data not shown). 3814 RAS-INDEPENDENT MAPK REQUIRED FOR NK LYSIS

FIGURE 3. Detection of Ras activation with GST-Raf-RBD in NK92 cells. NK92 cells, IL-2-starved for4hat37°C, were mixed with equal numbers of ﬁxed Raji tumor cells or with 100 U/ml of IL-2 for 0–30 min at 37°C. Cell lysates were then prepared and adsorbed with GST-Raf-RBD precoupled to glutathione-agarose beads for 2 h. The proteins bound to GST-Raf-RBD were resolved by gel electrophoresis and probed for the presence of active Ras-GTP by Western blotting with anti-pan-Ras. Whole cell lysates (WCL) from NK92 cells were included as a positive control. Downloaded from

Lack of effect of FTI-277 on perforin and granzyme B polarization in NK cells Perforin and granzyme B mobilization in NK cells is a hallmark of FIGURE 2. Analysis of active MAPK in NK92 cells after target liga- cytolytic events against target cells, and we have earlier demon-

tion. A, NK92 cells, rested in IL-2-free medium for4hat37°C, were strated that MAPK/ERK2 is critical for their movement (30). http://www.jimmunol.org/ mixed with equal numbers of fixed Raji tumor cells for 0–30 min at 37°C. Whether Ras is required for this event thus needs to be evaluated. Cell lysates were then prepared and analyzed by Western blotting with NK92 cells, pretreated 24 h at 37°C with 15 ␮M FTI-277 or anti-active MAPK (upper panel). The same membrane was stripped and DMSO at the concentration used to dissolve the Ras inhibitor, reprobed with anti-pan- ERK to check for equal loading (lower panel). B, were incubated with Raji tumor cells for 0–10 min at 37°C. NK92 ␮ NK92 cells, IL-2-starved for4hat37°C, were treated with 10–100 M cells were also pretreated for1hat37°C with 50 ␮M PD098059 PD098050 or the amount of DMSO used to dilute the highest concentration to inhibit MAPK activation before target ligation. The cells were of PD098059 for1hat37°C. The cells were then incubated with equal numbers of fixed Raji tumor cells for 0–5 min at 37°C before Western blot then cytocentrifuged onto slides for assessment of perforin and analysis with anti-active MAPK (upper panel). The same membrane was granzyme B polarization. Perforin or granzyme B, contained only stripped and reprobed with anti-pan- ERK to check for equal loading (low- in NK92 cells and not Raji tumor cells, was detected by FITC by guest on September 28, 2021 er panel). staining with specific mAbs. Raji tumor cells could be differentiated from NK92 cells by TRITC staining with anti-human IgM. Immunostaining demonstrated that tumor engagement for only 10 Lack of effect of farnesyl transferase inhibitor, FTI-277, on lytic min mobilized perforin and granzyme B unidirectionally toward function and MAPK activation in NK cells the target cell, and a representative NK92/Raji conjugate is shown It was essential to next link Ras activation to lytic function in under each treatment in Fig. 5. At 0 min of target binding, FITC- NK92 cells. An effective method to interfere with Ras function is labeled perforin was evenly distributed in the cytoplasm of NK92 to block its farnesylation, preventing it from localizing to the cells (Fig. 5A, upper panel). Upon 10-min incubation at 37°C, the ϩ plasma membrane and becoming functionally active. A specific DMSO-treated NK92 cell that had bound TRITC-labeled IgM farnesyl transferase inhibitor, FTI-277, well characterized to in- Raji target cells now showed almost complete polarization of hibit tumor cell growth in vitro and in vivo (35, 36), was thus FITC-labeled perforin at the site of tumor contact (Fig. 5B, upper employed. NK92 cells were treated for 24 h at 37°C with 7.5–30 panel). However, FTI-277 had no ability to interfere with perforin ␮M FTI-277, which represents the optimal range of concentrations polarization in NK92 cells (Fig. 5C, upper panel). PD098059, known to effectively inactivate Ras and inhibit tumor cell growth which specifically inhibited MAPK activation, clearly interrupted (Fig. 4A). The NK effector cells were then tested for NK function. this process (Fig. 5D, upper panel). Similar observations were It was notable that FTI-277, at all doses used, had no effect on NK made with granzyme B (Fig. 5, lower panel). Enumeration of NK/ lysis of 51Cr-labeled Raji tumor cells. This was reproduced in three Raji conjugates showed that the percentage of conjugates with po- other experiments. We also tested FTI-277 on freshly isolated larized perforin rose from 4.8 to 32% with 10 min of binding. human peripheral blood LGL and found in four donors tested that PD098059 decreased the percentage of polarized conjugates to it had no ability to interfere with NK lysis of tumor target cells 16.9%, but FTI-277-treated conjugates remained at 34%. Similar (Fig. 4B). percentages were obtained with granzyme B. Confocal microscopy We next examined whether FTI-277 could interfere with MAPK showed that all perforin was located at the contact point by 10 min activation (Fig. 4C). The same aliquots of NK92 cells from Fig. (data not shown). Thus, Ras does not appear to be essential for 4A, pretreated for 24 h with FTI-277, were IL-2-starved for 4 h granule movement in NK92 cells, but MAPK is. before incubation for 0–5 min at 37°C with fixed Raji tumor cells. Cell lysates prepared from these treatments were then probed with FTI-277 inhibits Ras farnesylation and activation in NK cells anti-active MAPK. Within 5 min of incubation with Raji tumor It is critical to ensure, however, that FTI-277 is able to block Ras cells, NK92 cells, either untreated or treated with DMSO or FTI- activation in our experiments. We thus treated NK92 cells with or 277, had equal ability to activate MAPK. FTI-277 therefore did not without FTI-277 for 24 h at 37°C, rested the NK92 cells in IL-2- block MAPK activation in NK92 cells starved medium for4hat37°C, and then activated them for 5 min The Journal of Immunology 3815

