Bcl-2 phosphorylation and apoptosis activated by damaged microtubules require mTOR and are regulated by Akt

Article metrics

Abstract

The serine/threonine kinase mTOR, the major sensor of cell growth along the PI3K/Akt pathway, can be activated by agents acting on microtubules. Damaged microtubules induce phosphorylation of the Bcl-2 protein and lower the threshold of programmed cell death, both of which are inhibited by rapamycin. In HEK293 cells expressing Akt mutants, the level of Bcl-2 phosphorylation and the threshold of apoptosis induced by taxol or by nocodazole are significantly modified. In cells expressing dominant-negative Akt (DN-Akt), Bcl-2 phosphorylation and p70S6KThr421/Ser424 phosphorylation induced by taxol or nocodazole were significantly enhanced as compared to cells expressing constitutively active Akt (CA-Akt) and inhibited by rapamycin. Moreover, DN-Akt cells were more sensitive to antitubule agents than CA-Akt cells. In nocodazole-treated HEK293 cells sorted according to cell cycle, the p70S6KThr421/Ser424 phosphorylation was associated to the G2/M fraction. More relevant, nocodazole inhibited, in a dose–response manner, mTOR phosphorylation at Ser2448. This activity, potentiated in DN-Akt cells, was not detectable in CA-Akt cells. Our results suggest that death signals originating from damaged microtubules in G2/M can compete with G1 survival pathways at the level of mTOR. These findings have implications for cancer therapy and drug resistance.

Introduction

The harmonious development of any organism requires a fine equilibrium between the processes of cell proliferation, differentiation and programmed cell death (PCD). These different functions are highly regulated and interdependent (Kozma and Thomas, 2002). Increasing evidence indicates that cell growth, cell cycle progression and apoptosis are sensitively coordinated in response to environmental and developmental conditions, and this may be perturbed in human diseases such as cancer and diabetes (Hara et al., 2002; Kim et al., 2002).

Apoptosis, the last controller of cell functions in regulating the final shape of multicellular organism, is in turn regulated by important signal transduction pathways (Brazil et al., 2002). A rationale has been suggested (Yamamoto et al., 1999; Ruvolo et al., 2001) for lowering the threshold of apoptosis at G2/M to ensure the elimination of cells with aberration of chromosome segregation. Both in yeast and in mammalian cells, withdrawal of nutrients and growth factors leads to the accumulation of cells in the G0/G1 phase by activating a starvation program that includes antiapoptotic signaling (Jacinto and Hall, 2003; Werlen et al., 2003).

In this complex of important mechanisms regulating apoptosis, bcl-2 plays a central role in monitoring cell well-being and in maintaining the genetic programs of the organisms (Cory and Adams, 2002; Borner, 2003). The Bcl-2 family of proteins possess both pro- and antiapototic molecules and their ratio determines, in part, the life or death of the cells. As regards the first member of this family, the antiapoptotic Bcl-2 protein is regulated at the transcriptional level by agents acting at the promoter, including p53 (Decary et al., 2002).

Most relevant is the control at post-transcriptional level including the degradation rate of the bcl-2 messenger RNA (Bevilacqua et al., 2003a, 2003b). The Bcl-2 protein is phosphorylated/inactivated by agents damaging microtubules, raising the possibility that this modification is related to the G2/M cell cycle block (Srivastava et al., 1999; Ling et al., 2002). Multiple kinases have been proposed to mediate the phosphorylation of Bcl-2 as a physiological process during normal cell cycle progression or as a defense mechanism following the activation by a variation of stimuli and stress (JNK, p38) (Ling et al., 1998; Du et al., 2003).

Our previous work has shown the serine/threonine kinase mTOR to be involved in the phosphorylation/inactivation of Bcl-2 (Calastretti et al., 2001a). mTOR, originally found in yeast and conserved in worm, flies and mammals, is a member of the phosphatidylinositol kinase-related protein family downstream in the PI3K/Akt cascade, which was discovered because it is the target of rapamycin (Brown et al., 1994; Sabatini et al., 1994; Sabers et al., 1995; Abraham, 1998).

mTOR can phosphorylate the ribosomal p70 S6 kinase that is known to regulate cell growth by inducing protein synthesis and to signal cell survival by phosphorylating the proapoptotic protein BAD (Harada et al., 2001).

Although displaying a variety of biochemical actions, mTOR was shown to sense nutrients in proliferating cells (Schmelzle and Hall, 2000) and, more recently, to regulate growth in nonproliferating cells such as neurons and muscle (Jacinto and Hall, 2003).

In addition to a plethora of effects as central controller of cell growth, a contrasting role for the mTOR kinase has been indicated in relevant steps of the G2/M phase (Pene et al., 2002). We have shown that the antibiotic rapamycin (Law et al., 2002), the potent and specific inhibitor of mTOR (Dudkin et al., 2001), in arresting cells in the G0/G1 phase, can augment the amount of the Bcl-2 protein by inhibiting its phosphorylation/degradation (Calastretti et al., 2001b). In these studies it has also been shown that mTOR is activated by damaged microtubules (Calastretti et al., 2001a).

These observations lead to the hypothesis that activated mTOR might simultaneously transmit survival and death signals. It is reasonable to argue that mTOR might switch its kinase activity to allow the passage of only one of these signals depending on the mechanism of upstream activation. In contrast to the action of rapamycin, selective downregulation of an upstream activator might not reduce the activity of the other pathways. For instance, inhibition of the survival cascade might allow the death signal to be transmitted, namely the phosphorylation/inactivation of Bcl-2. Indirect evidence suggests that downregulation of the PI3K cascade upstream of mTOR sensitizes tumor cells to cytotoxic agents and chemotherapy (Hill and Hemmings, 2002; Krystal et al., 2002).

