Inositol 1,3,4-Trisphosphate 5/6-Kinase Inhibits Tumor Necrosis Factor-induced Apoptosis*

Tumor necrosis factor receptor 1 (TNF-R1) signaling elicits a wide range of biological responses, including inflammation, proliferation, differentiation, and apoptosis. TNF-R1 activates both caspase-mediated apoptosis and NF-κB transcription of anti-apoptotic factors. We now report a link between the TNF-R1 and inositol phosphate signaling pathways. We observed that overexpression of inositol 1,3,4-trisphosphate 5/6-kinase (5/6-kinase) inhibited apoptosis induced by TNFα. The anti-apoptotic effect by 5/6-kinase is not attributable to NF-κB activation, as no changes were detected in the levels of NF-κB DNA binding, IκBα degradation, or anti-apoptotic factors, such as x-linked inhibitor of apoptosis protein. Decreased expression of 5/6-kinase by RNA interference rendered HeLa cells more susceptible to TNFα-induced apoptosis. Overexpression of 5/6-kinase in human embryonic kidney 293 cells inhibited TNFα-induced activation of caspases-8, -3, and -9, BID, and poly(ADP-ribose) polymerase. However, 5/6-kinase did not protect against Fas-, etoposide-, or cycloheximide-induced apoptosis. Further, 5/6-kinase protected against apoptosis induced by the overexpression of TNF-R1-associated death domain but not Fas-associated death domain. Therefore, we suggest that 5/6-kinase modifies TNFα-induced apoptosis by interfering with the activation of TNF-R1-associated death domain.

TNF␣, a member of a large class of cytokines, elicits a wide spectrum of cellular responses including differentiation, proliferation, inflammation, and cell death. TNF␣ signaling simultaneously triggers apoptotic and anti-apoptotic pathways (for review see Ref. 14). Although caspase activation results in apoptosis, NF-B mediates transcription of anti-apoptotic factors that can block the caspase cascade; integration of these events determines the cellular response to TNF␣ stimulation.
Upon TNF␣ binding to TNF-R1, interaction of the receptor with the TNF receptor-associated death domain (TRADD), a central adaptor protein for TNF-R1, leads to the formation of the death-inducing signaling complex (DISC) (15). TNF␣ stimulates both the mitochondrial-dependent (intrinsic) and -independent (extrinsic) apoptotic pathways. In response to extracellular stimuli, the extrinsic pathway of TNF␣ signaling induces the TRADD-dependent recruitment of Fas-associated death domain (FADD), which recruits and activates procaspase-8 leading to activation of the caspase cascade (16). Caspase-8 also stimulates the intrinsic pathway through the cleavage and activation of BID, resulting in cytochrome c release from mitochondria (17,18).
Binding of TRADD to TNF-R1 is also crucial for anti-apoptotic signaling events. Receptor-interacting protein and TNF receptor-associated factor-2 are recruited to the receptor, and this complex activates the IB-kinase (IKK) signalosome, which phosphorylates IB␣ (for review see Refs. 19 and 20). In resting cells, IB␣ inhibits NF-B by preventing translocation of NF-B subunits into the nucleus. However, upon TNF␣ stimulation, phosphorylation of IB␣ by the IKK signalosome, and subsequent ubiquitination, causes its degradation by the 26 S proteasome. As a result, NF-B translocates to the nucleus and activates the transcription of many anti-apoptotic factors, such as inhibitor of apoptosis protein 1, x-linked inhibitor of apoptosis protein (XIAP), and Bcl family members, which can inhibit the TNF␣-induced apoptotic pathway (21)(22)(23).
We now report a link between the inositol phosphate and TNF␣ signaling pathways. We show that 5/6-kinase inhibits TNF␣-induced apoptosis. In 5/6-kinase-overexpressing HEK 293 cells, activation of PARP, BID, caspases-8, -3, and -9 are inhibited in a time-and dose-dependent manner. Surprisingly, the protective effect of 5/6-kinase is not because of enhanced NF-B activation. In addition, 5/6-kinase does not protect against Fas-, etoposide-, or cycloheximide-induced cell death. Therefore, we conclude that 5/6-kinase inhibits apoptosis by blocking the TNF␣-induced activation of the caspase cascade.
