Vertebrate oocytes arrest at prophase of meiosis I as a result of high levels of cyclic adenosine monophosphate (cAMP) and protein kinase A (PKA) activity. In Xenopus, progesterone is believed to release meiotic arrest by inhibiting adenylate cyclase, lowering cAMP levels and repressing PKA. However, the exact timing and extent of the cAMP decrease is unclear, with conflicting reports in the literature. Using various in vivo reporters for cAMP and PKA at the single-cell level in real time, we fail to detect any significant changes in cAMP or PKA in response to progesterone. More interestingly, there was no correlation between the levels of PKA inhibition and the release of meiotic arrest. Furthermore, we devised conditions whereby meiotic arrest could be released in the presence of sustained high levels of cAMP. Consistently, lowering endogenous cAMP levels by >65% for prolonged time periods failed to induce spontaneous maturation. These results argue that the release of oocyte meiotic arrest in Xenopus is independent of a reduction in either cAMP levels or PKA activity, but rather proceeds through a parallel cAMP/PKA-independent pathway.
Full-grown vertebrate oocytes arrest in a G2-like state at prophase of meiosis I for prolonged periods of time, during which the oocyte grows and stores the macromolecular components needed for future development (Smith, 1989; Voronina and Wessel, 2003). Before ovulation, oocytes resume meiosis and arrest in metaphase of meiosis II, in a process termed ‘oocyte maturation’. Oocyte maturation encompasses drastic cellular remodeling, in line with meiosis progression, which prepares the egg for fertilization. This is characterized by the dissolution of the nuclear envelope [referred to as germinal vesicle breakdown (GVBD)], extrusion of the first polar body and chromosome condensation (Bement and Capco, 1990; Nader et al., 2013; Sadler and Maller, 1985; Smith, 1989; Voronina and Wessel, 2003).
Progesterone (P4) is the classical steroid used to mature Xenopus oocytes in vitro. Though a role for a classical nuclear steroid receptor cannot be ruled out (Bayaa et al., 2000; Tian et al., 2000), P4-induced maturation is thought to be achieved primarily through a non-classical plasma membrane P4 receptor (mPRβ) (Josefsberg Ben-Yehoshua et al., 2007). It is generally accepted that, upon hormone binding, mPRβ inhibits adenylate cyclase (AC) causing a transient dip in cyclic adenosine monophosphate (cAMP) levels, which is believed to act as the trigger to release oocyte meiotic arrest by lowering protein kinase A (PKA) activity (Cicirelli and Smith, 1985; Maller et al., 1979).
There is broad consensus that meiotic arrest in vertebrate oocytes is maintained by high levels of cAMP and PKA, with the most compelling evidence coming from studies of the constitutively active orphan G protein-coupled receptor (GPCR) GPR3 in mouse oocytes (Freudzon et al., 2005; Mehlmann et al., 2004). Knockout of Gpr3 leads to premature oocyte maturation in antral follicles and sterility. In Xenopus oocytes, injection of the catalytic subunit of PKA (PKAc) inhibits P4-induced oocyte maturation (Daar et al., 1993; Eyers et al., 2005; Matten et al., 1994). Interestingly, however, even injection of a catalytically dead PKA mutant inhibits maturation (Eyers et al., 2005; Schmitt and Nebreda, 2002), although it has been argued that this is due to residual PKA activity (Eyers et al., 2005). Furthermore, phosphodiesterase (PDE) inhibitors (Andersen et al., 1998; Han et al., 2006) and treatment with cholera toxin, which activates AC, repress maturation (Huchon et al., 1981; Mulner et al., 1979; Schorderet-Slatkine et al., 1978). Inhibition of AC accelerated P4-induced oocyte maturation (Sadler et al., 1986). Consistently, injection of PDE, the PKA regulatory subunit (PKAr), or the PKA pseudo-substrate inhibitor (PKI) all release meiotic arrest (Andersen et al., 1998; Daar et al., 1993; Eyers et al., 2005). However, others found that inhibition of PKA activity using H-89 or PKI injection failed to stimulate oocyte maturation (Noh and Han, 1998). Furthermore, induction of a drop in cAMP levels through stimulation of exogenously expressed Gi-coupled serotonin receptor did not stimulate oocyte maturation (Noh and Han, 1998).
Whereas some studies detect a 10-60% drop in cAMP levels from 15 s up to 6 h after P4 addition (Bravo et al., 1978; Cicirelli and Smith, 1985; Gelerstein et al., 1988; Maller et al., 1979; Mulner et al., 1979; Nader et al., 2014; Sadler and Maller, 1981; Schorderet-Slatkine et al., 1982), others failed to detect any changes in cAMP levels in response to P4 (Noh and Han, 1998; O'Connor and Smith, 1976; Schutter et al., 1975; Thibier et al., 1982). Furthermore, some studies found that a decrease in cAMP levels was not sufficient to induce spontaneous maturation, and that an increase in cAMP accelerates maturation (Gelerstein et al., 1988; Noh and Han, 1998). Consistently, exposing oocytes to P4 for <30 min induced a decrease in cAMP but did not induce oocyte maturation, which required ≥30 min incubation with P4 (Nader et al., 2014). Hence, there is strong support for the cAMP-PKA axis maintaining the prolonged oocyte prophase I meiotic arrest. However, the evidence supporting a drop of cAMP and PKA activity to release meiotic arrest is poor, with conflicting reports regarding the timing and extent of the cAMP decrease in response to P4.
