Abstract
Normal human cervical epithelium consists of a differentiating continuum of cell layers. It has been hypothesized that the most basal layer consists of stem cells and that cells migrate towards the surface as they mature and differentiate. Hence, cells in different layers of the epithelium are expected to be at different stages of maturity and express different genes. Carcinogenesis is a multi-step process frequently occurring over many years. Understanding the early genetic changes, most of which manifest as changes in this maturation process, will hopefully lead to better biomarkers for detecting precancers at highest risk to progress.
Objective: By studying differences in expression between cervical epithelial layers and across various grades of cervical intraepithelial neoplasia (CIN), we hope to gain a better understanding of the molecular basis of the carcinogenic process, which will hopefully lead to novel biomarkers for the early detection of cervical cancer.
Materials and Methods: CIN biopsies representing various histopathological grades of dysplasia were used for this study. These included some from a previous study that had been stored at −80°C in RNAlater, which we embedded in O.C.T., as well as newer biopsies that were fixed immediately in Tissue-Tek Xpress Molecular Fixative and embedded in paraffin (MFPE). Specimens were sectioned on to slides following standard protocols for their embedding medium. Microdissection to separate various layers of epithelium was performed manually with a needle. The epithelium was collected as 2 layers. RNA was extracted using the TRIzol reagent method for frozen specimens and the Qiagen RNeasy FFPE kit for MFPE specimens. mRNA was labeled using the Agilent Low Input Quick Amp Labeling kit and tested on the Agilent Whole Human Genome 4×44K Microarray. Data was normalized against the array median intensity and data from replicate probes was averaged. Genes that are differentially expressed in different layers of the epithelium and/or in different grades will be identified for further study and validation through immunohistochemistry.
Results and Conclusions: An earlier attempt to use laser capture microdissection was hampered by slow cutting speed and processing time, leading to low-quality RNA and weak expression array signal. Using manual microdissection, one CIN II case was done from the frozen samples. Duplicate samples of the stroma left over from microdissection were analyzed concurrently with the two epithelial layers. The scatter observed in comparing the duplicate samples was used to define what constituted differential expression between the epithelial layers, resulting in the identification of 114 genes overexpressed in the superficial layer and 2 genes overexpressed in the basal layer. Subsequent attempts to repeat this procedure showed that many of our frozen samples had poor quality RNA. Hence, we sought out newer MFPE specimens. Results from some of these samples will also be presented. A comparison of the results between the multiple dysplastic grades should be revealing as we hope to identify genes whose altered expression may play a role in cancer progression. Our success with manual microdissection suggests that this analysis is feasible. Work is ongoing to expand our sample size and to better comprehend the biochemical implications of our results.
Citation Information: Cancer Prev Res 2011;4(10 Suppl):A14.
RSS Feeds