ELAVL2‐directed RNA regulatory network drives the formation of quiescent primordial follicles
2019; Springer Nature; Volume: 20; Issue: 12 Linguagem: Inglês
10.15252/embr.201948251
ISSN1469-3178
AutoresYuzuru Kato, Tokuko Iwamori, Y Ninomiya, Takashi Kohda, Jyunko Miyashita, Mikiko Sato, Yumiko Saga,
Tópico(s)RNA regulation and disease
ResumoArticle28 October 2019free access Transparent process ELAVL2-directed RNA regulatory network drives the formation of quiescent primordial follicles Yuzuru Kato Corresponding Author [email protected] orcid.org/0000-0002-4001-3792 Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan Department of Genetics, SOKENDAI, Mishima, Shizuoka, Japan Search for more papers by this author Tokuko Iwamori Department of Biomedicine, Research Center for Human Disease Modeling, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan Search for more papers by this author Youichirou Ninomiya Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan Search for more papers by this author Takashi Kohda orcid.org/0000-0002-6856-5601 Faculty of Life and Environmental Sciences, University of Yamanashi, Kofu, Yamanashi, Japan Search for more papers by this author Jyunko Miyashita Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan Search for more papers by this author Mikiko Sato Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan Search for more papers by this author Yumiko Saga Corresponding Author [email protected] orcid.org/0000-0001-9198-5164 Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan Department of Genetics, SOKENDAI, Mishima, Shizuoka, Japan Search for more papers by this author Yuzuru Kato Corresponding Author [email protected] orcid.org/0000-0002-4001-3792 Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan Department of Genetics, SOKENDAI, Mishima, Shizuoka, Japan Search for more papers by this author Tokuko Iwamori Department of Biomedicine, Research Center for Human Disease Modeling, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan Search for more papers by this author Youichirou Ninomiya Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan Search for more papers by this author Takashi Kohda orcid.org/0000-0002-6856-5601 Faculty of Life and Environmental Sciences, University of Yamanashi, Kofu, Yamanashi, Japan Search for more papers by this author Jyunko Miyashita Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan Search for more papers by this author Mikiko Sato Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan Search for more papers by this author Yumiko Saga Corresponding Author [email protected] orcid.org/0000-0001-9198-5164 Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan Department of Genetics, SOKENDAI, Mishima, Shizuoka, Japan Search for more papers by this author Author Information Yuzuru Kato *,1,2, Tokuko Iwamori3, Youichirou Ninomiya1,†, Takashi Kohda4, Jyunko Miyashita1, Mikiko Sato1 and Yumiko Saga *,1,2 1Division of Mammalian Development, Genetic Strains Research Center, National Institute of Genetics, Mishima, Shizuoka, Japan 2Department of Genetics, SOKENDAI, Mishima, Shizuoka, Japan 3Department of Biomedicine, Research Center for Human Disease Modeling, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan 4Faculty of Life and Environmental Sciences, University of Yamanashi, Kofu, Yamanashi, Japan †Present address: Research Centre for Medical Bigdata, National Institute of Informatics, Tokyo, Japan *Corresponding author. Tel: +81 55 981 6832; Fax: +81 55 981 6828; E-mail: [email protected] *Corresponding author. Tel: +81 55 981 6829; Fax: +81 55 981 6828; E-mail: [email protected] EMBO Rep (2019)20:e48251https://doi.org/10.15252/embr.201948251 PDFDownload PDF of article text and main figures.AM PDF Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Formation of primordial follicles is a fundamental, early process in mammalian oogenesis. However, little is known about the underlying mechanisms. We herein report that the RNA-binding proteins ELAVL2 and DDX6 are indispensable for the formation of quiescent primordial follicles in mouse ovaries. We show that Elavl2 knockout females are infertile due to