Focal adhesion kinase (FAK) is important to cellular functions such as proliferation, migration, and survival of anchorage-dependent cells. We investigated the role of FAK in modulating normal cellular responses, specifically cell survival in response to inflammatory stimuli and serum withdrawal, using FAK-knockout (FAK−/−) embryonic fibroblasts. FAK−/− fibroblasts were more vulnerable to TNF-α-induced apoptosis, as measured by terminal deoxynucleotidyl transferase positivity. FAK−/− fibroblasts also demonstrated increased procaspase-3 cleavage to p17 subunit, whereas this was undetectable in FAK+/+ fibroblasts. Insulin receptor substrate-1 expression was completely abolished and NF-κB activity was reduced, with a concomitant decrease in abundance of the anti-apoptotic protein Bcl-xL in FAK−/− cells. Upon serum withdrawal, FAK+/+ cells exhibited marked attenuation of basal ERK phosphorylation, while FAK−/− cells, in contrast, maintained high basal ERK phosphorylation. Moreover, inhibition of ERK phosphorylation potentiated serum withdrawal-induced caspase-3 activity. This was paralleled by increased insulin receptor substrate (IRS)-2 expression in FAK−/− cells, although both insulin- and IGF-1-mediated phosphorylation of Akt/PKB and GSK-3 were impaired. This suggests that IRS-2 protects against apoptosis upon serum withdrawal via the ERK signaling pathway. The specific role of FAK to protect cells from apoptosis is regulated by activation and phosphorylation of NF-κB and interaction between activated growth factor anti-apoptotic signaling pathways involving both phosphatidylinositol 3-kinase/Akt and MAPK/ERK1/2. We demonstrate that FAK is necessary for upregulation of the anti-apoptotic NF-κB response, as well as for normal expression of growth factor signaling proteins. Thus we propose a novel role for FAK in protection from cytokine-mediated apoptosis.
there are several known mechanisms involved in the induction of apoptosis, depending on extracellular apoptotic signals. First, Fas-dependent signaling is considered the major mechanism for apoptosis (22). Second, when cells undergo genotoxic stress, stress-response MAP kinase is activated, resulting in p53-dependent release of cytochrome c from mitochondria and the activation of caspase-9 (33, 42, 46). Finally, apoptosis occurs when growth or survival factors are withdrawn from the culture medium (10, 68).
Tumor necrosis factor-α (TNF-α), an inflammatory cytokine, induces a range of biological responses associated with apoptosis/survival in many cells and is a potent initiator of both apoptosis and necrosis (11, 17, 36). Following binding to TNF-receptor-1 (TNFR-1), TNF-α recruits TNFR-associated death domain, which leads to the recruitment of either Fas-associated death domain (FADD) also known as TNFR-associated factor-2 or receptor interactive protein (RIP). FADD mediates apoptotic signals (via a Fas-dependent signaling pathway) through activation of initiator caspase-8 (18, 40). Caspase-8 subsequently activates the apoptotic caspase cascade in a mitochondrion-independent manner (9, 55). However, activation of TNFR-1 potentially activates NF-κB, which subsequently activates anti-apoptotic pathways, indicating that TNFR-1-induced apoptosis typically requires downregulation of anti-apoptotic NF-κB response (25, 64). FADD also mediates survival signals via activation of NF-κB, inducing a number of anti-apoptotic genes, including Bcl-2, Bcl-xL, inhibitory apoptosis proteins, and Gadd 45 (24).
Insulin and insulin-like growth factor-1 (IGF-1) are major circulating survival factors in vivo, and have been shown in several cell types in vitro to protect against apoptosis induced by various stimuli, including serum withdrawal and exposure to TNF-α (34, 67). Based on observations of 3T3 cells overexpressing IGF-1, IGF-1 positively regulates cell survival signaling pathways via the insulin receptor substrate (IRS)-phosphatidylinositol 3 (PI3)-kinase-PKB/Akt cascade (60). Recent studies using brown preadipocytes derived from different IRS knockout animals suggested that insulin receptor substrate-1 (IRS-1) has a critical function in mediating the anti-apoptotic effects of IGF-1 and insulin (59).