Here, we use multiple approaches, including cAMP- and PKA-sensitive channels and FRET sensors, to follow cAMP and PKA activity in individual oocytes in real time, and could not detect any change during the induction of oocyte maturation. Unexpectedly, increasing or buffering cAMP levels in the oocytes did not correlate with the ability of P4 to release meiotic arrest. Based on these findings, we propose an alternative model in which P4 stimulates oocyte maturation independently of a decrease in cAMP and/or PKA.
Decrease in cAMP levels does not correlate with the release of meiotic arrest
We and others have previously shown a decrease in cAMP levels after addition of P4 using an ELISA-based assay in a batch of oocytes. However, the early cAMP drop was not sufficient to induce maturation, because exposure to P4 from 15 s to 10 min was not sufficient to mature oocytes; rather, oocytes required a minimum of 30 min exposure to P4 to induce significant oocyte maturation (Nader et al., 2014). However, oocytes might require an integrated sustained lower level of cAMP to inhibit PKA activity and relieve meiotic arrest. To explore this possibility, we used the ELISA assay to measure cAMP levels in batches of 10 oocytes at time points ranging from seconds (Fig. 1B, left panel), up to 60 min (Fig. 1B, right panel) after P4 addition. Briefly, oocytes were treated transiently with 10−5 M P4 or ethanol as the carrier control for the indicated time points after which 10 oocytes were collected for the cAMP-ELISA assay and the rest of the oocytes were washed twice with the culture media (L15) to remove P4, and allowed to mature overnight.
When compared with oocytes treated with ethanol, cAMP levels showed a significant drop at 5, 15 s, 1, 4, 6, 10, 12, 30 and 60 min after P4 (Fig. 1B,C). Interestingly, cAMP levels at 1 min and 30 min after P4 were not significantly different (P=0.5071; Fig. 1C); however, oocytes failed to mature after a 1 min exposure to P4, yet achieved full maturation after a 30 min exposure to P4 (Fig. 1D). These results suggest that the cAMP dip observed within 30 min of P4 treatment is not the trigger for maturation.
Membrane channels do not detect changes in cAMP or PKA in response to P4
An important limitation of the correlation between cAMP levels and maturation is that the analysis was performed on a batch of 10 oocytes, as typically reported in the literature. This is problematic because the early stages of oocyte maturation (before GVBD) are not synchronized at the single-oocyte level; rather, individual oocytes reach the GVBD stage with disparate time courses. Therefore, changes in cAMP could be more pronounced at the single-oocyte level but are averaged out in a cell population, which could account for the absence of a correlation between the cAMP levels and maturation.
We therefore devised approaches to follow cAMP and PKA levels in individual oocytes in real time after P4 treatment. For that purpose, we used two membrane channels as sensors for cAMP and its effector PKA: the cyclic nucleotide gated channel (CNG), a cAMP-gated Ca2+ channel; and the cystic fibrosis transmembrane regulator (CFTR), a PKA-activated Cl− channel (Biel and Michalakis, 2009; Hanrahan et al., 1996). Both channels have been expressed and well characterized in Xenopus oocytes (Bear et al., 1991; Nache et al., 2012; Weber et al., 2001; Young and Krougliak, 2004). A rise in cAMP gates CNG channels, which permeate both mono- and divalent cations, including Ca2+. Ca2+, in turn, stimulates endogenous Ca2+-activated chloride channels (CaCCs), thus amplifying the signal (Fig. 2A,B).
Expressing CNG in oocytes was associated with a large CaCC (Fig. 2B), indicating that the channel is gated at resting cAMP levels. To test the sensitivity of CNG to changes in cAMP, we inhibited AC with 2′,5′-dideoxyadenosine (DDA), to lower basal cAMP levels (Sadler and Maller, 1983). DDA treatment significantly (P=0.021) decreased basal CNG-CaCC current (ICNG-CaCC) (Fig. 2C), indicating that CNG effectively detects a drop in cAMP. However, when CNG-expressing oocytes were treated with P4, we failed to detect any significant differences in ICNG-CaCC, up to 30 min after treatment with P4 compared with ethanol (Fig. 2D). This suggests that basal cAMP levels, at least in the sub-plasma membrane domain detected by the CNG channel, were unchanged up to 30 min after P4 treatment. However, the CNG channel clone used is gated by both cGMP and cAMP (Young and Krougliak, 2004), so the lack of response could be because the effect was masked by cGMP in the oocyte.
