Potent inhibition of small-cell lung cancer cell growth by simvastatin reveals selective functions of Ras isoforms in growth factor signalling

Abstract

The impact of the 3-hydroxy-3methylglutaryl CoA reductase inhibitor simvastatin on human small-cell lung cancer (SCLC) cell growth and survival was investigated. Simvastatin profoundly impaired basal and growth factor-stimulated SCLC cell growth in vitro and induced apoptosis. SCLC cells treated with simvastatin were sensitized to the effects of the chemotherapeutic agent etoposide. Moreover, SCLC tumour growth in vivo was inhibited by simvastatin. These responses correlated with the inhibition of stem cell factor (SCF)-stimulated activation of extracellular signal-regulated kinase (Erk), protein kinase B (PKB) and ribosomal S6 kinase by simvastatin. Constitutive activation of the Erk pathway was sufficient to rescue SCLC cell from the effects of simvastatin. The drug did not directly affect activation of c-Kit or its localization to lipid rafts, but in addition to its ability to block Ras membrane localization, it selectively downregulated H-Ras protein levels at the post-translational level. Downregulation of either H- or K-Ras by RNA interference (RNAi) did not impair Erk activation by growth factors, whereas an RNAi specific for N-Ras inhibited activation of Erk, PKB and SCLC cell growth. Together our data demonstrate that inhibiting Ras signalling with simvastatin potently disrupts growth and survival in human SCLC cells.

Introduction

Lung cancer is a major cause of death, and despite improvements in diagnosis and therapy, the overall 5-year survival rate is less than 5%. Small-cell lung cancer (SCLC) represents 20% of all cases of lung cancer and is strongly correlated with cigarette smoking. In recent years, considerable information concerning the molecular abnormalities involved in SCLC pathogenesis has emerged. These abnormalities include neuro-endocrine regulatory peptides, overexpression of Myc family oncogenes and genetic abnormalities in the tumour suppressor genes p53 and pRB (Wistuba et al., 2001). Ras mutations do not occur in human SCLC, whereas they are frequently found in non-small-cell lung cancer (NSCLC), where the K-Ras oncogene is mutated in about 30% of cases (Mitsudomi et al., 1991; Salgia and Skarin, 1998; Wistuba et al., 2001).

Polypeptide growth factors such as stem cell factor (SCF) and fibroblast growth factor-2 (FGF-2) induce a variety of responses in human SCLC cells, including growth and proliferation, chemoresistance and motility (Heasley, 2001; Pardo et al., 2001, 2002; Arcaro et al., 2002). Activation of two major intracellular signalling pathways, the mitogen-activated Erk kinase (MEK)/extracellular signal-regulated kinase (Erk) and phosphoinositide 3-kinase (PI3K)/protein kinase B (PKB) pathways have been shown to mediate SCLC responses to polypeptide growth factors and thus represent attractive targets for the development of new drugs targeting SCLC in patients (Pardo et al., 2001, 2002; Arcaro et al., 2002; Krystal et al., 2002). Indeed, recent studies have reported that interfering with PKB, ribosomal protein S6 kinase (S6K) and MEK/Erk activation results in negative effects on SCLC growth and chemoresistance (Moore et al., 1998; Pardo et al., 2001, 2002; Arcaro et al., 2002; Krystal et al., 2002). Since Ras family proteins transduce receptor tyrosine kinase signals to both the Erk and PI3K signalling cascades, inhibiting their function in SCLC cells may have a negative impact on tumour cell proliferation and survival.

3-Hydroxy-3methylglutaryl CoA (HMG-CoA) reductase inhibitors, commonly referred to as statins, block prenylation of the Ras superfamily of proteins, and have been found to have antiproliferative, proapoptotic and antimetastatic effects on a variety of human tumour cells (Chan et al., 2003; Graaf et al., 2004b). The exact mode of action of the various statins in human cancer cells remains the subject of debate. Some reports have shown that the negative effect of statins on tumour cell proliferation is potentially caused by a block in the G1–S transition in the cell cycle, attributable to increases in p21WAF1/CIP1 and p27KIP1, by effects on Rho family GTPase(s) (Hirai et al., 1997; Lee et al., 1998). Delocalization of RhoA from the cell membrane by statins also reduced invasion of human pancreatic cell lines (Kusama et al., 2001). Various reports have claimed that Ras inactivation by statins is not the main target of the drugs in their effects on proliferation and apoptosis, although the main line of evidence for this hypothesis is rather indirect, since it involved reversal of the effects of statins by exogenously added lipids (Chan et al., 2003). However, the MEK/Erk and PI3K pathways have also been reported to be targeted by statins in human tumour cells, but the evidence is somewhat conflicting (Graaf et al., 2004b).

Here we report that simvastatin blocks SCLC cell proliferation and present evidence that the PI3K/PKB and MEK/Erk signalling pathways mediate the effects of the drug on the tumour cells.

Results

Ras family proteins are upregulated in human SCLC cell lines

The protein expression levels of small GTPases of the Ras superfamily was compared between purified human Type II lung epithelial cells and a panel of eight human SCLC cell lines. Western blot analysis of total cell lysates revealed that all three H-, K- and N-Ras isoforms were expressed in the lung tumour cell lines (Figure 1). Moreover, the protein levels of H-Ras, N-Ras and Rac1 were elevated in the SCLC cell lines, as compared to Type II epithelial cells, while there was no increase in the levels of the guanine nucleotide exchange factors SOS-1 and Eps8. Together these data indicate that upregulation of Ras family GTPases may contribute to the abnormal growth and survival properties of human SCLC cells, which have not been reported to express oncogenically activated mutants of Ras (Mitsudomi et al., 1991; Wistuba et al., 2001).

