Targeting Heat Shock Transcription Factor 4 Enhances the Efficacy of Cabozantinib and Immune Checkpoint Inhibitors in Renal Cell Carcinoma

Abstract

Recently, immune checkpoint inhibitors (ICIs) and cabozantinib, a tyrosine kinase inhibitor (TKI), have been used to treat renal cell carcinoma (RCC); the combination of these agents has become a standard treatment for RCC. TKIs generally target vascular endothelial growth factor. However, cabozantinib is characterized by its targeting of MET. Therefore, cabozantinib can be used as a late-line therapy for TKI-resistant RCC. According to data from The Cancer Genome Atlas (TCGA), heat shock transcription factor 4 (HSF4) expression is higher in RCC tissues than in normal renal tissues. HSF4 binds to the MET promoter in colorectal carcinoma to enhance MET expression and promote tumor progression. However, the functional role of HSF4 in RCC is unclear. We performed loss-of-function assays of HSF4, and our results showed that HSF4 knockdown in RCC cells significantly decreased cell functions. Moreover, MET expression was decreased in HSF4-knockdown cells but elevated in sunitinib-resistant RCC cells. The combination of cabozantinib and HSF4 knockdown reduced cell proliferation in sunitinib-resistant cells more than each monotherapy alone. Furthermore, HSF4 knockdown combined with an ICI showed synergistic suppression of tumor growth in vivo. Overall, our strategy involving HSF4 knockdown may enhance the efficacy of existing therapies, such as cabozantinib and ICIs.

Keywords: MET; cabozantinib; heat shock transcription factor 4; immune checkpoint inhibitor; renal cell carcinoma.

Figure 1

Expression levels of HSF4 in RCC tissue and normal renal tissue. (A) In silico analysis of mRNA expression levels in the KIRC group from TCGA data suggested that HSF4 was a good target gene for our experiments. Mann–Whitney U tests were used to test differences between the two groups. Cliff’s delta was calculated for the effect size test. The analyses were conducted using R programming language version 4.4.2. (B) Overall survival in renal clear-cell carcinoma (KIRC) cohorts based on TCGA data. HSF4 expression was compared with that in normal samples (n = 156; Mann–Whitney U test). (C) HSF4 mRNA expression levels were compared using RT-qPCR of total RNA samples from normal kidneys and total RNA extracted from the human RCC cell lines 786-O, A498, Caki1, and Caki2. Each experiment was repeated at least three times. The Bonferroni/Dunn method was used as a multiple comparisons test.

Figure 2

The effects of HSF4 knockdown using si-HSF4 in RCC cells. (A) HSF4 protein expression levels were decreased in HSF4-knockdown A498 and Caki2 cells, as determined by Western blotting. The values listed are in comparison with β-actin expression. (B) Cell proliferation measured using XTT assays. (C) Cell migration was measured using wound healing assays. (D) Cell invasion was evaluated using Matrigel invasion assays. Infiltrating cells were counted and compared with parental and HSF4-knockdown RCC cells. (E) Tumor mass formation was measured using tumor spheroid formation assays. The percentage of viable cells was measured as luminosity in proportion with ATP production. Each of the experiments was repeated at least three times. In all experiments, si-HSF4 transfectants were compared with mock transfection. The Bonferroni/Dunn method was used as a multiple comparisons test.

Figure 3

RNA sequencing analysis and cell apoptosis assays. (A) RNA sequencing was performed using total RNA extracts from HSF4-knockdown RCC cells and parental cell lines to compare gene expression. Gene set analysis was performed using the BioPlanet gene set in GeneCordis 4. (B) Flow cytometry was used to assess the percentages of apoptotic cells among RCC cells transfected with si-HSF4 or si-Control (Ctr). Each of the experiments was repeated at least three times. The Bonferroni/Dunn method was used as a multiple comparisons test. (C) Western blotting analysis was performed to compare the expression levels of cleaved caspase 3 in HSF4-knockdown RCC cells and the parental cell lines.

Figure 3

RNA sequencing analysis and cell apoptosis assays. (A) RNA sequencing was performed using total RNA extracts from HSF4-knockdown RCC cells and parental cell lines to compare gene expression. Gene set analysis was performed using the BioPlanet gene set in GeneCordis 4. (B) Flow cytometry was used to assess the percentages of apoptotic cells among RCC cells transfected with si-HSF4 or si-Control (Ctr). Each of the experiments was repeated at least three times. The Bonferroni/Dunn method was used as a multiple comparisons test. (C) Western blotting analysis was performed to compare the expression levels of cleaved caspase 3 in HSF4-knockdown RCC cells and the parental cell lines.

Figure 4

HSF4 regulated MET expression and promoted tumor progression in RCC cells. (A) Western blotting analysis of MET expression in HSF4-knockdown RCC cells and the parental cell lines. (B) Western blotting analysis of MET and HSF4 expression in sunitinib-resistant RCC cells and the parental cells. (C) Cell proliferation assays were performed on sunitinib-resistant RCC cells using si-HSF4, si-Ctr, and cabozantinib. The percentages of viable cells were measured as luminosity in proportion to ATP production. The Bonferroni/Dunn method was used as a multiple comparisons test. The experiment was repeated at least three times. (D) MET mRNA expression levels were compared using RT-qPCR of total RNA extracted from cells in each group. Each experiment was repeated at least three times. The Bonferroni/Dunn method was used as a multiple comparisons test. (E) Western blotting analysis of MET expression in cells in each group.

Figure 5

Generation of shRNA-transfected Renca cells and xenograft assays in mice. (A) HSF4 protein expression levels were decreased in HSF4-knockdown Renca cells, as determined by Western blotting. (B) The animal model was set up as follows (n = 3 mice per group): group 1, mice injected with parental Renca cells and administered PBS; group 2, mice injected with parental Renca cells and administered the ICI; group 3, mice injected with HSF4-knockdown Renca cells and administered PBS; and group 4, mice injected with HSF4-knockdown Renca cells and administered the ICI. (C) Comparison of tumor volumes in mice. The Bonferroni/Dunn method was used as a multiple comparisons test. (D) Photograph showing Ki67-positive cells in the excised tumor stained using immunohistochemistry (400×). The percentages of Ki67-positive cells were analyzed using the Bonferroni/Dunn method.

Figure 5

Generation of shRNA-transfected Renca cells and xenograft assays in mice. (A) HSF4 protein expression levels were decreased in HSF4-knockdown Renca cells, as determined by Western blotting. (B) The animal model was set up as follows (n = 3 mice per group): group 1, mice injected with parental Renca cells and administered PBS; group 2, mice injected with parental Renca cells and administered the ICI; group 3, mice injected with HSF4-knockdown Renca cells and administered PBS; and group 4, mice injected with HSF4-knockdown Renca cells and administered the ICI. (C) Comparison of tumor volumes in mice. The Bonferroni/Dunn method was used as a multiple comparisons test. (D) Photograph showing Ki67-positive cells in the excised tumor stained using immunohistochemistry (400×). The percentages of Ki67-positive cells were analyzed using the Bonferroni/Dunn method.