HMGCS2 enhances invasion and metastasis via direct interaction with PPARα to activate Src signaling in colorectal cancer and oral cancer

Abstract

Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase (HMGCS2) is the rate-limiting enzyme of ketogenesis. Growing evidence indicates that HMGCS2 may be involved in cancer progression, but its exact role is largely unknown. In this study, we demonstrate that HMGCS2 mRNA expression is associated with poor clinical prognosis and outcomes in patients with colorectal cancer (CRC) and oral squamous cell carcinoma (OSCC). In vitro, ectopic expression of HMGCS2 enhanced cancer cell motility in a ketogenesis-independent manner. Moreover, HMGCS2 promoted Src activity by directly binding to peroxisome proliferator-activated receptor alpha (PPARα), a transcriptional activator of Src. Taken together, these results suggest that HMGCS2 may serve as a useful prognostic marker and vital target for future therapeutic strategies against advanced cancer.

Keywords: CRC; HMGCS2; OSCC; PPARα; metastasis.

Figure 1. HMGCS2 expression is positively correlated with TNM stage, survival rate, and lymph node metastasis in CRC and OSC patients

(A–C) Real-time quantitative RT-PCR was performed on CRC patients’ tumors. Of the 112 CRC patients analyzed, the distributions of demographic, clinical, and pathological features are presented. (D) Patients were divided into high (fold change > cutoff values) or low (fold change ≤ cutoff values) HMGCS2 expression categories. Kaplan–Meier survival curves show that patients with low HMGCS2 expression (n = 55) survived significantly longer than those with high HMGCS2 expression did (n = 57; *P < 0.001). (E–G) Real-time quantitative RT-PCR was performed on OSCC patients’ tumors. Of the 140 OSCC patients analyzed, the distributions of demographic, clinical, and pathological features are presented. (H) Survival curves show that patients with low HMGCS2 expression survived significantly longer than those with high HMGCS2 expression did (P < 0.001).

Figure 2. Overexpression and shRNA knockdown of HMGCS2 affect cell migration and invasion abilities in CRC and OSCC cells

(A) Western blot analysis of endogenous HMGCS2 protein expression in CRC and OSCC cell lines. β-actin was used as an internal loading control (upper panel). The Boyden chamber assay was used to evaluate the invasion ability of CRC and OSCC cell lines (lower panel). (B and C) Cells were transiently transfected with control plasmids or various dosages of shHMGCS2 expression plasmids (upper panel). The Boyden chamber assay was used to evaluate the migration and invasion ability and number of migratory cells in DLD1 and SAS after transient knockdown of HMGCS2. Quantification of migratory cell number in DLD1 and SAS cells treated with shHMGCS2 expression plasmids (lower panel; *P < 0.05; **P < 0.001). (D and E) Cells were transiently transfected with control plasmids or various dosages of HMGCS2 expression plasmids (upper panel). Boyden chamber assay was used to evaluate the migration and invasion ability of migratory cells in SW480 and Cal27 after transient overexpression of HMGCS2. Quantification of the migratory cell number in SW480 and Cal27 cells treated with HMGCS2 expression plasmids (lower panel; *P < 0.05).

Figure 3. Stable knock-down of HMGCS2 moderates cell migration and invasion ability in CRC and OSCC cells

(A) Migration and invasion ability of DLD1/pLKO and DLD1/shHMGCS2 was measured with the Boyden chamber assay. Cell migration and invasion toward the lower face of the filter were observed and quantified (lower left panel; *P < 0.05). (B) Migration and invasion ability of SAS/pLKO and SAS/shHMGCS2 was measured with the Boyden chamber assay. Cell migration and invasion toward the lower face of the filter were observed and quantified (lower right panel; *P < 0.05). (C) Migration and invasion ability of SW480/Neo and SW480/HMGCS2 was measured with the Boyden chamber assay. Cell migration and invasion toward the lower face of the filter were observed and quantified (lower left panel; * P < 0.05). (D) Migration and invasion ability of Cal27/Neo and Cal27/HMGCS2 was measured with the Boyden chamber assay. Cell migration and invasion toward the lower face of the filter were observed and quantified (lower right panel; *P < 0.05). (E) Mice were injected with DLD1/pLKO (n = 8) and DLD1/shHMGCS2 (n = 7) in the spleen and their livers were excised at the days indicated. (F) Survival curves of overall survival in the hepatic animal model.

