A novel role of gap junction connexin46 protein to protect breast tumors from hypoxia

Abstract

Connexin proteins are the principle structural components of the gap junctions. Colocalization and tissue-specific expression of diverse connexin molecules are reported to occur in a variety of organs. Impairment of gap junctional intercellular communication, caused by mutations, gain of function or loss of function of connexins, is involved in a number of diseases including the development of cancer. Here we show that human breast cancer cells, MCF-7 and breast tumor tissues express a novel gap junction protein, connexin46 (Cx46) and it plays a critical role in hypoxia. Previous studies have shown that connexin46 is predominantly expressed in lens and our studies find that Cx46 protects human lens epithelial cells from hypoxia induced death. Interestingly, we find that Cx46 is upregulated in MCF-7 breast cancer cells and human breast cancer tumors. Downregulation of Cx46 by siRNA promotes 40% MCF-7 cell death at 24 hr under hypoxic conditions. Furthermore, direct injection of anti-Cx46 siRNA into xenograft tumors prevents tumor growth in nude mice. This finding will provide an exciting new direction for drug development for breast cancer treatment and suggests that both normal hypoxic tissue (lens) and adaptive hypoxic tissue (breast tumor) utilize the same protein, Cx46, as a protective strategy from hypoxia.

Figure 1

Cx46 downregulated Human lens epithelial cells (HLEC) are susceptible to hypoxia induced death. (a) The upregulation of Cx46 in response to 1% O₂ in rabbit lens epithelial NN 1003A cells. Cells were subjected to hypoxia or normoxia and harvested at 1–7 days. Cx46 protein level was analyzed by western blot. Tubulin blot acted as a loading control. (b) siRNA knockdown of Cx46 in HLEC. 2 ×10⁶ cells were transfected with 512 ng anti-Cx46 siRNA (final conc. 10nM) or 512ng negative non-silencing siRNA and level of knockdown was determined by western blot at 24 and 48 hr after transfection. Cx50 blot was done to show specific action of anti Cx46-siRNA. (c) Densitometric analyses showed effective knockdown (more than 85%) of Cx46 was achieved using anti-Cx46 siRNA. (d) The effect of siRNA mediated downregulation of Cx46 on cell viability of HLEC under hypoxic conditions (1% O₂, 5% CO₂). 3 ×10⁴ HLEC (per well of 96 well micro-titer plate) were treated with 10 ng (per well, final conc.10nM) of anti-Cx46 siRNA or non-silencing siRNA or no siRNA (control) and pre-incubated for 24 hr at normoxia conditions (21% O₂, 5% CO₂). Following that the cells were incubated at 1% O₂ and cell viability was measured by fluorometric resazurin reduction method at 6, 12, 18 and 36 hr intervals. Data are represented as mean ± s.e.m of three independent experiments. Asterisk indicates significant statistical difference (P<0.01) between indicated data and control.

Figure 2

Overexpression of Cx46 in hypoxia sensitive neuronal N2A cells confers survival in hypoxia (1% O₂). (a) Western blot analyses of Cx46-GFP and Cx43-GFP proteins overexpressed in neuronal N2A cells. The blot was probed with antibody against GFP. Stable line of N2A cells overexpressing Cx46-GFP or Cx43-GFP were generated by transfecting the cells with Cx46 or Cx43 rat cDNA cloned in pEGFP-N3 vector and then growing the cells for six weeks in the presence of G418 antibiotic. (b) 3 ×10⁴ wild type N2A cells or N2A cells overexpressing Cx46 or Cx43 were incubated under hypoxic conditions (1% O₂, 5% CO₂). The viability was assessed every 4 h for up to 24 hr by fluorometric resazurin reduction assay. The data are plotted as mean ± s.e.m of three independent experiments. Asterisk indicates the statistical significance (P<0.01) between indicated data and control (N2A-WT).

