|Home | About | Journals | Submit | Contact Us | Français|
N-(4-hydroxyphenyl) retinamide (4HPR, fenretinide) is a synthetic retinoid that has been tested in clinical trials as a cancer therapeutic and chemopreventive agent. Although 4HPR has been shown to be cytotoxic to many kinds of cancer cells, the underlying molecular mechanisms are only partially understood. Until now, no direct cancer-related molecular target has been reported to be involved in the antitumor activities of 4HPR. Herein, we found that 4HPR inhibited mammalian target of rapamycin (mTOR) kinase activity by directly binding with mTOR, which suppressed the activities of both the mTORC1 and the mTORC2 complexes. The predicted binding mode of 4HPR with mTOR was based on a homology computer model, which showed that 4HPR could bind in the ATP-binding pocket of the mTOR protein through hydrogen bonds and hydrophobic interactions. In vitro studies also showed that 4HPR attenuated mTOR downstream signaling in a panel of non-small-cell lung cancer cells, resulting in growth inhibition. Moreover, knockdown of mTOR in cancer cells decreased their sensitivity to 4HPR. Results of an in vivo study demonstrated that i.p. injection of 4HPR in A549 lung tumor-bearing mice effectively suppressed cancer growth. The expression of mTOR downstream signaling molecules in tumor tissues was also decreased after 4HPR treatment. Taken together, our results are the first to identify mTOR as a direct antitumor target of 4HPR both in vitro and in vivo, providing a valuable rationale for guiding the clinical uses of 4HPR.
N-(4-hydroxyphenyl) retinamide (4HPR), also known as fenretinib, is a synthetic retinoid that has been widely tested in clinical trials as a cancer therapeutic and chemopreventive agent (1). 4HPR has been shown to inhibit carcinogenesis in a variety of cancer cells, including breast cancer (2), bladder cancer (3), lung cancer (4), prostate cancer (5) and leukemia (6–7). Clinical trials have shown that 4HPR induces a significant reduction of secondary breast cancers in premenopausal women (8). However, the mechanisms of the antitumor activity of 4HPR have not been fully elucidated. Previous studies demonstrated that induction of apoptosis is a key mechanism of 4HPR to inhibit tumor growth. 4HPR could induce apoptosis of cancer cells both in retinoic acid receptor-dependent and -independent manners (9–10). The activation of c-Jun N-terminal kinases and the mitochondrial apoptotic pathway through the generation of reactive oxygen species was reported to be involved in 4HPR-induced apoptosis (11–12). Moreover, studies also showed that anti-angiogenic effects mediated through vascular endothelial growth factor receptor (VEGFR) (13–14) and inhibition of tumor invasion by interfering with Matrix metalloproteinase (MMP) (15–16) also underlie the antitumor activity of 4HPR. Several molecules in different signaling transduction pathways, such as FAK/Akt/GSK3β, have also been reported to be involved in the antitumor activity of 4HPR (5,17). Recently, Rahmaniyan and colleagues identified dihydroceramide desaturase, an enzyme that is responsible for inserting the 4,5-trans-double bond into the sphingolipid backbone of dihydroceramide, as a direct in vitro target of 4HPR (18). However, a direct antitumor target of 4HPR has not yet been identified in cells or in vivo.
The mammalian target of rapamycin (mTOR) is a major component of the PI3-K/Akt/mTOR pathway. It is an evolutionarily conserved serine/threonine kinase and functions as a sensor of mitogen, energy and nutrient levels and is a central controller of cell growth and division (19). The PI3-K/Akt/mTOR pathway is deregulated in 50% of all human malignancies, and therefore, inhibition of mTOR is a promising strategy for the treatment of human cancers. mTOR has two functionally distinct multi-protein complexes, mTOR complex 1 (mTORC1) and mTORC2. mTORC1 contains Raptor and PRAS40 and regulates protein translation through phosphorylation of p70 ribosomal S6 kinase (p70S6K) and eukaryotic translation initiation factor binding protein (4E-BP) (20). mTORC2 contains Rictor and Protor and phosphorylates Akt on Ser473, thereby increasing Akt enzymatic activity (21–22). mTOR inhibitors are currently being developed as potential antitumor agents. Rapamycin and its derivatives (referred to as rapalogs) are the most well-characterized mTOR inhibitors. The rapamycins are allosteric inhibitors that, in complex with FKBP12, target the FKB domain of mTOR (23). They partially inhibit mTOR through allosteric binding to mTORC1, but not mTORC2 (24). However, inhibiting only mTORC1 may not be sufficient for achieving a broad and robust anticancer effect due to a failure to inhibit mTORC2 in some tumor types. A strong interest now exists in developing small-molecule mTOR kinase inhibitors, which target both mTORC1 and mTORC2.
