|Home | About | Journals | Submit | Contact Us | Français|
In nearly every vertebrate species, elevated intracellular cAMP maintains oocytes in prophase I of meiosis. Prior to ovulation, gonadotropins trigger various intra-ovarian processes, including the breakdown of gap junctions, the activation of EGF receptors, and the secretion of steroids. These events in turn decrease intracellular cAMP levels in select oocytes to allow meiotic progression, or maturation, to resume. Studies suggest that cAMP levels are kept elevated in resting oocytes by constitutive G protein signaling, and that the drop in intracellular cAMP that accompanies maturation is due to attenuation of this inhibitory G protein-mediated signaling. Interestingly, one of these G protein regulators of meiotic arrest is the Gαs protein, which stimulates adenylyl cyclase to raise intracellular cAMP in two important animal models of oocyte development: Xenopus leavis frogs and mice. In addition to Gαs, constitutive Gβ γ activity similarly stimulates adenylyl cyclase to raise cAMP and prevent maturation in Xenopus oocytes; however, the role of Gβ γ in regulating meiosis in mouse oocytes has not been examined. Here we show that Gβγ does not contribute to the maintenance of murine oocyte meiotic arrest. In fact, contrary to observations in frog oocytes, Gβγ signaling in mouse oocytes reduces cAMP and promotes oocyte maturation, suggesting that Gβγ might in fact play a positive role in promoting oocyte maturation. These observations emphasize that, while many general concepts and components of meiotic regulation are conserved from frogs to mice, specific differences exist that may lead to important insights regarding ovarian development in vertebrates.
Steroids trigger many transcription-independent, or nongenomic, effects by regulating G protein signaling [1, 2]. These G protein-coupled signals occur in a wide variety of cell types, including endothelial cells, neurons, and breast epithelial cells [3–5]. One fascinating and biologically relevant model of steroid-mediated G protein signaling is steroid-triggered oocyte maturation, or meiotic progression, in Xenopus leavis oocytes [6, 7]. In female Xenopus frogs, oocytes within the ovary are held in meiotic arrest at prophase I. Just prior to ovulation, gonadotropins stimulate ovarian steroid production in an unusual mechanism that appears to involve both follicular cells and germ cells . These newly synthesized steroids then trigger oocytes to re-enter the cell cycle and progress through to metaphase II, at which point the mature oocytes are ready for ovulation and subsequent fertilization.
Interestingly, evidence from many laboratories suggests that steroid-triggered oocyte maturation of Xenopus laevis oocytes occurs via a “release of inhibition“ model whereby oocytes are held in meiotic arrest by constitutive G protein signaling. Steroids then attenuate this inhibitory G protein signaling to promote maturation. The G proteins regulating the inhibitory signal include Gαs and Gβγ, both of which appear to act together to stimulate adenylyl cyclase and increase intracellular cAMP. Evidence supporting this model is plentiful, including the following: First, steroids trigger a rapid decrease in intracellular cAMP with a concomitant decrease in PKA activity [9–12]. Second, over-expression of either Gαs or Gβγ inhibits steroid-triggered oocyte maturation, while reduction of Gαs or Gβγ levels or activity leads to enhanced maturation in response to steroids [13–16]. Third, stimulation of over-expressed Gαs- or Gβγ-coupled receptors markedly inhibits steroid-triggered oocyte maturation [17, 18]. Finally, by using over-expressed Gβγ-regulated inward rectifying potassium channels (GIRKs) as markers of endogenous Gβγ signaling in oocytes, constitutive Gβγ signaling is detected that is reduced within minutes after steroid stimulation (Hammes, in press).
What are the physiologic steroids that mediate Xenopus leavis oocyte maturation in ovulating frogs? While multiple steroids are equally potent promoters of oocyte maturation in vitro, including progesterone, testosterone, and androstenedione [19–21], hCG stimulation of ovaries both in vitro and in vivo demonstrates that the androgens androstenedione and testosterone are produced at 10-fold higher amounts than progesterone [19, 20]. Furthermore, inhibition of androgen production downstream of progesterone in the steroidogenic pathway markedly inhibits both hCG-stimulated oocyte maturation and ovulation in live female frogs .
How do androgens suppress G protein signaling in frog oocytes? To start, androgens appear to activate extra-nuclear classical androgen receptors (ARs), as pharmacologic blockade of steroid binding to the AR, or reduction of AR expression by RNA interference, results in decreased steroid-triggered Gβγ signaling (Hammes, in press) and subsequent oocyte maturation [18, 20, 22]. While the detailed mechanisms are still not well-understood, androgen-activated ARs appear to attenuate G protein signaling through protein-protein complexes that may involve the scaffold molecule called the Modulator of Nongenomic Actions of steroid Receptors (MNAR) [23, 24].