DNA Constructs-An expressed sequence tag clone encoding human FADD (GenBank TM accession number BE794964) (ATCC), was used as template for amplifying full-length FADD open reading frame. The PCR reaction used primers 5Ј-GATCTGAATTCCATGGACCCGTTCCTGGT-GCTGCTG-3Ј and 5Ј-AGTCATGGATCCGGACGCTTCGGAGGTAGAT-GCGTCTGAG-3Ј. The product of the PCR reaction was digested with EcoRI (5-prime) and BamHI (3-prime) and subcloned into p3XFLAG-CMV-14 expression vector (Sigma). Full-length TRADD was obtained by PCR using a template from the human expressed sequence tag clone (GenBank TM accession number BE563501) (ATCC) with primers 5Ј-GATCTGAATTCCATGGCAGCTGGGCAAAA TGGGCACG-3Ј and 5Ј-AGTCATGGATCCGGCCAGGCCGCCATTGGGATCGGTC AGGCC-3Ј. The PCR product was digested with EcoRI and BamHI and inserted into p3XFLAG-CMV-14 expression vector (Sigma) to generate TRADD-FLAG. The sequence of each DNA construct was verified by PCR using ABI PRISM Big Dye Terminators v3.0 Cycle Sequencing Kit and analyzed at the Protein and Nucleic Acid Chemistry Laboratory (Washington University, St. Louis, MO).
Generation of 5/6-Kinase RNA Interference Cell Lines-HeLa cells (1 ϫ 10 6 cells) were transfected with a plasmid containing either pSuper vector (a gift from Dr. Reuven Agami) or pSuper 5/6-kinase using LipofectAMINE2000 (Invitrogen) according to the manufacturer's instructions. Three days post-transfection, HeLa cells were trypsinized and placed in puromycin-containing selection medium (Dulbecco's modified Eagle's medium supplemented with 2 mM glutamine, 10% fetal bovine serum (Invitrogen), and 1 g/ml puromycin). Isolated clones were harvested after 4 weeks and assayed for inositol 1,3,4 trisphosphate 5/6-kinase activity (5) and for protein expression by Western blot analysis using an antibody generated against a peptide encoding exons 1-3 of 5/6-kinase. The antibody was affinity purified on a peptide affinity column and eluted with 0.1 M glycine, pH 2.5, then neutralized with 1 M Tris, pH 8.0.
Apoptosis Assays-Trypan blue staining and APOPercentage Apoptosis Assay (Biocolor, Belfast, Northern Ireland) were used to measure the effect of 5/6-kinase overexpression on apoptosis. For APOPercentage assay, HEK 293 cells (2 ϫ 10 4 cells) stably transfected with expression plasmids encoding either 5/6-kinase or vector alone were plated into 24-well plates. After 48 h of tetracycline induction (0.1 g/ml), cell death was induced with either TNF␣/CHX (1 ng/ml and 0.5 g/ml, respectively) or CHX (0.5 g/ml) alone in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. Apoptotic cells were analyzed using APOPercentage Apoptosis Assay according to the manufacturer's instructions. For trypan blue staining, HEK 293 cells overexpressing 5/6-kinase or vector control cells were plated in triplicate in 48-well plates, and protein expression was induced with tetracycline for 48 h. After treatment with either TNF␣/CHX (1 ng/ml and 0.5 g/ml, respectively) or CHX (0.5 g/ml) alone, cells were trypsinized and counted with a hemocytometer. An average of three independent counts for each sample point was performed.
