The hypoxia-inducible factor 1 (HIF-1) plays a critical role for tumour adaptation to microenvironmental hypoxia, and represents an appealing chemotherapeutic target. Silibinin is a nontoxic flavonoid reported to exhibit anticancer properties. However, the mechanisms by which silibinin inhibits tumour growth are not fully understood. In this study, silibinin was found to inhibit hypoxia-induced HIF-1α accumulation and HIF-1 transcriptional activity in human cervical (HeLa) and hepatoma (Hep3B) cells. Neither HIF-1α protein degradation rate nor HIF-1α steady-state mRNA level was affected by silibinin. Rather, we found that suppression of HIF-1α accumulation by silibinin correlated with strong dephosphorylation of mammalian target of rapamycin (mTOR) and its effectors ribosomal protein S6 kinase (p70S6K) and eukaryotic initiation factor 4E-binding protein-1 (4E-BP1), a pathway known to regulate HIF-1α expression at the translational level. Silibinin also activated Akt, a mechanistic feature exhibited by established mTOR inhibitors in many tumour cells. Moreover, silibinin reduced hypoxia-induced vascular endothelial growth factor (VEGF) release by HeLa and Hep3B cells, and this effect was potentiated by the PI3K/Akt inhibitor LY294002. Finally, silibinin was found to be a potent inhibitor of cell proliferation. These results show that silibinin is an effective inhibitor of HIF-1 and provide new perspectives into the mechanism of its anticancer activity.
Hypoxia is a common feature of most solid tumours (Brown and Giaccia, 1998). As tumour cells proliferate, the oxygen and nutrient demand increases to the point at which the diffusion of oxygen from the blood vessels becomes limiting, resulting in the development of hypoxia. Cancer cells adapt to this hypoxic environment through the activation of a number of cellular pathways that stimulate glycolysis, proliferation, the upregulation of survival factors and angiogenesis, providing the tumour with adequate energy and blood supply to allow for continued growth (Brown and Giaccia, 1998; Hockel and Vaupel, 2001).
As the master regulator of the hypoxic transcriptional response, the hypoxia-inducible factor 1 (HIF-1) plays a central role in tumour growth and angiogenesis (Semenza, 2002). HIF-1 is a heterodimeric transcription factor consisting of subunits HIF-1α and HIF-1β (Wang and Semenza, 1995). HIF-1β is constitutively expressed and its levels are unaffected by changes in cellular pO2. In contrast, HIF-1α is tightly regulated by O2 tension and accumulates very rapidly in cells exposed to hypoxia (Wang et al., 1995; Jiang et al., 1996). Under non-hypoxic conditions, HIF-1α is rapidly degraded by the ubiquitin–proteasome system after the hydroxylation of proline residues 402 and 564 by a family of O2-dependent prolyl 4-hydroxylases, and subsequent binding of von Hippel–Lindau protein (Schofield and Ratcliffe, 2004). Under hypoxic conditions, prolyl hydroxylation of HIF-1α is impaired, leading to decreased von Hippel–Lindau protein ubiquitination and increased HIF-1α stability (Schofield and Ratcliffe, 2004). Although the inhibition of prolyl hydroxylases in hypoxia is the primary mechanism of HIF-1α accumulation, HIF-1α expression also depends on its rate of de novo synthesis. An increasing body of evidence indicates that some growth factors, cytokines and the induction of certain oncogenes can stimulate HIF-1α protein synthesis through the activation of the phosphatidylinositol-3-kinase (PI3K)/Akt/mammalian target of rapamycin (mTOR) pathway (Zhong et al., 2000; Laughner et al., 2001; Hudson et al., 2002; Xu et al., 2005). Indeed, recent data indicate that mTOR signalling is necessary for HIF-1α expression irrespective of oxygenation conditions (Thomas et al., 2006).
Silibinin, a polyphenolic flavonoid isolated from milk thistle (Silybum marianum), is used clinically to treat certain liver diseases (Saller et al., 2001), and its consumption is safe and nontoxic, even at high doses, in animals and humans (Wellington and Jarvis, 2001). Several studies have shown strong anticancer and chemopreventive efficacy of silibinin in diverse human cancer models (Zi et al., 1998; Bhatia et al., 1999; Zi and Agarwal, 1999; Varghese et al., 2005; Singh et al., 2008). Silibinin has been shown to potently suppress the proliferation of tumour (Zi et al., 1998; Bhatia et al., 1999; Zi and Agarwal, 1999; Varghese et al., 2005; Singh et al., 2008) and endothelial cells (Singh et al., 2005), and has exhibited promising anti-angiogenic properties in a variety of model systems (Singh et al., 2005, 2008; Gu et al., 2007). However, the rather pleiotropic effects of silibinin have hindered the elucidation of the mechanism of its antineoplastic activity. In this study, we have used cervical (HeLa) and hepatoma (Hep3B) human cancer cells to investigate the ability of silibinin to inhibit tumour growth by targeting HIF-1 activity.
Silibinin inhibits HIF-1α accumulation in Hep3B and HeLa cells
To investigate whether silibinin could affect the activity of HIF-1, we first performed time course experiments to examine HIF-1α and HIF-1β protein expressions under hypoxic conditions in the absence and presence of silibinin. As expected, hypoxia induced a robust accumulation of HIF-1α protein in both HeLa and Hep3B cells, which was detectable after 1 h. The addition of silibinin to the culture medium completely inhibited HIF-1α accumulation in both cell types (Figure 1a), and the inhibition persisted as long as the drug was present (at least up to 16 h, data not shown). In contrast, silibinin did not modify steady-state levels of HIF-1β protein (Figure 1a).
