RIP1 kinase-dependent Thr308 phosphorylation during necroptosis in multiple cell types. Furthermore, we found that downstream Akt signaling through mTORC1 and S6 contributes to the activation of necroptosis and TNFα production. We found that the Akt pathway serves as a critical link between RIP1 kinase and JNK activation in L929 cells. Further data suggested that in multiple other cell types including FADD deficient Jurkat cells, RAW and J774.1 macrophage cell lines, and mouse lung fibroblasts Akt provides a key link to TNFα production, but is dispensible for cell death per se. Overall, our results reveal a specific and novel role for the Akt pathway in regulated necrosis and necrosis-associated inflammatory signaling.

Results

Basic Fibroblast Growth Factor Promotes Necroptosis in L929 Cells
It has been established that mouse fibrosarcoma L929 cells undergo necroptotic cell death following stimulation with TNFα [10], [17]. In addition, inhibition of caspase-8 activity alone, either through siRNA knockdown or by using the pan-caspase inhibitor, zVAD.fmk, is sufficient to trigger necroptosis in these cells [10], [14]. Interestingly, while necroptosis was initially identified as a back-up form of cell death triggered by pro-apoptotic stimuli in the presence of apoptosis inhibitors [17], recent analysis of physiological cell death during mouse development has suggested that the loss of apoptotic regulators, such as caspase-8 and FADD [18], [19], [20], leads to robust induction of necroptosis and death of E10.5 embryos even though apoptosis is not normally induced in wild type embryos. These data are reminiscent of the observations in L929 cells where the loss of caspase activity in healthy cells is sufficient to trigger necroptosis and prompted us to explore the extrinsic or intrinsic cellular factors that promote necroptosis once caspase-8 activity, which cleaves and inactivates RIP1 kinase and the RIP1 deubiquitinase CYLD [21], [22], is removed in L929 cells. Consistent with a previous report [16], we found that serum starvation of L929 cells prevented necroptosis in response to zVAD.fmk (Fig. 1A). The addition of growth factors, such as bFGF, restored zVAD.fmk induced death under serum free conditions (Fig. 1B). Interestingly, this does not reflect a generic requirement for growth factor signaling, as only some growth factors (bFGF and IGF-1, but not EGF and PDGF) promoted death (Fig. 1B). Furthermore, growth factor-dependent necroptosis required the inhibition of caspase activity, as bFGF alone did not induce cell death (Fig. 1C). In contrast, TNFα triggered necroptosis equally efficiently in the absence of serum (Fig. 1A), suggesting that either growth factors and zVAD.fmk or TNFα are required for necroptotic death in L929 cells.
10.1371/journal.pone.0056576.g001 Figure 1  bFGF and IGF-1 promote necroptosis in concert with zVAD.fmk.
(A) L929 cells were treated with TNFα or zVAD.fmk under normal serum (10% FBS) or serum free conditions. Cell viability was determined after 24 hr using the CellTiter-Glo Viability assay. The concentrations of all necroptosis-inducing agents are listed in the  Materials and Methods  section or indicated in the figures. (B) Cells were treated with zVAD.fmk, the indicated growth factors, and Nec-1 under serum free conditions for 24 hrs followed by measurement of cell viability. (C) Cells under serum free conditions were treated with FGF, zVAD.fmk, or both for 24 hrs followed by viability assay. (D) Cell death was induced by zVAD.fmk or TNFα under full serum condition in the presence of 2 µM PD173074 and 20 µM PD166866. In all graphs, average±SD was plotted. Previously we described the development of 7-Cl-O-Nec-1 (Nec-1) as a potent and selective inhibitor of RIP1 kinase and necroptosis (Fig. S1A) [23], [24]. Recently, its selectivity has been further validated against a panel of more than 400 human kinases [15]. This inhibitor efficiently blocked growth factor/zVAD.fmk-induced necroptosis under serum free conditions in L929 cells and both zVAD.fmk and TNFα-induced necroptosis under full serum conditions (Fig. 1B, S1B). To further validate the role of RIP1, we used an inactive analog, 7-Cl-O-Nec-1i (Nec-1i) (Fig. S1A), which contains an extra N-methyl group that leads to almost complete loss of RIP1 kinase inhibitory activity in vitro [23]. Nec-1i was unable to protect L929 cell death under serum condtions treated with zVAD.fmk or TNFα (Fig. S1B) or serum free conditions treated with bFGF/zVAD.fmk (Fig. S1C). This confirms that RIP1 kinase is responsible for necroptosis in L929 cells under both serum and serum free conditions.