Similar to Xenopus oocytes, mouse oocytes are also held in meiotic arrest prior to ovulation [25, 26]. In addition, under some conditions, steroids appear to be capable of promoting mouse oocyte maturation [27, 28]. However, unlike frog oocytes, mouse oocytes do not require steroids to re-enter meiosis, as they mature spontaneously upon removal from follicles, and inhibition of steroidogenesis in mouse follicles does not appear to block gonadotropin-mediated maturation . These observations indicate that the inhibitory signals maintaining meiotic arrest in mouse oocytes are derived at least in part from the surrounding follicle cells rather than being endogenous to the oocytes themselves. Furthermore, these results suggest, while steroids are the primary regulators of oocyte maturation in frogs, they are likely one of many signals capable of triggering mouse oocyte maturation.
Despite differences in signals outside the germ cell that regulate meiosis in mouse versus frog oocytes, intracellular constitutive G protein signaling is a shared mechanism for maintaining meiotic arrest in both species. For example, as in frog oocytes, meiotic arrest in mouse oocytes is mediated by constitutive Gαs activity, as injection of antibodies against Gαs into follicle-enclosed oocytes enhances oocyte maturation . Intriguingly, the constitutive inhibitory Gαs activity in mice may be mediated at least in part through a constitutively active G protein-coupled receptor called GPR3, as over-expression of GPR blocks oocyte maturation, while knockout mice lacking GPR3 have a higher incidence of spontaneous oocyte maturation . Similarly, over-expression of mammalian GPR3 in frog oocytes inhibits steroid-triggered maturation, suggesting that a Xenopus isoform of GPR3 may also be regulating constitutive G protein signaling in Xenopus oocytes .
While the importance of Gαs in maintaining meiotic arrest is well established in mouse oocytes, the role of Gβγ in regulating meiosis has not been examined. Here we studied the effects of modulating Gβγ signaling on mouse oocyte maturation. We demonstrated that, in stark contrast to frog oocytes, Gβγ signaling in mouse oocytes decreased intracellular cAMP levels, and actually promoted maturation of follicle-enclosed oocytes. These studies underscore the concept that, while many of the general principals of meiotic regulation are conserved in vertebrates, species-specific differences exist that must be recognized.
Most of the plasmids used for this study are described elsewhere . The cDNA encoding transducin in the pFROG vector was a gift from S. Coughlin (UC San Francisco). The cDNAs encoding the bovine Gβ1 in pGEM-HE, Gγ2 in pFROG, and the G protein receptor kinase minigene (GRK) in pGEM-HE were gifts from L. Jan (UC San Francisco). The wild-type and constitutively active bovine Gαs cDNAs were gifts from S. Mumby (UT Southwestern), and were cloned into the pGEM-HE vector. The anti-Gαs antibody was also a gift from S. Mumby. The muscarinic-2 receptor (M2R) in pcDNA3.1 was purchased from UMR cDNA Resource Center (Rolla, MO), and the cDNA was cloned into the pGEM-HE vector. Finally, CL57BL/6J mice were purchased from Jackson laboratories (Bar Harbor, ME).
Plasmids containing the indicated cDNAs were linearized and transcribed in vitro using either SP6 or T7 as described . The purified and precipitated cRNA was re-suspended in 10 mM Hepes, pH 7.4, and quantified by spectrophotometry. Each cRNA was diluted to a final concentration of 200 ng/μl prior to injection.
Follicles were isolated from twenty-one day old C57BL/6J mice by puncturing isolated ovaries with 30 gauge needles. Intact follicles greater than 400 microns were collected and washed two times in M-2 media (Specialty Media, Phillipsburg, NJ). Using a Nikon PLI-188 microinjector and sterile injecting pipets (Humagen, Charlottesville, VA.), 10–15 picoliters of mRNA was microinjected into the cytoplasm of the follicle-enclosed oocyte in M-2 medium warmed to 37° C. Following injections, follicles were transferred to M-16 medium (Speciality Media) for overnight incubation at 37° C in 5% CO2. The next morning, follicles were punctured carefully to release the denuded oocytes, and meiotic progression was scored every 30 minutes for 4–5 hours using an inverted Nikon SMZ-1000 microscope. The zero time point was defined as the moment of follicle puncture. Oocyte maturation was scored by monitoring Germinal Vesicle Breakdown (GVBD). Oocytes showing clear nuclear membranes and nucleoli were classified as GV stage; those without visible nuclear structure were classified as GVBD.
For the frog oocyte injection, stage V - VI Xenopus laevis oocytes were isolated from frog ovaries, denuded by incubation with collagenase, and injected with 50 nanoliters of the indicated cRNAs (200 ng/μl) as described .