Dose–response experiments showed that silibinin decreased hypoxia-induced HIF-1α protein expression in a dose-dependent manner (Figure 1b), with comparable potency in both HeLa and Hep3B cells (IC50∼150 μM). HIF-1α accumulation induced by the prolyl hydroxylase inhibitor dimethyloxallylglycine (DMOG), a well-characterized hypoxia mimetic (Wright et al., 2003), could also be abrogated with silibinin (Figure 1b). Consistent with the inhibition of HIF-1α accumulation, silibinin also mediated dose-dependent inhibition of HIF-1 transcriptional activity in hypoxia- or DMOG-treated cells, as determined using a hypoxia-responsive reporter construct (Figure 1c).
In addition, the cells were subjected to the increments of hypoxia (6–0.1% O2) in the presence or absence of silibinin. As shown in Figure 1d, silibinin prevented HIF-1α accumulation at all O2 concentrations tested, showing its capacity to inhibit HIF-1α at any extent of cellular hypoxia.
Silibinin does not affect prolyl hydroxylase-driven HIF-1α degradation
HIF-1α is degraded mainly through the ubiquitin/proteasome pathway after the hydroxylation of prolines 402 and 564 by specific HIF prolyl hydroxylases (Schofield and Ratcliffe, 2004). To address the possibility that silibinin inhibited HIF-1α accumulation by promoting degradation and/or reducing its protein half-life, we first examined the effect of silibinin on the stability of the double-proline mutant of HIF-1α (P402A/P564A-HIF-1α), which cannot be hydroxylated by prolyl hydroxylases and accumulates in cells in an O2-independent manner (Masson et al., 2001). To avoid interference from endogenous HIF-1α, we expressed P402A/P564A-HIF-1α or wild-type HIF-1α constructs in HIF-1α-deficient chinese hamster ovary (CHO) Ka13 cells (Wood et al., 1998). Silibinin was equally potent to prevent the accumulation of wild-type HIF-1α and of P402A/P564A-HIF-1α (Figure 2a), ruling out the possibility that prolyl hydroxylation was involved in silibinin's effect. Similar results were obtained in HeLa cells (data not shown).
In addition, we measured steady-state levels of HIF-1α after reoxygenation. To this end, cells were exposed to hypoxia for a given time, and silibinin or vehicle was added during the last 15 min of hypoxia. At the end of the hypoxic treatment (time 0), cells were exposed to normoxia for increasing periods up to 30 min, and HIF-1α protein was monitored by western blotting. Under these conditions, HIF-1α protein levels predominantly reflect the rate of HIF-1α degradation. As shown in Figure 2b, the degradation of HIF-1α was slightly faster in silibinin-treated than in control cells, with half-life values of 10.5±1.0 and 13.5±1.2 min, respectively, this difference most likely due to the somewhat reduced HIF-1α accumulation in silibinin-treated cells. Collectively, these results indicate that the potent inhibitory effect of silibinin on HIF-1α accumulation is not accounted for by an increase in protein degradation.
Silibinin does not affect HIF-1α mRNA accumulation but decreases HIF-1α protein synthesis
We next addressed the possibility that silibinin may be acting on HIF-1α production. The reverse transcriptase PCR analysis revealed that HIF-1α mRNA levels remained unchanged by treatment with silibinin both in normoxia and hypoxia (Figure 2c), indicating that silibinin does not affect the accumulation of HIF-1α mRNA. Later, to examine whether silibinin-mediated inhibition of HIF-1α accumulation was due to reduced HIF-1α protein synthesis, we utilized MG132, a potent and specific proteasome inhibitor, to prevent ubiquitin-dependent HIF-1α degradation. Treatment of cells with MG132 resulted in a pronounced accumulation of high-molecular-weight ubiquitinated-HIF-1α protein species, both in normoxia and hypoxia (Figure 2d, top panel, lanes 2 and 6). In contrast, the inhibition of protein synthesis with cycloheximide almost completely prevented the accumulation of ubiquitinated HIF-1α in the presence of MG132 (lanes 3 and 7). Similarly, the addition of silibinin in the presence of MG132 resulted in a substantial reduction of ubiquitinated HIF-1α in normoxia and hypoxia (lanes 4 and 8). Moreover, silibinin produced a rapid decrease in HIF-1α under hypoxia, which was very similar in the absence or presence of MG132 (Supplementary Figure S1), all these results suggesting interference with the protein synthesis process. Thus, to analyse the rate of HIF-1α synthesis, HeLa cells were pretreated for 30 min with silibinin or vehicle and then pulse-labelled with [35S]Met for up to 60 min under hypoxia, followed by immunoprecipitation of HIF-1α. As shown in Figure 2d (bottom panel), silibinin treatment greatly decreased 35S-labelled HIF-1α accumulation at all times tested. These results indicated that silibinin decreases the HIF-1α translation rate.