In the present study, we report for the first time that mTOR is a direct antitumor target of 4HPR. 4HPR effectively targets both mTORC1 and mTORC2 by directly binding to mTOR, resulting in the inhibition of tumor growth both in cells and in vivo.
4HPR was obtained from the National Institutes of Health (NIH). Rapamycin was purchased from LC Laboratories (Woburn, MA). Recombinant active mTOR (1362-end) was purchased from Millipore (Billerica, MA). The inactive p70S6K protein was from SignalChem (Richmond, BC, CANADA) and CNBr-Sepharose 4B was purchased from GE Healthcare (Pittsburgh, PA).
All cell lines were purchased from American Type Culture Collection and were cultured in monolayers at 37°C in a 5% CO2 incubator according to American Type Culture Collection protocols. For transfection experiments, the jetPEI (Qbiogen, Inc.) transfection reagent was used following the manufacturer’s instructions.
Tumor cells were suspended in Basal Medium Eagle medium and added to 0.6% agar, with different concentrations of 4HPR in a base layer and a top layer of 0.3 % agar. For JB6 Cl41 cells, the procedure is similar, except that these cells were exposed to Epidermal growth factor (EGF) (20ng/ml) during treatment with 4HPR or vehicle. The cultures were maintained at 37°C in a 5% CO2 incubator for 1–2 weeks and then colonies were counted under a microscope using the Image-Pro Plus software (v.4) program (Media Cybernetics, Silver Spring, MD).
To estimate cytotoxicity, cells were seeded (8×103 cells per well) in 96-well plates and cultured overnight. Cells were then fed with fresh medium and treated with different doses of 4HPR. After culturing for various times, the cytotoxicity of 4HPR was measured using an MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2H-tetrazdium) assay kit (Promega, Madison, WI) according to the manufacturer’s instructions.
The three-dimensional structure of mTOR was obtained from the SWISS-MODEL Repository, which is a homology model based on the crystal structure of PI3K-delta (PDB id 2WXG). Protein–ligand docking was performed using the high-performance hierarchical docking algorithm, Glide. The final binding model structure of mTOR-4HPR was generated from Schrodinger Induced Fit Docking, which merges the predictive power of prime with the docking and scoring capabilities of Glide for accommodating the possible protein conformational change upon ligand binding.
Proteins were resolved by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes (Millipore, Billerica, MA), which were blocked with milk and hybridized with specific primary antibodies. The protein bands were visualized using an enhanced chemiluminescence reagent (GE Healthcare, Pittsburgh, PA) after hybridization with a horseradish peroxidase-conjugated secondary antibody.
Inactive p70S6K (1 µg) or inactive Akt1 (1 µg) proteins were used as the substrate, respectively, for an in vitro kinase assay with 250ng of active mTOR (1362-end). Reactions were carried out in 1 × kinase buffer (25mM Tris-HCl pH 7.5, 5mM beta-glycerophosphate, 2mM dithiothreitol, 0.1mM Na3VO4, 10mM MgCl2 and 5mM MnCl2) containing 100 µM ATP at 30°C for 30min. Reactions were stopped and proteins detected by western blotting.
The mTOR complexes mTORC1 and mTORC2 were immunoprecipitated with a polyclonal rictor or polyclonal raptor antibody, followed by western blotting to detect mTOR and raptor or rictor, as described previously (25).
