Induction of resistance to neurotrophic tropomyosin-receptor kinase inhibitors by HMGCS2 via a mevalonate pathway

Abstract

Introduction: A neurotrophic tropomyosin receptor kinase (NTRK)-tyrosine kinase inhibitor (TKI) has shown dramatic efficacy against malignant tumors harboring an NTRK fusion gene. However, almost all tumors eventually acquire resistance to NTRK-TKIs.

Method: To investigate the mechanism of resistance to NTRK-TKIs, we established cells resistant to three types of NTRK-TKIs (larotrectinib, entrectinib, and selitrectinib) using KM12 colon cancer cells with a TPM3-NTRK1 rearrangement.

Result: Overexpression of 3-hydroxy-3-methylglutaryl-CoA synthase 2 (HMGCS2) was observed in three resistant cells (KM12-LR, KM12-ER, and KM12-SR) by microarray analysis. Lower expression of sterol regulatory element-binding protein 2 (SREBP2) and peroxisome proliferator activated receptor α (PPARα) was found in two cells (KM12-ER and KM12-SR) in which HMGCS2 was overexpressed compared to the parental KM12 and KM12-LR cells. In resistant cells, knockdown of HMGCS2 using small interfering RNA improved the sensitivity to NTRK-TKI. Further treatment with mevalonolactone after HMGCS2 knockdown reintroduced the NTRK-TKI resistance. In addition, simvastatin and silibinin had a synergistic effect with NTRK-TKIs in resistant cells, and delayed tolerance was observed after sustained exposure to clinical concentrations of NTRK-TKI and simvastatin in KM12 cells. In xenograft mouse models, combination treatment with entrectinib and simvastatin reduced resistant tumor growth compared with entrectinib alone.

Conclusion: These results suggest that HMGCS2 overexpression induces resistance to NTRK-TKIs via the mevalonate pathway in colon cancer cells. Statin inhibition of the mevalonate pathway may be useful for overcoming this mechanistic resistance.

Keywords: HMGCS2; NTRK‐TKI; mevalonate pathway; resistance mechanism; statin.

FIGURE 1

(A) Establishment of cell lines resistant to NTRK‐TKIs (larotrectinib, KM12‐LR; entrectinib, KM12‐ER; selitrectinib, KM12‐SR). The results of the cell viability assay are shown. Each resistant cell line exhibited cross‐resistance to NTRK‐TKIs. (B) Phosphorylation of TRK and downstream signals of the parent and resistant cells were analyzed by western blotting method. Phosphorylation of TRK remained in KM12‐LR, but was lost in KM12‐ER and KM12‐SR. The downstream signals showed activation of ERK phosphorylation in KM12‐ER, but no similar trend was observed in the other resistant cells, and phosphorylation of AKT showed no apparent change in resistant cells compared to parental cell. Protein expression of KM12 parental cell and each resistance cell assessed using Western blotting method was quantified and confirmed, respectively. The band densities were semiquantitatively analyzed using Image J V.1.8.0 (*p < 0.05). (C) Heatmap showed genes with a log2 fold change greater than 1 or less than−1 in common two or more resistant cells in Microarray conducted in NTRK‐TKI–resistant cells compared with parental cells using a log2 ratio [log2 (NTRK‐TKI resistant KM12/parental KM12)] HMGCS2 was commonly upregulated in the three types of resistant cells after assuming the cutoff for log2 fold change was 2.0. (D) The expression of HMGCS2 was increased in NTRK‐TKI‐resistant cells compared with that in parental KM12 cells, as determined by quantitative real‐time reverse transcription–PCR (*p < 0.05). (E) The protein levels of factors related HMGCS2 were analyzed by western blotting. NTRK‐TKI‐resistant cells showed HMGCS2 overexpression. SREBP2 and PPARα related to mevalonate pathway were commonly decreased in KM12‐ER and KM12‐SR which showed stronger HMGCS2 overexpression. Factors related to ketogenesis and fatty acid synthesis showed no common trend or correlation with HMGCS2 expression in the three resistant cells. Protein expression of KM12 parental cell and each resistance cell assessed using Western blotting method was quantified and confirmed, respectively. the band densities were semiquantitatively analyzed using Image J V.1.8.0 (*p < 0.05).

