Synthetic small interfering RNA targeting heat shock protein 105 induces apoptosis of various cancer cells both in vitro and in vivo

Abstract

We previously reported that heat shock protein 105 (HSP105), identified by serological analysis of a recombinant cDNA expression library (SEREX) using serum from a pancreatic cancer patient, was overexpressed in various human tumors and in the testis of adult men by immunohistochemical analysis. In the present study, to elucidate the biological function of the HSP105 protein in cancer cells, we first established NIH3T3 cells overexpressing murine HSP105 (NIH3T3-HSP105). The NIH3T3-HSP105 cells acquired resistance to apoptosis induced by heat shock or doxorubicin. The small interfering RNA (siRNA)-mediated suppression of HSP105 protein expression induced apoptosis in human cancer cells but not in fibroblasts. By a combination of siRNA introduction and doxorubicin or heat shock treatment, apoptosis was induced synergistically in a human colon cancer cell line, HCT116. In vivo, siRNA inoculation into the human gastric cancer cell line KATO-3 established in the flank of an NOD SCID mouse suppressed the tumor growth. This siRNA-induced apoptosis was mediated through caspases, but not the p53 tumor suppressor protein, even though the HSP105 protein was bound to wild-type p53 protein in HCT116 cells. These findings suggest that the constitutive overexpression of HSP105 in cancer cells is involved in malignant transformation by protecting tumor cells from apoptosis. HSP105 may thus be a novel target molecule for cancer therapy and a treatment regimen using synthetic siRNA to suppress the expression of HSP105 protein may provide a new strategy for cancer therapy.

Figure 1

Anti‐apoptotic effect of HSP105 overexpression on NIH3T3 cells. (A) Western blot analysis of HSP105 expressed in NIH3T3 cells transfected with pCAGGS‐IRES‐neo‐R or pCAGGS‐IRES‐neo‐R‐HSP105. β‐Actin is shown as a control for the equal loading of protein. (B–D) Flow cytometric analyses of apoptotic cells. NIH3T3‐mock cells and NIH3T3‐HSP105 cells were treated at 45°C for 90 min (B) or with 200 ng/mL doxorubicin (C,D). To detect early apoptotic cells, the cells were harvested at the times indicated, stained with fluorescein‐isothiocyanate–annexin V and analyzed by flow cytometry (C,D). These data are representative of at least three independent experiments. Percentages shown in the panel indicate percentage of annexin V‐positive cells in heat‐treated cells (B) and doxorubicin‐treated cells (C). (D) Detection of DNA fragmentation by propidium iodide staining. The percentage of sub‐G₁ fractions at the times indicated is shown, and the representative data of flow cytometric analysis at 48 h is shown in the panel. Data are mean ± SD (n = 3). The asterisk indicates that the difference in the percentages of the sub‐G₁ fractions is statistically significant between the two values indicated by lines (P < 0.001).

Figure 2

Small interfering RNA (siRNA)‐mediated inhibition of HSP105 expression enhanced the apoptotic cell death of human cancer cell lines. (A) Reverse transcription–polymerase chain reaction analysis of HSP105 mRNA expression in normal human colon epithelium, and in human cancer cell lines HCT116 and SW620. (B) Light microscopic pictures of HCT116 cells introduced with or without siRNA and representative flow cytometric analysis data of apoptotic cells stained with annexin V at 48 h after transfection. (C,D) Western blot analysis of HSP105 protein expression and flow cytometric analysis of apoptotic cells. HCT116 cells and SW620 cells were treated with oligofectamine, control siRNA, HSP105‐siRNA (100 nM or 200 nM) or HSP105‐siRNA‐2 (B,C). (C) For western blot analysis, the cells were lysed at 24 or 48 h after transfection and analyzed. β‐Actin is shown as a quantitative control. For flow cytometric analysis, cells were harvested at 24 or 48 h after transfection and then stained with fluorescein‐isothiocyanate–annexin V and analyzed by flow cytometry. (D) Western blot analysis of HSP105 protein expression in cancer cell lines including SK‐Hep1, PK8, KATO‐3 and MKN28, and flow cytometric analysis of apoptotic cells at 48 h after transfection with 100 nM green fluorescent protein siRNA (□) or HSP105 siRNA (▪). Data are the mean of three independent experiments ± SD. The asterisks indicate that the differences in the percentages of annexin V‐positive cells are statistically significant between the two values indicated by lines (*P < 0.01; **P < 0.001).

