JAK-Mediated Signaling Inhibits Fas Ligand-Induced Apoptosis Independent of De Novo Protein Synthesis¹

Cytokines are proteins generally produced by and/or acting upon cells of the immune system to promote proliferation and function. They exert their effects by triggering specific cell surface receptors that initiate intracellular signaling pathways. Specifically, engagement of cytokine receptors activates receptor-associated JAKs that, in turn, recruit and phosphorylate Stat family members ( 1, 2, 3). Phosphorylated Stat proteins dimerize in certain combinations, translocate to the nucleus, and bind to specific DNA promoter elements to control gene expression ( 1, 2, 3).

Virtually all cell types are able to produce and respond to the cytokines IFN-α and IFN-β, collectively called type I IFNs. Elicited endogenously by infection, type I IFNs promote a wide range of biological effects ( 1, 4, 5, 6, 7). Many studies in a variety of cell types have demonstrated type I IFNs’ ability to inhibit proliferation and/or initiate apoptosis, generating much interest in exploiting these activities for clinical use in treating certain cancers ( 1). However, a few studies have also illustrated the potential of type I IFNs to act as negative regulators of programmed cell death by mechanisms that have yet to be defined ( 8, 9, 10). Indeed, the ability of type I IFN to prevent regulated cell death has considerable implications for its natural function in modulating immune responses. Thus, the effort continues to understand the many activities ascribed to type I IFNs so as to clinically exploit their functions for treating cancers and infectious diseases ( 1).

Receptor-mediated apoptosis is triggered by specific binding of ligand to receptor, i.e., binding of Fas ligand (FasL)³ to Fas receptor (FasR; CD95) ( 11, 12, 13, 14). A number of cell lines, in particular, the H9 T cell lymphoma line and the SKW6.4 lymphoblastoid B cell line, have been used to dissect the molecular events induced via the Fas/FasL pathway ( 12, 13). The death-inducing signaling complex (DISC) is formed in the cytoplasm by sequential recruitment of Fas-associated death domain protein (FADD) and procaspase 8 to the FasR upon its ligation, resulting in activation of procaspase 8 ( 11, 12, 13, 14). Subsequently, executioner caspases, such as caspase 3, are activated, cleaving target proteins, i.e., poly(ADP-ribose) polymerase (PARP), which is necessary for normal cell function and survival ( 11, 14). In the final stages of the apoptotic cascade, nuclear degradation and DNA cleavage are observed ( 11, 12, 13, 14).

Lymphocytes are continually exposed to opposing apoptotic and survival cues within the context of their dynamic environment. In fact, both cytokine and apoptotic signal transduction contribute to the homeostatic control of lymphoid populations, and evidence indicates that these conflicting pathways interact at the level of transcription and translation ( 15). For example, Stat-mediated signaling has been shown to result in the transcriptional activation of genes encoding antiapoptotic protein family members ( 16, 17, 18). However, the present studies were designed to test the hypothesis that cytokine signaling modulates cellular responses to apoptotic stimuli at yet another level, upstream of the effects on gene transcription. In a preliminary report we found that IFN-α-2α protected human T and B cell lines from FasL-induced apoptosis ( 19). We now extend these findings and present evidence that this IFN-α-2α effect is mediated by a novel mechanism, which is dependent on JAK activity and correlates with Stat phosphorylation, yet is independent of de novo protein synthesis. The implications of these findings on the ability of cells to interpret and respond to competing survival and death-signaling processes are discussed.

The human T cell lymphoma line, H9 (HTB 176), and the human B cell lymphoblastoid line, SKW6.4 (TIB-215), were originally obtained from American Type Culture Collection. Both cell lines were cultured in RPMI 1640 medium supplemented with 10% FBS, 2 mM MEM, and 2 mM l-glutamine (Invitrogen Life Technologies) at 37°C in a humidified incubator with 5% CO₂ and manipulated as described below. During all experiments, cells were maintained in the log phase of growth.

