Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
PLoS ONE. 2009; 4(4): e5223.
Published online 2009 Apr 29. doi:  10.1371/journal.pone.0005223
PMCID: PMC2671147

Antitumor Activity and Mechanism of Action of the Cyclopenta[b]benzofuran, Silvestrol

Thomas Preiss, Editor



Flavaglines are a family of natural products from the genus Aglaia that exhibit anti-cancer activity in vitro and in vivo and inhibit translation initiation. They have been shown to modulate the activity of eIF4A, the DEAD-box RNA helicase subunit of the eukaryotic initiation factor (eIF) 4F complex, a complex that stimulates ribosome recruitment during translation initiation. One flavagline, silvestrol, is capable of modulating chemosensitivity in a mechanism-based mouse model.

Methodology/Principal Findings

Among a number of flavagline family members tested herein, we find that silvestrol is the more potent translation inhibitor among these. We find that silvestrol impairs the ribosome recruitment step of translation initiation by affecting the composition of the eukaryotic initiation factor (eIF) 4F complex. We show that silvestrol exhibits significant anticancer activity in human breast and prostate cancer xenograft models, and that this is associated with increased apoptosis, decreased proliferation, and inhibition of angiogenesis. We demonstrate that targeting translation by silvestrol results in preferential inhibition of weakly initiating mRNAs.


Our results indicate that silvestrol is a potent anti-cancer compound in vivo that exerts its activity by affecting survival pathways as well as angiogenesis. We propose that silvestrol mediates its effects by preferentially inhibiting translation of malignancy-related mRNAs. Silvestrol appears to be well tolerated in animals.


Cyclopenta[b]benzofuran flavaglines are inhibitors of translation initiation isolated from Asian plants of the genus Aglaia of the family Meliacae [1][4]. These compounds show in vitro activity against tumor cell lines [1], [2], promising activity in xenograft cancer models [2], [5], [6] and appear to block G2/M cell cycle progression [7]. We have previously shown that the flavagline silvestrol can re-sensitize tumor cells to standard-of-care agents, such as doxorubicin, in the Eμ-myc lymphoma model [4]. Silvestrol inhibits translation initiation by targeting the RNA helicase, eukaryotic initiation factor (eIF) 4A, and prevents ribosome loading onto mRNA templates [4].

Translation initiation is regulated by eIF4F at the level of the ribosome recruitment step. eIF4F is composed of three subunits: eIF4E, which binds to the cap structure present at the 5′ end of mRNAs; eIF4A, a DEAD-box RNA helicase implicated in preparing a ribosome landing pad for 43S pre-initiation complexes (40S ribosomal subunit and associated factors) by unwinding 5′ mRNA structure; and eIF4G, a large scaffolding protein involved in recruiting the 43S pre-initiation complex via its interaction with 40S-associated eIF3 [8]. eIF4A is an abundant translation factor that exists in a free form (referred to herein as eIF4Af) and as a subunit of the heterotrimeric eIF4F complex (eIF4Ac) [9], [10]. The helicase activity of eIF4Ac is ∼20-fold more efficient than eIF4Af and during initiation, eIF4Af likely cycles through the eIF4F complex [11]. Silvestrol acts as a chemical inducer of dimerization (CID) to force an engagement between eIF4Af and RNA, although how this inhibits translation initiation is not known [4].

Levels of cellular eIF4F are regulated by the target of rapamycin, mTOR [12], [13]. The extent to which translation of specific mRNAs is altered in response to changes in mTOR activity and eIF4F levels varies substantially among different transcripts and is largely dependent upon sequence elements within each mRNA, such as the presence of discrete hairpin structures in the 5′ untranslated regions [14]. Many cellular mRNAs are characterized by relatively short, unstructured 5′ UTRs (e.g. β-actin, GAPDH) that require a minimal amount of eIF4F for 43S pre-initiation complexes recruitment. These mRNAs are efficiently translated when eIF4F activity is limiting. By comparison, a select group of mRNAs is extremely sensitive to, and dependent upon, eIF4F for translation. These mRNAs typically harbor lengthy, G+C rich, highly-structured 5′ UTRs that encumber efficient RNA unwinding by the eIF4F complex and subsequently prevent efficient ribosome loading [14]. Hence, altering levels with flavaglines can exert profound gene specific effects.

The deregulation of the PI3k/Akt/mTOR signaling axis in human cancers, the finding that ectopic expression of eIF4E is oncogenic [15], [16], and the demonstration that targeted down-regulation of eIF4E displays therapeutic benefit in xenograft mouse models [17] suggest that the process of translation initiation is a potential anti-cancer target. Herein, we report that silvestrol is effective against two human xenograft models as a single agent. We provide further insight into the mechanism of action of silvestrol and show that it depletes the eIF4F complex of eIF4A and this is associated with a preferential reduction in the translation of mRNAs with structured 5′untranslated regions. Our results are consistent with the idea that silvestrol's anticancer activity is linked to its ability to preferentially block translation of highly structured, malignancy-related mRNAs.


Structure-activity relationships of flavaglines

We have previously characterized silvestrol as an inhibitor of translation initiation in vitro and in vivo [4]. To determine how extensive this property is among flavaglines, we tested members of the cyclopenta[b]benzofurans, cyclopent[bc]benzopyrans, and benzo[b]oxepines family for their potential to inhibit protein synthesis. (Fig. 1 and Figs. S1 and S2). Among the compounds tested were rocaglaol (Fig. 1A, compound #4) [18], that has been shown to induce apoptosis and cell cycle arrest in LNCaP cells [7]; aglafolinformylester (compound #7) [19], which has been shown to inhibit cap specific translation initiation (4) and silvestrol, which can sensitize lymphomas to the cytotoxic action of doxorubicin (4). In vitro translation reactions in extracts programmed with the bicistronic mRNA, FF/HCV/Ren (Fig. S1A), revealed that although compound #7 inhibited FF translation to the greatest extent when compared to compounds 1–11 (Fig. 1B and Fig. S1B), it was not as potent as silvestrol in vitro (Figs. 1C) or in vivo (Fig. 1D). The closely related cyclopenta[bc]benzopyrans and benzo[b]oxepines flavagline-type compounds did not show significant activity as inhibitors of translation (Fig. S2). We focused the remainder of our studies on silvestrol since it displayed the most potent inhibition of protein synthesis among the flavaglines that we have identified to date.

Figure 1
Structure-activity relationship analysis of cyclopenta[b]benzofurans.

Mechanism of Action of Silvestrol

We have previously demonstrated that silvestrol acts as a chemical inducer of dimerization – promoting the interaction between eIF4A and RNA [4]. We confirmed that silvestrol increases the RNA binding properties of eIF4A using nitrocellulose binding assays (Fig. 2A). In this assay, 32P-labeled RNA is retained on nitrocellulose filters only when bound to proteins. Very little RNA was retained by nitrocellulose when RNA was incubated only in the presence of eIF4A (Fig. 2A, compare lane 2 to 1). In the presence of pateamine, a CID that induces eIF4Af-RNA interaction, a significant proportion of 32P-labeled RNA was retained (Fig. 2A, compare lane 3 to 2) [20]. Addition of silvestrol to the eIF4A-RNA binding reactions increased the retention of 32P-labeled RNA on nitrocellulose (compare lanes 5–8 to 2); an event that was inhibited by hippuristanol (compare lane 9 to 7) – a compound that binds to the C-terminal domain of eIF4A to inhibit RNA binding [21]. These results indicate that silvestrol enhances binding between eIF4Af and RNA.

Figure 2
Silvestrol stimulates binding of eIF4A to mRNA.

We have previously shown that RNA binding of eIF4Ac is also increased by silvestrol [4], and we now demonstrated that this was a cap-dependent phenomenon (Fig. 2B). Crosslinking of eIF4E and eIF4Ac from RSW to 32P-labeled mRNA cap structures was inhibited by the presence of m7GDP (Fig. 2B, compare lane 2 to 1). As previously documented [22], crosslinking of eIF4A, but not eIF4E, was ATP-dependent (Fig 2B, compare lane 3 to 1). The presence of silvestrol in the reaction stimulated the crosslinking of eIF4Ac, but not eIF4E (compare lane 4 to 1). Crosslinking of eIF4Ac in the presence of silvestrol was inhibited by m7GDP and required ATP (compare lanes 5–7 to 4).