To test for changes in PKA activity, we used the PKA-gated Cl− channel CFTR to follow PKA activity in real time in the oocyte. In order to test whether CFTR can detect changes in PKA activity, we injected the non-hydrolyzable cAMP analog db-cAMP to activate PKA, this showed a significant and reproducible, although short-lived, induction of the CFTR current (ICFTR) (Fig. S1A,B). Inhibition of PKA with PKI under these conditions, resulted in a significant decrease of ICFTR (P=0.015) (Fig. S1C,D). This shows that CFTR detects changes in PKA activity in real time in the oocyte.
ICFTR showed significant rundown within 1 min of establishing the voltage clamp (Fig. 2E), as previously reported (Button et al., 2001). To minimize this, we pre-treated oocytes with 3-isobutyl-1-methylxanthine (IBMX) (Ramu et al., 2007), a phosphodiesterase inhibitor, to raise cAMP and activate PKA (Sadler and Maller, 1987; Schorderet-Slatkine and Baulieu, 1982). IBMX pre-treatment significantly (P=0.0002) increased ICFTR, and limited its rundown to maintain a significant current for up to 25 min (Fig. 2F, lower panel). Although we cannot rule out the possibility that IBMX, by inhibiting PDE, might blunt changes in cAMP in response to P4, we failed to detect any significant changes in ICFTR, compared with the ethanol control, for up to 30 min after P4 treatment (Fig. 2G), in accordance with the CNG data. This argues that PKA activity, at least in the sub-plasma membrane compartment sampled by CFTR, does not change significantly in response to P4. Collectively, membrane-bound sensors did not detect significant changes in cAMP and PKA in response to P4.
Intracellular FRET sensors do not detect changes in cAMP or PKA in response to P4
Since we failed to detect changes in cAMP levels or PKA activity using membrane channels, we needed to rule out the possibility of P4 affecting cAMP or PKA levels in subcellular compartments that cannot be sampled by membrane channels. To do this, we expressed two widely used intracellular FRET-based sensors, TEPACVV and AKAR2, to monitor cAMP levels and PKA activity, respectively (Klarenbeek and Jalink, 2014; Zaccolo et al., 2000). TEPACVV employs the mTurquoise-YFP FRET pair, and has an improved dynamic range as a function of cAMP concentrations (Fig. 3A) (Klarenbeek et al., 2011), whereas the A kinase activity reporter (AKAR2) FRET sensor uses Clover-mRuby2, and has a phospho-amino acid binding domain, a defined docking site for recruiting PKA and a substrate domain, resulting in increased FRET as a function of PKA activity (Fig. 3D) (Lam et al., 2012; Ni et al., 2006). Confocal imaging using oocytes expressing TEPACVV and AKAR2 showed that both sensors are diffusely cytoplasmically distributed (Fig. 3A,D). We then calibrated both sensors in oocytes. To that end, TEPACVV-FRET efficiency and change in the AKAR2-mRuby/Clover ratio (ΔFRET%) were measured following injection with db-cAMP (Fig. 3B,E). Given that basal cAMP values in Xenopus oocytes averages ∼1.25 pmol/oocyte with a range of 0.7-2.7 pmol/oocyte (Maller et al., 1979; O'Connor and Smith, 1976; Schutter et al., 1975), we injected 0.2-0.8 pmol db-cAMP, which represents a 10-80% increase in cAMP levels. Although db-cAMP is a cell-permeable analog, we wanted to control the exact amounts of db-cAMP delivered to the oocyte; hence, we opted for injection rather than incubation.
The FRET efficiency of TEPACVV decreased significantly by ∼40% (P=0.0094) after injection of 0.8 pmol db-cAMP (Fig. 3B) and AKAR2-ΔFRET% increased significantly in a dose-dependent fashion after db-cAMP injection (0.2-0.8 pmol) (Fig. 3E).
However, when oocytes expressing either sensor were treated with P4 for up to 30 min, we were unable to detect any significant changes in cAMP or PKA levels compared with control ethanol-treated oocytes (Fig. 3C,F), although the expression of the FRET sensors did not affect the GVBD response (data not shown). These results support the data from the membrane-bound channel sensors, and argue that the release of meiotic arrest induced by P4 occurs independently of changes in cAMP or PKA.
PKA anchoring to AKAP is not essential for releasing meiotic arrest
Although we could not detect a repression of PKA using CFTR or AKAR2, we cannot rule out the possibility that changes in PKA activity localize to subcellular microdomains. To explore this possibility, we evaluated PKAc subcellular distribution by immunostaining (Fig. 4A-C). The specificity of the PKAc antibody was confirmed by pre-incubating the anti-PKA antibody with its peptide antigen (Fig. 4A, right panel). PKAc showed a diffuse cytoplasmic distribution with enrichment below the plasma membrane at the animal, but not vegetal, pole of the oocyte (Fig. 4A,B). Line-scan analyses confirm PKAc enrichment in the sub-plasma membrane domain on the animal but not the vegetal pole (Fig. 4C). To test whether sub-membrane PKA is specifically inhibited in response to P4, we prepared membrane and cytosolic fractions by ultracentrifugation before and 30 min after P4 treatment, and tested PKA kinase activity (Fig. 4). The efficiency of the biochemical separation of membrane and cytosolic fractions was confirmed by immunoblotting for the plasma membrane protein, Na+/K+ pump (Fig. 4D). We failed to detect any changes in PKA activity in either the cytosolic or membrane fractions 30 min after P4 treatment (Fig. 4E), or after normalizing for PKAc protein content in the two fractions (Fig. 4F).
