Loss of normal imprinting (1) occurs due to poor prenatal nutrition or exposure to heavy metals (e.g., cadmium, lead, arsenic) and (2) is linked to obesity, autism, and cancer [1,2,5,8,9,10,11,12,13,14,15]. The KCNK9 gene-product TASK3 is a pH-regulated, potassium channel membrane protein that we, and others [29,30,31], show regulates mitochondrial membrane potential and apoptosis. This study provides the first demonstration that KCNK9 is imprinted and monoallelically expressed in mammary epithelial cells. It also identifies a DMR that likely regulates imprinting at this locus in human breast tissue. In addition to the DMR identification, we demonstrated, by NOMe-Seq, a region of differential chromatin structure related to the methylation status of the DMR. The relationship between DNA methylation and condensed chromatin structure is consistent with a model in which methylation at the DMR silences gene expression by impacting chromatin accessibility and preventing transcription factor binding. This model also supports an epigenetic mechanism for KCNK9/TASK3 overexpression in breast cancer.

Hypomethylation of the KCNK9 DMR was observed more frequently in African-American women with TNBC (p = 0.006) and less frequently in European-American women with TNBC (p = 0.70). Hypomethylation of the KCNK9 DMR was observed concurrently in TNBC and normal-appearing adjacent breast tissue. Abnormal KCNK9 imprinting was associated with increased mitochondrial membrane potential in live TNBC cells and non-cancerous mammary epithelial cells from high-risk women (p < 0.001). These results are consistent with a study by Nagy et al. who showed that, in melanoma, TASK3 regulates apoptosis and mitochondrial function [30]. The finding that KCNK9-US1 DMR hypomethylation occurs preferentially in African-Americans suggests that KCNK9/TASK3 may provide a new target for prevention of TNBC.

While 50% of Ashkenazi European women with TNBC have a germline mutation of BRCA1, only 20% of African-American women with TNBC have a BRCA1 mutation [42]. This indicates that other mechanism(s) beyond germline mutation of BRCA1 are responsible for the etiology of TNBC in African-American women. African-American women experience disparities in income, access to care, and an unequal burden of environmental exposures [43]. Given that imprinting is dysregulated by poor nutrition and environmental toxicants, our findings provide a potential mechanistic link between disparities and TNBC in African-American women who do not have germline BRCA1 mutations. A limitation of this study is that it was conducted in a single institution in a restricted number of women; multi-institutional testing with an expanded test set and validation set is required to validate KCNK9/TASK3 as a potential risk biomarker.

The hypomethylation at the KCNK9 DMR was observed in both non-cancerous and cancerous breast tissue, but it is rare in the WBC of at-risk individuals. These findings indicate that epigenetic alterations occurring at the KCNK9 locus do not typically form at the time of fertilization and implantation. In contrast, they are consistent with alterations occurring in later epigenetically vulnerable developmental windows. These windows include during tissue differentiation, early childhood, or puberty. Identification of hypomethylation of the KCNK9 DMR in both breasts of at-risk individuals is indicative of a relatively early developmental change with large spatial distribution. Hypomethylation seen in only one breast would indicate epigenetic alteration occurring later in development, in a more specific location or cell type. The epialleles identified here are quantifiable markers for association studies between the environmental factors and the critical exposure timing that contribute to breast cancer risk.

The DMR methylation and NOMe-Seq chromatin data provide intriguing targets for future studies to better understand the origins and progression of TNBC. This will include further investigation of DMR methylation, chromatin structure, transcription factor binding, and gene/protein expression in this aggressive form of breast cancer. As TNBC has rapid-onset, aggressive growth, and resistance to treatment, patient survival could be improved through better classification of risk status, early detection, and better treatment targets. Furthermore, since genomic imprinting is dysregulated by poor nutrition and exposure to environmental toxicants, the results here provide support for the importance of good nutrition and a healthy environment in the prevention of TNBC.

Other recent work into the epigenetic regulation of KCNK9 has identified long distance cis-interactions between the promoter CpG island of KCNK9, which was found to be hypomethylated, and the PEG13 DMR [44]. As mentioned previously, parent-of-origin specific methylation in the KCNK9 CpG island was not detected, so any parental-specific regulation by this interaction would be due solely to the PEG13 DMR. Thus, it will be important to determine how CpG methylation and chromatin structure in the KCNK9 promoter influence long-range interactions with PEG13. Furthermore, it will be of interest to determine whether interactions between the KCNK9-US1 DMR and the KCNK9 promoter are involved in gene regulation. Such increased understanding of the epigenetic regulation of KCNK9 will be of great value in determining the role that expression of this gene has in the development and progression of cancers and in developing new treatment methods.