Defective expression of hematopoietic cell protein tyrosine phosphatase (HCP) in lymphoid cells blocks Fas-mediated apoptosis

Immunity,

Vol. 2, 353-362,

April, 1995, Copyright

0 1995 by Cell Press

Defective Expression of Hematopoietic Cell Protein Tyrosine Phosphatase (HCP) in Lymphoid Blocks Fas-Mediated Apoptosis Xiao Su,* Tong Zhou,’ Zheng Wang,’ PingAr Yang,’ Richard S. Jope,t and John D. Mountz** *Division of Clinical Immunology and Rheumatology Department of Medicine TDepartment of Psychiatry and Behavioral Neurobiology *Birmingham Veterans Administration Medical Center The University of Alabama at Birmingham 473 Lyons-Harrison Research Building University of Alabama at Birmingham Station Birmingham, Alabama 35294-0007

Summary Protein tyroslne dephosphorylation after Fas crosslinking occurred in Fas apoptosis-sensitive CEM-6 cells but not in Fas apoptosls-resistant MOLT-4 cells, and apoptosis in the CEM-6 cells could be inhibited by the protein tyroslne phosphatase inhibitor, psrvanadate. The time course and level of dephosphorylatlon were correlated with Increased hematopoietlc cell protein tyrosine phosphatase (HCP) activlty, but not with the activity of two other tyrosine phosphatases. The level of expression of HCP was correlated with Fas apoptosis function in eleven human and murine Faspositive lymphoid cell lines. Expression of recombinant HCP In the MOLT-4 cell line converted this Fas apoptosis-resistant cell line to Fas apoptosis sensitive. HCP-mutant meVmeV mice exhibited Increased expresslon of Fas but decreased Fas-mediated apop tosis function in lymphold organs after anti-mouse Fas antibody treatment In vivo. Thus, HCP-mediated protein dephosphorylatlon Is Involved In the delivery of the Fas apoptosls signal in lymphold cells.

Introduction Apoptosis is a physiological process of selective cell deletion that occurs during embryogenesis, metamorphosis, tissue atrophy, and tumor regression (Wyllie et al., 1980; Walkeret al., 1988; Dukeet al., 1983; Schmid etal., 1988). It is a critical mechanism by which the immune system maintains tolerance to self-antigens, such as clonal deletion of autoreactive T and B cells (Golstein et al., 1991; Cohen et al., 1992). It is also a mechanism for cytotoxic T cell-mediated target cell lysis, and corticosteroid-induced death of immature lymphocytes (Ucker, 1987; Wyllie, 1980; Kagi et al., 1994). One of the most distinctive features of apoptotic cell death is the extensive endonuclease degradation of chromosomal DNA into oligomers of about 180 bp (Wyllie et al., 1980). Apoptosis can be provoked by various stimuli that result in a sequence of metabolic events. These events include a sustained rise in cytosolic ionized calcium, which, in turn, activates latent enzymes that contribute to the structural changes associated with apoptosis in many cell types (Orrenius et al., 1989; Arends et al., 1990; Knight et al., 1991). Apoptosis usually requires

Cells

energy in the form of ATP and new gene expression and protein synthesis. This indicates that apoptosis, as op posed to necrosis, is an energydependent active process, which requires a cascade of signal-transducing events. It has long been appreciated that the growth and many functional responses of mammalian cells are regulated by kinase-mediated protein phosphoryiation, or phosphatase-mediated protein dephosphorylation signaling pathways, or both. The role of protein phosphorylation in apop tosis has been extensively studied. Activation of protein kinase C (PKC) by phorbol esters maintains cell viability in the absence of interleukin-2 (IL-2) in cytokinedependent CTLL cells, and the resistance to glucocorticoid-induced apoptosis in these cells requires a tyrosine kinase-dependent signal (Walker et al., 1993). Apoptosis induced by microtubuledisrupting drugs in the cultured human lymphoma cell line, BM13874, is inhibited by a phorbol ester, PDBu (Takano et al., 1993). PKC activation also appears to suppress spontaneous apoptosis in mouse splenic B lymphocytes (Illera et al., 1993) and chronic lymphocytic leukemia lymphocytes (Forbes et al., 1992). In contrast, protein phosphatases, which counteract the action of protein kinases, have been shown to enhance apoptosis in many systems. Recently, specific protein dephosphorylation has been observed to occur in apoptosis induced by ionizing radiation and heat shock in human lymphoid lines. Okadaic acid prevents this apoptosis by inhibiting protein dephosphorylation (Baxter and Lavin, 1992). Calyculin A, another potent inhibitor of phosphatase PPl and PP2A, also prevents apoptosis in Burkitt’s lymphoma cells exposed to y radiation or heat (Song and Lavin, 1993). It has been shown previously that T cell receptor (TCR)-CDSmediated induction of apoptosis in a murine T cell hybridoma is inhibited by the immunosup pressive drugs cyclosporin A (CsA) and FK508, owing to their inhibitory effect on the activity of calcineurin, a calcium- and calmodulindependent serinelthreonine phosphatase (Fruman el al., 1992). Taken together, these data suggest that protein phosphorylation and dephosphorylation regulate apoptosis in various systems. The Fas antigen (CD95) is a cell surface protein expressed on a variety of cells and tissues and belongs to the tumor necrosis factor/nenre growth factor (TNFINGF) receptor family (Itoh et al., 1991; Yonehara et al., 1989). Molecular cloning of the APO-l antigen showed that it is identical to the Fas antigen (Oehm et al., 1989). An anti-APO-l monoclonal antibody (MAb) induced apoptotic cell death in some sensitive human cell lines in vitro (Trauth et al., 1989). Recently, the ligand for Fas antigen has also been identified and characterized (Suda et al., 1993; Takahashi et al., 1994). Mutation of the fas and fes ligand gene in Ipr and g/d mice, respectively, causes lymphoproliferation and a generalized autoimmune disease resembling systemic lupus erythematosus (Watsnabe-Fukunagaet al., 1992; Lynchet al., 1994; Takahashi et al., 1994). Thus, Fas is an important apoptosis molecule that may be involved in the pathogenesis of autoimmune diseases.

Immunity 354

CEM-6

Cell Line anti-Fas

M

CEM-6

MOLT-4

-+

-+

Far +

MOLT-4

An anti-human Fas MAb induces apoptosis in some, but not all, cells that express the Fas antigen (Trauth et al., 1989; Yonehara et al., 1989). This suggests that Fasmediated apoptosis is not only regulated by interaction of Fas ligand with Fas, but also by a permissive Fas signaling pathway. Deletion analysis of the human Fas antigen from the C terminus indicated that the region homologous to TNF type I receptor is essential for the Fas antigen to transduce the signal for apoptosis into cells (Itoh and Nagata, 1993). This C terminus region of the Fas antigen does not contain domains for kinase or phosphatase activity, suggesting that this region might interact with cytoplasmic enzymes involved in the signal transduction. However, the mechanisms by which the intracellular death signal is transduced by Fas antigen remain unclear. Hematopoietic cell phosphatase (HCP), primarily expressed by hematopoieticderived cells, is a protein tyrosine phosphatase (PTP) belonging to a subfamily of intracellular PTPs that consists of a structure encompassing two N-terminal SH2 domains, a phosphatase domain, and a short basic amino acid-rich tail (vi et al., 1992). It has also been given several other names, including PTP-1C (Shen et al., 1991; Zhaoet al., 1993) Src homology region 2 (SH2) domain phosphatase (Matthews et al., 1992) Src homology PTP-1 (SHPTPl) (Plutzky et al., 1992a), and PTP nonreceptortype 8(Plutzkyet al., 1992b). Point mutations of the Hcph gene result in aberrant splicing of the Hcph transcript and cause severe immunodeficiency accompanied by systemic autoimmune disease in mice homozygous for the recessive allelic mutation motheaten (me) or viable motheaten (me? on chromosome 8 (Shultz et al., 1993; Tsui et al., 1993) but the role of HCP in the development of autoimmune phenotypes in me and me” mice is still unknown. In this study, Fas-mediated apop