defective primordial follicle formation. ELAVL2 associates with mRNAs encoding components of P-bodies (cytoplasmic RNP granules involved in the decay and storage of RNA) and directs the assembly of P-body-like granules by promoting the translation of DDX6 in oocytes prior to the formation of primordial follicles. Deletion of Ddx6 disturbs the assembly of P-body-like granules and severely impairs the formation of primordial follicles, indicating the potential importance of P-body-like granules in the formation of primordial follicles. Furthermore, Ddx6-deficient oocytes are abnormally enlarged due to misregulated PI3K-AKT signaling. Our data reveal that an ELAVL2-directed post-transcriptional network is essential for the formation of quiescent primordial follicles. Synopsis The RNA-binding proteins ELAVL2 and DDX6 drive the assembly of large P body-like cytoplasmic RNP granules in oocytes and the formation of quiescent primordial follicles. ELAVL2 is required for the formation of primordial follicles in mice. Large P body-like granules are assembled in oocytes prior to the formation of primordial follicles in an ELAVL2-dependent manner. ELAVL2-dependent DDX6 translation is crucial for the assembly of P body-like granules. DDX6 is required for maintaining the quiescent state of primordial follicles. Introduction Long-term reproduction of mammalian females is achieved by controlled sustainability and growth of the most primitive ovarian follicles called primordial follicles, which are composed of a single oocyte and a few of the supporting somatic cells, granulosa cells. Given the importance of primordial follicles as a finite reservoir of eggs in mammalian ovaries, understanding the molecular mechanisms underlying the formation and sustainability of primordial follicles is essential for reproductive biology and medicine. Mammalian oogenesis begins once primordial germ cells enter the embryonic ovary, where female gonocytes (or oogonia) proliferate with incomplete cytokinesis. Therefore, cells are connected to each other via intercellular bridges 1. Eventually, gonocytes initiate meiosis after receiving retinoic acid signaling 2, 3. Primordial follicles arise from oocytes in cysts by breaking down intercellular connections between oocytes, a process termed cyst breakdown, which occurs within a few days after birth in mice 4. Substantial effort has been made to understand the molecular mechanisms underlying the formation of primordial follicles. FIGLA was initially identified as an oocyte-specific transcription factor required for primordial follicle formation 5. Subsequently, several signaling pathways, Notch, TGFβ, and JNK, were reported to be involved in this process 6-9. More recently, in vitro production of functional eggs from primordial germ cells and pluripotent stem cells has been achieved in mice, which provides a great opportunity to understand the molecular basis of primordial follicle formation 10, 11. However, the production of quiescent primordial follicles remains challenging because follicle growth is simultaneously activated in vitro. In other words, these studies raised a new question of how primordial follicles become quiescent. In addition to transcriptional gene regulation, post-transcriptional RNA regulation also plays a fundamental role in the regulation of gene expression. Traditionally, studies on reproductive biology have documented integral contributions of post-transcriptional RNA regulation mediated by species-specific and evolutionarily conserved RNA-binding proteins (RBPs) to oocyte development 12. Consistent with this, many important processes of mammalian oogenesis, including sexual differentiation and meiosis in embryonic ovaries, and oocyte growth or maturation in postnatal ovaries, require the functions of RBPs 13-18. However, the role of post-transcriptional RNA regulation in primordial follicle formation remains unknown. In this regard, several studies have suggested some mechanistic differences in terms of post-transcriptional RNA regulation between oocytes in cysts and primordial follicles. The expression of MSY2, a germline-specific RBP involved