Pharmacological Inhibition of Phosphodiesterase 4 Triggers Ovulation ...

41 downloads 78 Views 187KB Size Report
Sep 24, 2004 - Sean D. McKenna, Michael Pietropaolo, Enrico Gillio Tos, Ann Clark, ...... Merz K-H, Marko D, Regiert T, Reiss G, Frank W, Eisenbrand G 1998.
0013-7227/05/$15.00/0 Printed in U.S.A.

Endocrinology 146(1):208 –214 Copyright © 2005 by The Endocrine Society doi: 10.1210/en.2004-0562

Pharmacological Inhibition of Phosphodiesterase 4 Triggers Ovulation in Follicle-Stimulating Hormone-Primed Rats Sean D. McKenna, Michael Pietropaolo, Enrico Gillio Tos, Ann Clark, David Fischer, David Kagan, Bagna Bao, P. Jorge Chedrese, and Stephen Palmer Serono Reproductive Biology Institute (S.D.M., M.P., A.C., D.F., D.K., B.B., S.P.), Rockland, Massachusetts 02370; Istituto di Ricerche Biomediche “A Marxer” (E.G.T.), LCG Bioscience, 10010 Colleretto Giacosa, Italy; and Department of Obstetrics, Gynecology, and Reproductive Sciences (P.J.C.), College of Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 0W8 Phosphodiesterases (PDEs) are a family of enzymes that hydrolyze cyclic nucleotides to render them biologically inactive. As such, these enzymes are critical regulators of signal transduction pathways that use cyclic nucleotides as second messengers. PDE4 is one such member that has been identified in ovarian tissue and purported to have a role in the regulation of gonadotropin action. In the present study, selective PDE4 inhibitors enhanced intracellular signaling in a human LH receptor-expressing granulosa cell line. In vivo, PDE4 inhibition in FSH-primed rats resulted in ovulation, indicating that the PDE4 inhibitors can substitute for LH and human chorionic gonadotropin (hCG) in this process. More-

over, when coadministered with a subeffective dose of hCG, PDE4 inhibitors acted synergistically to enhance the ovulation response. Inhibitors of PDE3 or PDE5 had no ovulatory effect under similar conditions. Oocytes that were ovulated after PDE4 inhibition could be fertilized in vitro at a rate similar to that of oocytes from hCG-induced ovulation. Moreover, such oocytes were fully capable of being fertilized in vivo and developing into normal live pups. These results indicate that small molecule PDE4 inhibitors may be orally active alternatives to hCG as part of a fertility treatment regimen. (Endocrinology 146: 208 –214, 2005)

T

HE PROCESSES OF follicular development and subsequent ovulation in mammals are controlled by the pituitary-derived glycoprotein hormones, FSH, and LH (1, 2). Both FSH and LH bind to their respective cell surface receptors expressed on the somatic cells of the developing follicle and mediate their biological activities through cAMPdependent signal transduction pathways. Whereas both FSH receptors and LH receptors (LHRs) can be expressed on the same cell types in the ovary, these gonadotropins stimulate distinct physiological responses (reviewed in Ref. 3). In granulosa cells, FSH induces aromatase expression, cell proliferation, and follicular growth, whereas LH triggers progesterone production, ovulation, and luteinization (4). Although recent reports have identified other potential signaling pathways for the two gonadotropins (3, 5), cAMP is recognized as the primary second messenger for gonadotropin action because increasing cAMP alone is sufficient to mediate many of the gonadotropin-associated effects in vitro, including estrogen production from granulosa cells, oocyte maturation, and germinal vesicle breakdown (6). Whereas the nature of the gonadotropin-induced signaling events that

determine which cellular responses are induced remain unclear, the divergence of such responses are believed to be controlled by coactivation of auxiliary signal transduction pathways, compartmentalization of regulatory components, influence of cell differentiation state, and/or qualitative differences in the cAMP signal itself (3, 5). Phosphodiesterases (PDEs) are a family of enzymes that regulate signal transduction events that use cyclic nucleotides as second messengers by catalyzing their hydrolysis into biologically inactive 5⬘-nucleotide monophosphate analogs. To date, at least 11 different PDE family members have been identified, each one having multiple subtypes encoded by unique genes (7). Murine ovarian follicles express multiple forms of PDEs, including at least two subtypes of PDE4, whereas the oocyte itself expresses PDE3 (8). Several lines of evidence indicate that one or more of the PDE4 subtypes expressed in follicles are responsible for regulating the biological activity of gonadotropins in granulosa cells. Notably, PDE4D mRNA expression is up-regulated in granulosa cells after treatment with both FSH and LH (9). Hence, lack of PDE4 function may significantly impact gonadotropininduced responses in vivo. Here we demonstrate, using pharmacological models, that selective inhibitors of PDE4 can stimulate ovulation.

