HER2+ breast cancers can be treated with several HER2-targeted inhibitors, but tumors often acquire or have de novo resistance to the therapies, limiting the long term efficacy of these treatments. To identify genes associated with resistance, we performed drug screens on a panel of HER2 positive breast cancer cell lines using lapatinib, then performed statistical analyses to identify differentially expressed genes. We procured siRNA against the top 20 targets expressed in the resistant subset for subsequent viability screening, both in the presence and absence of lapatinib. This screen identified INHBA, a member of the TGF-B superfamily, as a strong mediator of cell growth and lapatinib response in two resistant HER2+ cell lines. Both these lines were of the basal subtype, which tend to be more resistant in our cell line panel. We tested 8 different HER2+ breast cancer cell lines (4 basal and 4 luminal subtype) with siRNA against INHBA. We observed significant inhibition of growth in 3 out of 4 basal lines, with growth rates of 40-55% of scramble control lines. In contrast, none of the luminal lines showed significant growth inhibition when treated with the INHBA siRNA. Furthermore, treatment with INHBA siRNA sensitized cells to lapatinib, with GI50 (dose required to inhibit growth by 50%) dropping from 10uM to less than 2 uM in JIMT1 and 21-MT1 basal lines. We also tested the effects of knockdown in 3-d matrigel cultures, which recapitulated the 2-d findings with decreased growth and increased sensitivity. For the cell line 21-MT1, growth following INHBA siRNA treatment was 12.6% of scramble control, and lapatinib treatment further reduced this to 7.8%, in contrast to scramble control/lapatinib treated cells, which grew at 97.7% of control rate. We performed RNAseq on 21-MT1 and JIMT1 cells treated with siRNA and looked at both relative changes in expression as well as absolute magnitude changes in expression. There were a large number of mitochondrial genes induced in the INHBA siRNA knockdown cells and loss of glycolytic genes such as LDHA compared to scramble control cells, suggesting a change in metabolic activity. We tested this by treating cells with the oxidative phosphorylation inhibitor oligomycin or the glycolysis inhibitor 2-deoxyglucose (2-DG). We found that JIMT1 cells treated with INHBA siRNA were more sensitive to oligomycin but less sensitive to 2-DG compared to scramble treated controls, suggesting a shift from glycolytic to oxidative phosphorylation metabolism. Since this type of metabolic shift has also been associated with reduced invasive capacity, we performed invasion assays on cells treated with INHBA or scramble-control siRNA. We found that knockdown of INHBA reduced invasion of both 21-MT1 and JIMT1 cells, with invasion rates less than 10% of control cells. Finally, we mined expression data sets to determine associations between INHBA levels and outcome, and found that high levels of expression of INHBA were associated with poor outcome in both HER2+ and the basal subtype of breast tumors, but not in luminal tumors. In conclusion, we identified INHBA as a major regulator of metabolism and aggressiveness in HER2+ basal breast cancer cells that is associated with poor outcome in patients.

Citation Format: Korkola JE, Liu M, Smith R, Liby T, Heiser L, Gray JW. A directed siRNA screen identifies INHBA as a major regulator of tumor aggressiveness in basal HER2 breast cancer [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P6-08-01.