Chromosome Missegregation in Single Human Oocytes Is Related to the Age and Gene Expression Profile

Abstract

The growing trend for women to postpone childbearing has resulted in a dramatic increase in the incidence of aneuploid pregnancies. Despite the importance to human reproductive health, the events precipitating female age-related meiotic errors are poorly understood. To gain new insight into the molecular basis of age-related chromosome missegregation in human oocytes, we combined the transcriptome profiles of twenty single oocytes (derived from females divided into two groups according to age <35 and ≥35 years) with their chromosome status obtained by array comparative genomic hybridization (aCGH). Furthermore, we compared the transcription profile of the single oocyte with the surrounding cumulus cells (CCs). RNA-seq data showed differences in gene expression between young and old oocytes. Dysregulated genes play a role in important biological processes such as gene transcription regulation, cytoskeleton organization, pathways related to RNA maturation and translation. The comparison of the transcription profile of the oocyte and the corresponding CCs highlighted the differential expression of genes belonging to the G protein-coupled receptor superfamily. Finally, we detected the loss of a X chromosome in two oocytes derived from women belonging to the ≥35 years age group. These aneuploidies may be caused by the detriment of REEP4, an endoplasmic reticulum protein, in women aged ≥35 years. Here we gained new insight into the complex regulatory circuit between the oocyte and the surrounding CCs and uncovered a new putative molecular basis of age-related chromosome missegregation in human oocytes.

Keywords: GPCRs; REEP4; RNA-seq; array CGH; chromosome aneuploidy; cumulus cells; gene expression profile; ovarian aging.

Figure 1

Gene expression profiles of single human oocytes. (A) A total of 1852 genes were differentially expressed in age group <35 vs. ≥35 years. Volcano plot of gene expression changes in oocytes <35 vs. ≥35 years. The x-axis displays the log2 fold change value and the y-axis corresponds to the negative logarithm to the base 10 of the t-test p-values. The black dots correspond to genes with no significant difference, the red dots represent dysregulated genes with a range of values with p < 0.05 and the green dots are significant with p < 0.05 (adjusted p-values). (B) GO term enrichment analysis of biological processes that were significantly over-represented when considering DEGS in oocytes <35 vs. ≥35 years. In particular, pathways related to gene transcription, apoptosis, cytoskeleton organization and to mRNA maturation and translation were over-represented. (C) GO term enrichment analysis of biological processes that were significantly over-represented in genes expressed exclusively in oocytes <35 years. (D) GO term enrichment analysis showed that the “G-protein coupled receptor signalling pathway” (GO:0007186) was significantly over-represented in genes expressed exclusively in oocytes ≥35 years. (E) Heat map of cohesin genes differentially expressed when oocytes <35 vs. ≥35 years were compared. Their fold change ranges from −1 to +4. The expression of SMC1B and MAU2 increased whereas ESCO1, ESCO2, ESPL1, SMC1A and STAG3 genes decreased in the ≥35 years age group. (F) Array comparative genomic hybridization analysis in 20 oocytes from women ranging from 19 to 42 years old. We identified the loss of the X chromosome in two subjects belonging to the ≥35 years age group.

Figure 2

Interplay between oocyte and the surrounding CCs. (A) Volcano plot of gene expression changes in oocytes vs. CCs <35 years. (B) Volcano plot of gene expression changes in oocytes vs. CCs ≥35 years. (C) GO term enrichment analysis of biological processes that were significantly over-represented in oocytes vs. CCs <35 years. The pathways with the greatest number of annotated genes were related to gene transcription regulation. (D) GO term enrichment analysis of biological processes that were significantly over-represented in oocytes vs. CCs ≥35 years. The pathways with the greatest number of annotated genes were related to gene transcription regulation, mitotic nuclear division, cell cycle, and DNA repair. (E) Classification by molecular function of DEGs in oocytes <35 years according to PANTHER GO-slim. A total of 2.5% of DEGs (twenty genes) coded for receptors and were significantly more frequently expressed in oocytes <35 years. (F) Classification by molecular function of DEGs in oocytes ≥35 years according to PANTHER GO-slim. A total of 3.3% of DEGs (twenty-seven genes) encoded for receptors and were significantly more frequently expressed in oocytes from the ≥35 years age group.

Figure 3

RNA-seq data validation. A subset of seven genes, BMPR2, CACNG2, FZD3, GRM4, GRM6, LGR4 and REEP4, was validated in three technical replicates by RT-qPCR. These genes were chosen because they showed differential expression between oocytes <35 vs. ≥35 years. In addition, they are involved in important pathways for oocyte development and function (see Result section). Error bars represent standard deviation. * p < 0.05.

Figure 4

Effect of REEP4 depletion in human primary fibroblasts. (A) Western blotting showing the downregulation of REEP4 in human fibroblasts treated with 20 nM of smart pool siRNA (3) when compared with untreated (1) and mock siRNA-treated cells (2). (B) REEP4 silencing (3) led to high frequency of aneuploidy cells when compared with untreated (1) and mock siRNA-treated cells. (C) The expression of REEP4 was analyzed in three independent biological replicates of NIG (young and old) and GM08447 (young and old) cell lines. It was higher in young (p8 and p10 for NIG and GM08447, respectively) than in old fibroblasts (p28 and p31 for NIG and GM08447, respectively). Error bars represent standard deviation. * p < 0.05.