Recombinant human mTOR (1362-end) (200ng) or cell lysates (1mg) were incubated with 4HPR-Sepharose 4B beads (or Sepharose 4B beads alone as a control) (100 µl, 50% slurry) in the reaction buffer [50mM Tris (pH 7.5), 5mM ethylenediaminetetraacetic acid, 150mM NaCl, 1mM dithiothreitol, 0.01% Nonidet P-40, 2 µg/ml bovine serum albumin, 0.02mM phenylmethylsulfonyl fluoride and 1 µg/ml protease inhibitor mixture]. After incubation with gentle rocking overnight at 4°C, the beads were washed five times and proteins bound to the beads were analyzed using western blotting.
Athymic nude mice [Cr:NIH (S), NIH Swiss nude, 6- to 9-week old] were obtained from Harlan Laboratories and maintained under ‘specific pathogen-free’ conditions based on the guidelines established by the University of Minnesota Institutional Animal Care and Use Committee. Mice were divided into different groups (n = 10 of each group). A549 lung cancer cells (4×106/0.1ml) were injected subcutaneously into the right flank of each mouse. 4HPR was freshly prepared once a week and protected from light and kept at 4°C as described previously (26–27). 4HPR or vehicle was administered by i.p. injection three times a week for 29 days. Tumor volumes and body weights were measured. Tumor tissues from mice were embedded in a parafﬁn block and subjected to immunohistochemistry or hematoxylin and eosin (H&E) staining.
All quantitative data are expressed as mean values ± standard deviation, and significant differences were determined by Student’s t test or by one-way ANOVA. A probability value of P < 0.05 was used as the criterion for statistical significance.
In the present study, we first examined the effect of 4HPR (Figure 1A) on EGF-induced neoplastic transformation of JB6 Cl41 cells. Treatment of JB6 Cl41 cells with 4HPR significantly inhibited EGF-promoted neoplastic transformation in a dose-dependent manner (Figure 1B). 4HPR at 10 or 20 µM caused a decrease to 46 or 89% of control, respectively. The inhibition of colony formation by 4HPR was not due to cytotoxicity because the effective concentration range for suppressing cell transformation did not affect JB6 Cl41 cell viability (Figure 1C). Because anchorage-independent growth ability is an ex vivo indicator and a key characteristic of the transformed cell phenotype (28), these results suggest that 4HPR can reduce the malignant potential of JB6 Cl41 cells induced by EGF.
To identify a potential target of 4HPR, we examined several key-signaling molecules, including those in the RAS/RAF/MEK and PI3-K/Akt/mTOR pathways, which are frequently deregulated in human malignancies. Western blot results showed that 4HPR suppressed the phosphorylation of p70S6K and S6 ribosomal protein (S6) in a dose-dependent manner (Figure 1D) but had no inhibitory effect on the phosphorylation of other molecules, including EGFR, RAF, MEK and ERKs (data not shown).
The above results showed that the phosphorylation of p70S6K (T389), a direct downstream target of the PI3-K/Akt/mTOR pathway, was inhibited after 4HPR treatment, indicating that upstream molecules, such as PI3-K, Akt and mTOR, might be important targets of 4HPR. To this end, we first examined the effects of 4HPR on Akt1/2 or PI3-Kα/β in vitro kinase activity, but only weak inhibition was observed even at the highest concentration of 4HPR (20 µM) (data not shown). This raised the possibility that mTOR might be an important target of 4HPR. We performed an in vitro kinase assay with recombinant mTOR in the presence of various concentrations of 4HPR without FKBP12. Wortmannin, a well-known inhibitor of both PI3-K and mTOR, was used as a positive control in this assay. The phosphorylation of p70S6K (Thr389), an mTOR substrate, was strongly inhibited by 4HPR in a concentration-dependent manner (Figure 2A). For example, 1 µM 4HPR caused a 43% inhibition of mTOR kinase activity and 20 µM 4HPR resulted in an 88% inhibition. Moreover, we also used inactive Akt1 as a substrate for mTOR and results indicated that the phosphorylation of Akt on Ser473 was also dose-dependently suppressed by 4HPR treatment (Figure 2B). These results clearly support our hypothesis that mTOR is a direct target of 4HPR and, notably, the activity of both mTORC1 and mTORC2 might be inhibited by 4HPR.