One consequence of increased RNA binding of eIF4A in the presence of silvestrol is that the eIF4F complex could become depleted of eIF4Ac – thus reducing cap-dependent translation. We therefore tested this possibility by purifying eIF4E from RSW containing silvestrol (Fig. 2C; left panel) and from cell extracts that had been prepared from MDA-MB-231 silvestrol-treated cells (Fig. 2C; right panel) using m7GTP-Sepharose affinity chromatography. We then probed for the presence of eIF4E and co-purifying eIF4A in m7GTP eluates. In both cases, the levels of eIF4Ac in the eIF4F complex were reduced (Figs. 2C).

To determine if components of the eIF4F complex (i.e.-eIF4G) co-sequestered with eIF4A to RNA in the presence of silvestrol, we performed pull-down assays with poly(rG)-agarose (Fig. 2D). Increased eIF4A levels were retained on poly(rG)-agarose in the presence of silvestrol and pateamine, however, this did not result in a concomitant increase of eIF4G associating with poly(rG)-agarose (Fig. 2D, compare lanes 2 and 3 to 1). [The small amount of eIF4G binding to poly(rG)-agarose may be due to this proteins' intrinsic RNA binding.]

Inhibition of translation by silvestrol leads to mRNA discrimination

Silvestrol inhibited protein synthesis in MDA-MB-231 breast and PC-3 prostate cancer cell lines with approximately the same IC50 (∼60 nM) following a 1 h exposure (Fig. 3A). We next determined the extent to which 25 nM silvestrol would affect protein synthesis rates in these cells. Translation rates were monitored as a function of time-post-exposure to silvestrol by labeling proteins with 35S-Met 15 min before harvest (Fig. 3B). A biphasic response was noted with a precipitous drop occurring over the first 8 h, followed by a slower reduction in translation occurring from 8–72 h (Fig. 3B). The reduction in translation rates was not a consequence of silvestrol-induced apoptosis since this was only observed for MDA-MB-231 cells during the last 24 h of the experiment (40% reduction in viability) (Fig. 3B). We noticed that during the first 8 h following inhibition of protein synthesis by silvestrol there was an ∼2 fold decrease in 35S-met incorporation into newly synthesized proteins for MDA-MB-231 (Fig. S3A: between 0–4 h for MDA-MB-231 cells - compare lane 2 to 1) and PC-3 (Fig. S3A: between 4–8 h for PC-3 cells: compare lane 3 to 2). By 24 h, we noted a reduction in the labeling of specific proteins (Fig. S3A: denoted by open boxes). In contrast, cycloheximide, an inhibitor of elongation, completely blocked protein synthesis and reduced 35S-Met incorporation equivalently in all proteins (data not shown).

Figure 3
Silvestrol inhibits translation in vivo in MDA-MB-231 breast and PC-3 prostate cancer cells.

These results indicate that treatment of cells with silvestrol might lead to mRNA discrimination during translation. To explore this, we generated reporter constructs in which a G-quadruplex had been engineered into the 5′ untranslated region of the mRNA 6 nucleotides from the site of transcription initiation (Fig. S4A, G-Q(+6)/RL). The G-quadruplex was modeled on one previously described to be present within the 5′ UTR of NRAS RNA and shown to mediate translational repression [23]. A control construct, [CAA]10/RL, generated a reporter transcript having a 5′ UTR consisting of [CAA] tracts and harboring minimal secondary structure with lowered eIF4F dependency [24]. As an internal control, we utilized a construct in which the HCV IRES was driving expression of firefly luciferase, an element known to recruit ribosomes in an eIF4F-independent manner [25]. Translation of G-Q(+6)/RL was more cap-dependent than [CAA]10/RL in Krebs extracts, as judged by inhibition of translation in the presence of m7GDP (Fig. S4B). In addition, G-Q(+6)/RL was more sensitive to reduced levels in eIF4A activity, as determined by inhibition of translation in the presence of the eIF4A inhibitor, hippuristanol [26]. In the presence of increasing concentrations of silvestrol, translation of G-Q(+6)/RL was inhibited to a greater degree than [CAA]10/RL mRNA (Fig. S4C), being consistent with inhibition of translation by silvestrol leading to mRNA discriminatory effects. This discriminatory effect was also observed with another flavagline, compound #7 (Fig. 1A), as well as on another set of reporter constructs in which the highly structured HIV TAR element was placed upstream of the chloramphenicol acetyl transferase (CAT) coding region (Fig. S5A). The PLTAR element is inhibitory to translation initiation and renders translation more sensitive to ionic concentrations [27] (Fig. S5B; compare lane 7 to 4). PLTAR CAT mRNA translation proved to be more sensitive to compound #7, than the CAT reporter (Fig. S5B, compare lanes 8–11), indicating that the results obtained with silvestrol are not compound- or reporter- specific (Fig. S4).

We next asked whether exposure of cells to silvestrol would affect the expression of malignancy-related proteins (Fig. 3C). We chose to evaluate the oncogenes cyclin D1 and c-myc and the anti-apoptotic proteins Bcl-2, survivin, and Mcl-1 since these have all been reported to be eIF4E-dependent. Levels of cyclin D1, Bcl-2 and c-myc diminished within the first 8 h of exposing MDA-MB-231 cells to silvestrol (Fig. 3C, compare lanes 1–3). Levels of survivin and Mcl-1 decreased substantially after 48 h, whereas tubulin and GAPDH expression was largely unaffected (Fig. 3C), consistent with the notion that silvestrol selectively affects production of growth-related proteins. Treatment of cells with vehicle for 72 hrs did not change levels of any of these proteins (Fig. 3C, compare lane 7 to 8). The levels of tubulin and GAPDH seen at 48–72 hrs in silvestrol treated cells are also similar to levels observed in untreated cells and thus cannot be attributed to selective enhancement of their synthesis by silvestrol (Fig. 3C, compare lanes 5 and 6 to 1). RT-PCR analysis of RNA isolated from unbound and polysome-bound fractions from DMSO- and silvestrol-treated MDA-MB-231 cells indicated that silvestrol leads to a redistribution of mRNA from polysomes into unbound fractions for these malignancy-related mRNAs (Fig. S3B). This is consistent with silvestrol inhibiting translation initiation of these transcripts.

We have previously shown that eIF2α is not phosphorylated in response to exposure of cells to silvestrol [4], and wanted to assess if inhibition of translation by silvestrol induced stress granule formation, a common event associated with inhibition of translation initiation. Arsenite, a known inducer of SGs caused relocalization of eIF4A and eIF4E into SGs, as determined by co-localization with the SG marker, G3BP (Fig. S6A, B). Exposure of HeLa cells to silvestrol also induced colocalization of eIF4A, eIF4E and G3BP into SGs (Fig. S6).

Silvestrol reduces translation in normal mouse tissues without cytotoxicity

The cap-dependent translation process is a target of silvestrol. Injection of silvestrol for two consecutive days into mice caused a reduction in liver protein synthesis, as assessed by polysome analysis in liver tissue isolated 3 and 6 h after the second injection (Fig. 4A). These results clearly indicate that silvestrol can inhibit protein synthesis in vivo. We then administered silvestrol to non-tumor bearing mice once a day for 8 consecutive days and assessed if there was an effect on red blood cells, lymphocytes, or monocyte/granulocyte counts from both bone marrow (BM) and spleen (SP) of control and silvestrol-treated mice (Fig. 4B). Vehicle- and silvestrol-treated animals displayed similar blood cell profiles (Fig. 4B). As well, there was no change in body weight of mice treated with silvestrol (Fig. S7) or any signs of illness or distress. Additionally, there was no appreciable change in liver or spleen weight (Fig. 4C). In vivo toxicity can be monitored by the appearance of enzymes in the serum such as ALT and AST. However, rather than an elevation of ALT or AST levels as expected for liver or muscle damage, we observed a slight reduction in AST levels and a 50% decrease in ALT levels, likely reflecting the inhibition of protein synthesis exerted in vivo by silvestrol (Fig. 4C). Collectively, these data indicate that silvestrol appears well tolerated in normal tissues.

