Introduction
Oocyte maturation is the last phase of oogenesis, where the oocyte grows and acquires the competence to be ovulated and fertilized. Different mechanisms drive the oocyte into maturation including nuclear and cytoplasmic changes.
Nuclear maturation is underlain by the meiotic process and is defined as the period of progression from the first to the second meiotic arrest. In most of the species at the onset of meiosis, the oocyte undergoes a first meiotic arrest in the first prophase (PI) identified by the presence of a large nucleus known as germinal vesicle (GV). Meiosis is then resumed in response to a stimulus that is species-specific and the first sign of meiosis progression is the breakdown of the germinal vesicle (GVBD); after that, meiosis progresses up to metaphase I (MI) or II (MII) depending on the species. Here the meiosis arrests again up to the definitive removal of the block induced by the fertilizing spermatozoon (Tsafriri, Reference Tsafriri1979; Monroy, Reference Monroy1985; Voronina & Wessel, Reference Voronina and Wessel2003; Richard, Reference Richard2007). Resumption of meiosis after the first arrest in PI is induced by an external stimulus via hormones, neurotransmitters, pH increase and different molecules in echinoderms (Yamashita, Reference Yamashita1988; Mita, Reference Mita2000), mollusks (Dubè, Reference Dubè1992; Guerrier et al., Reference Guerrier, Leclerc-David and Moreau1993; Deguchi & Osanai, Reference Deguchi and Osanai1994; Deguchi & Morisawa, Reference Deguchi and Morisawa2003), coelenterates (Takeda et al., Reference Takeda, Kyozuka and Deguchi2006), amphibians (Schorderet-Slatkine, Reference Schorderet-Slatkine1972; Sánchez-Toranzo et al., Reference Sánchez-Toranzo, Oterino, Zelarayàn, Bonilla and Bühler2007) and mammals (Moor et al., Reference Moor, Osborne, Cran and Walters1981; Mattioli & Barboni, Reference Mattioli and Barboni2000; for review see Tripathi et al., Reference Tripathi, Kumar and Chaube2010).
Cytoplasmic maturation changes are related to many metabolic processes that involve cell cycle control proteins, mRNAs, plasma membrane permeability and calcium release (Ducibella et al., Reference Ducibella, Anderson, Alberini, Aalberg and Rangarajan1988; Whitaker & Patel, Reference Whitaker and Patel1990; Homa, Reference Homa1995; Masui, Reference Masui1996; Whitaker, Reference Whitaker1996; Dale & Elder Reference Dale and Elder1997; Wessel et al., Reference Wessel, Brooks, Green, Haley, Voronina, Wong, Zaydfudim and Conner2001; Tosti, Reference Tosti2006). The control of oocyte maturation involves the participation of numerous metabolic pathways; in particular, meiosis arrest and resumption rely on two different mechanisms: a positive stimulation and the removal of an inhibitory signal.
Oocyte maturation may be driven by either physiological inducers, collectively known as MIS (maturation inducing substances) (Kanatani, Reference Kanatani1983; Lambert, Reference Lambert2008), or inhibitory substances. Among these, a key role is played by intracellular concentrations of cyclic AMP (cAMP), whose action depends on interplay between the cAMP-synthesizing enzymes (adenylate cyclases) and the cAMP-degrading enzymes (phosphodiesterases) (Mehlmann, Reference Mehlmann2005; Sun et al., Reference Sun, Miao and Schatten2009). In many animal species, oocytes undergo spontaneous maturation when they are deprived of their accessory cells (Edwards, Reference Edwards1965; Foote & Thibault, Reference Foote and Thibault1969; Eppig, Reference Eppig1991; Sánchez-Toranzo et al., Reference Sánchez-Toranzo, Oterino, Zelarayàn, Bonilla and Bühler2007), or when they are removed from their ovarian environment (Lambert, Reference Lambert2008), suggesting an interplay among the oocyte, its accessory cells and the extracellular environment. In mammals, accessory cells may be responsible for the meiotic block by transferring small molecules to the oocyte through gap junctions connecting the two compartments (Eppig et al., Reference Eppig, Freter, Ward-Bailey and Schultz1983). In fact, it is known that loss of gap junctional communication triggers GVBD in hamster, bovine and mouse (Racowsky & Satterlie, Reference Racowsky and Satterlie1985; Dekel et al., Reference Dekel, Galiani and Sherlizy1988; Thomas et al., Reference Thomas, Armstrong and Gilchrist2004). In particular, cAMP produced in the follicle cells diffuses through the gap junctions, playing a fundamental role in maintaining meiotic arrest in PI (Downs et al., Reference Downs, Daniel, Bornslaeger, Hoppe and Eppig1989) and may even prevent premature spontaneous maturation (see Richard, Reference Richard2007 for a review). An alternative mechanism for meiotic arrest involving endogenous production of cAMP has been proposed by Vaccari et al. (Reference Vaccari, Corner, Mehlmann and Conti2008).