We then constructed a possible binding model of 4HPR with the mTOR protein using molecular docking. The three-dimensional structure of mTOR was obtained from the SWISS-MODEL Repository (29). The model is a homology model based on the crystal structure of PI3K-delta (PDB id 2WXG) (30). The binding pose of 4HPR-mTOR obtained from the docking result (Figure 2C) suggested that the binding of 4HPR to mTOR extended from the ATP site into the neighboring hydrophobic pocket. 4HPR formed three hydrogen bonds with the protein residues in the ATP-binding pocket—two with Ser2165 and the other with Lys2187. In addition, several residues, including Tyr2225, Val2240, Met2345, Leu2354 and Ile2356 showed strong hydrophobic interactions with the major retinal part of 4HPR.
The docking results indicated that 4HPR formed a good interaction with the mTOR active site, which might be the basis of the direct binding of 4HPR to mTOR. To confirm this idea, we performed an in vitro binding assay using 4HPR-conjugated Sepharose 4B beads or control Sepharose 4B beads. No obvious band was observed when the mTOR protein was incubated with Sepharose 4B beads, whereas a strong band was seen when mTOR was incubated with 4HPR-conjugated Sepharose 4B beads (Figure 2D), clearly indicating that 4HPR directly binds to recombinant mTOR. We then performed an ex vivo pull-down assay using A549 cells, and similar results were also obtained with an A549 lung cancer cell lysate (Figure 2E). Therefore, these results indicated that 4HPR binds directly to mTOR and inhibits mTOR kinase activity.
Previous studies suggested that the mTOR signaling pathway is highly activated in human lung cancer (31–34). Therefore, we examined the effect of 4HPR in a panel of non-small-cell lung cancer cells. First, we investigated the effects of 4HPR on anchorage-independent growth of different types of human lung cancer cells. The results showed that 4HPR significantly inhibited A549 cell growth in soft agar in a concentration-dependent manner (Figure 3A). Colony formation was inhibited by more than 30% after treatment with 4HPR at a concentration of 5 µM, and almost no colonies were formed at 20 µM (Figure 3A). Moreover, we also examined the effect of 4HPR on the growth of several other lung cancer cell lines, including H520, H1650 and HCC827. Results showed that 4HPR dose dependently inhibited the growth of each cell line on soft agar (Figure 3B–3D).
We then investigated the effect of 4HPR on downstream targets of mTOR, including the phosphorylation of p70S6K and S6, as well as phosphorylation of Akt, in both A549 cells and H520 cells, which were relatively more sensitive to 4HPR (Figure 3A–3D). Western blot results showed that in both of the cell lines, mTOR-mediated phosphorylation of p70S6K (T389), S6 (S235, 236), as well as Akt (S473), was substantially decreased dose dependently with 4HPR treatment (Figures 3E and and3F3F).
Next, we determined the effect of 4HPR on mTORC1 and mTORC2 complexes using immunoprecipitation. Results indicate that both raptor (Figure 4A) and rictor (Figure 4B), members of the mTORC1 and mTORC2 complexes, were decreased with 4HPR treatment in A549 cells. In contrast, the control compound rapamycin only suppressed raptor in mTORC1 but not rictor in mTORC2 at a concentration of 10nM, which is consistent with a previous report (24). These results indicated that 4HPR can target both the mTORC1 and mTORC2 complexes.
We then examined whether knocking down mTOR expression influences the sensitivity of A549 cancer cells to 4HPR. First, we determined the efficiency of shRNA knockdown, as well as the effect of shRNA transfection on anchorage-independent growth. The expression of mTOR was obviously decreased after shRNA transfection (Figure 5A). Moreover, the growth of cells on soft agar also decreased more than 30% after transfection compared with the mock group (Figure 5B). Next, A549 cells transfected with mTOR shRNA or mock control were treated with 4HPR or vehicle and subjected to a soft agar assay. The results showed that 4HPR (10 µM) inhibited anchorage-independent growth of A549 cells transfected with mock shRNA by about 55%. In contrast, the inhibition was only about 37% in A549 cells transfected with mTOR shRNA, indicating that A549 cells transfected with mTOR shRNA were more resistant to 4HPR treatment (Figure 5C). These results suggested that mTOR plays an important role in the sensitivity of A549 cells to the antiproliferative effects of 4HPR.