inhibited Ras farnesylation in NK92 cells (detected by a slower migrating unprenylated band in Western blots using anti-H-Ras), but did not interfere with Rap1A geranylation in NK92 cells (Fig. 6B). Thus, our results indicated that FTI-277 effectively inhibited Ras farnesylation and activation in NK cells, but could not interfere with MAPK activation or tumor cell lysis.

Inhibition of IL-2 or target cell-driven Ras activation by N17 Ras expression in NK cells To definitely determine whether MAPK activation and lytic function were independent of Ras in NK cells, we resorted to a molecular approach. We delivered dominant negative N17 Ras into NK92 effector cells via a vaccinia virus vector and assessed, first, Ras activity in the infected cells before analysis of MAPK activity and tumoricidal function. Infection was conducted for 6 h, which provided sufficient time for ample production of proteins from viral delivery without loss of cell viability. After viral infection, NK92 cells were then IL-2-starved 4 h before activation for 0–5

min with either IL-2 or ﬁxed Raji tumor cells. Wild-type Ras and Downloaded from an irrelevant gene, CD56, were introduced into separate NK92 cell pools as controls. Expression of the irrelevant gene, CD56, had no effect on Ras activation in NK92 cells, as measured by Ras-GTP binding to GST-Raf-RBD, and the levels of Ras activation by IL-2 and Raji tumor cells were similar to those by mock-infected con-

trol NK92 (Fig. 7). Expression of wild-type Ras caused a signif- http://www.jimmunol.org/ icant production of activated Ras even without IL-2 or Raji treatment. N17 Ras expression, in contrast, caused loss of detection of activated Ras in both IL-2- and Raji-stimulated NK92 cells. Thus, we have a working system to assess what fate MAPK will have in a Ras-defective NK92 cell.

Effect of dominant negative Ras expression on MAPK activation and lytic function in NK cells