Figure 4
Long-term administration of silvestrol is well tolerated.

Silvestrol suppresses HUVEC cell growth

The potential consequences of silvestrol on angiogenesis have not been explored, so we evaluated if it could suppress the response of endothelial cells to angiogenic stimuli. Silvestrol inhibited protein synthesis in HUVEC cells (Figs. 5A, B) and did not induce apoptosis even when cells were exposed to compound for 72 h (Fig. 5B). We also evaluated the ability of silvestrol to prevent or delay formation of tube or chord-like structures by HUVECs. Though this in vitro model does not fully recapitulate the features of in vivo angiogenesis, it is none-the-less useful for evaluating the response of endothelial cells to angiogenic stimuli. Silvestrol prevented the ability of HUVECs cultured on Matrigel to form vessel-like structures (Fig. 5C). This was quantitated by counting the number of tubes per field and demonstrated that silvestrol caused a dose-dependent reduction in tube formation (Fig. 5D).

Figure 5
Silvestrol inhibits protein synthesis and suppresses endothelial cell tube formation.

Silvestrol suppresses xenograft tumor growth

Deregulated eIF4F activity has been postulated to contribute to the oncogenic process in breast cancers [28]. We therefore chose to evaluate the impact of silvestrol administration on tumor growth in MDA-MB-231 breast cancer xenografts (Fig. 6). At 11 days after implantation, silvestrol was administered intraperitonally at 0.5 mg/kg once per day for 8 consecutive days. Following treatment, mice were monitored for up to 2 months and we noted that silvestrol dramatically suppressed the growth of these tumors (Fig. 6A and B). In these experiments, we used two different passages of MDA-MB-231 cells, an early passage that appeared to be slower growing in vivo (Fig. 6A) and a late passage, faster growing, line (Fig. 6B). In contrast to silvestrol treatment, administration of doxorubicin or rapamycin did not inhibit tumor growth (Fig. 6B). However, synergy with doxorubicin was observed for both rapamycin and silvestrol in vitro when tested against MDA-MB-231 cells (Table 1). An additional xenograft model that we used was the PC-3 human prostate cancer model, since the AKT/mTOR signaling pathway is frequently upregulated in these cancers [29], [30]. Growth of PC-3 xenografts was also significantly reduced in mice dosed with 0.5 mg/kg silvestrol for 8 consecutive injections following appearance of the tumors 24 days after implantation (Fig. S8). This observation contrasts with the results obtained with Dox at 5 mg/kg, which showed little effect.

Figure 6
Silvestrol suppresses xenograft tumor growth.
Table 1
Synergy of Doxorubicin with rapamycin and silvestrol on MDA-MB-231 cells.

MDA-MB-231 tumors harvested at the end of the study were analyzed to determine the fraction of apoptotic and dividing cells (Fig. 6C). TUNEL staining showed a significant increase in silvestrol-treated tumor cells staining positive for apoptosis and proliferation of silvestrol-treated tumor cells was greatly reduced as judged by staining for Ki67 (Fig. 6C).


Deregulated translation initiation through elevated eIF4E levels and increased eIF4F activity has been repeatedly implicated in malignancy through the altered translation of pro-survival and pro-growth encoding mRNAs that contribute to oncogenesis [31], [32], angiogenesis [17], and chemoresistance [33]. Recently, integrated genomic analysis of human glioblastoma multiforme, pancreatic cancers, breast and colon cancers revealed that human cancers contain a large number of genetic alterations (e.g. an average of 63 genetic alterations for pancreatic cancers) that define a core set of cellular signaling pathways and processes that are altered in a large number of tumors (67–100% in pancreatic cancers) [34]. In the case of pancreatic cancers, these include K-ras signaling, invasion, Small GTPase-signaling, apoptosis, integrin signaling, DNA damage control, notch signaling, and JNK signaling – all pathways that converge on eIF4E [35][41]. Hence inhibiting the eIF4F checkpoint suppresses tumor growth through multiple mechanisms and provides a rationale for development of broad-acting therapeutics. We show here that targeting the eIF4A subunit of eIF4F in human cancer xenograft tissues is sufficient to reduce tumor cell proliferation.

Silvestrol was the most active cyclopenta[b]benzofurans inhibiting protein synthesis among those that we tested. Interestingly, this compound has the same core structure as compound #2, which did not inhibit cap-dependent protein synthesis to the same extent as other compounds tested - notably #6, 7 and 8. The acetate moiety on compound #7 seems to be an important contributor to increasing this compound's activity relative to compound #2. These results indicate that silvestrol's activity may be improved by incorporating this functional group. Clearly, the dioxane moiety of silvestrol is a major contributor to its activity. Cyclopent[bc]benzopyrans and benzo[b]oxepines did not show any inhibition of protein synthesis (Fig. S2) indicating that the core benzofuran ring system is essential for this property. Cyclopent[bc]benzopyrans and benzo[b]oxepines have shown no antiproliferative activity for cancer cell lines in vitro, consistent with the idea that inhibition of protein synthesis by cyclopenta[b]benzofurans is responsible for their anti-proliferative activity [42].

In the current study, we also provide mechanistic insight into the mode of action of silvestrol. We confirm that RNA binding of eIF4Af is increased in the presence of silvestrol using an RNA filter binding assay and that this binding is inhibited by hippuristanol (Fig. 2A). The increased crosslinking of eIF4Ac to mRNA from RSW was shown to be cap-dependent (Fig. 2B). There are several possibilities to explain these results including stabilization of eIF4Ac in the eIF4F complex by silvestrol, increased recycling of eIF4Af through the eIF4F complex during initiation, or altered conformation of eIF4Ac induced by silvestrol that favors crosslinking (e.g.-exposure of an amide residue). We find less eIF4Ac associated with the eIF4F complex when eIF4F is isolated by m7GTP-Sepharose affinity column from RSW or silvestrol-treated cells (Fig. 2C), consistent with the idea that silvestrol depletes the eIF4F complex of eIF4A. We find no increase in eIF4G associated with RNA-bound eIF4A indicating that the eIF4F complex is not sequestered to RNA by silvestrol (Fig. 2D).

We have previously reported that administration of silvestrol to mice bearing Eμ-myc derived lymphomas synergizes with standard-of-care agents, such as doxorubicin, to induce apoptosis in vivo [4]. However, as a single agent silvestrol was not effective at curtailing lymphoma development in this model [4]. In the current setting, silvestrol was effective as a single agent against two different human xenograft models, resulting in tumor growth arrest associated with massive apoptosis and halt of cellular proliferation (Fig. 6). The reasons for these differences are not immediately apparent, but may be related to the presence of an activated c-myc allele in the Eμ-myc model. Alternatively, silvestrol's anti-angiogenic effects (Fig. 5) may be quite important for its anti-proliferative activity against solid tumors, but less so towards lymphomas. Alternatively, it may be that the tumors used in the current study are more “addicted” to altered translation initiation rates (due to increased numbers of altered signaling pathways or processes) than the Eμ-myc/PTEN tumors previously used [4], and hence respond to single agent treatment. Interestingly, rapamycin showed no effect in the MDA-MB-231 xenograft model used here demonstrating the superiority of silvestrol in the two models tested (Fig. 6B). Although high concentrations of doxorubicin (10 mg/kg) were sufficient to achieve an anti-tumor response, the mice lost weight and did not thrive (R.C., data not shown). A single dose of doxorubicin at 5 mg/kg was not effective against MDA-MB-231 breast cancer cells in vivo, although rapamycin and silvestrol did show synergy with doxorubicin in vitro (Table 1).