To further test whether P4 treatment modulates PKA activity away from the sub-plasma membrane space, we tested the role of A kinase-anchoring proteins (AKAPs) in oocyte maturation. AKAPs are scaffold proteins that bind the PKA regulatory subunit (PKAr), and anchor it close to its physiological substrates (Colledge and Scott, 1999; Malbon, 2005; Malbon et al., 2004). PKA interaction with AKAPs was found to be involved in mammalian oocyte meiotic arrest (Brown et al., 2002; Kovo et al., 2006, 2002; Newhall et al., 2006; Nishimura et al., 2013). Injection of an AKAP inhibitory peptide that blocks the AKAP-PKAr interaction in mouse oocytes stimulates oocyte maturation in the presence of high cAMP levels (Newhall et al., 2006). We therefore tested the effect of blocking AKAP-PKA interactions on Xenopus oocyte maturation. Oocytes injected with the AKAP inhibitory peptide matured significantly (P=0.0349) faster compared with oocytes injected with the control peptide (Fig. 4G,H). However, interfering with AKAP-PKA interaction did not affect maximal GVBD levels achieved in response to P4 (Fig. 4H). These results argue that PKA anchoring modulates the rate of P4-induced oocyte maturation without affecting its extent or promoting maturation independently of P4.
PKA activity does not change following the release of meiotic arrest
To further monitor total PKA activity at the single-oocyte level we used a quantitative enzymatic PKA assay (Fig. 5A). We first validated the sensitivity of the assay by injecting individual oocytes with the PKA inhibitor PKI or with PKAc (Fig. 5A). Control oocytes show a basal level of PKA activity at 37±4% phosphorylation of the substrate peptide (Fig. 5A, bottom panel). PKI injection led to a significant (P=0.0274) decrease in percent phosphorylated peptide to 16±1% (Fig. 5A). Consistently, PKAc injection increased the percentage of phosphorylated peptide to 89±3%, (P<0.0001) (Fig. 5A). We then checked the effect of P4 treatment on PKA activity in single oocytes at 15 s, 1, 2, 12 and 30 min and could not detect any change (Fig. 5A).
Concomitant with the PKA activity assay, we tested the effect of PKI and PKAc on oocyte maturation. This was done at sub-threshold (10−8 M) and saturating (10−5 M) P4 concentrations (Fig. 5B). Although PKI injection stimulates oocyte maturation, it was significantly less effective than 10−5 M P4 (Fig. 5B). However, PKI significantly (P=0.0044) enhanced the rate of P4-induced oocyte maturation (Fig. 5C,D), as previously reported (Noh and Han, 1998). The poor activation of oocyte maturation by PKI (Fig. 5B,C) is surprising, especially given the robust repression of PKA activity in the same batch of oocytes (Fig. 5A). These data show a lack of correlation between PKA activity and the ability of PKI and P4 to release meiotic arrest and induce oocyte maturation, hence bringing into question the accepted view that the release of meiotic arrest is mediated by a drop in PKA activity. It is worth noting here that previous studies using PKI to induce maturation in Xenopus oocytes, including a study from our lab, reported full maturation when compared with P4 treatment (Maller and Krebs, 1977; Sun and Machaca, 2004). By contrast, microinjection of PKI did not cause oocyte maturation, but accelerated GVBD in the presence of P4 in another study (Noh and Han, 1998). The release of meiotic arrest is an asynchronous process in oocytes from the same female and shows significant variability among different females based on whether they are lab bred, or captured in the wild and whether the females were hormonally stimulated to test for their ability to produce oocytes. All these factors could affect the effectiveness of various stimuli to induce oocyte maturation. However, notwithstanding this inherent variability, our results show a lack of correlation between the effectiveness of PKI or P4 in inducing oocyte maturation and their ability to inhibit PKA activity in the same batch of oocytes. PKI induces a robust inhibition of PKA activity (by 42.5%) with poor maximal maturation rates (33±7.3%). By contrast, P4 treatment did not induce any detectable change in PKA activity, yet fully induced oocyte maturation in the population (75±7.8%). Note that even P4 in this batch of oocytes did not fully stimulate maturation, consistent with the poor maturation levels with PKI.
Finally, PKAc injection abolished P4-induced maturation (Fig. 5B), as previously reported (Daar et al., 1993; Matten et al., 1994), indicating that high PKA activity is sufficient to block signals emanating from P4 stimulation that lead to oocyte maturation. Collectively, our data argue that the cAMP-PKA axis maintains oocyte meiotic arrest at steady state, but that it is not modulated downstream of P4 to release meiotic arrest, suggesting a parallel pathway that actively releases meiotic arrest despite constant high levels of cAMP and PKA.