Assessment of active MAPK in N17 Ras-expressing NK92 cells by guest on September 28, 2021 was next performed (Fig. 8A). NK92 cells, mock-infected or carrying the irrelevant gene, GB, showed similar enhanced levels of MAPK activated by Raji tumor cells. NK92 expressing wild-type FIGURE 4. Effect of FTI-277 on lytic function and MAPK activation in Ras had active MAPK even before Raji stimulation. Most impor- NK92 cells. A, NK92 cells, pretreated for 24 h at 37°C with 7.5–30 ␮M tantly, N17 Ras expressing NK92 cells showed equal ability as FTI-277 or the amount of DMSO used to dilute the highest concentration normal NK92 cells to activate MAPK upon target ligation. Thus, of FTI-277, were tested for lysis of 51Cr-labeled Raji tumor cells. B,NK MAPK activation triggered by target ligation is via a Ras-inde- cells, freshly isolated from the LGL population of peripheral blood from pendent mechanism. Using the same pool of cells, we tested their four normal volunteers, were pretreated with 15 ␮M FTI-277 or DMSO for tumorical capacity (Fig. 8B). We found that N17 Ras expressing 24 h at 37°C before testing for lytic function against 51Cr-labeled K562 NK92 cells had almost equal ability to lyse 51Cr-labeled tumor tumor cells. The results shown are from E:T cell ratios of 60:1, but similar cells as mock-infected and GB-infected NK92 cells. We also made results were obtained at lower ratios down to 7.5:1. The SEM of each the observation that overexpression of wild-type Ras consistently percent cytotoxicity value was Ͻ5% of the mean, and was not shown. C, caused a moderate decrease in NK lytic function. These results The same aliquots of NK92 cells from A were prepared as cell lysates for analysis by Western blotting with anti-active MAPK (upper panel), fol- were reproduced in two other experiments. It has been reported lowed by reprobing with anti-pan-ERK to check for equal loading (lower that MAPK activation by IL-2 in T cells such as Jurkat is a Ras- panel). dependent MAPK event. To ensure that our system was working appropriately, we also examined this process in NK92 cells. As in T cells, expression of N17 Ras completely abolished MAPK ac- at 37°C with either fixed Raji tumor cells or 500 U/ml of IL-2. IL-2 tivation in IL-2-stimulated NK cells (Fig. 9). Unlike target cell was included as a positive control. Cell lysates were prepared and ligation, IL-2 caused a Ras-dependent MAPK event. Thus, various incubated for 2 h with GST-Raf-RBD precoupled to glutathione- pathways exist to activate MAPK in NK cells, and we have evi- Sepharose beads, and the adsorbed complexes were resolved by dence to indicate that NK lytic function, activated by target en- gel electrophoresis. Detection of bound Ras-GTP by Western blot gagement, is controlled by a Ras-independent MAPK signal analysis with anti-pan-Ras indicated that both IL-2 and Raji tumor pathway. cells readily activated Ras (Fig. 6A). FTI-277 was able to inhibit this activation in a dose-dependent manner. To show specificity Discussion and effectiveness of FTI-277 in inhibiting only farnesyl trans- Cytotoxicity is a key effector function of NK cells, yet the molec- ferase, but not geranyl-geranyl transferase, we next performed ular mechanism driving it is unclear. Upon engagement of an ap- Western blot analysis of NK92 cells with anti-H-Ras and anti- propriate target cell, NK cells can rapidly mobilize lytic granule Rap1A after FTI-277 treatment. FTI-277 at 15 ␮g/ml specifically components, such as perforin and granzyme B, to the contact zone 3816 RAS-INDEPENDENT MAPK REQUIRED FOR NK LYSIS Downloaded from http://www.jimmunol.org/

FIGURE 5. Effects of FTI-277 and PD098059 on perforin and granzyme polarization in NK92 cells. NK92 cells were pretreated at 37°C either for 24 h with 15 ␮M FTI-277 or DMSO or for 1 h with 50 ␮M PD098059. The cells were then mixed with equal numbers of Raji tumor cells for 0–10 min at 37°C and cytospun onto microscope slides for staining with FITC-anti-perforin and TRITC-anti-IgM (upper panel) or with FITC-anti-granzyme B and TRITC- anti-IgM (lower panel).