In this study, by using chemical and molecular genetics (Mayer, 2003), we investigated whether Akt, the upstream activator of mTOR in survival signaling, might regulate the death signals originated by damaged microtubules. We found that genetically modified Akt significantly regulates the Bcl-2 phosphorylation and cell death induced by antitubule agents. These effects, inhibited by rapamycin, indicate that mTOR can regulate survival and death signals simultaneously.

Results

In previous studies we showed that mTOR can transmit death signals from damaged microtubules (Calastretti et al., 2001a, 2001b). The possibility that mTOR can transmit life or death signals immediately implies that Akt, the upstream kinase of mTOR in the survival pathway, might regulate the transmission of the death signals. The phosphorylation of the proteins Bcl-2 and p70S6K downstream of mTOR along the two alternative pathways and the apoptotic threshold were studied in cells treated with the antitubule agents nocodazole or taxol.

At the biochemical level these agents are known to have different mechanisms of action. Nocodazole exerts a depolymerising action (Jordan and Wilson, 1998) whereas taxol is a microtubule-stabilizing agent (Blagosklonny and Fojo, 1999), both altering the mitotic apparatus of cells and arresting in the G2/M phase (Haldar et al., 1997). The effects of the specific inhibitor of mTOR, rapamycin, on the biochemical events induced by antitubule agents downstream mTOR were also studied in cells expressing Akt mutants. Molecular genetics was used to study the effects of Akt on the kinase activity of mTOR by transfecting HEK293 cells with the constitutively active Akt gene (CA-Akt) or with the dominant-negative Akt gene (DN-Akt).

Preliminary studies aimed at regulating the kinase activity with the Akt inhibitors wortmannin or LY294002 (Walker et al., 2000) were unsuccessful because these agents were also effective on mTOR (Koyasu, 2003).

mTOR is required for the Bcl-2 phosphorylation

In the HEK293 cells, Bcl-2 phosphorylation induced by treatment with nocodazole was inhibited by rapamycin quite efficiently, as shown in Figure 1a. Besides p27 protein, inhibition of Bcl-2 phosphorylation by rapamycin caused a dose–response augmentation of the level of Bcl-2 protein, while cell cycle related proteins were not changed (Figure 1b). The activity of mTOR in HEK293 cells in which microtubules were damaged by taxol or nocodazole was measured by a kinase assay on the 4E-BP1 substrate. Figure 1c shows that nocodazole was able to activate mTOR in a dose-dependent manner and rapamycin inhibited kinase activity. In cells forced to express rapamycin-resistant mTOR, inhibition of the kinase activity by rapamycin was significantly reduced; this observation might support the argument for a role of mTOR in the transmission of the death signals from damaged microtubules.

Figure 1
figure1

Rapamycin inhibition of nocodazole activity in HEK293 cells. (a) Inhibition of nocodazole-induced Bcl-2 phosphorylation by rapamycin. HEK293 cells, pretreated overnight with rapamycin (100 ng/ml), were treated with nocodazole at the indicated doses for further 16 h. Cell extracts were analysed by immunoblotting using anti-Bcl-2 and anti-β-actin antibodies. Actin was used as control for loading same amount of proteins. (b) Increased level of Bcl-2 and p27 proteins by rapamycin. HEK293 cells were treated with rapamycin for 24 h at the indicated doses. Cell extracts were analysed by immunoblotting to value the expression of the apoptosis and cell cycle related proteins, as indicated. (c) Reduced inhibition by rapamycin of nocodazole-induced activation of mTOR in cells transfected with mTOR/RR. HEK293 cells were stably transfected with FLAG-mTOR or Myc-mTOR-RR(S2035) constructs (RR: rapamycin resistant). Cells, pretreated with rapamycin (100 ng/ml) as above, were treated with nocodazole at the indicated doses for further 16 h. The mTOR protein, immunoprecipitated with anti-FLAG or anti-c-Myc antibodies, was analysed for kinase activity with the [γ-32P]labeled ATP using human recombinant 4E-BP1 as substrate. An aliquot of the immunoprecipitated mTOR was immunoblotted and visualized with an anti-mTOR monoclonal antibody (lower panel). Experiments in A, B and C were repeated three times with similar results. Representative blots are reported in the figure

Akt regulates Bcl-2 phosphorylation by mTOR

The next step was to determine whether the level of Bcl-2 phosphorylation could be dependent on Akt, the kinase upstream of mTOR along the survival pathway. HEK293 cell clones in which the functions of Akt were up- or downregulated by stable transfection with CA-Akt or with DN-Akt were studied. The Akt kinase activity (Figure 2a) and the mTOR-dependent p70S6K phosphorylation as activated by insulin (Figure 2b) were evaluated in mutant cells.