H9 or SKW6.4 cells were seeded in 24-well tissue culture plates at a density of 1 × 10⁶ cells in 1 ml of medium (described above). To induce FasL-mediated apoptosis, recombinant human FasL (Kamiya Biomedical) was added to final concentrations of up to 400 ng/ml. After addition of FasL, cells were harvested at 3 h for Western blot analysis or at 24 h for flow cytometric analysis (methods are described below). For experiments requiring IFN-α-2a exposure, cells were incubated with human rIFN-α-2a (Roferon-A; Hoffmann-La Roche) at final concentrations of 10,000–500,000 IU/ml for the times indicated. To inhibit type I IFN-stimulated de novo protein synthesis, cells were incubated with 2 μM cycloheximide (CHX; Sigma-Aldrich) for 1 h before addition of exogenous IFN-α-2a. Inhibition of [³H]alanine incorporation validated that exposure to 2 μM CHX inhibited de novo protein synthesis by 90% under these experimental conditions (data not shown). For inhibition of type I IFN-stimulated JAK activity, the JAK inhibitors tyrphostin AG490, piceatannol, and tyrphostin A1, respectively, were purchased from Sigma-Aldrich, Calbiochem, and LC Laboratories ( 20, 21, 22, 23, 24). Inhibitors were reconstituted in sterile DMSO (Sigma-Aldrich). Cells were incubated with individual JAK inhibitors at final concentrations of 20 μM or with the equivalent volume of DMSO alone for 30 min before addition of IFN-α-2a. Under these conditions, each of the JAK inhibitors suppressed IFN-α-stimulated tyrosine phosphorylation of Stat1 and Stat3 as determined by Western blot (data not shown).

After the specific exposure conditions described, cells were collected, pelleted by centrifugation, washed once in cold PBS, and lysed in 75 μl of cell lysis buffer (30 mM Tris-HCl, 150 mM NaCl, 2 mM EDTA, 1% Triton X-100, 10% glycerol, and 1 mM PMSF, supplemented with Roche Complete Protease Inhibitors according to the manufacturer’s recommendations (Hoffmann-La Roche)). Total proteins from whole cell lysates (60 μg of protein/sample condition) generated from these aliquots were separated by SDS-PAGE, transferred to nitrocellulose, and probed with appropriate primary Abs, i.e., total Stat1 and Stat3 (Santa Cruz Biotechnology), tyrosine-phosphorylated Stat1 (Cell Signaling Technology), tyrosine-phosphorylated Stat3 (Upstate Biotechnology), procaspase 8 and Bcl-2 (BD Biosciences), PARP (Zymed Laboratories), FasR (Kamiya Biomedical), and β-tubulin (Sigma-Aldrich), by conventional methods. The resulting protein-Ab complexes were visualized by chemiluminescence techniques (Amersham Biosciences) after incubation with HRP-conjugated species- and Ig-specific secondary Abs (Amersham Biosciences).

After exposures to IFN-α-2a and/or FasL as described above, cells were collected, pelleted by centrifugation at room temperature, and resuspended in 400 μl of propidium iodide (PI) staining buffer (0.05 mg/ml PI, 0.1% Igepal CA-630, and 0.1% sodium citrate; all from Sigma-Aldrich). Within 1 h of staining, samples were acquired with a FACSCalibur flow cytometer (BD Biosciences) and analyzed for apoptosis, defined as the proportion of the cell population exhibiting a sub-G₀-G₁ DNA content, using CellQuest Pro and ModFit LT software packages (BD Biosciences).

