Analysis of infection-associated gene expression and their interaction with differentially expressed miRNA from the prospective of polymicrobial infections in corneal keratitis

Abstract

Keratitis is one of the leading causes of eye comorbidities. Bacteria, fungi, viruses, protozoans and helminths are the causative organisms of infectious keratitis. Simultaneous or sequential infection of more than one organism, known as polymicrobial infection, may lead to severe form of the disease. Few studies have highlighted the incidence and epidemiology of polymicrobial infections. However, no attempts were made to understand the changes in gene expression and identify pathways in polymicrobial infections, due to complex experimental design. This study utilizes the computational tools combined with experimental approaches to curate infection-associated genes that were expressed in cornea and secreted into tears. Gene Ontology and KEGG analyses revealed enrichment of pathways relevant to corneal host defense like epithelial polarity, leukocyte activation, antigen presentation, bacterial invasion and adherent junction. Network analysis identified highly interacting genes, including HSPA4, HSPA8, MAPK1, RHOA, STAT1, GSK3B, YWHAZ and TFRC. Integration of highly interacting genes with differentially expressed miRNAs from human fungal keratitis dataset identified regulatory hubs, such as miR-511-5p, miR-618 and miR-124-3p.1, with multiple infection-associated genes as targets. LPS treatment of human corneal epithelial cells modulated expression of several of these genes, further supporting their role in innate immune response. Connectivity map analysis identified perturbagens associated with key targets, like GSK3B and SYK, suggesting potential therapeutic relevance. Together, these findings reveal molecular mechanisms shared across microbial infections and highlight miRNA-gene interaction networks that may influence disease dynamics and outcomes in polymicrobial keratitis. While in vitro LPS treatment does not fully replicate the complexity of polymicrobial infection, this study provides a hypothesis-generating framework for future validation in patient derived samples and co-infection models.