We next examined whether bFGF contributes to zVAD.fmk-induced necroptosis under normal serum conditions (10% FBS). We used two bFGF receptor tyrosine kinase inhibitors (PD173074 and PD166866), and determined that inhibition of bFGF signaling strongly inhibited zVAD.fmk-induced necroptosis under normal serum conditions (Fig. 1D). In contrast, neither bFGF receptor inhibitor was able to attenuate TNFα-induced necroptosis (Fig. 1D), consistent with growth factors being dispensable for this pathway (Fig. 1A). Overall, these data suggest that the induction of necroptosis by zVAD.fmk is promoted by bFGF under both serum and serum free conditions. The induction of necroptosis, however, is not a simple consequence of growth factor signaling since not all growth factors allowed death to occur. Instead, specific signaling events mediated by particular growth factors appear to contribute to necroptotic death.

RIP1 Kinase-dependent Activation of Akt Contributes to Necroptosis
Given our observation that growth factors are important for zVAD.fmk induced death, we examined the contribution of several pathways, including MAPK pathways and Akt, which are known to be activated following growth factor receptor activation (Fig. 2A). Inhibition of Akt (Akt inhibitor VIII) strongly protected the cells from growth factor-sensitive necroptosis induced by zVAD.fmk [16] as well as cell death triggered by bFGF or IGF-1/zVAD.fmk under serum free conditions (Fig. 2B). Inhibition of Akt also protected the cells from growth-factor insensitive death by caused by TNFα (Fig. 2A). Consistent with previous reports, the JNK inhibitor SP600125 protected the cells from both zVAD.fmk and TNFα induced death (Fig. 2A,B and Fig. S2A) [12], [14]. In contrast, inhibition of two other MAPKs, p38 and ERK, previously reported not to be activated during necroptosis [14], did not protect from either zVAD.fmk or TNFα induced death (Fig. 2A).
10.1371/journal.pone.0056576.g002 Figure 2  Akt contributes to necroptosis induced by zVAD.fmk and TNFα.
(A,B) Necroptosis was induced by zVAD.fmk or TNFα (full serum, A) or growth factors/zVAD.fmk (serum free, B) in the presence of inhibitors of Akt (Akt inhibitor VIII), JNK (SP600125), p38 (PD169316), and Erk (UO126). Cell viability was determined after 24 hrs. (C) L929 cells transfected with Akt1, Akt2, and Akt3 siRNAs for 72 hrs were treated with zVAD.fmk or TNFα for 9 hrs. Cell viability and Akt expression levels were determined after 24 hrs. In all graphs, average±SD was plotted. Next, we used two approaches to further validate the role of Akt in necroptotic cell death. First, two additional Akt inhibitors, a highly specific, allosteric kinase inhibitor MK-2206 [25] and triciribine (TCN) [26], which blocks membrane translocation of Akt, both attenuated cell death (Fig. S2B). Secondly, simultaneous knockdown of Akt isoforms Akt1 and Akt2 using siRNAs protected cells from necroptosis induced by both zVAD.fmk and TNFα (Fig. 2C). No expression of Akt3 was seen in L929 cells (Fig. S2C) and, consistently, Akt3 siRNA had no additional effect on necroptosis. Our results confirmed that Akt plays a key role in necroptosis induced by multiple stimuli in L929 cells.
To understand the activation of Akt and JNK under necroptotic conditions, we examined the changes in Akt and JNK phosphorylation at 9 hrs post zVAD.fmk and TNFα stimulation. This time point was chosen because it reflects the early stage of cell death in our system (Fig. S3A, B). Following stimulation with either zVAD.fmk or TNFα we observed a robust increase in Akt phosphorylation at a known major activation site, Thr308 (Fig. 3A). Interestingly, we did not observe concomitant phosphorylation changes in the second major activation site of Akt, Ser473. We also observed an increase in the phosphorylation of both the p46 and p54 isoforms of JNK and its major substrate c-Jun (Fig. 3A). These data indicate that both Akt and JNK are activated under necroptotic conditions.
10.1371/journal.pone.0056576.g003 Figure 3  RIP1 kinase-dependent phosphorylation of Akt and JNK during necroptosis.
