尿蛋白标志物模型早期诊断糖尿病肾病的临床应用

Early diagnosis of diabetic nephropathy using protein pattern based on urinary biomarkers

目的寻找基于蛋白质组学技术早期、快速诊断糖尿病肾病的尿蛋白标志物模型，并探讨其临床应用价值。方法应用表面增强激光解析电离飞行时间质谱（surface-enhanced laser desorption-ionization time of flight mass spectrometry，SELDI-TOF-MS）技术及Au芯片（proteinchipgoldarray）检测292例患者尿蛋白质谱，包括129例糖尿病肾病和163例对照者（61例糖尿病及102名健康体检者）。获得的蛋白质谱数据用Biomaker Wizard3．1软件筛选差异蛋白，通过生物标志模型软件（biomarker patterns software，BPS）建立决策树辨别分析模型，评价其临床诊断价值。对部分筛选的差异蛋白通过比对标准蛋白质谱数据，根据分子量大小进行初步鉴定。结果糖尿病肾病患者与对照者尿液中差异表达的蛋白质峰有40个，其丰度值两组间比较差异均有统计学意义（t值为-9．81～24．52，P均〈0．05），通过BPS自动筛选66916质荷比（m／z）蛋白建立的模型诊断糖尿病肾病敏感度为98．7％（78／79），特异度98．2％（111／113）。对糖尿病和糖尿病肾病患者尿蛋白质谱图分析后得到24个差异蛋白质峰，其丰度值两组间比较差异均有统计学意义（t值为-6．95～14．45，P均〈0．05），BPS筛选4008、11619、66916m／z蛋白建立模型区分糖尿病与糖尿病肾病的敏感度（129／129）和特异度（61／61）均为100％。通过比对标准蛋白质谱数据，糖尿病肾病患者尿差异蛋白中m／z 11619、23529、66916和79378，可能为β2-微球蛋白、α1-微球蛋白、白蛋白和转铁蛋白。结论基于SELDI-TOF-MS及Au芯片技术检测尿蛋白质谱在鉴别蛋白尿来源、糖尿病肾病的早期快速诊断及肾脏损害评估具有重要应用价值。

Objective To search for protein markers in urine from patients with diabetic nephropathy by proteomic method and discuss its clinical significance in laboratory diagnosis of diabetic nephropathy. Methods This study included 129 patients with diabetic nephropathy, 61 diabetes mellitus patients, and 102 healthy volunteers. The urinary protein profiles were obtained using surface-enhanced laser desorption-ionization time of flight mass spectrometry （ SELDI-TOF-MS ） and Au Chip （ ProteinChip Gold Array）. The differential peaks were screened by Biomaker Wizard software and the decision tree pattern was developed by Biomarker Patterns Software （BPS）. The model was blindly tested to validate diagnostic efficieney. Some differentially expressed protein was preliminarily identified according to the molecular weight as compared with mass spectrometry data of standard proteins. Results Totally 40 distinguished protein peaks（ t value： - 9. 81-24. 52, P 〈 0. 05 ） were obtained after comparing the samples between diabetic nephropathy and the control groups. The peak with m/z 66 916 was automatically screened by BPS to develop decision tree pattern. The pattern was blindly tested and yielded a sensitivity of 98. 7% （78/79） and a specificity of 98. 2% （111/113 ）. After we compared results from diabetic nephropathy with those from diabetes mellitus, twenty-four differential peaks were obtained in diabetic nephropathy （t value： -6. 95-14. 45 ,P 〈0. 05）. The peaks with m/z 4 008, 11 619 and 66 916 were automatically screened by BPS to establish decision tree pattern. The model was blindly tested and yielded the sensitivity（129/129） and specificity（61/61 ） of 100%. After we compared our results with mass spectrometry data of standard proteins, the four differentially expressed proteins with m/z 11 619, 23 529, 66 916 and 79 378 were supposed to be β2-microglobulin, α1-microglobulin, albumin and transferrin. Conehmion The preliminary results suggest that these SELDI-TOF and Au chip have the potential application value in identification of protein source and early diagnosis of diabetic nephropathy, and evaluation of renal injury.