Deciphering molecular targets to enhance sensitivity to chemotherapy is becoming a priority for effectively treating cancers. Loss of function mutations of SMAD4 in colon cancer are associated with metastatic progression and resistance to 5-fluorouracil (5-FU), the most extensively used drug of almost all chemotherapy combinations used in the treatment of metastatic colon cancer. Here, we report that SMAD4 deficiency also confers resistance to irinotecan, another common chemotherapeutic frequently used alone or in combination with 5-FU against colon cancer. Mechanistically, we find that SMAD4 interacts with and inhibits RICTOR, a component of the mTORC2 complex, resulting in suppression of downstream effector phosphorylation of AKT at Serine 473. In silico meta-analysis of publicly available gene expression datasets derived from tumors indicates that lower levels of SMAD4 or higher levels of RICTOR/AKT, irrespective of the SMAD4 status, correlate with poor survival, suggesting them as strong prognostic biomarkers and targets for therapeutic intervention. Moreover, we find that overexpression of SMAD4 or depletion of RICTOR suppresses AKT signaling and increases sensitivity to irinotecan in SMAD4-deficient colon cancer cells. Consistent with these observations, pharmacologic inhibition of AKT sensitizes SMAD4-negative colon cancer cells to irinotecan in vitro and in vivo. Overall, our study suggests that hyperactivation of the mTORC2 pathway is a therapeutic vulnerability that could be exploited to sensitize SMAD4-negative colon cancer to irinotecan.

Implications:

Hyperactivation of the mTORC2 pathway in SMAD4-negative colon cancer provides a mechanistic rationale for targeted inhibition of mTORC2 or AKT as a distinctive combinatorial therapeutic opportunity with chemotherapy for colon cancer.

Despite preventive screening, colon cancer remains as the second most lethal cancer in men and women combined in the United States with more than 50,000 deaths estimated to occur every year, mostly attributed to metastasis and resistance to therapy (1). The majority of colon cancer cases are of sporadic origin, and surgery has limited therapeutic role in cases with metastatic colon cancer as only 10% to 15% of patients have resectable lesions (2). 5-Fluorouracil (5-FU) is one of the most extensively used drugs in the treatment of metastatic colon cancer and remains as the clinical backbone of almost all chemotherapy combinations. It is often used along with oxaliplatin (L-OHP) or irinotecan as FOLFOX [5-FU, leucovorin (folinic acid), oxaliplatin] or FOLFIRI [5-FU, leucovorin (folinic acid), irinotecan] respectively as the standard first-line and second-line chemotherapeutic options available to combat metastatic disease in improving patient survival (3, 4). Intriguingly, screening of biomarkers to predict response to these agents is not implemented, and an agent is only removed during subsequent regimens when intolerable toxicity occurs (5). Due to toxicities caused by chemotherapeutic agents, it would be more effective to combine single agents with therapies directed at biological targets to reduce the doses below toxic levels and to enhance sensitivity. Therefore, understanding the molecular basis of metastatic colon cancer will be beneficial to all affected patients in helping to design effective therapeutic strategies.

Loss of heterozygosity (LOH) at chromosome 18q has long been established as a late event during colon cancer progression (6, 7). Furthermore, several studies have suggested that LOH at 18q was an indicator of a poor prognosis in patients with tumors penetrating the bowel wall or involving regional lymph nodes (tumor–node–metastasis stages II and III, respectively) who succumbed to disease recurrence and died within 5 years of surgical removal of their primary tumor (8, 9). To identify the target gene(s) for 18q deletions in colon cancer, we found SMAD4 mutations or genomic deletions of this gene (10). This has been confirmed in numerous follow-up studies that a high frequency of LOH at 18q was associated with an increase in the frequency of SMAD4 mutations, which occur in about 10% to 30% of colon cancer and correlated to an advanced stage colon cancer (11–13). Furthermore, when tumors corresponding to different stages of colon cancer were interrogated for SMAD4 inactivation arising from deletions or point mutations, there was a strong correlation between increasing frequency of SMAD4 gene mutations and distant metastases (stage IV) relative to nonmetastatic colon cancer (14, 15). A strong correlation between loss of SMAD4 expression and liver metastasis with poor prognosis in colon cancers (the most common site for colon cancer metastases) has also been established from the examination of primary tumors and the corresponding metastatic tissues (15–17). In addition to colon cancer, a tumor-suppressive role corresponding to mutations, deletions, and low levels of SMAD4 has been associated with poor prognosis in several other cancers (18–22).

Moreover, credence to the contribution of SMAD4 defect in forming metastatic colon cancer was also derived from mouse models where a dramatic increase in malignant progression of intestinal polyps in cis-compound heterozygotes (i.e., Apc (±) Smad4 (±) compared with the simple Apc (±) heterozygotes) was observed (23). Subsequently, inactivation of SMAD4 in organoid models was crucial in showing tumor progression to the malignant and invasive stages of colon cancer (24). In vitro and xenotransplantation studies further supported the tumor-suppressive function of SMAD4, whereupon removal it promotes malignant phenotypes including cell migration, tumorigenesis, angiogenesis, aerobic glycolysis, and metastasis (25, 26). Clinically, loss of or low SMAD4 expression correlated with the presence of metastases and has been associated with poor response to 5-FU and worse survival post-5-FU treatment (27, 28). Although studies have shown that SMAD4-negative colon cancer is more resistant to 5-FU, whether and how SMAD4 inactivation confers resistance to standard chemotherapeutic regimens, such as FOLFIRI and FOLFOX, and strategies to enhance sensitivity to these therapeutic strategies remain elusive.

Here, we report that among the three established commonly used active chemotherapeutic agents that constitute the common therapeutic regimens to treat colon cancer, SMAD4-negative colon cancer cells exhibit resistance to both 5-FU and irinotecan but not to oxaliplatin. Furthermore, we found that inactivation of SMAD4 leads to overactivation of the mTORC2 pathway, thereby augmenting AKT signaling and resistance to irinotecan-mediated apoptosis. Consistent with these observations, targeting the mTORC2 pathway with SMAD4 overexpression, RICTOR depletion, or inhibition of the downstream effector, AKT, with MK2206 restores sensitivity of SMAD4-negative colon cancer cells to irinotecan.

Cell culture

HCT116 SMAD4+/+ and SMAD4−/− isogenic cell lines are a generous gift from Dr. Bert Vogelstein, whereas SW403, ASPC1, and CFPAC1 cells were obtained from the ATCC. HCT116 and SW403 cells were cultured in McCoy's medium. ASPC1 cells were maintained in RPMI medium, whereas CFPAC1 cells were cultured in IMDM. All cell lines were maintained in the presence of 10% FBS and 1% penicillin/streptomycin in a 37°C incubator with 5% CO2.