(A) L929 cells were treated with zVAD.fmk or TNFα for 9 hr, followed by western blotting with indicated antibodies. (B,C) L929 cells were treated with zVAD.fmk (B) or bFGF/zVAD.fmk (serum free conditions, C) and samples were collected at the indicated time points for western blot. (D) Nec-1 was added to the cells stimulated with bFGF or bFGF/zVAD (serum free conditions) for 15 min or 9 hr followed by western blot with the indicated antibodies. The RIP1 kinase inhibitor, Nec-1, completely prevented the increase in Thr308 Akt phosphorylation, while Nec-1i did not (Fig. 3A, Fig. S1D). Similarly, Nec-1 prevented the induction of JNK phosphorylation in response to zVAD.fmk and substantially reduced this change after TNFα addition. We observed some changes in total protein levels of JNK and c-Jun following necroptotic stimulation. Some of these changes, e.g. zVAD.fmk-induced increase in c-Jun, were also attenuated by Nec-1. Importantly, Nec-1 did not alter the basal phosphorylation levels of either Akt or JNK (Fig. 3A). This established that Akt Thr308 and JNK phosphorylation during necroptosis is RIP1 dependent.
Interestingly, we discovered that the phosphorylation of Akt Thr308, JNK and Jun are late events following zVAD.fmk stimulation (Fig. 3B) that coincide with the onset of necroptosis at 6 hr post-stimulation (Fig. S3A). To better understand the contributions of growth factors and RIP1 kinase to necroptotic regulation of Akt, we next analyzed the time course of these phosphorylation changes under serum free conditions. We found that the addition of bFGF alone or in combination with zVAD.fmk led to a substantial rapid and transient increase in both Thr308 and Ser473 phosphorylation of Akt as well as JNK and c-Jun at 15 minutes, reflecting the expected response to growth factor stimulation (Fig. 3C). Significantly, the combination of bFGF/zVAD.fmk, but not bFGF alone, also caused a robust, second, delayed increase in the phosphorylation of Thr308, but not Ser473, of Akt as well as a delayed increase in the phosphorylation of JNK and Jun. Furthermore, Nec-1 had no significant effect on the early increase in both Akt and JNK/c-Jun phosphorylation triggered by both bFGF and bFGF/zVAD, while Nec-1, but not its inactive analog Nec-1i (Fig. S1E), efficiently blocked the bFGF/zVAD increase at 6–9 hr (Fig. 3D), suggesting that only the delayed activation of Akt and JNK is specific for necroptosis and dependent on RIP1 kinase activity. Similarly, IGF/zVAD, which also promoted cell death under serum free conditions, produced a delayed increase in Thr308 phosphorylation on Akt, while IGF alone caused solely an early, transient increase in phosphorylation (Fig. S3C). We confirmed the kinetics of the Akt Thr308 and Ser473 phosphorylation changes using a quantitative ELISA assay, which also showed a robust delayed necroptosis-specific RIP1-dependent increase in Akt Thr308 phosphorylation (Fig. S3D, E). Taken together, these results indicate that the observed delayed increases in Akt and JNK phosphorylation, preceding the onset of cell death, represent specific consequences of necroptotic signaling downstream from RIP1 kinase.

TNFα Induces Delayed Akt Thr308 Phosphorylation and Necroptosis Independent of Growth Factor Stimulation
Consistent with TNFα inducing necroptosis independently of growth factors (Fig. 1A), FGFR inhibitors did not attenuate TNFα-induced changes in Akt or JNK phosphorylation, while efficiently preventing these changes in response to zVAD.fmk (Fig. S4A). Furthermore, addition of TNFα led to comparable late activation of Akt p308 signal under both normal and serum free conditions (Fig. S4B, C), indicating that TNFα signaling to Akt Thr308 is growth factor-independent. In contrast, activation of JNK by TNFα followed different kinetics from zVAD.fmk-induced changes. TNFα treatment caused an early and robust increase in the phosphorylation of JNK and c-Jun. Nec-1 did not affect this early increase, however, it reduced levels of pJNK/Jun at the late, 9 hr time point (Fig. S4B, C). This again separated early RIP1-independent changes, which likely reflect the ability of additional upstream kinases, such as Ask1 to activate JNK [27], from the late RIP1 kinase-dependent necroptotic signaling.