Increasing cAMP levels do not affect oocyte maturation
The above data suggest that the cAMP-PKA axis acts as the ‘brake’ that maintains meiotic arrest, but that the release of this meiotic arrest does not involve changes in cAMP and PKA levels. To further assess this model, we tested the effects of db-cAMP on the extent and kinetics of oocyte maturation. We used CNG-expressing oocytes to determine the level of endogenous cAMP in the oocyte and the effectiveness of db-cAMP. Young and Krougliak have previously characterized in Xenopus oocytes the cAMP dose dependence of the recombinant CNG channel, x-fA4, that we used here (Young and Krougliak, 2004). They showed a steep dose dependence of ICNG on cAMP concentrations with a K1/2 of 1 μM and a saturating current above 200 μM cAMP (see figure 2C in Young and Krougliak, 2004). Consistently injecting oocytes with 0.8 or 400 pmol db-cAMP (which corresponds to 400 μM final cAMP concentration in the oocyte) dose-dependently increased ICNG-CaCC (Fig. 6A). Assuming a linear response of the CaCC to Ca2+ flowing through CNG channels, which is reasonable given the large density of CaCCs in the oocyte and their Ca2+ dependence (Kuruma and Hartzell, 1998), we could translate ICNG-CaCC into cAMP concentration in the oocyte using the Young and Krougliak dose-response curve. We used the current induced by 400 pmol db-cAMP as the maximal current (Imax). With an I/Imax of 0.3 in control oocytes, this translates to a cAMP concentration of 0.9 μM, within the range of 0.7-2.7 μM/oocyte obtained from direct measurements of cAMP by others (Maller et al., 1979; O'Connor and Smith, 1976; Schutter et al., 1975). This validates the use of I/Imax of ICNG-CaCC to estimate oocyte cAMP concentration. Furthermore, ICNG-CaCC recordings show that the injected db-cAMP is stable and results in a sustained increase in cAMP levels for extended time periods. Injection of 0.8 and 400 pmol db-cAMP dose-dependently and significantly repressed oocyte maturation stimulated with 10−7 M P4 (Fig. 6B). Surprisingly, however, when oocytes were stimulated with the saturating P4 concentration of 10−5 M for only 1 h, neither concentration of db-cAMP produced any inhibition of oocyte maturation (Fig. 6C). Although maximal GVBD levels were not reduced following injection of 400 pmol db-cAMP, the time course of maturation was slowed down (Fig. 6D), requiring 6±0.62 h to reach 50% GVBD compared with 3±0.62 h in control oocytes (P=0.0146) (Fig. 6E). These data indicate that increasing cAMP levels modulate the rate and extent of maturation. However, increasing the P4-dependent signal overcomes the cAMP-mediated inhibition and releases meiotic arrest independently of high sustained levels of cAMP, of ∼400-fold higher levels than resting cAMP concentrations.
Buffering endogenous cAMP does not induce spontaneous maturation
To further test our hypothesis that P4 induces oocyte maturation through an alternative pathway rather than reducing the levels of cAMP and PKA, we wanted to test whether a reduction of endogenous cAMP levels is sufficient to induce oocyte maturation independently of P4. The highest reported P4-dependent reduction in cAMP levels is 60% (Maller et al., 1979). The FRET sensor TEPACVV binds cAMP and as such, when expressed at high concentrations in the oocyte (Fig. 7A), would be expected to buffer endogenous cAMP. Expression of TEPACVV decreased ICNG-CaCC by more than 10-fold (P=0.018), resulting in a 66.7% reduction of cAMP levels from 0.9 μM to 0.3 μM (Fig. 7B). This reduction in cAMP is in line with the maximal 60% reduction reported in response to P4 (Maller et al., 1979). Further attesting to the effectiveness of TEPACVV to buffer cAMP, cells expressing TEPACVV attenuate the stimulation of ICNG-CaCC following db-cAMP injection by 3-fold (P=0.0413) (Fig. 7B). Remarkably however, endogenous cAMP buffering by TEPACVV did not release oocyte meiotic arrest, nor it did enhance maximal GVBD when suboptimal or optimal concentration of P4 were used (Fig. 7C). Additionally, it did not accelerate the rate of oocyte maturation (Fig. 7D). These data support the conclusion that a dip in cAMP is not the trigger to release Xenopus oocyte meiotic arrest.
The current accepted model of releasing meiotic arrest in frog oocytes postulates that a drop in cAMP leads to inhibition of PKA. Our results challenge this model and argue that P4 releases meiotic arrest through a cAMP/PKA-independent pathway. There is no correlation between cAMP levels and oocyte maturation (Fig. 1). Furthermore, we could not detect any P4-mediated changes in cAMP or PKA at the single-oocyte level using CNG/CFTR channels (Fig. 2), FRET sensors (Fig. 3) or a PKA kinase assay (Figs 4,5). Despite these results, we cannot rule out the possibility that the dip in cAMP and PKA is too small, too localized or too transient to be detected. However, several lines of evidence argue against the need for a drop in cAMP and PKA levels to release meiotic arrest. (1) Injection of PKI inhibits PKA activity by 42.5% but results in poor maturation levels (Fig. 5). This argues that inhibition of PKA activity is not sufficient to release meiotic arrest in the absence of a P4-dependent signal. As discussed in the Introduction, whether PKA inhibition is sufficient to release meiotic arrest is controversial and there are conflicting reports (Andersen et al., 1998; Daar et al., 1993; Eyers et al., 2005; Noh and Han, 1998). This could be due to the priming state of the donor female as discussed above. (2) Maintaining sustained high levels of cAMP inhibited oocyte maturation but this inhibition was reversed by increasing P4 concentration (Fig. 6). A complete block of maturation could be achieved only with injection of excess PKA catalytic subunit (Fig. 5). (3) Buffering cAMP by over 65% was insufficient to induce maturation (Fig. 7).
