Analysis and Validation of Autophagy-Related Gene Biomarkers and Immune Cell Infiltration Characteristic in Bronchopulmonary Dysplasia by Integrating Bioinformatics and Machine Learning

Shuzhe Xiao,1,* Yue Ding,1,* Chen Du,1,* Yiting Lv,1 Shumei Yang,1,2 Qi Zheng,3 Zhiqiu Wang,3 Qiaoli Zheng,3 Meifang Huang,1 Qingyan Xiao,1 Zhuxiao Ren,2 Guangliang Bi,1 Jie Yang1,2 1Department of Neonatology, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, People’s Republic of China; 2Department of Neonatology, Guangdong Women and Children Hospital, Guangzhou, 511442, People’s Republic of China; 3The First Clinical Medical College of Southern Medical University, Guangzhou, 510515, People’s Republic of China*These authors contributed equally to this workCorrespondence: Jie Yang; Guangliang Bi, Email [email protected]; [email protected]: Autophagy and immunity play important regulatory roles in lung developmental disorders. However, there is currently a lack of bioinformatics analysis on autophagy-related genes (ARGs) and immune infiltration in bronchopulmonary dysplasia (BPD). We aim to screen and validate the signature genes of BPD by bioinformatics and in vivo experiment.Methods: GSE8586 was obtained from the Gene Expression Omnibus (GEO) database. The differentially expressed genes (DEGs) were identified using the R program. Using cell-type identification with CIBERSORT to analyze the inflammatory and immune status of BPD. Subsequently, the hub genes were identified by Lasso and Cytoscape with three machine-learning algorithms (MCC, Degree and MCODE). In addition, hub genes were validated with ROC, single-cell sequence and IHC in hyperoxia mice. Finally, we searched the drug targets of these hub genes, and established a nomogram model for predicting the risk of BPD.Results: There were 73 the differentially expressed and autophagy-related genes (DE-ARGs) by overlapping the DEGs in GSE8586 and ARGs. Five hub genes, BRIX1, JUN, PES1, NR4A1 and RRP9, were lowly expressed in the BPD group and had high diagnostic value in the diagnostic model. All hub genes are mainly located in B cell, epithelial cell, fibroblast, endothelial cell, smooth muscle cell and pneumocyte in lung single-cell sequencing. Moreover, immune infiltration analysis showed immune cells were higher in the BPD group and were closely associated with hub genes. We also predict the drug targets of the genes. Finally, the IHC result in mice showed that expression of PES1, BRX1, RRP9, JUN, NR4A1 was lower in the hyperoxia group compared to the normoxia group.Conclusion: BRIX1, JUN, PES1, NR4A1, RRP9, may be promising therapeutic targets for BPD. Our findings provided researchers and clinicians with more evidence regarding immunotherapeutic strategies for BPD treatment.Keywords: BPD, autophagy, immune cell infiltration, single cell sequencing, biomarkers