Many key proteins involved in malignancy are translationally controlled, including the potent angiogenic factors VEGF and FGF-2, the oncogenes cyclin D1 and c-myc, the antiapoptotic proteins of the Bcl family, as well as the inhibitor of apoptosis protein survivin (refs. 30, 31; see ref. 1 for a more extensive list of translationally controlled proteins involved in malignancy). Indeed, modulation of eIF4E can directly affect the expression of many of these malignancy related proteins (reviewed in refs. 1, 30). One prediction is that silvestrol should preferentially inhibit the translation of mRNAs whose expression is more dependent on eIF4F for ribosome recruitment. Indeed, silvestrol (and Compound #7) showed a preference for affecting the translation of “weaker” mRNAs (Figs. 3C, S3B, S4 and S5). This discriminatory effect is consistent with silvestrol targeting the eIF4F complex and inhibiting ribosome recruitment. Our results do not exclude an additional effect of silvestrol on protein stability. The ability of silvestrol to induce formation of stress granules (Fig. S6) is expected for an inhibitor of translation initiation.

The inhibition of protein synthesis that we observed with silvestrol for MDA-MB-231 and PC-3 cells is very different than that noted with 4E-antisense oligonucleotides. We observed a biphasic response with a precipitous drop occurring over the first 8 h, followed by a slower reduction in translation occurring from 8–72 h (Fig. 3B). This result is in contrast to what has been reported for eIF4E antisense oligonucleotides, where no reduction in global protein synthesis was observed over the course of 72 h [17]. These differences may reflect the different requirements of translation initiation for eIF4A versus eIF4E, since ribosome recruitment to mRNAs containing unstructured 5′UTRs can be mediated by eIF4G/eIF4A in the absence of eIF4E [24]. Nonetheless, it does seem that inhibition of translation with silvestrol can result in mRNA discriminatory effects (Fig. S4).

Administration of silvestrol appeared well tolerated in non-tumor bearing mice without inducing appreciable toxicity (Fig. 4 and Fig. S7). In normal resting cells translation initiation may be very low as a large fraction of eIF4E may be complexed with the inhibitory 4E-BPs in the absence of Akt/mTOR pathway stimulation [28]. As well, in normal cells stimulated to proliferate, the spectrum of mRNA whose translation is elevated may differ significantly from those elevated in transformed cells. Consistent with this hypothesis, increased phosphorylation of 4E-BP (presumably leading to elevated eIF4F activity) in liver from mice administered the branched chain amino acid (which activates mTOR) stimulates ribosomal protein mRNA translation but not global rates of protein synthesis in liver [43]. Yet, altering eIF4E levels by transcriptional activation [44], knock down using shRNAs [44], or inhibiting mTOR activity [45] in transformed cells can affect mRNA-selective translation and global rates of protein synthesis (∼2-fold decrease). The alteration in global translation rates may be due to a larger pool of mRNAs that are discriminated by eIF4E in tumor cells. Using anti-sense oligonucleotides to knockdown expression of eIF4E in the mouse is also well tolerated [17]. Our results support the idea that curtailing translation initiation by modulating eIF4A activity is a promising anti-cancer therapeutic approach that is well tolerated.

Materials and Methods

Ethics Statement

All animal studies were approved by the McGill University Faculty of Medicine Animal Care Committee

General Reagents

Silvestrol was resuspended in DMSO and stored at −70°C. Doxorubicin (Sigma) was dissolved in water and stored at 4°C. Rapamycin (LC Laboratories, Woburn, MA) was resuspended in 100% ethanol and stored at −70°C.

Cell culture

Malignant metastatic human mammary epithelial MDA-MB-231 cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD) and grown as monolayers in L15 medium (Invitrogen) supplemented with 10% fetal bovine serum and 100 U/ml penicillin/streptomycin. PC-3 cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD) and grown as monolayers in F12-K medium (Invitrogen) supplemented with 10% fetal bovine serum and 100 U/ml penicillin/streptomycin. HUVEC cells were obtained from Lonza (Walkersville, MD) and propagated in EBM-2 medium supplemented with EGM-2.

Recombinant DNA constructs and in vitro translations

The bicistronic reporter, FF/HCV/Ren, has been previously described [46]. For generating reporter constructs containing a G-quadruplex or a [CAA] tract for in vitro translation assays, the plasmid phRL-null was used (Promega). This plasmid was linearized with NheI/NcoI and gel purified. Two sets of annealed oligonucleotides were inserted into these sites. One set [sense: 5′CTAG[CAA]10C3′ and antisense: 5′CATGG[TTG]103′ ] generated [CAA]10/RL, whereas Set II [sense: 5′CTAGGGGAGGGGCGGGTCTGGG[CAA]4C3′ and antisense: 5′CATG[GTT]4CCCAGACCCGCCCCTCCC3′] generated G-Q(+6)/RL. Plasmid G-Q(+6)/RL contains the G quadruplex from NRAS positioned six nucleotides downstream from the T7 RNA polymerase transcription start site [23]. Plasmids encoding CAT or PLTARCAT have been previously described [47].

For in vitro translations, constructs G-Q (+6)/RL and [CAA]10/RL were linearized with XbaI followed by in vitro transciption with T7 RNA polymerase. Plasmid pSP72/HCV/Luc A+ (a gift from Dr. N. Sonenberg [McGill University]) was linearized with BamHI and subsequently in vitro transcribed with T7 RNA polymerase. In vitro translations where performed with 0.2 ng/µl in vitro transcribed RNAs in micrococcal nuclease treated Krebs extract as described previously [46] with the indicated concentrations of silvestrol for 1 h at 30°C. FF and Ren luc activity (RLU) were measured on a Berthold Lumat LB 9507 luminometer. In vitro translations of FF-HCV-Ren in RRL (Promega) were performed according to the manufacturer's instructions.

Filter Binding and Crosslinking Assays

Filter binding and chemical crosslinking assays were performed as previously described [22], [48]. For crosslinking assays, a 30 µL reaction containing 10 µL ribosomal salt wash (RSW) (1.2 µg/µL) was incubated under standard conditions (25 mM Hepes [pH 7.5], 70 µM GTP, 11 µM of each of the amino acids, 2 mM DTT, 60 µM PMSF and 0.5 mM Mg(OAc)2) with 0.9 mM ATP (unless indicated otherwise) in the presence of oxidized 32P-labeled CAT RNA (50,000 cpm). Reactions were incubated for 10 min at 30°C and then crosslinked using 20 mM NaBH3CN overnight at 4°C. After treatment with RNAse A, proteins were separated by 10% SDS-PAGE and visualized by autoradiography (Kodak X-Omat).

35S-methionine labeling and Western blotting

Malignant metastatic human mammary epithelial MDA-MB-231 cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD) and grown as monolayers in L15 medium (Invitrogen) supplemented with 10% fetal bovine serum and 100 U/ml penicillin/streptomycin. PC-3 cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD) and grown as monolayers in F12-K medium (Invitrogen) supplemented with 10% fetal bovine serum and 100 U/ml penicillin/streptomycin. HUVEC cells were obtained from Lonza (Walkersville, MD) and propagated in EBM-2 medium supplemented with EGM-2.

To measure the rate of 35S-Met incorporation into protein, 60,000 cells/well were seeded in a 24-well plate. The following day, the medium was removed, cells washed with PBS and exposed to silvestrol at the indicated concentrations in methionine-free DMEM supplemented with 10% dialyzed serum for 1 hr. For the last 15 min, cells were labeled with 35S-methionine. Medium was removed, cells washed in PBS and lyzed in RIPA buffer (20 mM Tris [pH 7.5], 100 mM NaCl, 1 mM EDTA, 1 mM EGTA, 0.1% NP-40, 0.5% sodium desoxycholate, 0.1% SDS, 20 mM ß-glycerophosphate, 10 mM NaF, 1 mM PMSF, 4 µg/ml aprotinin, 2 µg/ml leupeptin, 2 µg/ml pepstatin) for 20 min with shaking at 4°C. The protein was TCA precipitated and the radioactivity quantitated by scintillation counting. Protein content in the cell lysates was measured using the Bio-Rad DC ProteinAssay (Bio-Rad Laboratories) and used to standardize the counts obtained by TCA precipitation.

