利用蛋白芯片技术诊断和评估肾损伤的临床应用

摘要

Objective To research for protein fingerprints in urine from patients with kidney injury using Au protein chip (pro-teinchip gold array ) and discuss its diagnostic value .Methods Surface-enhanced laser desorpu'on-ronization time of flight mass spec-0 trometry (SELDI-TOF-MS) and Au Chip were used to detect the urinary protein profilings of 477 patients ,including 237 patients with nephropathy and 240 controls .The differential peaks were screened by biomarker patterns software (BPS) and the predictive pattern was developed by artificial neural network (ANN ) .The model was blindly tested to validate the diagnostic value .Some differential pro-teins were identified by molecular mass compared with the standard protein and electrospray ionization quadrupole time of flight mass spectrometry (ESI-Q-TOF-MS ) .Results Seventy five distinguished protein peaks (t value: —11.6—28.1, ρ＜0 ,05)were obtained between nephropathy and the control groups .The protein with 6 182、8 843、9 740、23 770、67 650、80 045 and 91 240 m/z were selected as biomarkers using BPS . A diagnostic model constructed by ANN with these protein markers could successfully diagnose nephropathy with sensitivity of 96 .6％ (84/87) and specificity of 94 .4％ (85/90) .Six specific protein peaks (m/z at 11 735 ,15 150,2 2871,23 770 ,67 650,80 045 ) were identified and considered as β2-microglobulin ,hemoglobin ,al-antitrypsin ,al-microglobu-lin ,albumin and transferrin ,respectively .Conclusion The protein fingerprints detected by SELDI-TOF-MS and Au-chip has the po-tentials in diagnosis of nephropathy ,identification of albuminuria and evaluation of renal injury .%目的 利用Au蛋白芯片(proteinchip gold array)技术检测肾损伤患者尿蛋白指纹图谱,探讨其灵敏、快速预测和评估肾损伤的临床应用价值.方法 利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蛋白芯片检测477例尿蛋白指纹图,包括237例肾病患者和240例对照者.利用生物标志模型软件(biomarker patterns software,BPS)筛选标志蛋白,结合人工神经网络(artificial neural network,ANN)技术建立模型,评价其在诊断和评估肾损伤中的应用价值.通过电喷雾四级杆飞行时间质谱(electrospray ionization quadrupole time of flight,ESI-Q-TOF)结合比对标准蛋白质谱,对部分差异蛋白进行鉴定.结果 肾病患者与对照者尿中差异表达的蛋白质峰有75个,其丰度值两组间比较差异均有统计学意义(t值为-11.6-28.1,P均<0.05),筛选质荷比(m/z)6 182、8 843、9 740、23 770、67 650、80 045和91 240蛋白质建立的ANN模型预测肾损伤的灵敏度为96.6%(84/87),特异度为94.4%(85/90).鉴定结果表明11 735、15 150、2 2871、23 770、67 650 h和80 045 m/z蛋白分别为β2-微球蛋白、血红蛋白、α1-抗胰蛋白酶、α1-微球蛋白、白蛋白和转铁蛋白.结论 基于Au蛋白芯片的SELDI-TOF-MS技术检测尿蛋白指纹图,在肾脏疾病的诊断、蛋白尿类型判断、肾损伤评估及治疗评价中具有潜在的应用价值.