Descazeaud Aurelien, Rubin Mark A, Hofer Matthias, Setlur Sunita, Nikolaief Nathalie, Vacherot Francis, Soyeux Pascale, Kheuang Laurence, Abbou Claude C, Allory Yves, de la Taille Alexandre
INSERM, Unité 841, IMRB, Faculté de Médecine, University Paris 12, France.
Diagn Mol Pathol. 2008 Dec;17(4):207-13. doi: 10.1097/PDM.0b013e31816f6352.
The aim of the current study was to analyze gene expression profiles in benign prostatic hyperplasia and to compare them with phenotypic properties. Thirty-seven specimens of benign prostatic hyperplasia were obtained from symptomatic patients undergoing surgery. RNA was extracted and hybridized to Affymetrix Chips containing 54,000 gene expression probes. Gene expression profiles were analyzed using cluster, TreeView, and significance analysis of microarrays softwares. In an initial unsupervised analysis, our 37 samples clustered hierarchically in 2 groups of 18 and 19 samples, respectively. Five clinical parameters were statistically different between the 2 groups: in group 1 compared with group 2, patients had larger prostate glands, had higher prostate specific antigen levels, were more likely to be treated by alpha blockers, to be operated by prostatectomy, and to have major irritative symptoms. The sole independent parameter associated with this dichotome clustering, however, was the prostate gland volume. Therefore, the role of prostate volume was explored in a supervised analysis. Gene expression of prostate glands <60 mL and >60 mL were compared using significance analysis of microarrays and 227 genes were found differentially expressed between the 2 groups (>2 change and false discovery rate of <5%). Several specific pathways including growth factors genes, cell cycle genes, apoptose genes, inflammation genes, and androgen regulated genes, displayed major differences between small and large prostate glands.
本研究的目的是分析良性前列腺增生中的基因表达谱，并将其与表型特征进行比较。从接受手术的有症状患者中获取了37份良性前列腺增生标本。提取RNA并与含有54000个基因表达探针的Affymetrix芯片杂交。使用聚类、TreeView和微阵列软件的显著性分析来分析基因表达谱。在最初的无监督分析中，我们的37个样本分别分层聚为18个样本和19个样本的两组。两组之间有五个临床参数在统计学上存在差异：与第2组相比，第1组患者的前列腺更大，前列腺特异性抗原水平更高，更有可能接受α受体阻滞剂治疗、接受前列腺切除术，并且有更严重的刺激性症状。然而，与这种二分聚类相关的唯一独立参数是前列腺体积。因此，在有监督分析中探讨了前列腺体积的作用。使用微阵列显著性分析比较了前列腺体积<60 mL和>60 mL的基因表达，发现两组之间有227个基因差异表达（变化>2倍且错误发现率<5%）。包括生长因子基因、细胞周期基因、凋亡基因、炎症基因和雄激素调节基因在内的几个特定途径，在小前列腺和大前列腺之间显示出主要差异。