Figure 2
figure2

Phosphorylation of Bcl-2 protein in HEK293 cells expressing Akt-mutants. (a) Analysis of Akt activity. Upper panel, serum-deprived (24 h) G7 neo, L10 CA-Akt and C3 DN-Akt clones were treated with insulin (10 μ M) for 30 min. Cell extracts were subjected to immunoprecipitation with anti-Akt antibody and analysed for Akt activity by in vitro kinase assay, using antiphospho-GSK-3 antibody. Lower panel, level of Myc-DN-Akt or Myc-CA-Akt was measured by immunoblotting with anti-Akt antibody, as control for loading same amount of proteins. (b) Insulin-induced p70S6KSer411 phosphorylation in HEK293 cells expressing Akt-mutants. Serum-deprived (24 h) G7 neo, C1 and C3 DN-Akt, L5 and L10 CA-Akt clones were treated with insulin (10 μ M) for 30 min or with rapamycin (100 ng/ml) for 1 h and insulin for further 30 min. Cell extracts were immunoblotted with antiphospho-p70S6KSer411 antibody. (c) Nocodazole-induced phosphorylation of Bcl-2. Left panel, Bcl-2 phosphorylation measured in HEK293 cells expressing Myc-tagged DN-Akt or Myc-tagged CA-Akt. The cell clones were treated with nocodazole (25, 50, 100, 200 nM) for 16 h. Total protein lysates from G7 neo, C1 and C3 DN-Akt, L5 and L10 CA-Akt clones were analysed by immunoblotting with anti-Bcl-2 MoAb. Right panel, level of expression of Myc-DN-Akt kinase in C1 or C3 clones and Myc-CA-Akt in L5 or L10 clones was measured by immunoblotting with anti-c-Myc MoAb. (d) Taxol-induced phosphorylation of Bcl-2. Bcl-2 phosphorylation was measured in HEK293 cells expressing Myc-tagged DN-Akt or CA-Akt kinase. The cells were treated with taxol (10, 30, 100 nM) for 16 h or treated overnight with rapamycin (100 ng/ml) before exposure to taxol. Cell extracts were analysed by immunoblotting with anti-Bcl-2 MoAb as above. (e) mTOR kinase activity in cells expressing Akt-mutants. C1 DN-Akt and L10 CA-Akt cell clones, pretreated with rapamycin (100 ng/ml) as above, were treated with nocodazole at the indicated doses for further 16 h. The mTOR protein, immunoprecipitated with anti-mTOR (mTAb1) antibody, was analysed for kinase activity with the [γ-32P]labeled ATP using human recombinant 4E-BP1 as substrate. An aliquot of the immunoprecipitated mTOR was immunoblotted and visualized with an anti-mTOR monoclonal antibody (lower panel). The experiments were repeated three times with similar results. Representative blots are reported in the figure

The level of Bcl-2 phosphorylation in the HEK293 cells expressing Akt mutants, treated with nocodazole, is shown in Figure 2c and, in those treated with taxol, in Figure 2d. Both pharmacological treatments, although damaging microtubules by opposite mechanisms, induced the same effects.

Bcl-2 phosphorylation was much higher in DN-Akt cells than in cells expressing the CA-Akt gene, supporting the notion that Akt plays an important role in transmitting the death signals by microtubules. One possibility is that by increasing the amount of mTOR engaged in transmitting survival signals might reduce the amount of mTOR in the configuration suitable for activation by microtubules. The upstream survival or death signals might compete for mTOR. According to the present findings activation of mTOR by one signaling cascade switches off the possibility to activate mTOR by the other cascade.

In the cells transfected with the empty vector, Bcl-2 is also phosphorylated quite efficiently. In normal conditions, mTOR is ready to be activated according to the upstream stimuli, including the death signals from damaged microtubules. As expected, rapamycin was able to inhibit Bcl-2 phosphorylation. The degree of inhibition was dependent on the level of phosphorylation. In any case, both pathways of mTOR activation were sensitive to rapamycin.

Moreover, the induction of the mTOR kinase activity by nocodazole was significantly higher in DN-Akt cells than in CA-Akt cells. Rapamycin inhibited mTOR activity in both cell clones, as shown in Figure 2e.

Cytotoxic effects of antitubule drugs in cells expressing Akt-mutants

Cells expressing Akt-mutants treated with antitubule agents were analysed by FACS in order to correlate the level of Bcl-2 phosphorylation with cell death (Nicoletti et al., 1991) and cell cycle modification. Figure 3 shows the DNA distribution in cells exposed to taxol. The sharpest alterations of cell cycle and the highest percentage of cells with sub-G1 DNA content, as shown in the histograms, were found in the DN-Akt cells, in keeping with the Bcl-2 hyperphosphorylation.

Figure 3
figure3

DNA distribution in Akt-mutated cells treated with taxol. HEK293 cell clones, G7 neo, C3 DN-Akt, L5 CA-Akt, were treated with taxol at 3, 10 or 30 nM for 24 h. Cells stained with propidium iodide were analysed by flow cytometry. Histograms are representative of five experiments. The sub-G1 DNA fraction, indicative of apoptosis, is reported as the percentages in each histograms

The cytotoxic effects of antitubule drugs were studied in the clones expressing Akt mutants by counting the cells under the microscope and by the MTT assay (data not shown). As shown in Table 1, the DN-Akt cells were the most sensitive to the effects induced by microtubule-damaging agents, whereas CA-Akt cells were less sensitive than the empty vector transfected cells. Rapamycin treatments did not significantly protect HEK293 cells from taxol death, possibly indicating that death pathways in these cells are not entirely dependent on the mTOR functions regulated by rapamycin (Ofir et al., 2002).

Table 1 Cytotoxic activity of taxol in HEK293 cells expressing Akt-mutants

p70S6K phosphorylation in cells expressing Akt-mutants

In the last part of the work (Figures 4 and 5), the phosphorylation of the p70S6K protein, the downstream target of mTOR along the survival pathway, was studied in some detail. P70S6K phosphorylation at position Ser411 was analysed because its level is increased by insulin or by insulin-like growth factor (IGF) and decreased by rapamycin (Han et al., 1995; Pullen and Thomas, 1997; Avruch, 1998). As expected, phosphorylation at Ser411 was constitutively higher in CA-Akt cells than in DN-Akt cells.

