尿蛋白组学筛选检测过敏性紫癜患者的肾损害

Urinary proteomics for the prediction of nephritis in patients with Henoch-Schoenlein purpura

目的表面增强激光解析电离飞行时间质谱技术（sELDI-TOF—MS）筛选早期过敏性紫癜肾炎患者尿液中的特异标志蛋白。方法CM10（弱阳离子交换表面）蛋白芯片结合SELDI-TOF-MS技术，检测60例过敏性紫癜患者（30例伴发紫癜性肾炎，30例不伴肾炎）尿液中的蛋白质质谱，用浙江大学肿瘤研究所蛋白芯片数据分析系统筛选过敏性紫癜肾炎特异的尿液蛋白标志物。结果共检测到154个经过信噪比和强调过滤的高质量质谱蛋白质峰，从中筛选出2个表达差异显著（P〈0．05）的蛋白标志物。用遗传算法结合支持向量机模型的方法筛选出最佳模型，用留一法评估模型的预测效果为敏感性84％，特异性71％。结论尿蛋白质谱与生物信息学分析方法对早期过敏性紫癜肾炎的检测具有一定的敏感性和特异性。

Objective To identify novel biomarkers from urinary protein profiles for the early diagnosis of nephritis in patients with Henoch-Schtinlein purpura by surface enhanced laser desorptiordionization time of flight-mass spectrometry （SELD1-TOF-MS） technique. Methods Urine samples were collected from 60 untreated patients with Henoch-Schoenlein purpura, including 30 patients with nephritis and 30 without. SELDI- TOF-MS technique was used to characterize the protein profile in these urine samples, and the Zhejiang University Cancer Institute - Protein Chip Data Analysis System （ZUCI-PDAS） to identify urine protein markers and construct diagnostic model for nephritis in patients with Henoch-Schoenlein purpura. Results Totally, 154 mass peaks were identified with high quality, and two proteins at a mass-to-charge ratio （m/z） of 2454.971 and 2439.686 showed significantly differential expression between the two groups of patients （P 〈 0.05）. Seven biomarkers were used to establish a diagnostic model. As estimated by the leave-one-out cross-validation, the diagnostic model distinguished patients with nephritis from those without with a specificity of 71% and sensitivity of 84%. Conclusions The developed diagnostic model based on SELDI-TOF-MS technique and bioinformatics is somewhat specific and sensitive for the prediction of nephritis in patients with Henoch-Schtoenlein purpura.