Coimmunoprecipitation

Cells were washed with PBS and incubated with 1 mmol/L dithiobis succinimidyl propionate at room temperature for 30 minutes. The cross-linking reaction was quenched using 10 mmol/L Tris for 15 minutes. Cells were then washed and lysed with Pierce IP buffer in the presence of protease and phosphatase inhibitors (Roche). Protein lysates were scraped from the dish using cell lifters and centrifuged for 15 minutes at 14,000 × g at 4°C. Anti-FLAG affinity gel (Sigma-Aldrich) was washed 3 times with Pierce IP buffer and mixed with cell lysate overnight at 4°C. The beads were then washed 3 times with Pierce IP buffer to remove unbound proteins. The beads were incubated with 3xFLAG peptide (Sigma-Aldrich) for 1 hour at room temperature to elute bound proteins.

Mass spectrometry analysis

Eluate from coimmunoprecipitation (co-IP) was mixed with 4xLDS buffer and 10x reducing agent before being loaded onto NuPAGE 4%–12% precast gels and separated for 15 minutes at 100 V. The gel area with trapped eluate was then excised and digested with trypsin. The digested samples were then analyzed with LC/MS/MS and subjected to Mascot database search for protein identification. Protein candidates detected as background in more than 10 experiments among the total 411 experiments curated by the CRAPome database were first filtered (29). Next, protein candidates that were enriched more than 5-fold based on spectral counts in the sample compared with control were then uploaded to Ingenuity Pathway Analysis for direct protein–protein interaction analysis to determine if they form any protein complexes.

Kaplan–Meier analysis

A database of patients with colon cancer was established as described previously (30). Survival curves were generated based on the transcript level of a candidate gene using the Cox proportional hazards regression analysis and plotting Kaplan–Meier plots as described (31). A P value below 0.05 was accepted as a significant correlation between gene expression and survival.

Tumor xenograft studies

The Institutional Animal Care and Use Committee at Boston University School of Medicine approved all animal experiments. Six-week-old female athymic nude mice (Nu/Nu) were purchased from Envigo and housed in a sterile environment with microisolator cages. The mice were subcutaneously (s.c.) injected with 2.5 × 106 HCT116 SMAD4−/− cells in 30% growth factor–reduced Matrigel (Corning). When the tumors reached around 5 mm in diameter, the mice were exposed to vehicle, MK2206 (MedChemExpress) alone, irinotecan (MedChemExpress) alone, or a combination of MK2206 and irinotecan. MK2206 (360 mg/kg) in 30% Captisol was administered on days 1, 8, 15, and 22 via oral gavage. Irinotecan (20 mg/kg) was administered on days 1, 8, 15, and 22 via intraperitoneal (i.p.) injections. Tumor volume was determined using (L × W2)/2, where L represents the length and W represents the width.

Statistical analysis

For two group comparisons, the Student t test (two-tailed, type two) was applied. Significance of multiple condition experiments was determined using one-way ANOVA. A P value below 0.05 was considered statistically significant. All data shown in the bar graphs are the mean ± SD of at least three biological replicates. Error bars represent SD.

Please refer to Supplementary File for additional Materials and Methods.

SMAD4-negative colon cancer cells exhibit higher migratory ability and resistance to 5-FU and irinotecan

To determine whether SMAD4 suppresses cancer progression in our colon cancer model, we first compared the migratory ability of a pair of isogenic SMAD4-positive and negative HCT116 cells. While the SMAD4 gene was knocked out using targeted homologous integration, TGFBRII was restored to reconstitute intact TGFβ signaling in HCT116 cells as previously described (25, 32). As expected, SMAD4-negative cells exhibited higher migratory potential than SMAD4-positive cells, which was in concurrence with previous reports that SMAD4 inactivation promotes malignant progression of colon adenoma to carcinoma (Supplementary Fig. S1; ref. 24). To assess whether SMAD4 expression level could serve as a prognostic biomarker in patients with colon cancer, we performed in silico Kaplan–Meier analyses and found that higher levels of SMAD4 associated with increased probability of overall survival (OS; HR = 0.62, P < 0.05), relapse-free survival (RFS; HR = 0.75, P < 0.015), and postprogression survival (PPS; HR = 0.42, P = 0.05) decreased significantly in patients with low levels of SMAD4 expression (33). Overall, these findings provide additional credence in support of the role of SMAD4 as a tumor-suppressor gene (Fig. 1A).

Figure 1.

SMAD4-negative colon cancer cells are resistant to 5-FU and irinotecan but not to oxaliplatin. A,In silico Kaplan–Meier analyses showing the correlation between SMAD4 expression and OS, RFS, and PPS in patients with colon cancer. The analyses ran on a cohort of 304 (OS), 1,045 (RFS), and 105 (PPS) patients, respectively. B,SMAD4+/+ and SMAD4−/− cells were treated with 5-FU, irinotecan, or oxaliplatin, and the viability of cells relative to DMSO-treated controls was determined after 72 hours (cell viability assay; mean ± SD, n = 3 biological replicates). C, The concentration at which 50% of growth was inhibited (IC50) was calculated using Prism for each drug (IC50 analysis; mean ± SD, n = 3; *, P < 0.05 and ***, P < 0.001).

Figure 1.

SMAD4-negative colon cancer cells are resistant to 5-FU and irinotecan but not to oxaliplatin. A,In silico Kaplan–Meier analyses showing the correlation between SMAD4 expression and OS, RFS, and PPS in patients with colon cancer. The analyses ran on a cohort of 304 (OS), 1,045 (RFS), and 105 (PPS) patients, respectively. B,SMAD4+/+ and SMAD4−/− cells were treated with 5-FU, irinotecan, or oxaliplatin, and the viability of cells relative to DMSO-treated controls was determined after 72 hours (cell viability assay; mean ± SD, n = 3 biological replicates). C, The concentration at which 50% of growth was inhibited (IC50) was calculated using Prism for each drug (IC50 analysis; mean ± SD, n = 3; *, P < 0.05 and ***, P < 0.001).

Close modal

Next, to test whether SMAD4 inactivation confers resistance to common chemotherapeutic agents used for treating colon cancer, we exposed the colon cancer model cells to increasing concentrations of 5-FU, irinotecan, or oxaliplatin. Interestingly, compared with SMAD4-positive, SMAD4-negative cells exhibited significant resistance to 5-FU (IC50 = 1.78 μmol/L vs. 4.7 μmol/L) and irinotecan (IC50 = 0.73 μmol/L vs. 6.5 μmol/L), but not to oxaliplatin (IC50 for both at around 0.37 μmol/L; Fig. 1B and C). Although resistance to 5-FU has been previously reported by others and us (25–28), here we also found that there was significant increase in viability of SMAD4-negative compared with SMAD4-positive cells upon exposure to irinotecan.