To visualize 35S-methionine labeled proteins, equal amounts of extracts were resolved on 10% SDS-polyacrlamide gels, stained with Coomassie Blue to verify equivalent loading, treated with En3Hance, dried and exposed to X-OMAT X-ray film (Kodak).

To monitor cellular viability, 200,000 cells/well were seeded in a 6-well plate and treated with 25 nM silvestrol for the indicated times. At the end of the treatment, the cell media was collected, cells were washed with 1 ml PBS and trypsinized with 200 µl trypsin. Cells were collected and pooled with the initial media and PBS wash. Samples were spun at 4°C for 2 min at 2000 rpm in a Sorval LegendRT table centrifuge. The pellet was resuspended in 2 ml cold PBS followed by another 2 min spin at 2000 rpm. The pellet was resuspended in 100 µl Annexin V binding buffer (10 mM HEPES [pH 7.4], 140 mM NaCl, 2.5 mM CaCl2) followed by the addition of propidium iodide to a final concentration of 5 µg/ml. After addition of 5 µl Annexin V-FITC (BD- Biosciences), samples were incubated for 15 min at RT in the dark followed by the addition of 400 µl of Annexin V binding buffer. FACS analyses were performed using a FACScan instrument from BD Biosciences and CELLQUEST software.

For Western blot analysis, cells were grown in 6-well plates, washed with PBS, harvested with a rubber policeman, and collected by brief centrifugation. Cell pellets were lyzed in RIPA buffer and separated on a 10% SDS-polyacrylamide gel followed by transfer to a PVDF membrane (Millipore). Primary Antibodies used were anti-Cyclin D1 (Cell Signaling, Danvers, MA), anti-Bcl-2 (Cell Signaling, Danvers, MA), anti-c-myc (Santa Cruz, Santa Cruz, CA), anti-Mcl-1 (Rockland, Gilbertsville, PA), anti-Survivin (Novus, Burlington, ON), and anti-tubulin (Sigma-Aldrich, Oakville, ON) and anti-GAPDH (Abcam, Cambridge, MA). Secondary antibodies were from Jackson Immuno Research (Burlington, ON).

Monitoring Stress Granule Formation

Anti-eIF4A and anti-G3BP antibodies have been previously described [49], [50]. Anti-eIF4E antibody was a gift of S. Kimball and has been described previously [51]. Cells were processed for immunofluorescence as previously described [50]. Essentially, cells were fixed in 3% paraformaldehyde and permeabilised with 0.1% Triton-X100/PBS. Slides were incubated with primary antibodies diluted in 0.1% normal goat serum for 1 h at RT. Following washing, the slides were incubated with goat anti-mouse/rabbit IgG (H+L) secondary antibodies coupled to goat Alexa Fluor 488/594. Fluorescence microscopy was performed using a Zeiss Axiovision 3.1 microscope equipped with Axiocam HR (Zeiss) digital camera. Images were compiled using Adobe Photoshop software.

Isolation of eIF4F from RSW and MDA-MB-231 cell extracts

RSW was incubated in the presence of 0.5% of DMSO or 50 µM silvestrol for 1 h at 30°C, followed by the addition of 50 µL of 50% m7GTP-Sepharose beads (G.E. Healthcare). The reactions were incubated for 2 h end-over-end at 4°C, after which the beads were washed three times with 240 µL of LCB (20 mM HEPES [7.5], 100 mM KCl, 0.2 mM EDTA), and twice with 240 µL of LCB+ 200 µM GDP. Proteins were eluted with 120 µl of LCB+ 200 µM m7GTP for 10 min on ice.

For eIF4F pulldown experiments from cell extracts, 8×106 MDA-MB-231 cells were seeded into 15 cm2 dishes and the next day treated with 25 nM silvestrol for 4 h. Cells were washed in cold PBS, scraped with a rubber policeman and spun down for 5 min at 2500 rpm. Cell pellets were resuspended in lysis buffer (20 mM Tris [7.5], 100 mM KCl, 1 mM DTT, 1 mM EDTA, 0.2% Tween20, 20 mM ß-glycerophosphate, 10 mM NaF, 1 mM PMSF, 4 µg/ml aprotinin, 2 µg/ml leupeptin, 2 µg/ml pepstatin) and immediately put on dry ice. Following 3 freeze-thaw cycles, extracts were centrifuged for 10 min at 14,000×g to remove cell debris. Pulldowns were performed with 1 mg of total protein extract and 50 µl of 50% m7GTP-Sepharose beads (GE Healthcare) for 2 h end-over-end at 4°C. Beads were washed 3 times with lysis buffer, once with LCB+ 500 µM GDP and proteins eluted with LCB+ 500 µM m7GTP for 10 min on ice. m7GTP elutions were analyzed on a 10% polyacrylamide gel, followed by transfer to a PVDF membrane.

For poly (rG) pulldown experiments, 500 µL of Krebs-2 extract was incubated with 0.5% DMSO, 10 µM pateamine or 50 µM silvestrol for 10 min at 4°C, followed by the addition of 50 µl of 50% Poly(rG) agarose (Sigma). Pulldowns were incubated for 1 h at 4°C. After incubation, the beads were washed twice with 10 volumes wash buffer (20 mM HEPES [7.5], 250 mM KOAc, 0.1% NP-40, 1 mM DTT). SDS elutions were resolved on a 10% SDS-polyacrylamide gel and transferred to PDVF membranes (Millipore). Antibodies used were anti-eIF4A [10] and anti-eIF4GI (Bethyl, Montgomery, TX), respectively. Secondary antibodies were from Jackson Immuno Research (Burlington, ON).

Drug Synergy Assessment

MDA-MB-231 cells were seeded at 10,000 cells/well in a 96- well plate. The cells were allowed to adhere to the bottom of the plates for 24 h in complete media and then exposed to drug or vehicle in fresh media for the indicated periods of time. At the end of treatment, cells were washed with PBS followed by addition of 200 µl PBS to each well. Cell proliferation was monitored using the sulforhodamine B (SRB) assay [52]. Drug interaction was assessed by the combination index method of Chou and Talalay using CalcuSyn software (BioSoft, Cambridge, UK) [53].

Animal Studies

For xenograft models, 5×106 MDA-MB-231 or PC-3 cells were injected with matrigel sub-cutaneously into the right flank of 4–6 weeks old female Balb/c nu/nu mice. Tumor growth was monitored every day using calipers. Treatments were started when tumors had reached 25–50 mm3. Nude mice bearing tumors were dosed I.P. with vehicle (5.2% PEG400/5.2% Tween80), silvestrol (0.5 mg/kg), doxorubicin (5 mg/kg), or rapamycin (4 mg/kg). Tumor growth was then monitored for the remainder of the experiment and no further injections were made. For fixation, tumors were fixed in 10% formalin and embedded in paraffin for further analysis. TUNEL staining was performed using the In Situ Cell Death Detection Kit, POD (Roche) following the manufacturer's protocol. For Ki67 staining, the rabbit monoclonal Ki67 (Clone SP6) antibody was purchased from Thermo Scientific. Antigen retrieval was performed by boiling samples for 15 min in 10 mM Citrate buffer [pH 6.5]. Sections were incubated with primary antibody for 1 h at RT. The Ultravision Detection System Anti Rabbit, HRP,DAB (Thermo Scientific) was used according to the manufacturer's instructions.