Late Increase in Akt Thr308 Phosphorylation Contributes to the Induction of Necroptotic Cell Death
We next investigated if the delayed RIP1 kinase-dependent increase in Akt Thr308 phosphorylation functionally contributes to the execution of necroptotic cell death. Firstly, PDGF/zVAD.fmk, which cannot induce necroptosis (Fig. 2A), triggered only the initial, rapid Akt and JNK phosphorylation changes and not the delayed activation (Fig. 4A), indicating that late, rather than early Akt phosphorylation correlates with necroptosis. Secondly, we saw that the ability of the Akt inhibitor to protect cells from necroptosis rapidly declined after 6 hrs of stimulation with zVAD.fmk, TNFα or bFGF/zVAD.fmk and no protection was observed when the inhibitor was added at 9 hrs (Fig. 4B,C). This time frame coincides with the timing of the secondary Akt Thr308 phosphorylation. Finally, we terminated the bFGF signal one hour after addition of bFGF by the addition of PD173074. This allowed us to retain early Akt activation, but to suppress the secondary increase (Fig. 4D). Both pre-addition and delayed addition of PD173074 fully prevented necroptosis (Fig. 4E). Overall, these data, while correlative, indicate that early Akt activation is insufficient to promote necroptosis and are strongly supportive of an important role for the delayed activation of Akt in the induction of necroptotic cell death.
10.1371/journal.pone.0056576.g004 Figure 4  Late Thr308 phosphorylation of Akt contributes to necroptosis.
(A) L929 cells were treated with zVAD.fmk and bFGF or PDGF, with or without Nec-1, for the indicated periods of time. (B,C) L929 cells were stimulated by zVAD.fmk or TNFα (B) or bFGF/zVAD.fmk under serum free conditions (C). Akt inh. VIII was added 15 min before necroptotic stimulation (Pre) or at indicated times after stimulation. Viability was measured 24 hr after activation of necroptosis. (D) L929 cells were stimulated with bFGF/zVAD under serum free conditions. PD173074 was added 15 min before or 1 hr after FGF/zVAD. Samples for western blot were collected at 15 min and 9 hr time points. (E) Cells were pretreated with PD173074 or it was added 1 hr after bFGF/zVAD.fmk, followed by viability assessment at 24 hr. In all graphs, average±SD was plotted.

The Akt Signaling Pathway Contributes to the Regulation of Necroptosis
We next determined whether the necroptosis-associated increase in Thr308 phosphorylation results in an increase in Akt kinase activity. Under necroptotic conditions, we observed an increase in the phosphorylation of multiple known Akt substrates (Forkhead box class O (FoxO) proteins, GSK-3 kinases and mouse double minute 2 (MDM2)) as well as downstream molecules (p70 ribosomal protein S6 Kinase (p70S6K), S6) (Fig. 5A). In some cases (FoxO1, FoxO4, MDM2), a robust increase was observed. In other cases (FoxO3a, GSK-3α/β, p70S6K and its substrate S6), the changes were less pronounced (Fig. 5A). The timing of the phosphorylation changes paralleled the increase in Akt phosphorylation (Fig. 5B, S5A, B). In the case of pFoxO1 we occasionally observed a shift in migration rather than an increase in band intensity (e.g. comparing Fig. 5A and B), suggesting that phosphorylation events in addition to Thr24 take place during necroptosis. Notably, in all cases the necroptosis-associated increases in Akt substrates were abrogated by Nec-1 (Fig. 5A, Fig. S5A, B). Overall, these data suggested that a significant part of the “canonical” Akt signaling network is activated at the onset of necroptotic cell death in a RIP1 dependent fashion.
10.1371/journal.pone.0056576.g005 Figure 5  mTORC1 contributes to the regulation of necroptosis.
(A) L929 cells were treated with zVAD.fmk or TNFα for 9 hr and harvested for western blot. (B) Cell under serum free condition were treated with bFGF or bFGF/zVAD.fmk for the indicated amounts of time, followed by western blotting using the indicated antibodies. (C) Necroptosis was induced by zVAD.fmk or TNFα in L929 cell in the presence of inhibitors of Akt(Akt inh. VIII) and mTOR (rapamycin, Torin-1 and PI-103). (D) L929 cells with mTOR siRNA knockdown were harvested for western blot or treated with zVAD.fmk or TNFα for 24 hrs. Cell viability was determined 24 hr after activation of necroptosis. In all graphs, average±SD was plotted. Akt kinase is considered to be a pro-survival protein that inhibits apoptosis through the control of multiple effectors including mTORC1, GSK-3 and others [28]. An important question is whether these same molecules reverse their pro-survival roles during necroptosis. We found that inhibition of mTORC1 by rapamycin, an inhibitor of the mTOR co-factor Raptor [29], protected cells from necroptosis (Fig. 5C). Similarly, the direct mTOR kinase inhibitor Torin1 [30] and the dual PI3K/mTOR inhibitor PI-103 [31] also efficiently inhibited necroptosis (Fig. 5C). Knockdown of mTOR using siRNA further validated the small-molecule inhibitor data indicating a role for mTOR in necroptosis by protecting cells from both zVAD.fmk and TNFα induced death (Fig. 5D).