Mass spectrometry reveals RICTOR as a novel SMAD4-interacting protein in colon cancer cells

Several reports have indicated that overexpression of hypoxia-inducible factor 1-alpha (HIF1α) is associated with poor prognosis in patients with colon cancer (33–35). Our previous studies found that one of the mechanisms for the tumor-suppressive role of SMAD4 in colon cancer is due to its interaction with and inhibition of tumor-promoting transcriptional activation mediated by HIF1α to suppress its target gene, VEGF, which promotes angiogenesis (25). These observations suggested that SMAD4 could similarly interact with other transcription factors or tumor-promoting pathway factors to suppress oncogenic events including metastasis and drug resistance. To dissect the SMAD4 interactome, we elected to use FLAG-tagged SMAD4 to capture proteins that interact to form complexes in colon cancer cells. We constructed a FLAG-SMAD4 overexpression plasmid and confirmed that the FLAG-SMAD4 protein was functional based on the ability to induce expression of the luciferase reporter gene downstream of a SMAD-binding element (SBE4) in response to TGFβ treatment (Fig. 2A and B). Next, co-IP of the FLAG-SMAD4 with other protein factors was performed and followed by mass spectrometry (MS) to identify the protein components of the complex. These analyses revealed 1,200 protein hits, which were subjected to CRAPome, spectral count enrichment, and ingenuity pathway analyses (Fig. 2C; Supplementary Table S1). Interestingly, three members of the mTORC2 complex, mTOR, RICTOR, and TELO2, were among the proteins bound by SMAD4 (36, 37). Therefore, we predicted that these interactions could be of functional relevance and decided to focus on the novel interaction between SMAD4 and RICTOR, where the latter is a unique component of the mTORC2 complex making it as a potential precision therapeutic target for colon cancer (Fig. 2C; ref. 38).

Figure 2.

MS analysis reveals novel SMAD4-interacting partners. A, Overexpression of FLAG-tagged SMAD4 in SMAD4−/− cells was monitored using Western blotting. B, Cells were transfected with SBE4-luciferase reporter plasmid for 72 hours, serum-starved overnight, and treated with 5 ng/mL of TGFβ for 4 hours prior to lysis (RLU, relative luminescence unit; mean ± SD, n = 3 biological replicates; ***, P < 0.001). C, FLAG-SMAD4 protein complexes were immunoprecipitated from cell lysates and identified using MS and Mascot database search. Schematic shows our strategy for the selection of RICTOR, which is part of mTORC2, as the top candidate targeted by SMAD4. D, The presence of phospho-SMAD2, a known SMAD4-interacting protein, and RICTOR, a novel SMAD4-interacting candidate, was determined in FLAG-SMAD4 complexes using Western blotting. β-Actin level was used as the loading control. E, Colon cancer cells were serum-starved and treated with or without 5 ng/mL of TGFβ prior to cell lysis and co-IP. Western blotting shows the relative levels of RICTOR in FLAG-SMAD4 complexes with or without TGFβ treatment.

Figure 2.

MS analysis reveals novel SMAD4-interacting partners. A, Overexpression of FLAG-tagged SMAD4 in SMAD4−/− cells was monitored using Western blotting. B, Cells were transfected with SBE4-luciferase reporter plasmid for 72 hours, serum-starved overnight, and treated with 5 ng/mL of TGFβ for 4 hours prior to lysis (RLU, relative luminescence unit; mean ± SD, n = 3 biological replicates; ***, P < 0.001). C, FLAG-SMAD4 protein complexes were immunoprecipitated from cell lysates and identified using MS and Mascot database search. Schematic shows our strategy for the selection of RICTOR, which is part of mTORC2, as the top candidate targeted by SMAD4. D, The presence of phospho-SMAD2, a known SMAD4-interacting protein, and RICTOR, a novel SMAD4-interacting candidate, was determined in FLAG-SMAD4 complexes using Western blotting. β-Actin level was used as the loading control. E, Colon cancer cells were serum-starved and treated with or without 5 ng/mL of TGFβ prior to cell lysis and co-IP. Western blotting shows the relative levels of RICTOR in FLAG-SMAD4 complexes with or without TGFβ treatment.

Close modal

Because mTORC2 phosphorylation of oncoprotein AKT at serine 473 activates the downstream events of the mTORC2 pathway to promote cell survival, we predicted that it could be a major mediator of chemoresistance and decided to undertake functional characterization of RICTOR as a potential target for therapeutic intervention in SMAD4-negative colon cancer (38, 39). First, we followed up with MS data and confirmed that SMAD4 interacted with RICTOR in vitro (Fig. 2D). Next, to evaluate if the interaction between SMAD4 and RICTOR is dependent upon TGFβ-mediated downstream effects, we performed Western blot analysis of the SMAD4 protein complexes formed in the presence or absence of TGFβ. Interestingly, this interaction appeared to be independent of TGFβ stimulation, indicating that SMAD4 may have other noncanonical roles in suppressing colon cancer progression that are not dependent upon active TGFβ signaling (Fig. 2E).

SMAD4-negative colon cancer exhibits hyperactivation of the mTORC2 pathway

To better understand whether mTORC2 is a major contributor to chemoresistance and could thereby serve as a precision therapeutic target in SMAD4-negative colon cancer, we first assessed the activation status of mTORC2 pathway by examining the level of phospho-AKTS473, a downstream oncogenic target activated by mTORC2. We found that SMAD4-negative cells displayed pronounced phospho-AKTS473 levels compared with SMAD4-positive cells, consistent with the notion that SMAD4 may play a role in suppressing this pathway (Fig. 3A; ref. 26). In line with these observations, SMAD4-negative cells were highly sensitive to AKT inhibition, indicated by significantly reduced viability upon treatment with MK2206, an allosteric AKT inhibitor (Fig. 3B and C).

Figure 3.

SMAD4-negative colon cancer cells exhibit hyper AKT signaling activity and sensitivity to MK2206. A, Western blotting shows the relative levels of phospho-AKTS473 in SMAD4+/+ and SMAD4−/− cells. B,SMAD4+/+ and SMAD4−/− cells were treated with MK2206, an allosteric AKT inhibitor, and the viability of cells relative to DMSO-treated controls was determined after 72 hours. C, The IC50 of MK2206 in each cell line was calculated using Prism (IC50 analysis; mean ± SD, n = 3 biological replicates; **, P < 0.01). D, Western blotting shows the relative levels of p-AKTS473 in cells overexpressing SMAD4. E, The indicated cell lines were treated with irinotecan (10 μmol/L) for 72 hours. Viability of cells was normalized to DMSO-treated controls (cell viability assay; mean ± SD, n = 3 biological replicates; *, P < 0.05). F, Western blotting shows the relative levels of cleaved caspase 3 in the indicated cell lines after treatment with irinotecan (50 μmol/L) for 18 hours.

Figure 3.

