MiR-200b and miR-429 Function in Mouse Ovulation and Are Essential for Female Fertility

See allHide authors and affiliations

Science  05 Jul 2013:
Vol. 341, Issue 6141, pp. 71-73
DOI: 10.1126/science.1237999

Female Infertility

Anovulation, the failure of a woman's ovary to release an oocyte, is a major cause of female infertility. The mechanisms of ovulation have been studied extensively, with the hypothalamicpituitary axis serving as a key player in its regulation. Hasuwa et al. (p. 71, published online 13 June) describe a mechanism by which anovulation can be caused by the disruption of two microRNAs that are expressed in the pituitary gland.


Ovulation in the mouse and other mammals is controlled by hormones secreted by the hypothalamo-pituitary-ovarian axis. We describe anovulation and infertility in female mice lacking the microRNAs miR-200b and miR-429. Both miRNAs are strongly expressed in the pituitary gland, where they suppress expression of the transcriptional repressor ZEB1. Eliminating these miRNAs, in turn, inhibits luteinizing hormone (LH) synthesis by repressing transcription of its β-subunit gene, which leads to lowered serum LH concentration, an impaired LH surge, and failure to ovulate. Our results reveal roles for miR-200b and miR-429, and their target the Zeb1 gene, in the regulation of mammalian reproduction. Thus, the hypothalamo-pituitary-ovarian axis was shown to require miR-200b and miR-429 to support ovulation.

MicroRNAs (miRNAs) are 21- to 22-nucleotide-long RNAs that silence gene expression posttranscriptionally by binding to and/or cleaving the 3′-untranslated regions (UTRs) of target mRNAs (13). They function in various biological processes, including development and tumorigenesis (46). To investigate the roles of miRNAs in reproduction, we produced miRNA-disrupted gene knockout (KO) mouse lines, choosing miR-200b as a candidate because it was initially detected in the mouse testis (7). The expression of miR-200b in the testis reached its peak at 2 weeks of age and persisted into adulthood (fig. S1A) but was also detected in other organs, including the kidney, lung, and thymus (fig. S1B). We produced a double-knockout mouse line for miR-200b and miR-429 (miR-DKO), which has the same seed sequence and resides in the genomic vicinity of miR-200b (figs. S2 and S3).

We did not detect abnormalities in the testis, nor did we observe fertility defects in the miR-DKO male mice (fig. S4). However, miR-DKO female mice displayed greatly reduced fertility. When the miR-DKO females were mated with heterozygous males, they seldom became pregnant—despite successful coitus demonstrated by vaginal plug formation. The pregnancy rate of the 8-week-old miR-DKO females was at most 9% after coitus (versus 85% in wild-type and heterozygous females) with an average litter size of 6.29 ± 0.57 (n = 7) versus 7.48 ± 0.45 (n = 26) in heterozygous mice. However, these miR-DKO mice did not become pregnant again even after another 3 months of pairing with males, including the postpartum estrus period (Fig. 1A). When estrous cyclicity was examined by taking daily vaginal smears, estrus was found to be prolonged in miR-DKO mice compared with wild-type females, but the cyclicity itself was maintained (fig. S5). To eliminate the possibility that the infertile phenotype in miR-DKO female mice was caused by unknown side effects reported in some gene KO experiments (8, 9), we produced miR-200b– and miR-429–expressing transgenic mouse lines using the H1 promoter (fig. S6) and crossed them with the miR-DKO mouse line (fig. S7). As shown in Fig. 1A, miR-DKO females regained their fecundity from the transgenically expressed miR-200b and/or miR-429. These results indicate that the sterility observed in miR-DKO females was caused directly by the lack of miR-200b and/or miR-429.

Fig. 1 Infertility phenotype of miR-DKO mice.

(A) Fecundity of female mice was analyzed by natural mating; n indicates the number of female mice used in each genotype. The numbers on the bar indicate pregnancy per vaginal plug formation. (**P < 0.01, ***P < 0.001 by Fisher’s exact test) (200bTG#6: miR-200b transgenic mouse line no. 6 and 429TG#10: miR-429 transgenic mouse line no. 10). (B) Number of ovulated oocytes from mated females. Superovulation was induced with 5 IU (international units) each of PMSG and hCG at 48-hour intervals; n shows the number of females analyzed, and error bars indicate ± SEM; ***P < 0.001. (C and D) Hematoxylin and eosin (HE)–stained ovarian sections of mated heterozygous and miR-DKO mice, respectively. No corpora lutea (CL) were found in miR-DKO ovaries. (E and F) Sections of ovaries from mated mice following superovulation. Many CL were observed in the miR-DKO mice. Scale bars, 0.4 mm.