Focal adhesion kinase (FAK) is a nonreceptor protein-tyrosine kinase that is activated upon integrin engagement (49). Initial studies in tumor cells showed that FAK positively regulates cell survival signaling (7, 19, 53, 66). It has been previously reported that there is crosstalk between FAK and growth factor signaling pathways (21). We recently showed that both the focal adhesion targeting property and tyrosine kinase activity of FAK are essential for normal insulin-mediated stimulation of glycogen synthesis in hepatoma cells (29) and glucose uptake in skeletal muscle cells (30). It is thus important to further define the role of FAK in growth factor signaling in various physiological processes including cell growth and survival.
In this study, we test the hypothesis that FAK is protective against apoptosis. We find that FAK positively regulates cell survival signaling to prevent apoptosis upon cytokine exposure, via activation of NF-κB and by maintaining expression of IRS-1 and Bcl-xL. However, in contrast, the presence of FAK enhances cell susceptibility to apoptosis following serum withdrawal.
Materials and supplies.
Antibodies to FAK, GSK-3β, and RIP were obtained from BD Biosciences (San Jose, CA). Anti-IR-1, anti-IRS-2, and anti-p85 polyclonal antibodies were obtained from Upstate Biotechnology (Lake Placid, NY). Antibody to Akt and anti-phospho-specific Akt polyclonal antibody, which recognizes serine473 phosphorylated Akt, as well as anti-phospho-specific GSK-3β polyclonal antibody which recognizes serine9 phosphorylated GSK-3β, anti-phospho-ERK, anti-ERK antibodies, anti-phosph-NF-κBp65 which recognizes serine536, anti-NF-κBp65, anti-NF-κBp105/50, PD-98059 (MEK1/2 inhibitor), and horseradish peroxidase (HRP)-conjugated anti-rabbit secondary antibodies were from Cell Signaling Technology (Beverly, MA). Anti-phospho-specific FAK polyclonal antibody which recognizes tyrosine397 phosphorylated FAK and anti-phospho-specific IRS-2 polyclonal antibody, which recognizes serine731 phosphorylated IRS-2 was obtained from Biosource International (Camarillo, CA). Antibody against phosphotyrosine (pY99) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Chemiluminescence kit, protein G-conjugated agarose beads, HRP-conjugated anti-mouse were all purchased from Pierce (Rockford, IL). CellTiter 96AQueous Non-Radioactive Cell Proliferation Assay kit was obtained from Promega (Madison, WI). All cell culture reagents were obtained from Invitrogen (Carlsbad, CA). Nitrocellulose membrane was from Schleicher and Schuell (Keene, NH). [U-14C]-d-glucose was from Amersham (Piscataway, NJ). [γ32P]-ATP was obtained from NEN Life Science Products (Boston, MA). UDP-[U-14C]glucose (303 mCi/mM) was from ICN (Irvine, CA). Z-VAD-FMK and all other chemicals were purchased from Sigma (St. Louis, MO) unless otherwise stated.
Measurement of programmed cell death in FAK+/+ and FAK−/− fibroblasts.
FAK−/− and FAK+/+ cells were seeded on 6 cm dishes and treated with 0, 0.015, 0.15, 1.5, 15, and 150 ng/ml TNF-α, 150 ng/ml TNF-α with 30 μM Z-VAD (a caspase-3 inhibitor), or 30 μM Z-VAD alone for 1 h. Cells were stained with flourescein isothiocyanate-annexin V and then analyzed by flow cytometry for quantitative cell death. For qualitative measurements, cell cultures were viewed under a light microscope. Samples were examined using an Axiophot microscope (Zeiss, Thornwood, NY).
Nonradioactive cell proliferation assay.
A modification of the technique of Mosmann et al. (41) was employed. For each condition, 105 cells were used, and the assay was performed according to the manufacturer's instructions (Promega). Briefly, FAK+/+ or FAK−/− cells were grown in 6-well plates for 24, 72, and 120 h. After treatment with or without 150 ng/ml TNF-α for 1 h, 200 μl of MTS/PMS solution were added into each well of 6-well assay plates containing 1 ml of culture medium and incubated at 37°C for 1 h. Plates were read on a Universal Microplate Reader (Bio-Tek Instruments), using a test wavelength of 490 nm. After various durations of culture, cells were trypsinized and counted in a hemocytometer.
FAK−/− and FAK+/+ fibroblasts were seeded and grown to 70–80% confluence in 60 mm dishes and then pretreated for 30 min with the ERK inhibitors: 40 nM PD-98059 (MEK1/2 inhibitor). The inhibitors were also present during subsequent exposure to TNF-α for 24 h.