Apoptosis Induction and Detection-HEK 293 cells overexpressing either 5/6-kinase or vector control cells were plated in 12-well plates (5 ϫ 10 5 cells/well) and induced with tetracycline (0.1 g/ml) for 48 h. Apoptosis was induced by the addition of TNF␣/CHX (1 ng/ml and 0.5 g/ml, respectively), CHX (0.5 g/ml), anti-Fas antibody/protein G (10 g/ml and 2.5 mg/ml, respectively), or etoposide (25 M) in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. After incubation for various times, cells were harvested and washed once in phosphate-buffered saline, pH 7.6, and lysed in buffer (20 mM HEPES, pH 7.6, 140 mM NaCl, 1 mM dithiothreitol, 0.1% Nonidet P-40, 10% (v/v) glycerol, 1 M microcystin (Sigma), and Complete TM protease inhibitor tablets (Roche Applied Science)). Total lysates were prepared by one freeze-thaw cycle on dry ice followed by brief sonication (3 ϫ 10 s). The insoluble fraction was removed by centrifugation at 15,800 ϫ g, and protein in the supernatant was measured with a Bio-Rad assay. Equal amounts of protein were then separated by SDS-PAGE followed by Western blot analysis. Data from multiple experiments were compared by densitometry of Western blots using an Eastman Kodak Co. Image Station 440 CF, and the data were analyzed using Kodak 1D V.3.5.4 (Scientific Imaging System). The averages Ϯ S.D. are compared for statistical significance using a student t test.
Northern Blot Analysis-For Northern analysis, total RNA extracts were prepared from HEK 293 cells (1 ϫ 10 8 ) using RNeasy Mini kits (Qiagen) according to the manufacturer's instructions. Agarose gel (1% w/v) was prepared with 0.6 M formaldehyde in 1ϫ MOPS buffer (40 mM MOPS, 10 mM sodium acetate, 1 mM EDTA, pH 7.2). Ten g of total RNA was denatured in loading buffer (18.5% formaldehyde, 50% formamide, 4% Ficoll400, and bromphenol blue in 1ϫ MOPS buffer) at 55°C for 15 min. Samples were separated by agarose gel electrophoresis at 50 mA for 3.5 h and transferred to GeneScreen Plus TM hybridization transfer membrane (Nylon; DuPont) overnight at room temperature. Upon completion of the transfer, membranes were cross-linked using Stratalinker (auto cross-link setting, 254 nm; Stratagene). Hybridization was performed using ExpressHyb TM hybridization solution (Clontech) according to the manufacturer's instructions.

5/6-Kinase Overexpression Protects against TNF␣-induced
Apoptosis-Previously, we reported that 5/6-kinase phosphorylated IB␣ in vitro (10). IB␣ is a key regulator of the NF-B response to TNF␣ stimulation (for review see Ref. 19); therefore we examined the effect of 5/6-kinase overexpression on TNF␣ signaling. TNF␣ can induce activation of NF-B and transcription of target genes that feed back to protect against TNF␣-induced cell death. However, if the NF-B pathway is blocked, cells become susceptible to TNF␣-induced apoptosis. Overexpression of 5/6-kinase protected cells against TNF␣induced cell death, as shown by APOPercentage staining (Fig.  1A). Treatment with TNF␣ alone (without CHX) did not increase cell death in either 5/6-kinase-overexpressing or vector control cells (data not shown). The anti-apoptotic effect of 5/6kinase was further analyzed using trypan blue staining of cells treated with either TNF␣/CHX or CHX alone. Although 80% of vector-transfected cells died within 20 h of TNF␣/CHX stimulation, cells expressing 5/6-kinase exhibited 50% cell death. Treatment by CHX alone caused only 20% cell death in either vector or 5/6-kinase-expressing cells (Fig. 1B). Similar results were observed in HeLa cells transiently transfected with an expression plasmid containing 5/6-kinase (data not shown). These results indicate that overexpression of 5/6-kinase inhibits TNF␣-induced apoptosis.
