Tet1 Deficiency Leads to Premature Ovarian Failure

Abstract

Tet enzymes participate in DNA demethylation and play critical roles in stem cell pluripotency and differentiation. DNA methylation alters with age. We find that Tet1 deficiency reduces fertility and leads to accelerated reproductive failure with age. Noticeably, Tet1-deficient mice at young age exhibit dramatically reduced follicle reserve and the follicle reserve further decreases with age, phenomenon consistent with premature ovarian failure (POF) syndrome. Consequently, Tet1-deficient mice become infertile by reproductive middle age, while age matched wild-type mice still robustly reproduce. Moreover, by single cell transcriptome analysis of oocytes, Tet1 deficiency elevates organelle fission, associated with defects in ubiquitination and declined autophagy, and also upregulates signaling pathways for Alzheimer's diseases, but down-regulates X-chromosome linked genes, such as Fmr1, which is known to be implicated in POF. Additionally, Line1 is aberrantly upregulated and endogenous retroviruses also are altered in Tet1-deficient oocytes. These molecular changes are consistent with oocyte senescence and follicle atresia and depletion found in premature ovarian failure or insufficiency. Our data suggest that Tet1 enzyme plays roles in maintaining oocyte quality as well as oocyte number and follicle reserve and its deficiency can lead to POF.

Keywords: Tet1; aging; epigenetics; oocyte; premature ovarian failure.

FIGURE 1

Tet1 knockout causes subfertility and premature ovarian failure. (A) Average litter size of wild-type (WT) and Tet1 knockout (Tet1^–/–) female mice at three different reproductive age (young, 2–4 months; middle-age, 7–9 months; old, 10–12 months). n > 10 mice for young and middle-age group and n = 7–8 mice for old group. (B) Linear regression analysis of the age effects on litter size between WT and Tet1^–/– mice. n > 10 mice for young and middle-age group and n = 7–8 mice for old group. (C) Average weight per ovary from WT and Tet1^–/– mice. n > 7. (D) Representative images of hematoxylin and eosin (H&E) staining of ovarian sections from young, middle-age and old groups. Scale bar, 500 μm. (E) Number of primordial and primary follicles, or secondary and antral follicles from WT and Tet1^–/– mice. n = 4-6. (F) Linear regression analysis of the age effects on the number of follicles between WT and Tet1^–/– mice. n = 3-4. ns, no significant difference. (G) AMH levels in serum of WT and Tet1^–/– mice by ELISA assay (n = 4). The bars show mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. The P value is labeled if P > 0.05 (A,E).

FIGURE 2

Differential gene expression of Tet1^–/– and WT oocytes. (A) Scatter-plots showing gene expression profile between young Tet1^–/– (n = 15) and WT (n = 19) oocytes from two biological repeats. (B) Boxplot showing gene expression on chromosome X and downregulated in young Tet1^–/– oocytes compared with young WT oocytes. ****P < 0.0001. (C) Boxplot showing expression of genes on mitochondrial function and Downregulated in young Tet1^–/– oocytes. (D) KEGG enrichment results showing genes downregulated or upregulated after Tet1 deficiency. X-axis represents the corrected p value (FDR) using Benjamini and Hochberg. (E) Pheatmap showing genes enriched in autophagy signaling pathway down regulated in Tet1 deficient oocytes. (F) Pheatmap showing genes enriched in Alzheimer’s disease upregulated in Tet1 deficient oocytes. (G) GO enrichment results showing genes downregulated or upregulated after Tet1 deficiency. X-axis represents the corrected p value (FDR) using Benjamini and Hochberg. (H) Pheatmap showing genes enriched in protein polyubiquitination in Tet1 deficient oocytes. (I) Pheatmap showing genes enriched in organelle fission upregulated in Tet1 deficient oocytes.

FIGURE 3

Differential gene expression of old Tet1^–/– and WT oocytes. (A) Scatter-plots showing gene expression profile between old Tet1^–/– (n = 14) and WT (n = 19) oocytes from two biological repeats. (B) KEGG enrichment results showing genes downregulated or upregulated after Tet1 deficiency. X-axis represents the corrected p value (FDR) using Benjamini and Hochberg. (C) Pheatmap showing genes enriched in insulin signaling pathway in Tet1 deficient oocytes. (D) Pheatmap showing genes enriched in spliceosome in Tet1 deficient oocytes. (E) GO enrichment results showing genes downregulated or upregulated after Tet1 deficiency. X-axis represents the corrected p value (FDR) using Benjamini and Hochberg. (F) Pheatmap showing genes enriched in proteasomal protein catabolic process in Tet1 deficient oocytes. (G) Pheatmap showing genes enriched in DNA repair in Tet1 deficient oocytes.

FIGURE 4

Differential gene expression between young and old Tet1^–/– oocytes. (A) Scatter plot displaying differential gene expression between young and old Tet1^–/– oocytes. (B) Venn plot showing the upregulated and downregulated genes between young and old WT oocytes, and young and old Tet1^–/– oocytes. (C) Slope plot showing the change between WT and Tet1^–/– oocytes. (D) KEGG enrichment results of differential genes between young and old Tet1^–/– oocytes. X-axis represents the p-value. (E) GO enrichment results of differential genes between young and old Tet1^–/– oocytes. X-axis represents the p-value.

FIGURE 5

Expression of transposon elements (TEs) of Tet1^–/– and WT oocytes. (A) Barplot showing the proportion of TEs in young Tet1^–/–, WT, old Tet1^–/–, and WT oocytes. (B) Barplot displaying the proportion of ERVs in young Tet1^–/–, WT, old Tet1^–/–, and WT oocytes. (C) Z-score showing expression of ERVL-MaLR between young Tet1^–/– and WT oocytes. Data are shown as mean ± SEM. **P < 0.01. (D) Volcano showing the differential TEs between young Tet1^–/– and WT oocytes. Orange represented upregulated TEs compared with young WT oocytes. (E) Expression of Full-length L1 family in young Tet1^–/– and WT oocytes, and old Tet1^–/– and WT oocytes.