To assess cytotoxicity of silvestrol, six male wt Balb/c mice, 8 weeks old, were treated with vehicle (5.2% PEG 400/5.2% Tween-80) or 0.2 mg/kg silvestrol for 8 consecutive days. Fresh cell suspensions of bone marrow (BM) and spleen (SP) were prepared in PBS+ 2% FBS. Nonspecific binding was blocked by incubation of the samples with purified anti-CD16/CD32 antibody (clone: 2.4G2; BD Biosciences) for 5 mins on ice before labeling of the cells with a combination of fluorochrome conjugated substrate specific antibodies. Antibodies used to identify monocytes and granulocytes were: Ly-6G/Ly-6C (Gr-1) PECy7 (clone RB6-8C5; Biolegend) and CD11b PE (clone M1/70; BD Biosciences). Antibodies used to identify T and B lymphocytes were: CD4 FITC (clone RM4-5; BD Biosciences), CD8 CyChrome (clone 53-6.7; BD Biosciences) and CD45R/B220 APC (clone RA3-6B2; BD Biosciences). Red blood cells were identified with the antibody Ter119 Biotin (clone ter119; Biolegend) followed by a streptavidin Pacific blue conjugated antibody (Invitrogen/Molecular Probes). Incubation was performed in the dark on ice for 30 min before data acquistion and analysis was conducted on a FACSAria machine (BD Biosciences) by using CELLQUEST (BD Biosciences) or FlowJo (Treestar) softwares.

Polysome profiling analysis on liver extracts was performed on male wt Balb/c mice, 6–8 weeks old. Mice were injected with vehicle (5.2% PEG 400/5.2% Tween-80) or 0.2 mg/kg silvestrol for 2 days. Three and six hours after the second injection, animals were sacrificed, and the livers excised and washed in cold PBS containing 100 µg/ml cycloheximide. The dissected livers were homogenized in a Eurostar Power-b homogenizer (IKA Liver Labortechnik, Staufen, Germany) in 3 volumes of lysis buffer (40 mM HEPES [7.5], 100 mM KCl, 5 mM MgCl2, 100 µg/ml cycloheximide). Homogenates were then centrifuged for 10 min at 1,200×g at 4°C. The supernatant was transferred into a new tube and to 150 µl of extract, 300 µl of lysis buffer containing 0.5% Triton X-100 and 0.5% sodium deoxycholate were added and the sample centrifuged briefly. Samples were loaded on 10–50% sucrose gradients and centrifuged in an SW40 rotor at 35,000 rpm for 2 h. Gradients were analyzed by piercing the bottom of the tubes with a Brandel tube piercer and passing 60% sucrose through the bottom. Recording of the data was performed using InstaCal Version 5.70 and TracerDaq Version (Measurement Computing Corporation, Norton, MA).


While this manuscript was under review, Lucas et al. (Lucas, DM, Edwards, RB, Lozanski, G, et al. The novel plant-derived agent silvestrol has B-cell selective activity in chronic lymphocytic leukemia and acute lymphoblastic leukemia in vitro and in vivo. Blood, 2009; [Epub ahead of print]) also showed reduction of Mcl-1 protein following silvestrol treatment of acute lymphoblastic and chronic lymphocytic leukemias.

Supporting Information

Figure S1

Relative potency of cyclopenta b benzofurans on cap- and HCV-mediated translation initiation. A. Schematic diagram of FF/HCV/Ren mRNA reporter used to program in vitro translation extracts. B. Krebs-2 translation extracts were programmed with FF/HCV/Ren mRNA and vehicle (MeOH) or 50 µM compound. A representative autoradiograph of an in vitro translations performed in the presence of [35S]-methionine is shown. Following separation of protein products on 10% SDS-polyacrylamide gels, the gels were treated with EN3Hance, dried, and subjected to autoradiography. The position of migration of FF and Ren luciferase proteins is indicated to the right.

(1.55 MB EPS)

Figure S2

Translation properties of cyclopenta bc benzopyrans and benzo b oxepines. A. Chemical structure of cyclopenta bc benzopyrans and benzo b oxepines tested in this study. B. Effect of compounds on cap- and HCV-mediated translation initiation. Krebs-2 translation extracts were programmed with FF/HCV/Ren mRNA and vehicle (MeOH) or 50 µM compound added. A representative autoradiograph of an in vitro translation reaction performed in Krebs-2 extracts with [35S]-methionine is shown. Following separation of protein products by SDS-PAGE, the gels were treated with EN3Hance, dried, and subjected to autoradiography. The position of migration of FF and Ren luciferase proteins is indicated to the right.

(6.70 MB EPS)

Figure S3

Effect of silvestrol exposure on global protein synthesis in MDA-MB-231 and PC-3 cells. A. Cells were exposed to 25 nM silvestrol for the indicated time points and radiolabeled with 35S-Met for the last 15 min of incubation. Cell extracts were prepared and equivalent protein amounts were loaded on 10% SDS-polyacrylamide gels. After electrophoresis, gels were stained with Coomassie blue to verify equal loading, treated with En3Hance, dried, and exposed to X-OMAT X-ray film (Kodak). Open squares identified some proteins whose abundance was reduced after 24 h. B. Six million cells (MDA-MB-231) were seeded in 15 cm2 dishes 24 h before treatment with 25 nM silvestrol or vehicle (DMSO) for 1 h. Cells were harvested by scraping with a rubber policeman in cold PBS containing 100 µg/ml cycloheximide and centrifuged for 10 min at 2000×g at 4°C. Pellets were resuspended in 425 µl hypotonic lysis buffer (5 mM Tris 7.5, 2.5 mM MgCl2, 1.5 mM KCl) followed by addition of 5 µl 10 µg/ml cycloheximide, 1 µM DTT, 0.5% Triton X-100 and 0.5% sodium deoxycholate. Samples were loaded on 10–50% sucrose gradients and centrifuged in an SW40 rotor at 35,000 rpm for 2 h. Gradients were analyzed by piercing the bottom of the tubes with a Brandel tube piercer and passing 60% sucrose through the bottom. Fractions were collected from the gradients and monitored with an ISCO UA-6 UV detector (left panel). Fractions were separated into unbound and polysome (Poly)-bound regions. RNA was isolated using Trizol according to the manufacturer's instructions (Invitrogen). One-twentieth of the pooled RNA fractions was reverse transcribed using SuperScript™ II (Invitrogen) primed with oligo(dT)12–18. PCRs were performed with undiluted, 1∶10 or 1∶100 diluted cDNA samples. Primers used were hCyclin D1 [5′ CTCCTCTCCGGAGCATTTTGAT 3′ and 5′ CACCGCTCAGGGTTATGCAAAT 3′], hBcl-2 [5′ TGATGGGATCGTTGCC 3′ and 5′ CGCGGAACACTTGATT 3′], hSurvivin [5′ GGCCCAGTGTTTCTTCTGCTT 3′ and 5′ TTGACAGAAAGGAAAGCGCAAC 3′], hMcl-1 [5′ TTCAGCGACGGCGTAACAAACT 3′ and 5′ CCCATCCCAGCCTCTTTGTTTA 3′], and hc-Myc [5′ AAGAAATTCGAGCTGCTGCCCA 3′ and 5′ AACTCTGGTTCACCATGTCTCC 3′]. Annealing temperatures for hBcl-2, hSurvivin and hMcl-1 were 55°C and annealing temperatures for hCyclin D1 and hc-Myc were 59°C and 63°C, respectively. PCRs were optimized to detect the exponential phase of amplification and analyzed on 1.5% agarose gels (right panel).

(10.21 MB EPS)

Figure S4

Silvestrol inhibition of translation leads to mRNA discrimination. A. Schematic diagram of constructs used in this study. The renilla and firefly luciferase ORFs are represented by grey and blackened boxes, respectively. B. In vitro translation of mRNAs performed in Krebs extracts programmed with 0.2 µg/ml HCV/FF and 0.2 µg/ml of either [CAA]10/RL or G-Q(+6)/RL mRNA. The concentrations of hippuristanol, m7GDP, or GDP present in the reactions are indicated. Luciferase values were read on a Berthold Lumat LB 9507 luminometer. Renilla luciferase values were normalized to the firefly luciferase values and calculated relative to the vehicle control (set at 1). The values represent the average of 2-4 reactions with the error of the mean shown. Renilla luciferase values from the vehicle control were ∼20,000 and 150,000 RLU for translations programmed with G-Q(+6)/RL and [CAA]10/RL, respectively. Firefly luciferase values for vehicle-containing translations programmed with HCV/FF were ∼250,000 RLU. C. In vitro translation of mRNAs in the presence of increasing concentrations of silvestrol. Translations were performed in Krebs extracts programmed with 0.2 µg/ml HCV/FF and 0.2 µg/ml of either [CAA]10/RL or G-Q(+6)/RL mRNA. The concentrations of silvestrol present in the translation reactions are indicated. Luciferase values were read on a Berthold Lumat LB 9507 luminometer. Renilla luciferase values were normalized to the firefly luciferase values and calculated relative to the vehicle control (which was set at 1). The values represent the average of 2–4 reactions, with the error of the mean shown. The average firefly [HCV/FF] and renilla [G-Q(+6)/RL] values obtained with the vehicle controls were ∼320,000 and ∼16,000 RLU, respectively. Average values for renilla obtained from translation of [CAA]10/RL mRNA were ∼100,000 RLU.