Figure 1
figure1

Expression of Ras superfamily proteins in human SCLC cell lines. Equal amounts of lysates of purified human Type II lung epithelial cells (from two donors: I and II) or human SCLC cell lines (as indicated) were analysed by SDS–PAGE and Western blotting with antibodies specific for the proteins indicated.

Simvastatin inhibits SCLC cell growth in vitro

To investigate the contribution of Ras superfamily GTPases to SCLC cell proliferation, the impact of the HMG-CoA reductase inhibitor simvastatin on H-69, H-209 and H-510 cell growth in liquid culture was studied. Simvastatin inhibited proliferation of all three SCLC cell lines grown in serum-containing medium, in a dose-dependent manner, with IC50s ranging from 3.5 μg/ml (8.4 μ M, H-510) to 19.0 μg/ml (45.4 μ M, H-69) (Figure 2a). Intriguingly, the relative resistance of H-69 cells to simvastatin treatment correlated with the previously reported enhanced activation of PKB in this cell line, as compared to H-209 and H-510 cells (Arcaro et al., 2002). In addition, simvastatin (3.3 μg/ml) completely blocked basal H-69 and H-209 cell growth in serum-free medium (Figure 2b). This effect was also observed when SCLC cell proliferation in response to SCF (10 ng/ml) was investigated (Figure 2b). Under this condition, increased cell proliferation in response to SCF was observed in H-69 but not in H-209 cells (Figure 2b), as previously reported (Arcaro et al., 2002). Since MEK/Erk signalling has been shown to be critical for SCLC cell growth in response to polypeptide growth factor stimulation (Pardo et al., 2001), we tested the impact of the specific MEK inhibitor PD098059 on H-69 growth in liquid culture. The inhibitor blocked basal and SCF-stimulated H-69 SCLC cell proliferation in serum-free medium (data not shown), which supports the model that Ras signalling to MEK/Erk is essential for SCLC proliferation.

Figure 2
figure2

Simvastatin blocks basal and growth factor-stimulated SCLC cell growth in vitro. (a) H-69 (diamonds) H-209 (squares), or H-510 (triangles) were incubated with increasing concentrations of simvastatin in serum-containing medium. Cell proliferation was assessed using the MTS assay after 5 days. The data are mean with s.d. from eight independent experiments. (b) H-69 or H-209 cells were incubated with simvastatin (3.3 μg/ml) in serum-free medium in the presence or absence of SCF (10 ng/ml). Cell counting was performed at 0 days (dark grey bars) and after 8 days (light gray bars). The data are mean with s.d. from four independent experiments. (c) H-69 cells treated with vehicle (open bars) or simvastatin (closed bars) were stained with PI and the percentage of cells in subG1, G1, S and G2/M phase determined by FACS analysis. Data are mean with s.d. from three independent experiments. (d) H-69 cells were incubated in the presence of simvastatin (3.3 μg/ml) in serum-free medium for 24 h. Total cell lysates were analysed by SDS–PAGE and Western blot with antibodies specific for PARP or activated (cleaved) caspase-3. The cleaved fragment of PARP (85 kDa) is shown. As a control, samples of cells treated with a caspase inhibitor (Z-VAD) were analysed in parallel.

In view of the ability of simvastatin to inhibit basal SCLC cell growth, we then investigated whether the drug induced apoptosis in H-69 cells. Cell cycle analysis by fluorescence-activated cell sorting (FACS) after cell staining with propidium iodide (PI) demonstrated an increase in cells in the sub-G1 phase, which is an indicator of apoptosis (Figure 2c). To complement these studies, the activation of caspase-3 and poly (ADP-ribose) polymerase (PARP) cleavage were investigated by Western blotting. H-69 cells treated with simvastatin displayed a marked activation of capase-3 and PARP cleavage (Figure 2d), demonstrating the induction of apoptosis in these cells. Together these data show that blocking Ras signalling with simvastatin impairs cell growth and induces apoptosis in human SCLC cells.

Simvastatin sensitizes human SCLC cells to chemotherapeutic agents and impairs tumour growth in vivo

In view of the ability of simvastatin to block SCLC cell proliferation and induce apoptosis, we next investigated whether the drug could enhance the effect of the chemotherapeutic agent etoposide in these cells. H-69 and H-510 cells were treated with different concentrations of etoposide in the presence of increasing concentrations of simvastatin. In both cell lines, simvastatin treatment markedly enhanced the antiproliferative effects of etoposide (Figure 3a, b).

Figure 3
figure3

Simvastatin sensitizes SCLC cells to the effects of etoposide and inhibits SCLC xenograft growth in vivo. (a and b) H-69 (a) or H-510 (b) cells were incubated in serum-free medium with increasing concentrations of simvastatin alone (diamonds), or in combination with etoposide at a concentration of 1.0 μ M (triangles), 10 μ M (squares) or 100 μ M (circles). Cell proliferation was assessed using the MTS assay after 5 days. The data are mean from eight independent experiments. (c) Xenografts of H-69 SCLC cells were grown subcutaneously in nu/nu mice. Simvastatin (50 mg/kg, triangles) or placebo (squares) was administered by oral gavage daily and tumour growth monitored over a 3-week period. Tumour volume was determined at the times indicated. Data are mean from seven animals in each group.

To investigate the potential ability of simvastatin to inhibit SCLC tumour growth in vivo, a xenograft of H-69 SCLC cells was established by subcutaneous injection in nu/nu mice. The mice were then given placebo or simvastatin (50 mg/kg) daily by oral gavage over a 3-week period. The xenografts of placebo-treated animals displayed a doubling in tumour volume after 3 weeks, while no increase was observed in mice treated with simvastatin (Figure 3c). Moreover, simvastatin induced a transient decrease in tumour volume between 10 and 15 days of treatment (Figure 3c).

