Department of Kidney Transplantation, Beijing Tsinghua Changgung Hospital, School of Clinical Medicine, Tsinghua University, Beijing, China.
Division of Nephrology, Department of Nursing, The University of Hong Kong-Shenzhen Hospital, Shenzhen City, Guangdong Province, China.
Transpl Immunol. 2024 Dec;87:102114. doi: 10.1016/j.trim.2024.102114. Epub 2024 Sep 5.
Glycosylation is a complex and fundamental metabolic biosynthetic process orchestrated by multiple glycosyltransferases (GT) and glycosidases enzymes. Functions of GT have been extensively examined in multiple human diseases. Our study investigated the potential role of GT genes in T-cell mediated rejection (TCMR) and possible prediction of graft loss of kidney transplantation.
We downloaded the microarray datasets and GT genes from the GEO and the HUGO Gene Nomenclature Committee (HGNC) databases, respectively. Differentially expressed GT genes (DE-GTGs) were obtained by differential expression and Venn analysis. A TCMR diagnostic model was developed based on the hub DE-GTGs using LASSO regression and XGboost machine learning algorithms. In addition, a predictive model for graft survival was constructed by univariate Cox and LASSO Cox regression analysis.
We have obtained 15 DE-GTGs. Both GO and KEGG analyses showed that the DE-GTGs were mainly involved in the glycoprotein biosynthetic process. The TCMR diagnostic model exhibited high diagnostic potential with generally highly correlated accuracies [aera under the curve (AUC) of 0.83]. The immune characteristics analysis revealed that higher levels of immune cell infiltration and immune responses were observed in the high-risk group than in the low-risk group. In particular, the Kaplan-Meier survival analysis revealed that renal grafts in the high-risk group have poor prognostic outcomes than the low-risk group. The predictive AUC values of 1-, 2- and 3-year graft survival were 0.76, 0.81, and 0.70, respectively.
Our results indicated that GT genes could be used for diagnosis of TCMR and prediction of graft loss in kidney transplantation. These results provide new perspectives and tools for diagnosing, treating and predicting kidney transplant-related diseases.
糖基化是由多种糖基转移酶（GT）和糖苷酶共同调控的复杂而基本的代谢生物合成过程。GT 的功能已在多种人类疾病中得到广泛研究。我们的研究调查了 GT 基因在 T 细胞介导的排斥反应（TCMR）中的潜在作用及其对肾移植移植物丢失的可能预测。
我们分别从 GEO 和人类基因命名委员会（HGNC）数据库下载了微阵列数据集和 GT 基因。通过差异表达和 Venn 分析获得差异表达的 GT 基因（DE-GTGs）。基于 LASSO 回归和 XGBoost 机器学习算法，利用 hub DE-GTGs 构建 TCMR 诊断模型。此外，通过单变量 Cox 和 LASSO Cox 回归分析构建了移植物存活的预测模型。
我们获得了 15 个 DE-GTGs。GO 和 KEGG 分析均表明，DE-GTGs 主要参与糖蛋白的生物合成过程。TCMR 诊断模型具有较高的诊断潜力，总体相关准确率较高（AUC 为 0.83）。免疫特征分析表明，高危组的免疫细胞浸润和免疫反应水平高于低危组。特别是，Kaplan-Meier 生存分析显示，高危组的肾移植物预后不良。1、2、3 年移植物存活率的预测 AUC 值分别为 0.76、0.81 和 0.70。
我们的研究结果表明，GT 基因可用于 TCMR 的诊断和肾移植移植物丢失的预测。这些结果为诊断、治疗和预测肾移植相关疾病提供了新的视角和工具。