To establish a model system for evaluating the functional relationship between cytokine and apoptosis signal transduction, initial studies determined the sensitivity of human lymphoid cell lines to FasL-induced apoptosis. The T cell lymphoma line H9 and the B lymphoblastoid cell line SKW6.4 were selected for these studies based on their reported ability to rapidly undergo caspase 8-dependent apoptosis upon incubation with a FasR-specific Ab ( 11, 12, 13, 25). To identify conditions of FasL-mediated apoptosis induction, cells were exposed to concentrations of up to 400 ng/ml FasL and examined for well-established indices of early, intermediate, and late apoptotic events, i.e., procaspase 8 cleavage, PARP cleavage, and DNA degradation, respectively. By all these criteria, H9 and SKW6.4 cells consistently showed a concentration-dependent induction of apoptosis. After 3-h incubation with concentrations of FasL between 100 and 400 ng/ml, cells exhibited decreases in procaspase 8 protein levels by Western blot analysis, reflecting its cleavage and activation (Fig. 1,A). Likewise, Western blot analysis of both intact (116-kDa) and cleaved (85-kDa) forms of the PARP protein revealed that FasL stimulated PARP cleavage in a FasL concentration-dependent manner (Fig. 1,A). Finally, the proportions of cells in the later stages of apoptosis were quantitated by flow cytometric analysis of DNA degradation; DNA of intact cells was stained with PI, and cells exhibiting a sub-G₀-G₁ DNA content were enumerated. Consistent with the idea that the extensive DNA fragmentation required for detection by this method is observed late in the apoptotic process, we found that longer incubation times were necessary to observe this effect. Specifically, PI staining and analysis after 3-, 6-, or 12-h incubation of H9 cells with 200 ng/ml FasL showed little, if any, increase in the apoptotic population (data not shown). However, at 24 h, a concentration-dependent effect of FasL on the proportions of apoptotic H9 cells was observed; concentrations of 0, 50, 100, and 200 ng/ml FasL resulted in 1.8 ± 0.7, 5.8 ± 1.5, 10.8 ± 1.0, and 47.5 ± 4.3% apoptosis, respectively (Fig. 1,B). Similarly, SKW6.4 cells incubated with increasing concentrations of FasL for 24 h showed increasing proportions of apoptotic cells; a concentration of 200 ng/ml FasL resulted in 8.5 ± 1.1% apoptosis, compared with 1.2 ± 0.5% apoptosis in unexposed SKW6.4 cells (Fig. 1,B). Additionally, we conducted analysis of DNA degradation by agarose gel electrophoresis; DNA laddering is indicative of degradation and characteristic of the apoptotic process. Although DNA laddering was clearly evident in H9 cells after 24-h exposure to 200 ng/ml FasL, identical exposure conditions for SKW6.4 cells did not result in a visible DNA ladder (data not shown). These results may reflect quantitative differences in the sensitivities of these cell lines to FasL for apoptosis, consistent with the results of PI staining (Fig. 1 B). Thus, although H9 cells appeared to be more sensitive, both the H9 and SKW6.4 cell lines were induced to undergo apoptosis by exposure to FasL.

Having defined incubation conditions to reproducibly induce FasL-stimulated apoptosis, studies next focused to determine the sensitivity of these cell lines to IFN-α-2a. Type I IFNs have been reported to induce apoptosis in selected cell lines ( 1, 26, 27). Therefore, an initial series of experiments assessed the proapoptotic activity of IFN-α-2a in both H9 and SKW6.4 cells. For these studies, cells were incubated in medium containing up to 500,000 IU/ml IFN-α-2a for 48 h. Proportions of apoptotic cells were determined by PI staining, followed by flow cytometric analysis, and were defined as the percentage of cells exhibiting sub-G₀-G₁ quantities of DNA. The results in H9 cells revealed that 48-h incubation with ≤100,000 IU/ml resulted in slight increases in the proportions of apoptotic cells: 1.9, 6.1, and 8.6% apoptosis was observed with exposures of 0, 10,000, and 100,000 IU/ml IFN-α-2a, respectively (Fig. 2,A). However, 21.4% apoptosis was observed upon 48-h incubation of H9 cells with 500,000 IU/ml IFN-α-2a (Fig. 2,A). In contrast, SKW6.4 cells incubated for 48 h in medium containing up to 500,000 IU/ml IFN-α-2a did not exhibit increased apoptosis (Fig. 2 A).