SMAD4-negative colon cancer cells exhibit hyper AKT signaling activity and sensitivity to MK2206. A, Western blotting shows the relative levels of phospho-AKTS473 in SMAD4+/+ and SMAD4−/− cells. B,SMAD4+/+ and SMAD4−/− cells were treated with MK2206, an allosteric AKT inhibitor, and the viability of cells relative to DMSO-treated controls was determined after 72 hours. C, The IC50 of MK2206 in each cell line was calculated using Prism (IC50 analysis; mean ± SD, n = 3 biological replicates; **, P < 0.01). D, Western blotting shows the relative levels of p-AKTS473 in cells overexpressing SMAD4. E, The indicated cell lines were treated with irinotecan (10 μmol/L) for 72 hours. Viability of cells was normalized to DMSO-treated controls (cell viability assay; mean ± SD, n = 3 biological replicates; *, P < 0.05). F, Western blotting shows the relative levels of cleaved caspase 3 in the indicated cell lines after treatment with irinotecan (50 μmol/L) for 18 hours.

Close modal

Next, to elucidate if restoration of SMAD4 could be directly involved in inhibiting RICTOR-mediated downstream signaling, we assessed the level of phospho-AKTS473 in colon cancer cells overexpressing FLAG-SMAD4. We found that SMAD4 overexpression resulted in suppressed levels of phospho-AKTS473, which serves as a functional readout of mTORC2 pathway activation (Fig. 3D). Importantly, SMAD4 overexpression resulted in enhanced sensitivity of the colon cancer cells to irinotecan, with a corresponding increase in the levels of the apoptotic marker, cleaved caspase 3 (Fig. 3E and F). Overall, these data suggested that SMAD4 deletion in colon cancer might lead to uninhibited mTORC2/AKT signaling activity, thereby promoting resistance to irinotecan-mediated apoptosis.

Depletion of RICTOR suppresses AKT signaling activity and increases sensitivity of SMAD4-negative colon cancer cells to irinotecan

To determine whether RICTOR is critical to mTORC2 functionality in SMAD4-negative colon cancer, we knocked down RICTOR and assessed the expression level of phospho-AKTS473 in SMAD4-negative HCT116 cells using two different shRNAs (Fig. 4A; Supplementary Fig. S2; ref. 36). We noticed a dramatic decrease in phospho-AKTS473 level upon RICTOR depletion, confirming that RICTOR is an essential component to the kinase function of mTORC2 required for activating AKT signaling pathway in our model system (Fig. 4B; Supplementary Fig. S2). In addition, we observed that knockdown of RICTOR increased sensitivity of SMAD4-negative colon cancer cells to irinotecan (Fig. 4C; Supplementary Fig. S2). Corresponding to RICTOR and phospho-AKTS473 depletion, the cells also displayed higher levels of cleaved caspase 3 in response to irinotecan treatment, suggesting that mTORC2/AKT signaling drives resistance to irinotecan by blocking apoptosis (Fig. 4D; Supplementary Fig. S2). Interestingly, depletion of RICTOR in SMAD4-negative HCT116 cells also impaired their migratory ability, suggesting that RICTOR could serve as a potential therapeutic target to suppress colon cancer progression (Supplementary Fig S3). To further support the universality of the phenomenon that mTORC2 activation correlated with SMAD4 deficiency in colon cancer, we examined an additional SMAD4-negative SW403 cell line and observed a decrease in pAKTS473 levels and sensitization of the cells to irinotecan upon depletion of RICTOR using siRNA (Supplementary Fig. S4).

Figure 4.

Knockdown of RICTOR suppresses AKT signaling and enhances sensitivity of SMAD4-negative colon cancer cells to irinotecan. A, RICTOR depletion in SMAD4-negative colon cancer cells using shRNA was measured by RT-qPCR (RT-qPCR; mean ± SD, n = 3 technical replicates; ***, P < 0.001). B, Western blotting shows the relative levels of RICTOR and p-AKTS473 in the indicated cell lines. C, The indicated cell lines were treated with irinotecan (10 μmol/L) for 72 hours. Viability of cells was normalized to DMSO-treated controls (cell viability assay; mean ± SD, n = 3; **, P < 0.01). D, Western blotting shows the relative levels of cleaved caspase 3 in the indicated cell lines after treatment with irinotecan (50 μmol/L) for 18 hours.

Figure 4.

Knockdown of RICTOR suppresses AKT signaling and enhances sensitivity of SMAD4-negative colon cancer cells to irinotecan. A, RICTOR depletion in SMAD4-negative colon cancer cells using shRNA was measured by RT-qPCR (RT-qPCR; mean ± SD, n = 3 technical replicates; ***, P < 0.001). B, Western blotting shows the relative levels of RICTOR and p-AKTS473 in the indicated cell lines. C, The indicated cell lines were treated with irinotecan (10 μmol/L) for 72 hours. Viability of cells was normalized to DMSO-treated controls (cell viability assay; mean ± SD, n = 3; **, P < 0.01). D, Western blotting shows the relative levels of cleaved caspase 3 in the indicated cell lines after treatment with irinotecan (50 μmol/L) for 18 hours.

Close modal

Despite the correlation between functional activation of mTORC2 pathway and malignant progression of colon cancer in SMAD4-negative cells, we wondered if overall levels of RICTOR or AKT1 could predict survival differences in all patients with colon cancer irrespective of their SMAD4 status. Interestingly, we found that higher levels of RICTOR associated with decreased OS (HR = 1.72, P < 0.01), RFS (HR = 1.34, P < 0.05), and PPS (HR = 3.02, P < 0.01) in patients with colon cancer (Supplementary Fig. S5A). Similarly, we found that higher levels of AKT1 corresponded with decreased OS (HR = 1.84, P < 0.01), RFS (HR = 1 0.56, P < 0.01), and PPS (HR = 1.8, P < 0.05) in patients with colon cancer (Supplementary Fig. S5B). Overall, these observations suggest that higher levels of RICTOR or AKT1 could predict worse prognosis of patients with colon cancer.

Targeting AKT with MK2206 sensitizes SMAD4-negative colon cancer cells to irinotecan

Because there are no drugs currently available to specifically target RICTOR and mTORC2 (40), we hypothesized that inhibiting their downstream effector target AKT could restore sensitivity of SMAD4-negative colon cancer cells to irinotecan. Unlike the mTOR inhibitor sirolimus, the addition of MK2206 was able to drastically deplete the level of phospho-AKTS473 in a manner similar to RICTOR knockdown (Figs. 4B and 5A; ref. 41). These observations are highly consistent with the notion that mTORC2 pathway activation is the primary mediator of AKTS473 phosphorylation in these cells. Interestingly, prolonged exposure to sirolimus did not affect the levels of phospho-AKTS473 in these cells, suggesting that mTORC2 is rapamycin-insensitive in our model (Fig. 5A). Subsequently, we found that the combination of MK2206 and irinotecan was able to suppress the viability of SMAD4-negative HCT116 cells 2-fold more effectively in vitro than irinotecan alone, on par with the sensitivity observed in SMAD4-positive cells treated with irinotecan alone (Fig. 5B). The addition of MK2206 also induced higher levels of cleaved caspase 3 in the presence of irinotecan, further supporting that AKT activation is a major driver of resistance to irinotecan-mediated apoptosis (Fig. 5C). To ensure that this phenomenon is not cell line–specific, we also examined the SMAD4-negative SW403 colon cancer cells and found that the cells exhibited enhanced suppression of viability upon combination treatment with a corresponding increase in cleaved caspase 3 levels (Supplementary Fig. S6). Furthermore, the use of a different allosteric AKT inhibitor (42), AKTi-1/2, in combination with irinotecan also exhibited enhanced sensitivity of SMAD4-negative cells in vitro, suggesting targeted inhibition of AKT is the common phenomenon responsible for the additive therapeutic effect (Supplementary Fig. S7).