Figure 4
figure4

Phosphorylation of p70S6K at Ser411 in HEK293 cells expressing Akt-mutants treated with nocodazole or taxol. Phosphorylation level of p70S6KSer411 was measured by immunoblotting in G7 neo, C1 DN-Akt, L10 CA-Akt clones. The cells were treated with (a) nocodazole at 50, 100, 200 nM for 16 h or with (b) taxol at 10, 30, 100 nM for 24 h. Cell extracts were analysed by immunoblotting with antiphospho-p70S6KSer411 and with anti-β-actin antibodies. Actin was used as control for loading same amount of proteins. Experiments were repeated three times with similar results. Representative blots are reported in the figure

Figure 5
figure5

Phosphorylation of p70S6K at Thr421/Ser424 in HEK293 cells expressing Akt-mutants treated with nocodazole or taxol. The phosphorylation level of p70S6KThr421/Ser424 was measured in extracts from clones treated with (a, c) nocodazole (50, 100, 200 nM) or (b) taxol (10, 30, 100 nM) for 16 h or treated overnight with rapamycin (100 ng/ml) before exposure to nocodazole and taxol. In (c), cells were serum-deprived for 24 h before the treatments. Cell extracts were analysed by immunoblotting with antiphospho-p70S6KThr421/Ser424 and anti-β-actin antibodies. Actin was used as control for loading same amount of proteins. Experiments in A, B and C were repeated three times with similar results. Representative blots are reported in the figure

In both types of clones phosphorylation at position Ser411 was not significantly modified by exposure to taxol or nocodazole (Figure 4). These studies indicate that the survival signals transmitted by mTOR are not obviously opposed by microtubule signals. Following up recent observations, we also studied p70S6K phosphorylation at Thr421/Ser424, which is induced by damaged microtubules and inhibited by rapamycin (Le et al., 2003). Here the p70S6KThr421/Ser424 phosphorylation by antitubule agents has been confirmed. Moreover, in Figure 5a,b it is shown that the level of p70S6KThr421/Ser424 phosphorylation was significantly higher in the DN-Akt cells. In sharp contrast, in the CA-Akt cells, p70S6KThr421/Ser424 phosphorylation was not significantly increased by the antitubule agents. Figure 5c shows that also upon starvation, DN-Akt cells were sensitive to nocodazole-induced phosphorylation at p70S6KThr421/Ser424. Partial inhibition of p70S6KThr421/Ser424 phosphorylation and low protection of taxol cytotoxicity by rapamycin might suggest additional pathways of phosphorylation not including mTOR.

This phosphorylation was more enhanced in DN-Akt cells than in CA-Akt cells, indicating the reverse relationship between p70S6KThr421/Ser424 phosphorylation and the level of Akt activation. In starved conditions rapamycin was most effective, whereas in starved CA-Akt cells nocodazole was almost not effective as expected.

These findings provide evidence for at least two kinase targets of mTOR upon its activation along the microtubule pathway, the phosphorylation/inactivation of Bcl-2 and the phosphorylation of p70S6K at position Thr421/Ser424. Both mechanisms are significantly potentiated by downregulating Akt, the upstream activator of mTOR in the survival pathway. Whether the microtubule-dependent mTOR kinase directly phosphorylates Bcl-2 and p70S6KThr421/Ser424 is not yet known.

p70S6KThr421/Ser424 phosphorylation occurs at G2/M

HEK293 cells treated with nocodazole were sorted according to the phase of the cell cycle. As shown in Figure 6, p70S6KThr421/Ser424 phosphorylation is decreased in the cells at the G1 phase, whereas abundantly phosphorylated p70S6KThr421/Ser424 was detected in the G2/M fraction. The insulin sensitive p70S6KSer411, not changed by nocodazole (Figure 4 and 6) was equally phosphorylated at G1 or at G2/M fraction. These results indicated that anticancer drugs affecting microtubule integrity induce p70S6KThr421/Ser424 phosphorylation in the G2/M phase of the cell cycle. In this cell fraction, death threshold is lowered further supporting the concept that cell death by antitubule agents is dependent on G2/M accumulation. It might be interesting to note that phosphorylation of Bcl-2 and p70S6KThr421/Ser424 occurs under the same conditions, is potentiated by the downregulation of Akt and is inhibited by rapamycin.

Figure 6
figure6

Phosphorylation of p70S6K in G1 and G2/M phase. HEK293 cells not treated or treated with nocodazole at 100 nM for 16 h were sorted in G1 and G2/M phases by a fluorescence-activated cell sorter. The separated fractions were analysed for p70S6K phosphorylation at Ser411, Thr421, Ser424. Actin was used as control to confirm that equal amounts of protein were loaded

Structural modification of mTOR

The structure of mTOR, according to the origin of the upstream stimuli, has been studied. In Figure 7 it is shown that phosphorylation at position Ser2448, highly increased by insulin or in CA-Akt cells, was decreased, in a dose–response manner, by nocodazole in HEK293 or in DN-Akt cells. These findings suggest that the structure of mTOR might be counter-modified by signals originating from opposing directions. The Akt-dependent phosphorylation of mTORSer2448, inhibited by nocodazole, is not in contrast with the life or death switching role we proposed for mTOR.

