Transcriptome profiling of sheep granulosa cells and oocytes during early follicular development obtained by laser capture microdissection

Abstract

Background: Successful achievement of early folliculogenesis is crucial for female reproductive function. The process is finely regulated by cell-cell interactions and by the coordinated expression of genes in both the oocyte and in granulosa cells. Despite many studies, little is known about the cell-specific gene expression driving early folliculogenesis. The very small size of these follicles and the mixture of types of follicles within the developing ovary make the experimental study of isolated follicular components very difficult.The recently developed laser capture microdissection (LCM) technique coupled with microarray experiments is a promising way to address the molecular profile of pure cell populations. However, one main challenge was to preserve the RNA quality during the isolation of single cells or groups of cells and also to obtain sufficient amounts of RNA.Using a new LCM method, we describe here the separate expression profiles of oocytes and follicular cells during the first stages of sheep folliculogenesis.

Results: We developed a new tissue fixation protocol ensuring efficient single cell capture and RNA integrity during the microdissection procedure. Enrichment in specific cell types was controlled by qRT-PCR analysis of known genes: six oocyte-specific genes (SOHLH2, MAEL, MATER, VASA, GDF9, BMP15) and three granulosa cell-specific genes (KL, GATA4, AMH).A global gene expression profile for each follicular compartment during early developmental stages was identified here for the first time, using a bovine Affymetrix chip. Most notably, the granulosa cell dataset is unique to date. The comparison of oocyte vs. follicular cell transcriptomes revealed 1050 transcripts specific to the granulosa cell and 759 specific to the oocyte.Functional analyses allowed the characterization of the three main cellular events involved in early folliculogenesis and confirmed the relevance and potential of LCM-derived RNA.

Conclusions: The ovary is a complex mixture of different cell types. Distinct cell populations need therefore to be analyzed for a better understanding of their potential interactions. LCM and microarray analysis allowed us to identify novel gene expression patterns in follicular cells at different stages and in oocyte populations.

Figure 1

Quality of tissue morphology with the four staining protocols. New born ovary staining section (100 × magnifications) produced by: A. Toluidin blue. B. Hematoxylin Eosin. C. Histogen^®. D. Cresyl Violet^®. Abbreviations: Oo: oocyte, GC: granulosa cells.

Figure 2

Summary of the LCM capture of primordial follicular compartments. Representative photographs of a primordial follicle before (a) and after (b) microdissection. GC were selected (a, arrow), dissected (b1) and collected in a cap (c1). The oocyte was dissected (b2) and collected (c2). Abbreviation: Oo: oocyte.

Figure 3

Gene expression profiles. Relative quantification throughout early follicular development and in the two follicular compartments (n = 3-4 except for Ant, n = 2). Profiles showed a significant increase in gene expression for BMP15 and AMH from the primary and secondary stage, respectively. By contrast, KL and VASA genes were expressed at all stages from the primordial follicular stage on, with no significant differential expression during the process of early folliculogenesis. The Y axis corresponds to the relative expression normalized by 2 reference genes (Actin β and RPL19). The X axis corresponds to the different follicular stages of the 2 follicular compartments. Abbreviations: Pd: primordial, Pm: primary, Sec: secondary, SA: small antrum. *< 0.05, **< 0.01.

Figure 4

SOLHL2, MAEL, SIRT7 and FST expression profiles. Relative quantification of SIRT7, SOLHL2, MAEL and FST mRNA throughout early follicular development (Pd, Pm, Sec) and small antrum (SA) and in the 2 follicular compartments (n = 3-4 except for SA n = 2). The Y axis corresponds to the relative expression normalized by 2 reference genes (Actin β and RPL19). The X axis corresponds to the different follicular stages of the 2 follicular compartments. Abbreviations: Pd: primordial, Pm: primary, Sec: secondary, SA: small antrum. *< 0.05, ***< 0.001.

Figure 5

Statistically significant enriched functions of oocyte and granulosa cells. The 2 specifically expressed gene lists (oocyte/GC) were evaluated in silico using Ingenuity Pathway Analysis (IPA). Analysis revealed 3 statistically significant enriched categories (p-value < 10^-3). The X axis corresponds to the ratio between the focus genes versus the genes of the class. GC data are in blue color and oocyte data are in red color.

Figure 6

Accuracy of the gene data sets. Comparison of A - oocyte data sets and B - GC data sets with Arraztoa [29], Dadé [53], Gallardo [14] and Pan [30] data sets.