Simvastatin impairs activation of signalling cascades downstream of c-Kit

To further elucidate the molecular mechanism of SCLC cell growth inhibition by simvastatin, the impact of the drug on early signalling events stimulated by SCF was investigated in H-69, H-209 and H-510 cells. Pretreatment of SCLC cells with simvastatin (3 μg/ml, 24 h) in serum-free medium completely abrogated SCF-stimulated activation of Erk1/2 in the three SCLC cell lines under study (Figure 4a–c). The drug also markedly impaired activation of PKB and ribosomal S6K (Figure 4a–c), which are targets of the PI3K pathway in these cells (Arcaro et al., 2002). The inhibitory phosphorylation of glycogen synthase kinase-3β (GSK-3β) induced by SCF was also blocked by simvastatin (Figure 4b). Similar results were obtained when PKB and Erk1/2 activation by other growth factors was studied in SCLC cells. Simvastatin inhibited activation of PKB and Erk1/2 by hepatocyte growth factor (HGF), FGF-2 and phorbol esters in the three cell lines under study (data not shown). The effect of simvastatin on SCF-stimulated Erk1/2 activation in H-69 cells was dose-dependent (Figure 4d), with a maximal effect at 3 μg/ml and was apparent after 24 h of pretreatment (Figure 4e). Together these results show that simvastatin blocks several signalling pathways implicated in SCLC cell growth, such as MEK/Erk and PI3K/PKB/S6K.

Figure 4
figure4

Impact of simvastatin on downstream targets of SCF/c-Kit signalling. (ac) H-69 (a), H-209 (b) or H-510 (c) SCLC cells were pretreated with or without simvastatin (3.3 μg/ml) for 24 h at 37°C in serum-free medium and stimulated with SCF (10 ng/ml) for 5 min at 37°C. The cells were lysed and samples analysed by SDS–PAGE and Western blotting with antibodies against activated Erk1/2 (p-Erk1/2) or total Erk1/2, activated PKB (p-PKB) or total PKB, Ser 9-phosphoralyed GSK-3β (p-GSK-3) or total GSK-3, or Ser 240/244-phosphorylated ribosomal S6 protein (p-S6) (where indicated). (d) H-69 SCLC cells were pretreated with increasing concentrations of simvastatin as above and stimulated with SCF (10 ng/ml) for 5 min at 37°C. Activation of Erk1/2 was analysed as in (ac). (e) H-69 SCLC cells were pretreated with simvastatin (3.3 μg/ml) for the times indicated and stimulated with SCF (10 ng/ml) for 5 min at 37°C. Activation of Erk1/2 was analysed as in (ac).

SCLC cells expressing constitutively active MEK are resistant to the effects of simvastatin

The above results indicated that inhibition of Ras signalling to MEK/Erk contributes to the ability of simvastatin to block SCLC cell growth. To further substantiate this hypothesis, H-69 cells stably expressing an activated mutant of MEK (Pardo et al., 2002) were compared to untransfected H-69 cells for their sensitivity to simvastatin in serum-free medium. Basal Erk1/2 activation was higher in the H-69/MEK cells than in control H-69 SCLC cells (Figure 5a). Simvastatin did not affect the increase in basal Erk1/2 activity induced by activated MEK, although it did promote H-Ras degradation in these cells (Figure 5b and data not shown). In contrast to H-69 cells, H-69/MEK cells grew normally in liquid culture in the presence of simvastatin (Figure 5c), demonstrating that the MEK/Erk pathway is both necessary and sufficient to sustain basal H-69 cell growth in liquid culture.

Figure 5
figure5

An activated mutant of MEK restores H-69 SCLC cell growth in the presence of simvastatin. (a) H-69 cells or H-69 cells stably expressing activated MEK1 were lysed and samples analysed by SDS–PAGE and Western blotting with antibodies against activated Erk1/2 (p-Erk1/2) or total Erk1/2. (b) H-69 cells or H-69 cells stably expressing activated MEK1 cells were pretreated with or without simvastatin (3.3 μg/ml) for 24 h at 37°C and stimulated with SCF (10 ng/ml) for 5 min at 37°C. The cells were lysed and samples analysed by SDS–PAGE and Western blotting with antibodies against activated Erk1/2 (p-Erk1/2) or actin. (c) H-69 cells or H-69 cells stably expressing activated MEK1 were pretreated with or without simvastatin as above in serum-free medium in the presence or absence of SCF (10 ng/ml). Cell counting was performed at 0 days (dark grey bars) and after 8 days (light gray bars). The data are mean with s.d. from four independent experiments.

Simvastatin does not impair c-Kit activation or its lipid raft association

In order to confirm the model that simvastatin blocks the activation of various signalling cascades downstream of c-Kit by targeting Ras isoforms, we first tested the impact of the drug on receptor activation. Comparable amounts of c-Kit were present in anti-phosphotyrosine immunoprecipitates from control or simvastatin-treated H-69 cells after addition of SCF (Figure 6a), demonstrating that the drug does not directly inhibit receptor activation in these cells.

Figure 6
figure6

Simvastatin does not block c-Kit autophosphorylation or localization to lipid rafts. (a) H-69 SCLC cells were pretreated with or without simvastatin (3.3 μg/ml) for 24 h at 37°C and stimulated with SCF (10 ng/ml) for 5 min at 37°C. The cells were lysed and samples immunoprecipitated with anti-phosphotyrosine (pY) antibody, analysed by SDS–PAGE and Western blotting with anti-c-Kit antibody. (b) H-69 SCLC cells were pretreated with or without simvastatin as in (a). The cells were lysed in Triton X-100 and fractionated on discontinuous sucrose gradients. Fractions from the gradients were analysed by SDS–PAGE and Western blotting for the proteins indicated. The DIM fraction (2–4) contains lipid raft-associated molecules and the M fraction (6–9) detergent-soluble cellular components.