Figure 7
figure7

Analysis of mTOR phosphorylation at Ser2448. Serum-starved (24 h) C1 DN-Akt, L10 CA-Akt and G7 neo clones were treated with insulin (10 μ M) for 30 min or with nocodazole for 16 h at the indicated doses. Cell lysates were analysed by immunoblotting using antiphospho-mTORSer2448 and anti-mTOR antibodies

Discussion

Recent studies have addressed the biochemical and molecular mechanisms regulating life and death, the two essential functions of cells and organisms. A picture is emerging in which the same metabolic cascades may regulate both pathways (Fisher, 2001; Green and Evan, 2002; Pelengaris et al., 2002; Chau and Wang, 2003). For instance, a growth factor-dependent cascade can lead to cell survival if the receptor is activated by the cognate ligand (Vincent and Feldman, 2002; Chau and Wang, 2003). The same metabolic cascade can activate a death program if the ligand is withdrawn (Ferrara, 2002) or the signaling is interrupted (Church, 2003; Kurzrock et al., 2003). The tremendous achievements in the molecular biochemistry of life and death processes have facilitated and simplified the understanding of these essential functions in cells. Several lines of evidence implicate a limited number of molecular functions that regulate basic pathways, the disruption of which contributes to most, if not all, cancers (Hahn and Weinberg, 2002; Johnstone et al., 2002).

The most important system for signal transduction of the cell growth pathway is the phosphatidylinositol 3-kinase (PI3K/Akt) cascade that includes the downstream terminal kinase mTOR (Sekulic et al., 2000; Chen and Fang, 2002). However, death signals are surprisingly also regulated by mTOR and inhibited by rapamycin (Calastretti et al., 2001a). We have previously shown that the Bcl-2 phosphorylation and apoptosis induced by damaged microtubules can be inhibited by prior treatment of the cells with rapamycin. Those findings indicated that death signals activated by damaged microtubules are dependent on mTOR. Thus, one protein kinase can transmit the entirely opposite signals, growth or death. As it is widely agreed that survival signals are mainly transmitted in the G1/S phase (Agami and Bernards, 2002) and death signals in the G2/M (Taylor and Stark, 2001), it might be predicted that mTOR might switch the signaling in accordance with the phase of the cell cycle.

This work was designed to obtain more insight into the molecular mechanisms regulating the death or survival signals as mediated by mTOR. The concept to prove was whether the inhibition of the Akt kinase upstream of mTOR might potentiate Bcl-2 phosphorylation and the death program initiated by damaged microtubules. Studies in constitutively active Akt cells were expected to produce findings in the opposite direction.

The role of mTOR in transmitting survival or death signals was studied by analyzing its site of phosphorylation or those in downstream proteins. It was found that:

(a):

nocodazole and taxol activate mTOR, phosphorylate Bcl-2 and p70S6K at Thr421/Ser424;

(b):

these effects are augmented in DN-Akt cells and decreased in CA-Akt cells;

(c):

phosphorylation of mTOR at Ser2448 is induced by insulin and inhibited by nocodazole;

(d):

phosphorylation of p70S6K at Thr421/Ser424 induced by nocodazole is augmented at G2/M and decreased at G1.

These findings provide evidence that the mTOR kinase can transmit the two opposing signals. Moreover, studies modulating Akt, the upstream activator of mTOR, indicated the molecular mechanisms by which the survival signaling can inhibit Bcl-2 phosphorylation and cell death. In the model depicted in Figure 8, mTOR functions as a molecular switch whereby the transmission of one signal excludes the other. The model is compatible with the assumption that survival signals are transmitted in the G1 phase and the death signals in the G2/M. Rapamycin, by occupying the activation site of mTOR, can arrest both signals.

Figure 8
figure8

Model depicting the switch role of mTOR. mTOR can control growth pathways at G1 or PCD pathways at G2/M, respectively. At structural level, mTOR is phosphorylated at Ser2448 by activated Akt or dephoshorylated at the same position by damaged microtubules

In addition to lowering the death threshold by acting on Bcl-2, mTOR has a role to phosphorylate p70S6K at position Thr421/Ser424. We show here that Bcl-2 (Haldar et al., 1997, 1998) and p70S6KThr421/Ser424 are both significantly over-phosphorylated in the G2/M phase, which argues for a coordinated action. A recent paper (Le et al., 2003) shows that taxol phosphorylates the protein kinase p70S6K, the immediate target of mTOR, at position Thr421/Ser424. This metabolic event might modify the kinase activity of p70S6K, in sharp contrast with the effects induced by the phosphorylation of p70S6K at position Ser411. The role of mTOR is once more inferred because the phosphorylation of the p70S6KThr421/Ser424 is reduced by rapamycin. Thus, microtubules can activate death signals by a kinase cascade including mTOR and p70S6KThr421/Ser424.

The mTOR-requiring kinase cascade that phosphorylates Bcl-2 and p70S6KThr421/Ser424 after microtubule damage is not yet known. It seems likely that the positions phosphorylated by the kinase inherent in the microtubule cascade are distinct from those phosphorylated by Akt, and that each kinase cascade induces, by site-specific phosphorylation, a molecular configuration of mTOR suitable to transmit the message to the desired targets. This hypothesis might be supported by the observation that mTOR phosphorylation at position Ser2448, required for the transmission of the synthetic program (Sekulic et al., 2000), is induced by growth signals, while is inhibited by death signals from microtubules.

It is shown here that two mTOR configurations, mutually exclusive, are possible. The phosphorylation at Ser2448 opens the gate for the growth signals and closes the gate to the death signals. The inhibition of Ser2448 phosphorylation can allow the transmission of death signals. Since rapamycin can inhibit both pathways of mTOR activation, the sites of phosphorylation triggered by microtubules might be located close to the Akt phosphorylation sites in the FKBP12-rapamycin binding domain (Chen and Fang, 2002; McMahon et al., 2002). Future studies will strive to understand the upstream kinases and determine the precise site of mTOR activation by microtubules.