Collectively, these results argue that meiotic arrest is released through a positive signal downstream from the membrane P4 receptor (mPRβ, also known as PAQR8), while the cAMP-PKA pathway acts a ‘brake’ (see model in Fig. 8). It is the balance between the ‘positive’ P4-dependent signal and the ‘negative’ cAMP-PKA inhibitory signal that defines whether the oocyte commits to maturation. PKA inhibits the maturation signaling pathway through at least two points: mRNA translation and Cdc25C activation (Duckworth et al., 2002; Matten et al., 1994), consistent with the idea that it is acting as a safety mechanism to eliminate spontaneous maturation in the absence of a positive signal. Although challenging the accepted dogma, our model is supported by earlier studies (Gelerstein et al., 1988; Nader et al., 2014; Noh and Han, 1998; Schmitt and Nebreda, 2002) and the yin-yang balance between the mPRβ and cAMP pathways effectively explains the discrepancies in the literature.
cAMP and PKA levels are maintained at high levels for the duration of the meiotic arrest through the action of a constitutively active Gαs-coupled GPCR (Ríos-Cardona et al., 2008). GPR185 is the Xenopus homolog of mammalian GPR3, which has been shown to maintain meiotic arrest in mouse oocytes (Freudzon et al., 2005; Mehlmann et al., 2002, 2004; Norris et al., 2009). Surprisingly, knockdown of GPR185 is not sufficient to induce oocyte maturation (Deng et al., 2008; Ríos-Cardona et al., 2008), arguing that other GPCRs are involved in maintaining meiotic arrest. We have previously shown that (1) blocking exocytosis, while maintaining endocytosis, releases meiotic arrest, (2) P4 results in the internalization of GPR185 and (3) a GPR185 mutant that does not internalize is more effective at maintaining meiotic arrest (El-Jouni et al., 2007; Nader et al., 2014). These data argue that P4 induces internalization of GPR185 and potentially other GPCRs that are involved in maintaining high levels of cAMP and PKA and thus meiotic arrest. Consistent with the regulation of GPR185 in Xenopus, a drop in cAMP has been shown to correlate with the release of meiotic arrest in mouse oocytes, through a cGMP signal from the surrounding somatic cells (Norris et al., 2009). Furthermore, there is evidence in the mouse oocyte for a positive signal through an EGF-like pathway downstream of LH stimulation (Ashkenazi et al., 2005; Park et al., 2004), which might be the equivalent of the P4 signal in the frog.
mPRβ has seven-transmembrane domains and belongs to the progestin and adiponectin receptor family (PAQR) (Tang et al., 2005). It is still unclear whether mPRβ is a GPCR (Moussatche and Lyons, 2012; Thomas et al., 2007). For example, mPRβ does not work through Gαi to inhibit AC since pertussis toxin, a specific Gαi inhibitor, failed to block P4-induced maturation (Mulner et al., 1985; Olate et al., 1984; Sadler et al., 1984). This is consistent with the idea that mPRβ acts through other pathways to release the meiotic arrest. mPRs and AdipoQ receptors can functionally couple to the same signal transduction pathway in yeast, suggesting a common mechanism of action (Kupchak et al., 2007). Adiponectin receptors signal through multiple pathways, including the adaptor protein APPL1, which links to the trafficking GTPase Rab5 (Buechler et al., 2010; Mao et al., 2006) or p38 MAPK (MAPK14) (Heiker et al., 2010). Adiponectin receptors also possess alkaline ceramidase activity to generate sphingosine 1-phosphate (S1P) (Moussatche and Lyons, 2012). S1P is known to modulate activity of GPR3, the homolog of GPR185 in mammals (Uhlenbrock et al., 2002; Zhang et al., 2012) and ceramide is a potential mediator of P4-induced maturation in Xenopus oocytes (Strum et al., 1995).
In summary, our results challenge the generally accepted initial signaling pathway downstream of P4, which assumes a drop in cAMP and repression of PKA activity to release meiotic arrest. Rather, they argue for the existence of a positive signal downstream of mPRβ to overcome the negative inhibitory signal from cAMP and PKA to release meiotic arrest. As such, in the future it would be of great interest to better define the mPRβ signaling pathway.