Figure 5.

Targeting AKT with MK2206 suppresses p-AKTS473 level and sensitizes SMAD4-negative colon cancer cells to irinotecan. A, Western blotting shows the relative levels of p-AKTS473 in SMAD4-negative cells after treatment with MK2206 (1 μmol/L) or of sirolimus (10 μmol/L), an mTOR inhibitor, for 24 and 48 hours. B, SMAD4-positive and -negative cells were treated with MK2206 (1 μmol/L), irinotecan (1 μmol/L), or both for 72 hours (cell viability assay; mean ± SD, n = 3 biological replicates; **, P < 0.01). C, The levels of cleaved caspase 3 were determined in SMAD4-positive and -negative cells after treatment with MK2206 (1 μmol/L), irinotecan (50 μmol/L), or both for 18 hours.

Figure 5.

Targeting AKT with MK2206 suppresses p-AKTS473 level and sensitizes SMAD4-negative colon cancer cells to irinotecan. A, Western blotting shows the relative levels of p-AKTS473 in SMAD4-negative cells after treatment with MK2206 (1 μmol/L) or of sirolimus (10 μmol/L), an mTOR inhibitor, for 24 and 48 hours. B, SMAD4-positive and -negative cells were treated with MK2206 (1 μmol/L), irinotecan (1 μmol/L), or both for 72 hours (cell viability assay; mean ± SD, n = 3 biological replicates; **, P < 0.01). C, The levels of cleaved caspase 3 were determined in SMAD4-positive and -negative cells after treatment with MK2206 (1 μmol/L), irinotecan (50 μmol/L), or both for 18 hours.

Close modal

Having demonstrated the effects of drug treatment in vitro, we next examined the efficacy of the antitumor activities of combination therapy in vivo using nude mice harboring HCT116 SMAD4−/− xenograft tumors. The tumor-bearing mice were randomized to receive vehicle, MK2206, irinotecan, or a combination of MK2206 and irinotecan. Compared with single agents, we found that the combinatorial treatment with chemotherapeutic agent irinotecan and targeted inhibition of the mTORC2 pathway was the most effective in remarkable tumor growth suppression in vivo (Fig. 6A and B). Interestingly, the more dramatic tumor suppression with combination treatment was not associated with increase in host toxicity (Supplementary Fig. S8).

Figure 6.

Combination treatment with MK2206 suppresses growth of SMAD4-negative xenografts. A, HCT116 SMAD4−/− xenografts in nude mice were treated with vehicle (n = 4 biological replicates), MK2206 (360 mg/kg, n = 4 biological replicates), irinotecan (20 mg/kg, n = 5 biological replicates), or a combination of MK2206 and irinotecan (n = 5 biological replicates). Tumors were monitored twice a week using a caliper (relative tumor volume; mean ± SD; *, P < 0.05 by ANOVA). Representative images of tumors at the end of the experiment are shown. B, Waterfall plot shows the relative volume of tumors between treatment arms on day 30. C, Our working model shows that SMAD4 inactivation leads to uninhibited mTORC2/AKT signaling activity and resistance to irinotecan-mediated apoptosis.

Figure 6.

Combination treatment with MK2206 suppresses growth of SMAD4-negative xenografts. A, HCT116 SMAD4−/− xenografts in nude mice were treated with vehicle (n = 4 biological replicates), MK2206 (360 mg/kg, n = 4 biological replicates), irinotecan (20 mg/kg, n = 5 biological replicates), or a combination of MK2206 and irinotecan (n = 5 biological replicates). Tumors were monitored twice a week using a caliper (relative tumor volume; mean ± SD; *, P < 0.05 by ANOVA). Representative images of tumors at the end of the experiment are shown. B, Waterfall plot shows the relative volume of tumors between treatment arms on day 30. C, Our working model shows that SMAD4 inactivation leads to uninhibited mTORC2/AKT signaling activity and resistance to irinotecan-mediated apoptosis.

Close modal

Despite the use of irinotecan-based (FOLFIRI) and oxaliplatin-based (FOLFOX) chemotherapy consisting of the 5-FU backbone to significantly improve survival in metastatic colon cancer, 5-year OS of patients remains at 5% to 15% and resistance to chemotherapy is mounting. As such, it warrants the development of new strategies to combat this disease a priority. At this time, an exponential increase in our knowledge base on the identity of genetic alterations in colon and other cancers is still unable to deliver precision medicine as it has lagged due to the difficulties in pinpointing biological targets that become functionally active in the various tumor types. SMAD4 mutation or loss of expression, which occurs frequently in late-stage colon cancer (10–13), correlates with poor OS, RFS, and PPS (Fig. 1A). Interestingly, several studies reported that loss of SMAD4 functionality corresponded to resistance to 5-FU in the clinic, a standard first-line treatment for the disease (27, 28). Although 5-FU resistance has been associated with SMAD4-defective colon cancers, the applicability of this resistance phenomenon to other standard chemotherapeutics, such as oxaliplatin and irinotecan as well as potential biological targets for inhibition to enhance the therapeutic benefit, has remained elusive. Here, we report that although SMAD4-negative colon cancer exhibits resistance to both 5-FU and irinotecan, sensitivity to oxaliplatin is unaffected by the SMAD4 status.

On the contrary to blinded use of inhibitors to common oncogenic signaling pathways, such as MEK-ERK, p38-MAPK, and PI3K/AKT alone or in combination with chemotherapy with uncertain therapeutic benefit for patients with colon cancer, here we present an attempt to identify specific biological targets to sensitize chemoresistant SMAD4-negative colon cancer to reap the maximum benefit with minimal side effects (25, 26). Previously, our group reported that SMAD4 interacts with HIF1α to suppress the expression of VEGF, a well-established HIF1α target gene that promotes angiogenesis (25). Based on these observations, we hypothesized that SMAD4 may also act by inhibiting other critical protein factors involved in conferring resistance to chemotherapeutic agents such as irinotecan. MS analysis revealed candidate proteins bound by SMAD4, including mTOR and TELO2, both of which are common in mTORC1 and mTORC2, as well as RICTOR, which is an essential constituent of mTORC2, a protein complex that primarily phosphorylates and fully activates the oncogene AKT at Serine 473 (36). Because the activation of AKT at Serine 473 has been shown to promote colon cancer cell migration and antagonizes apoptosis (38, 39), we decided to characterize the role of RICTOR, which is unique to mTORC2, in SMAD4-negative colon cancer (38). Indeed, we found that RICTOR depletion not only impairs AKT signaling and cell migration but also sensitizes the cells to irinotecan-mediated cell death. Interestingly, Kaplan–Meier analyses also revealed that high RICTOR/AKT1 expression, independent of the SMAD4 status, significantly correlated with worse OS, RFS, and PPS in patients with colon cancer, indicating the roles of these two genes in promoting disease progression and thus could serve as potential therapeutic targets for colon cancer in general (Supplementary Fig. S5).

