Identification of cuproptosis-related biomarkers and analysis of immune infiltration in allograft lung ischemia-reperfusion injury

Allograft lung ischemia-reperfusion injury (ALIRI) is a major cause of early primary graft dysfunction and poor long-term survival after lung transplantation (LTx); however, its pathogenesis has not been fully elucidated. Cell death is a mechanism underlying ALIRI. Cuproptosis is a recently discovered form of programmed cell death. To date, no studies have been conducted on the mechanisms by which cuproptosis-related genes (CRGs) regulate ALIRI. Therefore, we explored the potential biomarkers related to cuproptosis to provide new insights into the treatment of ALIRI. Materials and

NFE2L2, NLRP3, LIPT1, and MTF1 may serve as predictors of cuproptosis and play an important role in the pathogenesis of cuproptosis in ALIRI.

Datasets containing pre- and post-LTx lung biopsy samples and CRGs were obtained from the GEO database and previous studies. We identified differentially expressed CRGs (DE-CRGs) and performed functional analyses. Biomarker genes were selected using three machine learning algorithms. The ROC curve and logistic regression model (LRM) of these biomarkers were constructed. CIBERSORT was used to calculate the number of infiltrating immune cells pre- and post-LTx, and the correlation between these biomarkers and immune cells was analyzed. A competing endogenous RNA network was constructed using these biomarkers. Finally, the biomarkers were verified in a validation set and a rat LTx model using qRT-PCR and Western blotting.

Fifteen DE-CRGs were identified. GO analysis revealed that DE-CRGs were significantly enriched in the mitochondrial acetyl-CoA biosynthetic process from pyruvate, protein lipoylation, the tricarboxylic acid (TCA) cycle, and copper-transporting ATPase activity. KEGG enrichment analysis showed that the DE-CRGs were mainly enriched in metabolic pathways, carbon metabolism, and the TCA cycle. NFE2L2, NLRP3, LIPT1, and MTF1 were identified as potential biomarker genes. The AUC of the ROC curve for each biomarker was greater than 0.8, and the LRM provided an excellent classifier with an AUC of 0.96. These biomarkers were validated in another dataset and a rat LTx model, which exhibited good performance. In the CIBERSORT analysis, differentially expressed immune cells were identified, and the biomarkers were associated with the immune cells.