Figure 1. Fas Expression and Apoptosis Function in CEM-5 and MOLT-4 Cell Lines (A) Cells (109 were stained with mouse antihuman Fas MAb followed by anti-mouse IgM fluorescein isothiocyanate-conjugatsd second antibody. Cells (10,ooO) were then analyzed by FACScan. The histograms of fluorescein isothiocyanate fluorescence density are presented. Dotted lines represent the second antibody control showing no nonspecific binding by the second antibody as compared wlth the unstained cells (solii line). The densely dotted line represents staining with the Fas MAb with a separated positive peak. (6) CEM-5 and MOLT-4 cells (5 x 1CP) were incubated in 1 ml of 10% FCS in RPM1 1840 medium and 1 us/ml of either control antibody or mouse anti-human Fas MAb. DNA was extracted from the cells 4 hr later and analyzed in a 1% agarose gel.

tosis signaling pathway was investigated. Our results indicated that the expression of HCP is necessary for delivering Fas apoptosis signal in lymphoid cells. ReStlIt

Disassociation of Far Exprassion from Its Function We screened six human and five murine lymphoid cell lines for expression of Fas antigen. Of the eleven cell lines, two, CEM-8 and MOLT-4, expressed equivalent levels of cell surface Fas antigen (Figure 1A). The capability of the Fas molecule to transduce the apoptotic signal in these two cell lines was investigated by analysis of anti-fasinduced internucleosomal DNA cleavage (Figure 1B). A DNA ladder indicating extensive internucleosomal DNA cleavage was obsenred 4 hr after incubation with anti-Fas antibody in CEM-8 cell line, but not in MOLT-4 cell line. These data indicated that Fas expressed on MOLT-4 cells was not functional and there might be a defect in the Fas signaling pathway in the MOLT-4 cell line. The apoptosis defect in MOLT-4 cells was Fas specific, as both CEMB and MOLT-4 cells have the same sensitivity to corticosteroid-induced apoptosis (data not shown). Role of Protein Phosphorylation and bphosphotylatlon Protein phosphorylation and dephosphorylation have been demonstrated previously to regulate apoptosis signaling (Walker et al., 1993; lllera et al., 1993; Baxter and Lavin, 1992). To investigate their role in Fas-mediated apoptosis, CEM-8 cells were incubated with anti-Fas antibody in the presence of different protein phosphorylation and dephosphorylation modulators. Pervanadate, a protein tyrosine phosphatase inhibitor, dramatically inhibited

Fas Apoptosis 355

Signaling

by HCP

B 110

Genistein

IO

.I

-1

Anti-Fas + Pervanadate

Figure 2. Effects of Inhibitors and Stimulators of Protein Kinase and Phosphataee on Fas-Mediated Apoptosis in CEM-6 Cell Line CEM6 cells (IV) were incubated with 1 pg/ml of anti-human Fas antibody in the presence of different concentrations of pervanadate, genistein. or PMA. The cells were analyzed for apoptosis 4 hr later by the in situ nick translation technique. Cells undergoing apoptosis were quantitated using a standardized eyepiece grid at a magnification of 25.8 x . The results represent the mean f SEM of apoptotic cells in 25 randomly selected fields of view. (A) Fas-mediated apoptosis in CEM6 cells incubated with pervanadate. (6) Fas-mediated apoptosis in CEM-5 cells incubated with genistein. (C) Fas-mediated apoptosis in CEM-5 cells incubated with PMA. (D) Inhibition of Fas-mediated apoptosis in CEM-6 cells by pervanadate (1 mM).

anti-Fas-induced apoptosis in dose-dependent manner (10-‘-lW3 M), and 50% inhibition occurred at 0.01 mM (Figures 2A and 2D). Preincubation of CEM-6 cells with pervanadate for 10 min to 2 hr prior to Fas stimulation was required for maximal inhibitory effect, whereas incubation with pervanadate after Fas cross-linking or washing of cells after pervanadate treatment did not inhibit apoptosis (data not shown). This indicated that pervanadate blocks Fas apoptosis signaling at an early stage. In contrast, the protein kinase inhibitors, genistein (Figure2B), herbimycin A (1 @ml), and staurosporine (20 nfvl) (data not shown) had little effect on Fas-mediated apoptosis. Identical concentrations of these protein kinase inhibitors were capable of inhibiting antiCD3induced proliferation (data not shown). Stimulation of PKC by phorbol myristic acetate (PMA) inhibited apoptosis by approximately 40% at 0.2 uM (Figure 2C). Fas expression was increased in pervana-

dateor PMA-treated CEM-6 cells and, therefore, the inhibition of Fas-mediated apoptosis could not be attributed to a decrease in Fas expression (data not shown). These results suggest that protein tyrosine dephosphorylation might be a critical biochemical event during Fas-mediated apoptosis. HCP-Msdlatsd Protein Tyroslne Dephosphorylatlon In Fas-Msdlated Apoptosls Todetermine whether protein tyrosine dephosphorylation occurs in Fas-mediated apoptosis, CEM6 and MOLT4 cells were prelabeled with “PI, incubated with anti-human Fas or control MAb, followed by anti-mouse immunoglobulin M (IgM) treatment to enhance the cross-linking for various periods of time. The tyrosine-phosphorylated proteins were immunoprecipitated by an anti-phosphotyrosine agarose and then analyzed in SDS-PAGE gels. A rapid increase in protein tyrosine dephosphorylation occurred in lysates of CEM-6 cells after Fas cross-linking, but not in lysates of MOLT4cells(Figure3A). Spontaneousdephosphorylation did not occur in either CEM6 or MOLT4 cells treated with the control antibody (Figure 3A). In CEM-6 cells, antiFascross-linking signaling led to dephosphorylation of two proteins with molecular masses of approximately 100 kDa and 90 kDa in 5 min. Two other proteins with molecular masses of approximately 60 kDa and 58 kDa were also tyrosine dephosphorylated in 15 min after anti-Fas antibody treatment. These results suggest that protein tyrosine phosphatase(s) activation was induced by apoptosis signaling through Fas. To determine which tyrosine phosphatase was activated by Fas signaling and was responsible for protein tyrosine dephosphorylation, phosphatase activity for HCP, protein tyrosine phosphatase-I B (PTP-1 B), and SH-protein tyrosine phosphataseZ/Syp (SHPTP2/Syp) was assayed after immunoprecipitation with specific antibodies. HCP activity was rapidly induced 1 min after anti-Fas stimulation and reached its maximal activity in 15 min in the CEM-6 cells, but not in the MOLT4 cells (Figure 38). In contrast, there was neither early induction of the PTP-1 B activity (Figure 38) nor detectable SHPTP2/Syp activity (data not shown) in both cell lines, indicating that the initial increase in tyrosine phosphatase activity in CEM-6 cells was attributable to an increased activity of HCP. These results indicate that the Fas apoptosis signal induces increased HCP activity and that HCP-mediated protein tyrosine dephosphorylation may play a critical role in the Fas signaling pathway. As HCP activity was induced during Fas-mediated apoptosis, we questioned whether the cells that are resistant to Fas-mediated apoptosis have defective expression of HCP. Therefore, Fas-mediated apoptosis function and HCP expression were examined in eleven human and murine Fas-positive lymphoid cell lines. Fas apoptosis correlated with Hcph gene expression, but not with expression of Fas on the cell surface (Table 1). There was abundant Hcph mRNA in the CEM-6 cell line, whereas no Hcph mRNA was detected in the MOLT4 cell line (Figure 4A), and both CEM6 and MOLT4 cell lines expressed equivalent levels of PTP-1 B, and both lacked detectable levels of SHPTP2/Syp (Figure 4B). These results suggest that