to initiate cell lysis by caspase-dependent (41) and/or caspase-in- cells. Irrelevant CD56 gene delivery, used as a control, had no dependent pathways (42). Based on our results, MAPK is a critical adverse effect on Ras activation. Of key importance was that al- by guest on September 28, 2021 component in direct NK lytic function, and it is essential for driv- though N17-Ras was able to suppress Ras, it could not interfere ing perforin and granzyme B polarization in NK cells upon en- with MAPK activation in NK92 cells triggered by Raji tumor cells. gagement of a target cell. The goal of the present study was to To ensure that N17-Ras could block a Ras-dependent MAPK path- further define this signal pathway and to examine whether Ras was way, we also evaluated its effect on IL-2 activation in NK92 cells. a key upstream effector for MAPK activation in NK cells. Using IL-2 activation of MAPK was completely dependent on intact Ras. tumor cells and IL-2 as separate triggers, we identified divergent These results suggest that the tumoricidal process triggered in NK pathways used by NK92 cells to activate MAPK. Whereas IL-2 cells by target ligation is controlled via a Ras-independent MAPK- triggered a Ras-dependent MAPK pathway, tumor cells essentially dependent signal event. It is thus important to note that lympho- did not require Ras for MAPK activation. This Ras-independent cytes may actually use separate signal pathways in the initial ac- MAPK signal critically controlled lytic function and perforin/gran- tivation phase and the later effector phase. Initially, Ag ligation zyme B polarization in NK cells during target lysis. This conclu- may drive a Ras-dependent MAPK mechanism that results in gene sion was supported by results obtained by biochemical, molecular, expression that is required for cell proliferation and differentiation and biological means. Using a pharmacological reagent, FTI-277, (1, 2, 20). Later Ag ligation in the effector lymphocyte calls upon which inhibits farnesyl transferase, we determined whether FTI- a Ras-independent MAPK mechanism, which now rapidly mobi- 277 inhibition of Ras farnesylation could interfere with NK func- lizes lytic components, such as perforin and granzyme B, without tion. Target ligation readily activated Ras in NK92 cells, as mea- the need for new gene expression. Another key difference is that sured by capture of Ras-GTP with GST-Raf RBD. We showed, IL-2-driven Ras/MAPK activation persists for up to 30 min, while however, that the range of FTI-277 concentrations that was suffi- target-induced MAPK activation is very transient. This suggests cient to prevent Ras activation did not interfere with the ability of that a longer period of MAPK activation is required for the induc- NK cells to lyse 51Cr-labeled tumor cells or activate MAPK upon tion of cell cycle control-related gene expression in response to tumor target engagement. FTI-277 also could not inhibit perforin IL-2, as opposed to a short burst of MAPK activation for rapid or granzyme B polarization in NK92 cells during tumor cell en- granule mobilization, which is accomplished within 5 min upon gagement. In contrast, PD098059, which inactivates MAPK func- contact with target tumor cells. tion, could readily disrupt granule polarization. Several candidates could serve in place of Ras to activate Pharmacological inhibition of Ras is one means to analyze Ras MAPK in the lytic process. Phosphoinositol 3-kinase (PI 3-kinase) participation in NK function. To definitely identify the role of Ras interfaces with Ras in insulin and G protein-coupled receptor sig- in MAPK activation and lytic function, we undertook the genetic naling (43, 44), but in integrin- or IL-2-mediated signaling, PI approach of introducing dominant negative N17-Ras into NK92 3-kinase can act downstream of Ras to regulate MAPK (45, 46). cells via vaccinia delivery. Viral delivery of N17-Ras abolished Syk70 overexpression in JCAM1 Jurkat T cells can enable TCR Ras activation in NK92 cells triggered by either IL-2 or Raji tumor cross-linking to activate MAPK/ERK2, but two distinct pathways The Journal of Immunology 3817 Downloaded from http://www.jimmunol.org/

FIGURE 6. Inhibition of Ras farnesylation and activation by FTI-277 in NK92 cells. A, NK92 cells, pretreated with 7.5–30 ␮M FTI-277 or the FIGURE 8. Effect of N17 Ras expression on MAPK activation and lytic amount of DMSO used to dissolve the highest concentration of FTI-277, function triggered by target ligation in NK92 cells. A, NK92 cells, mock- were IL-2-starved for4hat37°C and then incubated with equal numbers infected or infected for6hat37°C with vaccinia viral vectors encoding of ﬁxed Raji tumor cells or 100 U/ml of IL-2, for 0–5 min at 37°C. Cell GB, wild-type Ras, or N17 Ras, were IL-2-starved for4hat37°C. The lysates were prepared for adsorption to GST-Raf-RBD for 2 h, followed by cells were then incubated with equal numbers of ﬁxed Raji tumor cells for resolution of the adsorbed active Ras-GTP by gel electrophoresis and 0–5 min at 37°C before Western blot analysis of the cell lysates with Western blot analysis with anti-pan-Ras. B, Whole cell lysates from NK92 anti-active MAPK (upper panel). Reblotting with anti-pan-ERK was per- cells, similarly treated with 15 ␮M FTI-277 or DMSO, were analyzed by formed to check for equal presence of the MAPK proteins (lower panel). by guest on September 28, 2021 Western blotting with anti-H-Ras to detect the farnesylated (faster migrat- B, Aliquots of the same NK92 cells from A were tested for lytic function ing band) and unfarnesylated (slower migrating band) forms of Ras (upper against 51Cr-labeled Raji tumor cells. panel). Aliquots of these cells were also Western blotted with anti-rap 1A to detect the geranylated and ungerylated forms of rap 1A (lower panel). possible that either of these molecules could control MAPK in NK cells during the lytic process. can emerge (47, 48). For MAPK activation, Lck-negative Jurkat T Other potential pathways include protein kinase C and protein cells require Ras, but normal Jurkat cells do not. Thus, Syk70 can kinase A. Ca-dependent isoforms of protein kinase C control potentially bypass Ras to activate MAPK. Because PI 3-kinase can be triggered by either CD94 or FcR cross-linking in NK cells (49, 50), and Syk70 is critical for signaling NK activation (24–26), it is