In selected cell culture model systems, exposure to type I IFNs may also result in cytostasis ( 1, 26). Thus, the experiments examined the effects of exposure to IFN-α-2a on the proliferation rates of these cell lines. H9 and SKW6.4 cells were exposed to increasing IFN-α-2a concentrations between 10,000 and 500,000 IU/ml, and the cell proliferation rate was determined by monitoring the cell density at 24-h intervals. At all concentrations tested, IFN-α-2a decreased H9 cell proliferation between 0 and 48 h (Fig. 2,B). In contrast, exposure to IFN-α-2a had no effect on SKW6.4 cell proliferation between 0 and 24 h. Only the highest concentration of IFN-α-2a examined, 500,000 IU/ml, resulted in decreased proliferation of SKW6.4 cells (Fig. 2 B).

It is well established that type I IFNs promote cellular responses via the JAK/Stat signal transduction pathway ( 2, 3, 28). Therefore, the ability of IFN-α-2a to stimulate Stat1 and Stat3 tyrosine phosphorylation in H9 and SKW6.4 cells was examined next. Based on type I IFN responses reported for certain mouse models of viral infection, preliminary experiments examined recombinant human IFN-α-2a concentrations in the range of 10–10,000 IU/ml ( 29). After 1-h of IFN-α-2a exposure, concentrations as low as 10 IU/ml IFN-α-2a induced Stat1 tyrosine phosphorylation in H9 and SKW6.4 cells and Stat3 tyrosine phosphorylation in H9 cells (data not shown). However, reproducible induction of Stat3 tyrosine phosphorylation in SKW6.4 cells required 1,000–10,000 IU/ml. Therefore, subsequent studies were conducted exposing cells to 10,000 IU/ml IFN-α-2a. Cells were incubated with 10,000 IU/ml IFN-α-2a for 0.5, 1, or 4 h, and analyzed by Western blot techniques using phosphotyrosine-specific Abs for these Stat proteins. This study revealed that, as expected, the expression of tyrosine-phosphorylated forms of these molecules was low to undetectable in unexposed cells, but was readily observed in both cell lines within 30 min of IFN-α-2a exposure (Fig. 2,C). By 4 h, phosphorylated forms of Stat were declining (Fig. 2 C). IFN-α-2a stimulation under these conditions did not alter levels of total Stat1 or Stat3 proteins (data not shown).

Taken together, these data demonstrate that in H9 and SKW6.4 cell lines, JAK/Stat-mediated signal transduction in response to type I IFN is intact. Under some of the conditions examined in this study, IFN-α-2a-exposed H9 cells exhibited decreased proliferation and increased apoptosis. However, it is important to note that all subsequent experiments were conducted within 4–24 h after exposure to IFN-α-2a at a final concentration of 10,000 IU/ml. These conditions were specifically selected because they 1) are not conducive to IFN-α-2a-stimulated apoptosis (Fig. 2,A), and 2) coincide with Stat phosphorylation (Fig. 2 C).

The next series of experiments was designed to test the effect of pre-exposing H9 and SKW6.4 cells to IFN-α-2a on subsequent sensitivity to FasL for apoptosis. H9 or SKW6.4 cells were incubated in the presence or the absence IFN-α-2a (10,000 IU/ml) for 1 h before FasL exposure (200 ng/ml). Thereafter, cells were harvested for Western blot analysis or flow cytometry. In both cell lines, IFN-α-2a pre-exposure reduced all hallmarks of FasL-induced apoptosis compared with cells incubated with FasL alone. Western blot analysis of cell lysates generated after 3-h FasL exposure showed that preincubation with IFN-α-2a inhibited the apoptosis-associated degradation of procaspase 8 and PARP proteins (Fig. 3,A). These findings were fully supported by flow cytometry data generated 24 h after addition of FasL, which revealed that IFN-α-2a pre-exposure reduced the proportions of apoptotic cells by >50%. H9 or SKW6.4 cells incubated with or without IFN-α-2a for 1 h and then exposed to 200 ng/ml FasL for 24 h were PI-stained for analysis of apoptosis-induced DNA degradation. The proportion of H9 cells stimulated to undergo apoptosis was reduced from 46.6 ± 3.6% (n = 4) to only 22.5 ± 2.5% (n = 4) by IFN-α-2a pre-exposure (Fig. 3,B). Likewise, the percentage of SKW6.4 cells stimulated to undergo apoptosis in response to FasL was reduced from 7.9 ± 0.8 to 2.8 ± 1.2% by IFN-α-2a exposure (Fig. 3,B). Interestingly, although H9 and SKW6.4 cell populations exhibited different sensitivities to Fas-induced apoptosis (Fig. 1 B), these data indicate that IFN-α-2a treatment reduced the percent apoptosis proportionately, by ∼50%, in both cell lines. These experiments demonstrate that under these conditions, in two independent human cell lines, IFN-α-2a inhibits FasL-induced apoptosis.

