BACKGROUND: Chronic inflammation is well recognized as a key factor related to renal function deterioration in patients with diabetic kidney disease (DKD). Neutrophil extracellular traps (NETs) play an important role in amplifying inflammation. With respect to NET-related genes, the aim of this study was to explore the mechanism of DKD progression and therefore identify potential intervention targets. METHODS: Hub NET-related DEGs were screened via differential expression analysis and three machine learning methods, namely, LASSO, SVM-RFE and random forest. Consensus clustering was performed to analyze NET-related subtypes in DKD patients. KEGG enrichment analysis, GSEA, GSVA, ssGSEA and ESTIMATE were conducted to explore the molecular features of DKD patient subtypes. Leveraging single-nucleus RNA-seq datasets, the "scissor" and "bisqueRNA" algorithms were applied to identify the composition of renal cell types in DKD patient subtypes. Soft clustering analysis was performed to obtain gene groups with similar expression patterns during the development and progression of DKD. The correlations between hub NET-related DEGs and clinical parameters were mined from the Nephroseq V5 database. The core gene among the hub NET-related DEGs was selected by calculating semantic similarity. "Cellchat" algorithm, immunostaining, ELISA and flow cytometry were performed to explore the expression and function of the core gene. The Drug-Gene Interaction Database (DGIdb) was searched to identify candidate drugs. RESULTS: Six hub NET-related DEGs, namely, ACTN1, ITGB2, IL33, HRG, NFIL3 and CLEC4E, were identified. On the basis of these 6 genes, DKD patients were classified into 2 clusters. Cluster 1 patients, with higher NET scores, were evidently more immune-activating than those of cluster 2. Markedly increased numbers of immune cells, fibroblasts and proinflammatory proximal tubular cells were observed in cluster 1 but not in cluster 2. Cluster 1 also represented a more clinically advanced disease state. Among the 6 hub NET-related DEGs, the mRNA expression of ACTN1, ITGB2, IL33 and HRG was correlated with the eGFR. By semantic similarity analysis, IL33 was considered a central gene among the 6 genes. Cell-cell communication analysis further indicated that intercellular interactions via IL-33 were enhanced in DKD. Serum IL-33 concentration was negatively correlated with eGFR. IHC staining revealed that IL-33 expression was upregulated in the tubular epithelium in DKD patients. Supernatants from inflammatory tubular epithelial cells can increase MPO in neutrophils, whereas addition of anti-IL-33 antibody attenuated this phenotype. CONCLUSIONS: We identified 2 distinct NET-related subtypes in DKD patients, in which one subgroup was apparently more inflammatory and associated with a more severe clinical state. A significantly increased level of IL-33 in this inflammatory patient subgroup may play a role in aggravating inflammation via the IL-33-ST2 axis.