FAK phosphorylation in detached cells.
Cells were washed twice with phosphate-buffered saline (PBS) containing 1 mM EDTA, and detached gently from the dishes using a rubber policeman. They were washed once in PBS and then maintained in suspension in PBS for 10 min at 37°C. Cells were dipped into ice and centrifuged at 3,000 g for 3 min at 4°C. Cell solubilization and immunoprecipitation of lysates were performed as described above using anti-phospho-specific FAK polyclonal antibody which recognizes tyrosine397 phosphorylated FAK or with antibodies to RIP.
Immunoprecipitation and Western blot analysis.
For cell lysate preparation, RIPA lysis buffer was used (50 mM Tris·HCl, pH 7.4, 150 mM NaCl, 0.5% deoxycholate, and 1% Nonidet P-40) containing 20 μg/ml aprotinin, 20 μg/ml leupeptin, 1 mM sodium orthovanadate, 10 mM NaF, 10 mM sodium pyrophosphate, and 1 mM phenylmethylsulfonylfloride. The protein concentration of lysates was determined by using the DC protein assay kit (Bio-Rad). For immunoprecipitation, lysates (500 μg of total protein) were precleared with protein G-agarose beads (Pierce) at 4°C for 1 h and then incubated with 10 μl of antibody for 2 h, followed by overnight incubation with protein G-agarose beads at 4°C. Precipitates were washed three times in lysis buffer, and beads were resuspended in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample loading buffer, boiled for 5 min, and resolved by SDS-PAGE. Proteins were transferred to polyvinylidene difluoride membrane, probed with appropriate antibody, and detected by Supersignal chemiluminescence kit (Pierce). All experiments were repeated at least 3 times.
Preparation of nuclear extracts from FAK−/− and FAK+/+ fibroblasts.
FAK−/− and FAK+/+ fibroblasts were seeded on 6 cm dishes and washed with 5 ml ice-cold PBS/phosphatase inhibitors containing 10× PBS, distilled water, and phosphatase inhibitors (Active Motif, Carlsbad, CA). Cells were transferred to 15-ml conical tubes and centrifuged for 5 min at 500 rpm at 4°C. Supernatant was then discarded and the pellet resuspended in 500 μl 1× hypotonic buffer (Active Motif). After 15 min of incubation on ice, 25 μl of detergent was added and cells were vortexed for 10 s at the highest setting. This was followed by another centrifugation for 30 s at 14,000 g at 4°C. The supernatant was then removed and the pellet resuspended in 50 μl of Complete lysis buffer (Active Motif) (10 mM DTT, lysis buffer AM1, protease inhibitor cocktail) and vortexed for 10 s. The resulting suspension was incubated for 30 min on ice followed by a 30-s vortex and then centrifuged for 10 min at 14,000 g at 4°C. Nuclear fractions were collected in the supernatant.
Northern blot analysis.
Total RNA was extracted using RNA STAT-60 reagent (Tel-Test). Total RNA (30 μg) was denatured with formaldehyde/formamide and resolved by electrophoresis on 1% agarose gels containing formaldehyde. RNA was transferred onto a Hybond-N+ membrane (Amersham Pharmacia Biotech), that was hybridized at 65°C for 18 h. Probes used were cDNA fragments of IRS-2 (3.5 kDa) and GAPDH (1.3 kDa) (Sigma) and were labeled with [α-32P]dCTP using a random primer labeling kit (Amersham Pharmacia Biotech).
Electrophoretic mobility shift assays.
A double-stranded oligonucleotide containing the NF-κB DNA-binding consensus sequence: 5′-AGT TGA GGG GAC TTT CCC AGG C-3′ and 3′-TCA ACT CCC CTG AAA GGG TCC G-5′ was used to study the DNA binding activity of NF-κB by electrophoretic mobility shift assays (EMSA), as described elsewhere (14). For supershift assays, 5 μg of nuclear extract were incubated with 1 μg of specific antibody against the p65 component of NF-κB for 30 min at room temperature and analyzed by EMSA.
Statistical comparisons were conducted using Student's two-tailed t-test for paired or unpaired samples as indicated, with appropriate correction for multiple comparisons with the use of statistical software (SigmaStat version 2.0, SPSS). Data are reported as means ± SE.
TNF-α treatment induces caspase-dependent programmed cell death in FAK−/− fibroblasts.
