ABSTRACT Diabetic kidney disease (DKD) is the primary cause of end‐stage renal disease globally, yet reliable biomarkers for its diagnosis and progression assessment are lacking. This study employed artificial intelligence techniques, including weighted gene co‐expression network analysis (WGCNA) and machine learning, to identify crucial genes associated with DKD. Validation was conducted using online databases such as Nephroseq and KIT, alongside biological samples from human serum, urine, peripheral blood mononuclear cell (PBMC) mRNA, kidney tissues, DKD rat models, and high glucose‐treated HK‐2 cells. Statistical analyses evaluated the correlations. The study revealed that C‐X‐C motif chemokine ligand 3 (CXCL3) was markedly upregulated in the serum and urine of DKD patients compared to healthy controls and those with type 2 diabetes mellitus, primary glomerulonephritis (e.g., IgA nephropathy, membranous nephropathy, minimal change disease). Immunohistochemistry showed significantly higher CXCL3 in both the glomeruli and tubulointerstitium of DKD patient kidneys than in those from controls. Elevated CXCL3 mRNA levels were also noted in PBMCs from DKD patients, STZ‐induced DKD rat kidneys, and high glucose‐treated HK‐2 cells. Furthermore, urinary CXCL3 protein levels positively correlated with the pathological grade, serum blood urea nitrogen (BUN), serum creatinine, and HbA1c percentage, while inversely correlating with estimated glomerular filtration rate (eGFR) in DKD patients. Mechanically, high glucose stimulation significantly upregulates the expression of inflammatory factors (including IL‐6 and IL‐1β) and fibrosis markers (α‐SMA and CTGF) in HK‐2 cells overexpressing CXCL3. Conversely, CXCL3 knockout in HK‐2 cells led to substantial downregulation of these inflammatory and fibrotic markers in the same high glucose conditions. Elevated CXCL3 levels in the serum, urine, and kidney tissues, alongside increased mRNA in PBMCs, suggest its potential as a biomarker for diagnosing and monitoring DKD. Correlations of urinary CXCL3 with disease severity indicators further support its diagnostic and prognostic utility. Mechanically, CXCL3 promotes inflammation and fibrosis in DKD.
Su et al. (Mon,) studied this question.