炎症性肠病患者血浆差异蛋白质的筛选

目的 应用表面增强激光解析电离飞行质谱技术(SELDI-TOF-MS)寻找溃疡性结肠炎(UC)和克罗恩病(CD)患者的血浆差异表达蛋白,建立筛选模型.方法 采用CM10芯片对24例UC、25例CD和25例正常对照者血浆进行分析,从蛋白表达图谱中发现差异蛋白,建立区分CD和正常对照、UC和正常对照、UC和CD以及炎症性肠病(IBD)和正常对照的决策树分析模型,并进行盲法筛选.结果 在质荷比(m/z)2000～30 000范围内,CD和正常对照之间、UC和正常对照之间、IBD和正常对照之间具有2倍以上差异的蛋白峰数分别为9个、5个和11个(P＜0.05),软件自动选取m/z为8208、8837的蛋白质建立区分CD和正常对照的决策树模型,m/z为6985的蛋白质建立区分UC和正常对照的决策树模型,m/z为8208、1752、28 840和1702的4个蛋白质建立IBD和正常对照的决策树模型,3个决策树模型的灵敏度分别为96%、82%、91%,特异度分别为100%、85%、100%.结论 m/z 8208对于CD筛选价值高,特异度好,值得进一步探索.

Analysis of plasma proteomic spectra in patients with inflammatory bowel disease

Objective To screen the plasma differential expressed proteins in patients with Crohn's disease(CD)and ulcerative colitis(UC)using surface-enhanced laser desorption ionizationtime of flight mass spectrometry,and to establish decision trees algorithms.Methods The plasma samples from 24 UC patients,25 CD patients and 25 healthy controls were analyzed with CM10 protein chip.The proteomic spectra of CD,UC and inflammatory bowel disease(IBD)were compared with controls respectively.The differential proteins that significantly altered expression levels were selected to establish decision trees algorithms of CD.UC and IBD and then blind validations were tested.Results In the range of m/z 2000-30 000,differential expressed proteins that changed at least 2-fold between CD and controls were 9.between UC and controls were 5,and between IBD and controls were 11(P＜0.05).The software automatically picked up the m/z 8208 and 8837 as decision trees algorithms for differentiating CD from controls and m/z 6985 for differentiating UC from controls as well as m/z 8208,1752,28840 and 1702 for differentiating IBD from controls.The sensitivities of decision trees algorithms for CD,UC and IBD were 96%,82%and 91%respectively,and the specificities were 100%,85%and 100%respectively.Conclusions The protein of m/z 8208 which has high sensitivity in differentiating CD from controls is worthy of further study.