(1.44 MB EPS)

Figure S5

mRNA discrimination by flavaglines is neither silvestrol- nor reporter- specific. A. Predicted secondary structure of PLTAR. B. Translation products of the indicated mRNAs from Krebs extracts containing KOAc concentrations of 75 mM (lanes 4 and 8), 100 mM (lanes 5 and 9), 125 mM (lanes 6 and 10), and 150 mM (lanes 7 and 11). The presence of vehicle (lanes 4–7) or 1 µM Compound 7 (lanes 8–11) is indicated. Fluorographs of the dried gels are presented.

(5.02 MB EPS)

Figure S6

Silvestrol induces the formation of SGs in HeLa cells. A. Distribution of eIF4A in HeLa cells upon exposure to arsenite or silvestrol. Cells were exposed to DMSO (0.5%) (top panel), arsenite (0.5 mM for 1 h) (middle panel), or silvestrol (5 µM for 1 h) (bottom panel), fixed and stained for eIF4A and G3BP. B. Distribution of eIF4E in HeLa cells upon exposure to arsenite or silvestrol.

(3.62 MB EPS)

Figure S7

Silvestrol does not cause weight loss over time. Eight Balb/c mice were administered silvestrol (0.2 mg/kg) on a daily basis for 8 days and their weights monitored at the indicated times.

(0.65 MB EPS)

Figure S8

Silvestrol suppresses tumor growth in a PC-3 prostate cancer xenograft model. Nude mice bearing human PC-3 prostate cancer (∼35 mm3) were dosed I.P. with vehicle (5.2% PEG400/5.2% Tween80), silvestrol (0.5 mg/kg), doxorubicin (5 mg/kg), or rapamycin (4 mg/kg). Dosing schedule is indicated at the start of Day 24.

(1.06 MB EPS)


We are grateful to Dr. Murray Tait (Cerylid Biosciences, Ltd.) for the generous supply of silvestrol.


Competing Interests: The authors have declared that no competing interests exist.

Funding: RC was supported by a CIHR Cancer Consortium Training Grant Award and a Cole Foundation Award. GG-V and MEB were supported by a CIHR Cancer Consortium Training Grant Award. AB was supported by a US Lymphoma Foundation Award. This work was supported by an NIH RO1 grant (GM073855) to J.A. Porco Jr. and an NCIC grant (#017099) and a Canadian Breast Cancer Research Alliance Translational Acceleration grant (#16512) to J. Pelletier. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.