in messenger RNA (mRNA) stability and translational suppression, was reported to precede primordial follicle formation 5, 15, 19, 20. In addition, another germline-specific RBP, DAZL, is translationally suppressed coinciding with the formation of primordial follicles 21. These observations prompted us to investigate the roles of post-transcriptional RNA regulation in the formation of primordial follicles. In this study, we explored RBPs involved in the formation of primordial follicles in the mouse and identified ELAVL2 (embryonic lethal and abnormal vision-like 2) as an essential RBP. ELAVL2 (also known as HuB) is a member of the ELAVL RNA-binding proteins, which promote translation by associating with mRNAs containing AU-rich elements 22-24. We found that ELAVL2 associates with mRNAs encoding components of stress granules and processing bodies (P-bodies), cytoplasmic ribonucleoprotein (RNP) granules involved in mRNA decay, storage, or translational stalling 25, in the ovary. Indeed, we discovered that large P-body-like granules were assembled in oocytes prior to the formation of primordial follicles in an ELAVL2-dependent manner. To investigate the potential importance of the P-body-like granules for primordial follicle formation, we focused on Ddx6, a target gene of ELAVL2 that encodes a central component of P-body assembly 26, 27. We found that Ddx6 is required for the assembly of the P-body-like granules and the formation of quiescent primordial follicles. Our study provides a framework to understand the molecular basis of the post-transcriptional network underlying the formation of primordial follicles. Results Identification of ELAVL2 as a candidate RNA-binding protein involved in the formation of primordial follicles Primordial follicles are formed in neonatal ovaries. To explore the genes encoding RNA-binding proteins (RBPs) involved in the formation of primordial follicles, we performed a microarray analysis using total RNA isolated from embryonic day (E) 15.5, 17.5, and newborn (P0) ovaries, and selected relevant genes whose expression increased by more than twofold from E15.5 to P0. As a result, 55 gene probes encoding RBPs (GO: 0003723) were selected (Fig EV1A). We further focused on 17 genes among the 55 genes whose expression increased by more than twofold from E17.5 to P0, of which three genes, G3bp2, Elavl2, and Dppa5, had the strongest expression in the P0 ovary (Fig EV1B). As Dppa5 knockout mice do not exhibit any abnormalities 28, G3bp2 and Elavl2 were selected to examine female-specific expression. Reverse transcription and quantitative polymerase chain reaction (RT–qPCR) revealed that Elavl2 was predominantly expressed in the ovary, whereas G3bp2 was increased toward birth not only in the ovary but also in the testis (Fig EV1C). As primordial follicle formation is a female-specific developmental event, we selected Elavl2 as a strong candidate. Immunostaining analysis demonstrated that ELAVL2 was specifically expressed in oocytes marked by c-KIT (Fig EV1D). The specificity of the antibodies is shown in Fig 1C. Click here to expand this figure. Figure EV1. Screening of female-specific genes encoding RNA-binding proteins Heat map of genes encoding RNA-binding proteins (n = 3 independent ovary samples at each developmental stage). Genes whose expression levels were increased by more than twofold in ovaries between E15.5 and P0 are listed. Red and green indicate high- and low-processed signals, respectively. Genes whose expression increased by more than twofold in ovaries between E17.5 and P0 were selected from 55 gene probes in (A). Reverse transcription and quantitative polymerase chain reaction (RT–qPCR) analysis of Elavl2 and G3bp2 in XX and XY gonads from E12.5 to P1 (n = 3–5 animals). Error bars, ±SD. Immunostaining of ELAVL2 in P0 ovaries (n = 3 animals). c-KIT was used as a germ cell marker. DNA was counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Download figure Download PowerPoint Figure 1. Loss of oocytes shortly after birth in Elavl2-deficient ovaries Periodic acid–Schiff (PAS) staining in wild-type (Elavl2+/+) and Elavl2 knockout (Elavl2Δ/Δ) adult