mTORC1 regulates translation through activation of p70S6 kinase and, subsequently, ribosomal protein S6 [32]. Notably, a genome-wide siRNA screen [10] suggested an important role for protein translation in necroptosis. Consistently, we found that the small molecule inhibitor of p70S6K PF-4708671 [33] attenuated necroptosis at the concentrations required to block S6 phosphorylation (Fig. S5C, D). Partial siRNA knockdown of S6 protein attenuated necroptosis as well (Fig. S5E), suggesting that translational control by p70S6K/S6 may play a role in necroptosis. Overall, while the full repertoire of Akt targets during necroptosis remains to be fully explored, our data provide evidence that the activity of an anti-apoptotic branch of Akt signaling can promote necroptosis.
RIP1 kinase, Akt, mTORC1 and JNK control the upregulation of TNFα accompanying necroptosis. Hitomi et al. [10] have recently reported that the induction of necroptosis by zVAD.fmk in L929 cells is associated with increased synthesis of TNFα, which potentiates cell death. Therefore, we examined whether Akt and its effectors contribute to TNFα synthesis. Consistent with a RIP1-dependent increase in TNFα protein (Fig. S6A, B), we found that TNFα mRNA levels increased during necroptosis in L929 cells in a RIP1 (Fig. S6C. Under serum free conditions, bFGF alone triggered some induction of TNFα mRNA, while its combination with zVAD.fmk (but not zVAD.fmk alone) caused a pronounced further increase (Fig. 6A). Conversely, PDGF caused a modest upregulation of TNFα mRNA, which was not further increased in the presence of zVAD.fmk (Fig. 6A), demonstrating that activation of necroptosis is specifically accompanied by a marked increase in autocrine TNFα synthesis.
10.1371/journal.pone.0056576.g006 Figure 6  Akt and mTORC1 control autocrine TNFα synthesis and JNK activation during necroptosis.
(A) Cells were treated under serum free conditions with bFGF or PDGF with or without zVAD.fmk for 9 hr, followed by qRT-PCR analysis of mTNFα. Data was normalized to mouse 18S RNA. (B) Necroptosis was induced by zVAD.fmk or TNFα in cells treated with Nec-1, rapamycin (rapa), or Akt inh. VIII inh. followed by qRT-PCR analysis of TNFα mRNA levels. (C-F) L929 cells with siRNA knockdown of Akt isoforms (C,E) or mTOR (D,F) were stimulated with zVAD.fmk or TNFα for 9 hr, followed by qRT-PCR analysis of mTNFα (C,D) or western blot (E,F). In all graphs, average±SD was plotted. Further analysis suggested that both Akt and mTORC1 contribute to the upregulation of TNFα mRNA during necroptosis as both small-molecule inhibition and siRNA knockdown of Akt and mTOR reduced TNFα mRNA levels in necroptotic cells (Fig. 6B,C,D). Notably, RIP1 and Akt inhibitors had no effect on the levels of TNFα mRNA in control cells or in the cells stimulated with bFGF alone (Fig. 6A,B, Fig. S6C), suggesting that these kinases specifically mediate necroptosis-dependent increase in TNFα synthesis.

Akt and mTORC1 Control the Activation of JNK during Necroptosis
JNK is a well-established regulator of TNFα synthesis in a variety of systems [13], [14], [15], [34]. Therefore, the ability of Akt and mTORC1 inhibitors to block the increase in TNFα mRNA lead us to examine their role in the activation of JNK during necroptosis. Knockdown of Akt isoforms Akt1 and Akt2 or inhibition of Akt prominently suppressed the necroptosis dependent increase in JNK and c-Jun phosphorylation (Fig. 6E, S6D,E) suggesting that Akt may provide a link between RIP1 and JNK activation. Importantly, inhibition of Akt only inhibited the delayed, but not the early, increase in bFGF/zVAD.fmk induced JNK and c-Jun phosphorylation (Fig. S6F). Knockdown of mTOR, rapamycin and the p70S6K inhibitor PF-4708671 also attenuated the necroptosis-associated increase in JNK and c-Jun phosphorylation (Fig. 6F, S6E,G, Fig. S5D). Overall, these data suggested that the Akt-mTORC1-S6K axis, acting downstream from RIP1 kinase, is required for the increase in JNK activity during necroptosis in L929 cells.