Figure 4. Ketogenesis activity does not affect cancer progression in HMGCS2 transfectants

(A) Diagram of ketogenesis. (B) Ketone body assay was used for the quantitative determination of 3-HB in stable HMGCS2/shHMGCS2 transfectants. (C) DLD1/pLKO, DLD1/shHMGCS2, SAS/pLKO, and SAS/shHMGCS2 cells were treated with 3-HB and evaluated for cell invasion by the Boyden chamber assay. The quantification of migratory cell number of DLD1/pLKO and DLD1/shHMGCS2 was conducted in a transwell (lower left panel; *P < 0.05). (D) Nucleotide sequence of the HMGCS2 with substrate binding site and active site. (E) Cells were transiently transfected with control plasmids or nonenzymatic HMGCS2 expression plasmids. The Boyden chamber assay was used to evaluate the migration and invasion ability and the number of migratory cells in SW480 and Cal27 after transient transfection with nonenzymatic HMGCS2 expression plasmids. The migratory cell number in SW480 and Cal27 cells treated with nonenzymatic HMGCS2 expression plasmids was quantified (*P < 0.05).

Figure 4. Ketogenesis activity does not affect cancer progression in HMGCS2 transfectants

(A) Diagram of ketogenesis. (B) Ketone body assay was used for the quantitative determination of 3-HB in stable HMGCS2/shHMGCS2 transfectants. (C) DLD1/pLKO, DLD1/shHMGCS2, SAS/pLKO, and SAS/shHMGCS2 cells were treated with 3-HB and evaluated for cell invasion by the Boyden chamber assay. The quantification of migratory cell number of DLD1/pLKO and DLD1/shHMGCS2 was conducted in a transwell (lower left panel; *P < 0.05). (D) Nucleotide sequence of the HMGCS2 with substrate binding site and active site. (E) Cells were transiently transfected with control plasmids or nonenzymatic HMGCS2 expression plasmids. The Boyden chamber assay was used to evaluate the migration and invasion ability and the number of migratory cells in SW480 and Cal27 after transient transfection with nonenzymatic HMGCS2 expression plasmids. The migratory cell number in SW480 and Cal27 cells treated with nonenzymatic HMGCS2 expression plasmids was quantified (*P < 0.05).

Figure 5. Src plays a role as an essential downstream effector in HMGCS2-induced cancer cell motility

(A) Heatmap of the mRNA expression profile in DLD1/pLKO and DLD/shHMGCS2 stable clones. (B) GeneGo pathway maps. Canonical pathway maps represent a set of approximately 650 signaling and metabolic maps covering human biology (signaling and metabolism) comprehensively. All maps are drawn from scratch by GeneGo annotators and manually curated and edited. The height of the histogram corresponds to the relative expression value for a particular gene/protein. (C) Top-scoring network from DLD1/shHMGCS2 versus DLD1/pLKO. Thick cyan lines indicate the fragments of canonical pathways. Upregulated genes are marked with red circles, whereas downregulated genes are marked with blue circles. The “checkerboard” color indicates mixed expression for the gene between files or between multiple tags for the same gene. (D) RT-PCR analysis of Src expression in stable HMGCS2/shHMGCS2 transfectants. GAPGH was used as an internal control for RNA quantity. (E) Cells were transiently transfected with control plasmids or various dosages of shSrc expression plasmids (upper panel). The Boyden chamber assay was used to evaluate the migration and invasion ability and number of migratory cells in SW480/HMGCS2 and Cal27/HMGCS2 after transient knockdown of Src. The migratory cell number in SW480/HMGCS2 and Cal27/HMGCS2 cells treated with shSrc expression plasmids was quantified (lower panel; *P < 0.05). (F) Quantitative RT-PCR analysis was performed to detect HMGCS2 and Src mRNA expression in CRC and OSCC patients. The data are shown as Log10 of relative quantification, and B2M was used as an endogenous normalization control.