Studies next focused on H9 cells for elucidating the mechanism by which IFN-α-2a protected cells from FasL-induced apoptosis. It is well known that sensitivity to apoptosis can be reduced by growth factor- and/or cytokine-stimulated production of antiapoptotic proteins. For example, evidence indicates that under certain conditions Bcl-2 is up-regulated by type I IFNs ( 16, 17, 18). Thus, the expression of this antiapoptotic protein in the presence or the absence of IFN-α-2a was examined. H9 cells were incubated in medium containing 10,000 IU/ml IFN-α-2a for up to 4 h before harvest for Western blot analysis. With IFN-α-2a exposure, the expression of Bcl-2 increased modestly relative to that in unexposed cells (Fig. 4,A). Next, experiments were conducted in the presence of CHX to determine whether the ability of IFN-α-2a to suppress FasL-induced apoptosis required the synthesis of antiapoptotic proteins. Exposure of H9 cells to 2 μM CHX for 1 h inhibited 90% of new protein synthesis, as measured by [³H]alanine incorporation (data not shown). Indeed, this exposure was sufficient to abrogate an IFN-α-2a-stimulated change in Bcl-2 protein expression (Fig. 4,B). However, this same CHX exposure had no effect on the ability of IFN-α-2a to induce Stat1 phosphorylation (Fig. 4,B). Importantly, under these conditions, permitting IFN-α-2a-induced protein phosphorylation, but inhibiting protein synthesis, the ability of IFN-α-2a to suppress FasL-mediated apoptosis was intact. In the presence of CHX, IFN-α-2a retained its ability to reduce apoptosis-associated changes in both procaspase 8 and PARP expression (Fig. 4 C). Thus, these results support a mechanism of action for IFN-α-2a that is independent of new protein synthesis.

Cytokines, including type I IFNs, exert their effects mainly through activation of JAK family kinases that, in turn, directly phosphorylate specific members of the Stat family of transcription factors ( 2, 24, 28, 30). The results presented above suggested that Stat phosphorylation correlated with the ability of IFN-α-2a to inhibit FasL-induced apoptosis. Therefore, experiments in H9 cells were next designed to examine the requirement for JAK activity in the observed IFN-α-2a-mediated inhibition of apoptosis. H9 cells were evaluated for IFN-α-2a effects on FasL-induced apoptosis after being incubated with selected chemical inhibitors of JAK activity, AG490, piceatannol, or tyrphostin A1. Cells were either unexposed or exposed to the JAK inhibitors (20 μM) for 30 min before addition of 10,000 IU/ml IFN-α-2a for 1 h. Preliminary experiments revealed that these exposure conditions were sufficient to inhibit JAK-mediated Stat phosphorylation after IFN-α-2a exposure (data not shown). Cells then were tested for sensitivity to apoptosis induction by FasL exposure at 200 ng/ml for 3 h. Western blot analysis of PARP cleavage demonstrated that exposure to AG490 (Fig. 5,A), piceatannol (Fig. 5,B), or tyrphostin A1 (Fig. 5 C) abrogated all effects of IFN-α-2a observed. Therefore, IFN-α-2a inhibited Fas-mediated apoptosis by a pathway dependent on JAK activity. In total, the findings presented above reveal that IFN-α-2a inhibits FasL-induced apoptosis by a novel mechanism through JAK activation without requiring new protein synthesis.