Figure 3. Protein Tyrosine Dephosphorylation and Tyrosine Phosphatase Activities in FasMediated Apoptosis (A) Tyrosinaphosphorylated proteins were analyzed on SDB-polyacrylamide gels after being immunoprecipitated by anti-phosphotyrosine agarose from CEM-6 and MOLT-4 cells, which were prelabled with =P, and then crosslinked with anti-Fas or control antibody for 0,5, and ibmin. Thesizesandpositionsofstandard markers are indicated. (B) HCP and PTP-I B were immuncprecipitated from GEM-6 and MOLT-4 cells at the specified periods of time after anti-Fas cross-linking. PTP activity in the immune complexes was de tected by an in vitro PTP assay using tyrosine phosphorylated raytide as a substrate. The activity of HCP and PTP-1B is indicated as the percentage of control.

A MOLT-4

M fKD) CCllltTOl

0

5

16

0

AntWas 5 15

CWltrd

g

5

15

0

Anti-Fes 5 15 Timefmfn)

defective HCP expression in MOLT4 cells might result in resistance to Fas-mediated apoptosis.

Fas Apoptosis Function in the MOLT-4 Cell Llne after Trsnsfectlon with a Hcph Expression Vector To determine further whether HCP is required for transducing the Fas-mediated apoptosis signal, HCPdeficient MOLT4 cells were transfected with an expression vector that contains full-length Hcp/r cDNA (Capone et al., 1993) or with an empty vector as control. After 48 hr, Hcph was expressed in MOLT4 cells transfected with Hcph expression vector as demonstrated by immunoblot analysis (Figure 5A). The transfected MOLT4 cells were incubated with several concentrations of mouse anti-human Fas (0.05-l ug/ml) for different time periods (2-l 2 hr) and apoptosis was assessed by the internucleosomal DNA fragmentation assay and in situ staining. Anti-Fas antibody induced apoptosis in MOLT4 cells transfected with the Hcph expression vector, but ‘not the control vector (Figure 58). Increased sensitivity to Fas-mediated apoptosis in Hcphtransfected MOLT4 cells was demonstrated by dose and time course response analysis. In Hcph-transfected MOLT4 cells, there was more than a l,CKXt-fold increase in the dose response to anti-Fas antibody (Figure 5C), and a 5-fold decrease in the time required for the induction of apoptosis(Figure5D). These results indicate that recombinant expression of HCP restored Fas apoptosis function in MOLT4 cells. Abnormelitles of Fas Apoptosls Function In HCP Mutant Motheaten Mice Mice homozygous for the autosomal recessive motheaten (me), or the less severe allelic viable motheaten (me”) mutations develop severe combined immunodeficiency and

systemic autoimmunity (Sidman et al., 1989; Green and Shultz, 1975; Shultz, 1988; Shultz et al., 1984). Recent evidence suggests that mutations in the Hcph gene are directly responsible for the motheaten phenotype (Shultz et al., 1993; Tsui et al., 1993). To determine whether there is a defect of Fas apoptosis function in HCPdeficient moth eaten mice, thymocytes and splenic lymphocytes from both C57BL/8J-+/+ and HCP-mutant mev/mev mice were first examined for their Fas expression by flow cytometry

Table 1. Correlation of Fas Expression, HCP Expression in Human and Murine

Apoptosis Lymphoid

Function, and Cell Lines

Name

Cell Type

FeS Expression

Human Cell Lines CEM-6 MOLT-4 MOLT-3 H9 LBW-2 u937

T T T T B Monocyte

55 57 55 120 72 36

>95% <5% <5% X5% 22% 10%

1.0 0 0.05 0.8 0.4 0.1

Murine Cell Lines EL-4 YAC-1 WEHI 231 WEHI 279 A20

T T B 0 B

67 34 55 60 72

12% <5% 44% 92% X5%

0.4 0 0.7 0.9 1.2

Cell Line

Fas Apoptosis

HCP&actin Ratio

Fas expression was determined by flow cytometry analysis and is expressed as the mean fluorescence intensity. Fas apoptosis was determined by the in situ apoptosis staining method and is expressed as the percent of cells undergoing apoptosis as described in Figure 2. HCP expression was determined by Northern blot analysis using the human or murine HcpIr cDNA fragment as probe, and is expressed as the ratio ( x 102) of HCP expression relative to 6-actin expression quantitated with a Phosphorlmager.

Fas Apoptosis 357

Signaling

by HCP

Fiiure4. Expression of Protein Tymaine Phoaphatases in CEM-6 and MOLT4 Cells (A) Total RNA (5 pg) from CEM-6 and MOLT4 cells was electrophoreeed, traneferred to a nitmcellulose membrane, and hybridized with a

A 28s

CEY+

18s

MOLT-4 :... “II _.

HCP

Hcph

65KD

are preknt

PTP-1 B

lane. The 26S and 18s

from CEM-6 and MOLT-4 cells, and eeparated by SDS-polyacrylamide gels, blotted to the nitrooellubee membranes, and deteoted byantiHCP, m-1 B, and SHPTP2/Sypantlbody. The sizes and posltiis of the PTP protein bands are indicated to the right of the flgure.

FActin i.

analysis. Fas was expressed at equivalent levels in thymocytes, but at significantly higher levels in splenic lymphocytes in me%rW mice compared with control mice (Figure 6A). Fas-mediated apoptosis function was then determined in vivo by injecting 100 pg of hamster anti-mouse Fas antibody and control hamster antibody intraperitoneally into C57BL&l-+/+ and meWW and control mice (Ogasawara et al., 1993). Apoptosis was assessed by in situ apoptosis staining of frozen sections. The number of lymphocytes that had undergone apoptosis was reduced in the thymus, spleen, and lymph node of me%W compared with control mice after anti-Fas treatment (Figure 68; Figure 7A). In the thymus of mev/mev mice as well as

CEM-6 MOLT-4 MOLT-4

in eack

riboeomal bands are shown to the iefl of the figure. (B)Total cellular protein lysateawere pmpared

28s 18s

spP-labeW human !@#I probe. An kkntkal Motwashybridizedwithapmbefor@actin, indicating that nearly identical amount8 of RNA

Control pHcph

,001 31 .1 1 10 Anti-b (WJlrnl,

(Cl (M,) (M,)

control mice, apoptosis was distributed predominantly in a focal pattern in the outer regions of the cortex (Figure 78). The number of apoptotic foci was equivalent in these two strains of mice. However, the average number of apoptotic cells per foci in control mice was 28 f 10 compared with 8 f 3 in the me%w mice. Low levels of apoptmis were observed in the thymus, spleen, and peripheral lymph node after control antibody treatment, and there was no difference between me%Wand control mice (data not shown). These data indicated that there was defertive Fas apoptosis signaling in lymphoid organs in HCPdeficient motheaten mice. It was also noted that although anti-Fas-induced apoptosis was decreased in the lym-

B

MC M”

Figure 5. Fas Apoptosis Function Cells after Transfeotion with

in MOLT-i Expregdar

MOLT-4 cells (2 x IV) were bansfected with either the pSlX@lE vector containing the human Ho@ gene insert or a wntrol p913CBlE empty vector. Tranafeclion efficiency was determined to be 40% by immunohistoohemical staining using anti-HCP antibody. (A) CEM-6 cells (C), MOLT-4 cells transfeoted with a control pgl3CplE empty vector (MC),

0 7. . 6 I 10 I* 14 Tima (W

conhol (MC)transfected MOLT-4 cells after 4 hr incubation with the mouse anti-human Fas MAb (1 @ml). (C) CEM-6 (C) and Hcph (M,,) or control (k&) transfected MOLT4 cells were incubated with mouse anti-human Fas MAb at different concentrations (0.05-I &ml) for 6 hr. Apoptosis was assessed and quantitakl by in situ apoptosis staining as detailed in Figure 2. (0) CEM-6 and transfected cells were incubated with mouss anti-human Fas MAb at 1 &ml for different periods of time (2-12 hr). &OPtOSiS WaS assessed and quantitated by in situ apoptosis staining as detailed in Figure 2.