Figure 3

No apoptosis‐inducing effects of HSP105 small interfering RNA (siRNA) on human fibroblasts. (A) Western blot analysis of HSP105. The lysates of human fibroblasts Turu and Mori, and human cancer cell lines HCT116, SW620, SK‐Hep1, PK8, KATO‐3 and MKN28 were used and blotted with an anti‐HSP105 antibody. β‐Actin is shown as a quantitative control. (B) Light microscopic pictures of Turu at 72 h after transfection with 100 nM green fluorescent protein (GFP) siRNA or HSP105 siRNA. (C) Effects of siRNA on Turu and Mori. Western blot analysis of HSP105 and flow cytometric analysis of apoptotic cells detected by annexin V staining at 72 h after transfection of 100 nM GFP siRNA or HSP105 siRNA. These data are representative of at least three independent experiments. The percentages shown in the panel indicate percentage of annexin V‐positive cells in HSP105 siRNA‐treated cells.

Figure 4

The inhibitory effect of HSP105 small interfering RNA (siRNA) on the growth of established tumors in mice. (A) KATO‐3 cells (2 × 10⁶) were implanted subcutaneously into the dorsal skin of NOD SCID mice to establish growing tumors, and siRNA was injected into the tumors every 3 days (indicated by arrows). The tumor volume was measured and plotted (luciferase siRNA, ▴ HSP105‐siRNA, ▪). Data are mean ± SD (n = 4). The asterisk indicates that the difference in the tumor volume on day 15 is statistically significant between the two values as indicated by lines (P < 0.01).

Figure 5

The synergistic effect of HSP105 small interfering RNA (siRNA) with doxorubicin or heat shock regarding the induction of apoptotic cell death in HCT116 cells. At 12 h after transfection with 100 nM siRNA, the cells were incubated with 200 ng/mL doxorubicin (A) or treated with heat shock at 45°C for 30 min (B). Subsequently, the cells were stained with fluorescein‐isothiocyanate–annexin V and analyzed by flow cytometry. Data are the mean of three independent experiments ± SD (n = 3). The asterisks indicate that the differences in the percentages of annexin V‐positive cells are statistically significant between the two values as indicated by lines (P < 0.001).

Figure 6

Caspase‐dependent and p53‐independent induction of apoptosis in HCT116 cells administered HSP105 small interfering RNA (siRNA). (A) Western blot analysis of poly ADP‐ribose polymerase (PARP) expression and (B) flow cytometric analysis of apoptosis induced in HCT116 cells transfected with siRNA in the presence of dimethylsulfoxide (DMSO) or Z‐VAD‐FMK. HCT116 cells treated with luciferase siRNA + DMSO, HSP105 siRNA + DMSO or HSP105 siRNA + 100 µM Z‐VAD‐FMK were cultured for 48 h and apoptotic cells were stained with annexin V. Cells were lysed and blotted with either anti‐HSP105 or anti‐PARP antibody. Data are the mean values of three independent experiments ± SD. The asterisks indicate that the differences in the percentages of annexin V‐positive cells were statistically significant between the two values as indicated by lines (P < 0.001). (C,D) Western blot analysis of HSP105 and p53. HCT116 cells were lysed and immunoprecipitated with an anti‐HSP105 antibody or an anti‐p53 antibody (DO‐1), and the proteins were blotted with either anti‐HSP105 antibody or a biotin‐labeled DO‐1. The immunoprecipitates with rabbit polyclonal IgG and mouse monoclonal IgG2a were used as negative controls for anti‐HSP105 antibody and DO‐1, respectively (C). HCT116 cells transfected with luciferase siRNA or HSP105 siRNA were lysed at 48 h after transfection and blotted with the anti‐HSP105 antibody DO‐1, anti‐phospho‐p53 (Ser46) and antiphospho‐p53 (Ser15). Ultraviolet light‐irradiated HCT116 cell lysates were used as a positive control for p53 phosphorylation. (D) β‐Actin is shown as a quantitative control. (E) Reverse transcription–polymerase chain reaction (RT‐PCR) analysis of HSP105, p53, Bax, NOXA and PUMA expression in HCT116 cells transfected with siRNA. HCT116 cells transfected with luciferase siRNA or HSP105 siRNA were harvested at 24 h after transfection and the cDNAs were used for PCR analysis. cDNA extracted from the untreated HCT116 cells was used as a negative control. β‐Actin is shown as a quantitative control.