Currently, there are no drugs that specifically and effectively target RICTOR or mTORC2 with precision (39, 40, 43, 44). Therefore, to inhibit mTORC2 signaling activity in SMAD4-negative colon cancer in our proof of principle experiments, we opted to block its downstream effector target AKT using MK2206, a commercially available, most clinically advanced, and well-tolerated allosteric inhibitor of AKT, which blocks S473 phosphorylation of AKT, the primary target of mTORC2 pathway (41). We found that SMAD4-negative colon cancer cells are more sensitive to MK2206 treatment compared with SMAD4-positive cells, and that MK2206 can further suppress the growth of SMAD4-negative cells in the presence of irinotecan (Fig. 5B). Importantly, the additive effect of these two drugs resulted in increased apoptosis of the treated SMAD4-negative cells in vitro (Fig. 5C). The suppression of tumor growth using the combination therapy was also confirmed in xenograft models derived from SMAD4-negative colon cancer cells, and it was not associated with increase in host toxicity (Supplementary Fig. S8).

Interestingly, we also found that the use of MK2206 with irinotecan could also significantly enhance suppression of the viability of SMAD4-negative pancreatic cancer cell lines (ASPC1, CFPAC1) displaying active AKT signaling in vitro, suggesting that the combination therapy could be of general applicability for cancers exhibiting loss of SMAD4 (Supplementary Fig. S9). In addition, we also noted that high RICTOR or AKT1 expression corresponded significantly to poor OS in patients with pancreatic cancer, indicating these as potential therapeutic targets for the affected individuals (Supplementary Fig. S10).

The Cancer Genome Atlas data have indicated that KRAS mutations frequently occur with SMAD4 defects in metastatic colon cancers (45). These observations are consistent with previous studies that correlated SMAD4 defects to advanced stages of colon cancer and as such the probability of simultaneously finding KRAS mutations and SMAD4 alterations in these tumors is very high (10, 25 and references therein). On the other hand, clinical benefit for targeting EGFR with the use of humanized monoclonal antibodies, such as cetuximab or panitumumab, has been restricted to patients with wild-type KRAS metastatic colon cancers (46). Therefore, our finding of mTORC2 pathway activation (i.e., AKT activation) with SMAD4 loss of function provides a rationale for RICTOR/AKT as potential precision therapeutic targets in colon cancers with low levels of SMAD4 with activated EGFR. Further credence to this notion is also derived from the recent finding that patients carrying SMAD4 mutations had a higher possibility of a less effective response to EGFR blockade with a shorter progression-free survival (47). Thus, targeting mTORC2 pathway activation as suggested from our studies is likely to be beneficial to patients exhibiting poor response to therapy using antibody therapy targeting EGFR, and clinical trials in the future are required to take advantage of these findings.

In conclusion, our observations suggest that overactivation of the mTORC2 pathway, which has been associated with poor survival in a growing number of cancers (48, 49), may be the driver of metastatic cancer progression and resistance to apoptosis induced by chemotherapeutic agents. We report here for the first time that SMAD4 interacts with RICTOR to suppress mTORC2 functionality and therefore the loss of SMAD4 function results in oncogenic activation of the mTORC2 pathway, leading to enhancement in malignant colon cancer progression and resistance to chemotherapeutic agents such as irinotecan (Fig. 6C). Thus, inactivation of AKTS473 phosphorylation or more specifically its upstream regulator, RICTOR, emerged as legitimate strategies to enhance the sensitivity of SMAD4-negative colon cancer cells to irinotecan as shown in our studies. Interestingly, our studies also found that overexpression of RICTOR or AKT1 could serve as biomarkers for poor prognosis, independently of the SMAD4 status. Overall, we suggest that design of therapies involving established chemotherapeutic agents such as irinotecan might be highly effective when combined with targeted inhibitors for RICTOR/AKT when the colon cancer cells are either SMAD4-negative or exhibit overexpression of RICTOR/AKT.

No potential conflicts of interest were disclosed.

Conception and design: C.K. Wong, S. Ozturk, H. Feng, S. Thiagalingam

Development of methodology: C.K. Wong, S. Ozturk, D. Lopez, Z. Sen, S. Thiagalingam

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): C.K. Wong, A.W. Lambert, P. Papageorgis, D. Lopez, H.M. Abdolmaleky, B. Győrffy, S. Thiagalingam

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): C.K. Wong, A.W. Lambert, P. Papageorgis, B. Győrffy, S. Thiagalingam

Writing, review, and/or revision of the manuscript: C.K. Wong, A.W. Lambert, S. Ozturk, P. Papageorgis, H.M. Abdolmaleky, S. Thiagalingam

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): N. Shen, H. Feng, S. Thiagalingam