Since simvastatin interferes with cholesterol biosynthesis, it may disrupt lipid rafts by lowering cellular cholesterol levels, which in turn could impair signalling by SCF/c-Kit. To investigate this possible effect of the statin, H-69 cells were lysed in Triton X-100 and fractionated on discontinuous sucrose gradients to isolate the detergent-insoluble membrane (DIM) fraction that contains lipid rafts (Arcaro et al., submitted). Lipid rafts isolated from H-69 cells were enriched in Src and contained a fraction of total cellular c-Kit and the p85α regulatory subunit of PI3K (Figure 6b and data not shown). The association of these proteins with lipid rafts was not impaired upon simvastatin pretreatment (Figure 6b), but intriguingly, H-Ras was completely absent from the DIM fraction. Total cellular levels of H-Ras were also lower in simvastatin-treated SCLC cells (Figure 6b).

The impact of simvastatin on Ras isoform localization to the plasma membrane was then investigated. SCLC cells were lysed in a hypotonic buffer and fractionated into the plasma membrane, low-density microsomes and cytosol. The receptor c-Kit was mainly localized at the plasma membrane, while caspase-3 was cytosolic (Figure 7a). As expected, all three Ras isoforms were mainly localized at the plasma membrane, but N-Ras was also present to some extent in the low-density microsomes and cytosol (Figure 7a). Simvastatin treatment severely impaired association of H-, K- and N-Ras with the plasma membrane, but did not significantly affect Src localization (Figure 7b). Simvastatin induced the redistribution of the K- and N-Ras isoforms to the low-density microsomes (Supplementary information). Cholesterol replenishment of simvastatin-treated cells differentially affected association of Ras isoforms with the plasma membrane. While H-Ras association with the plasma membrane was not reversed, association of N-Ras was almost completely restored upon cholesterol treatment (Figure 7c). Cholesterol also induced reassociation of a minor fraction of K-Ras in simvastatin-treated cells (Figure 7c).

Figure 7
figure7

Simvastatin treatment impairs Ras localization to the plasma membrane and selectively downregulates cellular H-Ras protein levels in SCLC cells. (a) H-69 SCLC cells were fractionated into the plasma membrane (PM), low-density microsomes (LDM) and cytosol (CYT). The fractions were analysed by Western blot for the proteins indicated. (bc) H-69 SCLC cells were pretreated with or without simvastatin, in the presence or absence of cholesterol (0.6 mg/ml) (c) and fractionated as in (a). The fractions were analysed by Western blot for the proteins indicated. (d) H-69 SCLC cells were pretreated with or without simvastatin (3.3 μg/ml) for 24 h at 37°C. The cells were lysed and samples analysed by SDS–PAGE and Western blotting with antibodies against the proteins indicated. (e) H-69 SCLC cells were pretreated with or without simvastatin as in (d). Total mRNA was extracted as described in experimental procedures and real-time RT–PCR for H-Ras was performed on a LightCycler.

H-Ras protein levels are selectively downregulated by simvastatin

We then investigated whether the ability of simvastatin to decrease total H-Ras protein levels was specific and involved effects on gene transcription. Under conditions where H-Ras protein levels were decreased by over 50% in H-69 cells treated with the drug, no changes in the levels of K-Ras, N-Ras, RhoA, Rac1 or actin were observed (Figure 7d).

To investigate whether the simvastatin-induced downregulation of H-Ras was caused by an effect on gene transcription, levels of H-Ras mRNA were quantified by quantitative RT–PCR. The levels of H-Ras mRNA were unaffected by simvastatin, as assessed by quantitative RT–PCR (Figure 7e), indicating that the drug acts post-transcriptionally by inhibiting H-Ras mRNA translation, or enhancing protein degradation.

Simvastatin selectively inhibits PKB activation and cell growth in NSCLC cells

To investigate whether simvastatin inhibits NSCLC cell proliferation through the same mechanisms as in SCLC cells, the impact of the inhibitor on A549 cell responses was studied. Simvastatin significantly impaired growth of A549 cells with an IC50 of approximately 4.5 μg/ml (Supplementary information). Various growth factors such as epidermal growth factor (EGF), FGF-2 and SCF were then tested for their ability to activate PKB and Erk1/2 in A549 cells. EGF was found to be the most potent activator of both pathways in this NSCLC cell line (data not shown). Simvastatin (3.0 μg/ml) significantly inhibited EGF-stimulated PKB activation (Supplementary information), but did not affect Erk1/2 activation. Moreover, basal Erk1/2 activation was enhanced by simvastatin treatment in A549 cells (Supplementary information). The drug also very significantly decreased total H-Ras protein levels in A549 cells, confirming the data obtained in SCLC cells (Supplementary information).

Specific downregulation of Ras isoforms by RNAi differentially affects SCLC cell responses to polypeptide growth factors

The inhibitory effect of simvastatin on growth factor-stimulated activation of MEK/Erk and PKB suggested the involvement of Ras family proteins in both signalling pathways. To gain further insight into the specific functions of individual Ras isoforms in the control of signalling cascades downstream of polypeptide growth factor receptors, we developed an RNAi approach to selectively downregulate cellular H-, K- or N-Ras protein levels. The three RNAi constructs were first tested for their ability to selectively inhibit expression of Ras isoforms, by transient cotransfections with expression plasmids encoding H-, K- or N-Ras into HEK 293 cells (data not shown).