FAK−/− cells showed dose-dependent programmed cell death to TNF-α, which was attenuated by addition of Z-VAD (the cell-permeable inhibitor of caspase that has been shown to inhibit caspase-mediated apoptosis) (4), whereas FAK+/+ cells showed little response to increased concentrations of TNF-α or to the addition of Z-VAD vs. nontreated controls (Fig. 1A). Light micrographs were used to view necrotic cells of the same cultures and showed similar qualitative results (data not shown). TNF-α-treated cells alone showed a dose-dependent altered morphology, with many cells rounding up. Combined TNF-α and Z-VAD treatment had a minimal effect on cell morphology (data not shown). To further document this synergistic effect, we determined the level of one of the key executioners of apoptosis, caspase-3, in cells exposed to increasing doses of TNF-α alone or in combination with Z-VAD (Fig. 1B). Cleaved caspase-3 fragments appeared at higher TNF-α concentrations but disappeared when Z-VAD was added to FAK−/− cells. Almost no cleavage fragments were observed in any of the conditions tested in FAK+/+ cells. We conclude that FAK protects against TNF-α-mediated apoptosis. To determine whether TNF-α also affects cell proliferation rates differently in the two cell lines, we performed a cell proliferation assay. Although the growth rate and proliferation of FAK−/− cells was significantly reduced compared with FAK+/+ cells, TNF-α itself did not affect cell growth rate and proliferation in either cell line (Fig. 1C). This confirms that the increased rate of apoptosis seen in FAK−/− cells was not due to TNF-α-mediated inhibition of cell growth and/or proliferation.
Growth factor treatment reduces caspase-3 activity in FAK−/− fibroblasts in response to TNF-α.
We examined the ability of FAK to protect cells from activation of caspase-3 after 12, 48, and 72 h in the presence or absence of TNF-α. As shown in Fig. 2A, cleavage fragments were undetectable in FAK−/− cells but were detected in FAK+/+ cells that were serum starved for >24 h. Furthermore, cleavage fragments appeared in a time-dependent manner in FAK−/− cells that were TNF-α treated for >24 h. Also, insulin and IGF-1 were analyzed for their ability to protect cells from TNF-α induced activation of caspase-3. As shown in Fig. 2B, FAK−/− cells once again showed caspase-3 cleavage fragments in response to overnight TNF-α treatment. However, this caspase-3 activity was reduced by addition of 100 nM insulin or 50 ng/ml IGF-1, while FAK+/+ fibroblasts showed no caspase-3 activity in response to overnight TNF-α treatment. These results indicate that FAK protects against apoptotic processes induced by cytokines and that, in the absence of FAK, growth factors, such as insulin and IGF-1, can protect cells against cytokine-induced death.
NF-κB phosphorylation and activation are impaired and the anti-apoptotic protein Bcl-xL is downregulated in FAK−/− fibroblasts in response to TNF-α.
We tested whether FAK knockout diminishes NF-κB activation, as well as altering the balance between pro- and anti-apoptotic events, by measuring expression of the anti-apoptotic protein Bcl-xL in FAK−/− cells. TNF-α induces the phosphorylation of NF-κB p65 at serine536. The phosphorylation of NF-κB p65 leads to its acetylation, which plays a key role in the NF-κB transcriptional activity (8). The phosphorylation of NF-κB p65 was detected in the cells after 12, 48, and 72 h in the presence or absence of TNF-α. As shown in Fig. 2A, TNF-α-induced phosphorylation of NF-κB at serine536 was markedly decreased in FAK−/− cells. An electromobility shift assay (EMSA) with a radiolabeled probe containing a consensus NF-κB binding site was performed to measure the binding activity of NF-κB. Upon stimulation by TNF-α, NF-κB normally translocates from the cytoplasm into the nucleus and its DNA-binding activity increases (38). As shown in Fig. 3A, FAK−/− cells showed impaired activation and translocation of NF-κB, which was unaffected by insulin or IGF-1 treatment. To further confirm the specificity of this NF-κB-DNA complex, a supershift assay was performed on nuclear extracts.
The results were similar to those seen in the positive controls (Fig. 3B). Translocation of NF-κB to the nucleus was reduced in nucleus extracts from FAK−/− cells compared with FAK+/+ cells. Moreover, this attenuation was partially reversed by addition of insulin (100 nM) or IGF-1 (50 ng/ml) as shown in Fig. 3C. As shown in Fig. 4, TNF-α treatment induced a much greater reduction of Bcl-xL expression in FAK−/− cells, which was again unaffected by insulin or IGF-1 treatment. These results indicate that FAK regulates TNF-α-induced NF-κB phosphorylaion and activation and the balance between pro- and anti-apoptotic events.
