Targeting TGF-β-Smad2/3-JNK1-mediated SIRT1 activity overcomes the chemoresistance of KRAS mutation lung cancer

Abstract

Patients with lung cancer harboring a KRAS oncogenic driver mutation have a very poor prognosis. Recently, we reported that SIRT1 is upregulated by the KRAS^Mut-c-Myc axis, and that KRAS^Mut-induced SIRT1 is stably deacetylated at lysine 104, which in turn increases KRAS^Mut activity and enhances chemoresistance. Notably, SIRT1 activity as well as SIRT1 levels are more elevated in KRAS^Mut cells compared with EGFR^Mut, KRAS^Mut- and EGFR^Mut-negative cells, and nontumorigenic cells. This prompted us to investigate the mechanism by which SIRT1 activity was increased and the role of pSIRT1 in the chemoresistance of KRAS^Mut lung cancer cells. The activated MEK-ERK pathway under KRAS^Mut increased AP-1 transcription activity, which in turn enhanced TGF-β1 secretion. The secreted TGF-β1 activated the Smad2/3-JNK1 signaling pathway in an autocrine manner, increasing pSIRT1^S27 and pSIRT1^S47, ultimately enhancing KRAS^Mut activity through KRAS deacetylation and affecting chemoresistance. We identified a small molecule from the natural compound library-Kuwanon C (KWN-C), a SIRT1 activity inhibitor-which reduced pSIRT1^S27 and pSIRT1^S47 levels via a decrease in the activity of the TGF-β1--Smad2/3-JNK1 signaling pathway. Treatment with the SIRT1 activity inhibitor triggered the anticancer effects of cisplatin and pemetrexed in human lung cancer cells, lung orthotopic tumors and a spontaneous in vivo model of KRAS^Mut lung cancer. Our findings reveal a novel pathway critical for the regulation of SIRT1 activity in KRAS^Mut lung cancer and provide important evidence for the potential application of SIRT1 activity inhibitors as an adjuvant chemotherapy, overcoming chemoresistance in patients with KRAS^Mut lung cancer.

Fig. 1. KRAS^Mut regulates SIRT1 expression in NSCLC cells both in vitro and in vivo.

a Normal human bronchial epithelial cell (BEAS-2B), KRAS^Mut cell lines (H358 and H460), KRAS^WT and EGFR^WT cell lines (HCC1666 and H522), and EGFR^Mut cell lines (HCC827 and PC9) were collected with lysis buffer, and immunoblotted with anti-SIRT1 and β-actin antibody. b Immunohistochemical staining for SIRT1 with the lung from LSL-Kras^G12D Tg mouse at 16 weeks after administration of adenovirus Cre recombinase induction. Representative images are shown. Scale bar, 100 μm. High-magnification images correspond to the areas marked by the black box. c SIRT1 mRNA expression was measured by RT–qPCR with the same cell lines as in a. RPL32 was used as internal control and for normalization. Student’s t-test, mean ± s.d.; n = 6, *P < 0.05. d The mRNA expression of Sirt1 was analyzed by RT–qPCR with the cancerous and adjacent noncancerous lung tissues of LSL-Kras^G12D Tg mouse. Rpl32 was used as internal control and for normalization. Student’s t-test, mean ± s.e.m.; n = 6, *, P < 0.05. e H358 and H460 cells were transfected with the 2-μg plasmids of pcDNA, KRAS^WT and KRAS^G12D. The cells were collected with cell lysis buffer and subjected to western blotting with anti-KRAS, anti-SIRT1 and β-actin antibodies. f H358 and H460 cells were transfected with siCon and siKRAS (80 nM). The cells were collected with cell lysis buffer and subjected to western blotting with same antibodies in e. g H358 and H460 cell extracts were immunoprecipitated with anti-KRAS and anti-SIRT1, and immunoblotted with anti-KRAS, anti-SIRT1 and β-actin antibodies.

Fig. 2. KRAS^Mut-induced SIRT1 contributes to chemoresistance in KRAS^Mut NSCLC cells.

a,b H358 and H460 cells were transfected with siCon and siSIRT1 and then treated with CP and MTA (0.001–100 μM). Cell proliferation was assessed using the MTS assay 3 days after the drug treatment. IC₅₀ of CP and MTA was calculated on the basis of cell viability. Student’s t-test, mean ± s.d.; n = 6, *P < 0.05. c,d H358 and H460 cells were transfected and treated with CP (H358 1 μM and H460 5 μM) and MTA (H358 10 μM and H460 5 μM). The cells were then seeded with 0.5% top agar and cultured in a mixture of medium, CP and MTA. Cell colonies were stained with crystal violet and counted per 3.8 cm². Student’s t-test, mean ± s.d.; n = 6, *P < 0.05.