Immunity 356

C57BUG+l+

Figure 6. Fas Expression and Apoptosis Function in Motheaten and Normal Mice (A) Single-cell suspensions (lq from thymus a (Thy) and spleen (Sp) of C57BLfBJ-+/+ and 0 mc”lme* me%n# mice were stained with phycoerythrtnconjugated hamster anti-mouss Fas antibody or irrelevant control antibody. Viabte cell0 (10,000) were then anatyzed by FACScan. The solid lines represent the staining wlth the antimouse Fas antibody, and the dotted lines rep & resent the control antlbody statning. (B) For in vtvo anafysts, C57BL/5&+/+ and meVme~ mice were injscted with 100 ug (lntraperttoneally) of anti-mouse Fas antibody. ThyThymus Spleen LN mu8 (Thy), spleen (Sp), and lymph node (LN) secttons were prepared 2 hr after injsction of three 5week-old mice. The percent of apoptosis cells was determined as described in Figure 2 and quantitated by counting apoptotic and nonapoptotic cells in 25 randomly selected fields of view using a standardized eyepiece grid. The MEAN f SEM for three different mice are

phoid organs from HCPdeficient rrW/rrW mice, there were equivalent levels of Fas-mediated apoptosis in the hepatocytes (approximately 75%), which caused equivalent mortality in both strains of mice. These results imply that HCP-mediated Fas apoptosis signaling is limited in lymphoid cells, whereas other signaling molecules may be utilized in Fas-induced apoptosis pathways in the liver.

Discussion Our interest in the mechanism of Fas/APO-1 apoptosis signaling arose from the observation that anti-Fas/APO-1 antibody could mediate apoptosis in some, but not all, cell lines that express the Fas molecule. After analyzing Fas expression and function in several cell lines, we found that the MOLT-4 cell line expressed equivalent levels of the Fas antigen as the CEM-9 cell line, but was resistant to anti-Fas-induced apoptosis. Therefore, the presence of Fas/APO-1 antigen on the cell surface was not sufficient to execute the anti-Fas-induced signal for cell death. Several signal transduction modulators were assessed for their effects on Fas-mediated apoptosis in CEM-9 cell line. Pervanadate, a protein tyrosine phosphatase inhibi-

A meVme”

Thy

LN

tor, dramatically reduced Fas-mediated apoptosis in CEM-6 cells, suggesting that Fas-mediated apoptosis in this cell line is dependent on protein tyrosine phosphatase activity. Analysis of tyrosine-phosphotylated proteins by immunoprecipitation using anti-phosphotyrosine agarose indicated that several proteins in cytosolic extracts of CEM-9 cells were rapidly dephosphorytated after incubation with anti-Fas antibody, whereas no protein dephosphorylation was observed in cytosolic extracts from the Fas apoptosis-resistant MOLT-4 cells. This protein tyrosine dephosphorylation correlated with rapid induction of in CEM-9 HCP activity in CEM-9 cells. Expression of Hcph as well as other Fas-positive lymphoid cell lines correlated with their Fas-mediated apoptosis function. Transfectfon of a Hcph expression vector into Hcph negative MOLT4 cells partially restored the Fas-mediated apoptosis function. Based on these observations, we proposed that HCP might be responsible for Fas-mediated apoptosis signaling transduction in certain cell lines. This is consistent with previous observations that HCP negatively regulates cell growth through association of its SH2 domains with tyrosine phosphorylated c-Kit as well as the IL-3 receptor (vi and Ihle, 1993; Yi et al., 1993). The extracellular and intracellular signaling domains of

Figure 7. Decreased Fas Apoptosis Function in me%ne” Mice (A) Apoptosis in the thymus, spteen, and lymph node of C57BLBJ-+/+ and rrW/rrW mice was determined by the in situ apoptceis statning method as detailed in Figure 2 and Ftgure 68 (magnification, 6x). (B) Cells undergoing apoptosts in the outer cortex of the thymus of C57BUBJ-+/+ and me”/ rnw mice were anafvzed as described in Future 2 and Figure 6B (magnification, 24x). -

Fas Apoptosis 359

Signaling

by HCP

Fas antigen are similar to those of TNF/NGF receptor family, which includes the low affinity NGF receptor, TNF receptors, CD27, CD39 CD49, and OX40 (Itoh et al., 1991). No intrinsic kinaseor phosphataseactivity has been identified in the cytoplasmic region of Fas. However, a domain has been identified in the cytoplasmic region that is re quired for transduction of the signal for cell death (Itoh and Nagata, 1993). Whether this apoptotic signal-transducing domain exerts its function through the interaction with HCP remains to be clarified. Based on the recent descriptlon of TNF receptor-associated factors (TRAFl and TRAF2) (Rothe et al., 1994), it is possible that HCP is indirectly activated through other proteins that coupled to the Fas intracellular domain. In addition, a sphingomyelinase-ceramide signaling mechanism for phosphatase activation by the TNF-R has been described (Kolesnick and Golde, 1994; Kim et al., 1991; Mathiaset al., 1991; Ballou, 1992) and this pathway has been demonstrated to be important in ceramide-induced apoptosis (Jarvis et al., 1994; Obeid et al., 1993). Whether HCP is activated through sphingomyelinase-ceramide pathway is under investigation. The Hcph gene has been defined as the homozygous recessive atlelic mutation in the motheaten (me) mice on chromosome 6 (Shultzet al., 1993; Tsui et at., 1993). These mice develop abnormalities in T cells, B ceils, and macrophages, increased autoantibody production, and immune deficiency (Sidman et at., 1989; Green and Shultz, 1975; Shultz, 1988; Shultz et al., 1964). Although HCP has been identified as a mutant gene in this mouse strain, the function of HCP and the mechanism(s) by which autoimmune disease developed are still unknown. Our results indicated that despite increased Fas expression in the spleen of me%ney mice, there was defective Fas apoptosis function in thymocytes and peripheral lymphocytes. This defective Fas apoptosis function may be caused by defective HCP expression that leads to defective Fas signaling in lymphoid cells of these mice. The observation that the HCP defect reduces lymphoid cell apoptosis but not hepatocyte apoptosis in meVmeY mice suggests that Fas uses different signaling pathways in different cell types. This regulation of Fas-mediated apoptosis at the signal transduction level may be an important mechanism to induce and regulate Far+mediated apoptosis selectively in the presence of equivalent levels of Fas and Fas ligand expression. This demonstrates that HCP may function as a Fas apoptosis signaling molecule and that this HCP deficiency is linked to a Fas-mediated apoptosis defect in the immune system in this mouse model. Although Fas-deficient /prnpr and HCP-deficient me/me mice develop different cellular phenotypes, both of them share some common features of autoimmunity, such as splenomegaly, autoantibody production, and glomerulonephritis (Cohen and Eisenberg, 1991; Samelson et al., 1996; Budd et al., 1992). The differences between the autoimmune disease of Ipr mice and me mice indicate that defective expression of the Fas apoptosis molecule in Ipf immune features identical to those observed with low expression of the protein tyrosine phosphatase in me mice. One explanation may be that Fas signaling leads to signal

