应用MALDI-TOF-MS检测原发性肝癌骨转移患者血清的多肽差异谱

目的:应用基质辅助激光解析电离飞行时间质谱(matrix-assisted laser desorption ionization-time of flight mass spectrometry,MALDI-TOF-MS)系统分析原发性肝细胞癌(hepatocellular carcinoma,HCC)骨转移患者血清多肽差异谱,寻找具有潜在诊断意义的血清分子标志物.方法:收集50例HCC骨转移患者和50例HCC未骨转移患者的血清,分成训练组(76例)和验证组(24例).所有样本分别经ClinProt磁珠纯化、MALDI-TOF-MS检测及ClinProTools软件进行血清多肽差异谱分析.应用液相色谱质谱联用的方法对差异多肽进行序列鉴定.应用径向基神经网络(radial basis function neural network,RBFNN)算法建立诊断模型,并对诊断模型进行单盲法实验验证.结果:HCC骨转移患者中,共获得10条差异有统计学意义[P (Wilcoxon-test)＜0.001]的多肽峰,并成功鉴定了其中7条肽段(质荷比分别为1 780.7、1 866.5、2 131.6、2 880.4、1 532.4、2 489.8和2 234.3)的氨基酸序列,这些肽段的来源蛋白分别为甲胎蛋白(alpha-fetoprotein)、凝血酶原(prothrombin)、丝甘蛋白聚糖(serglycin)、交联-α-胰蛋白酶抑制物H4重链异构体2(isoform 2 of inter-alpha-trypsin inhibitor heavy chain H4)、自噬相关蛋白16-2异构体1(isoform 1 of autophagy-related protein 16-2)、转甲状腺素蛋白(transthyretin)和纤维蛋白β链(fibrinogen beta chain).选取2组间统计学差异最显著的6条多肽峰(质荷比分别为1 535.4、1 780.7、1 866.5、2 131.6、2 880.4和2 901.9)建立的诊断模型的识别率为89.47％,预测率为82.89％;单盲法验证模型的灵敏度为83％,特异度为92％.结论:筛选获得的差异血清多肽可能成为潜在的诊断HCC骨转移的分子标志物.

Differential serum peptide profiling of patients with bone metastasis from primary hepatocellular carcinoma using MALDI-TOF-MS

Objective:To investigate differential serum peptide profiling of patients with bone metastasis from primary HCC (hepatocellular carcinoma) using MALDI-TOF-MS (matrix-assisted laser desorption ionization-time of flight mass spectrometry), and to seek the potential serum biomarkers for diagnosis of bone metastasis. Methods:The serum samples of 50 patients with bone metastasis from primary HCC and 50 primary HCC patients without bone metastasis were collected. The 100 serum samples were randomly assigned into training set (n = 76) and test set (n = 24). The differential serum peptide profiling of all serum samples were examined by using ClinProt magnetic bead-based purification followed by MALDI-TOF-MS, and the data were analyzed by ClinProTools software. The sequences of serum peptides differentially expressed were identified using liquid chromatography-mass spectrometry. The diagnostic model was established using a learning algorithm of RBFNN (radial basis function neural network) and verified in a single-blind trial. Results:Ten peptide peaks in primary HCC patients with bone metastasis were significantly different from those in primary HCC patients without bone metastasis (P < 0.001, Wilcoxon-test). Of these ten peptides, the sequences of amino acids of seven peptides were successfully identified [the mass-to-charge ratios (m/z) for these seven peptides were 1 780.7, 1 866.5, 2 131.6, 2 880.4, 1 532.4, 2 489.8 and 2 234.3, respectively]. These seven peptides were derived from alpha-fetoprotein, prothrombin, serglycin, isoform 2 of inter-alpha-trypsin inhibitor heavy chain H4, isoform 1 of autophagy-related protein 16-2, transthyretin and fibrinogen beta chain. The recognition rate and the predictive power of the diagnostic model established on the basis of six significant peptides between patients with and without bone metastasis (the m/z for these six peptides were 1 535.4, 1 780.7, 1 866.5, 2 131.6, 2 880.4 and 2 901.9, respectively) were 89.47% and 82.89%, respectively. The sensitivity and specificity of this model were 83% and 92%, respectively, in a single-blind trial. Conclusion:The differential serum peptide profiling may offer many potential molecules which can be used in the diagnosis of bone metastasis from primary HCC.