FIGURE 7. Efﬁciency of N17 Ras expression in inhibiting Ras activa- FIGURE 9. Inhibition of IL-2-driven MAPK activation in NK92 cells tion in NK92 cells. NK92 cells were mock-infected or infected for6hat by N17 Ras expression. NK92 cells, mock-infected or infected with GB, 37°C with vaccinia viral vectors carrying CD56, wild-type Ras, or N17 wild-type Ras, or N17 Ras, were IL-2-starved for4hat37°C. The cells Ras. The cells were then IL-2-starved for4hat37°C and incubated for 0–5 were then incubated with or without 100 U/ml of IL-2 for 5 min at 37°C min with either equal numbers of ﬁxed Raji tumor cells or 100 U/ml of before Western blot analysis of the cell lysates with anti-active MAPK IL-2. Cell lysates were then prepared and probed for active Ras-GTP by (upper panel). Reblotting with anti-pan-ERK was performed to check for binding to GST-Raf-RBD. equal presence of the MAPK proteins (lower panel). 3818 RAS-INDEPENDENT MAPK REQUIRED FOR NK LYSIS

MAPK by bypassing Ras to activate Raf-1 (51–53), while protein 1994. A complex of Grb2 adaptor protein, Sos exchange factor, and a 36-kDa kinase A uses Rap1 (instead of Ras) to activate Raf, resulting in membrane-bound tyrosine phosphoprotein is implicated in Ras activation in T cells. J. Biol. Chem. 269:9019. MEK activation and subsequent MAPK activation (54, 55). Work 19. Franklin, R. A., A. Tordai, H. Patel, A. M. Gardner, G. L. Johnson, and is being planned to determine whether there is a specific upstream E. W. Gelfand. 1994. Ligation of the T cell receptor complex results in activation of the Ras/Raf-1/MEK/MAPK cascade in human T lymphocytes. J. Clin. Invest. pathway in NK cells that is critical for MAPK activation to attain 93:2134. functional lysis of tumor cells. 20. Cantrell, D. 1996. T cell antigen receptor signal transduction pathways. Annu. In terms of the clinical significance of our study, the finding that Rev. Immunol. 14:259. 21. Motto, D. G., S. E. Ross, J. Wu, L. R. Hendricks-Taylor, and G. A. Koretzky. NK cells use a Ras-independent pathway to kill tumor cells might 1996. Implication of the GRB2-associated phosphoprotein SLP-76 in T cell re- be of therapeutic advantage. A major recent focus in pharmaceu- ceptor-mediated interleukin 2 production. J. Exp. Med. 183:1937. ticals is in the targeting of oncogenes that are mutated in cancer. 22. Wu, J., D. G. Motto, G. A. Koretzky, and A. Weiss. 1996. Vav and SLP-76 interact and functionally cooperate in IL-2 gene activation. Immunity 4:593. Ras has been a prime target for the development of new therapeu- 23. Froelich, C. J., V. M. Dixit, and X. Yang. 1998. Lymphocyte granule-mediated tic agents, because of the high frequency of Ras mutations in hu- apoptosis: matters of viral mimicry and deadly proteases. Immunol. Today 19:30. man tumors (36). If Ras is not involved in NK lytic function, a 24. Brumbaugh, K. M., B. A. Binstadt, D. D. Billadeau, R. A. Schoon, C. J. Dick, R. M. Ten, and P. J. Leibson. 1997. Functional role for Syk tyrosine kinase in rational drug design targeted specifically against Ras could gener- natural killer cell-mediated natural cytotoxicity. J. Exp. Med. 186:1965. ate new therapeutic agents that might spare a key immune system 25. Lanier, L. L., B. Corliss, J. Wu, and J. H. Phillips. 1998. Association of DAP12 with activating CD94/NKG2C NK cell receptors. Immunity 8:693. in cancer patients. 26. Lanier, L. L., B. C. Corliss, J. Wu, C. Leong, and J. H. Phillips. 1998. Immu- noreceptor DAP12 bearing a tyrosine-based activation motif is involved in activating NK cells. Nature 391:703. 27. Billadeau, D. D., K. M. Brumbaugh, C. J. Dick, R. A. Schoon, X. R. Bustelo, and Acknowledgments P. J. Leibson. 1998. The Vav-Rac1 pathway in cytotoxic lymphocytes regulates Downloaded from We thank the staff of the Analytical Microscopy Core and the Molecular the generation of cell-mediated killing. J. Exp. Med. 188:549. Imaging Core facilities of the H. Lee Moffitt Cancer Center for their tech- 28. Galandrini, R., G. Palmieri, M. Piccoli, L. Frati, and A. Santoni. 