To examine whether the miR-DKO mice could ovulate, we tried to collect oocytes from the oviducts in the morning of the day that the vaginal plug was found. Oocytes were recovered in only 2 out of 20 females (Fig. 1B). To examine the ovulation of miR-DKO mice more precisely, we prepared histology sections of miR-DKO mouse ovaries on the day that the plug was formed after natural mating (Fig. 1, C and D). When sections from heterozygous mice were examined, we found many early corpora lutea, as shown in Fig. 1C, which indicated successful ovulation. In contrast, they were basically not found in miR-DKO mice, except in ~10% of the miR-DKO females, which ovulated (Fig. 1B) and formed corpora lutea. However, when we superovulated the mice by injecting pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hCG), the miR-DKO females ovulated about 27.6 ± 6.0 (mean ± SEM) oocytes per female, similar to heterozygous females (Fig. 1B). When the oocytes from superovulated miR-DKO females were subjected to in vitro fertilization, the oocytes were fertilized at a normal rate and produced phenotypically normal pups after the resulting embryos were transplanted into the oviducts of pseudopregnant female wild-type foster mothers (fig. S8). The ovaries from prepubertal 3-week-old miR-DKO mice showed normal morphology (fig. S9), and normal corpora lutea were formed in the superovulated mature miR-DKO ovaries (Fig. 1, E and F). This indicates that the miR-DKO females can undergo normal oogenesis and ovulate when hormones are supplemented exogenously, which suggests that impaired hormonal regulation prevents these mice from ovulating naturally.

We examined the hormone levels in the sera of wild-type and miR-DKO female mice. Although decreases in follicle-stimulating hormone (FSH) and estradiol were not apparent, a statistically significant decrease in serum LH, together with a decrease in serum progesterone, was observed in the female miR-DKO mice (Fig. 2, A to D). The serum LH levels in male miR-DKO mice were unaffected and were similar to that of heterozygous mice (fig. S10). Thus, the miR-DKO male mice were fertile, whereas females were infertile. We do not know the reason for this sex-dependent phenotype. Female-specific infertility caused by a decrease in LH secretion was reported for Ptprn and Ptprn2 (dense-core vesicle proteins, which relate to LH secretion) DKO mice (10) and Lhb-regulating transcription factor Egr1 KO mice (11). A decrease in LH protein content in the pituitary of miR-DKO female mice was also demonstrated by immunostaining, whereas tissue sections appeared histologically normal (Fig. 2, E to H, and fig. S11). To examine the integrity of pituitary function in miR-DKO mice, we analyzed the LH surge induced by an exogenous estradiol administration after ovariectomy (mouse LH surge protocol) (12). Both the basal and estradiol-induced LH release levels were significantly impaired in ovariectomized miR-DKO compared with control mice (Fig. 2I).

Fig. 2 Pituitary defects in miR-DKO female mice.

(A to D) The FSH (A) and LH (B) serum concentrations were measured by radioimmunoassay (RIA), and estradiol (C) and progesterone (D) levels were measured by liquid chromatography–tandem mass spectrometry (LC-MS/MS). (E to H) Histology of pituitaries from 12-week-old mice heterozygous (E and G) and miR-DKO mice (F and H). (E and F) Paraffin wax sections of pituitaries were stained with HE. (G and H) Paraffin wax sections of pituitaries were stained with a rabbit antibody against LH (Biogenesis) and an Alexa 488–conjugated secondary antibody (Invitrogen). Scale bars, 0.2 mm. (I) Induction of LH surges by estradiol injection in miR-DKO mice. Bars indicate serum LH concentrations with or without estradiol injection in ovariectomized (OVX) control mice (wild-type or heterozygous) and OVX miR-DKO mice (error bars indicate ± SEM; *P < 0.05, **P < 0.01). (J) Expression levels of miR-200b and miR-429 in the kidney, ovary, hypothalamus, and pituitary were analyzed by Northern blotting (top) using 4-μg aliquots of total RNA from wild-type female mice. Ethidium bromide–stained tRNAs were used as loading controls (bottom).

We examined the expression of miR-200b and miR-429 in the ovary, hypothalamus, and pituitary (Fig. 2J) by Northern blotting and found a high expression of miRNAs solely in the pituitary. (Kidney was also examined as a high-expression control.) The expression level changed significantly during the estrous cycle, but the absolute values remained within a narrow range (fig. S12). Working on the hypothesis that these miRNAs might act in regulating pituitary function, we searched for their target mRNAs in public databases using a computer algorithm named miGTS (miRNA Global Target Search; developed by Kyowa Hakko Kirin Co., Ltd., Tokyo, Japan). Among the putative target mRNAs listed, the 3′-UTRs of both mouse and human Zeb1 and Zeb2 have the highest number of loci that were complementary to the “seed sequence” of miR-200b and miR-429 (table S1).

