Luo Hongmin, Cao Yuxuan, Guo Liping, Li Hui, Yuan Yingying, Lu Fan
Department of Nephrology, Hebei Medical University Third Hospital, Shijiazhuang, Hebei, China.
Department of Epidemiology and Statistics, School of Public Health, Hebei Medical University, Hebei Key Laboratory of Environment and Human Health, Shijiazhuang, Hebei, China.
PLoS One. 2025 Jun 3;20(6):e0321636. doi: 10.1371/journal.pone.0321636. eCollection 2025.
Cuproptosis may be a new clue to illustrate the pathogenesis of the disease. There was no study focused on the relationship between the cuproptosis genes and diabetic nephropathy (DN). This study aimed to reveal the relationship between cuproptosis genes and the immune microenvironment in DN and distinguish different phenotypes to describe disease heterogeneity through consensus clustering based on cuproptosis genes.
We downloaded RNA sequencing data sets of DN glomerular and normal renal tissue samples (GSE142025, GSE30528, and GSE96804) from the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) between DN and control samples were screened. Immune cell subtype infiltration and immune score were figured out via different algorithms. Consensus clustering was performed by Ward's method to determine different phenotypes of DN. Key genes between phenotypes were identified via a machine-learning algorithm. Logistic regression analysis was applied to establish a nomogram for assessing the disease risk of DN. The role of related genes was verified by cell experiments.
In DN samples, NOD-like receptor thermal protein domain associated protein 3(NLRP3) and cyclin-dependent kinase inhibitor 2A Gene(CDKN2A) were positively correlated to immune score. Nuclear factor erythroid 2-related factor 2(NFE2L2), Lipoic Acid Synthetase(LIAS), Lipoyltransferase 1(LIPT1), Dihydrolipoamide dehydrogenase(DLD), Dihydrolipoamide Branched Chain Transacylase E2(DBT) and Dihydrolipoamide S-Succinyltransferase(DLST) were negatively correlated to immune score. Via Consensus clustering based on cuproptosis genes, the DN samples were divided into cluster C1 and cluster C2. Cluster C1 was characterized by low cuproptosis gene expression, high immune cell subtype infiltration, and high enrichment of immune-related pathways. Cluster C2 was on the contrary. Dicarbonyl/l-xylulose reductase (DCXR) and heat-responsive protein 12 (HRSP12) were key genes related to clinical traits and immune microenvironment, negatively correlated with most immune cell subtypes. The nomogram constructed based on DCXR and HRSP12 showed good efficiency for DN diagnosis.
Immune microenvironment imbalance and metabolic disorders may lead to the occurrence of DN. Cuproptosis genes, with the ability to regulate the immune microenvironment and metabolism, can be used for disease clustering to describe the heterogeneity and characterize the immune microenvironment. HRSP12 and DCXR, as key genes related to disease phenotypes and immune microenvironment characteristics, were jointly constructed as nomograms for DN diagnosis with high accuracy and reliability. HRSP12 and DCXR may be potential biological markers and renal protective factors.
铜死亡可能是阐明该疾病发病机制的新线索。目前尚无关于铜死亡基因与糖尿病肾病（DN）关系的研究。本研究旨在揭示DN中铜死亡基因与免疫微环境之间的关系，并通过基于铜死亡基因的一致性聚类区分不同表型以描述疾病异质性。
我们从基因表达综合数据库（GEO）下载了DN肾小球和正常肾组织样本的RNA测序数据集（GSE142025、GSE30528和GSE96804）。筛选DN样本与对照样本之间的差异表达基因（DEG）。通过不同算法计算免疫细胞亚型浸润和免疫评分。采用沃德法进行一致性聚类以确定DN的不同表型。通过机器学习算法确定表型之间的关键基因。应用逻辑回归分析建立用于评估DN疾病风险的列线图。通过细胞实验验证相关基因的作用。
在DN样本中，NOD样受体热蛋白结构域相关蛋白3（NLRP3）和细胞周期蛋白依赖性激酶抑制剂2A基因（CDKN2A）与免疫评分呈正相关。核因子红细胞2相关因子2（NFE2L2）、硫辛酸合成酶（LIAS）、硫辛酰胺转移酶1（LIPT1）、二氢硫辛酰胺脱氢酶（DLD）、二氢硫辛酰胺支链转酰酶E2（DBT）和二氢硫辛酰胺S-琥珀酰转移酶（DLST）与免疫评分呈负相关。通过基于铜死亡基因的一致性聚类，DN样本被分为C1簇和C2簇。C1簇的特征是铜死亡基因表达低、免疫细胞亚型浸润高以及免疫相关通路富集高。C2簇则相反。二羰基/l-木酮糖还原酶（DCXR）和热反应蛋白12（HRSP12）是与临床特征和免疫微环境相关的关键基因，与大多数免疫细胞亚型呈负相关。基于DCXR和HRSP12构建的列线图对DN诊断显示出良好的效能。
免疫微环境失衡和代谢紊乱可能导致DN的发生。铜死亡基因具有调节免疫微环境和代谢的能力，可用于疾病聚类以描述异质性并表征免疫微环境。HRSP12和DCXR作为与疾病表型和免疫微环境特征相关的关键基因，联合构建的DN诊断列线图具有较高的准确性和可靠性。HRSP12和DCXR可能是潜在的生物学标志物和肾脏保护因子。