ovaries (left) and testes (right) (n = 3 animals for each genotype). Immunostaining of a germ cell marker, MVH/DDX4, in neonatal ovaries (n = 3 animals for each genotype). DNA was counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Immunostaining of ELAVL2 newborn ovaries (n = 2 animals for each genotype). Negative; negative control. Download figure Download PowerPoint Elavl2 is indispensable for primordial follicle formation ELAVL2 is a member of the ELAVL RNA-binding proteins (ELAVL1-4), which contain three RNA recognition motifs and associate with mRNAs containing AU-rich elements. ELAVL family proteins are involved in several post-transcriptional events, including alternative splicing, mRNA stability, and translation in somatic cells 29. To examine the role of ELAVL2 in the formation of primordial follicles, we generated Elavl2 knockout mice (Appendix Fig S1A and B). Whereas a previous report indicates that ELAVL2 has oocyte-specific isoforms 30, Western blot analysis revealed that our Elavl2 knockout mice were null mutants (Appendix Fig S1C). Elavl2 knockout mice (Elavl2Δ/Δ) were born according to the Mendelian inheritance (Appendix Fig S1D). Although more than 80% of homozygous mutants progressively died by the weaning stage due to growth retardation (Appendix Fig S1E), the surviving pups grew to adulthood. We found that Elavl2 knockout ovaries were markedly small in size and oocytes were lost in adults, whereas mutant testes exhibited no clear abnormalities in morphology (Fig 1A). To determine the cause of the ovarian defects, a time-course immunostaining analysis was conducted on Elavl2 knockout neonatal ovaries using mouse vasa homolog (MVH, also known as DDX4) as a germ cell marker. The specificity of the antibody was shown (Appendix Fig S2A). We found that Elavl2 knockout ovaries progressively lost oocytes after birth and oocytes were hardly observed in mutant ovaries at a week after birth (Fig 1B), indicating that Elavl2 is required for the survival of oocytes in neonatal ovaries. Immunostaining analysis confirmed that ELAVL2 expression was not detected in mutant ovaries (Fig 1C). We then asked whether cyst breakdown occurs in Elavl2 knockout ovaries. To distinguish oocytes in cysts from those in primordial follicles, we performed immunostaining for an extracellular matrix protein, LAMININ (Fig 2A). The specificity of the antibodies was shown (Appendix Fig S2B). Primordial follicle formation began in the medullary area indicated by the LAMININ layer that encloses a single oocyte and surrounding granulosa cells (Fig 2A, asterisks), whereas oocytes in the cortical area were still connected with each other, represented by LAMININ-enclosed cysts in wild-type newborn ovaries (Fig 2A, dashed circles). On the other hand, primordial follicles were hardly observed in the medullary area, represented by LAMININ-enclosed oocytes that were broadly observed in Elavl2 knockout ovaries (Fig 2A). We next examined the number of intercellular bridges (ICBs), cell junctions that connect oocytes in cysts, by performing immunostaining for the ICB markers TEX14 and MKLP1 31 (Fig 2B). ICBs were detected as small dot- or ring-like signals as indicated by white arrowheads (Fig 2B). We confirmed that these small foci were specifically observed in a first antibody-dependent manner (Appendix Fig S2C). Quantification of TEX14 and MKLP1 foci revealed that the number of ICBs was approximately doubled in Elavl2 knockout ovaries (Fig 2C). These results indicated that cyst breakdown is severely compromised in Elavl2 knockout ovaries. Figure 2. Elavl2 is indispensable for primordial follicle formation Immunostaining of an extracellular matrix protein, LAMININ, and a germ cell marker, c-KIT in wild-type (Elavl2+/+) and Elavl2 knockout (Elavl2Δ/Δ) newborn ovaries (n = 3 animals for each genotype). Immunostaining of TEX14 and MKLP1 in newborn ovaries (n = 3 animals for each genotype). Oocytes were marked by DAZL. Scale bars, 50 μm. Quantification of TEX14 and MKLP1 foci (n = 3 animals) in (B). Microarray analysis of wild-type (WT) and Elavl2 knockout ovaries (n = 3 independent ovary samples). Gene probes