To explore the antitumor activity of 4HPR in vivo, A549 cancer cells were injected into the right flank of individual athymic nude mice. Mice were then administered vehicle or 4HPR by i.p. injection of 10 or 40mg/kg three times a week for 29 days. The results showed that treatment of mice with 10 or 40mg/kg BW of 4HPR significantly suppressed A549 tumor growth by 52 and 73%, respectively, relative to the vehicle-treated group (Figure 6A, P < 0.01). Moreover, mice seemed to tolerate treatment with these doses of 4HPR without overt signs of toxicity or significant loss of body weight compared with vehicle-treated group (Figure 6B). The effect of 4HPR on mTOR protein targets was evaluated by immunohistochemistry and H&E staining of A549 tumor tissues after 29 days of treatment. The expression of phosphorylated p70S6K, a direct target of mTOR, was markedly decreased after treatment with 4HPR at either 10 or 40mg/kg B.W. (Figure 6C and and6D). In6D). In addition, the phosphorylation of S6 on S235,236 and the phosphorylation of Akt on S473 in tumor tissues was also strongly suppressed by 4HPR at 10 or 40mg/kg B.W. (Figure 6E). These data indicated that A549 lung tumor development was suppressed by 4HPR through inhibition of the both the mTORC1 and mTORC2 signaling pathways.
4HPR is a highly active and promising therapeutic and chemo preventive agent (1). However, the underlying mechanism explaining its
anticancer activity has not yet been elucidated. The present study identified mTOR as a
direct target of 4HPR both in vitro and
Previous reports revealed that the PI3-K/Akt/mTOR pathway may be
involved in the antitumor activity of 4HPR, and the phosphorylation of Akt (Ser473) is
reportedly involved in 4HPR-mediated apoptosis (5,17). Our results indicated that 4HPR
effectively suppressed EGF-induced transformation of JB6 Cl41 cells that was accompanied by
decreased phosphorylation of Akt (Ser473) and p70S6K (Thr389) (Figure 1D). Thus, the PI3-K/Akt/mTOR pathway is likely to play an important role
in the antitumor activity of 4HPR. We first determined whether 4HPR could affect the kinase
activity of Akt1/2 or PI3-Kα/β using an in vitro kinase
assay. However, only a weak inhibition was observed even at the highest concentration
(20 µM) (data not shown). These results indicated that Akt and PI3-K are probably not the major or direct targets of 4HPR, because the plasma concentration of 4HPR reportedly ranges from 0.7 to 10 µM at steady state (35). Previous studies have shown that phosphorylated p70S6K (Thr389) and Akt (Ser473) are downstream targets of mTORC1 and mTORC2, respectively (21,36). Therefore, we hypothesized that mTOR, which is located downstream in this pathway, might be an important and potential direct target of 4HPR. Thus, we determined the effect of 4HPR on mTOR in vitro kinase activity. Consistent with our idea, the results clearly showed that 4HPR could potently suppress mTOR activity in vitro in a concentration-dependent manner. This compound also reduced the phosphorylation of mTOR downstream molecules in cancer cells resulting in the inhibition of growth of several types of human lung cancer cells. In addition, knocking down mTOR expression decreased the sensitivity of A549 cells to 4HPR treatment. Moreover, results of an in vivo study using a xenograft mouse model further confirmed that 4HPR inhibited mTOR’s protein targets in tumor tissues resulting in inhibition of tumor growth in vivo. Overall, these results clearly demonstrated that mTOR is a direct and important antitumor target of 4HPR.