FAK-RIP complex disassociates upon dephosphorylation of FAK.
It has been reported that FAK binds to RIP and provides survival signals to cells and suppresses RIP-dependent pro-apoptotic signaling (35). To test the hypothesis that the phosphorylation of FAK play a key role in FAK-RIP association, immunoprecipitation was performed on both suspended and attached FAK+/+ and FAK−/− cells. As shown in Fig. 5, the phosphorylation of FAK at tyrosine 397 was totally abolished in FAK+/+ cells suspended in PBS for 10 min compared with that in the attached cells. FAK-RIP association was also not detected in suspended FAK−/− cells. This result supports the findings of Kurenova et al. (35), suggesting that FAK provides anti-apoptotic signals to cells via suppression of fas-dependent signaling pathways.
Insulin-mediated Akt/PKB-ser473 and GSK-3β-Ser9 phosphorylation is impaired in FAK−/− fibroblasts.
The activation of PKB/Akt by serine phosphorylation plays a major role in cell survival by inhibiting apoptosis induced by several stimuli, including serum withdrawal (13). PKB/Akt modulates the activation of transcription factors such as NF-κB and Akt phosphorylates GSK-3β and affects other transcription factors such as AP-1 (13). We therefore analyzed phospho-Akt and phospho-GSK-3β by Western blot analysis in FAK−/− and FAK+/+ cells. As shown in Fig. 6, dose-dependent increases in phospho-serine9 and serine473 were seen in FAK+/+ fibroblasts after 1-h exposure to 0.1 nM-1.0 μM insulin, which was markedly attenuated in FAK−/− cells. There was no difference in basal phosphorylation of Akt or GSK-3β between the two cell lines. This data suggests that FAK is required for normal insulin-mediated phosphorylation of Akt/PKB and GSK-3β.
FAK−/− increases ERK phosphorylation after serum deprivation.
Because several lines of evidence suggest that integrin activation of signaling pathways involving PI 3-kinase, ERK, and c-Jun NH2-terminal kinase requires FAK (2, 12, 32, 51, 56, 69), and that the basal levels of activated Akt and/or ERK seem to be important for protecting cells against apoptosis induced by serum removal (23), we measured the phosphorylation of the cell survival signaling protein, ERK, in FAK−/− cells. As shown in Fig. 7A, we found that after serum deprivation for 4 h, ERK phosphorylation was unchanged compared with that seen in normal media culture. ERK phosphorylation, however, was markedly reduced in FAK+/+ fibroblasts. This indicates that cellular adaptations occurring in the absence of FAK expression lead to upregulation of ERK phosphorylation and protection against apoptosis caused by serum withdrawal. To further confirm this finding, we treated cells with or without TNF-α in the presence or absence of PD-98059 (MEK1/2 inhibitor) and then analyzed caspase-3 and phosphorylation of ERK. As shown in Fig. 7B, caspase-3 activity was markedly increased in both FAK−/− and FAK+/+ cells treated with PD-98059 alone after serum deprivation, in parallel with the suppression of ERK phosphorylation.
IRS-1 is not expressed but IRS-2 expression is upregulated in FAK−/− fibroblasts.
Since our data suggests that FAK knockout inhibits apoptosis induced by serum withdrawal, we hypothesized that another IRS, primarily acting via the ERK/MAPK cascade, may be upregulated in response to the previously observed abolition of IRS-1 in FAK−/− cells (37). We therefore analyzed mRNA expression and protein abundance of IRS-1 and IRS-2 in both FAK−/− and FAK+/+ cell lines through Northern and Western blot analysis, respectively. First, as shown in Fig. 8, IRS-1 mRNA and protein were not expressed in FAK knockout cells, confirming the findings of Lebrun et al. (37). However, IRS-2 protein (Fig. 9A) and mRNA (Fig. 9B) exhibited a significant increase in FAK−/− cells, while the insulin-stimulated phosphorylation of IRS-2 was not different between two cell lines (Fig. 9C). This result is very similar to that observed in IRS-1 knockout mice (62). This suggests that, in the presence of FAK (and therefore normal levels of IRS-1) cells are susceptible to apoptosis mediated by serum withdrawal, but when FAK and IRS-1 are absent, the increase in IRS-2 confers resistance to apoptosis, presumably by recruitment of the ERK/MAPK signaling pathway.
