The Role of Biological Sex in Modulating Human Gene Expression and Immune Responses to Infection and Autoimmune Risk

Abstract

Biological sex shapes nearly every aspect of human health—from height and immune response to disease risk and treatment outcomes. Despite these widespread differences, the mechanisms driving sex-specific disease physiology remain poorly understood. Through integrated computational and experimental approaches highlighted single-cell RNA-seq in lymphoblastoid cell lines (LCLs; scHi-HOST method), my work reveals how sex chromosomes and genetic variation intersect to shape gene regulation and disease risk. I found that >9% of all expressed genes show significant sex bias and are enriched for sex-differential transcription factor activityWhile it is well accepted that sex-biased expression is abundant throughout the genome, the existence and relevance of sex-biased expression quantitative trait loci (sb-eQTL) is much more contentious. However, by focusing on conditionally independent eQTL identified in each sex separately, I discovered over 2,000 unique sb-eQTL. Enrichment of sb-eQTL in GWAS traits revealed that these variants are enriched in over 100 phenotypes, including many loci associated with female-biased autoimmune diseases such as multiple sclerosis (MS). One identified sb-eQTL (rs7117261) for the gene H2AX strongly colocalizes with MS risk in females (posterior probability=0.94), but not in males (posterior probability=0.01) suggesting a sex-specific risk factor of H2AX expression. In addition to baseline analyses, I examined sex differences during IAV infection using the single-cell RNA-seq scHi-HOST method developed in my lab, which quantifies infection at the cellular level in LCLs. Differential expression analyses testing Sex × IAV interaction across three IAV strains—the 2009 pandemic strain (CA09) and two seasonal strains (HI19 and WY03)—revealed a striking interaction for the long noncoding RNA TSIX in HI19 and WY03, but not CA09. TSIX was strongly induced in males infected with HI19 or WY03 and correlated with overall infection but remained unchanged in females or during CA09 infection. In mouse models, TSIX regulates XIST, the long noncoding RNA responsible for X-chromosome inactivation, but its function in human male cells was unknown. RNAi knockdown in A549 lung epithelial cells confirmed that TSIX promotes viral replication during WY03 infection (p = 0.006). Bulk RNA-seq of TSIX knockdown further validated its proviral role and revealed suppression of interferon responses. Using the sb-eQTL pipeline I developed and sex × IAV interaction GWAS of cellular viral phenotypes, I also identified single genetic variants associated with sex-biased responses to IAV, including variants in known antiviral genes such as DDX58 (RIG-I) and CGAS (cGAS) and novel risk loci in SLC24A2. Thus, this work uncovers how IAV reprograms sex-biased gene expression and highlights new targets for understanding—and potentially mitigating—sex differences in viral disease. Lastly, through investigation of Epstein-Barr virus (EBV) transcription in the uninfected scHi-HOST data, I identified variability in spontaneous lytic reactivation that is significantly variable by continental ancestry (p=0.0002). I further found that risk loci for the autoimmune diseases MS, systemic lupus erythematosus (SLE), Sjögren’s syndrome, and rheumatoid arthritis are associated with EBV transcription in LCLs. These variants provide a crucial connection to host regulation of EBV, a known risk factor in autoimmunity. These discoveries challenge the traditional “one-size-fits-all” model in genomics and underscore the importance of sex as a fundamental biological variable. By decoding how sex shapes gene expression, my research provides new frameworks for understanding sex-biased diseases and guiding more equitable biomedical strategies.