Injected oocytes from mouse follicles were gently denuded. Those expressing the M2R were treated for 10 minutes with or without 30 μM carbachol, followed by lysis in 100 μl of O.1M HCl/1% Triton-X-100 (20 oocytes for each point). Oocytes expressing all other proteins were denuded and put directly in lysis buffer. All samples were then centrifuged to remove debris, and cAMP levels in the supernatants were measured in duplicate using an acetylated cAMP competitive ELISA kit (Endogen, Rockford, IL) per the manufacturer's protocol.
For Xenopus laevis, five oocytes were lysed in 500 μl of the aforementioned buffer. Samples were then diluted 1:10 and analyzed using the acetylated cAMP assay.
For each condition, twenty injected mouse oocytes were permeabilized in 20 μl of 2x SDS-sample buffer [18, 23]. The entire lysate was then loaded in an individual lane for SDS-PAGE, transferred to Immobilon-P membranes (Millipore, Billerica, MA), blocked in 5% TBST-milk for 1 hour, then incubated with our anti-Gαs at 1:10,000 overnight at 4°C. Membranes were then washed, incubated with HRP-conjugated secondary antibody (BioRad, Hercules, CA) for 1 hour, and signal detected by ECL-Plus (Amersham, Piscataway, NJ)
As mentioned in the introduction, constitutive Gαs-mediated stimulation of adenylyl cyclase plays a critical role in maintaining meiotic arrest in both Xenopus leavis and in mice. In addition, constitutive Gβγ-mediated signaling is important for maintaining meiotic arrest in Xenopus oocytes, perhaps by working in conjunction with Gαs to enhance adenylyl cyclase activity. To determine whether constitutive Gβγ signaling is also important for maintaining meiotic arrest in mice, follicle-enclosed mouse oocytes were injected with cRNAs encoding either transducin or the carboxyl tail of the G protein-coupled receptor kinase 1 (GRK-1), both of which are known to sequester Gβγ heterodimers and reduce Gβγ signaling . Unlike in Xenopus oocytes, where injection of cRNAs generated from these same plasmids resulted in enhanced steroid-triggered oocyte maturation , expression of transducin or the carboxyl GRK protein in mouse oocytes had no effect on the rate or magnitude of spontaneous GVBD once oocytes were removed from follicles (Fig. 1A and 1B). These results indicate that, unlike in frogs, constitutive Gβγ signaling is not likely to be involved in maintaining meiotic arrest in mouse oocytes.
The observation that constitutive Gβγ signaling was not involved in maintaining mouse oocytes in meiotic arrest suggested that either constitutive Gβγ signaling simply does not occur in mouse oocytes, or that constitutive Gβγ signaling is present, but does not enhance adenylyl cyclase activity to increase cAMP and maintain meiotic arrest. To differentiate between these two possibilities, follicle-enclosed oocytes were injected with cRNAs encoding both Gβ and Gγ. Notably, when Gβγ is over-expressed in Xenopus oocytes, steroid-triggered oocyte maturation is markedly reduced . In contrast, over-expression of Gβγ in follicle-enclosed mouse oocytes resulted in 70% spontaneous GVBD, even while the oocytes were still enclosed in follicles (Fig. 2A). As a comparison, the rate of spontaneous GVBD of mock-injected oocytes measured directly after removal from follicles was only 30%. Even more remarkably, intracellular cAMP in mouse oocytes over-expressing Gβγ was reduced by nearly 10-fold (Fig. 2B). This contrasted with over-expression of Gβγ in Xenopus leavis oocytes, which led to nearly a five-fold induction of cAMP (Fig. 2B). Together, these data demonstrate that Gβγ signaling has completely opposite effects on cAMP levels in mouse versus Xenopus leavis oocytes. Furthermore, since Gβγ signaling actually promotes oocyte maturation in mouse oocytes, these results confirm that little to no constitutive Gβγ signaling is likely occurring in immature oocytes.
To confirm that Gβγ signaling increases intracellular cAMP levels and promotes mouse oocyte maturation, follicle-enclosed oocytes were injected with cRNA encoding the Gβγ-coupled M2R receptor. Stimulation of the M2R receptor with its agonist carbachol should activate endogenous Gβγ signaling, thus eliminating concerns related to over-expression of exogenous G proteins. As mentioned, activation of over- expressed M2R in Xenopus leavis oocytes markedly inhibits steroid-triggered oocyte maturation, most likely due to Gβγ-mediated stimulation adenylyl cyclase . As anticipated, activation of the M2R had the opposite effect in mouse oocytes. Follicle- enclosed mouse oocytes were injected with cRNA encoding the M2R. Twenty-four hours after injection, oocytes were removed from follicles and treated with either saline or 30 μM carbachol, and GVBD was measured. Interestingly, denuded M2R-expressing mouse oocytes treated with carbachol consistently progressed through GVBD more rapidly than mock-treated cells (Fig. 3A). In addition, carbachol stimulation of mouse oocytes over-expressing the M2R decreased intracellular cAMP by approximately twofold (Fig. 3B). In contrast, stimulation of the M2R in Xenopus oocytes markedly increased intracellular cAMP by approximately six-fold.