Inhibition of HIF-1α by silibinin correlates with the repression of the mTOR/p70S6 kinase/4E-BP1 pathway
The PI3K/Akt/mTOR pathway is implicated in the regulation of HIF-1α expression at the translational level (Zhong et al., 2000; Laughner et al., 2001; Hudson et al., 2002; Thomas et al., 2006). To address the potential involvement of this pathway in silibinin-mediated inhibition of HIF-1α accumulation, we measured the phosphorylation status of mTOR and its effectors p70S6K and 4E-BP1, as well as that of Akt. Interestingly, treatment of HeLa and Hep3B cells with silibinin under hypoxia resulted in a dose-dependent dephosphorylation of mTOR (Ser2448), which correlated with the inhibition of phosphorylation of p70S6K and 4E-BP1, and with the reduction of HIF-1α accumulation in both cell types (Figure 3a). Hypoxia has been shown to downregulate mTOR activity through the induction of Regulated in Development and DNA damage responses (REDD1) in a process that requires TSC1/TSC2 (tuberous sclerosis complex), an upstream mTOR regulator (Brugarolas et al., 2004). Here, we obtained identical results under normoxia (Supplementary Figure S2a), indicating that the effect of silibinin on mTOR is independent on a hypoxia-induced mechanism.
It can be noted that concomitant with the suppression of HIF-1α and the mTOR/p70S6K/4E-BP1 pathway, silibinin also induced the activation of Akt, as measured by phosphorylation at Ser473 (Figure 3a). Similarly, the mTOR inhibitor rapamycin induced a pronounced increase in phospho-Akt comparable with that produced by silibinin, both under normoxic and hypoxic conditions (Figure 3b). Silibinin did not affect the total protein levels of the corresponding phosphoproteins analysed (data not shown), indicating that this effect was specific to protein phosphorylation. Interestingly, we observed a differential behaviour between both cell types regarding the effect of silibinin on Akt. Although in HeLa cells, Akt phosphorylation increased progressively with increasing silibinin concentrations, in Hep3B cells high doses of silibinin (500 μM) inhibited Akt phosphorylation (Figures 3a, c and Supplementary Figure S2a). However, this difference had no implications for the inhibition of HIF-1α and mTOR, which remained maximal in both cell lines. It is worth noting that although silibinin-induced mTOR/p70S6K/4E-BP1 suppression was independent on the TSC1/TSC2 function, the activation of Akt was dependent on the presence of intact TSC1/TSC2, as shown by knocking down TSC2 expression with siRNA (Supplementary Figure S3). Furthermore, the activation of p-Akt by silibinin required the activity of its upstream regulator PI3K, as the PI3K inhibitor LY294002 blocked this activation (Figure 3b, right panel, and Figure 3c). LY294002 also enhanced the inhibitory effect of silibinin on the HIF-1α expression and mTOR pathway (reflected by the loss of phospho-p70S6K) (Figure 3c). Together, these results show that silibinin affects both, mTOR/p70S6K/4E-BP1 and PI3K/Akt pathways.
To further investigate the dynamic changes produced by silibinin on mTOR and Akt signalling in relation to the inhibition of HIF-1α expression, we exposed the cells to hypoxia and subsequently treated them with silibinin for the final 10–60 min of hypoxic incubation. As shown in Figure 4a, silibinin induced a rapid and potent inhibition of HIF-1α accumulation, which was evident in both cell lines as early as 10 min after treatment and was complete by 60 min. Silibinin-induced dephosphorylation of mTOR and its effectors was detectable within 10 min of treatment, and resulted in total loss of phospho-p70S6K after 20–30 min (Figure 4a). Phospho-Akt increased after 20 min of exposure to silibinin (500 μM) in HeLa cells, whereas it decreased below hypoxic basal levels in Hep3B cells, which is consistent with the dose–response experiments. Similarly, rapamycin produced a rapid inhibition of mTOR/p70S6K/4E-BP1 and the induction of p-Akt that was comparable with silibinin (Figure 4b). However, rapamycin did not modify HIF-1α levels appreciably in the same time course, with HIF-1α inhibition being apparent only after more prolonged treatments (Figure 3b, left panel), which is consistent with earlier reports (Zhong et al., 2000; Hudson et al., 2002; Thomas et al., 2006). Interestingly, the inhibition of HIF-1α was rapidly reversed after silibinin removal (Figure 4, lanes 7 and 8). Likewise, silibinin washout reverted all the phosphoproteins analysed to their basal phosphorylation levels. Again, identical results were obtained in normoxia (Supplementary Figure S2b). Overall, these results suggest that the suppression of mTOR signalling might be involved, at least in part, in the inhibition of HIF-1α protein synthesis by silibinin.
Silibinin inhibits hypoxia-induced VEGF secretion
Given that HIF-1 is the main regulator of VEGF expression in hypoxia (Ferrara, 2004), we next analysed the effect of silibinin on VEGF production in vitro. Compared with normoxic conditions, exposure of HeLa and Hep3B cells to hypoxia for 12 h induced a two- to three-fold increase in VEGF secretion to the extracellular medium (Figure 5). As anticipated, silibinin could inhibit hypoxia-induced VEGF release in both cell types, that is at a concentration of 250 μM silibinin resulted in ∼25% inhibition in HeLa and ∼40% inhibition in Hep3B cells, whereas at 500 μM the inhibition reached >90% in both cell lines (Figure 5a). No significant effects were observed at lower concentrations of 50 and 100 μM.
In a comparative analysis, we examined the effects of silibinin on the hypoxia-induced VEGF release in combination with maximal doses of the inhibitors of PI3K/Akt and mTOR pathways (Alvarez-Tejado et al., 2002). Addition of the PI3K/Akt inhibitor LY294002 produced a similar inhibition of VEGF release (∼30%) to that induced by a submaximal dose of silibinin (250 μM), whereas the mTOR inhibitor, rapamycin, produced a weak inhibition of ∼15% (Figure 5b). Interestingly, combined treatment of silibinin and LY294002 produced a synergistic effect, reducing VEGF release to near normoxic control levels. In contrast, the combined treatment of silibinin and rapamycin produced no additive effect (Figure 5b). At the maximal silibinin dose (500 μM), the inhibition of VEGF release was >90% irrespective of the presence of additional inhibitors (Figure 5b).