5/6-kinase Does Not Enhance NF-B Activation as a Mechanism for Protection-Signaling triggered by TNF-R1 can also cause activation of NF-B signaling that leads to the expression of anti-apoptotic genes. To determine whether 5/6-kinase mediates its anti-apoptotic effect(s) by inhibiting the TNF␣induced apoptotic pathway and/or by promoting the TNF␣induced anti-apoptotic pathway, the stability of IB␣ was measured after TNF␣ stimulation. Both vector control cells and 5/6-kinase-expressing cells showed IB␣ degradation within 10 min ( Fig. 2A). Although IB␣ stability in response to TNF␣ signaling is not increased with the overexpression of 5/6-kinase, this does not directly show whether NF-B activity is modulated in response to 5/6-kinase and TNF␣. To determine whether 5/6-kinase directly affects NF-B activity, the DNA binding activity of NF-B was measured by EMSA as described under "Materials and Methods." Treatment with TNF␣ for 30 min resulted in similar DNA binding activity of NF-B in both vector control cells and 5/6-kinase-expressing cells (Fig. 2B). Furthermore, no significant differences were detected upon supershifting with an anti-p50 (NF-B) antibody in 5/6-kinase or vector-expressing cells. Therefore, the ability of NF-B to bind DNA does not seem to be increased by 5/6-kinase in response to TNF␣.
To confirm that TNF␣-induced NF-B activity is not affected by 5/6-kinase, the expression of NF-B anti-apoptotic target genes were measured in mRNA extracts from 5/6-kinase-or vector-transfected cells treated with TNF␣ for 6 h. As shown in Fig. 2C, in the absence of TNF␣, mRNA levels of IB␣ and NF-B2 are low, and RelA is undetectable. After TNF␣ treatment, IB␣, RelA, and NF-B2 mRNA levels increase significantly and to similar levels in both vector control cells and 5/6-kinase-overexpressing cells. The mRNA levels of ␤-actin and 5/6-kinase are shown as controls. Activation of NF-B increases the synthesis of many anti-apoptotic proteins, such as inhibitor of apoptosis protein 1 and XIAP (21). However, the levels of XIAP are unchanged in cells expressing 5/6-kinase treated with TNF␣/CHX (Fig. 2D). The mRNA levels of other NF-B targets genes including TNF receptor-associated factor-2 and TNF receptor-associated factor-6 and Bcl family members, such as Bax, Bcl-2, and Bim were also not changed by the overexpression of 5/6-kinase (data not shown). Consistent with the lack of NF-B up-regulation in response to 5/6kinase overexpression, we have observed that 5/6-kinase continues to exert a protective effect against TNF␣-induced apoptosis in the presence of inhibitors of NF-B pathway, such as MG132, BAF (28), and BAY 11-7082 (data not shown). Therefore, these results argue that 5/6-kinase protects against TNF␣-induced apoptosis through inhibition of the protease pathway and not through enhanced NF-B activation and expression of anti-apoptotic genes. RNA interference plasmid for 5/6-kinase (5/6-kinase RNAi) or a control vector (pSuper) were treated with TNF␣/CHX. Two stable cell lines (KD1 and KD2) were isolated that exhibit marked decreases in 5/6-kinase expression as shown by Western blot analysis (Fig. 3A). The activity of 5/6-kinase was reduced by 90 -95% in these cell lines (Fig. 3B). To determine the effect of decreased 5/6-kinase expression on TNF␣-induced apoptosis, 5/6-kinase RNAi cells (KD1) and control cells were treated with TNF␣/CHX. As shown in Fig. 3C, the control cells exhibited proteolysis of apoptotic substrate PARP within 4 h of TNF␣ treatment but not of BID. In comparison, 5/6-kinase RNAi cells demonstrated decreased levels of full-length BID and increased levels of cleaved PARP in response to TNF␣. Similar results were obtained using the KD2 cell line. We also showed that RNAi of 5/6-kinase increased apoptosis as measured by APOPrecentage stain and trypan blue exclusion as shown (Fig. 3, D and E). These results support the hypothesis that 5/6-kinase affects TNF␣-induced cell death as the 5/6kinase RNAi cells are more susceptible to TNF␣-induced apoptosis.