First Published Online September 24, 2004 Abbreviations: DC-TA-46, 7-Benzylamino-6-chloro-2-piperazino-4pyrrolidino-pteridine; hCG, human chorionic gonadotropin; IBMX, 3-isobutyl-1-methylxanthine; IVF, in vitro fertilization; LHR, LH receptor; PDE, phosphodiesterase; PMSG, pregnant mare serum gonadotropin; r-h, recombinant human. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.

Materials and Methods Reagents Recombinant human (r-h) FSH and recombinant human chorionic gonadotropin (hCG) were supplied by Laboratoires Serono Aubonne (Aubonne, Switzerland). PDE4 inhibitors piclamilast, mesopram, and

208

McKenna et al. • PDE4 Inhibitors Induce Ovulation

7-benzylamino-6-chloro-2-piperazino-4-pyrrolidino-pteridine (DC-TA46) were synthesized based on published synthetic methods (10 –12). The PDE3 inhibitor milrinone was obtained from Sigma Chemical Co. (St. Louis, MO).

cAMP production in human LHR-expressing JC-410 granulosa cells JC-410, porcine granulosa cells lacking functional endogenous gonadotropin receptors (13, 14), were maintained in DMEM/F12 supplemented with 5% newborn calf serum (Life Technologies, Inc., Grand Island, NY) and 5 ␮g/ml of insulin (Life Technologies, Inc.). The human LHR coding sequence was subcloned into pN␣, a mammalian expression vector containing the mouse metallothionein promoter and a simian virus 40 enhancer/early region selection cassette. The construct sequence was verified with the ThermoSequenase radiolabeled terminator cycle sequencing kit (Amersham Biosciences, Piscataway, NJ). Stable cell lines were established by transfecting the LHR plasmid into JC-410 cells using Effectene (Qiagen, Valencia, CA) according to the manufacturer’s recommended DNA to lipid ratio. The cells were allowed to recover for 48 h before the monolayer was trypsinized and replated in culture medium supplemented with 300 ␮g/ml Geneticin (Life Technologies) for selection. The cells were routinely maintained in 300 ␮g/ml Geneticin. For cAMP determinations, the cells were plated at a density of 25,000 cells/well in 96-well plates 1 d before the assay. The following day, the cells were stimulated for 1 h with increasing doses of the inhibitor molecules in the presence or absence of 1 nm hCG, as indicated. All compounds were diluted in assay buffer (DMEM/F12, 0.1% BSA; Sigma) containing 4% dimethylsulfoxide (0.5% final concentration in the assay). After 1 h stimulation, the cells were lysed and cAMP was assayed using the Tropix cAMP-screen assay (Applied Biosystems, Foster City, CA), according to the manufacturer’s protocol.

Animals All animal studies were approved by the Institutional Animal Care and Use Committee. Sprague Dawley CD BR rats (Charles River Laboratories, Wilmington, MA) were housed under the following constant environmental conditions: temperature 22 C ⫾ 2, relative humidity 55 ⫾ 10%, 15–20 air changes per hour, and artificial light with a 12-h circadian cycle (0700 –1900 h). For the entire duration of the study, the rats were provided with standard pelleted diet and water ad libitum. In some studies, hypophysectomized rats (Charles River Laboratories) were substituted for intact rats. In this case, rats were 26 d of age at the initiation of the study and were provided 5% sucrose/water during the acclimatization and study periods. Hypophysectomized rats with a body weight greater than 55 g were excluded from the studies. Unmanipulated littermate controls had an average body weight of 87 g (⫾7 g).