An in vitro reporter assay allowed us to focus on Zeb1 as a potential target for miR-200b and miR-429 (fig. S13). Consistent with the in vitro experiments, the amount of ZEB1 in the pituitary increased in miR-DKO females, whereas ZEB1 remained unaffected in the hypothalamus, where the miR-200b and miR-429 expression levels were low (Figs. 2J and 3A). Given that the Zeb1 mRNA level in the pituitary was not altered by deficiencies in these miRNAs, the amount of ZEB1 must be regulated by a posttranscriptional mechanism (fig. S14). Moreover, the increased level of ZEB1 in miR-DKO mice was decreased to near the wild-type level in the miR-200b and miR-429 transgenic mouse lines (fig. S15). The transcriptional repressor ZEB1 has CACCT-binding ability (13), and the Lhb 5′ upstream region has multiple CACCT sites. Therefore, we performed an electrophoretic mobility shift assay (EMSA) and found that ZEB1 bound to at least three sites upstream of Lhb (fig. S16). The influence of Zeb1 on Lhb was also confirmed in a clonal gonadotroph LβT2 cell line (14). When Zeb1 was knocked down by small interfering RNA in LβT2 cells, the Lhb mRNA level increased about 80% (fig. S17). The same line of evidence was obtained in vivo. To test the hypothesis that increased ZEB1 expression in the pituitaries of miR-DKO mice might cause LH suppression, we produced transgenic mice overexpressing Zeb1 under the control of the Fshb promoter (Fig. 3, B and C). In transgenic females, the Lhb expression was decreased, and the mice were anovulatory as in miR-DKO mice (Fig. 3, D and E, and fig. S18). One possible explanation for the infertility could be the aberrant expression of Egr1, but we found no difference in Egr1 expression between wild-type and miR-DKO pituitaries (fig. S19). Combining these results, it appears that the loss of miR-200b and miR-429 led to an increase in the expression of ZEB1 in the pituitary, resulting in a decrease in Lhb expression (fig. S20).

Fig. 3 Lhb expression is regulated by ZEB1 as a potential target of miR-200b and miR-429.

(A) Total lysates (20 μg) of pituitary and hypothalamic tissues from wild-type and miR-DKO mice were subjected to Western blotting. Hypothalamic tissue with little miR-200b or miR-429 expression was used as a control. (B) The Zeb1 transgene was constructed with an Fshb promoter, Zeb1 CDS and rabbit β-globin poly(A)+ sequence. (C) Expression of Zeb1 in transgenic mouse lines. Western blotting was performed with total lysates (20 μg) of pituitaries from each mouse line (Zeb1-TG#1 and #2). (D) Fecundity of Zeb1 transgenic females was analyzed by natural mating; n indicates the number of female mice used for each genotype. The numbers on the bars indicate litter deliveries per vaginal plug formation. (***P < 0.001 by Fisher’s exact test) (E) Expression of Lhb in the pituitaries of wild-type, miR-DKO, and Zeb1-TG#1 and #2 were analyzed by quantitative reverse transcription polymerase chain reaction (normalized to Actb; error bars indicate ± SEM; ***P < 0.001).

There are reports that miR-200b and miR-429 affect the malignancy of tumors by regulating E-cadherin expression (15, 16). Although susceptibility to oncogenesis was not examined here, the homozygous mutant mice did not show any overt abnormality in organs such as the lung, thymus, or kidney (where miR-200b and miR-429 are highly expressed), and the mice appeared normal during more than a year of observation. Because the miR-200b cluster is also present in humans and the upstream regions of human LHB also contain ZEB1-binding sites, the roles of these miRNAs in maintaining normal ovulation in the mouse could also be applicable to human reproductive physiology.

Supplementary Materials

Materials and Methods

Figs. S1 to S21

Tables S1 and S2

References (1720)

References and Notes

  1. Acknowledgments: We thank the late N. B. Hecht for critical reading of the manuscript and Y. Esaki and Y. Maruyama for preparing the KO and TG mice. We also thank T. Miyazawa and T. Yoshida of Kyowa Hakko Kirin Co., Ltd. for analyzing the miRNA target predictions. This work was supported by Ministry of Education, Culture, Sports, Science and Technology of Japan and the Japan Society for the Promotion of Science KAKENHI; grant numbers 20062008, 20700364, 18700397, and 24115710. The miR-DKO mice (RBRC05917) are available from RIKEN BioResource Center with a materials transfer agreement.
View Abstract

Stay Connected to Science

Navigate This Article