Notably, stimulation of the M2R will also activate Gαi, which could then contribute to the observed decrease in intracellular cAMP and enhanced maturation associated with carbachol treatment. However, pertussis toxin does not affect LH-induced oocyte maturation of mouse oocytes , nor does over-expression of either wild type or constitutively active Gαi (data not shown); thus, as in frog oocytes , Gαi-mediated signaling is not likely a major regulator of meiosis in mouse oocytes.
Together, these studies using Gβγ over-expression and M2R stimulation confirm that Gβγ functions very differently in mouse versus frog oocytes. In frogs, evidence suggests that Gβγ directly activates adenylyl cyclase activity to increase intracellular cAMP . In fact, certain isoforms of adenylyl cyclase, including ACII and ACVII, are activated in concert by both Gβγ and Gαs , and ACVII has been identified in Xenopus laevis oocytes .
In contrast, we show here that Gβγ signaling in mouse oocytes decreases intracellular cAMP and promotes spontaneous maturation. These observations suggest that either mouse oocytes do not express ACII or ACVII, or that these cyclases cannot be activated by Gβγ in these cells. Furthermore, the ability of Gβγ to reduce intracellular cAMP indicates that Gβγ and Gαs signaling are directly antagonistic in mouse oocytes.
To determine whether Gβγ can block Gαs-mediated stimulation of adenylyl cyclase and inhibition of maturation, follicle-enclosed mouse oocytes were injected with cRNA encoding Gαs with or without Gβγ. As expected, over-expression of Gαs significantly inhibited spontaneous germinal vesicle breakdown after oocytes were removed from follicles (Fig. 4A). Furthermore, intracellular cAMP was increased in these Gαs-injected oocytes relative to mock-injected oocytes (Fig. 4A). Interestingly, over-expression of Gβγ rescued the inhibitory effects of Gαs on oocyte maturation, as well as the stimulatory effect on cAMP production. Notably, co-expression of Gβγ with Gαs did not alter Gαs expression levels, as determined by Western blot (Fig. 4B). These results suggest that the over-expressed Gβγ might directly bind to over-expressed Gαs to prevent activation of adenylyl cyclase
To determine whether Gβγ was in fact suppressing Gαs activity through binding and sequestration, similar studies were performed using a constitutively activated form of Gαs, which presumably binds poorly to Gβγ. As expected, over-expression of constitutively activated Gαs significantly raised intracellular cAMP (10-fold) and inhibited spontaneous oocyte maturation relative to mock-injected oocytes (Fig. 4C). Co-expression of Gβγ with constitutively activated Gαs did not increase the rate or extent of oocyte maturation relative to oocytes only expressing constitutively activated Gαs, most likely due to continued elevation of intracellular cAMP that was still four-fold higher than in mock-injected oocytes (Fig. 4C). However, co-expression of Gβγ still reduced constitutively activated Gαs-stimulated cAMP by approximately three-fold (Fig. 4C). Since Gβγ presumably binds poorly to constitutively activated Gαs, this result suggests that some of Gβγ ‘s ability to lower intracellular cAMP may be mediated independent of Gαs sequestration. Other possibilities include direct inhibition of adenylyl cyclase or activation of phosphodiesterases by Gβγ. Further biochemical studies will be necessary to definitively differentiate between these possibilities; however, it is intriguing to postulate that activation of Gβγ signaling in mouse oocytes, perhaps via a Gβγ-coupled receptor, may contribute to the decrease in intracellular cAMP that regulates maturation.
These studies and others highlight the many differences between G protein-mediated signaling in oocytes from different vertebrate species. For example, in fish, oocyte maturation is pertussis toxin sensitive, and Gαi appears to be the primary G protein triggering oocyte maturation. In contrast, Gαi does not appear to be playing a significant role in mediating oocyte maturation in frogs and mice [14, 33]. Instead, both Gαs and Gβγ play critical roles in maintaining meiotic arrest in frogs oocytes, while only Gαs seems to maintain meiotic arrest in mouse oocytes. In fact, we find that Gβγ signaling can actually lower intracellular cAMP and promote maturation in mouse oocytes, suggesting that Gβγ may be a positive regulator of meiosis in mice. Further studies using intact FSH-primed follicles will be necessary to address whether activation of Gβγ signaling is a true physiologic mediator of LH-induced mouse oocyte maturation.
We thank Lisa Halvorson and Bruce Carr at UT Southwestern for generously allowing us to use their injection equipment. S.R.H. is a W. W. Caruth, Jr. Endowed Scholar in Biomedical Research. Work from our laboratory was supported by the NIH (DK59913) and the March of Dimes (FY05-78).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.