transduction through multiple pathways. Recent evidence that supports this hypothesis is the description of other potential signaling pathways that can be induced by Fas stimulation (D. Green, personal communications). This concept is also consistent with our observations that equivalent levels of Fas-mediated apoptosis occur in liver from normal double-positive and HCPdeficient me/me mice, and that okadaic acid, a serine/threonine phosphatase inhibitor, can also partially inhibit Fas-mediated apoptosis in CEM-6 cells (data not shown). The signaling pathway utilizing the HCP phosphatase is important for apoptosis signaling, but other signaling pathways may be critical for other functions mediated by the Fas molecule. This concept is consistent with the finding that HCP is highly expressed in the tracheobronchial cells, lung, and kidney (Plutzky et al., 1992a). This multifunctional view of Fas is supported by the obsenration that the Fas molecule can transduce a costimulatory signal in developing thymocytes (Alderson et al., 1993). Also, Fas signaling might utilize different pathways at different stages of lymphocyte development. Fas signaling mediated through the HCP phosphatase pathway may be important at an early stage of hematopoietic cell development (Shultz, 1991; Hayes et al., 1992), whereas defective expression of the Fas molecule in Ipr mice would prevent signaling through the Fas molecule at all stages of lymphocyte development (Watanabe-Fukunagaet al., 1992). Finally, HCP may be involved in signal transduction initiated by cell surface molecules in addition to the Fas molecule. Therefore, it is not surprising that the consequence of the Hcph gene defect in me mice is not equivalent to the fas gene defect in Ipr mice, despite the utilization of HCP as a second messenger in Fas-mediated apoptosis by some cells. The present study provides promising intervention points for apoptosis signaling. The apoptosis defect observed in MOLT-4 cells could conceivably be overcome by supplying a phosphatase activity or inhibiting a phosphorylation activity. Further understanding of the interplay between phosphorylation and dephosphorylation events mediated by protein tyrosine kinase and phosphatase will be critical to the identification of pharmacologic targets for modulation of apoptosis. Experimental

Procedure8

Ylce The C57BU5J-+/+ and C57BUW viable homozygous motheaten (meVAne’) mice were purchased from the Jackson Laboratory (Bar Harbor, Maine). Mice were used at 0 weeks of age. Cell and Cell Culture The CEM-O, MOLT-t, MOLT-3, LBW-2, U937, H9, EL-l, YAC-1, WEHI 231, WEHI 279, and A20 cell lines were obtained from American Type Culture Collection (Rockvtlle, Maryland) and were cultured at 37OC and 5% CQ in RPM1 1840 medium supplemented with 19% heatinactivated fetal calf serum (FCS), 10 mM sodium bicarbonate, 10 mM HEPES, 100 U/ml penicillin, 199 pg/mi streptomycin, and 2 mM glutamine. Antlbodiw and Magenta Mouse anti-human Fas MAb. rabbit anti-human SHPTPl (HCP), PTP18, and SHPTPVSyp polycionai antibody were purchased from Up state Biotechnology, Incorporated (Lake Placid, New York). Goat antirabbit @G-horseradish peroxidase was obtained from Boehringer

Immunity 360

Mnnheim Corporation (Indiiapolia, India) and an&mouse IgM antlbodywas obtained from Phamrigen (San Dtego, California). The hamster admouse Fas antibody was generated in our laboratory (T. 2. et al., unpublished data). Monocbnal mouse antwne agarose. gentstein, PMA, and sodium om te were purchased from Sigma (St. Louis, Mlesourii. Pervanadate (vanadyi hydroperoxMe) was prepared as described (Pumtglta et al., 1992). Raytide and Peep- protein tyrosine kinase were purchased from Oncogene Science, lnaxporated (Uniondale. New York). DNAPlokr The DNA fragments of human and murine Ncp/r gene (nucleotides 251-105Oand nucleotldes551-128O)werederivedfrom revemetranscrtptbn+olymerase chain reaction (RT-PCR) ampliication of cDNA prepared from CEM-8 and EL4 total RNA (2-4 ug) using an RNA PCR kit (Perkin-Elmer Coqmmtbn, Norwalk, Connecticut) (vi et al., 1992) reepectfvely. Sequenceswere confirmed by DNAsequencing analysis using the didsoxy chain terminatlon method (vi et al., 1992). The pmkrbe was obtained from Oncor, Incorporated (Gaithersburg,

For Fas expression analysis of human cell lines, cells (lg were stainedwithoptimal concentrationsofmoussanti-human Fasantibody in PBS with 5% FCS and 0.1% sodium azide (NaN,) at room temperatumfor2Omin, followedbyincubation with eitherfluorescein isothiocyanate-conjugatad anti-mouse IgM or control antibody with irrelevant epec&ky at room temperatun for 20 min. For Fas expression analysis of murine lymphoid organs, single-cell suspensions (1 @were stained wkh phycoerymrinoonjugated hamster anti-mouse Fas antibody or wntrol antibody in PBS with 5% FCS and 0.1% sodium azide (NaNs) at mom temperature for 20 min. Viable cells (10,000) were analyzed by flow cytometry on a FAGS-Scan (Becton Dickinson, Mountain View, California) with logarithmic scales. Gel Ekmophoreats DNA Fragmentatlon Aesay CelN (5 x lorlrnl) were cultured in 8-well sterile dishes (Corning Glass Works, Corning, New York) with mouse anti-human Fas MAb or control antibody with irrelevant specificity (1 rig/ml) for 4 hr. The cells were then wllected and tysed with 500 ul of lysis buffer containing 50 mM NaCI, 10 mM Tris, 5 mM EDTA, and 0.1% SDS. Lysates were then subjected to proteinase K digestion (109 rig/ml) at 42*C for 1 hr. The DNAwas extracted with phenol(pH 7) phenol:chloroform:isoamylalcohol(5Oz49:l) ethanol precipitated, and then dried in a Speedvac wncentrator(Savant Instruments, Incorporated, Farmingdale, New York), resuspended in Tris-EDTA buffer and analyzed by electrophoresis on a 1% agaroee gel. f?egulatlon of Apoptoels by Inhlbltom and Stlmulaton of Proteln Klneee and Phoephateee CEM-8 cells (lOt cells/ml) were incubated with mouse anti-human Fas antibody (1 @ml final concentration) in the presence of different wncentrations of pervanadate (0.0001-l .O mM), genistein (0.03-0.5 ug/ ml), or PMA (0.12-2.0 CM). Cell viability was analyzed by the in situ apoptosis staining method after 4 hr. In Situ Apoptoele Stalnlng In situ apoptosis staining was carried out as described previously, with modification (Gavrieli et al., 1992). Cells were harvested and cytospun onto poly+lysine precoated sliies and fixed in 10% formalin at room temperature for 1 hr. Tissue sections for slides were snap frozen, sectioned, and treated similarly. After rinsing with distilled water, proteins were stripped from cells by incubation with 20 fig/ml proteinase K at room temperature for 15 min, and the slides were then washed four times with distilled water for 2 min each. Slides were immersed inTdTbuffer(3OmMTrizrnabase(pH7.2], 14OmMsodiumcacodyiate, 1 mM cobalt chloride). TdT (0.3 U/ml) and digitonigen-modified dUTP in TdT buffer were added and the slides were incubated in a humid atmosphere at 37OC for 1 hr. The reactbn was terminated by washing the slides with PBS at mom temprature. After the slides were dipped in 10% FCS in PBS at room temperature for 30 min and dried, they were covered with alkaline phosphatase-m njugated antidigitonigen antibody and incubated at room temperature for 80 min. The slides

were then washed with PBS and statned with nitrobiue tetrazoliuml 5bromo4chbroSindolyl phosphate (Sigma, St. Louis, Missouri) at mom temperature for approximately 30 min.