Silibinin potently inhibits cell proliferation
Examination of the effect of silibinin on cell proliferation revealed a strong dose-dependent inhibition in HeLa and Hep3B cells (Figure 6a). Compared with controls, the treatment of HeLa cells with silibinin for 8 h under hypoxic conditions resulted in the reduction of total cell number that amounted 10% at 100 μM, 45% at 250 μM and 75% at 500 μM. Identical treatment of Hep3B cells inhibited growth by 25% at 100 μM, 45% at 250 μM and 72% at 500 μM. Very similar results were obtained at incubation times of 4 and 6 h, and under normoxic conditions (data not shown). In contrast to silibinin, rapamycin did not affect cell proliferation in the same time course (Figure 6a); the effects of rapamycin were significant only after 24 h of treatment (data not shown). Silibinin exhibited a similar anti-proliferative behaviour in normal epithelial cells (Supplementary Figure S4).
Given that inhibition of cell proliferation might be mediated by apoptosis (Varghese et al., 2005; Singh et al., 2008), we examined the apoptosis-inducing capacity of silibinin by measuring the activation of caspases 3 and 7. We did not find a direct correlation between impaired cell growth and apoptosis: activation of apoptosis by silibinin in these experiments was evident only after 8 h of treatment, after the effect on cell proliferation (Figure 6b). Again, identical results were obtained in hypoxia and normoxia.
Overexpression of HIF-1α has been shown in many human cancers and their metastases, and its expression correlates with increased vascularity, resistance to chemotherapy and radiotherapy and poor prognosis (Brown and Giaccia, 1998; Zhong et al., 1999). The so-called ‘angiogenic switch’ is essential for tumour growth and metastasis (Hanahan and Folkman, 1996). Given that HIF-1-mediated VEGF expression plays a pivotal role for tumour angiogenesis (Ferrara, 2004), factors that modulate HIF-1 activity are potential targets for anticancer therapy (Giaccia et al., 2003; Semenza, 2003). In this study, we have shown that silibinin inhibited HIF-1 activity and VEGF production in human cervical and hepatoma cells under hypoxia. Silibinin was also found to downregulate the mTOR/p70S6K/4E-BP1 pathway, uncovering a novel mechanism for its anticancer activity.
The expression of HIF-1α is tightly regulated through both protein degradation and protein synthesis. We found that silibinin inhibited HIF-1α accumulation in HeLa and Hep3B cells, without affecting the prolyl hydroxylase-driven HIF-1α degradation. Neither did silibinin affect HIF-1α mRNA levels. Rather, silibinin was found to decrease the rate of HIF-1α protein synthesis, as shown by strong reduction of HIF-1α accumulation in metabolic labelling experiments, or despite the inhibition of proteasomal protein degradation.
Earlier studies have shown that the PI3K/Akt/mTOR pathway regulates HIF-1α protein translation through mTOR (Zhong et al., 2000; Laughner et al., 2001; Hudson et al., 2002; Thomas et al., 2006). mTOR, a central serine/threonine kinase, controls protein translation by the phosphorylation of two downstream effectors, namely ribosomal protein S6 kinase (p70S6K), which activation is thought to enhance the translation of mRNAs containing 5′-terminal oligopyrimidine tract (TOP) sequences in their 5′-UTR (untranslated region), and eukaryotic initiation factor 4E-binding protein-1 (4E-BP1), which inactivation induces eIF4E (eukaryotic initiation factor 4E), promoting cap-dependent mRNA translation (Bjornsti and Houghton, 2004; van den Beucken et al., 2006). Interestingly, the 5′-UTR of the mRNA encoding HIF-1α contains 5′-TOP sequences that might regulate its translation in response to mTOR activation (Laughner et al., 2001; Thomas et al., 2006). We have shown that treatment of HeLa and Hep3B cells with silibinin suppressed the phosphorylation of mTOR and its effectors p70S6K and 4E-BP1, which paralleled with the loss of HIF-1α expression. Moreover, we observed a total correlation between the recovery of the mTOR/p70S6K/4E-BP1 pathway and the expression of HIF-1α upon silibinin removal. Therefore, given the key role of this pathway in the regulation of HIF-1α translation, our results strongly suggested that silibinin-induced suppression of mTOR/p70S6K/4E-BP1 pathway might be involved in the inhibition of HIF-1α translation rate. Nonetheless, the fact that silibinin was more effective inhibiting HIF-1α accumulation than rapamycin (used at maximal dose) suggests the participation of an additional mechanism apart from mTOR/p70S6K/4E-BP1 repression. The identification of other potential mechanisms is currently under investigation in our laboratory.