5/6-Kinase Does Not Protect against Apoptosis Induced by Etoposide or Cycloheximide-
To determine whether the antiapoptotic effect exerted by 5/6-kinase is a specific response to TNF-R1-induced signaling, other death-inducing agents including CHX and etoposide were examined. Overexpression of 5/6-kinase was shown previously not protective against CHX-induced apoptosis (Fig. 1B). Another apoptotic agent, etoposide, induces DNA damage leading to the activation of p53 and the mitochondrial pathway of apoptosis (24). To determine whether 5/6-kinase overexpression could inhibit the mitochondrial pathway, cells expressing either 5/6-kinase or vector control cells were treated with etoposide for 20 h, and caspase levels were determined by Western blot analysis. As shown in Fig. 4A, etoposide-induced cleavage of caspase-9 was not inhibited by 5/6-kinase overexpression. These results suggest that 5/6-kinase overexpression does not inhibit the intrinsic apoptotic pathways.
Fas, another member of the TNF-R superfamily, shares a similar mechanism of caspase activation. Both TNF-R1 and Fas-mediated signaling involve FADD recruitment to the DISC for the activation of caspase-8 (25,26). To determine whether 5/6-kinase protected against Fas-induced apoptosis, vector control cells or 5/6-kinase-overexpressing cells were treated with Fas for 20 h. As shown in Fig. 4B, BID and PARP were cleaved in both 5/6-kinase and vector-transfected cells, suggesting that 5/6-kinase does not inhibit Fas-induced cell death. The relative cleavage of PARP and the remaining full-length BID were determined by densitometry and are shown below the respective bands. Therefore, these experiments show that 5/6-kinase does not protect against Fas-induced apoptosis and suggest that the mechanism(s) of 5/6-kinase action is upstream of FADD recruitment.

FIG. 2. Effects of 5/6-kinase overexpression on NF-B activation.
A, HEK 293 cells stably transfected with expression plasmids of vector or 5/6-kinase were treated with TNF␣ (1 ng/ml) for the indicated times. Western blot analysis was performed using anti-IB␣ antibody. B, HEK 293 cells overexpressing 5/6-kinase or vector control cells were stimulated with TNF␣ (1 ng/ml) for 30 min. Nuclear extracts were prepared, and EMSA was performed as described under "Materials and Methods." C, HEK 293 cells overexpressing 5/6-kinase or vector control cells were treated with TNF␣ for 6 h, and total mRNA extracts were prepared. Northern blot analysis was performed using 32 Plabeled probes for IB␣, NF-B2, RelA, 5/6-kinase, and actin. D, HEK 293 cells overexpressing 5/6-kinase or vector control cells were stimulated with TNF␣/ CHX (1 ng/ml and 0.5 g/ml, respectively) for the indicated times. Total lysate (20 g) was applied to SDS-PAGE, and Western blot analysis was performed using antibodies against XIAP. The results are representative of three independent experiments.