For ‘P, labeling, CEM-8 and MOLT-4 cells (5 x l(Plml) were preincubated in phosphate-free RPM1 1840 medium supplemented with 10% dialyzed FCS for 90 min, followed by incubation with 0.2 mCi of =PI per ml for 2 hr. The cells were then washed twice with ice&d RPM1 1840 medium and anti-human Fas MAb (1 uglml) or control antibody was added. After 10 min on ice, anti-mouse IgM (0.5 u@ml) was added to promote the cross-linking, and the cells wereincubated at 37OC for 0, 5, and 15 min. The cells were then lysed at 2.5 x 10 per 500 ul of Nonidet P-40 (NP-40) buffer (150 mM NaCI, 50 mM Tris-HCI [pH 8.0].0.5% NP4gO.5 mM ZnC12, 1 mM NasVO,, 50 mM NaF) containing leupeptin (1 ug/ml), sprotinin (1 nglml), pepstatin A (1 @ml), antipain (1 &ml), and phenyimethylaulfonyl fluoride (20 uglml). Cell lysates were clarified by centrifugation for 30 min at 14,008 rpm at 4OC, followed bytheadditionof 1OOuloffixedStaphybcoccusaureus(Pansorbin; Calbiochem) for 1 hr at 4OC. The immunoprecipitations were carded out by incubating the cell lysates with anti-phosphotyrosine agarose at 4OC for 2 hr with gentle agitation. The immune complexes were then washed five timeswith cold NP-40 lysis buffer and denatured in SDS sample buffer. The samples were then analyzed in 10% SDSPAGE gels, followed by autoradiography at -7O*C. Phoephateee Activity Auay After CEM-8 and MOLT-4 cells were treated with anti-Fas antibody for different periods of time, total cellular protein lysates were prepared by treating the cells (5 x Iv with lysis buffer (20 mM Tris-HCI [pH 8.0],150 mM NaCI, 1 mM EDTA, 1% [v/v] Triton X-100,1 mM PMSF, 0.04% bovineserumalbumin)at4°Cfor30min, followed bycentrifuge at 4OC for 20 min. Anti-HCP, anti-PTPlB, or antiSHPTP2/Syp antibody (2 pg) was incubated with each cell lysate sample at 4OC for 80 min with gentle agitation. Immune complexes were absorbed with anti-rabbit IgG protein A-agarose (Sigma, St. Louis, Missouri) at 4OC for 30 min and washed four times with cold phosphatase assay buffer (25 mM HEPES [pH 7.31, 5 mf.4 EDTA, 10 mM dithiothreitol). The synthetic peptide, raytide, was labeled at its tyrosine residue using [ys1P]ATP and p80protein tyrosine kinase as described (Streuli et al., 1990). The labeled protein was precipitated with 20% (w/v) trichloroacetic acid in 20 mM NaHtPD,, and dissolved in 0.2 M Tris-HCI (pH 8.0). The phosphatase activity assays were performed as previously described (Streuli et al., 1990). Northern Slot Analyels of Hcph Expmeelon Human or murine cells(2 x 103 were homogenized and total RNAwas extracted from the homogenates by the guanidinium-isothiocyanate method (Chirgwin et al., 1979). Equivalent amounts of total RNA (5 pg) from each cell line were denatured at 85OC for 5 min in electrophoresis buffer(0.4 M 3morpholinopropanesulfonicacid. 0.1 M sodium acetate, 2 mM EDTA [pH 7.01) containing 8% formaldehyde and 50% formamide, and size fractionated by electrophoresis through 1% agarose gel containing 0.6% formaldehyde. RNA was transferred to a nitrocellulose membrane (Nitroplus 2OCQ Micron Separations, Incorporated. Eastboro, Massachusetts) and baked for 2 hr at 80°C in a vacuum oven. The membrane was prehybridized and then hybridized with a PP-labeled 808 bp human or 730 bp murine Hcph cDNA probe. The membrane was then washed with 2 x SSC plus 0.1% SDS at room temperature for 30 min and then with 0.1 x SSC plus 0.1% SDS at 80°C for 30 min. and exposed to XAR-2 film (Eastman Kodak, Rochester, New York) with an intensifying screen at -7OOC. The membranes were then extensively washed to remove residual probe and rehybridized with a probe for 8-actin under identical conditions. Expreeelon of Proteln Tyroelne Phoephataeee CEM-8 and MOLT-4 cells (10s) were fysed in lysis buffer (1 M TrisHCI [pH 7.51, 0.5% Triton X-100, 1 mM PMSF, 1 ug/ml leupeptin, 1 &ml aprotinin, and 1 mM sodium orthovanadate) and equivalent amounts of total cellular protein lysates (20 j4g) were separated on 10% SDS-polyactylamide gels, blotted to nitrocellulose membranes (BieRad Laboratories, Hercules, California), and incubated with rabbit anti-human SHPTPl (HCP)or PTP-1 B or SHPTP2/Syp polyclonal anti-

Fas Apoptosis 381

Signaling

by HCP

body. The blots were counterstained with goat anti-rabbit IgG conjugated with horseradish peroxidase, and incubated with chemiluminescence detection reagents (Bcehringer Mannheim Corporation, Indianapolis, Indiana) including lumind, Qiodophenol, and Ha* at room temperature for 1 min, followed by exposure to XAR-2 film (Eastman Kodak, Rochester, New York) with an intensifying screen at room temperature for 1 min. Construct and Expmulon of Recombinant Hcph Gene The full-length 1937 bp human HCP cDNA fragment corresponding to nucleotides 28-1984 was amplified by RT-PCR from total RNA isolated from CEM-8 cells as follows: 30 cycles at 94% for 1 min, at 58% for 1.5 min, and at 72% for 2 min. The primers used for RTPCR were S-AGACTAGCTGCACCTCCT-3’ (nucleotides 28-45) and 5’~CGACATCTCCTAGGTTCA-3’ (complementary to nucleotides 1947-1984). The amplifted HCP cDNA fragment was directly subcloned into pCR vector (Invitrogen, San Diego, Caltfomia), following the procedures recommended by the manufacturer, and DNA sequencing analysis revealed that the insert cDNA sequence was identical to the published sequence (Yi et al., 1992). The insert was then excised by EcoRl and subcloned into the BamHl site of the p913C51 E vector (provlded by Dr. C. Weaver at University of Alabama at Birmingham) in front of TCR Vjl gene and 8 chain enhancer, which has been described to allow T cell-specific expression (Capone et at., 1993). MOLT4cells(2 x lq were transfected with 5 pg of pSlX8lE vector containing human Hcph gene insert or empty p913Cj3lE vector ascontrolbylipofectin(lOug)in 1 mlofserumfreeRPMll84tImedium at 37%. After 48 hr, the expression of recombinant Hcph gene in transfected cellswas detected by immunoblot using anti-HCP antibody as described above. Transfection efficiency was determined by immunohistochemical staining using the anti-HCP antibody following the procedures suggested by the manufacturer. In Vlvo Afmptosfs Auay Hamster anti-mouse Fas antibody and control antibody (100 pg) purified under identical conditions were injected intraperitoneally into three 6weekold C57BL/8J-+/+ and me%nev mice, respectively. Thymus, spleen, and lymph node sections were prepared 2 hr after injection. The apoptosis was analyzed by in situ apoptosis staining as described above.