up- or down-regulated by more than twofold in Elavl2 knockout ovaries are shown. Expression profile of selected genes involved in follicular and early zygotic development in (D). Reverse transcription and quantitative polymerase chain reaction (RT–qPCR) of selected genes (n = 5–7 animals). Scatter plot analysis of gene probes whose expression increased by more than twofold from E17.5 to P0 in WT ovaries in (D). Probes demonstrating a greater than 1.5-fold difference between wild-type and Elavl2 knockout ovaries are shown. Data information: (C, F, and G) Circles indicate individual data. Error bars, ±SD. The significance is indicated (two-tailed Student's t-test; ***P < 0.0005, **P < 0.005, *P < 0.05). Download figure Download PowerPoint To evaluate the influence of Elavl2 deficiency on gene expression, we performed a time-course microarray analysis by extracting total RNA from Elavl2 knockout E15.5, E17.5, and P0 ovaries. We found that gene expression (more than twofold difference) was greatly changed in mutant P0 ovaries (946 gene probes, Fig 2D), consistent with the timing of primordial follicle formation, in which the number of decreased gene probes (646) was approximately double that of increased gene probes (300) (Datasets EV1 and EV2). Among the decreased gene probes, representative genes involved in folliculogenesis (Gdf9), zona pellucida formation (Zp1, Zp2, and Zp3), and early zygotic development (Npm2 and Zar1) were included (Fig 2E) 32-37. These genes were reported to be directly or indirectly up-regulated by the oocyte-specific transcription factors FIGLA or NOBOX 5, 38. RT–qPCR data revealed that Figla and Nobox expression was also decreased in Elavl2 knockout ovaries together with a greater reduction of their downstream genes, including Bmp15 and Rfpl4 38 (Fig 2F). We normalized the data using Mvh because its expression level in Elavl2 knockout P0 ovaries was similar to wild-type in our microarray data (fold change: 1.02) and RT–qPCR (Ct values: 25.27 ± 0.39 in wild-type and 25.23 ± 0.64 in Elavl2 knockout ovaries). As expression of these genes increased toward birth (Fig 2E), we further examined the influence of Elavl2 deficiency on the expression of cognate genes. Among 735 gene probes whose expression increased from E17.5 to P0 in wild-type ovaries, more than half (418 gene probes, 56.9%) demonstrated decreased expression in Elavl2 knockout P0 ovaries (Fig 2G). These data suggest that Elavl2 deficiency globally influences oocyte gene expression. Based on the results of immunostaining and gene expression analysis, we concluded that ELAVL2 is an indispensable RBP involved in the formation of primordial follicles. ELAVL2 associates with mRNAs encoding components of P-bodies, which assembled in oocytes prior to the formation of primordial follicles To search for possible targets of ELAVL2, we performed RNA immunoprecipitation followed by sequencing (RIP-seq) analysis using wild-type newborn ovaries. Western blot analysis confirmed successful immunoprecipitation of ELAVL2 (Fig 3A). As a result, 2519 genes were identified as putative ELAVL2-associating mRNAs according to our stringent criteria (IP/input > 2, q-value < 10−10, Dataset EV3). Using this gene list, gene ontology analysis was carried out to explore the cellular events that ELAVL2-associating mRNAs are involved in. Although ELAVL2-associated mRNAs were varied in terms of molecular or cellular events, we found that four terms related to RNA processing were included in the top 10 of the list (Fig 3B), among which three terms (CCR4-NOT, cytoplasmic stress granule, and cytoplasmic processing body) were related to cytoplasmic ribonucleoprotein (RNP) granules (Dataset EV4). Figure 3. ELAVL2 associates with mRNAs encoding components of P-bodies that assembled in oocytes prior to the formation of primordial follicles A. RNA immunoprecipitation of ELAVL2 using WT newborn ovaries. Immunoprecipitated ELAVL2 was confirmed by Western blotting. MVH was used as a negative control (n = 3 independent ovary samples). B. Gene ontology analysis (DAVID ver. 6.8) of ELAVL2-associating mRNAs. C. Immunostaining of DDX6 in E17.5 (n = 3 