The Ras RNAi constructs were then stably expressed in H-510 SCLC cells using recombinant retroviruses. Downregulation of the protein expression of each individual Ras isoforms by the corresponding RNAi was confirmed in these cells (Figure 8a). The impact of the individual Ras RNAi constructs on early signalling events was then investigated. No difference in Erk1/2 activation in response to polypeptide growth factor (FGF-2) or phorbol dibutyrate (PDB) stimulation was observed between the control and Ras RNAi-transfected cells (Figure 8b). However, PKB activation induced by PDB was significantly impaired by the N-Ras RNAi (Figure 8b). Moreover, the H-Ras RNAi significantly increased basal PKB activity. To confirm these data in another cell line, we transiently transfected human SW2 SCLC cells with the various Ras RNAi constructs. The choice of this cell line was justified by the superior transfection efficiency achievable, in comparison to other human SCLC cell lines. In these cells, SCF-stimulated activation of PKB and Erk1/2 was significantly inhibited by the N-Ras RNAi (Figure 8c). The H- and K-Ras RNAi constructs failed to cause any significant effect on PKB and Erk1/2 activation by SCF, when transiently transfected into these SCLC cells (data not shown).

Figure 8
figure8

An RNAi targeting N-Ras impairs PKB activation and cell growth in SCLC cells. (a) H-510 SCLC cells were stably transduced with RNAi constructs targeting H-Ras, K-Ras or N-Ras or with empty vector. Total cell lysates were analysed by SDS–PAGE and Western blot analysis for individual Ras isoforms or lamin B as a loading control. P: parental (untransfected) cell line. V: vector-transfected cells. R: RNAi-transfected cells. (b) H-510 cells stably expressing the vector (V), RNAi against H-Ras (H), K-Ras (K) or N-Ras (N) were stimulated with FGF-2 (0.1 ng/ml) or phorbol dibutyrate (PDB, 400 nM) for 5 min at 37°C. The cells were lysed and samples analysed by SDS–PAGE and Western blotting with antibodies against the proteins indicated. (c) SW2 SCLC cells were transiently transfected with vector or RNAi against N-Ras. The cells were then stimulated with SCF (10 ng/ml) for 5 min at 37°C, lysed and samples analysed as in (b). (d) H-510 cells stably expressing the vector (closed circles), RNAi against H-Ras (open circles), K-Ras (triangles), or N-Ras (crosses) were grown in serum-free or serum-containing (10% FCS) medium where indicated. Cell numbers were analysed by counting after the times indicated. The data are mean with s.d. from four independent experiments. (e) H-510 cells stably expressing the vector (V), RNAi against H-Ras (H) or K-Ras (K) were stained with PI and the percentage of cells in S phase determined by FACS analysis. The data are mean with s.d. from four independent experiments.

The growth properties of the various Ras RNAi-transfected H-510 cell lines were compared in serum-free medium and medium supplemented with 10% FCS. The N-Ras RNAi-transfected H-510 cell line failed to grow either in the presence or absence of serum, while the K-Ras RNAi-transfected cells grew in the presence but not in the absence of serum (Figure 8d). Surprisingly, the H-Ras RNAi stimulated the growth of H-510 cells, as compared to the vector-transfected cells, independently of the presence of serum (Figure 8d). The effect of the H-Ras RNAi on cell growth was caused by increased DNA synthesis in the H-Ras RNAi-expressing cells (Figure 8e), which correlated with the enhanced PKB activation of these cells (Figure 8b).

We next investigated whether H-Ras RNAi-expressing cells remained sensitive to simvastatin. Indeed, the drug inhibited proliferation of the H-Ras RNAi-expressing cells slightly more potently than vector-transfected cells (IC50 1 vs 3 μ M) (Figure 9).

Figure 9
figure9

SCLC cells expressing H-Ras RNAi remain sensitive to simvastatin. H-510 SCLC cells stably expressing the RNAi construct targeting H-Ras (triangles) or cells transduced with empty vector (diamonds) were incubated with increasing concentrations of simvastatin. Cell proliferation was measured after 72 h. Data are mean with s.d. from eight experiments.

Discussion

The potential of statins as anticancer agents has recently emerged because of their antiproliferative, proapoptotic and antimetastatic effects on a variety of cancer cell lines (Chan et al., 2003; Graaf et al., 2004b). Their exact mechanism of action, however, remains unclear since inhibition of the mevalonate pathway can have effects on membrane integrity, cell signalling, protein synthesis and cell cycle progression (Chan et al., 2003; Graaf et al., 2004b). One of the main targets of statins in the context of intracellular signalling was shown to be RhoA (Hirai et al., 1997; Lee et al., 1998), which could explain the effect of these drugs on tumour cell cycle and motility.

Our data demonstrate that simvastatin is a potent inhibitor of SCLC cell growth in vitro and in vivo, and that the drug targets both the MEK/Erk and PI3K/PKB pathways downstream of polypeptide growth factor receptors such as c-Kit. Indeed, constitutive activation of the MEK/Erk pathway reversed the effects of simvastatin on SCLC cell growth and pharmacological inhibition of MEK inhibited SCLC cell proliferation, as previously described (Pardo et al., 2001). Moreover, cells with enhanced PKB activation (H-69) were more resistant to the effects of simvastatin. This model is supported by the findings that PI3K inhibitors or dominant-negative mutants of PI3Ks or PKB block SCLC cell growth (Moore et al., 1998; Arcaro et al., 2002; Krystal et al., 2002). The MEK/Erk and PI3K pathways have previously been reported to be targeted by statins in human tumour cells, but the evidence is somewhat conflicting (Graaf et al., 2004b). Moreover, farnesyl transferase inhibitors (FTIs) have also been shown to target the PI3K/PKB pathway (Jiang et al., 2000) and a recent report showed that a constitutively active mutant of PKB reversed the effect of the FTI SCH66336 on squamous carcinoma cell growth (Chun et al., 2003). The ability of simvastatin to block activation of both the PI3K/PKB and MEK/Erk pathway was specific for SCLC cells, since the drug affected only PKB activation in NSCLC cells. This difference may be caused by the fact that A549 cells express an activated mutant of K-Ras, in contrast to SCLC cell lines. However, the sensitivities of the growth responses of SCLC and NSCLC to simvastatin were not significantly different, indicating that A549 cells may be more dependent on PKB for growth.

