Maternally recruited DCP1A and DCP2 contribute to messenger RNA degradation during oocyte maturation and genome activation in mouse

Abstract

The oocyte-to-zygote transition entails transforming a highly differentiated oocyte into totipotent blastomeres and represents one of the earliest obstacles that must be successfully hurdled for continued development. Degradation of maternal mRNAs, which likely lies at the heart of this transition, is characterized by a transition from mRNA stability to instability during oocyte maturation. Although phosphorylation of the oocyte-specific RNA-binding protein MSY2 during maturation is implicated in making maternal mRNAs more susceptible to degradation, mechanisms underlying mRNA degradation during oocyte maturation remain poorly understood. We report that DCP1A and DCP2, proteins responsible for decapping mRNA, are encoded by maternal mRNAs recruited for translation during maturation via cytoplasmic polyadenylation elements located in their 3' untranslated regions. Both DCP1A and DCP2 are phosphorylated during maturation, with CDC2A being the kinase likely responsible for both, although MAPK may be involved in DCP1A phosphorylation. Inhibiting accumulation of DCP1A and DCP2 by RNA interference or morpholinos decreases not only degradation of mRNAs during meiotic maturation but also transcription of the zygotic genome. The results indicate that maternally recruited DCP1A and DCP2 are critical players in the transition from mRNA stability to instability during meiotic maturation and that proper maternal mRNA degradation must be successful to execute the oocyte-to-zygote transition.

FIG. 1

Maturation-associated increase of DCP1A and DCP2 protein and relative transcript abundance. A) Immunofluorescent analysis of decapping complex components DCP1A and DCP2 in full-grown GV oocytes (GV), MI eggs, and MII eggs. Bar = 25 μm. B) Immunoblot analysis of decapping complex components DCP1A and DCP2 during oocyte maturation and early embryonic development. The blot was stripped and reprobed with TUBB antibody as a loading control. The experiment was conducted two times, and similar results were obtained in each case. C) Temporal expression profile of Dcp1a and Dcp2 transcripts during oocyte maturation and preimplantation embryogenesis. Relative abundance of Dcp1a and Dcp2 transcripts was determined by qPCR. 1C, 1-cell embryo; 2C, 2-cell embryo; 8C, 8-cell embryo. The experiment was performed three times, and data are expressed as the mean ± SEM.

FIG. 2

Maturation-associated phosphorylation of DCP1A and DCP2. A) Immunoblot analysis of changes in electrophoretic mobility of DCP1A and DCP2 during oocyte maturation and early embryonic development. For DCP1A, 50 oocytes/eggs/embryos were loaded for each lane, and the blot was stripped and reprobed with TUBB antibody as a loading control. For DCP2, Flag-DCP2 was analyzed instead of endogenous DCP2, because large numbers of oocytes/eggs (n = 400) were required to detect endogenous DCP2. Fifty oocytes/eggs/embryos were loaded for each lane, and the blot was stripped and reprobed with TUBB antibody as a loading control. The experiment was conducted three times, and representative immunoblots are shown. GV, GV-intact oocyte; MI, MI egg; MII, MII egg; 1C, 1-cell embryo; 2C, 2-cell embryo. B) Phosphatase treatment of DCP1A and Flag-DCP2 in MII eggs. An MII egg extract was incubated with lambda protein phosphatase (+) before probing for DCP1A or Flag-DCP2, respectively. As a control, an MII egg lysate without phosphatase treatment (−) was run in parallel. TUBB served as loading control. The experiment was conducted two times, and representative immunoblots are shown. C) Time course of DCP1A and DCP2 phosphorylation during oocyte maturation. GV oocytes were injected with either Flag-Dcp1a or Flag-Dcp2 cRNA, then cultured overnight before allowing them to mature in vitro for the indicated times, at which point oocytes were removed for either immunoblot or kinase assays. CDC2A and MAPK activities were measured using histone H1 and myelin basic protein (MBP), respectively. The experiment was conducted two times, and representative immunoblots are shown. TUBB or ACTB served as loading controls. D) Effect of CDC2A and MAPK inhibitors on phosphorylation of DCP1A and DCP2. GV oocytes were injected with either Flag-Dcp1a or Flag-Dcp2 cRNA, allowed to in vitro mature for 16 h, then transferred to CZB medium with roscovitine or U0126 for additional 6 h before harvesting for immunoblot analysis. TUBB or ACTB served as loading controls. In parallel, dual kinase assays were performed to demonstrate CDC2A and MAPK activities were inhibited by adding the inhibitor or inhibitors. The decrease in MAPK activity following addition of roscovitine is a consequence of the decrease in CDC2A activity that is required to maintain elevated MAPK activity.