animals) and P1 (n = 10 animals) ovaries. DAZL was used as an oocyte marker. D. Quantification of oocytes containing DDX6 foci (n = number of oocytes from two animals at each stage). E. Images analysis of DDX6 foci. Images of DDX6 from P0 and P2 ovaries (left) were used to create masks of DDX6 foci (middle). Merged images are shown on the right. F. Quantification of DDX6 foci (n = number of oocytes from two animals at each stage). G–I. Immunostaining of DCP1A (G) (n = 6 animals), AGO2 (H) (n = 3 animals), and TIAR (I) (n = 5 animals) together with DDX6 in newborn ovaries. DNA was counterstained with DAPI. Download figure Download PowerPoint Cytoplasmic RNP granules are membraneless biomolecular condensates induced by liquid–liquid phase separation 39. Accumulating evidence has demonstrated that these RNP granules are closely associated with numerous biological processes, including cell survival, neurodegenerative diseases, and germline development 40-42. We thus hypothesized that primordial follicle formation involves ELAVL2-directed regulation of cytoplasmic RNP granules. To address this question, we first investigated the subcellular localization of cytoplasmic RNP granules in perinatal oocytes by focusing on a DEAD-box RNA helicase, DDX6, because DDX6 is a component of both P-bodies 26, 27 and stress granules 43. We found that DDX6 formed aggregate-like foci in newborn oocytes (Fig 3C, arrowheads), whereas it was diffused in the oocyte cytoplasm at E17.5 (Fig 3C). The stage-specific existence of granular signals was confirmed by immunostaining of negative control (Appendix Fig S3A). The number of oocytes containing clear DDX6 foci increased toward birth (Fig 3D), suggesting that DDX6-containing RNP granules are assembled prior to the formation of primordial follicles. DDX6 foci were also observed in oocytes after primordial follicles had formed (Fig 3E, arrows); however, their size was slightly smaller than those in cysts (Fig 3E, arrowheads). Thus, we quantified their size by image analysis (Fig 3E, middle) and found that the proportion of large foci decreased in primordial follicles (Fig 3F, magenta). Instead, smaller foci became more abundant in primordial follicles (Fig 3F, blue and green). We next performed co-immunostaining using antibodies for P-body components, DCP1A 44, 45 and AGO2 46, 47, and a stress granule component, TIAR 48, together with DDX6. As a result, DCP1A and AGO2, but not TIAR, colocalized to DDX6 foci in newborn oocytes (Fig 3G–I). The specificity of the antibodies was represented by DCP1A (Appendix Fig S3B). These results demonstrate that large P-body-like granules were assembled in oocytes prior to the formation of primordial follicles. ELAVL2-dependent DDX6 translation is required for the assembly of P-body-like granules We next analyzed the assembly of P-body-like granules in Elavl2 knockout newborn ovaries. Immunostaining of P-body components revealed that the assembly of P-body-like granules was impaired in Elavl2 knockout ovaries (Fig 4A and B). To quantitatively analyze the results of immunostaining, we performed image analysis and measured the area of DDX6 foci per unit area of oocytes (1 μm2). The area of DDX6 foci was decreased to approximately one-fourth in Elavl2 knockout ovaries (Fig 4C), indicating that massive assembly of P-body-like granules requires ELAVL2 function. The expression changes in mRNAs related to P-body were less than twofold in Elavl2 knockout P0 ovaries (Fig 4D), suggesting that ELAVL2 has little effect on mRNA stability. We then examined protein expression by focusing on DDX6 as a model. According to Western blot analysis, DDX6 expression in Elavl2 knockout ovaries decreased to 34% of that in wild-type (Fig 4E), suggesting that ELAVL2 is required for the translation or stability of DDX6. In addition, DCP1A and AGO2 expression appeared to be weaker in Elavl2 knockout ovaries (Fig 4A and B). Although we could not detect their expression changes by Western blotting due to high backgrounds, it is possible that ELAVL2 broadly influences for protein expression of the components of P-body-like granules. Figure 4. ELAVL2-dependent DDX6 