These findings fit pretty well into the biochemical mechanisms belonging to the survival cascade. Activation of Akt–mTOR pathway induces a survival program by upregulating Bcl-2 (inhibition of Bcl-2 phosphorylation/degradation) and stimulates a synthetic program by activating p70S6KSer411 and 4E-BP1.

Our findings have relevance for cancer therapy because they indicate the molecular mechanisms by which inhibition of the survival cascade upstream of mTOR can potentiate the death cascade that starts from damaged microtubules. The development of a specific inhibitor of the Akt kinase might increase the antitumor activity of cytotoxic drugs and provide a new mode of reducing drug resistance. The action of mTOR on Bcl-2 is most promising when one considers the role of Bcl-2 protein in tumor transformation and in resistance to antitumor drugs.

Materials and methods

Plasmids and chemicals

FLAG-mTOR cDNA or Myc-mTOR/RR cDNA, containing S2035T (RR) mutation inducing resistance to a broad range of rapamycin concentrations (Brown et al., 1995), kindly supplied by Dr Jie Chen (Kim and Chen, 2000), were inserted in the plasmid pCDNA3. Plasmid pUSEamp+ (Upstate Biotechnology, Lake Placid, NY, USA), kindly supplied by Dr A Gulino, was used to obtain cells expressing Akt mutants. The cDNA encoding for Myc-His Akt1, tagged at the 3′ end of the Akt1 ORF and with methionine substituted for lysine at residue 179 (K179M), expresses a mutated catalytic domain devoid of kinase activity (dominant-negative Akt, DN-Akt). Myc-His 3′ tagged Akt1 cDNA containing sequences corresponding to aminoacids 1–11 of avian c-src at 5′, which are required for protein myristoylation and activation, was also inserted into the plasmid (constitutively active Akt, CA-Akt). Vacant plasmid was used as control (neo). Rapamycin, taxol and nocodazole were purchased from Sigma-Aldrich, solubilized as indicated by the manufacturer and maintained at −20°C.

Cell culture, transfection, sorting

Cell culture

Human embryonic kidney (HEK) 293 cells were grown in Dulbecco's modified Eagle's medium containing 10% heat-inactivated FCS (HyClone Laboratories, Logan, UT, USA), 2 mM glutamine, 50 IU/ml penicillin and 50 μg/ml streptomycin (Sigma-Aldrich, Milan, Italy) at 37°C with 5% CO2. The cells were seeded in 60-mm Petri dishes at a density of 3 × 105 cells per dish. After 24 h, the clones were treated with taxol or nocodazole at indicated doses and times and rapamycin overnight, when indicated. Growth rate and viability of the cells were determined by the Trypan Blue exclusion assay and by the 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyl tetrazolium bromide (MTT)-based colorimetric assay (Mosmann, 1983).

Transfection

Plasmid DNA (2 μg) was mixed in serum-free medium Optimem (GIBCO, BRL) with 20 μl of lipofectin reagent (GIBCO, BRL) and incubated for 5 h with 5 × 105 HEK 293 cells the day after seeding, according to the manufacturer's instructions. Cells were incubated with DMEM fresh medium supplemented with 10% FCS for 48 h and selected for 3 weeks in the presence of G418 (GIBCO, BRL). Resistant cells, cloned in the medium supplemented with G418 (0.8 mg/ml), gave rise to a battery of pUSE/DN-Akt (C1 and C3 DN-Akt), pUSE/CA-Akt (L5 and L10 CA-Akt) or pUSE/neo (G7 neo) clones. The expression of Myc-tagged dominant-negative or Myc-tagged constitutively active Akt was analysed in each clone by Western blot.

Cell sorting

HEK293 cells, treated with nocodazole at 100 nM for 16 h, were stained with Hoecst 33342 (Sigma) at a concentration of 15 μg/ml for 30 min at 37°C, washed in PBS and sorted in G1 or G2/M fractions by a fluorescence-activated cell sorter (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA).

Western blot analysis

Western blots were performed as follows. Samples of 5 × 106 cells were collected by centrifugation at 1000 r.p.m. for 10 min, washed with cold PBS, resuspended in lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.2% Triton X-100, 0.3% Nonidet P-40, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, 10 μg/ml leupeptin, 10 μg/ml aprotinin) and incubated on ice for 30 min. The lysates were centrifuged at 13 000 r.p.m. for 15 min at 4°C. Equivalent amounts of proteins were analysed by SDS–polyacrylamide gel electrophoresis (PAGE). After electrophoretic separation, the proteins were transferred onto nitrocellulose membrane (Amersham Biosciences). After 1 h incubation in blocking solution (5% dried milk in PBS), filters were incubated with the appropriate antibodies to: mTOR (Oncogene Research Products), Bcl-2 (Santa Cruz Biotechnology), c-myc (Sigma), p27 (DAKO), p21 (Oncogene Research Products), phospho-p70S6KSer411 (S. Cruz Biotechnology), all MoAbs; phospho-mTORSer2448, phospho-p70S6KThr421/Ser424 (Cell Signaling Technology), CDK2, cyclin E, Bcl XL, Bax (S. Cruz Biotechnology), or β-actin (Sigma-Aldrich) were obtained in rabbit.

Proteins were visualized with peroxidase-coupled secondary antibody (Amersham Biosciences), using enhanced chemiluminescence (ECL) for detection (Amersham Biosciences).