In vivo ovulation induction assay Immature female rats were weaned at 21–22 d of age and randomly sorted into the experimental groups (n ⫽ 6 – 8/group). The rats were sc injected in the scruff of the neck twice per day (0900 and 1600 h) for 2 d with r-hFSH (606 ng total dose) to induce maturation of multiple ovarian follicles. On the second day and at the same time as the final injection of r-hFSH, all groups received hCG, PDE inhibitor, or a combination of both to induce ovulation of the matured follicles. The vehicle for hCG was PBS, whereas that for the PDE inhibitors was NP3S (5% N-methyl2-pyrrolidone, 30% polyethylene glycol 400, 25% polyethylene glycol 200, 20% propylene glycol, 20% saline). Preliminary studies using this vehicle demonstrated that at the volumes used, this vehicle neither induced ovulation by itself nor inhibited hCG-induced ovulation. On the morning after ovulation induction, rats were killed by CO2 asphyxiation. The ovaries, uterine horns, and uterus body were collected and placed in PBS. The oviducts were removed from the ovaries and placed between two glass microscope slides. The oviducts were then examined by light microscopy under phase-contrast conditions and the ova present in the ampullae of the oviducts were counted. For studies involving hypophysectomized rats, the rats were primed with FSH twice per day for 4 d before ovulation induction. Because the hypophysectomized rats required higher doses of gonadotropins to generate and ovulate follicles, FSH was dosed at 800 ng/d and hCG at

Endocrinology, January 2005, 146(1):208 –214

209

1600 ng (administered concurrently with last FSH injection). Piclamilast was limited to 1 mg/kg due to toxicity observed at higher doses in these animals.

Serum progesterone determination Intact immature rats were primed with FSH (606 ng total dose) and sc treated with hCG (2860 ng), PDE4 inhibitor (2, 10, and 50 mg/kg), or NP3S vehicle, as described for the ovulation induction assay. Thirty-six hours after treatment, sera were collected and stored at ⫺20 C until assay. Sera were assessed for progesterone levels using the active progesterone enzyme immunoassay (Diagnostic Systems Laboratories, Webster, TX) according to manufacturer’s protocol.

In vitro fertilization Immature female rats (22 d of age) were superovulated as described above. The method for in vitro fertilization (IVF) of rat gametes was the protocol described by Toyoda and Chang (15) with minor modifications. All chemicals were from Sigma unless specified. M16 medium was supplemented with 25.1 mm NaHCO3, 26 mm Na lactate, 0.5 mm Na pyruvate (16), and 50 IU penicillin/50 ␮g streptomycin per milliliter (Life Technologies, Rockville, MD). The IVF medium was prepared under sterile conditions using endotoxin-screened water (Life Technologies) and was not filter sterilized due to an unknown toxin contributed by the filter. Supplements were dissolved in the endotoxin-screened water and filter sterilized through a washed syringe filter (Gelman Sciences, Ann Arbor, MI) before being added to the medium. Fresh medium was prepared once per week, the day before the IVF was performed, and was equilibrated overnight at 37 C in a humidified atmosphere of 5% CO2/95% air. Microdrops of IVF medium (100 ␮l for ova and 500 ␮l for sperm) were placed in 35 ⫻ 10-mm culture dishes (Corning, Corning, NY) that were subsequently flooded with light white mineral oil and equilibrated overnight along with the medium. Also equilibrated were 24-well culture dishes containing IVF medium covered by a thin layer of mineral oil; these were used for collection and dissection of oviducts. Throughout the following procedures, culture dishes containing medium or microdrops were maintained at 37 C on heated stage warmers on the dissecting microscopes. Adult male rat sperm donors (3– 6 months of age) were caged with an adult female rat for 2 wk before collection of sperm (16). One male rat was used for each IVF procedure and was killed by CO2 asphyxiation. Cauda epididymides were removed, rinsed in 37 C IVF medium, and trimmed of adipose tissue. Multiple punctures were made in the organs and drops of sperm were transferred to four microdrops using sterile microdissecting forceps. Sperm were allowed to capacitate for 2–3 h at 37 C in a humidified atmosphere of 5% CO2/95% air. Percent motility of the sperm in each microdrop was determined, as was the concentration of sperm using a hemacytometer. Superovulated female rats were killed by CO2 asphyxiation 18 –24 h after ovulation triggering. Oviducts were removed, placed in the medium of the 24-well culture dishes for rinsing, and transferred to fresh wells of medium. The swollen ampulla of each oviduct was torn open with sterile microdissecting forceps, and the clot of ova in their cumulus mass was expressed into the medium. Ovum masses were transferred to microdrops using pipettors fitted with sterile MultiFlex microcapillary pipette tips (VWR, South Plainfield, NJ). Microdrops were inseminated with 25 ␮l sperm for a final average concentration of 1.3 ⫻ 105 sperm/ml. At 18 –20 h after insemination, ova/embryos were rinsed through three microdrops. Using phase-contrast microscopy, the number of dead, unfertilized, and fertilized ova and cleaved embryos were recorded. Ova were considered to be fertilized if a sperm tail was seen inside the plasma membrane, two polar bodies were present, or two pronuclei were visible. Dead cells were identified by fragmentation of the cytoplasm. Cleaved embryos were at the 2- to 4-cell stage.