Initiation of meiosis may therefore result from a negative signal, such as the removal of inhibitory factors (cAMP) whose degradation has been shown to stimulate AMP-activated kinase (AMPK), a well documented inducer of GVBD in mice (Chen & Downs, Reference Chen and Downs2008). This is also supported by the spontaneous maturation that in marine animals occurs when oocytes are removed from the ovary and put in sea water (Cuomo et al., Reference Cuomo, Silvestre, Santis and Tosti2006; Stricker & Smythe, Reference Stricker and Smythe2006; Lambert, Reference Lambert2008).
Alternatively, GVBD may be induced by a positive signal, such as growth factors provided by accessory cells and transmitted to the oocyte through gap junctions (O'Donnel et al., Reference O'Donnel, Hill and Gross2004; Park et al., Reference Park, Su, Ariga, Law, Jin and Conti2004).
In the ascidian Ciona intestinalis, the ovary contains immature oocytes of various size and stages of oogenesis, but only full vitellogenic oocytes were shown to be meiotic competent (Silvestre et al., Reference Silvestre, Cuomo and Tosti2009). Although the natural signal responsible for triggering maturation in ascidians remains unknown, we showed that GVBD is modulated by an interplay between pH, trypsin-like molecules and calcium, whereas we excluded a role of the follicle cells in transmitting the signal for oocyte maturation (Silvestre et al., Reference Silvestre, Cuomo and Tosti2009).
In this study, we show that treatment that elevates cAMP levels in immature oocytes of C. intestinalis inhibits the onset of GVBD, in a manner that is more similar to vertebrates than to other ascidian species.
Materials and methods
Chemicals
All the materials were purchased from Sigma-Aldrich.
Animals and oocytes
The C. intestinalis ascidians were collected from the Bay of Naples and kept in tanks with running sea water until use. The single bag-like ovary was removed with forceps and fine scissors, cut, and transferred directly to Petri dishes containing a solution of artificial sea water (ASW: 400 mM NaCl; 50 mM MgCl2; 10 mM KCl; 10 mM CaCl2; 10 mM HEPES) at pH 5.0. Among all the stages present in the ovary (Silvestre et al., Reference Silvestre, Cuomo and Tosti2009), we selected the fully grown oocytes showing a clear GV and characterized by a diameter of the nude oocyte >120 μm, a brown cytoplasm, and a layer of cube-shaped follicle cells up to 35 μm ca. width. The size of nude oocytes and follicle layers was evaluated with a millimetre grid.
Spontaneous maturation and cAMP level maintenance
In vitro GVBD was performed according to Cuomo et al. (Reference Cuomo, Silvestre, Santis and Tosti2006) by incubating GV oocytes in ASW at pH 8.2 as control.
In order to maintain high cAMP levels, the immature oocytes at the GV stage were incubated with the following substances:
• the activator of adenylate cyclase, forskolin at different final concentrations from 50 to 500 μM in ASW at pH 8.2;
• three inhibitors of phosphodiesterases: theophylline at 2 mM final concentration, caffeine at 25 mM final concentration and isobutylmethylxanthine (IBMX) at 1 mM final concentration, all in ASW at pH 8.2;
• the membrane permeable cAMP analogue, 8-bromo cAMP at 4 and 8 mM final concentration in ASW at pH 8.2.