The immune system has evolved to integrate a multitude of temporal and environmental cues that promote survival, function, or regulatory apoptosis. Toward survival, cytokines are important stimulators of immune cell proliferation and differentiation, whose effects are exerted mainly through JAK/Stat signaling pathways ( 2, 3, 28, 30, 31). Alternatively, for apoptosis, lymphocyte sensitivity to Fas-dependent signaling is an important mechanism for regulating immune cell population size and responses ( 15). Within this context, lymphocytes must continuously maintain this plasticity to quickly increase proliferation and activity in response to pathogens in a tightly controlled manner so as to limit immunopathology ( 18, 32). Therefore, the ability of type I IFNs to desensitize lymphocytes to apoptosis and presumably reinforce proliferative stimuli may be of critical importance because type I IFNs are among the first cytokines produced in response to microbial stimuli ( 6, 7). In this report studies were specifically designed to examine the cytokine IFN-α-2a for its ability to alter lymphocyte sensitivity to FasL-induced apoptosis. The data showed that IFN-α-2a decreased sensitivity to FasL for apoptosis in human cell lines of either T or B cell lineage (Fig. 3). Desensitization to apoptotic stimuli did not correlate directly with changes in Bcl-2 expression (Fig. 4,A). Moreover, these studies show that IFN-α-2a-stimulated protein expression is not necessary for its inhibitory effect on apoptosis (Fig. 4,C). Rather, IFN-α-2a-mediated desensitization to FasL-induced apoptosis did require JAK activity and did correlate with JAK-mediated phosphorylation of Stat protein (Figs. 4,B and 5). Taken together, our data reveal a distinct and previously unappreciated mechanism by which a cytokine, IFN-α-2a, through JAK activation, can attenuate FasL-mediated apoptosis.

Type I IFNs are pleiotropic cytokines whose reported activities include inhibition of proliferation or induction of blastogenesis, promotion of apoptosis, or survival, contrasting effects that appear to be highly cell type specific and context dependent ( 1, 6, 7, 8, 9, 10, 26, 27, 28). We present data in this study demonstrating a role for IFN-α-2a in inhibiting FasL-stimulated apoptosis. These results are surprising given type I IFNs’ reported antiproliferative effect. However, the mechanisms underlying the cytostatic effects of type I IFNs are not well understood ( 1, 28). Reported in this study are results demonstrating that within a few hours of IFN-α-2a exposure, these T and B cell lines were relatively resistant to FasL-induced apoptosis. It required relatively longer incubation times and higher concentrations, i.e., 48 h and 500,000 IU/ml, for observation of reduced proliferation and increased apoptosis in response to IFN-α-2a alone (Fig. 2, A and B). Thus, it is important to emphasize that our studies, revealing an antiapoptotic effect of IFN-α-2a, focused on rapid events, occurring within 4 h of an exposure to a physiological concentration, concomitant with peak IFN-α-2a-simulated signaling activity and independent of new protein synthesis. Understanding how these opposing responses to physiological concentrations of type I IFN-α are regulated to shape the numerous biological functions for this cytokine family awaits a more thorough definition of the molecular mediators involved. Interestingly, we have found that, similar to IFN-α-2a, IL-2 inhibited FasL-mediated apoptosis of H9 cells independent of de novo protein synthesis (L. P. Cousens and J. W. Darnowski, unpublished observations). Thus, the paradigm established in this study for apoptosis modulation downstream of JAK activity may broadly apply to other selected cytokines as well.