FIG. 3

Functional analysis of 3′ UTR of Dcp1a and Dcp2 mRNA. A) Schematic representation of Dcp1a and Dcp2 3′ UTR and luciferase reporter construct (top). Luc reporter cRNAs with the terminal 0.5 kb of Dcp1a or Dcp2 3′ UTRs were injected into GV oocytes. Following maturation, luciferase activity was analyzed in individual eggs. Injected oocytes cultured in milrinone-containing medium to inhibit maturation served as controls. The experiment was performed three times, and data are expressed as the mean ± SEM. B) Schematic of the Luc reporter cRNA used to identify functional CPEs in Dcp1a 3′ UTR. Three putative CPEs and one HEX were mutated. C) Schematic of the Luc reporter cRNA used to identify functional CPEs in Dcp2 3′ UTR. Two putative CPEs and one HEX were mutated. For these experiments, firefly luciferase reporter activities were normalized to a coinjected Renilla luciferase control and are expressed relative to the activity in oocytes injected with the same nonmutated reporter cRNA. The experiment was conducted three times, and data are presented as the mean ± SEM.

FIG. 4

Luc cRNA is less stable in MII eggs than in GV-intact oocytes. GV-intact oocytes or MII eggs were injected with Luc cRNA, and one portion of each sample was immediately frozen (t = 0). The remaining portion of each group was then cultured for 3 or 6 h, at which times a portion was removed for qPCR; GV-intact oocytes were cultured in milrinone-containing medium to inhibit maturation. Data are expressed relative to the amount of Luc cRNA present at t = 0 and represent the mean ± range (n = 2). GV-intact oocytes, solid diamonds; MII-arrested eggs, solid circles.

FIG. 5

Effects of Dcp1a and Dcp2 knockdown on maternal mRNA stability. A) The effect of siRNA treatment on inhibiting the maturation-associated increase in DCP1A and DCP2 as assessed by immunocytochemistry and immunoblotting. GV, control uninjected oocytes; MII egg, oocytes injected with scrambled siRNA or with Dcp1a and Dcp2 targeting siRNAs and matured to MII. Bar = 25 μm. B) Analysis by qPCR of selected transcripts to assess effect of Dcp1a and Dcp2 knockdown. GV oocytes were microinjected with siRNAs targeting Dcp1a and Dcp2, cultured for 20 h in milrinone-containing medium, and then transferred to milrinone-free medium to initiate maturation; MII eggs were then collected for qPCR analysis. Transcript abundance is expressed relative to that of uninjected GV-intact oocytes. The experiment was performed three times, and data are expressed as the mean ± SEM. Gray bars, GV-intact oocytes; open bars, oocytes injected with scrambled siRNAs and matured to MII; black bars, oocytes injected with siRNAs targeting Dcp1a and Dcp2 mRNAs and matured to MII. *P < 0.05, **P < 0.01.

FIG. 6

A) Principal component analysis of microarray profiles of MII eggs depleted of Dcp1a and Dcp2 transcripts. Each color-coded circle represents a projection of a complete microarray data set in a two-dimensional space formed by the top two principal components. The amount of variation covered by the top two principal components is indicated along each corresponding axis. Double-knockdown samples cluster apart from the controls, suggesting that knockdown of both Dcp1a and Dcp2 transcripts cause significantly distinct transcriptome changes. B) MII egg transcriptome changes upon Dcp1a and Dcp2 knockdown. The x-axis shows the average probe set hybridization signal, and the y-axis shows the relative expression change upon inhibition of decapping. Probe sets for which the relative signal was significantly increased or decreased (>1.5-fold; FDR < 0.05) in MII eggs following knockdown of Dcp1a and Dcp2 are shown as red and blue points, respectively. C) Extensive Dcp1a and Dcp2-dependent degradation of maternal mRNAs during maturation. The graph shows the relationship between transcriptome changes during oocyte maturation (shown on the y-axis) and changes following Dcp1a and Dcp2 double-knockdown (shown on the x-axis). Each point displays relative behavior of one Affymetrix probe set. Microarray data for oocyte maturation were taken from the literature [18]. The contribution of DCP1A and DCP2 to the maturation-associated mRNA degradation is clearly visible in the lower right quadrant. Transcripts for which degradation is independent of DCP1A and DCP2 are in the lower left quadrant. Probe sets for which the relative signal was increased or decreased in MII eggs following knockdown of Dcp1a and Dcp2 (see also B) are shown as red and blue points, respectively.

FIG. 7

Effect of inhibiting maturation-associated increase in DCP1A and DCP2 on genome activation. A) Increased levels of maternal mRNAs Ppil3 and Upp1 following inhibition of the maturation-associated increase in DCP1A and DCP2. The qPCR experiment was performed two times, and data are presented as the mean ± range. B) Global transcription in 2-cell embryos following inhibiting the maturation-associated increase in DCP1A and DCP2 using morpholinos. Representative staining images of control and experimental 2-cell embryos and quantification of the data are shown. The experiment was performed four times. The difference between the two groups is significant (P < 0.001). Bar = 25 μm. C) H3K4me3 methylation is reduced following inhibition of the maturation-associated increase in DCP1A and DCP2. Representative staining images of control and experimental 2-cell embryos and quantification of the data are shown. The experiment was performed two times. The difference between the two groups is significant (P < 0.001). Bar = 25 μm.