5/6-Kinase Inhibits TNF␣-induced PARP, BID, Caspase-3, and Caspase-8 Cleavage-
To characterize the mechanism(s) by which 5/6-kinase mediates protective effect(s) against TNF␣induced apoptosis, we sought to determine whether 5/6-kinase directly affects caspase activation. One substrate of caspases is PARP, a DNA repair enzyme that is cleaved by caspase-3 (27). Cells overexpressing 5/6-kinase or vector control cells were treated with TNF␣/CHX for 6 h, and full-length and cleaved PARP levels were measured. In vector control cells treated with CHX, PARP degradation is observed with increasing doses of TNF␣. In contrast, cells expressing 5/6-kinase exhibited lower levels of degraded PARP in both a time-and dose-dependent manner (Fig. 5, A and B). The antibody against PARP preferentially recognizes the cleaved form of PARP, which explains the increase in the intensity of cleaved PARP as compared with that of full-length PARP in Fig. 5A. At this early time point (6-hr), the protein levels of major signaling intermediates in this pathway, including the p50 subunit of NF-B (Fig. 5A)
Caspase-8 is a critical initiator of the apoptotic pathways triggered by TNF-R1 and is responsible for processing caspase-3 (29). To determine whether caspase-8 activation is affected by 5/6-kinase, caspase-8 levels were measured in TNF␣-treated vector control and 5/6-kinase cells. As shown in Fig. 5D, vector control cells exhibit caspase-8 activation within 4 h, with the majority of caspase-8 cleaved by 8 h. However, 5/6-kinase-expressing cells show retarded activation of caspase-8, with only low levels of cleaved caspase-8 appearing within 8 h (20% of that seen in vector cells as shown in the graph.) Activation of caspase-8 induces BID truncation, which activates the mitochondrial pathway of apoptosis (30). Therefore, inhibition of BID cleavage is expected in response to 5/6-kinase overexpression. Consistently, 5/6-kinase-expressing but not vector control cells maintained full-length BID levels (4-fold greater than in vector cells as shown in the graph) (Fig.  5E). This result concurs with the observation that the levels of BID in 5/6-kinase RNAi cells were lower than that of control cells (Fig. 3). Consistent with the role of truncated BID in intrinsic apoptotic pathways, caspase-9 activation was reduced in 5/6-kinase-expressing cells (data not shown). These results demonstrate that the protective effect of 5/6-kinase on TNF␣induced apoptosis is because of inhibition of the TNF-R1-induced apoptosis upstream of caspase-8 activation.

5/6-Kinase Protects against Cells Overexpressing TRADD but
Not FADD-Caspase-8 can be triggered by the interaction of TNF-R1 and Fas with their respective ligands. In the case of TNF-R1, caspase-8 activation requires the recruitment of the adaptor proteins TRADD and FADD. To delineate further the molecular events mediating the anti-apoptotic effects of 5/6kinase, plasmids encoding TRADD or FADD were transiently transfected into either vector control cells or 5/6-kinase-overexpressing cells. In response to TNF␣, FADD is recruited to the DISC by its interaction with TRADD (14). To block apoptosis during the transfection of TRADD and FADD, cells were incubated with an irreversible caspase inhibitor, Z-Glu-Val-Aspfluoromethyl ketone (20 M) prior to TNF␣/CHX stimulation. In mock transfected cells, caspase 8 cleavage was blocked in 5/6-kinase-expressing cells (Fig. 6A). Cells transfected with TRADD showed increased levels of cleaved caspase-8 as compared with mock-transfected cells, but 5/6-kinase cells continued to exhibit less caspase-8 cleavage than the control cells as shown in the graph (Fig. 6B). Therefore, 5/6-kinase mediates its anti-apoptotic effect downstream of TRADD-induced FADD activation. If 5/6-kinase inhibits FADD-induced activation of caspase-8, overexpression of FADD would result in an inhibition of 5/6-kinase's protective effect and a higher level of activated caspase-8. Cells expressing FADD exhibited more cleaved caspase-8 than mock-transfected cells, indicating that FADD overexpression is sufficient to activate caspase-8 (Fig. 6, compare A with C). FADD overexpression nearly eliminated the protective effect of 5/6-kinase, as demonstrated by the elevated levels of cleaved caspase-8 in both 5/6-kinase and vector control cells, and there was no significant protection by 5/6-kinase expression as shown in the graph (p value Ͼ 0.1) (Fig. 6C). Therefore, 5/6-kinase may mediates its protective effect by interfering with the activation of TRADD. DISCUSSION We have shown that 5/6-kinase inhibits TNF␣-induced cell death by specifically blocking apoptotic-signaling pathways. This phenotype is specific for TNF␣, as 5/6-kinase did not protect cells against death induced by cycloheximide, etoposide, that 5/6-kinase may exert its inhibition of the caspase cascade through TRADD.