The phosphorylated form of Bcl-2 protein was measured as previously indicated (Calastretti et al., 2001b). Cells, 5 × 106, were collected, lysed in Laemmli buffer (Laemmli, 1970) and sonicated for 20 s. Supernatants, added to a volume of 1 : 10 of BMG (glycerol: β-mercaptoethanol : bromo-phenol-blue, 1%/2 : 5 : 5), were boiled for 3 min and aliquots of 150 μg were separated by 12% SDS–PAGE. Proteins, blotted onto nitrocellulose membrane were immunodetected with anti-Bcl-2 (DAKO) MoAb. Visualization was carried out by means of anti-mouse peroxidase-coupled secondary antibody, using ECL (Amersham Biosciences).

Kinase assays

mTOR kinase assay

Cells treated as indicated were washed in cold PBS, lysed in lysis buffer and then immunoprecipitated with indicated antibodies as described (Brunn et al., 1997). The immune complexes, suspended in 50 μl of kinase buffer (Calastretti et al., 2001a, 2001b), were incubated with 5 μCi of [γ-32P]ATP and analysed by SDS–PAGE electrophoresis. Radiolabeled 4E-BP1 was detected by autoradiography and quantified with an Ambis imaging system.

Akt kinase assay

The Akt Kinase Assay Kit (Cell Signaling Technology) was used according to the manufacturer's instructions. Cells were washed and lysed for 30 min on ice. The cell extracts were centrifuged at 13 000 r.p.m. for 15 min at 4°C. Immunoprecipitates, obtained with Immobilized α-Akt antibody, were incubated for 30 min at 30°C in 40 μl of kinase buffer supplemented with 200 μ M ATP and 1 μg of GSK-3 fusion protein (IndexTermCGPKGPGRRGRRRTSSFAEG), as substrate. Samples, boiled for 5 min, were analysed by 12% SDS–PAGE, blotted and the level of phosphorylated GSK-3 detected with α-phospho-GSK-3 α/β (Ser21/9) antibody.

Analysis of cellular DNA content by flow cytometry

Cells were collected, washed and stained for 30 min at 37°C with 1 ml of DNA-staining solution in 0.1%. Nonidet P-40 (Sigma-Aldrich, Milan, Italy), 0.5 mg/ml RNAse (type IIIA, Sigma-Aldrich) and 25 μg/ml propidium iodide (Sigma-Aldrich)(Nicoletti et al., 1991). The cellular DNA content was analysed by FACScalibur (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA) using Cell Quest software system for histograms of propidium iodide fluorescence intensity vs cell frequency (Becton Dickinson).

References

  1. Abraham RT . (1998). Curr. Opin. Immunol., 10, 330–336.

  2. Agami R and Bernards R . (2002). Cancer Lett., 177, 111–118.

  3. Avruch J . (1998). Mol. Cell. Biochem., 182, 31–48.

  4. Bevilacqua A, Ceriani MC, Canti GF, Asnaghi L, Gherzi R, Brewer G, Papucci L, Schiavone N, Capaccioli S and Nicolin A . (2003a). J. Biol. Chem., 278, 23451–23459.

  5. Bevilacqua A, Ceriani MC, Capaccioli S and Nicolin A . (2003b). J. Cell Physiol., 195, 356–372.

  6. Blagosklonny MV and Fojo T . (1999). Int. J. Cancer, 83, 151–156.

  7. Borner C . (2003). Mol. Immunol., 39, 615–647.

  8. Brazil DP, Park J and Hemmings BA . (2002). Cell, 111, 293–303.

  9. Brown EJ, Albers MW, Shin TB, Ichikawa K, Keith CT, Lane WS and Schreiber SL . (1994). Nature, 369, 756–758.

  10. Brown EJ, Beal PA, Keith CT, Chen J, Shin TB and Schreiber SL . (1995). Nature, 377, 441–446.

  11. Brunn GJ, Fadden P, Haystead TA and Lawrence JC . (1997). J. Biol. Chem., 272, 32547–32550.

  12. Calastretti A, Bevilacqua A, Ceriani MC, Viganò S, Zancai P, Capaccioli S and Nicolin A . (2001a). Oncogene, 20, 6172–6180.

  13. Calastretti A, Rancati F, Ceriani MC, Asnaghi L, Canti G and Nicolin A . (2001b). Eur. J. Cancer, 37, 2121–2128.

  14. Chau BN and Wang JY . (2003). Nat. Rev. Cancer, 3, 130–138.

  15. Chen J and Fang Y . (2002). Biochem. Pharmacol., 64, 1071–1077.

  16. Church AC . (2003). QJM, 96, 91–102.

  17. Cory S and Adams JM . (2002). Nat. Rev. Cancer, 2, 647–656.

  18. Decary S, Decesse JT, Ogryzko V, Reed JC, Naguibneva I, Harel-Bellan A and Cremisi CE . (2002). Mol. Cell. Biol., 22, 7877–7888.

  19. Du L, Lyle CS, Hall T and Chambers TC . (2003). Proceedings of the 94th AACR Annual Meeting.Toronto, Ontario, Canada.

  20. Dudkin L, Dilling MB, Cheshire PJ, Harwood FC, Hollingshead M, Arbuck SG, Travis R, Sausville EA and Houghton PJ . (2001). Clin. Cancer Res., 7, 1758–1764.

  21. Ferrara N . (2002). Nat. Rev. Cancer, 10, 795–803.

  22. Fisher DE . (2001). Apoptosis, 6, 7–15.

  23. Green DR and Evan GI . (2002). Cancer Cell., 1, 19–30.

  24. Hahn WC and Weinberg RA . (2002). Nat. Rev. Cancer, 2, 331–341.

  25. Haldar S, Basu A and Croce CM . (1997). Cancer Res., 57, 229–233.

  26. Haldar S, Basu A and Croce CM . (1998). Cancer Res., 58, 1609–1615.

  27. Han JW, Pearson RB, Dennis PB and Thomas G . (1995). J. Biol. Chem., 270, 396–403.

  28. Hara K, Maruki Y, Long X, Yoshino K, Oshiro N, Hidayat S, Tokunaga C, Avruch J and Yonezawa K . (2002). Cell, 110, 177–189.

