Multi‐Omics Analysis by Machine Learning Identified Lysophosphatidic Acid as a Biomarker and Therapeutic Target for Porcine Reproductive and Respiratory Syndrome

Abstract

Abstract As a significant infectious disease in livestock, porcine reproductive and respiratory syndrome (PRRS) imposes substantial economic losses on the swine industry. Identification of diagnostic markers and therapeutic targets has been a focal challenge in PPRS prevention and control. By integrating metabolomic and lipidomic serum analyses of clinical pig cohorts through a machine learning approach with in vivo and in vitro infection models, lysophosphatidic acid (LPA) is discovered as a serum metabolic biomarker for PRRS virus (PRRSV) clinical diagnosis. PRRSV promoted LPA synthesis by upregulating the autotaxin expression, which causes innate immunosuppression by dampening the retinoic acid‐inducible gene I (RIG‐I) and type I interferon responses, leading to enhanced virus replication. Targeting LPA demonstrated protection against virus infection and associated disease outcomes in infected pigs, indicating that LPA is a novel antiviral target against PRRSV. This study lays a foundation for clinical prevention and control of PRRSV infections.

Keywords

• antiviral approach
• diagnostic biomarker
• immune suppression
• lysophosphatidic acid (LPA)
• machine learning
• multi‐omics