In vivo fertility assay Three cohorts of rats (n ⫽ 6) were induced to ovulate as described above with the following modifications. Rats (28 d of age) were treated with pregnant mare serum gonadotropin (PMSG) (3 IU) on d 1 and with FSH (700 ng/d) on d 2 and 3 to induce follicular maturation and enable mating behavior. In previous studies, rats injected with FSH only would

210

Endocrinology, January 2005, 146(1):208 –214

not exhibit lordosis (data not shown). After ovulation induction, one cohort of rats was killed for oocyte counting as described above, whereas two cohorts were caged singly with adult male rats of proven fertility. Successful mating was determined by the presence of vaginal plugs on the morning after mating, after which the females were removed to separate cages. On d 11 post mating, the second cohort of animals was killed and the number of viable embryo implantation sites was visualized in the uterine horns. The final cohort of rats was allowed to progress to parturition at which time the number of live born pups was determined.

Statistics Significance differences in ovulation induction were determined by one-way ANOVA using Origin 7.0 (OriginLab Corp., Northampton, MA). Tukey’s honestly significant difference method was used for P value adjustment for multiple comparisons. Statistical significance was set at P ⬍ 0.05 for all assays.

Results PDE4 inhibitors increase hCG-induced cAMP production in LHR-expressing cells

To investigate the potentiation of hCG bioactivity through the selective inhibition of PDE4, cells from the porcine granulosa line JC410, stably transfected to express the human LHR, were treated with the PDE4 inhibitors mesopram or piclamilast in the presence or absence of hCG. hCG itself was capable of inducing a modest increase in cAMP in these cells when added in the absence of a PDE inhibitor (Fig. 1). As expected, the cAMP levels were dramatically increased when the nonselective PDE inhibitor 3-isobutyl-1-methylxanthine (IBMX) was added along with the hCG. hCG, either with or without IBMX, had no effect on the cAMP levels in parental or FSH receptor-expressing JC410 cells (not shown). When the JC410 cells were treated with either of the PDE4 inhibitors alone, no response was observed at concentrations up to 10 ␮m. In contrast, when piclamilast or mesopram were added to the cells with 1 nm hCG, a dose-dependent increase in cAMP was observed, with EC50 values of 23 and 156 nm, respectively.

FIG. 1. Effect of PDE4 inhibition on cAMP generation in stable granulosa cells expressing the human LHR. Porcine granulosa cells expressing the human LHR were treated for 1 h with 1 nM hCG (⫾100 ␮M IBMX) or varying doses of PDE4 inhibitors (⫾1 nM hCG). Data are expressed as the mean cAMP concentration from duplicate wells ⫾ SD.

McKenna et al. • PDE4 Inhibitors Induce Ovulation

PDE4 inhibitors in triggering ovulation

Based on the in vitro demonstration of the synergy between hCG and PDE4 inhibitors, the effect of these inhibitors on hCG/LH-mediated functions in vivo was assessed. Because the midcycle LH surge is responsible for inducing final oocyte maturation and triggering ovulation of mature ovarian follicles in vivo, the impact of PDE4 inhibition on the ovulation process was investigated by treating FSH-primed immature rats with PDE4 inhibitors. Subcutaneous administration of the PDE4 inhibitors mesopram and piclamilast triggered ovulation in all of the rats tested when administered at 10 and 2 mg/kg, respectively (Fig. 2A). That this is a general phenomenon mediated by selective PDE4 inhibitors is supported by additional studies in which three other selective PDE4 inhibitors, namely WAY-PDA-641 (17), LAS31025 (18), and CI1044 (19), were also found to induce ovulation when administered to FSH-primed rats (data not shown). Because an increase in hCG/LH-induced progesterone is indicative of the formation of normally functioning corpora lutea in the ovary (20), the level of progesterone was evaluated in the sera of FSH-primed rats after treatment with either hCG or the PDE4 inhibitor piclamilast. As demon-

FIG. 2. Effect of treatment with PDE4 inhibitors in ovulation induction. FSH-primed, immature female rats were injected sc with hCG (2860 ng/rat), the indicated amount of PDE4 inhibitors, or the NP3S vehicle. Treated rats were killed either on the morning after ovulation induction and the number of ovulated oocytes determined for each rat (n ⫽ 8) (A) or after 36 h for evaluation of serum progesterone levels (n ⫽ 8) (B). Data are expressed as mean number of ovulated oocytes per rat or serum progesterone concentration ⫾ SEM. Significant differences vs. vehicle control (P ⬍ 0.05) are indicated by an asterisk (*).