Study supervision: H. Feng, S. Thiagalingam

This work was supported by a grant from NIH/NCI (CA165707), an integrated pilot grant funded by the Boston University Clinical and Translational Science Institute (NIH/NCATS award 1UL1TR001430), Carter Pilot Award for Diversity and Cancer Equity from Boston University Cancer Center, and a seed grant from the Boston University Genome Science Institute to S. Thiagalingam. C.K. Wong is a recipient of the Boston University Cross-Disciplinary Training in Nanotechnology for Cancer (XTNC), BUnano, and Susan G. Komen Mentoring and Training in Cancer Health Disparities (MATCH) fellowships. We thank Dr. Bert Vogelstein (Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins), Dr. William C. Hahn (Dana-Farber Cancer Institute), Dr. Anuragh Singh (Boston University School of Medicine), and Drs. Jian Yu and Lin Zhang (University of Pittsburgh Cancer Institute) for generously providing reagents and cell lines. We would like to acknowledge support from the Boston University School of Medicine Biomedical Genetics Section, the Boston University School of Medicine Molecular Genetics Core Facility, the Boston University Analytical Instrumentation Core Facility, and the University of Massachusetts Medical School Mass Spectrometry Facility. We also thank Drs. Herbert T. Cohen, Marc Lenburg, and Shoumita Dasgupta for their valuable suggestions for the research project.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Siegel
RL
,
Miller
KD
,
Jemal
A
. 
Cancer statistics, 2019
.
CA Cancer J Clin
2019
;
69
:
7
34
.
2.
Hollande
F
,
Pannequin
J
,
Joubert
D
. 
The long road to colorectal cancer therapy: searching for the right signals
.
Drug Resistance Updates
2010
;
13
:
44
56
.
3.
Douillard
JY
,
Cunningham
D
,
Roth
AD
,
Navarro
M
,
James
RD
,
Karasek
P
, et al
Irinotecan combined with fluorouracil compared with fluorouracil alone as first-line treatment for metastatic colorectal cancer: a multicentre randomised trial
.
Lancet
2000
;
355
:
1041
7
.
4.
Neugut
AI
,
Lin
A
,
Raab
GT
,
Hillyer
GC
,
Keller
D
,
O'Neil
DS
, et al
FOLFOX and FOLFIRI use in stage IV colon cancer: analysis of SEER-medicare data
.
Clin Colorectal Cancer
2019
;
18
:
133
40
.
5.
Benson
AB
 III
,
Venook
AP
,
Cederquist
L
,
Chan
E
,
Chen
YJ
,
Cooper
HS
, et al
Colon cancer, version 1.2017, NCCN clinical practice guidelines in oncology
.
J Natl Compr Canc Netw
2017
;
15
:
370
98
.
6.
Jen
J
,
Kim
H
,
Piantadosi
S
,
Liu
ZF
,
Levitt
RC
,
Sistonen
P
, et al
Allelic loss of chromosome 18q and prognosis in colorectal cancer
.
N Engl J Med
1994
;
331
:
213
21
.
7.
Jernvall
P
,
Makinen
MJ
,
Karttunen
TJ
,
Makela
J
,
Vihko
P
. 
Loss of heterozygosity at 18q21 is indicative of recurrence and therefore poor prognosis in a subset of colorectal cancers
.
Br J Cancer
1999
;
79
:
903
8
.
8.
Popat
S
,
Houlston
RS
. 
A systematic review and meta-analysis of the relationship between chromosome 18q genotype, DCC status and colorectal cancer prognosis
.
Eur J Cancer
2005
;
41
:
2060
70
.
9.
Watanabe
T
,
Wu
TT
,
Catalano
PJ
,
Ueki
T
,
Satriano
R
,
Haller
DG
, et al
Molecular predictors of survival after adjuvant chemotherapy for colon cancer
.
N Engl J Med
2001
;
344
:
1196
206
.
10.
Thiagalingam
S
,
Lengauer
C
,
Leach
FS
,
Schutte
M
,
Hahn
SA
,
Overhauser
J
, et al
Evaluation of candidate tumour suppressor genes on chromosome 18 in colorectal cancers
.
Nat Genet
1996
;
13
:
343
6
.
11.
Miyaki
M
,
Iijima
T
,
Konishi
M
,
Sakai
K
,
Ishii
A
,
Yasuno
M
, et al
Higher frequency of Smad4 gene mutation in human colorectal cancer with distant metastasis
.
Oncogene
1999
;
18
:
3098
103
.
12.
Maitra
A
,
Molberg
K
,
Albores-Saavedra
J
,
Lindberg
G
. 
Loss of Dpc4 expression in colonic adenocarcinomas correlates with the presence of metastatic disease
.
Am J Pathol
2000
;
157
:
1105
11
.
13.
Cancer Genome Atlas Network
. 
Comprehensive molecular characterization of human colon and rectal cancer
.
Nature
2012
;
487
:
330
7
.
14.
Reinacher-Schick
A
,
Baldus
SE
,
Romdhana
B
,
Landsberg
S
,
Zapatka
M
,
Mönig
SP
, et al
Loss of Smad4 correlates with loss of the invasion suppressor E-cadherin in advanced colorectal carcinomas
.
J Pathol
2004
;
202
:
412
20
.
15.
Tanaka
T
,
Watanabe
T
,
Kazama
Y
,
Tanaka
J
,
Kanazawa
T
,
Kazama
S
, et al
Chromosome 18q deletion and Smad4 protein inactivation correlate with liver metastasis: A study matched for T- and N- classification
.
Br J Cancer
2006
;
95
:
1562
7
.
16.
Tanaka
T
,
Watanabe
T
,
Kitayama
J
,
Kanazawa
T
,
Kazama
Y
,
Tanaka
J
, et al
Chromosome 18q deletion as a novel molecular predictor for colorectal cancer with simultaneous hepatic metastasis
.
Diagn Mol Pathol
2009
;
18
:
219
25
.
17.
Mizuno
T
,
Cloyd
JM
,
Vicente
D
,
Omichi
K
,
Chun
YS
,
Kopetz
SE
, et al
SMAD4 gene mutation predicts poor prognosis in patients undergoing resection for colorectal liver metastases
.
Eur J Surg Oncol
2018
;
44
:
684
92
.
18.
Hahn
SA
,
Schutte
M
,
Hoque
AT
,
Moskaluk
CA
,
da Costa
LT
,
Rozenblum
E
, et al
DPC4, a candidate tumor suppressor gene at human chromosome 18q21.1
.
Science
1996
;
271
:
350
3
.
19.
Iacobuzio-Donahue
CA
,
Fu
B
,
Yachida
S
,
Luo
M
,
Abe
H
,
Henderson
CM
, et al
DPC4 gene status of the primary carcinoma correlates with patterns of failure in patients with pancreatic cancer
.
J Clin Oncol
2009
;
27
:
1806
13
.
20.
Aitchison
AA
,
Veerakumarasivam
A
,
Vias
M
,
Kumar
R
,
Hamdy
FC
,
Neal
DE
, et al
Promoter methylation correlates with reduced Smad4 expression in advanced prostate cancer
.
Prostate
2008
;
68
:
661
74
.
21.
He
S
,
Zhao
Z
,
Wang
Y
,
Zhao
J
,
Wang
L
,
Hou
F
, et al
Reduced expression of SMAD4 in gliomas correlates with progression and survival of patients
