# Integrating Mendelian randomization and multi-omics analysis unravels gut microbiota-driven metabolic mechanisms in sepsis and identifies diagnostic biomarkers through experimental validation

## TL;DR

This study combines genetic and multi-omics data to uncover how gut microbes like Prevotella 9 influence sepsis through fatty acid metabolism and identifies potential biomarkers for diagnosis.

## Contribution

Novel integration of Mendelian randomization and multi-omics data to identify causal gut microbiota-metabolite interactions and validate diagnostic biomarkers in sepsis.

## Key findings

- Prevotella 9 shows protective effects in sepsis through regulation of fatty acid metabolism.
- Three key diagnostic genes (ABCC1, CYP1B1, PPARG) are identified and validated for sepsis.
- Monocytes are highlighted as potential cellular targets in sepsis via immunometabolic pathways.

## Abstract

Sepsis, a life-threatening systemic inflammatory syndrome, remains a leading cause of global mortality due to its complex pathophysiology and the lack of specific diagnostic biomarkers. Recent evidence highlights intricate interactions between the gut microbiota, metabolites, and host inflammatory responses; however, the causal relationships and underlying mechanisms remain poorly understood. We integrated Mendelian randomization (MR) with multi-omics approaches (including transcriptomics, untargeted metabolomics, and single-cell transcriptomics) to elucidate the causal relationships and underlying mechanisms between gut microbiota and their associated metabolites in the inflammatory response of sepsis. Building on this analysis, we employed machine learning algorithms to identify sepsis-specific diagnostic biomarkers derived from Prevotella 9, fatty acids, and PANoptosis-related genes. These key diagnostic genes were experimentally validated using a pulmonary sepsis organoid model. Seven gut microbiota taxa were identified as causally associated with sepsis, with Prevotella 9 demonstrating significant protective effects (odds ratio = 0.89, P = 0.01). The protective role of Prevotella 9 appears to be mediated through its regulation of fatty acid metabolism. Machine learning algorithms pinpointed three key diagnostic genes for sepsis: ABCC1, CYP1B1, and PPARG. Validation in an independent cohort (area under the receiver operating characteristic curve = 0.93) and the lung-derived organoid model confirmed their relevance. Functional analyses revealed that these genes are involved in immunometabolic pathways, including neutrophil regulation, oxidative stress, and macrophage polarization, and are predominantly expressed in monocytes. This study integrates MR and multi-omics analyses to reveal that Prevotella 9 may regulate sepsis through lipid metabolism. Additionally, three key genes (ABCC1, CYP1B1, and PPARG) were identified based on Prevotella 9, fatty acids, and PANoptosis, contributing to sepsis progression via the regulation of neutrophils, oxidative stress, and macrophage polarization. Monocytes may serve as potential cellular targets for sepsis.