  29. Harada H, Andersen JS, Mann M, Terada N and Korsmeyer SJ . (2001). Proc. Natl. Acad. Sci. USA, 98, 9666–9670.

  30. Hill MM and Hemmings BA . (2002). Pharmacol. Ther., 93, 243–251.

  31. Jacinto E and Hall MN . (2003). Nat. Rev. Mol. Cell. Biol., 4, 117–126.

  32. Johnstone RW, Ruefli AA and Lowe SW . (2002). Nat. Rev. Drug. Discovery, 1, 287–299.

  33. Jordan MA and Wilson L . (1998). Curr. Opin. Cell. Biol., 10, 123–130.

  34. Kim JE and Chen J . (2000). Proc. Natl. Acad. Sci., 97, 14340–14345.

  35. Kim DH, Sarbassov DD, Ali SM, King JE, Latek RR, Erdjument-Bromage H, Tempst P and Sabatini DM . (2002). Cell, 110, 163–175.

  36. Koyasu S . (2003). Nat. Immunol., 4, 313–319.

  37. Kozma SC and Thomas G . (2002). Bioessays, 24, 65–71.

  38. Krystal GW, Sulanke G and Litz J . (2002). Mol. Cancer Ther., 1, 913–922.

  39. Kurzrock R, Kantarjian HM, Druker BJ and Talpaz M . (2003). Ann. Intern. Med., 138, 819–830.

  40. Laemmli UK . (1970). Nature, 227, 680–685.

  41. Law BK, Chytil A, Dumont N, Hamilton EG, Waltner-Law ME, Aakre ME, Covington C and Moses HL . (2002). Mol. Cell. Biol., 22, 8184–8198.

  42. Le XF, Hittelman WN, Liu J, McWatters A, Li C, Mills GB and Bast Jr RC . (2003). Oncogene, 22, 484–497.

  43. Ling YH, Liebes L, Ng B, Buckley M, Elliott PJ, Adams J, Jiang JD, Muggia FM and Perez-Soler R . (2002). Mol. Cancer Ther., 1, 841–849.

  44. Ling YE, Tornos C and Perez-Soler R . (1998). J. Biol. Chem., 273, 18984–18991.

  45. Mayer TU . (2003). Trends Cell Biol., 13, 270–277.

  46. McMahon LP, Choi KM, Lin TA, Abraham RT and Lawrence Jr JC . (2002). Mol. Cell. Biol., 22, 7428–7438.

  47. Mosmann T . (1983). J. Immunol. Methods, 65, 55–63.

  48. Nicoletti I, Migliorati G, Pagliacci MC, Grignani F and Riccardi C . (1991). J. Immunol. Methods, 139, 271–279.

  49. Ofir R, Seidman R, Rabinski T, Krup M, Yavelsky V, Weinstein Y and Wolfson M . (2002). Cell Death Differ., 9, 636–642.

  50. Pelengaris S, Khan M and Evan G . (2002). Nat. Rev. Cancer, 2, 764–776.

  51. Pene F, Claessens YE, Muller O, Viguie F, Mayeux P, Dreyfus F, Lacombe C and Bouscary D . (2002). Oncogene, 21, 6587–6597.

  52. Pullen N and Thomas G . (1997). FEBS Lett., 23, 78–82.

  53. Ruvolo PP, Deng X and May WS . (2001). Leukemia, 15, 515–522.

  54. Sabatini DM, Erdjument-Bromage H, Lui M, Tempst P and Snyder SH . (1994). Cell, 78, 35–43.

  55. Sabers CJ, Martin MM, Brunn GJ, Williams JM, Dumont FJ, Wiederrecht G and Abraham RT . (1995). J. Biol. Chem., 270, 815–822.

  56. Schmelzle T and Hall MN . (2000). Cell, 103, 253–262.

  57. Sekulic A, Hudson CC, Homme JL, Yin P, Otterness DM, Karnitz LM and Abraham RT . (2000). Cancer Res., 60, 3504–3513.

  58. Srivastava RK, Mi QS, Hardwick JM and Longo DL . (1999). Proc. Natl. Acad. Sci. USA, 96, 3775–3780.

  59. Taylor WR and Stark GR . (2001). Oncogene, 20, 1803–1815.

  60. Vincent AM and Feldman EL . (2002). Growth Horm IGF Res., 12, 193–197.

  61. Walker EH, Pacold ME, Perisic O, Stephens L, Hawkins PT, Wymann MP and Williams RL . (2000). Mol. Cell., 6, 909–919.

  62. Werlen G, Hausmann B, Naeher D and Palmer E . (2003). Science, 299, 1859–1863.

  63. Yamamoto K, Ichijo H and Korsmeyer SJ . (1999). Mol. Cell. Biol., 19, 8469–8478.

Download references

Acknowledgements

We thank Dr J Chen for plasmids and critically reading the manuscript, Dr A Gulino for plasmids, Dr P Woodford for revision of the manuscript and E Fontanella for helpful FACS assistance. This work was supported by grants from AIRC, Fondazione CARIPLO, Milan, Italy; MIUR, CNR-Project Oncology, Rome, Italy.

LA and AC were supported by a fellowship from FIRC.

Author information

Correspondence to Angelo Nicolin.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • mTOR
  • BCL-2
  • microtubules
  • Akt
  • p70S6K
  • apoptosis

Further reading