Title: Identification of differential gene expression in human endometriosis by subtractive hybridization and gene expression profiling with real-time PCR
Abstract: Objective: Endometriosis is one of the most common diseases in women at reproductive age. It is also one of the major causes of infertility. There is yet no conclusion on the etiology and pathogenesis of endometriosis. It is hypothesized in this study that specific genetic alterations and/or aberrant expression of specific genes in endometrial cells is the underlying molecular mechanism for successful implantation and growth of the disseminated endometrial cells to form endometriosis. This project focused on the differential gene expression in human endometriosis compared with the paired normal uterine endometrium. Design: By making comparisons of gene expression in endometrisis against normal tissue counterparts from the same patients, we took advantage of the 'subtractive' effect to minimize the influence of different genetic and physiological conditions on gene expression patterns, and focused on genes that distinguished normal and ectopic tissues. Materials/Methods: Subtractive hybridization using cDNAs from the ectopic and normal uterine endometrial tissues was employed to screen the differentially expressed genes in endometriosis. The subtracted cDNA was globally amplified by PCR, and cloned into a vector to generate a subtractive library. Some 750 colonies were random-selected from the libraries for differential screening with colony-lift and/or DNA dot blot hybridization. Northern blot analysis was used to confirm the differential gene expression. Real-time PCR was applied to transcriptional profiling of the identified candidate genes in multiple tissue samples. Results: Dozens of cDNA clones were isolated and analyzed with DNA sequencing. About 20 clones were confirmed to be differentially expressed by Northern blot analysis. To identify the deregulated genes contributing to the pathogenesis of endometriosis from these potential candidates, gene expression profiling in 15 pairs of clinical tissue biopsies has been performed by using Real-time PCR. The preliminary data obtained from about 30 genes are analyzed by hierarchical clustering algorithm. The genes and experimental samples are grouped on the basis of similarities of gene expression patterns. On the other hand, gene expression profile analysis has confirmed the prevalence of differential expression of several candidate genes in the endometriotic tissues, including a protooncogene and a homeobox gene. Conclusions: Further study on the roles and mechanism of these identified genes in the development and progression of human endometriosis are warranted. Supported By: National Medical Research Council, Ministry of Health, Singapore (NMRC/0311/1998).