Accumulating studies have demonstrated that apoptosis is a key mechanism of 4HPR antitumor activity. However, in the present study, no obvious apoptosis was observed in JB6 C141 cells and A549 cells after exposure to 4HPR for 48h at the concentration (20 µM or less) that required for mTOR inhibition and anchorage-independent growth inhibition (data not shown). In addition, we also examined the apoptosis-inducing activity of 4HPR in leukemia cells and breast cancer cells, and results demonstrated that the sensitivity of these cells to 4HPR varied obviously (data not shown). Therefore, these results indicated that 4HPR might induce apoptosis in a cell-line-dependent manner and mTOR might play more important role in the growth inhibitory activity of 4HPR than that in apoptosis induction, at least in the cells tested in the present study. Our molecular docking results showed that 4HPR might bind to the ATP-binding pocket of mTOR through hydrogen bonding with Ser2165 and Lys2187 and hydrophobic interactions with several amino acid residues, including Tyr2225, Val2240, Met2345, Leu2354 and Ile2356. A previous report (37) showed that most of these amino acid residues are important for the binding between mTOR and ATP or mTOR inhibitors. Thus, the predicted binding model between 4HPR and mTOR is very similar to that of other reported mTOR inhibitors. Subsequently, we performed an in vitro binding and ex vivo pull-down assays to determine whether 4HPR can bind directly with mTOR. The results confirmed the direct binding between 4HPR and the mTOR protein. Therefore, these results demonstrated that 4HPR inhibited mTOR kinase activity through its direct binding to the mTOR protein.
Rapamycin and its analogues have shown that mTOR is an attractive target in cancer. mTORC1 is sensitive to the selective inhibitor, rapamycin, and is activated by growth factor stimulation by the canonical PI3-K/Akt/mTOR pathway. However, mTORC2 is not rapamycin sensitive, mainly due to Akt activation by disruption of a negative feedback loop (38). The negative feedback, which is dependent on IGF1R/insulin receptor substrate 1, involves the S6K-mediated suppression of upstream signaling. mTORC1 inhibitors abrogate this feedback suppression, resulting in Akt activation. Indeed, Cloughesy and colleagues (39) reported that rapamycin treatment induces an increase in phosphorylation of Akt in a subset of patients with phosphatase and tension homolog (PTEN)-deficient glioblastoma. Therefore, the use of mTORC1 inhibitors risks the possible activation of Akt by disruption of a negative feedback loop, and patients treated with these drugs eventually become refractory. In the present study, we demonstrated that 4HPR was able to inhibit the binding of mTOR with both rictor of mTORC2 and raptor of mTORC1, suggesting its ability to silence the rictor-dependent positive feedback loop on Akt activation. Consistent with this point, our results confirmed that 4HPR can inhibit the phosphorylation of Akt (Ser473) at a dose that suppressed the phosphorylation of both p70S6K and S6, in different types of lung cancer cell lines. In contrast, rictor downregulation was not achieved in A549 cells treated with rapamycin at a concentration that can dramatically inhibit p70S6K phosphorylation and S6 phosphorylation, which is consistent with previous reports. These data provide strong evidence that 4HPR is an effective inhibitor that targets both mTORC1 and mTORC2, possessing an advantage over rapalogues in Akt inhibition.
Taken together, the results of present study identify mTOR as a direct and important target of 4HPR, which offers useful evidence for the rational use and combinational treatment of 4HPR in cancer therapy. Meanwhile, we cannot exclude the possibility that 4HPR treatment, besides decreasing mTOR activity, leads to antitumor activity by affecting other pathways, which have been reported previously by several groups (5,11–13,15). Moreover, in order to further validate the functional role of the binding between 4HPR and mTOR, experiments utilizing overexpression of wild type and mTOR mutant cancer cells will be performed in the near future. Numerous preclinical studies and clinical trials of 4HPR are still ongoing in different countries (1), and more work needs to be performed in the future to provide more valuable evidence for guiding the clinical use of 4HPR in cancer prevention and therapy (34).
The Hormel Foundation and National Institutes of Health (CA027502, CA120388, R37 CA081064 and ES016548) and NCI Contract Number HHSN-261200533001C-NO1-CN-53301 and N01-CN-43309-18018-01WA 13B.
We would like to thank Tonya Poorman for her help in submitting our manuscript.