1999. Role for nical assistance. the Rac1 exchange factor Vav in the signaling pathways leading to NK cell cytotoxicity. J. Immunol. 162:3148. 29. Gismondi, A., J. Jacobelli, F. Mainiero, R. Paolini, M. Piccoli, L. Frati, and A. Santoni. 2000. Cutting edge: functional role for proline-rich tyrosine kinase 2 References in NK cell-mediated natural cytotoxicity. J. Immunol. 164:2272. 1. Davis, R. J. 1994. MAPKs: new JNK expands the group. Trends Biochem. Sci. 30. Wei, S., A. M. Gamero, J. H. Liu, A. A. Daulton, N. I. Valkov, J. A. Trapani, http://www.jimmunol.org/ 19:470. A. C. Larner, M. J. Weber, and J. Y. Djeu. 1998. Control of lytic function by mitogen-activated protein kinase/extracellular regulatory kinase 2 (ERK2) in a 2. Johnson, G. L., and R. R. Vaillancourt. 1994. Sequential protein kinase reactions human natural killer cell line: identification of perforin and granzyme B mobi- controlling cell growth and differentiation. Curr. Opin. Cell Biol. 6:230. lization by functional ERK2. J. Exp. Med. 187:1753. 3. Seth, A., E. Alvarez, S. Gupta, and R. J. Davis. 1991. A phosphorylation site 31. Alessi, D. R., A. Cuenda, P. Cohen, D. T. Dudley, and A. R. Saltiel. 1995. PD located in the NH2-terminal domain of c-Myc increases transactivation of gene 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase expression. J. Biol. Chem. 266:23521. kinase in vitro and in vivo. J. Biol. Chem. 270:27489. 4. Pulverer, B. J., J. M. Kyriakis, J. Avruch, E. Nikolakaki, and J. R. Woodgett. 32. Milella, M., A. Gismondi, P. Roncaioli, L. Bisogno, G. Palmieri, L. Frati, 1991. Phosphorylation of c-Jun mediated by MAP kinases. Nature 353:670. M. G. Cifone, and A. Santoni. 1997. CD16 cross-linking induces both secretory 5. Gille, H., M. Kortenjann, O. Thomae, C. Moomaw, C. Slaughter, M. H. Cobb, and extracellular signal-regulated kinase (ERK)-dependent cytosolic phospho- and P. E. Shaw. 1995. ERK phosphorylation potentiates Elk-1-mediated ternary lipase A2 (PLA2) activity in human natural killer cells: involvement of ERK, but complex formation and transactivation. EMBO J 14:951. not PLA2, in CD16-triggered granule exocytosis. J. Immunol. 158:3148. by guest on September 28, 2021 6. Nakajima, T., S. Kinoshita, T. Sasagawa, K. Sasaki, M. Naruto, T. Kishimoto, 33. Trotta, R., K. A. Puorro, M. Paroli, L. Azzoni, B. Abebe, L. C. Eisenlohr, and and S. Akira. 1993. Phosphorylation at threonine-235 by a Ras-dependent mito- B. Perussia. 1998. Dependence of both spontaneous and antibody-dependent, gen-activated protein kinase cascade is essential for transcription factor NF-IL6. granule exocytosis-mediated NK cell cytotoxicity on extracellular signal-regu- Proc. Natl. Acad. Sci. USA 90:2207. lated kinases. J. Immunol. 161:6648. 7. Cheng, J. T., M. H. Cobb, and R. Baer. 1993. Phosphorylation of the TAL1 34. Klingemann, H. G., and B. Miyagawa. 1996. Purging of malignant cells from oncoprotein by the extracellular-signal-regulated protein kinase ERK1. Mol. Cell. blood after short ex vivo incubation with NK-92 cells. Blood 87:4913. Biol. 13:801. 35. Lerner, E. C., Y. Qian, M. A. Blaskovich, R. D. Fossum, A. Vogt, J. Sun, A. D. Cox, C. J. Der, A. D. Hamilton, and S. M. Sebti. 1995. Ras CAAX pep- 8. Lin, L. L., M. Wartmann, A. Y. Lin, J. L. Knopf, A. Seth, and R. J. Davis. 1993. tidomimetic FTI-277 selectively blocks oncogenic Ras signaling by inducing cPla2 is phosphorylated and activated by Map kinase. Cell 72:269. cytoplasmic accumulation of inactive Ras-Raf complexes. J. Biol. Chem. 270: 9. Sturgill, T. W., L. B. Ray, E. Erikson, and J. L. Maller. 1988. Insulin-stimulated 26802. MAP-2 kinase phosphorylates and activates ribosomal protein S6 kinase II. Na- 36. Sebti, S. M., and A. D. Hamilton. 