This study demonstrates that FAK regulates the TNF-α-induced apoptotic signaling pathway via modulation of NF-κB in embryonic fibroblasts. Activation of caspase-3 and suppression of Bcl-xL were also modulated by this pathway. To our knowledge, this is the first report demonstrating the involvement of FAK in cell survival via growth factor and Fas-family death receptor signaling.
FAK-dependent signaling events promote cell survival by offering protection against death induced by stressors such as cytotoxic drugs and TNF-α (7, 15, 31, 48, 54). Antisense attenuation of FAK expression and antibodies to FAK lead to cell detachment and apoptotic death in various cell types (32, 33). In addition, FAK-related non-kinase (FRNK) is expressed as an independent protein and consists of the carboxyl-terminal noncatalytic domain of FAK (50). It has been utilized as a dominant-negative mutant to inhibit FAK signaling (57). FRNK is selectively expressed in smooth muscle cells, with particularly high levels being observed in conduit blood vessels, and FRNK expression is dramatically increased during vascular development and following vessel injury (50). FRNK overexpression results in inhibition of cell proliferation, loss of cell adhesion, and activation of caspase-mediated cell death (34). TNF-α is a potent activator of the transcription factor NF-κB and is involved in the immune responses central to inflammation as well as pro- and anti-apoptotic processes (25, 64). We now demonstrate that the absence of FAK results in the activation of TNF-α-induced caspase-mediated cell death. NF-κB activation was severely impaired in FAK−/− cells, and these results support the findings of Funakoshi-Tago et al. (20, 35). Moreover, TNF-α induced phosphorylation of NF-κB at serine 536 was also markedly decreased in FAK−/− cells, suggesting that FAK plays a role in TNF-α-induced NF-κB activation and its transcriptional activity.
As expected, FAK−/− fibroblasts were hypersensitive to apoptosis provoked by TNF-α alone, since they showed an attenuated response of NF-κB and upregulation of Bcl-xL expression. TNF-α also elicited caspase-3-dependent cell death in FAK−/− cells. Unexpectedly, TNF-α-induced caspase-3 activation was attenuated by insulin or IGF-1 treatment in the absence of FAK. This may be due in part to the upregulation of IRS-2 expression. Interestingly, insulin-stimulated phosphorylation of Akt/PKB and GSK-3β were also impaired, suggesting that upregulation of IRS-2 cannot totally compensate for loss of the IRS-1 function. Other studies on the metabolic functions of the IRS proteins also suggest that IRS-1 and IRS-2 cannot substitute functionally for one another (6, 16).
Hoeflich et al. (27) reported that fibroblasts from GSK-3β-deficient embryos were hypersensitive to TNF-α and showed reduced NF-κB function. Their data indicated that GSK-3β is required for protection from TNF-α-mediated cytotoxicity. The activation of Akt/PKB by serine phosphorylation plays a major role in cell survival by inhibiting apoptosis induced by a number of stimuli, including serum withdrawal (13). Akt/PKB modulates the activation of transcription factors such as NF-κB, phosphorylates GSK-3β, and also regulates other transcription factors, such as AP-1 (13). Our data show that phosphorylation of both Akt/PKB and GSK-3β were impaired in response to insulin stimulation in FAK−/− cells, suggesting that FAK is involved in mediating cell survival via Akt/PKB-GSK-3β.
TNF-α plays a dual role by triggering both cellular survival and induction of apoptosis, depending on complex environmental factors (3). While activation of NF-κB and its target genes are essential for the suppression of TNF-induced apoptosis, it is also equally important for the propagation of action of pro-inflammatory cytokines, such as IL-6 and IL-8 (5, 52, 58, 63, 65). Regulation of these genes may result in either activation or suppression of NF-κB, contributing to the balance between pro- and anti-apoptotic events (61). In our study, we demonstrated that the expression of the anti-apoptosis protein Bcl-xL was reduced in response to TNF-α treatment in FAK−/− cells, indicating an imbalance between pro-and anti-apoptotic events.