.
J Exp Clin Cancer Res
2011
;
30
:
70
.
22.
Haeger
SM
,
Thompson
JJ
,
Kalra
S
,
Cleaver
TG
,
Merrick
D
,
Wang
XJ
, et al
Smad4 loss promotes lung cancer formation but increases sensitivity to DNA topoisomerase inhibitors
.
Oncogene
2016
;
35
:
577
86
.
23.
Takaku
K
,
Oshima
M
,
Miyoshi
H
,
Matsui
M
,
Seldin
MF
,
Taketo
MM
. 
Intestinal tumorigenesis in compound mutant mice of both Dpc4 (Smad4) and Apc genes
.
Cell
1998
;
92
:
645
56
.
24.
Drost
J
,
van Jaarsveld
RH
,
Ponsioen
B
,
Zimberlin
C
,
van Boxtel
R
,
Buijs
A
, et al
Sequential cancer mutations in cultured human intestinal stem cells
.
Nature
; 
2015
;
521
:
43
7
.
25.
Papageorgis
P
,
Cheng
K
,
Ozturk
S
,
Gong
Y
,
Lambert
AW
,
Abdolmaleky
HM
, et al
Smad4 inactivation promotes malignancy and drug resistance of colon cancer
.
Cancer Res
2011
;
71
:
998
1008
.
26.
Zhang
B
,
Zhang
B
,
Chen
X
,
Bae
S
,
Singh
K
,
Washington
MK
, et al
Loss of Smad4 in colorectal cancer induces resistance to 5-fluorouracil through activating Akt pathway
.
Br J Cancer
2014
;
110
:
946
57
.
27.
Boulay
JL
,
Mild
G
,
Lowy
A
,
Reuter
J
,
Lagrange
M
,
Terracciano
L
, et al
SMAD4 is a predictive marker for 5-fluorouracil-based chemotherapy in patients with colorectal cancer
.
Br J Cancer
2002
;
87
:
630
4
.
28.
Alhopuro
P
,
Alazzouzi
H
,
Sammalkorpi
H
,
Dávalos
V
,
Salovaara
R
,
Hemminki
A
, et al
SMAD4 levels and response to 5-fluorouracil in colorectal cancer
.
Clin Cancer Res
2005
;
11
:
6311
6
.
29.
Mellacheruvu
D
,
Wright
Z
,
Couzens
AL
,
Lambert
JP
,
St-Denis
NA
,
Li
T
, et al
The CRAPome: a contaminant repository for affinity purification-mass spectrometry data
.
Nat Methods
2013
;
10
:
730
6
.
30.
Sztupinszki
Z
,
Győrffya
B
. 
Colon cancer subtypes: concordance, effect on survival and selection of the most representative preclinical models
.
Sci Rep
2016
;
6
:
37169
.
31.
Lanczky
A
,
Nagy
A
,
Bottai
G
,
Munkacsy
G
,
Paladini
L
,
Szabo
A
, et al
miRpower: a web-tool to validate survival-associated miRNAs utilizing expression data from 2,178 breast cancer patients
.
Breast Cancer Res Treat
2016
;
160
:
439
46
.
32.
Zhou
S
,
Buckhaults
P
,
Zawel
L
,
Bunz
F
,
Riggins
G
,
Dai
JL
, et al
Targeted deletion of Smad4 shows it is required for transforming growth factor beta and activin signaling in colorectal cancer cells
.
Proc Natl Acad Sci USA
1998
;
95
:
2412
6
.
33.
Wong
MP
,
Cheung
N
,
Yuen
ST
,
Leung
SY
,
Chung
LP
. 
Vascular endothelial growth factor is up-regulated in the early pre-malignant stage of colorectal tumour progression
.
Int J Cancer
1999
;
81
:
845
50
.
34.
Cao
D
,
Hou
M
,
Guan
YS
,
Jiang
M
,
Yang
Y
,
Gou
HF
. 
Expression of HIF-1alpha and VEGF in colorectal cancer: association with clinical outcomes and prognostic implications
.
BMC Cancer
2009
;
9
:
432
.
35.
Rasheed
S
,
Harris
AL
,
Tekkis
PP
,
Turley
H
,
Silver
A
,
McDonald
PJ
, et al
Hypoxia-inducible factor-1alpha and -2alpha are expressed in most rectal cancers but only hypoxia-inducible factor-1alpha is associated with prognosis
.
Br J Cancer
2009
;
100
:
1666
73
.
36.
Sarbassov
DD
,
Guertin
DA
,
Ali
SM
,
Sabatini
DM
. 
Phosphorylation and regulation of Akt/PKB by the rictor-mTOR complex
.
Science
2005
;
307
:
1098
101
.
37.
Kaizuka
T
,
Hara
T
,
Oshiro
N
,
Kikkawa
U
,
Yonezawa
K
,
Takehana
K
, et al
Tti1 and Tel2 are critical factors in mammalian target of rapamycin complex assembly
.
J Biol Chem
2010
;
285
:
20109
16
.
38.
Saxton
RA
,
Sabatini
DM
. 
mTOR signaling in growth, metabolism, and disease
.
Cell
2017
;
168
:
960
76
.
39.
Franke
TF
,
Hornik
CP
,
Segev
L
,
Shostak
GA
,
Sugimoto
C
. 
PI3K/Akt and apoptosis: size matters
.
Oncogene
2003
;
22
:
8983
98
.
40.
Walters
HE
,
Cox
LS
. 
mTORC inhibitors as broad-spectrum therapeutics for age-related diseases
.
Int J Mol Sci
2018
;
19
:
pii
:
E2325
.
41.
Brown
JS
,
Banerji
U
. 
Maximising the potential of AKT inhibitors as anti-cancer treatments
.
Pharmacol Ther
2017
;
172
:
101
15
.
42.
Barnett
SF
,
Defeo-Jones
D
,
Fu
S
,
Hancock
PJ
,
Haskell
KM
,
Jones
RE
. 
Identification and characterization of pleckstrin-homology-domain-dependent and isoenzyme-specific Akt inhibitors
.
Biochem. J
2005
;
385
:
399
408
.
43.
Werfel
TA
,
Wang
S
,
Jackson
MA
,
Kavanaugh
TE
,
Joly
MM
,
Lee
LH
, et al
Selective mTORC2 inhibitor therapeutically blocks breast cancer cell growth and survival
.
Cancer Res
2018
;
78
:
1845
58
.
44.
Jebali
A
,
Dumaz
N.
The role of RICTOR downstream of receptor tyrosine kinase in cancers
.
Mol Cancer
2018
;
17
:
39
.
45.
Mehrvarz Sarshekeh
A
,
Advani
S
,
Overman
MJ
,
Manyam
G
,
Kee
BK
,
Fogelman
DR
, et al
Association of SMAD4 mutation with patient demographics, tumor characteristics, and clinical outcomes in colorectal cancer
.
PLoS One
2017
;
12
:
e0173345
.
46.
De Roock
W
,
Claes
B
,
Bernasconi
D
,
De Schutter
J
,
Biesmans
B
,
Fountzilas
G
, et al
Effects of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in chemotherapy-refractory metastatic colorectal cancer: a retrospective consortium analysis
.
Lancet Oncology
2010
;
11
:
753
62
.
47.
Mei
Z
,
Shao
YW
,
Lin
P
,
Cai
X
,
Wang
B
,
Ding
Y
, et al
SMAD4 and NF1 mutations as potential biomarkers for poor prognosis to cetuximab-based therapy in Chinese metastatic colorectal cancer patients
.
BMC Cancer
2018
;
18
:
479
.
48.
Sparks
CA
,
Guertin
DA
. 
Targeting mTOR: prospects for mTOR complex 2 inhibitors in cancer therapy
.
Oncogene
2010
;
29
:
3733
44
.
49.
Gkountakos
A
,
Pilotto
S
,
Mafficini
A
,
Vicentini
C
,
Simbolo
M
,
Milella
M
, et al
Unmasking the impact of Rictor in cancer: novel insights of mTORC2 complex
.
Carcinogenesis
2018
;
39
:
971
80
.

Supplementary data