Figure 3

Cx46 is expressed in human breast cancer cells and human breast tumors. Breast cancer cell MCF-7 express Cx46 protein as determined by western blot (a) and RT-PCR study (b). HMEC: Human mammary epithelial cells, (c) Immunohistochemistry showing the expression of Cx46 protein in pre-metastatic breast tumors. The blue arrow indicates Cx46 staining. The characterization of this tumor breast tissue is infiltrating ductal carcinoma, grade 3, T2NOMO. (d) Western blot analyses showing Cx46 protein is upregulated in pre-metastatic breast tumor tissue. Tumor breast tissue 1: Infiltrating Ductal Carcinoma, grade 2, stage IIA. T2N0M0, source: female, 42 years; Tumor breast tissue 2: Invasive Ductal Carcinoma, grade 2, stage IIA. T2N0M0, source: female, 54 years.

Figure 4

Downregulation of Cx46 in breast cancer MCF-7 cells confers susceptibly to hypoxia induced death. (a) The siRNA mediated selective downregulation of Cx46 protein in MCF-7 cells. 2 × 10⁶ MCF-7 cells were transfected with 512 ng anti-Cx46 siRNA (final conc. 10nM) or negative non-silencing siRNA (b). Level of knockdown was determined by western blot by probing with anti-Cx46 antibody at 12, 24 and 48 h after transfection. The same blots were probed with anti-tubulin antibody to show equal loading of protein in each lane. Cx26 blot was done to show the absence of any non-specific action of anti-Cx46 siRNA in MCF-7 cells. (c) Densitometric analyses shows maximum knockdown was achieved between 24–48 hr time period. (d) Downregulation of Cx46 in MCF-7 cells decreased cell viability under hypoxic conditions 3 ×10⁴ MCF-7 cells (per well of 96 well micro-titer plate) were pre-incubated with 10ng (per well, final conc. 10nM) negative non-silencing siRNA or anti-Cx46 siRNA for 24 h at normoxic conditions (21% O₂, 5% CO₂) and then cells were kept at hypoxic conditions (1% O₂, 5% CO₂). Cell viability was assessed every 12 h intervals for up to 24 h. The data are represented as mean ± s.e.m of three independent experiments. The asterisk indicates significant statistical difference (P<0.01) between indicated data and control.

Figure 5

siRNA mediated downregulation of Cx46 in mouse tumor xenografts inhibits the growth of breast cancer tumors. (a) Mice were implanted with 17β-estradiol (1.7 mg/pellet) followed by the injection of 1 × 10⁷ MCF-7 cells subcutaneously into the inguinal region of mammary fat pads to develop breast tumors. After two weeks, animals were injected with 7.5 ug Cx46 siRNA (n=10 tumor) or 7.5 ug negative control siRNA (n=7 tumors) every 48 h for 10 days. The siRNAs were injected directly at two-three different locations of each tumor. Animals with no siRNA treatment represent control (n=6 tumor). The tumors size was measured in two dimensions by a caliper every alternate day prior to injection. The results after 10 days (total five injections) show a significant decrease in tumor growth of Cx46 siRNA treated mice. Data are plotted as mean ± s.e.m of two different independent experiments. Asterisk indicates statistical difference (P<0.001) between anti-Cx46 siRNA treated tumors and control (no siRNA) tumors. (b) The images of control (no treatment), anti-Cx46 siRNA and negative control siRNA treated MCF-7 tumor-bearing xenograft mice at day 0 and day 10. The reduction of tumor size at day 10 is observed only for anti-Cx46 siRNA treatment. (c) The gross images of control (no siRNA), anti-Cx46 siRNA treated or non-silencing siRNA treated (negative control) xenograft tumors in Nu/Nu mice. Mice were euthanized and tumors were isolated after 10 days of siRNA treatment. (d) Western blot analysis shows knockdown of Cx46 in xenograft tumors by anti-Cx46 siRNA. All tumors were dissected out at day 10. Tumors were harvested and western blot was performed on whole tissue lysate using rabbit anti-Cx46 antibody. The same blot was reprobed with anti-Cx26 antibody to show the specificity of anti-Cx46 siRNA action in vivo.