1997. Inhibition of Ras prenylation: a novel ture 334:715. approach to cancer chemotherapy. Pharmacol. Ther. 74:103. 10. Dubois, M. F., V. T. Nguyen, M. E. Dahmus, G. Pages, J. Pouyssegur, and 37. Trapani, J. A., K. A. Browne, M. Dawson, and M. J. Smyth. 1993. Immunopu- O. Bensaude. 1994. Enhanced phosphorylation of the C-terminal domain of RNA rification of functional Asp-ase (natural killer cell granzyme B) using a mono- polymerase II upon serum stimulation of quiescent cells: possible involvement of clonal antibody. Biochem. Biophys. Res. Commun. 195:910. MAP kinases. EMBO J. 13:4787. 38. Burshtyn, D. N., A. M. Scharenberg, N. Wagtmann, S. Rajagopalan, K. Berrada, 11. Rane, M. J., S. L. Carrithers, J. M. Arthur, J. B. Klein, and K. R. McLeish. 1997. T. Yi, J. P. Kinet, and E. O. Long. 1996. Recruitment of tyrosine phosphatase Formyl peptide receptors are coupled to multiple mitogen-activated protein ki- HCP by the killer cell inhibitor receptor. Immunity 4:77. nase cascades by distinct signal transduction pathways: role in activation of re- 39. Pazmany, L., O. Mandelboim, M. Vales-Gomez, D. M. Davis, H. T. Reyburn, duced nicotinamide adenine dinucleotide oxidase. J. Immunol. 159:5070. and J. L. Strominger. 1996. Protection from natural killer cell-mediated lysis by 12. Pillinger, M. H., A. S. Feoktistov, C. Capodici, B. Solitar, J. Levy, T. T. Oei, and HLA-G expression on target cells. Science 274:792. M. R. Philips. 1996. Mitogen-activated protein kinase in neutrophils and enucle- 40. de Rooij, J., and J. L. Bos. 1997. Minimal Ras-binding domain of Raf1 can be ate neutrophil cytoplasts: evidence for regulation of cell-cell adhesion. J. Biol. used as an activation-specific probe for Ras. Oncogene 14:623. Chem. 271:12049. 41. Andrade, F., S. Roy, D. Nicholson, N. Thornberry, A. Rosen, and 13. Schlaepfer, D. D., S. K. Hanks, T. Hunter, and P. van der Geer. 1994. Integrin- L. Casciola-Rosen. 1998. Granzyme B directly and efficiently cleaves several mediated signal transduction linked to Ras pathway by GRB2 binding to focal downstream caspase substrates: implications for CTL-induced apoptosis. Immu- adhesion kinase. Nature 372:786. nity 8:451. 14. Cox, M. E., and S. J. Parsons. 1997. Roles for protein kinase C and mitogen- 42. Sarin, A., M. S. Williams, M. A. Alexander-Miller, J. A. Berzofsky, activated protein kinase in nicotine-induced secretion from bovine adrenal chro- C. M. Zacharchuk, and P. A. Henkart. 1997. Target cell lysis by CTL granule maffin cells. J. Neurochem. 69:1119. exocytosis is independent of ICE/Ced-3 family proteases. Immunity 6:209. 43. Yamauchi, K., K. Holt, and J. E. Pessin. 1993. Phosphatidylinositol 3-kinase 15. Qian, D., and A. Weiss. 1997. T cell antigen receptor signal transduction. Curr. functions upstream of Ras and Raf in mediating insulin stimulation of c-fos tran- Opin. Cell Biol. 9:205. scription. J. Biol. Chem. 268:14597. 16. Sieh, M., A. Batzer, J. Schlessinger, and A. Weiss. 1994. GRB2 and phospho- 44. Lopez-Ilasaca, M., P. Crespo, P. G. Pellici, J. S. Gutkind, and R. Wetzker. 1997. ␥ lipase C- 1 associate with a 36- to 38-kilodalton phosphotyrosine protein after Linkage of G protein-coupled receptors to the MAPK signaling pathway through T-cell receptor stimulation. Mol. Cell. Biol. 14:4435. PI 3-kinase ␥. Science 275:394. 17. Zhang, W., J. Sloan-Lancaster, J. Kitchen, R. P. Trible, and L. E. Samelson. 1998. 45. King, W. G., M. D. Mattaliano, T. O. Chan, P. N. Tsichlis, and J. S. Brugge. LAT: the ZAP-70 tyrosine kinase substrate that links T cell receptor to cellular 1997. Phosphatidylinositol 3-kinase is required for integrin-stimulated AKT and activation. Cell 92:83. Raf-1/mitogen-activated protein kinase pathway activation. Mol. Cell. Biol. 17: 18. Buday, L., S. E. Egan, P. Rodriguez Viciana, D. A. Cantrell, and J. Downward. 4406. The Journal of Immunology 3819