Figure 5. Src plays a role as an essential downstream effector in HMGCS2-induced cancer cell motility

(A) Heatmap of the mRNA expression profile in DLD1/pLKO and DLD/shHMGCS2 stable clones. (B) GeneGo pathway maps. Canonical pathway maps represent a set of approximately 650 signaling and metabolic maps covering human biology (signaling and metabolism) comprehensively. All maps are drawn from scratch by GeneGo annotators and manually curated and edited. The height of the histogram corresponds to the relative expression value for a particular gene/protein. (C) Top-scoring network from DLD1/shHMGCS2 versus DLD1/pLKO. Thick cyan lines indicate the fragments of canonical pathways. Upregulated genes are marked with red circles, whereas downregulated genes are marked with blue circles. The “checkerboard” color indicates mixed expression for the gene between files or between multiple tags for the same gene. (D) RT-PCR analysis of Src expression in stable HMGCS2/shHMGCS2 transfectants. GAPGH was used as an internal control for RNA quantity. (E) Cells were transiently transfected with control plasmids or various dosages of shSrc expression plasmids (upper panel). The Boyden chamber assay was used to evaluate the migration and invasion ability and number of migratory cells in SW480/HMGCS2 and Cal27/HMGCS2 after transient knockdown of Src. The migratory cell number in SW480/HMGCS2 and Cal27/HMGCS2 cells treated with shSrc expression plasmids was quantified (lower panel; *P < 0.05). (F) Quantitative RT-PCR analysis was performed to detect HMGCS2 and Src mRNA expression in CRC and OSCC patients. The data are shown as Log10 of relative quantification, and B2M was used as an endogenous normalization control.

Figure 6. PPARα is a transcriptional co-activator of HMGCS2's up-regulation of Src expression in CRC and OSCC cells

(A) Whole-cell lysates of SW480/Neo, SW480/HMGCS2, Cal27/Neo, and Cal27/HMGCS2 were prepared for immunoprecipitation with anti-PPARα followed by immunoblotting with anti-HMGCS2. PPARα was used for the internal control. (B) Schematic representation of the Src promoter design used in the dual luciferase reporter assay. DLD1/pLKO, DLD1/shHMGCS2, SAS/pLKO, and SAS/shHMGCS2 cells were transiently cotransfected with various promoter constructs for 48 h. Luciferase activity was normalized for transfection efficiency and cell numbers against thymidine kinase (TK) activity from cotransfected TK plasmids. Results are expressed as the mean ± SD, and each experiment was performed three times in duplicate. (C) Cells were transiently transfected with control plasmids or various dosages of shPPARα expression plasmids (upper panel). The Boyden chamber assay was used to evaluate the invasion ability and migratory cell number in SW480 and Cal27 after transient transfection with various plasmids. The migratory cell number in SW480 and Cal27 cells treated with HMGCS2 expression plasmids was quantified (lower panel; * P < 0.05). (D) Real-time quantitative RT-PCR was performed on CRC (N = 43) and OSCC (N = 48) patients’ tumors. Of the CRC and OSCC patients analyzed, the distributions of demographic, clinical, and pathological features are presented. Patients were divided into high (fold change > cutoff values) or low (fold change ≤ cutoff values) HMGCS2 expression categories. Kaplan–Meier survival curves show that patients with low HMGCS2 expression survived significantly longer than those with high HMGCS2 expression did in both CRC and OSCC patients.

Figure 6. PPARα is a transcriptional co-activator of HMGCS2's up-regulation of Src expression in CRC and OSCC cells

(A) Whole-cell lysates of SW480/Neo, SW480/HMGCS2, Cal27/Neo, and Cal27/HMGCS2 were prepared for immunoprecipitation with anti-PPARα followed by immunoblotting with anti-HMGCS2. PPARα was used for the internal control. (B) Schematic representation of the Src promoter design used in the dual luciferase reporter assay. DLD1/pLKO, DLD1/shHMGCS2, SAS/pLKO, and SAS/shHMGCS2 cells were transiently cotransfected with various promoter constructs for 48 h. Luciferase activity was normalized for transfection efficiency and cell numbers against thymidine kinase (TK) activity from cotransfected TK plasmids. Results are expressed as the mean ± SD, and each experiment was performed three times in duplicate. (C) Cells were transiently transfected with control plasmids or various dosages of shPPARα expression plasmids (upper panel). The Boyden chamber assay was used to evaluate the invasion ability and migratory cell number in SW480 and Cal27 after transient transfection with various plasmids. The migratory cell number in SW480 and Cal27 cells treated with HMGCS2 expression plasmids was quantified (lower panel; * P < 0.05). (D) Real-time quantitative RT-PCR was performed on CRC (N = 43) and OSCC (N = 48) patients’ tumors. Of the CRC and OSCC patients analyzed, the distributions of demographic, clinical, and pathological features are presented. Patients were divided into high (fold change > cutoff values) or low (fold change ≤ cutoff values) HMGCS2 expression categories. Kaplan–Meier survival curves show that patients with low HMGCS2 expression survived significantly longer than those with high HMGCS2 expression did in both CRC and OSCC patients.