Development of Laser Capture Microdissection to study the gene expression of the sheep early ovarian folliculogenesis

Background Correct achievement of early ovarian folliculogenesis is a crucial phase for further ovarian function. This process is closely regulated by cell-cell interactions and coordinated expression of genes from oocyte and granulosa cells. But, despite of the large number of studies, little is known about the precise gene expression patterns driving early folliculogenesis. The experimental limitations concerned the very small size of these follicles and the mixture of the different developmental stages within an ovary that make the study of isolated follicular components much more difficult. The recently developed laser capture microdissection (LCM) technique coupled with microarrays experiments is promising in addressing the molecular specificity of each follicular compartment. Nevertheless, the isolation of unique cells or group of cells is still challenging to maintain RNA quality during this process and to obtain sufficient amount of RNA. In this study, we described a method allowing the analysis of oocyte and granulosa cells gene expression during the first stages of sheep early folliculogenesis. Results First we developed a new fixation protocol using a frizzed 70% ethanol fixation solution that ensures correct single cell capture and RNA integrity during microdissection time. After LCM capture of the compartments and follicular stages, RNA extraction and amplification, the expression of 6 oocyte-specific genes (SOHLH2, MAEL, MATER, VASA, GDF9, BMP15) and 3 granulosa cell-specific genes (KITLG, GATA4, AMH) confirmed the purity of the samples and documented their ovine expression profiles. Then, using bovine Affymetrix chip, we identified for the first time, a global gene expression for each follicular compartment during early developmental stages. Particularly the granulosa cell data set is quite unique. 1050 granulosa cell specific transcripts compared to oocyte and 759 oocyte specific transcripts were detected. The analysis of the expression of 2 genes (SIRT7, FST) confirmed this specificity of expression. Finally, the integration of the data stated the 3 main physiological events involved in early folliculogenesis and provided descriptive elements that confirmed the relevance and the potential of the LCM-derived RNAs. Conclusions This method should contribute through an additional genome wide expression profiling to give insights on molecular mechanisms involved in stage transitions and cell type interplays. The 2 ovine follicular compartments (i.e. granulosa cells (G) and oocytes (O) were captured using LCM technology for each early stage (primordial (Pd), primary (Pm), secondary (Sec) follicles. The RNA of each group was extracted using Picopure RNA Isolation kit (Arcturus) and subjected to 2 round T7 amplification (RiboAmp®HS PLUS kit, Arcturus). Ovine microarray experiments were performed using the Affymetrix Bovine Expression Array. First the quality of the cross-species hybridizations was checked by comparison of hybridization data of ovine fetal ovary RNA with bovine fetal ovary ones, generated with the Affymetrix standard protocol (protocole 1). Then, three biotin-labeling protocols were compared from ovine fetal ovary total RNA: protocol 1; protocol 2 (Biotin-labeled cRNAs were synthesized following Affymetrix protocol using the second-round cDNAs from RiboAmp®HS kit as templates); protocol 3 (Arcturus biotin turboTM labeling kit from aRNA after the 2 round amplification using RiboAmp®HS kit). Last , one LCM-derived aRNA sample of each group was labeled using the Arcturus biotin turboTM labeling kit (protocol 3) and hybridized to Affymetrix Bovine Expression arrays. Images were interpreted using Microarray Suite version 5.0 (MAS 5.0) in GCOS with scaling (100) and without normalization.