In this study, silibinin was also found to induce the activation of the pro-survival kinase Akt. This result is consistent with recent findings describing the induction of Akt (Ser473) as the result of mTOR inhibition in several human cancer cells (Sun et al., 2005; O'Reilly et al., 2006; Sarbassov et al., 2006). It is well established that mTOR can form two multi-protein complexes that regulate different aspects of mTOR signalling (Guertin and Sabatini, 2007). The mTOR complex 1 (mTORC1) binds to raptor (regulatory-associated protein of mTOR), and controls cellular growth and proliferation by regulating ribosomal biogenesis and protein translation through p70S6K and 4E-BP1. Importantly, in addition to promoting protein translation, p70S6K can inhibit the PI3K/Akt pathway by phosphorylating and inactivating IRS1 (insulin receptor substrate-1) (Manning, 2004). Therefore, the inhibition of mTORC1 will activate Akt by relieving the negative feedback produced by p70S6K. The mTOR complex 2 (mTORC2) binds to rictor (rapamycin-insensitive companion of mTOR), and upregulates Akt activity by direct phosphorylation of Ser473 (Sarbassov et al., 2005). Strikingly, the activation of Akt by mTORC2 has been recently shown to depend on TSC1/TSC2 function (Huang et al., 2008), a well-established mTORC1 repressor (Guertin and Sabatini, 2007). Currently, rapamycin is recognized as a universal inhibitor of mTORC1 and a cell-type specific inhibitor of mTORC2 (Sabatini, 2006). On the basis of these mechanisms, tumour cells in which mTOR suppression results in Akt inhibition will express rapamycin-sensitive mTORC2, whereas cells in which Akt is activated or unaffected by mTOR suppression will express rapamycin-insensitive mTORC2 (Sabatini, 2006; Sarbassov et al., 2006). This distinction has become critical to predict whether a tumour will respond to treatment with mTOR inhibitors or not, given that Akt activation, by promoting cell survival, is considered to be a side effect (Sabatini, 2006). In our study, silibinin-induced mTOR suppression resulted in Akt activation in HeLa cells, whereas in Hep3B resulted in Akt activation or inhibition depending on the concentration used. Recent studies have shown the ability of silibinin to inhibit (Chen et al., 2005; Lah et al., 2007; Singh et al., 2008) and also to activate (Deep et al., 2008) Akt in different cancer cells, but the involvement of mTOR was not examined. Our finding that LY294002 blocked silibinin-induced Akt activation indicates that the PI3K activity is required for this activation after mTOR inhibition, which implies that silibinin induces Akt phosphorylation as a consequence of p70S6K inhibition (that is, mTORC1 inhibition). Moreover, the fact that in TSC2 knockdown cells silibinin is unable to induce Akt phosphorylation, regardless of mTOR/p70S6K/4E-BP1 inhibition, also suggests a role for mTORC2 in silibinin-induced Akt phosphorylation. Taken together, our results provide the first demonstration that silibinin behaves as an mTOR inhibitor, and suggest that similarly to rapamycin, silibinin may target mTORC1 (HeLa and Hep3B) and/or mTORC2 (Hep3B) depending on the cell type and treatment conditions. This possibility needs to be examined in more detail to define precisely the mechanisms underlying silibinin-mediated mTOR suppression, and thus evaluate the anticancer capacity of silibinin in different tumour models.
Silibinin has shown anti-angiogenic properties in several in vitro and in vivo tumour models (Singh et al., 2005, 2008; Gu et al., 2007), but the mechanisms of its anti-angiogenic action have not been fully elucidated. We report here that silibinin inhibits hypoxia-induced VEGF production in HeLa and Hep3B cells, suggesting that this is the consequence of HIF-1 inhibition. Unexpectedly, our studies revealed a lack of correlation between the profiles of HIF-1 and VEGF inhibition; although silibinin inhibited HIF-1 activity at all concentrations tested, it only decreased VEGF production at concentrations of 250 μM or above. It is known that VEGF expression is regulated by HIF-1-dependent and HIF-1-independent mechanisms (Ferrara, 2004), and recent data have shown that the activation of the PI3K/Akt pathway can trigger VEGF expression by promoting Sp1 phosphorylation (Pore et al., 2004). Our results suggest that the activation of Akt induced by silibinin (and also by rapamycin) can stimulate VEGF expression, compensating to some extent for its reduction due to HIF-1 inhibition. Importantly, our study shows that LY294002, by blocking Akt activation, potentiates silibinin's anti-angiogenic capability. These results provide a mechanistic basis for cancer therapy using silibinin in combination with other anti-tumoral drugs that target PI3K or Akt, similar to proposals for established mTOR inhibitors (Sun et al., 2005; Fan et al., 2006; O'Reilly et al., 2006).
Our study also confirms the potent anti-proliferative effect of silibinin, shown by strong inhibition of HeLa and Hep3B cell growth. Diverse mechanisms have been proposed to account for the inhibition of cancer cell growth by silibinin, such as cell-cycle arrest and the induction of apoptosis (Zi et al., 1998; Bhatia et al., 1999; Zi and Agarwal, 1999; Varghese et al., 2005; Singh et al., 2008). mTOR is well known as a regulator of cell cycle progression and cell proliferation, and its inhibition causes strong G1-phase cell-cycle arrest (Fingar and Blenis, 2004). In addition, mTOR activity can regulate apoptotic death in some cellular situations (Castedo et al., 2002). Our data here suggest that the inhibition of mTOR signalling induced by silibinin might account greatly for its anti-proliferative and pro-apoptotic effects. However, the fact that silibinin is more effective than rapamycin inhibiting cell proliferation again suggests an additional mechanism besides mTOR/p70S6K/4E-BP1 inhibition not shared by rapamycin.
In summary, this study shows, for the first time, that silibinin inhibits the mTOR/p70S6K/4E-BP1 signalling pathway and HIF-1 activity in HeLa and Hep3B cells. Thus, we have elucidated important mechanisms of the anticancer activity of silibinin, related to cell survival and angiogenesis, which are essential for the adaptation of cancer cells to microenvironmental hypoxia and hence for tumour progression. These mechanisms may in part explain the broad spectrum of silibinin's anticancer effects, and provide a rationale for the development of silibinin as an anticancer drug alone or in combination with PI3K/Akt inhibitors.
