A Longitudinal and Transancestral Analysis of DNA Methylation Patterns in Systemic Lupus Erythematosus

Abstract

Objective: Systemic lupus erythematosus (SLE or lupus) is a complex heterogeneous autoimmune disease that can affect multiple organ systems. Lupus predominantly affects women, with African-American women carrying higher disease burden and poorer outcomes than European-American women. Disruption in the epigenetic landscape of immune cells, including DNA methylation, is a recognized feature of lupus. Global hypomethylation is consistently detected across lymphoid and myeloid immune cells from lupus patients, with the most extensive hypomethylation occurring in type I interferon response genes. Genetic background plays a role in shaping the epigenome of immune cells and in the risk of developing lupus. The aim of this dissertation was to measure the influence of genetics on DNA methylation differences between African-American and European-American lupus patients and those between lupus patients and healthy controls. Further, we sought to measure the variation in DNA methylation longitudinally over time and with disease activity in lupus. Results: We first analyzed granulocyte DNA methylation over time in a longitudinal cohort of African-American and European-American lupus patients followed for up to four years. Ethnicity-associated differential methylation in lupus granulocytes included genes associated with the type I interferon response and NFκB pathways. Approximately 16% of differentially methylated sites between ethnicities were associated with a nearby (≤ 1kb) genetic variant known as methylation quantitative trait loci (meQTL). An meQTL involving the promoter region of TREML4, a regulator of TLR7 response in myeloid cells, was among the strongest meQTL detected. Genotype-specific mRNA expression of TREML4 negatively correlated with promoter methylation in this locus. Methylation levels in two CpG sites were associated with changes in disease activity in African-American, but not European-American patients. Hypomethylation in GALNT18 was associated with the development of lupus nephritis during follow up. The lupus-associated epigenotype in granulocytes was not explained by genetics in lupus patients. To further examine the role of genetics in lupus-associated epigenetic changes, we studied DNA methylation in naïve CD4+ T cells from lupus patients and controls. CpG-associated genes in meQTL of lupus patients showed overlap with lupus risk genes. Some meQTL were shared between lupus granulocytes and naïve CD4+ T cells, including HLA-DQB1 and IRF7. However, confirming the results in lupus granulocytes, less than 1% of differentially methylated sites in lupus T cells were associated with meQTL. We then used a trend deviation analysis to compare disease-associated DNA methylation differences between lupus patients and controls in naïve CD4+ T cells, with methylation correlation patterns in over 16,000 tissue samples. The promoter of the miR-17-92 cluster, which regulates T cell proliferation and differentiation, was hypomethylated in lupus. Expression of two member microRNAs in this cluster were positively correlated with lupus disease activity. Conclusions: We have identified significant differences in the DNA methylome between African-American and European-American lupus patients, which are in part associated with nearby genetic variants. We have determined that granulocyte DNA methylation is predominantly stable over time and across disease activity levels in both African-American and European-American lupus patients. Furthermore, using DNA methylation profiles in granulocytes and naïve CD4+ T cells, we have determined that the DNA methylation signature of lupus, which is defined by robust hypomethylation of type I interferon genes, has little association with nearby genetic variants. These findings suggest that non-genetic factors play a predominant role in the DNA methylation signature of lupus.