Since statins block cholesterol biosynthesis, one potential explanation for the effects of simvastatin on intracellular signalling events downstream of c-Kit is that the drug affects lipid raft integrity by lowering intracellular cholesterol levels. Indeed, lipid rafts were recently shown to play a critical role in the activation of PI3K/PKB by growth factors in SCLC cells (Arcaro et al., submitted). However, lipid raft association of c-Kit and other raft-associated molecules such as Src was unaffected by simvastatin in SCLC cells, indicating that lipid microdomains are not the primary target of the drug in SCLC cells. Moreover, activation of MEK/Erk by SCF was independent of lipid rafts (Arcaro et al., submitted), indicating that simvastatin affects Ras superfamily activation, leading to effects on both MEK/Erk and PI3K. Since simvastatin targets prenylation and membrane association of all Ras family GTPases, and SCLC cells express H-, K- and N-Ras, the effects of the drug on activation of intracellular signalling by SCF/c-Kit could result from inhibition of all three Ras isoforms. Indeed, simvastatin impaired association of all three Ras isoforms with the plasma membrane. In simvastatin-treated cells, K- and N-Ras associated with the low-density microsome fraction, which was possibly mediated by the alternative localization signals present in Ras family proteins, such as palmitoylation in the case of N-Ras, or the polybasic sequence in the case of K-Ras. Cholesterol replenishment of simvastatin-treated cells induced the reassociation of N-Ras with the plasma membrane, but a similar response was not observed with H- and K-Ras. Together these data show that membrane localization of Ras isoforms is controlled by different molecular mechanisms and that the effects of simvastatin cannot be accounted for only by lowering of cellular cholesterol. Intriguingly, H-Ras protein levels were markedly reduced by simvastatin treatment of H-69 SCLC and A549 cells, indicating that lipid modification of H-Ras controls its stability. Simvastatin has been reported to inhibit malignant transformation by the Ha-ras oncogene (Furst et al., 2002), which correlates with the findings presented here. It is conceivable that H-Ras downregulation reflects decreased mRNA translation or increased protein degradation, since there were no changes in H-Ras mRNA levels. Cellular transformation by H-Ras has been reported to be inhibited by expression of caspase-2 (Hiwasa and Nakagawara, 1998), which may indicate that H-Ras downregulation by simvastatin in SCLC cells involves caspase-dependent degradation of H-Ras. Together these data suggest that differences in subcellular localization of individual Ras isoforms may control their turnover.

The present report also reveals surprising differences in the contributions of individual Ras isoforms to the activation of MEK/Erk and PI3K/PKB. Indeed, N-Ras appears to be the major Ras isoform involved in the control of the PI3K/PKB pathway in SCLC cells, while the three Ras isoforms probably have overlapping roles in the activation of MEK/Erk signalling by FGF-2. Moreover, downregulation of N-Ras by RNAi in SCLC cells blocked cell growth, in accordance with the critical role of PI3K/PKB signalling in this response (Arcaro et al., 2002; Krystal et al., 2002). N-Ras has previously been reported to selectively control PKB activity and cell survival (Wolfman and Wolfman, 2000), which is confirmed by the findings presented here. This model is further supported by the observation that mutations in N-Ras and PTEN are functionally overlapping in human melanoma (Tsao et al., 2000). Interestingly, H-Ras downregulation by RNAi did not have the same impact on SCLC cell responses, as simvastatin treatment, although the drug also significantly decreased H-Ras protein levels. This is most likely caused by the fact that simvastatin also impairs K- and N-Ras function by inhibiting their membrane localization. The observation that RNAi-mediated downregulation of H-Ras promotes SCLC cell growth and PKB activation may reflect an indirect activation of N-Ras, if both Ras isoforms can be activated by the same guanyl nucleotide exchange factors in SCLC cells. In support of this notion, H-Ras RNAi-expressing cells remained fully sensitive to simvastatin treatment.

A recent report showed that FTIs have no significant activity as single agents in relapsed SCLC (Johnson and Heymach, 2004). However, statins may represent more potent inhibitors of tumour cell proliferation. Recent epidemiologic studies have revealed that statin use reduced the risk of cancer incidence, when different human cancers were studied, including lung cancer (Graaf et al., 2004a). Our data demonstrate that simvastatin may represent a potential new drug to inhibit SCLC proliferation and sensitize tumour cells to chemotherapy.

Materials and methods

Antibodies and reagents

Antibodies against actin, c-Kit, Eps8, GSK-3β, Lamin B, PARP, phosphotyrosine, PKB, Rac1, H-, K-, N-Ras, RhoA, SOS-1, c-Src were obtained from Santa Cruz Biotechnology. Antibodies against active (cleaved) caspase-3, activated PKB (Ser473), Ser240/244-phosphorylated protein S6, Ser9-phosphorylated GSK-3β were from Cell Signalling Technology. Antibodies against activated Erk1/2 and water-soluble cholesterol were from Sigma Aldrich. EGF, SCF, simvastatin and the caspase inhibitor (Z-VAD) were from Calbiochem.