Materials and methods
Cell culture and reagents
Details describing cell culture and reagents are provided in the Supplementary information.
Cell extracts and western blotting
Cells were seeded into 60-mm culture dishes and allowed to attach for 24 h. Immediately after treatments, cells were washed with ice-cold phosphate-buffered saline and scraped in 1-ml ice-cold phosphate-buffered saline supplemented with protease inhibitor cocktail (Roche Diagnostics, Barcelona, Spain). Cell pellets were homogenized in 50 μl lysis buffer consisting of the CytoBuster protein extraction reagent (Novagen, Madison, WI, USA) supplemented with protease (Roche) and phosphatase (Sigma-Aldrich, Madrid, Spain) inhibitor cocktails. After incubation on ice (15 min), lysates were vortexed and centrifuged (16 000 × g for 10 min at 4 °C), and the supernatants collected as whole cell extracts. Cell extracts (40–80 μg) were resolved by SDS–polyacrylamide gel electrophoresis, transferred to nitrocellulose membranes and western blotting carried out as described earlier (Mateo et al., 2003). When required, the intensity of the bands was quantified with the Quantity One software (Bio-Rad, Madrid, Spain).
HIF-1α pulse labelling and immunoprecipitation
Cells were plated into 10-cm dishes. After 24 h, the cells were exposed to the methionine-free Dulbecco's modified essential medium for 30 min and then pulse labelled with 250 μCi/ml of [35S]L-methionine (Perkin-Elmer, Madrid, Spain) under hypoxia. Total cell lysates were prepared with RIPA buffer. Lysates (1 mg) were precleared with 20 μl of protein A/G-Plus agarose beads (Santa Cruz Biotechnology, Santa Cruz, CA, USA) for 1 h, and thereafter subjected to immunoprecipitation using a polyclonal anti-HIF-1α antibody (Santa Cruz Biotechnology) in constant rotation at 4 °C overnight. The protein A/G-Plus agarose beads were added, rotated for 2 h at 4 °C, pelleted and washed five times with RIPA buffer. An equal volume of 2 × sample buffer was added, and the samples were boiled and separated by SDS–polyacrylamide gel electrophoresis. The gel was vacuum-dried and autoradiographed.
Detailed description of plasmids construction is given in the Supplementary information.
Transient transfection and luciferase reporter assay
Transfection of CHO Ka13 cells was performed in 60-mm culture dishes using Lipofectamine 2000 (Invitrogen, Barcelona, Spain) and 1 μg of plasmid DNA (HIF-1α wild-type, P402A/P564A-HIF-1α or empty vector). Luciferase reporter assays were performed according to Alvarez-Tejado et al. (2002). Details provided in Supplementary Information.
Human VEGF was measured with an ELISA kit (Pierce, Rockford, IL, USA). Cells were plated in six-well plates and cultured to 80–90% confluence. The medium was replaced and cultures treated as indicated. Secreted VEGF in extracellular medium (50 μl) was quantified 12 h later, according to the manufacturer's instructions. Results were normalized to the amount of protein per well.
Total RNA, isolated using the RNeasy Mini kit (Qiagen, Valencia, CA, USA), was reverse transcribed with the SuperScript II reverse transcriptase (Invitrogen) using an oligo-(dT) primer according to the manufacturer's protocol. The cDNA was subjected to PCR amplification using the following forward and reverse primer sets: human HIF-1α, 5′-IndexTermCGTTGTGAGTGGTATTATTCAGCA-3′ and 5′-IndexTermCAGTTTCTGTGTCGTTGCTGCC-3′ (GenBank NM_001530); human glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 5′-IndexTermAGTGGGGTGATGCTGGTGCTG-3′ and 5′-IndexTermCGCCTGCTTCACCACCTTCTT-3′ (GenBank NM_002046). PCR conditions were established in pilot experiments to ensure linear reaction rates. GAPDH was used as the internal standard. PCR products were separated on 1.5% agarose gels and visualized by ethidium bromide staining. Gels were photographed using a Gel DOC 2000 image analyzer (Bio-Rad).
Cell proliferation and apoptosis
Cell proliferation was analysed with the fluorometric Cell-Blue Cell Viability Assay (Promega, Barcelona, Spain). For apoptosis analysis, we used the luminescent Caspase-Glo 3/7 Assay (Promega), which measures caspase-3 and caspase-7 activities. Both assays were performed in 96-well plates according to the manufacturer's recommendations.
Where indicated, experimental data were analysed using the Prism GraphPad software (San Diego, CA, USA). Significant differences were determined by Student's t-test (two-tailed). P-values below 0.05 were considered significant.
Alvarez-Tejado M, Alfranca A, Aragonés J, Vara A, Landázuri MO, del Peso L . (2002). Lack of evidence for the involvement of the phosphoinositide 3-kinase/Akt pathway in the activation of hypoxia-inducible factors by low oxygen tension. J Biol Chem 277: 13508–13517.
Bhatia N, Zhao J, Wolf DM, Agarwal R . (1999). Inhibition of human carcinoma cell growth and DNA synthesis by silibinin, an active constituent of milk thistle: comparison with silymarin. Cancer Lett 147: 77–84.
Bjornsti MA, Houghton PJ . (2004). The TOR pathway: a target for cancer therapy. Nat Rev Cancer 4: 335–348.
Brown JM, Giaccia AJ . (1998). The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. Cancer Res 58: 1408–1416.
Brugarolas J, Lei K, Hurley RL, Manning BD, Reiling JH, Hafen E et al. (2004). Regulation of mTOR function in response to hypoxia by REDD1 and the TSC1/TSC2 tumor suppressor complex. Genes Dev 18: 2893–2904.