We wish to thank Drs. W. Koopman and F. Hunter for review of the manuscript and B. K. Bunn for expert secretarial assistance in typing the manuscript. We also thank Dr. C. Weaver for kindfy providing the p913C8lE vector and Dr. X. Li for assistance in signaling modulator experiments. This work was supported by the National Institute of Arthritis, Musculoskeletal, and Skin Diseases (ROI-AR42547, POlAR03555), the National Institute of Allergy and Infectious Diseases (UOl-Al34588, ROl-Al30744), and theNational Instituteon Aging(ROlAG11853). X. S. is the recipient of a Gina Finri Memorial Fellomrhip from the Lupus Foundation of America, Incorporated, and a Glenn/ American Federation for Aging Research scholarship from the Glenn Foundation and the American Federation for Aging Research. T. Z. is supported by a grant from the Arthritis Foundation. Received

January

5, 1995; revised

February

3, 1995.

Alderson, M. R., Armitage, R. J., Maraskovsky, E., Tough, T. W., Roux, E., Schooley, K., Ramsdell, F.. and Lynch, D. H. (1993). Fas transduces activation signals in normal human T lymphocytes. J. Exp. Med. 178, 2231-2235. Arends, M. J., Morris, R. G., and Wyllie, A. H. (1990). Apoptosis: the role of the endonuclease. Am. J. Pathol. 738, 593-808. Ballou, L. R. (1992). Sphingolipids and cell function. Immunol. Today 73, 339-341. Baxter, G. D., and Lavin, M. F. (1992). Specific protein dephosphorylation in apoptosis induced by ionizing radiation and heat shock in human lymphoid tumor lines. J. Immunol. 748, 1949-1954.

Budd, R. C., Russell, J. Q., van Houten, N., Cooper, S. M., Yagtta. H., and Wolfe, J. (1992). CD2 expression correlates wlth prottferattve capacity of a8’ or yl* CD4CD8T cells in fpr mice. J. Immunot. 148, 1055-1084. Capone. M., Watrin, F., Fernex, C., Horvat, B., Krippl, B., Wu. L., Scollay, R., and Ferrler, P. (1993). TCR8 and TCRa gene enhancers confer tissue- and stage-specificity on V(D)J recombination events. EMBO J. 72,4335-4348. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J., and Rutter, W. J. (1979). Isolation of biotogically active ribonucleic acid from sources enriched in ribonuctease. Biochemistry 18,529&5299. Cohen, J. J., Duke, R. C., Fadok, V. A., and Sellins, K. S. (1992). Apoptosis and programmed cell death in immunity. Annu. Rev. Immunol. 10, 287-293. Cohen, P. L., and Eisenberg, R. A. (1991). fpr and @? singte gene modets of systemic autoimmuntty and lyrnphopmliferative disease. Annu. Rev. Immunol. 9.243-287. Duke, R. C., Chervenak, R., and Cohen, J. J. (1983). Endogenous endonuclease-induced DNA fragmentation: an early event in cellmediated cytolyais. Proc. Natl. Acad. Sci. USA 80, 8381-8385. Forbes, I. J.. Zalewski, P. D., Giannakis, C.. and Cowled, P. A. (1992). Induction of apoptosis in chronic lymphocytic leukemia cells and tts prevention by phorbol ester. Exp. Cell Res. 198,387-372. Fruman, D.A., Mather, P. E., Burakoff, S. J., and Bierer, B. E. (1992). Correlation of calcineurin phosphataes actMty and programmed cell death in murine T cell hybridomas. Eur. J. Immunot. 22, 2513-2517. Gavrteli, Y., Sherman, Y., and Ben-Basson, S. A. (1992). ldentiftcatlon of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J. Cell Biil. 179, 493-501. Golstein, P., Ojcius. D. M., and Young, J. D. (1991). Cell death mechanisms and the immune system. Immunol. Rev. 721,29-85. Green, M. C.. and Shultz, L. D. (1975). Motheaten, an immunodeficient mutant of the mouse. I. Genetics and pathology. J. Hered. 88, 250258.

Hayes, S. M., Shultx, L. D., and Greiner, D. L. (1992). Thymic involution in viable motheaten (me(v)) mice is associated with a loss of intrathymic precursor activity. Dev. Immunol. 2, 191-205. Illera, V. A., Perandones, C. E., Stunz, L. L.. Mower, D. A., Jr., and Ashman. R. F. (1993). Apoptosis in splenic B lymphocytes: regulation by protein kinaae C and IL-4. J. Immunol. 157,2985-2973. Itoh, N., and Nagata. S. (1993). A novel protein domain required for apoptosis: mutational analysis of human Fas antigen. J. Biol. Chem. 288, 10932-l 0937. Itoh, N., Yonehara, S., Ishii, A., Yonehara, M., Mizushima, S.-l., Sameshima, M., Hase, A., Seto, Y., and Nagata, S. (1991). The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis. Cell 88, 233-243. Jarvis, W. D., Kolesnick, R. N., Fornari, F. A., Traylor, R. S., Gewirtx, D. A., and Grant, S. (1994). Induction of apoptotic DNA damage and cell death by activation of the sphingomyelin pathway. Proc. Natl. Acad. Sci. USA 91,73-77. Kagi, D., Vignaux, F., Ledermann, B., Burki, K., Depraetere, V., Nagata, S., Hengartner, H., and Golstein, P. (1994). Fas and perforin pathways as major mechanisms of T cell-mediated cytotoxicky. Science 265528-530. Kim, M. Y., Linardic, C., Obeid, L., and Hannun, Y. (1991). Identification of sphingomyelin turnover as an effector mechanism for the action of tumor necrosis factor alpha and gamma-interferon: specific role in cell differentiation. J. Biol. Chem. 288, 484-489. Knight, C. R., Rees, R. C., and Griffin, M. (1991). Apoptosis: a potential role for cytosolic transglutaminase and tts importance in tumor pro gression. Biochem. Biophys. Acta. 7098, 312918. Kolesnick, R., and Golde, D. W. (1994). The sphingomyelin pathway in tumor necrosis factor and interleukin-1 signaling. Cell 77,32!%328. Lynch, D. H., Watson, M. L., Alderson, M. R., Baum, P. R.. Miller, R. E., Tough, T., Gibson, M., Davis-Smith, T., Smith, C. A., Hunter, K.. Bhat, D., Din, W., Goodwin, R. G., and Seldin. M. F. (1994). The mouse Fas-ligand gene is mutated in gkf mice and is part of a TNF family gene cluster. Immunity 1, 131-138.