The JAK/Stat pathway is only one of many signal transduction pathways promoting cell survival and proliferation. It is not surprising, then, that certain other growth factor-mediated signaling pathways, i.e., those transduced by MAPK and PI3K cascades, are also reported to suppress apoptosis ( 33, 34, 35, 36, 37). Particularly relevant to the findings presented in this study is the evidence that MAPK and PI3K also mediate their protective effects independent of new protein synthesis ( 15, 34, 35). The data reported in this study are the first to describe a signaling-based mechanism for modulating FasL-induced apoptosis specifically dependent upon cytokine-stimulated JAK activity (Fig. 5). We acknowledge the possibility that chemical inhibitors of the JAK family may affect other signaling pathways. Thus, we present data for three such JAK inhibitors used in three independent experiments. These data consistently support the conclusion that JAK activity is required for IFN-α-2a-mediated inhibition of FasL-induced apoptosis. Moreover, we have conducted preliminary experiments testing the ability of IFN-α-2a to inhibit FasL-stimulated apoptosis in the presence of either PD98059 or wortmannin, chemicals that inhibit MAPK and PI3K cascades, respectively. That IFN-α-2a suppressed apoptosis in the presence of these inhibitors (L. P. Cousens and J. W. Darnowski, unpublished observations), but not in the presence of JAK inhibitors (Fig. 5), was additional proof that we have uncovered a mechanism for attenuation of apoptosis specifically mediated by the JAK/Stat pathway.

Studies in this laboratory continue to focus on characterizing the link between JAK/Stat- and the FasR-mediated signal transduction pathways. In this report we show evidence that IFN-α-2a inhibits activation of procaspase 8. Experiments aimed at characterizing JAK-mediated effects on DISC assembly are ongoing, but preliminary findings indicate that in the presence of IFN-α-2a, FasL-induced recruitment of procaspase 8 to DISC is decreased (L. P. Cousens and J. W. Darnowski, unpublished observations). There is some evidence in the literature that the functions of both FasR and Fas-associated death domain protein can be modulated by phosphorylation of critical serine and/or threonine residues ( 25, 38, 39, 40). It is interesting to speculate that the activity of these death complex components can likewise be regulated by tyrosine phosphorylation, and perhaps they are direct targets of JAKs. Alternatively, a role for activated Stat1 interacting directly with death-associated signaling components downstream of the TNF receptor has been reported ( 41). Thus, IFN stimulation of Stat phosphorylation may prevent efficient transmission of the death signal by direct physical interference in DISC assembly. These hypotheses are currently under investigation.

Lymphocyte proliferation and function are tightly controlled in preparation for initiation of an immune response. For these cells to quickly integrate transient and/or overlapping signals induced by immunological stimulation, the ability to interrupt apoptosis without requiring new protein synthesis may be of particular importance. For example, recent evidence supports a role for FasR/FasL interactions in promoting certain aspects of T cell development and function ( 42, 43, 44, 45, 46, 47, 48). For Fas-mediated signaling to impact upon nonapoptotic functions of viable cells, regulatory mechanisms limiting the apoptotic effects of FasL exposure must be considered. For a regulatory mechanism to be effective in this context, it must be triggered before a cell is committed to death, and it must specifically compete with the enzyme-driven apoptotic cascade to inhibit cell death selectively. Over an extended time, cytokines are clearly capable of stimulating the production of numerous antiapoptotic proteins to promote survival ( 15, 16, 17, 18). However, a signaling-based mechanism that rapidly disrupts an apoptotic cascade independent of new protein synthesis, as described in this report, may provide a window of opportunity for the cell to more rapidly respond to multiple and diverse external cues. Based on the data presented in this report, we propose that the ability of JAK activity to inhibit FasL-mediated apoptosis without new protein synthesis may enable lymphocytes to maintain a low apoptotic threshold that can be raised quickly and transiently by cytokines elicited upon immunological challenge. Because a number of human pathologies have been associated with inhibition of apoptosis, including certain neoplasias and autoimmune disorders, it is important to continue to increase our understanding of the interrelationships of the signaling pathways involved ( 49, 50, 51). Understanding these signaling pathway relationships will be important for dissecting molecular mechanisms by which the distinct cascades for cytokine-mediated responses, on the one hand, and caspase activation, on the other, are, in fact, linked to integrate life and death signals at the cellular level.