1. Ohse T, Ohba S, Yamamoto T, Koyano T, Umezawa K. Cyclopentabenzofuran lignan protein synthesis inhibitors from Aglaia odorata. J Nat Prod. 1996;59:650–652. [PubMed]
2. Lee SK, Cui B, Mehta RR, Kinghorn AD, Pezzuto JM. Cytostatic mechanism and antitumor potential of novel 1H-cyclopenta[b]benzofuran lignans isolated from Aglaia elliptica. Chemico-Biol Interact. 1998;115:215–228. [PubMed]
3. Gerard B, Cencic R, Pelletier J, Porco JAJ. Enantioselective synthesis of the complex rocaglate (-)-silvestrol. Angew Chem Int Ed Engl. 2007;46:7831–7834. [PubMed]
4. Bordeleau ME, Robert F, Gerard B, Lindqvist L, Chen SM-H, et al. Therapeutic suppression of translation initiation modulates chemosensitivity in a mouse lymphoma model. J Clin Inv. 2008;118:1–11. [PMC free article] [PubMed]
5. Hwang BY, Su BN, Chai H, Mi Q, Kardono LB, et al. Silvestrol and episilvestrol, potential anticancer rocaglate derivatives from Aglaia silvestris. J Org Chem. 2004;69:3350–3358. [PubMed]
6. Meurer-Grimes BM, Yu J, Vairo GL. Therapeutic compounds and methods. USA: The Government of the State of Sarawak, Malaysia; 2004. p. 28.
7. Mi Q, Su BN, Chai H, Cordell GA, Farnsworth NR, et al. Rocaglaol induces apoptosis and cell cycle arrest in LNCaP cells. Anticancer Res. 2006;26:947–952. [PubMed]
8. Pestova TV, Kolupaeva VG, Lomakin IB, Pilipenko EV, Shatsky IN, et al. Molecular mechanisms of translation initiation in eukaryotes. Proc Natl Acad Sci U S A. 2001;98:7029–7036. [PMC free article] [PubMed]
9. Grifo JA, Tahara SM, Leis JP, Morgan MA, Shatkin AJ, et al. Characterization of eukaryotic initiation factor 4A, a protein involved in ATP-dependent binding of globin mRNA. J Biol Chem. 1982;257:5246–5252. [PubMed]
10. Edery I, Humbelin M, Darveau A, Lee KA, Milburn S, et al. Involvement of eukaryotic initiation factor 4A in the cap recognition process. J Biol Chem. 1983;258:11398–11403. [PubMed]
11. Rogers GW, Jr, Komar AA, Merrick WC. eIF4A: The godfather of the DEAD box helicases. Prog Nucleic Acid Res Mol Biol. 2002;72:307–331. [PubMed]
12. Raught B, Gingras A-C, Sonenberg N. In: Regulation of Ribosome Recruitment in Eukaryotes. Sonenberg N, Hershey JWB, Mathews MB, editors. Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 2000. pp. 245–293.
13. Dorrello NV, Peschiaroli A, Guardavaccaro D, Colburn NH, Sherman NE, et al. S6K1- and betaTRCP-mediated degradation of PDCD4 promotes protein translation and cell growth. Science. 2006;314:467–471. [PubMed]
14. De Benedetti A, Graff JR. eIF-4E expression and its role in malignancies and metastases. Oncogene. 2004;23:3189–3199. [PubMed]
15. Lazaris-Karatzas A, Montine KS, Sonenberg N. Malignant transformation by a eukaryotic initiation factor subunit that binds to mRNA 5′ cap. Nature. 1990;345:544–547. [PubMed]
16. Wendel H-G, de Stanchina E, Fridman JS, Malina A, Ray S, et al. Survival signalling by Akt and eIF4E in oncogenesis and cancer therapy. Nature. 2004;428:332–337. [PubMed]
17. Graff JR, Konicek BW, Carter JH, Marcusson EG. Targeting the eukaryotic translation initiation factor 4E for cancer therapy. Cancer Res. 2008;68:631–634. [PubMed]
18. Jiang L, Suri AK, Fiala R, Patel DJ. Saccharide-RNA recognition in an aminoglycoside antibiotic-RNA aptamer complex. Chem Biol. 1997;4:35–50. [PubMed]
19. Greenberg JR. The polyribosomal mRNA–protein complex is a dynamic structure. Proc Natl Acad Sci U S A. 1981;78:2923–2926. [PMC free article] [PubMed]
20. Bordeleau ME, Cencic R, Lindqvist L, Oberer M, Northcote P, et al. RNA-mediated sequestration of the RNA helicase eIF4A by Pateamine A inhibits translation initiation. Chem Biol. 2006;13:1287–1295. [PubMed]
21. Lindqvist L, Oberer M, Reibarkh M, Cencic R, Bordeleau M-E, et al. Selective Pharmacological Targetting of a DEAD box RNA Helicase. PLoS ONE. 2008;3:e1583. [PMC free article] [PubMed]
22. Sonenberg N. ATP/Mg++-dependent cross-linking of cap binding proteins to the 5′ end of eukaryotic mRNA. Nucleic Acids Res. 1981;9:1643–1656. [PMC free article] [PubMed]
23. Kumari S, Bugaut A, Huppert J, Balasubramanian S. An RNA G-quadruplex in the 5′ UTR of the NRAS proto-oncogene modulates translation. Nat Chem Biol. 2007;3:218–221. [PMC free article] [PubMed]
24. Pestova TV, Kolupaeva VG. The roles of individual eukaryotic translation initiation factors in ribosomal scanning and initiation codon selection. Genes Dev. 2002;16:2906–2922. [PMC free article] [PubMed]
25. Pestova TV, Shatsky IN, Fletcher SP, Jackson RJ, Hellen CU. A prokaryotic-like mode of cytoplasmic eukaryotic ribosome binding to the initiation codon during internal translation initiation of hepatitis C and classical swine fever virus RNAs. Genes Dev. 1998;12:67–83. [PMC free article] [PubMed]
26. Bordeleau M-E, Mori A, Oberer M, Lindqvist L, Chard LS, et al. Functional Characterization of IRESes by an inhibitor of the RNA helicase eIF4A. Nat Chem Biol. 2006;2:213–220. [PubMed]
27. Parkin NT, Cohen EA, Darveau A, Rosen C, Haseltine W, et al. Mutational analysis of the 5′ non-coding region of human immunodeficiency virus type 1: effects of secondary structure on translation. EMBO J. 1988;7:2831–2837. [PMC free article] [PubMed]
28. Avdulov S, Li S, Michalek V, Burrichter D, Peterson M, et al. Activation of translation complex eIF4F is essential for the genesis and maintenance of the malignant phenotype in human mammary epithelial cells. Cancer Cell. 2004;5:553–563. [PubMed]
29. Majumder PK, Sellers WR. Akt-regulated pathways in prostate cancer. Oncogene. 2005;24:7465–7474. [PubMed]
30. Graff JR, Konicek BW, McNulty AM, Wang Z, Houck K, et al. Increased AKT activity contributes to prostate cancer progression by dramatically accelerating prostate tumor growth and diminishing p27Kip1 expression. J Biol Chem. 2000;275:24500–24505. [PubMed]
31. Rajasekhar VK, Viale A, Socci ND, Wiedmann M, Hu X, et al. Oncogenic Ras and Akt signaling contribute to glioblastoma formation by differential recruitment of existing mRNAs to polysomes. Mol Cell. 2003;12:889–901. [PubMed]
32. Mamane Y, Petroulakis E, Martineau Y, Sato TA, Larsson O, et al. Epigenetic activation of a subset of mRNAs by eIF4E explains its effects on cell proliferation. PLoS ONE. 2007;2:e242. [PMC free article] [PubMed]
33. Wendel HG, Malina A, Zhao Z, Zender L, Kogan SC, et al. Determinants of sensitivity and resistance to rapamycin-chemotherapy drug combinations in vivo. Cancer Res. 2006;66:7639–7646. [PubMed]
34. Jones S, Zhang X, Parsons DW, Lin JC, Leary RJ, et al. Core signaling pathways in human pancreatic cancers revealed by global genomic analyses. Science. 2008;321:1801–1806. [PMC free article] [PubMed]
35. Graff JR, Zimmer SG. Translational control and metastatic progression: Enhanced activity of the mRNA cap-binding protein eIF-4E selectively enhances translation of metastasis-related mRNAs. Clin Exp Met. 2003;20:265–273. [PubMed]
36. Mavrakis KJ, Zhu H, Silva RL, Mills JR, Teruya-Feldstein J, et al. Tumorigenic activity and therapeutic inhibition of Rheb GTPase. Genes Dev. 2008;22:2178–2188. [PMC free article] [PubMed]
37. Polunovsky VA, Rosenwald IB, Tan AT, White J, Chiang L, et al. Translational control of programmed cell death: eukaryotic translation initiation factor 4E blocks apoptosis in growth-factor-restricted fibroblasts with physiologically expressed or deregulated Myc. Mol Cell Biol. 1996;16:6573–6581. [PMC free article] [PubMed]
38. Maeshima Y, Sudhakar A, Lively JC, Ueki K, Kharbanda S, et al. Tumstatin, an endothelial cell-specific inhibitor of protein synthesis. Science. 2002;295:140–143. [PubMed]
39. Morley SJ, McKendrick L. Involvement of stress-activated protein kinase and p38/RK mitogen-activated protein kinase signaling pathways in the enhanced phosphorylation of initiation factor 4E in NIH 3T3 cells. J Biol Chem. 1997;272:17887–17893. [PubMed]
40. Tee AR, Proud CG. DNA-damaging agents cause inactivation of translational regulators linked to mTOR signalling. Oncogene. 2000;19:3021–3031. [PubMed]
41. Mungamuri SK, Yang X, Thor AD, Somasundaram K. Survival signaling by Notch1: mammalian target of rapamycin (mTOR)-dependent inhibition of p53. Cancer Res. 2006;66:4715–4724. [PubMed]
42. Kim S, Salim AA, Swanson SM, Kinghorn AD. Potential of cyclopenta[b]benzofurans from Aglaia species in cancer chemotherapy. Anticancer Agents Med Chem. 2006;6:319–345. [PubMed]
43. Anthony TG, Anthony JC, Yoshizawa F, Kimball SR, Jefferson LS. Oral administration of leucine stimulates ribosomal protein mRNA translation but not global rates of protein synthesis in the liver of rats. J Nutr. 2001;131:1171–1176. [PubMed]
44. Lin C-J, Cencic R, Mills JR, Robert F, Pelletier J. c-Myc and eIF4F are Components of a Feedforward Loop that Link Transcription and Translation. Cancer Res. 2008;68:5326–5334. [PubMed]
45. Mills JR, Hippo Y, Robert F, Chen SMH, Malina A, et al. mTORC1 promotes survival through translational control of Mcl-1. Proc Natl Acad Sci U S A. 2008;105:10853–10858. [PMC free article] [PubMed]
46. Novac O, Guenier AS, Pelletier J. Inhibitors of protein synthesis identified by a high throughput multiplexed translation screen. Nucleic Acids Res. 2004;32:902–915. [PMC free article] [PubMed]
47. Svitkin YV, Pause A, Haghighat A, Pyronnet S, Witherell G, et al. The requirement for eukaryotic initiation factor 4A (elF4A) in translation is in direct proportion to the degree of mRNA 5′ secondary structure. RNA. 2001;7:382–394. [PMC free article] [PubMed]
48. Merrick WC, Sonenberg N. Assays for eukaryotic translation factors that bind mRNA. Methods. 1997;11:333–342. [PubMed]
49. Mazroui R, Huot ME, Tremblay S, Filion C, Labelle Y, et al. Trapping of messenger RNA by Fragile X Mental Retardation protein into cytoplasmic granules induces translation repression. Hum Mol Genet. 2002;11:3007–3017. [PubMed]
50. Mazroui R, Sukarieh R, Bordeleau ME, Kaufman RJ, Northcote P, et al. Inhibition of ribosome recruitment induces stress granule formation independently of eukaryotic initiation factor 2alpha phosphorylation. Mol Biol Cell. 2006;17:4212–4219. [PMC free article] [PubMed]
51. Kimball SR, Horetsky RL, Ron D, Jefferson LS, Harding HP. Mammalian stress granules represent sites of accumulation of stalled translation initiation complexes. Am J Physiol Cell Physiol. 2003;284:C273–284. [PubMed]
52. Skehan P, Storeng R, Scudiero D, Monks A, McMahon J, et al. New colorimetric cytotoxicity assay for anticancer-drug screening. J Natl Canc Inst. 1990;82:1107–1112. [PubMed]
53. Chou T, Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multiple durgs or enzyme inhibitors. Adv Enzyme Regul. 1984;22:27–55. [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem chemical compound records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records. Multiple substance records may contribute to the PubChem compound record.
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem chemical substance records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records.

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...