FIGURE 2

(A) Cell viabilities were significantly reduced in KM12‐ER and KM12‐SR, which had decreased HMGCS2 protein levels in response to HMGCS2 siRNA, after NTRK‐TKI treatment. NTRK‐TKIs were used at 100 nM, respectively, for 72 h (*p < 0.05). (B) After transfection with the two types of siHMGCS2, a decrease in HMGCS2 protein expression and the effect of HMGCS2 knockdown by siRNA to NTRK‐TKI treatment was investigated by evaluating the expression of related protein factors by western blotting. Each NTRK‐TKI treatment was administered at a concentration of 1 μM for 24 h. After HMGCS2 knockdown by siRNA, cleaved PARP levels increased in resistant cells treated with NTRK‐TKIs. (C) An increase in cell viability assay after mevalonolactone and NTRK‐TKI treatment of resistant cells after HMGCS2 knockdown compared to treatment with NTRK‐TKI alone (*p < 0.05). NTRK‐TKIs were used at 100 nM, and mevalonolactone was used at 25 μM for 72 h. (D) After mevalonolactone treatment, cleaved PARP protein levels decreased in resistant cells after HMGCS2 knockdown and NTRK‐TKI treatment. NTRK‐TKIs were used at 100 nM, and mevalonolactone was used at 25 μM for 24 h.

FIGURE 3

(A) Cell viabilities were significantly reduced in NTRK‐TKI resistant cells after simvastatin and NTRK‐TKIs treatment compared to NTRK‐TKIs alone. Treatments were with 100 nM NTRK‐TKIs (KM12‐LR: larotrectinib, KM12‐ER: entrectinib and KM12‐SR: selitrectinib), 5 μM simvastatin for 72 h (*p < 0.05). (B) Western blotting showed cleaved PARP protein levels were increased in each resistant cell line after treatment with simvastatin and NTRK‐TKIs. Treatments were: 1 μM NTRK‐TKIs (KM12‐LR: larotrectinib, KM12‐ER: entrectinib and KM12‐SR: selitrectinib) and 5 μM simvastatin for 72 h. (C) Cell viabilities were significantly reduced in NTRK‐TKI resistant cells after silibinin and NTRK‐TKIs treatment compared to NTRK‐TKIs alone. Treatments were with 100 nM NTRK‐TKIs (KM12‐LR: larotrectinib, KM12‐ER: entrectinib and KM12‐SR: selitrectinib) and 20 μM silibinin for 72 h (*p < 0.05). (D) Western blotting showed cleaved PARP protein levels were increased in each resistant cell line after treatment with simvastatin and NTRK‐TKIs. Treatments were: (KM12‐LR: larotrectinib, KM12‐ER: entrectinib and KM12‐SR: selitrectinib) 1 μM NTRK‐TKIs and 20 μM silibinin for 72 h. NC, negative control.

FIGURE 4

(A–C) Cell viabilities in parental KM12 cells incubated with each NTRK‐TKIs alone and each NTRK‐TKIs plus simvastatin or silibinin for one month and two months were shown. (A: Larotrectinib, B: entrectinib and C: selitrectinib) Cells were continuously incubated with each NTRK‐TKI (50 nM) and simvastatin (25 nM), respectively, which are close to the clinical C max concentration used in human. Silibinin were administered to cells continuously with each NTRK‐TKI at 40 μM. KM12 parental cells treated with combination therapy used each NTRK‐TKIs and simvastatin or silibinin showed a delayed tolerance compared to cells treated with each NTRK‐TKI alone.

FIGURE 5

(A) In mouse xenograft models injected with KM12 parental cells, HMGCS2 expression in tumors in vehicle control group (n = 3), and those in entrectinib group (n = 3) that acquired resistance to entrectinib were compared by immunohistochemistry. Tumors that acquired resistance to entrectinib were shown to HMGCS2 overexpression. (B) Western blot analysis showed HMGCS2 overexpression in KM12 parental cells which acquired resistance to entrectinib in xenograft mouse model compared to the vehicle control. (C) Representatives of mouse xenograft models and the gross appearance of tumors excised at the end of the experiment. Mice of vehicle control group were excised tumor at Day 8, and the other three groups were excised tumor at Day 15. (D) Tumor volumes in xenograft mice transplanted with KM12‐ER cells was significantly lower in the simvastatin and entrectinib combination and entrectinib treatment groups when comparing tumor volumes on Days 7 and 14 (Vehicle control group n = 8, Simvastatin group n = 6, Entrectinib group n = 6, Simvastatin, and entrectinib group n = 8). At Day 7, 14 after started treatment, combination treatment with entrectinib and simvastatin significantly reduced tumor volume compared to treatment with entrectinib alone (*p < 0.05). E, Entrectinib group; S, Simvastatin group; SE Simvastatin and entrectinib group; V, Vehicle control group.

FIGURE 6

Schematic result in this study. HMGCS2 induces resistance to NTRK‐TKI via the mevalonate pathway, suggesting that simvastatin may overcome this resistance.