All the incubations were performed for 3 h, after that oocytes were scored for GVBD occurrence at the inverted microscope.
Spontaneous maturation and fertilization test
We tested potential toxicity of the substances used to maintain high cAMP levels. GV oocytes after the incubation time were transferred in ASW pH 8.2 in order to allow the spontaneous maturation. In vitro GVBD oocytes were followed at the inverted microscope in order to assess the disappearance of the GV, the extrusion of follicle cells and the chorion elevation. After 3 h of incubation in ASW, GVBD oocytes were then fertilized by adding 106 spermatozoa/ml. After the assessment of post-fertilization oocyte contraction the zygotes were put in an incubator at 18 °C and followed for embryo development up to hatched larvae (18–24 h). Vitality and morphology of the hatched larvae were observed at the inverted microscope.
As a control we submitted to spontaneous maturation and fertilization the untreated GV oocytes.
Statistical analysis
Differences between GVBD percentages were analyzed with ANOVA (SAS, 1988). In the case of values expressed as percentages, we proceeded to analyze data after arcsine transformation.
Pairwise comparisons of means were analyzed by the least significant difference (LSD) test.
Results
The results were expressed as mean ± standard error (SE).
Adenylate cyclase activator
GV stage oocytes in ASW treated with forskolin gave rise to GVBD % in a dose-dependent manner as follows: 88.8 ± 4.2 at 50 μM (n = 5), 69.3 ± 5.0 at 100 μM (n = 5), 58.3 ± 4.1 at 250 μM (n = 5) with a maximum inhibition value of 29.4 ± 3.6 at 500 μM (n = 8), compared with 91.3 ± 3.6 in control oocytes (n = 8; Fig. 1). The data obtained was significant (p < 0.01) starting from a concentration of 100 μM.
Forskolin-induced GVBD percentage (mean ± SE) in ASW pH 8.2 after 3 h. A vs. B vs. C (p < 0.01).
Phosphodiesterase inhibitors
GV stage oocytes incubated in ASW with phospho-diesterase inhibitors showed significant differences in maturation occurrence.
Theophylline gave rise to 25.0 ± 2.0% GVBD vs. 85.9 ± 3.4% in the control (p < 0.01; n = 9; Fig. 2). Similar results were obtained also by using caffeine that reduced GVBD to 26.0 ± 4.0% vs. 95.0 ± 5.0% of the control (p < 0.01; n = 8; Fig. 2).
Occurrence (mean ± SE) of GVBD percentage in GV stage after 3 h of incubation in: ASW pH 8.2 (control, light shading), theophylline 2 mM, caffeine 25 mM and IBMX 1 mM (dark shading). A vs. B (p < 0.01).
The last inhibitor tested, IBMX, also decreased the maturation percentage leading to 52.4 ± 4.7% GVBD vs. 93.4 ± 2.7% of the control (p < 0.01; n = 11; Fig. 2).
The cAMP analogue
The cAMP analogue, 8-bromo cAMP, was used at two different concentrations, but in both cases we did not observe any significant reduction in GVBD percentage (n = 7; data not shown).
Spontaneous maturation and fertilization test
Oocytes treated with cAMP-elevating substances remained blocked at the GV stage up to the removal from the treatment bath and the incubation in ASW. Here, they retained the ability to undergo spontaneous maturation with higher GVBD percentage in IBMX and caffeine treated oocytes (Fig. 3).
Occurrence (mean ± SE) of GVBD percentage in oocytes removed from the treatment bath and incubated in sea water pH 8.2. The concentrations of substances in the treatment bath were: theophylline 2 mM, caffeine 25 mM, IBMX 1 mM and forskolin 500 μM. The reversible effect was evaluated after 3 h of incubation.
Figure 4 reports time lapse sequence of spontaneous maturation in control and treated oocytes after incubation in ASW. Both the two groups extruded follicle cells within 7 h, but the shape and length of the follicle cells were different from those of ovulated MI oocytes. By fertilizing control and treated oocytes after spontaneous maturation, we observed 2-cell stage and hatched larvae 50 min and 24 h after fertilization, respectively (Table 1). However, hatched larvae were obtained only from oocytes treated with caffeine or IBMX showing 100% abnormal coiled tails with respect of 70% of the control (Table 1 and Fig. 4).