MATERIALS AND METHODS
Human CFTR – a gift from John Riordan (University of North Carolina) – was amplified using primers, 5′-CTGCAGGAATTCGATATGCAGAGGTCGCCTCTGGAAAAGGCC3′ (F) and 5′-ATCGATAAGCTTGATCTAAAGCCTTGTATCTTGCACCTCTTCTTC-3′ (R), and subcloned in the Xenopus oocyte expression vector pSGEM. TEPACVV [mTurq2Del-EPAC(dDEPCD)Q270E-tdcp173Venus(d)EPAC-SH187] was a gift from Kees Jalink (Netherlands Cancer Institute) (Klarenbeek et al., 2011) and was subcloned into the NotI-XbaI sites of pSGEM. The CNG channel chimeric clone X-fA4 was a gift from Edgar Young (Simon Fraser University, Canada) (Young and Krougliak, 2004). AKAR2 (Lam et al., 2012) was purchased from Addgene and subcloned into the BamHI-EcoRI sites of pSGEM. All constructs were verified by DNA sequencing and by analytical endonuclease restriction digestion. mRNAs for all the clones were produced by in vitro transcription after linearizing the vectors with NheI (CFTR, AKAR2), NotI (CNG) or SphI (TEPACVV) using the mMessage mMachine T7 kit (Ambion).
Stage VI Xenopus oocytes were obtained as previously described (Machaca and Haun, 2002). The donor females were not hormonally stimulated prior to use. Animals were handled according to Weill Cornell Medicine College IACUC approved procedures (protocol #2011-0035). The oocytes were used 24-72 h after harvesting and digestion with collagenase to remove follicular cells surrounding the oocytes. To study the role of AKAPs, cells were treated with 100 µM AKAP St-Ht31 inhibitor peptide (Promega, V8211) or its control peptide (Promega, V8221). Oocytes were injected with RNA and kept at 18°C for 1-2 days after injection to allow for protein expression.
ELISA cAMP assay
We used the cAMP Complete Enzyme Immunometric Assay kit (Assay Designs, 900-163). Ten oocytes were lysed by forcing them through a pipette tip in 250 μl of ice-cold 95% ethanol. Extracts were centrifuged at 15,000 g for 15 min at 4°C. The supernatants were transferred to new tubes and dried under vacuum. The residue was dissolved and cAMP was measured according to the kit's protocol.
CNG and CFTR recording
Ionic currents were recorded using standard two-electrode voltage-clamp recording technique. Recording electrodes were filled with 3 M KCl and coupled to a Geneclamp 500B controlled with pClamp 10.5 (Axon instruments). The CNG currents were measured indirectly by monitoring the activation of endogenous CaCCs. Those chloride channels work as biological sensors and give a very precise and amplified indication of the sub-membrane Ca2+ concentration. The currents were recorded at a 0.1 Hz frequency using a previously described ‘triple-jump’ protocol that allows the measurement of Ca2+ influx (Courjaret and Machaca, 2014; Machaca and Hartzell, 1998). The CNG-CaCC current was measured as the difference in the amplitude of the currents at +40 mV before and after a voltage pulse to −140 mV that increases the driving force for Ca2+ entry (Fig. 2B). For CNG recordings using DDA and P4, the extracellular Ca2+ concentration was lowered to 0.9 mM to limit Ca2+ influx. The standard extracellular saline contained (in mM) 96 NaCl, 2.5 KCl, 1.8 CaCl2, 2 MgCl2, 10 HEPES, pH 7.4. CFTR currents were recorded at a steady-state membrane potential of −80 mV. For CNG currents normalization was done on the last current trace before treatment, for CFTR experiments normalization was performed using the average current over a 1 min period before treatment.
TEPACVV and AKAR2 FRET imaging
Confocal imaging of live cells was performed using a LSM710 (Zeiss, Germany) fitted with a Plan Apo 40×/1.3 oil immersion objective. z-stacks were taken in 0.45 µm sections using a 1 Airy unit pinhole aperture. When using TEPACVV, FRET efficiency was calculated using the FRET acceptor bleaching technique. Briefly, this was done by comparing donor fluorescence intensity in the same sample before and after photobleaching the acceptor. If FRET was initially present, a resultant increase in donor fluorescence will occur after photobleaching the acceptor, and the FRET efficiency can be calculated as follow FRET efficiency=(Dpost−Dpre)/Dpost where Dpost is the fluorescence intensity of the donor after acceptor photobleaching, and Dpre the fluorescence intensity of the donor before acceptor photobleaching. The animal pole of TEPACVV-expressing oocytes was imaged with the pinhole fully open. mTurquoise was excited at 458 nm and emission detected at 462-520 nm and YFP was photobleached at 514 nm and detected at 520-620 nm. YFP photobleaching was done by selecting at least five regions and FRET efficiency calculated using ZEN 2008 (Zeiss) software. When using AKAR2 the excitation was performed at 488 nm and emission detected at 495-560 nm (Clover) and at 588-702 nm (mRuby). AKAR2 fluorescence was analyzed using ImageJ software (Schneider et al., 2012) and the mRuby/Clover fluorescence ratio change (ΔFRET%) was calculated.