McKenna et al. • PDE4 Inhibitors Induce Ovulation

strated in Fig. 2B, 36 h after treatment with either hCG or the PDE4 inhibitor, significantly elevated progesterone levels were detected, indicating that corpora lutea generated after PDE4 inhibitor treatment were fully mature. Correspondingly, corpora lutea were clearly visible on ovaries collected from the hCG and PDE4 inhibitor treated but not on vehicletreated FSH-primed rats. Unlike the PDE4 inhibitors piclamilast (IC50 1.5 nm) (21) and mesopram, the PDE4 inhibitor DA-TC-46 (IC50 16 nm) (11) was not able to trigger ovulation at even the highest dose tested. However, in combination with a subeffective dose of hCG, even this inhibitor was capable of inducing more than 25 oocytes/rat at 50 mg/kg (Fig. 3). Similar synergy could be observed between the subeffective dose of hCG and low doses of piclamilast and mesopram (not shown). These results demonstrate synergy between PDE4 inhibition and hCG in triggering ovulation, but treatment with potent PDE4 inhibitors alone is sufficient to mediate this response. This does not exclude the possibility that in hCG-untreated immature rats, there may be a low level of endogenous LH that can be potentiated by inhibition of the regulatory PDE4. In contrast to the ovulatory effects of PDE4 inhibitors, treatment with either the PDE3-selective inhibitor milrinone or the PDE5-selective inhibitor sildenafil citrate had no effect on the number of oocytes that were ovulated when administered either alone or with the subeffective dose of hCG under conditions identical with those in which the PDE4 inhibitors were tested (data not shown). To determine whether the PDE4 inhibitors were acting through either the hypothalamus or pituitary, possibly by stimulating the release of endogenous LH, ovulation induction studies were performed in FSH-primed hypophysectomized rats. As observed in intact rats, treatment of the hypophysectomized rats with the PDE4 inhibitor piclamilast together with a suboptimal dose of hCG could increase the number of ovulated oocytes, compared with those receiving hCG only (Fig. 4). Additionally, treatment with piclamilast in the absence of exogenous hCG induced ovulation in the rats,

FIG. 3. Synergy between hCG and PDE4 inhibitor in ovulation induction. FSH-primed immature female rats were injected sc with a subeffective dose of hCG (110 ng), the PDE4 inhibitor DC-TA-46, a combination of the two, or the NP3S vehicle. On the morning after ovulation induction, the number of ovulated oocytes was determined for each rat (n ⫽ 8). Data are expressed as the number of oocytes ovulated per rat (mean ⫾ SEM). Significant differences vs. vehicle control (P ⬍ 0.05) are indicated by an asterisk (*).

Endocrinology, January 2005, 146(1):208 –214

211

FIG. 4. Inhibition of PDE4 in hypophysectomized rats enhances hCG-induced ovulation. Hypophysectomized rats (26 d old) were primed for 4 d with r-hFSH (800 ng/d) before ovulation induction with a sc injection of hCG (1600 ng), piclamilast (1 mg/kg), or a combination of the same doses of hCG and piclamilast. On the morning after ovulation induction, the number of ovulated oocytes was determined for each rat (n ⫽ 6). Data are expressed as the number of oocytes ovulated per rat (mean ⫾ SEM). Significant differences for piclamilast treatment vs. respective control group (P ⬍ 0.05) are indicated by an asterisk (*).

whereas no ovulation was observed in vehicle-treated rats. These results are consistent with the target cells of PDE4 inhibition being downstream of the pituitary. Coupled with the in vitro and in vivo synergies observed between hCG and PDE4 inhibitors, it is likely that the target for the PDE4 inhibitors are the LH receptor-expressing ovarian cells. Fertility potential of oocytes ovulated with PDE4 inhibitors