Castedo M, Ferri KF, Kroemer G . (2002). Mammalian target of rapamycin (mTOR): pro- and anti-apoptotic. Cell Death Differ 9: 99–100.
Chen PN, Hsieh YS, Chiou HL, Chu SC . (2005). Silibinin inhibits cell invasion through inactivation of both PI3K–Akt and MAPK signalling pathways. Chem Biol Interact 156: 141–150.
Deep G, Oberlies NH, Kroll DJ, Agarwal R . (2008). Isosilybin B causes androgen receptor degradation in human prostate carcinoma cells via PI3K–Akt–Mdm2-mediated pathway. Oncogene 27: 3986–3998.
Fan QW, Knight ZA, Goldenberg DD, Yu W, Mostov K, Stokoe D et al. (2006). A dual PI3 kinase/mTOR inhibitor reveals emergent efficacy in glioma. Cancer Cell 9: 341–349.
Ferrara N . (2004). Vascular endothelial growth factor: basic science and clinical progress. Endocr Rev 25: 581–611.
Fingar DC, Blenis J . (2004). Target of rapamycin (TOR): an integrator of nutrient and growth factor signals and coordinator of cell growth and cell cycle progression. Oncogene 23: 3151–3171.
Giaccia A, Siim BG, Johnson RS . (2003). HIF-1 as a target for drug development. Nat Rev Drug Discov 2: 803–811.
Gu M, Singh RP, Dhanalakshmi S, Agarwal C, Agarwal R . (2007). Silibinin inhibits inflammatory and angiogenic attributes in photocarcinogenesis in SKH-1 hairless mice. Cancer Res 67: 3483–3491.
Guertin DA, Sabatini DM . (2007). Defining the role of mTOR in cancer. Cancer Cell 12: 9–22.
Hanahan D, Folkman J . (1996). Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell 86: 353–364.
Hockel M, Vaupel P . (2001). Tumor hypoxia: definitions and current clinical, biologic, and molecular aspects. J Natl Cancer Inst 93: 266–276.
Huang J, Dibble CC, Matsuzaki M, Manning BD . (2008). The TSC1–TSC2 complex is required for proper activation of mTOR complex 2. Mol Cell Biol 28: 4104–4115.
Hudson CC, Liu M, Chiang GG, Otterness DM, Loomis DC, Kaper F et al. (2002). Regulation of hypoxia-inducible factor 1alpha expression and function by the mammalian target of rapamycin. Mol Cell Biol 22: 7004–7014.
Jiang BH, Semenza GL, Bauer C, Marti HH . (1996). Hypoxia-inducible factor 1 levels vary exponentially over a physiologically relevant range of O2 tension. Am J Physiol 271: C1172–C1180.
Lah JJ, Cui W, Hu KQ . (2007). Effects and mechanisms of silibinin on human hepatoma cell lines. World J Gastroenterol 13: 5299–5305.
Laughner E, Taghavi P, Chiles K, Mahon PC, Semenza GL . (2001). HER2 (neu) signaling increases the rate of hypoxia-inducible factor 1alpha (HIF-1α) synthesis: novel mechanism for HIF-1-mediated vascular endothelial growth factor expression. Mol Cell Biol 21: 3995–4004.
Manning BD . (2004). Balancing Akt with S6K: implications for both metabolic diseases and tumorigenesis. J Cell Biol 167: 399–403.
Masson M, Willam C, Maxwell PH, Pugh CW, Ratcliffe PJ . (2001). Independent function of two destruction domains in hypoxia-inducible factor-alpha chains activated by prolyl hydroxylation. EMBO J 20: 5197–5206.
Mateo J, García-Lecea M, Cadenas S, Hernández C, Moncada S . (2003). Regulation of hypoxia-inducible factor-1alpha by nitric oxide through mitochondria-dependent and -independent pathways. Biochem J 376: 537–544.
O'Reilly KE, Rojo F, She QB, Solit D, Mills GB, Smith D et al. (2006). mTOR inhibition induces upstream receptor tyrosine kinase signaling and activates Akt. Cancer Res 66: 1500–1508.
Pore N, Liu S, Shu HK, Li B, Haas-Kogan D, Stokoe D et al. (2004). Sp1 is involved in Akt-mediated induction of VEGF expression through and HIF-1-independent mechanism. Mol Biol Cell 15: 4841–4853.
Sabatini DM . (2006). mTOR and cancer: insights into a complex relationship. Nat Rev Cancer 6: 729–734.
Saller R, Meier R, Brignoli R . (2001). The use of silymarin in the treatment of liver diseases. Drugs 61: 2035–2063.
Sarbassov DD, Ali SM, Sengupta S, Sheen J, Hsu P, Bagley A et al. (2006). Prolonged rapamycin treatment inhibits mTORC2 assembly and Akt/PKB. Mol Cell 22: 159–168.
Sarbassov DD, Guertin DA, Ali SM, Sabatini DM . (2005). Phosphorylation and regulation of Akt/PKB by the rictor–mTOR complex. Science 307: 1098–1101.
Schofield CJ, Ratcliffe PJ . (2004). Oxygen sensing by HIF hydroxylases. Nat Rev Mol Cell Biol 5: 343–354.
Semenza GL . (2002). HIF-1 and tumor progression: pathophysiology and therapeutics. Trends Mol Med 8: S62–S67.
Semenza GL . (2003). Targeting HIF-1 for cancer therapy. Nat Rev Cancer 3: 721–732.