PI3-kinase and PDK1 Mediate the Increase in Akt Thr308 Phosphorylation Under Necroptotic Conditions
Typical regulation of Akt by growth factors involves its recruitment to the plasma membrane, which is mediated by the binding of the pleckstrin homology (PH) domain of Akt to the product of PI3K, phosphatidylinositol-3,4,5-triphosphate (PIP3). In the membrane, Akt is phosphorylated on Thr308 and Ser473 by 3-phosphoinositide dependent protein kinase-1 (PDK1) and mTORC2 (or DNA-PK), respectively [35]. Since our results showed that only Thr308 Akt phosphorylation is increased during necroptosis, we next examined whether it is still dependent on PI3K and PDK1. Inhibition of PI3K and PDK1 using the specific inhibitors LY249002 and BX912 [36] resulted in the efficient inhibition of cell death and Akt Thr308 phosphorylation (Fig. S7A–D). Likewise, siRNA knockdown of PDK1 protected cells from death and inhibited Akt Thr308 phosphorylation (Fig. S7E,F) Therefore, PI3K and PDK1 activity is still required for non-canonical Akt activation during necroptosis.

Expression of Constitutively Active Akt, Rescues Necroptosis Under Serum Free Conditions
We used L929 cells stably expressing constitutively active wild type Akt1 (Myr-Akt) or the catalytically inactive mutant K179M in order to further understand the contribution of growth factors and RIP1 kinase to Akt activation during necroptosis. Constitutively active Akt1 (Myr-Akt) was generated as previously described [37] by the addition of a myristoylation signal which provides constitutive localization to the plasma membrane and by the deletion of the auto-inhibitory PH domain (Fig. 7A) resulting in an Akt that is active under serum free. It is important to note that the cells expressing Myr-Akt were viable, grew in a manner indistinguishable from the empty vector control cells, and were not triggered to induce necroptosis by serum starvation (Fig. 7B). This indicates that active Akt alone is not sufficient to induce necroptotic cell death. Under serum free conditions Myr-Akt, but not the K179M mutant, fully restored zVAD.fmk-induced necroptosis (Fig. 7A,B). Nec-1 prevented both Myr-Akt dependent cell death and the necroptosis-specific delayed increase in Akt Thr308 phosphorylation (Fig. 7B,C). Myr-Akt also allowed other zVAD.fmk-dependent events, including activation of JNK and c-Jun phosphorylation (Fig. 7C) and upregulation of TNFα mRNA (Fig. 7D) to occur under serum free conditions, confirming an important role for Akt at the apex of necroptotic signaling. These data demonstrated that the presence of active and membrane localized Akt is sufficient to uncouple Akt activation during necroptosis from growth factor signaling. RIP1 kinase was still able to regulate Akt activation during necroptosis, suggesting that growth factors and RIP1 kinase provide two independent inputs required for Akt changes during necroptosis.
10.1371/journal.pone.0056576.g007 Figure 7  Over expression of constitutively active Akt restores necroptosis under serum free conditions.
(A,B) L929 cells were stably infected with empty MSCV retrovirus or viruses encoding Myr-Akt or the catalytically inactive Myr-Akt K179M. Necroptosis was induced by the addition of zVAD.fmk under serum free conditions (A) or serum or serum free conditions with Nec-1 (B). Viability assays were performed after 24 hr. (C) Myr-Akt and Myr-Akt K179M cells were treated with zVAD.fmk and/or Nec-1 under serum free conditions for 9 hr, followed by western blot using the indicated antibodies. Endogenous Akt (∼) and Myr-Akt (*) bands are indicated. (D) L929 cells, stably infected with Myr-Akt and Myr-Akt K179KM, were stimulated with zVAD.fmk for 9 hr under serum free conditions. TNFα mRNA levels were determined by qRT-PCR and normalized using mouse 18S RNA. (E-G) L929 cells expressing Myr-Akt and Ala and Asp mutants of Thr308 and Ser473 were treated with zVAD.fmk under serum free conditions, followed by viability assay at 24 hr (E), western blot at 9 hr (F), or evaluation of TNFα mRNA levels by qRT-PCR at 9 hrs (G). In all graphs, average±SD was plotted. RIP1 kinase-dependent Thr308 phosphorylation of Myr-Akt during necroptosis increased Myr-Akt activity as it did