Background: Different patients have different sensitivities to lifestyle interventions (LSI) and thus different outcomes (improved or unchanged). In the present study, we aim to find key markers that can identify patients sensitive to lifestyle interventions.
Methods: Raw data were obtained from the Liver Clinic and Obesity Clinic of the Antwerp University Hospital that divided patients into LSI responders and non-responders based on the effectiveness of the intervention at the pathologic level. The limma / voom algorithm was used for difference analysis; Over representation analysis was used for functional enrichment analysis. Functional class scoring was used for gene set enrichment analysis. WGCNA and PPI / MCODE were used to identify valuable modules and genes. Western blot and immunohistochemistry were used for experimental validation.
Results: Specific features of LSI responders were extracted based on differential expression and functional enrichment. The CIBERSORT and ESTIMATE algorithm reconfirmed that there were differences in the immune cell composition of LSI responders versus non-responders. Next, the WGCNA algorithm was used to integrate analyses of associations between immune cell phenotypes and genes. PPI and MCODE algorithms were used to screen for gene clusters with higher weights. The test set and validation set data were subjected to correlation analysis of the filtered genes to obtain the hub gene - ANT1. ANT1 was highly expressed in LSI responders, while it was lowly expressed in non-responders. Mechanistically, ANT1 responds to lifestyle intervention by activating Panx1, which is key to lesion improvement in LSI responder populations.
Conclusions: In the study, we confirm that LSI responders populations respond to exercise intervention via the ANT1 / Panx1 pathway.