Singh RP, Dhanalakshmi S, Agarwal C, Agarwal R . (2005). Silibinin strongly inhibits growth and survival of human endothelial cells via cell cycle arrest and downregulation of survivin, Akt and NF-kappaB: implications for angioprevention and antiangiogenic therapy. Oncogene 24: 1188–1202.
Singh RP, Gu M, Agarwal R . (2008). Silibinin inhibits colorectal cancer growth by inhibiting tumor cell proliferation and angiogenesis. Cancer Res 68: 2043–2050.
Sun SY, Rosenberg LM, Wang X, Zhou Z, Yue P, Fu H et al. (2005). Activation of Akt and eIF4E survival pathways by rapamycin-mediated mammalian target of rapamycin inhibition. Cancer Res 65: 7052–7058.
Thomas GV, Tran C, Mellinghoff IK, Welsbie DS, Chan E, Fueger B et al. (2006). Hypoxia-inducible factor determines sensitivity to inhibitors of mTOR in kidney cancer. Nat Med 12: 122–127.
van den Beucken T, Koritzinsky M, Wouters BG . (2006). Translational control of gene expression during hypoxia. Cancer Biol Ther 5: 749–755.
Varghese L, Agarwal C, Tyagi A, Singh RP, Agarwal R . (2005). Silibinin efficacy against human hepatocellular carcinoma. Clin Cancer Res 11: 8441–8448.
Wang GL, Jiang BH, Rue EA, Semenza GL . (1995). Hypoxia-inducible factor 1 is a basic–helix–loop–helix–PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci USA 92: 5510–5514.
Wang GL, Semenza GL . (1995). Purification and characterization of hypoxia-inducible factor 1. J Biol Chem 270: 1230–1237.
Wellington K, Jarvis B . (2001). Silymarin: a review of its clinical properties in the management of hepatic disorders. BioDrugs 15: 465–489.
Wood SM, Wiesener MS, Yeates KM, Okada N, Pugh CW, Maxwell PH et al. (1998). Selection and analysis of a mutant cell line defective in the hypoxia-inducible factor-1 α-subunit (HIF-1α). Characterization of hif-1alpha-dependent and -independent hypoxia-inducible gene expression. J Biol Chem 273: 8360–8368.
Wright G, Higgin JJ, Raines RT, Steenbergen C, Murphy E . (2003). Activation of the prolyl hydroxylase oxygen-sensor results in induction of GLUT1, heme oxygenase-1, and nitric-oxide synthase proteins and confers protection from metabolic inhibition to cardiomyocytes. J Biol Chem 278: 20235–20239.
Xu Q, Briggs J, Park S, Niu G, Kortylewski M, Zhang S et al. (2005). Targeting Stat3 blocks both HIF-1 and VEGF expression induced by multiple oncogenic growth signaling pathways. Oncogene 24: 5552–5560.
Zhong H, Chiles K, Feldser D, Laughner E, Hanrahan C, Georgescu MM et al. (2000). Modulation of hypoxia-inducible factor 1alpha expression by the epidermal growth factor/phosphatidylinositol 3-kinase/PTEN/AKT/FRAP pathway in human prostate cancer cells: implications for tumor angiogenesis and therapeutics. Cancer Res 60: 1541–1545.
Zhong H, De Marzo AM, Laughner E, Lim M, Hilton DA, Zagzag D et al. (1999). Overexpression of hypoxia-inducible factor 1α in common human cancers and their metastases. Cancer Res 59: 5830–5835.
Zi X, Agarwal R . (1999). Silibinin decreases prostate-specific antigen with cell growth inhibition via G1 arrest, leading to differentiation of prostate carcinoma cells: implications for prostate cancer intervention. Proc Natl Acad Sci USA 96: 7490–7495.
Zi X, Feyes DK, Agarwal R . (1998). Anticarcinogenic effect of a flavonoid antioxidant, silymarin, in human breast cancer cells MDA-MB 468: induction of G1 arrest through an increase in Cip1/p21 concomitant with a decrease in kinase activity of cyclin-dependent kinases and associated cyclins. Clin Cancer Res 4: 1055–1064.
This work was supported by grants from ISCIII (FIS03-0924 and FIS07-1168 to JM). PG-M holds a fellowship from the Centro Nacional de Investigaciones Cardiovasculares (CNIC)-Bancaja predoctoral program. CNIC is supported by the Spanish Ministry of Health and Consumer Affairs and the Pro-CNIC Foundation.
About this article
Cite this article
García-Maceira, P., Mateo, J. Silibinin inhibits hypoxia-inducible factor-1α and mTOR/p70S6K/4E-BP1 signalling pathway in human cervical and hepatoma cancer cells: implications for anticancer therapy. Oncogene 28, 313–324 (2009). https://doi.org/10.1038/onc.2008.398
- hypoxia-inducible factor-1α
Effects of HIF-1α on Spermatogenesis of Varicocele Rats by Regulating VEGF/PI3K/Akt Signaling Pathway
Reproductive Sciences (2021)
Immunoregulatory Effects of Silymarin on Proliferation and Activation of Th1 Cells Isolated from Newly Diagnosed and IFN-ß1b-Treated MS Patients
Intravitreal Ets1 siRNA alleviates choroidal neovascularization in a mouse model of age-related macular degeneration
Cell and Tissue Research (2019)
Journal of Biomedical Science (2016)
Suppression of c-Myc induces apoptosis via an AMPK/mTOR-dependent pathway by 4-O-methyl-ascochlorin in leukemia cells