血清miR-146a／b实时荧光定量分析及其作为分子标志物的初步评价

目的：以特异性调节Toll样受体信号通路的miR-146a／b为检测靶标，建立血清miRNAs分子荧光定量检测方法，并对其作为血清炎症分子标志物的潜在价值进行初步评价。方法：采集正常体质量儿童血清20例（对照组）、超重儿童血清20例（超重组）、肥胖儿童血清20例（肥胖组）及健康成年人血清50例（健康成人组），酚-氯仿法提取血清总RNAs，SYBRGreen实时荧光定量技术定量检测血清中miR-146a／b拷贝数，GraphpadPrism5．0软件进行统计分析并绘图。结果：对照组血清中miR-146a的拷贝数显著低于超重组（P=-0．0061）和肥胖组（P=-0．0262），超重组和肥胖组之间差异无统计学意义（P=0．0656）。miR-146a表达量的受试者工作特征曲线下面积（AUC）分析显示：对照组vs超重组AUC=O．8475，P=-0．0002；对照组vs肥胖组AUC=0．6050，P=-0．2560；超重组vs肥胖组AUC=0．5475，P=0．6073。miR-146b与miR-146a呈不同表达趋势，超重组儿童血清miR-146b拷贝数显著高于对照组（P=-0．0090）和肥胖组儿童（P=0．0023），肥胖组儿童血清中miR-146b的拷贝数均值虽略低于对照组，但差异无统计学意义（P=-0．1556）。miR-146b表达量的AUC分析显示：正常组vs超重组AUC=0．7425，P=O．0087；正常组vs肥胖组AUC=0．6325，P=0．1517；超重组vs肥胖组AUC=0．7825，P=0．0023。miR．146a／b拷贝数值变异系数（CV）大：对照组、超重组和肥胖组儿童血清中miR。146a拷贝数的cv值分别为80．94％、110．94％、175．88％；miR-146b拷贝数的cV值分别为164．11％、189．72％、152．00％。在健康成人血清中miR-146a／b呈现相近表达模式，miR-146a和miR-146b的cV值分别是37．86％、74．82％。结论：miR-146a在对照组、超重组和肥胖组的表达量变化显示其与儿童肥胖具有-定相关性，可以从整体水平说明miR-146a与体内炎症水平呈正相关关系，但是由于其在血清中拷贝数值变异较大且各组问无明确分界，不能确定其参考值范围。因此，血清miR-146a／b拷贝数差异不足以用于临床个体化诊断，血清miR-146a／b作为炎症相关分子标志物的价值尚需进一步探讨。

Analysis of serum miR-146a/b by real-time fluorescent quantitative PCR and its preliminarily evalua-tion as biomarker

To establish method for the quantitative fluorescence detection of serum miRNAs,and to preliminarily evaluate the potential value of miR-146a/b as serum inflammatory biomarkers. Using serum miR-146a/b, specifically regulating the Toll-like receptor signalling pathway, as the targets. Methods： Serum samples were collected from 20 cases of normal weight children （control group）, 20 cases of overweight children （overweight）, 20 cases of obese children （obese group） and 50 cases of healthy adults （healthy adult group）. Serum total RNAs were extracted by phenol-chloroform extracting methods, copy numbers of serum miR-146a/b were analyzed by SYBR Green real-time fluorescent quantitative PCR techniques, followed by statistical analysis using Graphpad Prism5.0 software and graphics. Results： Copy numbers of serum miR-146a in control group were significantly lower than that in overweight （P=0.0061） and obese groups （P=0.0262）, whereas no significantdifference was found between overweight group and obesity group （P=0.0656）. Area under the ROC curves （AUC） of miR-146a expression in different groups showed as follows： control group vs overweight group AUC=0.8475, P=0.0002; control group vs obese group AUC=0.6050, P=0.2560; overweight group vs obese group AUC=0.5475, P=0.6073. Expression of miR-146b and miR-146a exhibited different trends, copy numbers of serum miR-146b in overweight children were significantly higher than that of normal weight children （P=0.0090） and obese children （P=0.0023）; even though the copy numbers of serum miR-146b in obese children were slightly lower than that in control group, the difference had no statistically significant （P=0.1556）. The area under the ROC curves （AUC） of miR- 146b expression were analyzed as shown in following data： normal group vs overweight group AUC=0.7425, P=0.0087; normal group vs obese group AUC=0.6325, P=0.1517; overweight group vs obese group AUC=0.7825, P=0.0023. The coefficient of variation （CV） values of miR-146a/b copy numbers was largely diverse： CV values of miR-146a copy numbers in control, overweight and obese children were 80.94%, 110.94% and 175.88%, respectively; and that of miR-146b copy numbers were 164.11%, 189.72% and 152.00%, respectively. In healthy adults, the expression patterns of serum miR-146a/b were similar; CV values of miR-146a and miR-146b were 37.86% and 74.82%. Conlusion： The different expression levels of miR-146a in control, overweight and obese groups indicate that miR-146a is correlated with childhood obesity and positively correlated with levels of inflammation in the body. However, the reference range of miR-146a could not be determined on account of its widely varied serum copy numbers and no clear dividing line among different groups. Therefore, the difference of serum miR-146a/b copy number is insufficient to be used in clinical diagnosis, the values of serum miR-146a/b as biomarkers of inflammation still need to be explored further.