Abstract TP275: Early Epigenetic Changes Associated with Increased Risk of Post-Stroke Infection

Objective: Post-stroke infections account for one-third of all deaths from acute ischemic stroke. Epigenetic modifications have been shown to regulate the inflammatory and immune response in stroke and may play an important role in development of post-stroke infection. The objective of this study was to identify early whole blood epigenetic modifications associated with post-stroke infection. Methods: Whole blood was collected at ED presentation within 72 hours of last known well for 95 adult patients participating in the University of Colorado Emergency Medicine Specimen Bank with a final diagnosis of acute ischemic stroke. Post-stroke infection was determined by the clinical team as any infection within the first 90-days, not present on ED presentation. We used the Infinium Methylation EPIC Bead Chip array to characterize DNA methylation patterns in subjects with and without a post-stroke infection. Differentially methylated genes were identified using methylKit to analyze the data, and ggplot2 to plot the data in R. Results: The mean age of subjects with a post-stroke infection (n=22, 23.2%) was 71.1 years (stdev 14.1 years) compared to 63.3 years (stdev 13.8 years) among those without post-stroke infection (n=73, 76.8%) (p=0.02). The median time to infection was one day (IQR 1,8) with 86% (n=19) patients developing a PSI within 14 days. The top 20 differentially methylated genes included those involved in TGF-b signaling (FOXH1, SMAD2, SMAD2/3, SMAD4), regulation of transcription (GRHL3, ZNF44, ZNF534, ZNF57, ATF7IP, LMTK3, POLR3D, NR3C1, TP63, and DNMT3A), regulation of the cell cycle (CDCA2, HCFC1R1), and RNA-mediated gene silencing (AGO2). Beyond TGF-b signaling, GPS2, NR3C1, and POLRD3D also have roles in the inflammatory and immune response. Conclusions: Following an acute ischemic insult, several genes are differentially methylated among patients who develop a post-stroke infection, most notably multiple genes involved in regulating transcription, TGF-beta signaling, and inflammation and the immune response. In this preliminary cohort, it is not clear whether these changes are due to preclinical infection or truly identify risk. In addition, ongoing work aims to identify the predicted effects of these modifications and replicate the findings in a larger, independent cohort. Understanding the potential role of epigenetic modifications in risk of post-stroke infection may help identify therapeutic targets to help prevent poor outcomes.