translation is required for the assembly of P-body-like granules A, B. Immunostaining of DDX6 together with DCP1A (A) (n = 6 animals for each genotype) and AGO2 (B) (n = 3 animals for each genotype) in wild-type (Elavl2+/+) and Elavl2 knockout (Elavl2Δ/Δ) newborn ovaries. C. Quantification of the area of DDX6 foci in newborn ovaries. Twenty (wild-type) and 21 (Elavk2 knockout) ovarian sections from two animals were immuno-stained by anti-DDX6 antibody and analyzed. D. Expression changes of mRNAs encoding components of P-bodies. Blue bars, fold enrichment of ELAVL2-associating mRNAs in Fig 3B. Red bars, fold expression changes of cognate mRNAs by microarray (P0 ovaries). E. Western blotting of DDX6 in newborn ovaries. The expression level of DDX6 in Elavl2 knockout ovaries was normalized by MVH and represented as a ratio to WT. F. Co-transfection assay of full-length and mutant ELAVL2 with gfp reporters in HeLa cells. G. Detection of immunoprecipitated ELAVL2 by Western blotting. ACTB was used as a negative control. H. RT–qPCR analysis of gfp reporters. The vertical axis indicates relative quantity of immunoprecipitated gfp to inputs (n = number of experiment). Data information: (C and H) Circles represent individual data. Error bars, ±SD. Boxes and horizontal bands in boxes represent quartile deviations and median, respectively. The significance of changes is indicated (Wilcoxon rank sum test in (C) and two-tailed Student's t-test in (H); ***P < 0.0005, **P < 0.005). Download figure Download PowerPoint To further clarify the interaction between ELAVL2 and Ddx6 mRNA, Flag-tagged ELAVL2 was transfected into HeLa cells together with a gfp reporter (Fig 4F). Successful expression and immunoprecipitation of ELAVL2 was confirmed by Western blotting (Fig 4G). RT–qPCR analysis revealed that interaction between ELAVL2 and the gfp reporter was greatly increased in a Ddx6 3′-UTR sequence-dependent manner (Fig 4H). ELAVL family proteins have three conserved RNA recognition motifs (RRM1-3), among which RRM1 and RRM2 cooperatively bind to AU-rich elements 49. To test the role of RRM1 and RRM2 in the binding of the 3′-UTR sequence of Ddx6, mutant ELAVL2, which carried a large deletion in RRM1 and RRM2, was transfected (Fig 4F and G). We found that the interaction was diminished to one-tenth in the RRM mutant (Fig 4H). These data suggest that ELAVL2 is involved in the translation of DDX6 by directly associating with its 3′-UTR sequence. DDX6 is required for the assembly of P-body-like granules and the formation of primordial follicles The defective assembly of P-body-like granules in Elavl2 knockout ovaries prompted us to ask the importance of the ELAVL2-directed assembly of P-body-like granules for the formation of primordial follicles. To address this question, we knocked out Ddx6 in an oocyte-specific manner using Mvh-Cre 50 based on the requirement of DDX6 for P-body assembly in somatic cells 26 (Appendix Fig S4A and B). Hereafter, we refer to this mouse line as Ddx6 cKO. On immunostaining analysis of P-body components, the assembly of P-body-like granules was severely compromised in Ddx6 cKO newborn oocytes (Fig 5A and B). Quantification of DCP1A foci by image analysis revealed that the mean area of foci was reduced to 15% in Ddx6 cKO ovaries (Fig 5C). As ELAVL2 was robustly expressed in Ddx6 cKO oocytes (Fig 5D), it is unlikely that the defective assembly of P-body-like granules was due to the reduction of ELAVL2. These results indicate that Ddx6 is required for the assembly of P-body-like granules. Figure 5. DDX6 is required for the assembly of P-body-like granules and the formation of primordial follicles A, B. Immunostaining of DDX6 together with DCP1A (A) (n = 3 animals for each genotype) and AGO2 (B) (n = 3 animals for each genotype) in newborn ovaries. DNA was counterstained with DAPI. C. Quantification of the area of DCP1A foci in newborn ovaries. Twenty-eight (control) and 25 (Ddx6 cKO) ovarian sections from two animals were immuno-stained by anti-DCP1A antibody and analyzed. D. Immunostaining of ELAVL2 in newborn ovaries (n = 3 animals for each genotype). E. PAS staining of P7
Referência(s)