Integrated machine learning and multi-omics analysis identifies ALOX5 as a potential therapeutic target for tubulointerstitial inflammation in diabetic kidney disease

Diabetic kidney disease (DKD) is a leading cause of renal failure. Inflammation of the renal tubules and interstitium is a critical factor in the progression of DKD; however, the key regulatory genes and characteristics of the immune microenvironment remain poorly understood. This study aims to identify key inflammatory biomarkers in the renal tubule tissues of DKD patients and to elucidate their potential immunoregulatory mechanisms. By integrating multiple GEO transcriptome datasets and employing differential expression analysis, weighted gene co-expression network analysis (WGCNA), and machine learning algorithms (LASSO, Random Forest), we identified arachidonate 5-lipoxygenase (ALOX5) as a crucial feature gene of renal tubular inflammation in DKD. Clinical correlation analysis revealed that ALOX5 is significantly upregulated in DKD tissues, with high expression closely associated with decreased glomerular filtration rate and infiltration of M1 macrophages. Additionally, combining single-cell sequencing pseudotime analysis and multiplex immunohistochemistry (mIHC), we demonstrated that ALOX5 and its partner protein ALOX5AP are primarily expressed in CD68⁺ macrophages infiltrating the renal interstitium. They exhibit a high degree of co-localization with NF-κB/p65, iNOS, and CYSLTR1, suggesting that they may mediate the pro-inflammatory polarization of macrophages through the leukotriene-NF-κB axis. Finally, based on molecular docking and ADMET analysis, we screened the natural small molecule honokiol as a potential inhibitor of ALOX5, which possesses favorable pharmacokinetic properties. This study suggests that ALOX5 is a potential biomarker of immune microenvironment imbalance in DKD and provides a rationale for further investigation of targeted anti-inflammatory strategies, with honokiol as a candidate compound.