According to Sonoda et al. (21), the PI3K-Akt survival pathway, activation of NF-κB, and increase of inhibitory apoptosis proteins are all involved in FAK-mediated resistance to apoptosis. In addition, several growth factors have been identified as regulators of cell survival (47). Among these, IGF-1 and insulin have been reported to protect a broad range of cells from a variety of proapoptotic stimuli (45). Tyrosine phosphorylation of IRS proteins is an early step in signal transduction by both IGF-IR and insulin receptors to activate several distinct signaling pathways, such as the PI3-K and MAPK cascades (39). We investigated these signaling pathways to elucidate the relationship between FAK and growth factor-mediated anti-apoptosis. The absence of FAK unexpectedly induced cells to completely resist apoptosis caused by serum deprivation. This involved in upregulation of IRS-2 expression and enhancement of both basal and stimulated ERK activation. The inhibition of ERK phosphorylation with PD-98059 markedly attenuated the basal high level of phosphorylation of ERK and potentiated serum withdrawal induced caspase-3 activity.
In FAK−/− cells, in which the expression of IRS-1 mRNA and protein was totally undetectable, we found a twofold increase in IRS-2 mRNA and protein, similar to observations in adipocytes of IRS-1−/− mice, in which IRS-2 protein expression was 15–25% higher (62). Interestingly, when FAK−/− cells were treated with insulin or IGF-1 for 4 h, ERK phosphorylation was attenuated and accompanied by a downregulation of IRS-2 expression. Hirashima et al. (26) reported that insulin represses IRS-2 gene expression via a PI3-kinase/Akt but not a MAP kinase-ERK kinase pathway. Moreover, Huang et al. (28) recently reported that in L6 myotubes, in which IRS-2 cognate protein expression was reduced by 75% when using small interfering RNA-mediated specific gene silencing, insulin-induced ERK phosphorylation was much more sensitive to loss of IRS-2 than IRS-1. Valverde et al. (62) also found that insulin failed to rescue IRS-2−/− hepatocytes from serum withdrawal-induced apoptosis. According to Nagle et al. (43), a higher incidence of apoptosis was observed in IRS-2−/− tumors than in wild-type tumors, both in the presence and absence of growth factors. In contrast, cells expressing only IRS-2 (IRS-1−/−) were more resistant to stress-induced apoptosis (43). Thus it appears that IRS-2 is more critical than IRS-1 in providing resistance to serum withdrawal-induced apoptosis. Since it is known that IGF-1 and insulin have a potent ability to protect embryonic fibroblasts from apoptosis due to serum deprivation via the IRS-2-MAP kinase/ERK pathway (43). Ajenjo et al. (1) reported that constitutively active ERK2 localized in the cytoplasm inhibits apoptosis induced by serum starvation of chronic myelogenous leukemia cells. Also, Lilian et al. (44) reported that ERKs, which have been phosphorylated by bisphosphonates, then phosphorylate their cytoplasmic substrates, the pro-apoptotic proteins BAD and C/EBP-β. Phosphorylation of BAD renders it inactive, whereas phosphorylation of C/EBP-β leads to binding of pro-caspases, thus inhibiting apoptosis-mediated osteocyte survival (44). It is probable that FAK is involved in growth factor-mediated protection of embryonic cells from apoptosis induced by serum withdrawal via its essential role to permit normal IRS-1 expression. When FAK is absent, compensatory upregulation of IRS-2 expression protects against apoptosis due to serum withdrawal by preferential activation of the ERK/MAPK signaling pathway.
In summary, this study demonstrates for the first time that FAK offers cell protection by favorably modulating anti-apoptosis mediated by growth factors and is involved in growth factor-mediated survival via the IRS-MAP kinase/ERK signaling pathway, as well as protecting against TNF-α-induced cell death via NF-κB activation. We found that FAK indirectly modulates expression of the anti-apoptotic mediator PKB/Akt-NF-κB, known to control inflammatory processes and to determine whether cells enter into apoptosis or survive. Taken together, our data and those of others (5, 11) show that IRS-1 and IRS-2 mediate different downstream pathways and that IRS-2 is unable to fully substitute functionally for concomitant abolition of IRS-1. Future studies investigating this compensatory overexpression of IRS-2 may elucidate its importance in the cell survival process.
M. Bryer-Ash was supported in part by a Merit Review Award from the Veterans′ Administration Research Service and by funds from the Gonda Family Endowment.
Present address for D. Ilic: StemLifeLine, 1300 Industrial Rd., Suite 13, San Carlos, CA 94114.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society