46. Karnitz, L. M., L. A. Burns, S. L. Sutor, J. Blenis, and R. T. Abraham. 1995. 51. Kolch, W., G. Heidecker, G. Kochs, R. Hummel, H. Vahidi, H. Mischak, Interleukin-2 triggers a novel phosphatidylinositol 3-kinase-dependent MEK ac- G. Finkenzeller, D. Marme, and U. R. Rapp. 1993. Protein kinase C␣ activates tivation pathway. Mol. Cell. Biol. 15:3049. RAF-1 by direct phosphorylation. Nature 364:249. 47. Williams, S., C. Couture, J. Gilman, T. Jascur, M. Deckert, A. Altman, and 52. Izquierdo, M., S. J. Leevers, D. H. Williams, C. J. Marshall, A. Weiss, and T. Mustelin. 1997. Reconstitution of T cell antigen receptor-induced Erk2 kinase D. Cantrell. 1994. The role of protein kinase C in the regulation of extracellular activation in Lck-negative JCaM1 cells by Syk. Eur. J. Biochem. 245:84. signal-regulated kinase by the T cell antigen receptor. Eur. J. Immunol. 24:2462. 48. von Willebrand, M., S. Williams, P. Tailor, and T. Mustelin. 1998. Phosphory- 53. Schonwasser, D. C., R. M. Marais, C. J. Marshall, and P. J. Parker. 1998. Acti- lation of the Grb2- and phosphatidylinositol 3-kinase p85-binding p36/38 by Syk vation of the mitogen-activated protein kinase/extracellular signal-regulated ki- in Lck-negative T cells. Cell. Signalling 10:407. nase pathway by conventional, novel, and atypical protein kinase C isotypes. Mol. 49. Brumbaugh, K. M., J. J. Perez-Villar, C. J. Dick, R. A. Schoon, M. Lopez-Botet, Cell. Biol. 18:790. and P. J. Leibson. 1996. Clonotypic differences in signaling from CD94 (kp43) on 54. Vossler, M. R., H. Yao, R. D. York, M. G. Pan, C. S. Rim, and P. J. Stork. 1997. NK cells lead to divergent cellular responses. J. Immunol. 157:2804. cAmp activates Map kinase and Elk-1 through a B-Raf- and Rap1-dependent 50. Kanakaraj, P., B. Duckworth, L. Azzoni, M. Kamoun, L. C. Cantley, and pathway. Cell 89:73. B. Perussia. 1994. Phosphatidylinositol-3 kinase activation induced upon 55. Wan, Y., and X. Y. Huang. 1998. Analysis of the Gs/mitogen-activated protein Fc␥RIIIA-ligand interaction. J. Exp. Med. 179:551. kinase pathway in mutant S49 cells. J. Biol. Chem. 273:14533. Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021