Immunity 362

Matthews, Ft. J., Bowne, D. B., Flares, E., and Thomas, M. L. (1992). Characterization of hematopoietic intracellular protein tyrosine phosphatases:description of a phosphatase containing an SH2 domain and another enriched in proline-, glutamicacid-, serfne-, and threonine-rich sequences. Mol. Cell. Biol. 12, 2396-2465. Mathias, S., Dressier, K. A., and Kolesnick, R. N. (1991). Characterization of a ceramide-activated protein kinase: stimulation by tumor necrosis factor a. Proc. Natl. Acad. Sci. USA 88, lC069-10013. Obeid, L. M., Linardic, C. M., Karolak. L. A., and Hannun, Y. A. (1993). Programmed cell death induced by ceramide. Science 259, 1769 1771. Oehm, A., Behrmann, I., Falk, W., Pawlita, M., Maier, G., Klas, C., Li-Weber, M., Richards, S., Dhein, J., Trauth, B. C., Ponstingl, H., and Krammer, P. H. (1969). Purification and molecular cloning of the APO-1 cell surface antigen, a member of the tumor necrosis factor/nerve growth factor receptor family. J. Biol. Chem. 267, 10709-10715. Ogasawara, J., WatanabeFukunaga, R., Adachi, M., Matsuzawa, A., Kasugai, T., Kitamura, Y., Itoh, N., Suda, T., and Nagata, S. (1993). Lethal effect of the anti-Fas antibody in mice. Nature 364, 866-6g9. Orrenius, S., McConkey, D. J., Bellomo, 0.. and Nicotera, P. (1969). Role of Ca% in toxic cell killing. Trends Pharmacol. Sci. IO, 261-265. Plutzky, J.. Neel, B. G., and Rosenberg, R. D. (1992a). Isolation of a src homology P-containing tyrosine phosphatase. Proc. Natl. Acad. Sci. USA 8gs 1123-l 127. Plutzky. J., Neel, B. G., Rosenberg, R. D., Eddy, R. L., Byers, M. G., JaniSait, S., and Shows, T. B. (1992b). Chromosomal localization of an SH2containingtyrosine phosphatase(PTPN6). Genomics 13,669672. Pumiglia, K. M., Lau, L., Huang, C., Burroughs, S., and Feinstein, M. B. (1992). Activation of signal transduction in platelets by the tyrosine phosphatase inhibitor pervanadate (vanadyi hydroperoxide). Biothem. J. 288441-449. Rothe, M., Wong. S. C., Henzel, W. J., and Goeddel, D. V. (1994). A novel family of putative signal transducers associated with the cytoplasmic domain of the 75 kDa tumor necrosis factor receptor. Cell 78, 661-692. Sam&on, L. E.. Davidson, W. F., Morse, M. C., Ill, and Klausner, R. D. (1986). Abnormal tyrosine phosphorylation of T cell receptor in lymphoproliferatiie disorders. Nature 324, 674-676. Schmid, D. S., Tite, J. P.. and Ruddle, N. H. (1966). DNA fragmentation: manifestation of target cell destruction mediated by cytotoxic T-cell lines, lymphotoxin-secreting helper T-cell clones, and cell-free lymphotoxincontaining supernatant. Proc. Natl. Acad. Sci. USA 83, 1661-1665. Shen, S.-H., Bastien. L., Posner, B. I., and Chretien, P. (1991). A protein tyrosine phosphatase with sequence similarity to the SH2 domain of the protein tyrosine kinases. Nature 352, 736-739. Shultz, L. D. (1966). Pleiotropic effects of deleterious alleles at the “motheaten” locus. Curr. Top. Microbial. Immunol. 737, 216-222. Shultz, L. D. (1991). Hematopoiesis and modelsof immunodeficiency. Sem. Immunol. 3.39748. Shultz, L. D., Coman, D. R., Bailey, C. L., Beamer, W. G., and Sidman, C. L. (1984). viable motheaten”: a new allele at the motheaten locus. I. Pathology. Am. J. Pathol. 718, 179-192. Shultz, L. D.. Schweitzer, P. A., Rajan, T. V., Yi, T., Ihle, J. N., Matthews, R. J., Thomas, M. L., and Beier, D. R. (1993). Mutations at the murtne motheaten locus are within the hematopoietic cell protein tyrosine phosphatase (Hcph) gene. Cell 73, 1445-1454. Sidman, C. L., Marshall, J. D., and Allen, R. D. (1969). Murine “viable motheaten” mutation reveals a gene critical to the development of both B and T lymphocytes. Proc. Natl. Acad. Sci. USA 86,6279-6262. Song, Q., and Lavin, M. F. (1993). Calyculin A, a potent inhibitor of phosphatase-1 and+A, prevents apoptosis. Biochem. Biophys. Res. Comm. 19847-55. Streuli, M., Krueger, N. X., Thai, T., Tang, M., and Saito, H. (1996). Distinct functional roles of the two intracellular phosphatase like do. mains of the receptor-linked protein tyrosine phosphatase LCA and LCR. EMBO J. 9.2399-2467.

Suda, T., Takahashi, T., Golstein, P., and Nagata, S. (1993). Molecular doning and expression of the Fas ligand, a novel member of the tumor necrosis factor family. Cell 75, 1169-I 176. Takahashi, T., Tanaka, M., Brannan, C. I., Jenkins, N. A., Copeland, N. G., Suda, T., and Nagata, S. (1994). Generalized lymphoprotiferative disease in mice, caused by a point mutatton in the Fas ligand. Cell 76, 969-976. Takano, Y., Okudaira, M., and Harmon, B. V. (1993). Apoptosis is induced by microtubule disrupting drugs in cultured human lymphoma cells: inhibitory effects of phorbol ester and zinc sulphate. Pathot. Reg. Practice 789,197-203. Trauth, B. C, Ktas, C., Peters, A. M. J., Matxku, S., Moller, P., Falk, W., Debatin, K.-M., and Krammer, P. H. (1969). Monoclonal antibodymediated tumor regression by induction of apoptosis. Science 245. 301305. Tsui, H. W., Siminovitch, K. A., de Sousa, L., and Tsui, F. W. (1993). Motheaten and viable motheaten mice have mutations in the hemato poietic cell phosphatase gene. Nature Genet. 4124-129. Ucker, D. S. (1967). Cytotoxic T lymphocytes and glucocorticoids activate an endogenous suicide process in target cells. Nature 327, 6264. Walker, N. I., Harmon, B. V., Gobe, G. C., and Kerr, J. F. (1966). Patterns of cell death. Meth. Achiev. Exp. Pathol. 73, 16-54. Walker, P. R., Kwast-Welfeld, J., Gourdeau, H., Leblanc, J., Neugebauer, W., and Sikorska. M. (1993). Relationship between apoptceis and the cell cycle in lymphocytes: roles of protein kinase C, tyrosine phosphorylation, and API. Exp. Cell Res. 207, 142-151. Watanabe-Fukunaga, R., Brannan, C. I., Copeland, N. G., Jenkins, N. A., and Nagata, S. (1992). Lymphoproliferatlon disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 358, 314-317. Wyllie, A. H. (196Ct). Glucocorticoid-induced thymocyte apoptosis is associated with endogenous endonuclease activation. Nature 284, 555-556. Wyllie, A. H., Kerr, J. F., and Currie, A. R. (196C1). Cell death: the significance of apoptosis. Int. Rev. Cytol. 68, 251-308. Yi, T., and Ihle, J. N. (1993).Asscdationofhematopoieticcellphoephatase with c-Kit after stimulation with oKit ligand. Mol. Cell. Bid. 13, -. Yi, T., Cleveland, J. L., and Ihle, J. N. (1992). Protein tyroaine phcephatase containing SH2 domains: characterization, preferential expression in hematopoietic cells, and localization to human chromosome 12p12-~13. Mol. Cell. Biol. 12, 836-646. Yi, T., Mui, A. L.-F., Krystal, G.. and Ihle, J. N. (1993). Hematopoietic cell phosphatase associates with the interteukin-3 (11-3) receptor (3 chain and down-regulates IL-3-induced tyrosine phosphoryfatton and mitogenesis. Mol. Cell. Biol. 73, 7577-7566. Yonehara, S., Ishii, A, and Yonehara, M. (1969). A cell-killing mono clonal antibody (anti&s) to a cell surface antigen codownregulated with the receptor of tumor necrosis factor. J. Exp. Med. 769, 17471456. Zhao, Z., Bouchard, P., Diltz, C. D., Shen, S.-H., and Fischer, E. H. (1993). Purification and characterization of a protein tyrosine phoephatase containing SH2 domains. J. Biol. Chem. 288, 2616-2820.