To assess the effects of PDE4 inhibitors on the fertility potential of oocytes, FSH-primed rats were induced to ovulate with either hCG or the PDE4 inhibitor piclamilast (10 mg/kg), and harvested oocytes were incubated with capacitated sperm and evaluated 18 –20 h later for evidence of fertilization, cleavage, or fragmentation. As demonstrated in two independent studies, the oocytes derived from piclamilast-induced ovulation were fully capable of being fertilized in vitro (Table 1). The overall fertility and cleavage rates of these oocytes were similar to those that were induced to ovulate with hCG. Because no oocytes were ovulated in control rats that received neither piclamilast nor hCG, all of the collected oocytes were associated with one of these treatments. To determine the developmental potential of oocytes resulting from ovulation induced by PDE4 inhibition, 28-d-old rats were mated after ovulation induction with either piclamilast or hCG. To induce follicle maturation as well as support lordosis behavior, the rats were primed with a combination of PMSG and FSH. This priming protocol, in the absence of an ovulatory trigger, resulted in a low level of ovulation and embryonic development after mating (Table 2). In contrast, treatment with the PDE4 inhibitor induced a relatively high level of ovulation in all of the rats evaluated. Moreover, the oocytes from the PDE4 inhibitor-treated group were capable of being fertilized in vivo and developing into grossly normal pups at full term. Notably, hCG triggered ovulation in all of the rats in that group, but no viable embryos or live pups were observed after mating.

212

Endocrinology, January 2005, 146(1):208 –214

McKenna et al. • PDE4 Inhibitors Induce Ovulation

TABLE 1. In vitro fertility of oocytes harvested from ovulations induced with hCG or PDE4 inhibitor Ovulation triggera

Study

n

Nonviable

Viable/unfertilized

Fertilized

Cleaved

1 2 1 2

47 85 64 103

6.3 17.6 14.7 15.5

54.2 24.7 44.1 27.2

39.5 57.6 41.2 57.3

10.4 10.6 22.1 21.3

hCG (1430 ng/rat) Piclamilast (10 mg/kg)

FSH-primed immature female rats were induced to ovulate with hCG or piclamilast. Collected oocytes were subjected to IVF procedures. Results represent number of oocytes observed (n) and the percentage of oocytes observed in each category in two independent studies. Total percentage exceeds 100 because fertilized embryos includes those at the cleavage stage as well as uncleaved oocytes with penetrated sperm, two polar bodies present, or two pronuclei. a No ovulations were observed in rats receiving vehicle only.

Discussion

Gonadotropins are important therapeutic agents in the treatment of infertility. A major drawback with their use is the need for multiple sc injections. It is therefore of potential clinical interest to identify targets for small molecules that may enhance or substitute for gonadotropin therapy. Several lines of evidence support a role for PDE4 in the regulation of gonadotropin action in granulosa cell function, including: 1) gonadotropin regulation of PDE4 activity in vitro (22), 2) changes in PDE4 activity that parallel the estrous cycle (23), and 3) detection of PDE4 in granulosa cell compartments of ovarian follicles (8). In the latter study, the in vitro treatment of follicle cultures with the PDE4 selective inhibitor rolipram resulted in resumption of meiosis at low concentrations and maturation of follicle-enclosed oocytes at higher concentrations. Previous in vivo studies are consistent with a link between PDE4 inhibition and LH signaling. Treatment of mice with rolipram induced premature luteinization of granulosa cells in ovarian follicles (3). Similarly, PDE4D-null mice showed a reduced capacity to ovulate, a small litter size, and a reduced response to superovulation due to the capture of oocytes within luteinized follicles (24). These published results and those of the present study are indicative of the critical timing of PDE4 inhibition in determining the nature of the ovarian response. Chronic PDE4 inhibition, particularly during follicular maturation, may be more likely to mimic a premature LH-like response, whereas inhibition of PDE4 after FSH-induced follicular maturation may mimic the normal LH surge. PDE4 enzymes are encoded by four homologous genes, PDE4A to PDE4D, each of which may express numerous isoforms (7). Genetic deletion of PDE4D results in a clear reproductive phenotype, and both PDE4B and PDE4D have been found to be expressed in ovarian follicles by in situ TABLE 2. Impact of PDE4 inhibition on in vivo fertility Vehicle

Ovulation Viable embryos (d 11) Live pups (d 23)

HCG (1430 ng/rat)

Piclamilast (10 mg/kg)

Avg.

No./6

Avg.

No./6

Avg.

No./6

3.8 5.8 1.5

4 1 2

13.3 0 0

6 0 0

22.3 23.2 15.5

6 5 4

Three cohorts of PMSG/FSH-primed rats (n ⫽ 6 per group) were treated sc with hCG or piclamilast to induce ovulation. The first cohort was assessed for ovulation number, whereas the second and third cohorts were assessed for preterm and full-term pregnancy after mating. Data are represented as mean for all rats in a given group (Avg.) and the number of responding rats per group (No./6).