Cell culture

Human SCLC lines were cultured as described (Pardo et al., 2002). For experimental purposes the SCLC cells were diluted into serum-free medium (Pardo et al., 2002) and grown for 3–5 days. A549 cells were grown in DMEM (Sigma Aldrich) with 10% (v/v) heat-inactivated foetal calf serum (HIFCS, Life Technologies) and penicillin/streptomycin/L-glutamine (Sigma Aldrich), and passaged every 3 days by trypsinization. For serum-starving, the cells were incubated for 16 h in DMEM containing 0.5% (v/v) HIFCS.

Cell proliferation and apoptosis

SCLC cells (2 × 105/ml) were grown for 5 days in serum-containing medium in the presence or absence of simvastatin. Cell proliferation was analysed by MTS assay using the CellTiter 96® Aqueous One Solution Cell Proliferation Assay (Promega). Alternatively, cell counting experiments were performed. SCLC cells grown in RPMI containing 10% FCS were washed 4 times in RPMI and resuspended either in serum-containing medium or in SFM. Cells were then aliquoted in 24-well Falcon plates at a density of 1.5 × 104 cells/ml of corresponding medium. After the relevant incubation period, cell clumps within the cell suspension were disaggregated by passing the well content 5 times through a 19-gauge needle and cell number determined using a haemocytometer. Each condition was performed in quadruplicate and results obtained from averaging the values of three counts per replicate.

For detection of apoptosis, SCLC cells (2 × 106/ml) were incubated for 24 h in the presence or absence of simvastatin. The cells were then lysed and samples analysed by SDS–PAGE and Western blot with anti-poly (ADP-ribose) polymerase (PARP) or anti-caspase-3 (active fragment) antibodies.

In vivo experiments

The effects of simvastatin on growth of established H-69 tumours in vivo were assessed in adult BALB/c nu/nu mice (Cancer Research-UK BSU, Clare Hall, South Mimms, UK). The xenograft was established by injecting 107 H-69 SCLC cells in 0.2 ml PBS subcutaneoulsy into one flank and subsequently passaged. Simvastatin (50 mg/kg) or placebo were administered daily by oral gavage (seven animals per group). Tumour growth was monitored over a 3-week period.

Subcellular fractionation and isolation of lipid rafts

SCLC cells (108) were lysed in 400 μl of MNE buffer containing 1% Triton X-100, and ultracentrifugation on discontinuous sucrose gradients performed as described (Arcaro et al., 2000) to isolate lipid rafts. Fractionation of SCLC cell into the plama membrane, low-density microsomes and cytosol was performed as previously reported (Arcaro et al., 1998).

RNA extraction and RT–PCR

H69 cells (4 × 106) grown in serum-free medium were treated with or without 3.3 μg/ml simvastatin for 24 h. The cell pellet was resuspended in ice-cold RNA lysis buffer (150 mM NaCl, 10 mM Tris pH 7.4, 1 mM MgCl2 and 0.5% NP40) and incubated on ice for 5 min. After centrifugation, Tris-buffered phenol and 10% SDS was added to the supernatant and centrifuged at room temperature for 5 min. The aqueous phase was then transferred to Tris-buffered phenol (pH 7.4) and briefly centrifuged. The aqueous phase was then mixed with 2 M Na acetate (pH 5.2) and 1 ml absolute ethanol and allowed to precipitate at –20°C overnight. The RNA was pelleted down and washed in 70% ethanol, vacuumed dry and resuspended in sterile water. cDNA was synthesized from 2 μg of extracted RNA. The RNA was heated to 65°C for 10 min and cDNA synthesis was carried out in 21 μl of a cDNA mix containing 1.9 × Invitrogen MMLV buffer, 1.9 mM DTT, 1.9 mM 4 × dNTPs, 200 ng/μl random primers, 21 μl 0.5 M KCl and 372 μl dH2O. The final volume of the cDNA synthesis reaction was made to 40 μl with dH2O. The mixture was heated to 37°C for 1 h, heated again to 65°C for 10 min and then cooled to 4°C. cDNA (2 μl) was used for each PCR on LightCycler™ real-time PCR machine using 2 μl of the SYBR1 Green fast start PCR kit (Roche Diagnostics). The reaction contained 12.5 mM Tris pH 8.3, MgCl2 62.5 mM, KCl 0.25 mM each of dATP, dCTP, dTTP and dGTP and 30 units/ml Taq ploymerase) plus 0.5 μ M of each H-Ras primer (IndexTerm5′-ATGACGGAATATAAGCTGGTG-3′ and IndexTerm5′-TCAGGAGAGCACACACTTCAG-3′) in a total reaction volume of 20 μl. The PCR conditions used were followed in accordance to the manufacturer's instructions and keratin mRNA was used as an internal standard.

Immunoprecipitations, SDS–PAGE and Western blotting

These assays were performed exactly as described (Arcaro et al., 1998).

Transient expression in SCLC cells

Human SW2 SCLC cells were transiently transfected with Ras RNAi constructs using Lipofectamine 2000 (Invitrogen). Cell responses were assessed 72 h post-transfection.