Percentage of 2-cell stage and hatched larvae obtained from GV oocytes matured in vitro and then fertilized (control) or from GV oocytes blocked by the substances, rinsed and incubated in sea water for 3 h and then fertilized.
Percentage of 2-cell and hatched larvae was calculated over the total of oocytes fertilized, while percentage in parentheses was calculated over the total of hatched larvae.
Representative experiment of time lapse sequence of spontaneous maturation in control and treated (caffeine 25 mM) oocytes after incubation in ASW: (a) GV oocytes at the beginning of incubation time; (b) GVBD oocytes after 1–2 h of incubation showing the disappearance of GV; (c) oocytes after 2–3 h showing early follicle cells extrusion; (d) oocytes after 3–5 h with longer follicle cells and beginning of chorion elevation; (e) oocytes after 5–7 h reaching the maximum extrusion of follicle cells in vitro; (f) 2-cell stage 50 min after fertilization of GVBD oocytes (row d); (g) abnormal hatched larvae 24 h after fertilization; (h) mature oocytes collected from the oviduct (on the left) and normal hatched larva (on the right) showed for comparison. Scale bar is 120 μm for the oocytes and 2-cell stage, and 240 μm for the larvae.
Discussion
In this paper we have shown that high cytoplasmic levels of cAMP in C. intestinalis immature oocytes maintain the meiotic arrest at the PI stage.
Many studies have provided evidence of the involvement of cyclic nucleotides in the maintenance of meiotic arrest (Sun et al., Reference Sun, Miao and Schatten2009 for a review) and in particular, it is known that the intracellular second messenger cAMP plays a significant role in the regulation of mammalian oocyte maturation (Gilchrist & Thompson, Reference Gilchrist and Thompson2007). High levels of cAMP, some analogues, cAMP-dependent protein kinase (PKA) and related substances such as GPR3, act by preventing spontaneous maturation and/or blocking GVBD in vitro or, on the contrary, may release oocyte from meiotic arrest (Mehlmann et al., Reference Mehlmann, Saeki, Tanaka, Brennan, Evsikov, Pendola, Knowles, Eppig and Jaffe2004; Richard, Reference Richard2007 for a review; Vaccari et al., Reference Vaccari, Corner, Mehlmann and Conti2008).
In fact, contrasting results exist in the literature regarding the inhibitory role of cAMP elevation (Eppig, Reference Eppig1989; Conti et al., Reference Conti, Andersen, Richard, Shitsukawa and Tsafriri1998; Nogueira et al., Reference Nogueira, Albano, Adriaenssens, Cortvrindt, Bourgain, Devroey and Smitz2003; Voronina & Wessel, Reference Voronina and Wessel2003; Mehlmann, Reference Mehlmann2005) or its role in induction of GVBD (Freeman & Ridgway, Reference Freeman and Ridgway1988; Yamashita, Reference Yamashita1988; Yi et al., Reference Yi, Lefevre, Gagnon, Anctil and Dube2002; Stricker & Smythe, Reference Stricker and Smythe2006; Takeda et al., Reference Takeda, Kyozuka and Deguchi2006; Lambert, Reference Lambert2008).
In ascidian oocytes, spontaneous maturation occurs in natural sea water, resulting in very fast morphological changes such as extrusion of test cells, elevation of the chorion, and extrusion of follicle cells.
In previous papers (Cuomo et al., Reference Cuomo, Silvestre, Santis and Tosti2006; Silvestre et al., Reference Silvestre, Cuomo and Tosti2009), we showed that the largest stage of the Ciona GV oocytes was the only immature stage that is competent for meiosis progression and was able to undergo spontaneous maturation.