PKA kinase activity was measured in single oocytes using the PepTag non-radioactive protein kinase assay kit from Promega according to the provided protocol. Briefly, the PepTag assay uses a highly specific fluorescent PKA peptide substrate that changes the peptide's net charge when phosphorylated, allowing easy electrophoretic separation of the phosphorylated and non-phosphorylated peptide. This allows for quantitative measurement of PKA catalytic activity in oocyte lysates. Phosphorylated and non-phosphorylated bands were imaged using the Geliance 600 Imaging system and the intensities of the bands were analyzed and corrected using ImageJ (NIH) software, allowing calculation of percentage phosphorylation with correction for the negative control value in the absence of lysates or PKA.
Xenopus oocytes (∼100) were lysed in Tris-HCl, pH 8 (25 mM, EDTA 0.5 mM, EGTA 0.5 mM, protease inhibitor 1:100) using 5 µl per oocyte, followed by centrifugation at 1000 g for 10 min. Supernatants were collected and centrifuged for 1 h at 150,000 g. The supernatant was saved as the cytosolic fraction and the pellet (membrane fraction) was dissolved with 50 µl PKA extraction buffer (with freshly added 10 mM β-mercaptoethanol, protease inhibitor and PMSF). All centrifugation steps were done at 4°C, and the tubes were kept on ice during the whole procedure. The equivalent of two oocytes was used for the PepTag assay and also for the western blot to detect PKAc.
Cells were ground using a Dounce homogenizer in MPF lysis buffer [0.08 M β-glycerophosphate, 20 mM Hepes (pH 7.5), 15 mM MgCl2, 20 mM EGTA, 1 mM Na-Vanadate, 50 mM NaF, 1 mM DTT, 1 mM PMSF and 0.1% protease inhibitor (Sigma)] and centrifuged twice at 1000 g for 10 min at 4°C to remove yolk granules. The lysates were then incubated with 4% NP40 at 4°C for 2 h followed by centrifugation at maximum speed for 15 min at 4°C and the supernatant was stored. Supernatants were resolved on 4-12% SDS-PAGE gels, transferred to polyvinylidene difluoride (PVDF) membranes (Millipore), blocked for 1 h at room temperature with 5% milk in TBS-T buffer (150 mM NaCl and 20 mM Tris-HCl, pH 7.6, 0.1% Tween) and then incubated overnight at 4°C in 3% BSA in TBS-T with one of the following primary antibodies: anti-PKAc (1:1000, sc-903, Santa Cruz), anti-GFP (1:1000, 2955S, Cell Signaling), anti-actin (1:10,000, A1978, Sigma) and anti-Na+/K+ pump (1:1000, 3010S, Cell Signaling). Blots were washed three times with TBS-T and probed for 1 h with infrared fluorescence, IRDye 800 and 680 secondary antibodies (1:10,000) and the western blots were revealed using the quantitative LiCor Odyssey Clx Infrared Imaging system.
Oocytes were fixed using 4% paraformaldehyde, washed with phosphate-buffered saline containing 30% sucrose and incubated overnight in 50% OCT (WVR, clear frozen section compound) with gentle shaking. The oocytes were then transferred to 100% OCT and frozen into a plastic mold prior to slicing into 7 µm sections on a cryostat. For immunostaining, the oocyte slices were first saturated with 3% BSA and 1% goat serum for 1 h and incubated with anti-PKAα antibody (Santa Cruz, sc-903) at a 1:50 dilution for 2 h. For experiments using the blocking peptide (Santa Cruz, sc-903p) the primary antibody was incubated overnight with the blocking peptide at 5× concentration according to the manufacturer's instructions. The secondary antibody was anti-rabbit IgG coupled to Alexa Fluor 488 (A11008, Fisher Scientific) in a 1:500 dilution for 2 h. Imaging was performed on a Leica SP5 microscope controlled by Leica LAS software using a 40×/1.3 lens and the pinhole slightly open to an optical slice of 1.3 µm.
Values are given as means±s.e.m. Statistical analysis was performed when required using Student's paired and unpaired t-tests.
We thank Dr Lu Sun for helping with the TEPACVV FRET confocal imaging and analysis, and Drs John J. Riordan, Kees Jalink, Edgar C. Young and Jin Zhang for sharing the CFTR, TEPACVV, CNG and AKAR clones, respectively. We thank the Histology and Microscopy Cores of Weill Cornell Medicine Qatar for contributing to these studies. Both Cores are supported by the BMRP program funded by Qatar Foundation.
N.N. designed and performed experiments, analyzed data and wrote the paper. R.C. designed and performed experiments. M.D. and R.P.K. performed experiments. K.M. developed the concepts, analyzed data and wrote the paper.
This work was funded by the Qatar National Research Fund (QNRF) [NPRP 7-709-3-195]. The statements made herein are solely the responsibility of the authors. Additional support for the authors comes from the Biomedical Research Program (BMRP) at Weill Cornell Medical College in Qatar, a program funded by Qatar Foundation.
The authors declare no competing or financial interests.