Integrative multi-omic analyses identify novel biomarkers for recurrent stroke in participants from the Vitamin Intervention for Stroke Prevention (VISP) clinical trial
Davis Armstrong, Nicole Marie
This item will be available on: 2020-05-01
Background: Recurrent strokes are deadlier and more disabling than primary strokes and over 185,000 individuals in the US suffer a recurrent event each year. Current diagnostic and at-risk identification tools are inadequate in rapidly detecting ischemic stroke and furthermore, recurrent stroke and stroke-prone populations. Integrative multi-omics approaches can detect novel biomarkers that can be used as therapeutics to improve the quality of life post-stroke and to identify at-risk individuals. Although genomic analyses have identified numerous variants that confer some stroke risk, as well as risk for stroke comorbidities, other -omics approaches are vital in identifying biological markers for stroke and recurrent stroke prevention and treatment. This dissertation is aimed to identify epigenetic and metabolite biomarkers that are associated with stroke outcomes, time to event, and related phenotypes. Methods: We analyzed a subset of 204 individuals from the Vitamin Intervention for Stroke Prevention (VISP) clinical trial. DNA was extracted from whole blood samples from these individuals and DNA methylation data was generated using the Illumina Infinium HumanMethylation 450K BeadChip microarray. Upon quality control and quantile normalization, our methylation subset consisted of 180 individuals, comprising of both individuals of African descent (n=76) and European descent (n=104). In order to detect potential epigenetic biomarkers, differential methylation, survival, and network analyses were performed on each ethnicity stratum. Furthermore, the global, untargeted metabolite profiles were generated for a subset of 50 African Americans from the original methylation subset of 204 at Metabolon. Over 900 metabolites, both endogenous and exogenous, were detected from plasma samples, normalized, log-transformed, and used in subsequent statistical analyses. Univariate tests comparing the mean metabolite concentrations across phenotypically different groups, survival analysis, and network analyses were used to detect associations between metabolites and stroke outcomes. Matched pairs analyses were performed on 44 of the 50 individuals and were selected based on age, sex, cigarettes smoked per day, and modified Rankin score indicative of VISP enrollment stroke severity. Integrative analyses employing both methylation and metabolite data included multiple linear regression models and network analyses. Results: We detected two statistically significant methylation loci in the European cohort that were associated with the number of strokes suffered prior to VISP enrollment. cg22812874 was located within an exon of the ASB10 gene (p=3.40e-09), while cg00340919 was an intronic locus of the TTC37 gene (p=8.74e-08). Cox Proportional Hazards survival analyses resulted in seven statistically significant methylation loci associated with time to event. Four of these loci were identified in individuals of African descent in which time to VISP recurrence was the outcome: cg04059318 (p=4.52e-08), ch.2.81927627R (p=9.11e-08), cg03584380 (p=9.66e-08) and cg24875416 (p=9.82e-08). Two loci, cg00076998 (p=7.87e-08) and cg16758041 (p=1.04e-07) were significant in the African descent analyses for time to composite vascular endpoint, while cg02365967 (p=8.08e-08) was significant in the European descent analysis for composite vascular endpoint. One locus, cg03584380, was validated in our Barcelona-based replication cohort. In our metabolomics analyses, we identified six metabolites associated with baseline smoking status and stroke recurrence in males (p=9.25e-13 to 3.30e-05), as well as one metabolite associated with time to recurrent stroke in survival analysis (tricosanoyl sphingomyelin (d18:1/23:0); HR=0.20, 95% CI: 0.01-3.49, p=1.56e-05). Over 2,600 suggestive associations were observed between metabolite profiles and DNA methylation loci (p=4.40e-11 to 1.05e-07). Matched pairs analyses identified four metabolites, gamma-glutamylhistidine (p=1.80e-03), behenoyl sphingomyelin (d18:1/22:0) (p=2.10e-03), trimethylamine N-oxide (p=2.10e-03), and lignoceroyl sphingomyelin (d18:1/24:0) (p=2.20e-03) that moderately differed between individuals suffering a VISP recurrent stroke and those who did not. Conclusion: These studies suggests that DNA methylation and metabolic data offer critical insight into complex disease pathophysiology and provide analytical methodologies to detect disease biomarkers. The epigenetic and metabolite biomarkers identified in subsequent chapters provide a foundation for future mechanistic studies into epigenetic and metabolic marks that are associated with recurrent stroke, as well as serve as potential biomarkers that confer risk of recurrence in stroke-prone populations.
Davis Armstrong, Nicole Marie. (May 2019). Integrative multi-omic analyses identify novel biomarkers for recurrent stroke in participants from the Vitamin Intervention for Stroke Prevention (VISP) clinical trial (Doctoral Dissertation, East Carolina University). Retrieved from the Scholarship. (http://hdl.handle.net/10342/7228.)
Davis Armstrong, Nicole Marie. Integrative multi-omic analyses identify novel biomarkers for recurrent stroke in participants from the Vitamin Intervention for Stroke Prevention (VISP) clinical trial. Doctoral Dissertation. East Carolina University, May 2019. The Scholarship. http://hdl.handle.net/10342/7228. December 05, 2019.
Davis Armstrong, Nicole Marie, “Integrative multi-omic analyses identify novel biomarkers for recurrent stroke in participants from the Vitamin Intervention for Stroke Prevention (VISP) clinical trial” (Doctoral Dissertation., East Carolina University, May 2019).
Davis Armstrong, Nicole Marie. Integrative multi-omic analyses identify novel biomarkers for recurrent stroke in participants from the Vitamin Intervention for Stroke Prevention (VISP) clinical trial [Doctoral Dissertation]. Greenville, NC: East Carolina University; May 2019.
East Carolina University