Here, we show that treatment of these immature oocytes with cAMP-elevating substances maintained the meiotic block at the GV stage in a reversible manner. In fact, after removal of the inhibiting substances and incubation in sea water, the GV-arrested oocytes were able to undergo in vitro GVBD and extrusion of the follicle cells. However, the resulting size and morphology of the follicle cells was different from those in vivo matured. We demonstrated that the substances used, even at high concentrations, did not interfere with the acquired meiotic competence of the oocyte: in fact, all the oocytes matured after removal of the substances showing a reversible effect; furthermore, they were successfully fertilized showing high percentage of first cleavage. Hatched larvae were obtained in good percentage especially from caffeine or IBMX-incubated oocytes, but in a lower percentage than the control and showing an abnormal morphology. All together, these results suggest that although spontaneous oocyte maturation occurs in sea water, a different follicle cell formation and abnormal larval development may be due to a long-term effect of the lack of the physiological stimulus and related process of meiosis resumption.
Cyclic AMP has been defined as a paradox (Tsafriri & Dekel, Reference Tsafriri, Dekel, Tosti and Boni2010) since its action is differs with species. In the ascidian Boltenia villosa (Lambert, Reference Lambert2008) and in the echinoderm Amphipholis kochii (Yamashita, Reference Yamashita1988), it has been observed that different substances elevating cAMP levels, i.e. forskolin, theophylline, caffeine and IBMX, stimulate GVBD even at low pH values normally used in the ascidians to fully block spontaneous maturation. Here, we demonstrated that, in the C. intestinalis, as in Halocynthia roretzi (Sakairi & Shirai, Reference Sakairi and Shirai1991), the same compounds exerted an opposite effect as they inhibited GVBD.
These data provide support that some physiological mechanisms of Ciona are different from some other invertebrates, but similar to those of some vertebrates such as amphibians and mammals (Kren et al., Reference Kren, Ogushi and Miyano2004; Sánchez-Toranzo et al., Reference Sánchez-Toranzo, Oterino, Zelarayàn, Bonilla and Bühler2007; Ozawa et al., Reference Ozawa, Nagai, Somfai, Nakai, Maedomari, Fahrudin, Karja, Kaneko, Noguchi, Ohnuma, Yoshimi, Miyazaki and Kikuchi2008; Chen et al., Reference Chen, Chi, Moley and Downs2009).
Although Ciona shows many similarity with mammals, in the latter as in other species, the maintenance of elevated cAMP levels rely on a complex communication between oocyte and follicle cells. In Ciona, we previously demonstrated that defolliculated oocytes could directly respond to external stimuli and that functional gap junctions between oocyte and follicle cells were absent, indicating that follicle cells do not play a role in oocyte spontaneous maturation (Silvestre et al., Reference Silvestre, Cuomo and Tosti2009). Although a similar mechanism has been recently proved in marine worms suggesting that GVBD may be induced by a deactivation of AMPK (Stricker et al., Reference Stricker, Swiderek and Nguyen2010), in Ciona an unresolved question remains: how the signal coming from cAMP is transduced to the oocyte. A possible mechanism could involve the test cells that in Ciona exhibit tachykinin receptors, whose activation induces expression of several proteases. These enzymes were found to be responsible for multiple biological events in the growth and maturation of oocytes and follicle cells in various animal species, including Ciona (Aoyama et al., Reference Aoyama, Kawada, Fujie, Hotta, Sakai, Sekiguchi, Oka, Satoh and Satake2008; Silvestre et al., Reference Silvestre, Cuomo and Tosti2009). The lack of action of cAMP analogue 8-bromo cAMP may be attributed to its limited inhibitory capability as shown by Downs & Chen (Reference Downs and Chen2006). Moreover, similar results were obtained by using another cAMP analogue in the H. roretzi oocytes (Sakairi & Shirai, Reference Sakairi and Shirai1991).
Although we show that cAMP plays a role in the process inducing GVBD, at present, the mechanism that influences cAMP increase and hence its role in maintaining maturation arrest remains to be clarified.
In this paper we indicate a variability of biological mechanisms existing among the ascidian species, and on the other hand we support increasing evidence that tunicates share many common biological mechanisms with vertebrates (Delsuc et al., Reference Delsuc, Brinkmann, Chrout and Philippe2006). This close relationship with vertebrates makes ascidians a suitable model for developmental and genomic studies.
Acknowledgements
We thank Prof. Raffaele Boni for helpful comments, Mr Giuseppe Gargiulo and Mr Giampiero Lanzotti for computer graphics.
