Aicardi-Goutières syndrome gene Rnaseh2c is a metastasis susceptibility gene in breast cancer

Abstract

Breast cancer is the second leading cause of cancer-related deaths in the United States, with the majority of these deaths due to metastatic lesions rather than the primary tumor. Thus, a better understanding of the etiology of metastatic disease is crucial for improving survival. Using a haplotype mapping strategy in mouse and shRNA-mediated gene knockdown, we identified Rnaseh2c, a scaffolding protein of the heterotrimeric RNase H2 endoribonuclease complex, as a novel metastasis susceptibility factor. We found that the role of Rnaseh2c in metastatic disease is independent of RNase H2 enzymatic activity, and immunophenotyping and RNA-sequencing analysis revealed engagement of the T cell-mediated adaptive immune response. Furthermore, the cGAS-Sting pathway was not activated in the metastatic cancer cells used in this study, suggesting that the mechanism of immune response in breast cancer is different from the mechanism proposed for Aicardi-Goutières Syndrome, a rare interferonopathy caused by RNase H2 mutation. These results suggest an important novel, non-enzymatic role for RNASEH2C during breast cancer progression and add Rnaseh2c to a panel of genes we have identified that together could determine patients with high risk for metastasis. These results also highlight a potential new target for combination with immunotherapies and may contribute to a better understanding of the etiology of Aicardi-Goutières Syndrome autoimmunity.

Fig 1. Knockdown of Rnaseh2c reduces spontaneous pulmonary metastasis.

(A) Mice from the DO Generation 5 were bred to the MMTV-PyMT transgenic model to induce tumorigenesis and metastasis in the offspring, which were used to identify strain-specific haplotypes associated with metastasis. A tumor expression microarray was used to identify genes within the identified haplotypes exhibiting changes in expression associated with changes in metastasis. Rnaseh2c was one of the identified candidate genes. (B) RNASEH2C protein expression by western blot upon shRNA-mediated knockdown. One representative blot is shown, average relative densitometry for three independent experiments is reported below. (C-E) Pulmonary metastases (C) and tumor mass (D) from orthotopically injected scramble control (scr) and knockdown cells were quantified at euthanasia and normalized (metastases per gram of tumor, E); average ± standard deviation, n = 25 mice per group in three independent experiments. (F) Tail vein injection (TVI) experimental metastasis experiment using scramble and knockdown cells. Pulmonary metastases were counted at euthanasia; average ± standard deviation, n = 10 mice per group. NS—not significant.

Fig 2. Overexpression of Rnaseh2c increases pulmonary metastasis.

(A) RNASEH2C protein expression by western blot following exogenous overexpression (OE). One representative blot is shown, average relative densitometry for three independent experiments is reported below; OE lane is combined endogenous and exogenous tagged RNASEH2C expression. (B-D) OE cell line was orthotopically injected along with empty vector control (Ctrl) as described. Pulmonary metastases (B) and tumor mass (C) were quantified at euthanasia and normalized (metastases per gram of tumor, D); average ± standard deviation, n = 19 mice per group in two independent experiments. NS—not significant.

Fig 3. The metastatic effect of Rnaseh2c knockdown is not through enzyme activity.

(A) Percent ribonucleotide excision repair (RER) activity was measured in Mvt1 scr and sh4 Rnaseh2c knockdown cells. Average ± standard deviation of three experiments. (B) RNASEH2A protein expression by western blot. One representative blot is shown, average relative densitometry for two independent experiments is reported below. (C) Percent ribonucleotide excision repair (RER) activity was measured in Mvt1 scr and Rnaseh2a knockdown cells. Average ± standard deviation of three experiments. (D-F) Control and Rnaseh2a knockdown cells were used in spontaneous metastasis assays. Pulmonary metastases (D) and tumor mass (E) were quantified at euthanasia and normalized (metastases per gram of tumor, F); average ± standard deviation, n = 20 mice per group in two independent experiments. (G) Quantification of Ki67 (top) and Cleaved Caspase 3 (bottom) staining by IHC of tumor sections as the average ± standard deviation of one section from each of three independent tumors. (H) Unsynchronized cells were labeled with BrdU (left) and stained with propidium iodide (PI, right) to analyze cell cycle progression. Average ± standard deviation of duplicate samples. (I) Cells ± doxorubicin treatment (1μM for 24 hr) were analyzed for p21 protein expression in response to DNA damage. NS—not significant.

Fig 4. T cells mediate the effect of Rnaseh2c knockdown on metastasis.

(A) GSEA analysis of RNA-sequencing data comparing sh4 with scr tumors in hallmark pathways related to inflammation and interferon signaling. nom. p: nominal p-value; NES: normalized enrichment score; FDR q: false discovery rate q-value. (B) Tumor mass in athymic nude mice that received Mvt1 cells with a knockdown of Rnaseh2c; average ± standard deviation. (C) Metastasis counts for the mice from (B); average ± standard deviation, n = 20 mice per group in two independent experiments. (D) Proportion of Interferon γ (IFN-γ)-producing CD8+ T cell that were isolated from tumors of FVB mice that received orthotopic injection with either scr or sh4 cells; average ± SEM. (E) Proportion of Interferon γ (IFN-γ)-producing CD8+ T cell that were isolated from metastatic lungs of the mice shown in (D); average ± SEM. n = 5 mice per group with samples from each mouse analyzed in triplicate. NS—not significant.

Fig 5. IRF3 signaling, the proposed AGS mechanism, is not activated in Mvt1 Rnaseh2c knockdown cells.

(A) qPCR analysis of a panel of interferon stimulated genes in response to Rnaseh2c knockdown. Arrows indicate genes that are upregulated in AGS. Average ± standard error of three experiments. (B) Western blot analysis of IRF3 phosphorylation and nuclear translocation in Rnaseh2c knockdown cells following fractionation of the cytoplasmic (Cyto) and nuclear (Nuc) fractions. Sting activation using treatment with DMXAA was used as a positive control.