The Epitranscriptome as a Novel Mechanism of Arsenic-Induced Diabetes.

SUMMARY In the United States, the prevalence of type 2 diabetes mellitus (T2DM) is particularly high among American Indian (AI) communities. Arsenic (As), a pervasive environmental contaminant disproportionately affecting AI communities, may explain this increased risk. Arsenic induces oxidative stress and systemic low-grade

inflammation leading to β-cell dysfunction and insulin resistance in target tissues. However, the impact of As on T2DM has been disputed due to a lack of coherent mechanism for these findings. Previous studies have focused on epigenomic mechanisms (e.g., DNA methylation, histone modifications), overlooking downstream regulatory

mechanisms that can more directly shape phenotypes. We propose to investigate the RNA modification N6- methyladenosine (m6A), the most prevalent epitranscriptomic modification on messenger RNA, which is directly involved in the cellular stress response. In experimental systems, arsenic induces a m6A response. m6A also

modulates key processes underlying T2DM pathogenesis, including immune response and systemic inflammation. m6A is controlled by a group of proteins called reader, writer, and erasers (RWEs), responsible for adding, interpreting, and removing m6A marks. Fat mass and obesity-associated protein (FTO) is one example

of an arsenic-sensitive m6A eraser with strong ties to T2DM and glucose homeostasis. Our pilot study in elderly men exposed to low-level arsenic supported these findings. We propose to test the hypothesis that altered m6A and RWEs are plausible mechanisms for As-related T2DM in the Strong Heart Study (SHS). The SHS is an

ongoing longitudinal study in AI communities in Arizona, Oklahoma, and North/South Dakota with detailed clinical data for T2DM and metabolic syndrome (MetS). The SHS has measured speciated As exposure data covering childhood and adult exposure windows, both independently associated with T2DM in previous research.

Leveraging the cohort design, exposure and phenotypic data, infrastructure, and study team, we propose to conduct epitranscriptomic analysis of mRNA m6A profiles via m6A sequencing and measure mRNA expression of 20 RWEs using whole blood from 1100 participants at the upcoming SHS follow up visit (scheduled for 2022-

23). Our specific aims are to: 1) determine the association of past and current As exposure with epitranscriptomic profiles of m6A and RWEs mRNA expression levels in blood; 2) determine the association of blood m6A epitranscriptomic profiles with metabolic markers and MetS, clinical T2DM prevalence, and T2DM control

(glycated hemoglobin, albuminuria); 3) develop a predictive m6A fingerprint that quantifies the risk of T2DM due to As exposure using machine learning approaches. For aims 1 and 2 we will further use Mendelian randomization to assess causal relationships. Characterization of m6A profiles in a population of AI adults highly

impacted by T2DM will reveal biological features linking a pervasive toxicant such as As to diabetes. In addition to leading to interventions to reduce As exposure in the US and globally, defining the roles of m6A and RWEs in T2DM may contribute to new targets for future diabetes therapies.