Fig. 3. SIRT1 activity is increased by JNK1 in KRAS^Mut cells.

a KRAS^Mut cell lines (H358, A427, NCIH727, NCIH23 and SKLU-1), EGFR^Mut cell lines (HCC827, HCC2279, H1650 and H1975), KRAS^WT and EGFR^WT cell lines (H322M, H522, Calu-3 and HCC1666) and nontumorigenic cells (HEK-293T and BEAS-2B) were collected with lysis buffer, and SIRT1 activity was measured with cell lysates. Student’s t-test, mean ± s.d.; n = 6, *P < 0.05. b Normal human bronchial epithelial cell (BEAS-2B) and KRAS^Mut cell lines used in a were collected with lysis buffer, and immunoblotted with anti-pMKK4, MKK, pMKK7, MKK, pJNK1, JNK1 and β-actin antibodies. c KRAS^Mut cell lines (H358, A427 and NCIH727) were treated with anisomycin 38 μM (10 μg/ml) (JNK1 activator) and SP600125 20 μM (JNK1 inhibitor) for 2 h. The protein levels of pJNK1, JNK, pSIRT1^S27, pSIRT1^S47, pSIRT1^T530, SIRT1 and β-actin were measured by western blot analysis. d H358 cells were treated with anisomycin and SP600125 under the same condition as in b and then whole-cell lysates were subjected to immunoprecipitation with an anti-JNK1 antibody. Immunoblot analysis was performed using antibodies against SIRT1, pSIRT1^S27, pSIRT1^S47, pSIRT1^T530, pJNK1, JNK and β-actin antibodies. e The recombinant proteins, SIRT1 and JNK1 were incubated in the reaction mixture for phosphorylation at 32 °C for 4 h, and then Ser- and Thr-phosphorylated peptides were identified using anti-pSIRT1^S27, pSIRT1^S47, pSIRT1^T530 antibody. f H358 cells were transfected with SIRT1^WT, SIRT1^T530A, SIRT1^S27A&S47A, and then H358 cell extracts were immunoprecipitated with anti-KRAS antibody and RAF-1 agarose beads and immunoblotted with anti-acetylation, anti-SIRT1, anti-KRAS, anti-KRAS–GTP-bound and β-actin antibodies. g KRAS^Mut cell lines (H358, A427 and H727) were transfected with SIRT1^WT, SIRT1^S27A, SIRT1^S47A, SIRT1^T530A and SIRT1^S27A,S47A (2 μg), then collected with lysis buffer, and SIRT1 activity was assessed with cell lysates. Student’s t-test, mean ± s.d.; n = 6, *P < 0.05.

Fig. 4. SIRT1 activity in KRAS^Mut cells is dependent on MEK–ERK–AP-1 pathway-mediated TGF-β1–Smad2/3–JNK signaling, which is blocked by KWN-C.

a Normal human bronchial epithelial cell (BEAS-2B) and KRAS^Mut cell lines (H358, A427 and H727) were collected with lysis buffer and subjected to western blotting with anti-pERK, ERK and β-actin antibodies. b A luciferase assay was performed to assess the AP-1-mediated transcriptional regulatory activity with cell lysates in Fig. 4a. Student’s t-test, mean ± s.d.; n = 6, *P < 0.05. c TGFB1 mRNA expression was measured by RT–qPCR with same cell lines as in a. RPL32 was used as internal control and for normalization. Student’s t-test, mean ± s.d.; n = 6, *P < 0.05. d The medium of four cell lines were changed by FBS-free medium before cell collection at 24 h. Conditioned medium was collected and concentrated using an Amicon Ultra-15 tube, and total TGF-β1 levels were measured by ELISA. e KRAS^Mut cell lines (H358, A427 and H727) were transfected with pcDNA, KRAS G12C, G12D and G12V plasmids (2 μg). TGF-β1 levels were measured under the same method as in d. f H358, A427 and H727 cells were transplanted with pcDNA, KRAS G12C, G12D and G12V plasmids, siCon and siSmad2/3 (80 nM) for 48 h, and then the activity of Smad2/3, JNK1 and KRAS was measured. g The cell lysates of each different KRAS^Mut cell lines (H358, A427 and H727) under KWN-C with indicated dosage for 24 h were transferred by immunoblotting assay with anti-pSmad2/3, anti-Smad2/3, pJNK1, JNK1, anti-pSIRT1^Ser27, pSIRT1^Ser47, SIRT1, KRAS–GTP-bound and β-actin antibodies. h, H358, A427 and H460 cells were treated with DMSO or KWN-C (10 μM), and cell extracts were then immunoprecipitated using immunoglobulin G, anti-KRAS, and RAF-1 agarose bead antibodies. Immunoblotting was performed using anti-acetyl, anti-KRAS–GTP-bound, anti-SIRT1, anti-KRAS and β-actin antibodies.