We showed previously (10) that 5/6-kinase could phosphorylate IB␣. It was hypothesized that 5/6-kinase may inhibit apoptosis through the regulation of NF-B. This phosphorylation event was predicted to decrease the stability of IB␣ and therefore enhance the activation of NF-B, which then feeds back to block the apoptotic pathway. However, our results from IB␣ and NF-B functional assays show that 5/6-kinase does not significantly alter the NF-B pathway. NF-B nuclear translocation, transcriptional activity, and levels of anti-apoptotic proteins regulated by NF-B did not change significantly. However, NF-B targets many different genes, and we cannot definitively exclude the possibility of 5/6-kinase affecting another NF-B-regulated gene as a further mechanism for apoptosis inhibition.
The formation of DISC is an early and important stage of both Fas and TNF-R1 signaling (15). There are several antiapoptotic factors that can inhibit both death-receptor pathways. Inhibitors of apoptosis, such as inhibitor of apoptosis protein 1, are recruited to the DISC and block caspase activation by interfering with the prodomain or the catalytic domain of caspases. (see Ref. 31, and for review see Ref. 32). In addition, FLIP (FLICE/caspase-8 inhibitor protein), a caspase-8like protein that lacks the death effector domain, functions as a dominant negative inhibitor of caspase-8 and affects both Fas and TNF␣ signaling (33,34). In contrast to 5/6-kinase, these known inhibitors of caspase-8 activation affect both Fas-and TNF␣-mediated cell death and therefore differ from the TNF␣specific protection by 5/6-kinase. Another possible target of 5/6-kinase anti-apoptotic activity is silencer of death domain (SODD) (35). SODD specifically interacts with TNF-R1 and competes with TRADD for binding to the receptor. Overexpression of SODD inhibits TNF␣ signaling, resulting in decreased NF-B activation and apoptosis. However, 5/6-kinase specifically affects the apoptotic arm of TNF signaling. Therefore an effect on SODD is not likely a mechanism by which 5/6-kinase inhibits apoptosis.
We propose that 5/6-kinase can interfere the interaction of FADD and TRADD. Because only the apoptotic pathway is affected, inhibition of TNF-R1 and TRADD activation are not likely mechanisms for the 5/6-kinase protective effect; both are required for the NF-B arm that is unaffected. In addition, we have shown that 5/6-kinase-mediated protection can be abrogated by FADD overexpression, thus FADD must act downstream of 5/6-kinase. The most likely mechanism is that 5/6-kinase inhibits the recruitment of FADD to TRADD. It is possible that 5/6-kinase phosphorylates a specific domain required for apoptosis but not for NF-B activation. Furthermore, phosphorylation of FADD could inhibit its recruitment to TRADD but play no role for clustering upon overexpression.
There are two possible mechanisms for which 5/6-kinase could exert its protective effect(s). 5/6-Kinase may exhibit protein kinase activity toward subunits within the DISC. Phosphorylation has only been recently reported to affect DISC function. For instance, FADD is phosphorylated on Ser-194 in a cell cycle-dependent manner (36). Phosphorylation of FADD can enhance apoptosis induced by etoposide (37). Also, PKC II may regulate cellular FLIP expression and phosphorylation in response to Fas (38). It is conceivable that 5/6-kinase could selectively phosphorylate a component within DISC and alter its function specific to TNF␣ signaling. Another possibility is an inositol phosphate product of 5/6-kinase could affect binding and function of DISC. Based on in vitro studies by Wilson and Majerus (5), 5/6-kinase is responsible for generating inositol 1,3,4,5-tetrakisphosphate and inositol 1,3,4,6-tetrakisphosphate, precursors for inositol 1,3,4,5,6-pentakisphosphate and inositol hexakisphosphate. Elevation of cellular inositol phosphates or depletion of substrates, such as inositol 1,3,4trisphosphate, could affect the formation of DISC and would represent a novel link between inositol phosphate and cell death signaling pathways. Future studies are needed to dissect the protein and inositol phosphate kinase activities of 5/6-kinase and determine the activity required for TNF␣protection.