hybridization, with PDE4B localized to the thecal and interstitial cells and PDE4D to the mural (but not cumulus) granulosa cells of antral follicles (8). Because LHR expression has been demonstrated in both of these cell types (25), it is unclear whether inhibition of PDE4B, PDE4D, or both is necessary for triggering ovulation. Rolipram, the inhibitor reported to induce premature luteinization in mice, has only a 2-fold selectivity for PDE4B vs. PDE4D (26). Similarly, piclamilast, the most potent inducer of ovulation tested in our model, has little selectivity for the individual PDE4 gene products (27), although this inhibitor is highly selective for the PDE4 family members (28). Exact identification of the PDE4 subtype(s) critical for ovulation must await the identification of subtype-selective inhibitors or the availability of mice genetically deficient in the other PDE4 genes. In addition to the expression of PDE4 in follicular cells, PDE3 isoforms are also expressed in the ovarian follicle (29). As seen in PDE4 knockout mice, mice lacking PDE3 display fertility deficiencies. However, because expression of PDE3 appears to be localized in the oocyte, which lacks gonadotropin receptor expression, it should not be unexpected that an inhibitor of PDE3 was incapable of substituting for or potentiating gonadotropin activity. It should be noted, however, that although infertile, the PDE3 knockout mice were fully capable of ovulating oocytes under normal gonadotropin stimulation. Similarly, inhibition of PDE5, a cGMP-specific phosphodiesterase found in brain, vascular smooth muscle cells, and other tissues (30), also did not induce ovulation under the current experimental conditions. Although other non-type 4 phosphodiesterases have been reported in granulosa cells (3), no evidence that PDE5 is expressed in ovarianspecific tissues has been reported. Whereas in vitro and in vivo evidence supports a direct role for PDE4 in regulating CG/LH-mediated granulosa cell function, an indirect role for PDE4 in triggering ovulation cannot be ruled out because the PDE4 selective inhibitor rolipram can increase secretion of GnRH from the GT1 neural cell line in vitro (31), and intracerebroventricular administration of the PDE4 inhibitor denbufylline into adult male rats induces the release of LH (32). In the latter study, release of LH may have been mediated by cross-reactivity of the inhibitor with a non-PDE4 target because intracerebroventricular administration of rolipram did not induce LH release. To determine whether the ovulation that was induced with PDE4 inhibitors in the present study was mediated through the release of LH from the pituitary, the use of hypophysectomized rats was employed. Whereas the overall sensitivity of these rats to gonadotropins was less than that

McKenna et al. • PDE4 Inhibitors Induce Ovulation

observed in intact rats, the fact that the PDE4 inhibitor piclamilast was capable of inducing ovulation supports a direct action of the inhibitors on the ovarian cells themselves. Although PDE4 inhibitors strongly synergize with subeffective levels of hCG in inducing cAMP in rat (not shown) and porcine granulosa cells in vitro, it should be noted that we were unable to detect a change in ovarian cAMP levels after treatment of FSH-primed rats in vivo with the PDE4 inhibitors at ovulatory doses, possibly supporting an indirect role of the inhibitors in ovarian function. However, because we also could not detect ovarian cAMP changes after hCG treatment in vivo, it is possible that a biologically sufficient cAMP response took place but was undetected, either due to the transient nature of the response or masking by the FSHinduced cAMP. Recently Park et al. (33) provided evidence that LH signaling through the LH receptor induces the downstream release of epithelial growth factor-like growth factors which, in turn, mediate many of the LH-associated follicular responses such as cumulus expansion and oocyte maturation. This paracrine mechanism cascades the LH signal throughout the ovarian follicle. Whereas results presented in our study are consistent with a direct role for PDE4 in regulating LH receptor activity, it remains possible that the inhibitors are in fact involved in the regulation of the signaling of some or all of these downstream peptide mediators. In the clinic, hCG is typically administered sc to induce ovulation as part of clinical ovulation induction programs in the treatment of infertility. The identification of orally active small molecules that can substitute for gonadotropin activity may make such injectable therapeutics unnecessary. From the present results, we suggest that inhibitors of PDE4 may be suitable for the induction of ovulation as part of an assisted reproduction technology protocol. Aside from practical and economical considerations, such a treatment regimen could potentially reduce the risk of ovarian hyperstimulation syndrome, a rare but serious complication of a clinical ovarian stimulation regimen. The development of this syndrome has been linked to the prolonged activity of hCG in treated women (34). Therefore, the use of small molecules that have relatively rapid clearance rates and/or affect only one aspect of the hCG/LH signal transduction pathway may be advantageous. As demonstrated in the present fertility studies, oocytes recovered after PDE4 inhibition were competent to be fertilized in vitro, and in vivo fertility, as measured by the development of viable embryos and normal full-term pups, was markedly greater than that after hCGinduced ovulation. Of potential relevance, Mattheij et al. (35) reported that administration of 20 IU hCG to rats during late diestrus or proestus before mating results in embryonic mortality at some point after d 3 of pregnancy but before implantation. The authors speculate that the hCG disrupted steroidogenesis, which may have then interfered with implantation. However, the LHR has been demonstrated to be expressed in the oviduct and stromal cells of the uterine endometrium of mice and rats (36), so a direct role for hCG on extragonadal tissue in mediating this mortality cannot be ruled out. Based on our results, the hCG-mediated embryo toxicity may have been circumvented by the use of PDE4 inhibitors in place of hCG. In summary, the results presented herein demonstrate the