Fig. 5. SIRT1 inhibitor synergistically decreased KRAS^Mut lung cancer proliferation combined with CP and MTA.

a H358, H460, NCIH23, SKLU-1 and SW900 cells were treated with CP (H358 1 μM, H460, NCIH23 and SKLU-1 5 μM), MTA (H358 10 μM, H460, NCIH23 and SKLU-1 5 μM) and/or KWN-C (10 μM), and cell proliferation was measured by the MTS assay 3 days after drug treatment. Student’s t-test, mean ± s.d.; n = 6, *P < 0.05. b Top: KRAS^Mut cells were seeded with 0.5% top agar and cultured in a mixture of fresh medium with drugs as described in Supplementary Fig. 5a. Cell colonies were stained with crystal violet and counted per 3.8 cm². Bottom: representative colony images are shown. Student’s t-test, mean ± s.d.; n = 6, *P < 0.05. c H358 and H460 cells were treated with CP (H358 1 μM and H460 5 μM), MTA (H358 10 μM and H460 5 μM) and/or KWN-C (10 μM). Cell lysates from drug-treated cells were incubated with Raf-1–RBD to pull down KRAS–GTP (the active form of KRAS), followed by western blotting with an anti-KRAS antibody. Expression levels of KRAS, pERK, ERK, pAkt, Akt, PARP, cleaved PARP, pro-caspase-3, cleaved-caspase-3 and β-actin in total lysates were analyzed by western blotting. d H358 and H460 cells treated as in a were assessed for apoptosis by TUNEL assay (middle row), and their nuclei were stained with DAPI (top row; scale bar, 25 μm). All figures are representative of at least three separate experiments. e H358 and H460 cells treated as in a were stained with Annexin V/PI staining for apoptosis using flow cytometric analysis. Representative flow cytometry plots. All figures are representative of at least three separate experiments.

Fig. 6. Evaluation of in vivo toxicity of SIRT1 activity inhibitor KWN-C from natural product compound screening.

a Nu/Nu nude mice were treated with increasing doses of KWN-C (0, 7.5, 15 and 30 mg/kg per day, i.p.) for 21 days (n = 6 mice per group). Body weight of mice was measured every 3 days during treatment with various doses of KWN-C. b Blood analysis of mice after treatment with various doses of KWN-C for 21 days. Student’s t-test, mean ± s.e.m.; n = 6, *P < 0.05. c H&E staining histology of brain, heart, liver, spleen, kidney, lung and intestine from mice after treatment with KWN-C dose dependently for 21 days. Scale bar, 50 μm.

Fig. 7. The combination CP, MTA and SIRT1 activity inhibitor synergistically decreases KRAS^Mut lung orthotopic tumorigenesis.

a H358 cells harboring stably expressed luciferase plasmid were intratracheally injected into nude mice (1 × 10⁶ cells per mouse). Top: representative bioluminescence images 2 months after the injection. The mice were euthanized 2 months after the injection, and lungs were excised and stained with Bouin’s fixative. Bottom: the lung tumor images. Therapeutic candidates were treated with CP (5 mg/kg per day, i.p.), MTA (150 mg/kg twice a week, i.p.) and/or KWN-C (30 mg/kg per day, i.p.). b The photon emission values represent the mean ± s.e.m. of the indicated number of mice. Student’s t-test, mean ± s.e.m.; n = 5, *P < 0.05. c The lung tumor weight from combination CP, MTA and/or KWN-C-treated mice was measured and compared with nontreatment, each single treatment and combined treatment. Student’s t-test, mean ± s.e.m.; n = 5, *P < 0.05. d The number of colonies formed in the lungs were measured under microscopy under the same conditions as in c. Student’s t-test, mean ± s.e.m.; n = 5, *P < 0.05. e KRAS–GTP (active form of KRAS) was pulled down by Raf-1–RBD from tumor tissue lysates, followed by western blot using KRAS antibody. Expression levels of anti-pSIRT1^S27, pSIRT1^S47, SIRT1, KRAS–GTP-bound, KRAS, pERK, ERK pAkt, Akt, PARP, cleaved PARP, pro-caspase-3, cleaved-caspase-3 and β-actin were analyzed by western blot in tumor tissues.

Fig. 8. SIRT1 activity inhibitor synergistically sensitizes the anticancer effect of CP and MTA in a KRAS^G12D spontaneous lung tumor model.

a After administration of adenovirus Cre recombinase in KRAS^G12D mice for 10 weeks, mice were treated using the same method as in Fig. 7. Representative H&E staining images and pJNK1, cleaved caspase-3, cleaved PARP, TUNEL and Ki-67 were analyzed by immunohistochemical staining in tumor tissues at the end of experiments. b Tumor tissue from each drug-treated group was collected with lysis buffer, and SIRT1 activity was measured with cell lysates. Student’s t-test, mean ± s.e.m.; n = 10, *P < 0.05. c Tumor area was quantified using ImageJ software. Student’s t-test, mean ± s.e.m.; n = 10, *P < 0.05. d Tumor numbers per lung area were counted under the microscope in specimens collected from mice treated with drugs. Student’s t-test, mean ± s.e.m.; n = 10, *P < 0.05. e Survival rates of mice treated with drugs (log-rank test). f Median survival days and P values were calculated using the log-rank test and the Gehan–Breslow–Wilcoxon test, respectively, based on Student’s t-test. g A schematic overview of the mechanism of enhanced SIRT1 activity in KRAS^Mut lung cancer and definition of a rational combination strategy between SIRT1 activity inhibitor and conventional chemotherapy.