Ras RNAi vectors and retrovirus production

The sequences used were as follows:

for N-Ras, IndexTermGATCCCCGACTCGGATGATGTACCTATTCAAGAGATAGGTACATCATCCGAGTCTTTTTGGAAA (forward) and IndexTermAGCTTTTCCAAAAAGACTCGGATGATGTACCTATCTCTTGAATAGGTACATCATCCGAGTCGGG (reverse); for H-Ras, IndexTermGATCCCCTCTCGGCAGGCTCAGGACCTTCAAGAGAGGTCCTGAGCCTGCCGAGATTTTTGGAAA (forward) and IndexTermAGCTTTTCCAAAAATCTCGGCAGGCTCAGGACCTCTCTTGAAGGTCCTGAGCCTGCCGAGAGGG (reverse); for K-Ras, IndexTermGATCCCCGAGTTAAGGACTCTGAAGATTCAAGAGATCTTCAGAGTCCTTAACTCTTTTTGGAAA (forward)b and IndexTermAGCTTTTCCAAAAAGAGTTAAGGACTCTGAAGATCTCTTGAATCTTCAGAGTCCTTAACTCGGG (reverse).

Forward and reverse sequences were annealed in 100 mM potassium acetate, 30 mM HEPES–KOH pH 7.4, 2 mM Mg-acetate by incubating oligos for 4 min at 95°C, then 10 min at 70°C followed by slow cooling to 4°C. The annealed oligonucleotides were phosphorylated and cloned between the BglII and HindIII sites of the pRETRO-SUPER vector. Successfully cloned vectors were transfected into Phoenix packaging cells using Lipofectamine Plus (Invitrogen). Virus-containing medium was collected 24 and 48 h post-transfection, filtered, mixed with 8 μg/ml Polybrene and applied overnight on H-510 SCLC cells stably expressing the mouse ecotropic receptor (Pardo et al., 2002). Infected H-510 cells were then selected in the presence of 2 μg/ml for 14 days prior to experiments.

References

  1. Arcaro A, Gregoire C, Boucheron N, Stotz S, Palmer E, Malissen B et al. (2000). J Immunol 165: 2068–2076.

  2. Arcaro A, Khanzada UK, Vanhaesebroeck B, Tetley TD, Waterfield MD, Seckl MJ . (2002). EMBO J 21: 5097–5108.

  3. Arcaro A, Volinia S, Zvelebil MJ, Stein R, Gout I, Layton MJ et al. (1998). J Biol Chem 273: 33082–33090.

  4. Chan KKW, Oza AM, Siu LL . (2003). Clin Cancer Res 9: 10–19.

  5. Chun K-H, Lee H-Y, Hassan K, Khuri F, Hong WK, Lotan R . (2003). Cancer Res 63: 4796–4800.

  6. Furst J, Haller T, Chwatal S, Woll E, Dartsch PC, Gschwentner M et al. (2002). Cell Physiol Biochem 12: 19–30.

  7. Graaf MR, Beiderbeck AB, Egberts ACG, Richel DJ, Guchelaar H-J . (2004a). J Clin Oncol 22: 2388–2394.

  8. Graaf MR, Richel DJ, van Noorden CJF, Guchelaar H-J . (2004b). Cancer Treat Rev 30: 609–641.

  9. Heasley LE . (2001). Oncogene 20: 1563–1569.

  10. Hirai A, Nakamura S, Noguchi Y, Yasuda T, Kitagawa M, Tatsuno I et al. (1997). J Biol Chem 272: 13–16.

  11. Hiwasa T, Nakagawara A . (1998). Biochem Biophys Res Commun 250: 741–746.

  12. Jiang K, Coppola D, Crespo NC, Nicosia SV, Hamilton AD, Sebti SM et al. (2000). Mol Cell Biol 20: 139–148.

  13. Johnson BE, Heymach JV . (2004). Clin Cancer Res 10: 4254s–4257s.

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

  15. Kusama T, Mukai M, Iwasaki T, Tatsuta M, Matsumoto Y, Akedo H et al. (2001). Cancer Res 61: 4885–4891.

  16. Lee SJ, Ha MJ, Lee J, Nguyen P, Choi YH, Pirnia F et al. (1998). J Biol Chem 273: 10618–10623.

  17. Mitsudomi T, Viallet J, Mulshine JL, Linnoila RI, Minna JD, Gazdar AF . (1991). Oncogene 6: 1353–1362.

  18. Moore SM, Rintoul RC, Walker TR, Chilvers ER, Haslett C, Sethi T . (1998). Cancer Res 58: 5239–5247.

  19. Pardo OE, Arcaro A, Salerno G, Raguz S, Downward J, Seckl MJ . (2002). J Biol Chem 277: 12040–12046.

  20. Pardo OE, Arcaro A, Salerno G, Tetley TD, Valovka T, Gout I et al. (2001). Oncogene 20: 7658–7667.

  21. Salgia R, Skarin AT . (1998). J Clin Oncol 16: 1207–1217.

  22. Tsao H, Zhang X, Fowlkes K, Haluska FG . (2000). Cancer Res 60: 1800–1804.

  23. Wistuba II, Gazdar AF, Minna JD . (2001). Semin Oncol 28: 3–13.

  24. Wolfman JC, Wolfman A . (2000). J Biol Chem 275: 19315–19323.

Download references

Acknowledgements

We thank Dr R Marais and Dr U Zangemeister-Wittke for reagents. This work was supported by a grant from the Association for International Cancer Research to AA and a Program Grant from Cancer Research UK to MJS.

Author information

Correspondence to A Arcaro.

Additional information

Supplementary Information accompanies the paper on Oncogene website (http://www.nature.com/onc)

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Khanzada, U., Pardo, O., Meier, C. et al. Potent inhibition of small-cell lung cancer cell growth by simvastatin reveals selective functions of Ras isoforms in growth factor signalling. Oncogene 25, 877–887 (2006). https://doi.org/10.1038/sj.onc.1209117

Download citation

Keywords

  • lung cancer
  • statins
  • ras
  • protein kinase B
  • receptor tyrosine kinase

Further reading