Endocrinology, January 2005, 146(1):208 –214

213

utility of PDE4-selective inhibitors in triggering ovulation. This notion supports the concept that targeting components of gonadotropin signal transduction pathways with small molecules may provide an alternative to the use of injectable protein therapeutics in the treatment of infertility. Acknowledgments Received May 4, 2004. Accepted September 9, 2004. Address all correspondence and requests for reprints to: Sean D. McKenna, Serono Reproductive Biology Institute, One Technology Place, Rockland, Massachusetts 02370. E-mail: sean.mckenna@serono. com.

References 1. Richards JS 1980 Maturation of ovarian follicles: actions and interactions of pituitary and ovarian hormones on follicular cell development. Physiol Rev 60:51– 89 2. Knobil E 1980 The neuroendocrine control of the menstrual cycle. Recent Prog Horm Res 36:53– 88 3. Conti M 2002 Specificity of the cyclic adenosine 3⬘,5⬘-monophosphate signal in granulosa cell function. Biol Reprod 67:1653–1661 4. Richards JS 2001 Perspective: the ovarian follicle—a perspective in 2001. Endocrinology 142:2184 –2193 5. Houslay MD, Adams DR 2003 PDE4 cAMP phosphodiesterases: modular enzymes that orchestrate signaling cross-talk, desensitization and compartmentalization. Biochem J 370:1–18 6. Brannstrom M, Koos RD, Le Maire WJ, Janson PO 1987 Cyclic adenosine 3⬘,5⬘-monophosphate-induced ovulation in the perfused rat ovary and its mediation by prostaglandins. Biol Reprod 37:1047–1053 7. Soderling SH, Beavo JA 2000 Regulation of cAMP and cGMP signaling: new phosphodiesterases and new functions. Curr Opin Cell Biol 12:174 –179 8. Tsafriri A, Chun S-Y, Zhang R, Hsueh AJW, Conti M 1996 Oocyte maturation involves compartmentalization and opposing changes of cAMP levels in follicular somatic and germ cells: studies using selective phosphodiesterase inhibitors. Dev Biol 178:393– 402 9. Park J-Y, Richard F, Chun S-Y, Park, J-H, Law E, Horner K, Jin CS-L, Conti M 2003 PDE4D regulation is critical for the differentiation and pattern of gene expression in granulosa cells of the follicle. Mol Endocrinol 17:1117–1130 10. Ashton MJ, Cook DC, Fenton G, Karlsson JA, Palfreyman MN, Rabur D, Ratcliffe AJ, Souness JE, Thurairatnam S, Vicker N 1994 Selective type IV phosphodiesterase inhibitors as antiasthmatic agents. The syntheses and biological activities of 3-(cyclopentyloxy)-4-methoxybenzamides and analogues. J Med Chem 27:1696 –1703 11. Merz K-H, Marko D, Regiert T, Reiss G, Frank W, Eisenbrand G 1998 Synthesis of 7-benzylamino-6-chloro-2-piperazino-4-pyrrolidinopteridine and novel derivatives free of positional isomers. Potent inhibitors of cAMP-specific phosphodiesterase and of malignant tumor cell growth. J Med Chem 41:4733– 4743 12. Laurent H, Ottow E, Kirsch G, Wachtel H, Schneider H, Faulds D, Dinter H 1997 Chiral methyl phenyl oxazolidinones. WO 97/15561 Patent 13. Chedrese PJ, Rodway MR, Swan CL, Gillio-Meina C 1998 Establishment of a stable steroidogenic porcine granulosa cell line. J Mol Endocrinol 20:287–292 14. Rodway MR, Swan CL, Gillio-Meina C, Crellin NK, Flood PF, Chedrese PJ 1999 Regulation of steroidogenesis in JC-410, a stable cell line of porcine granulosa origin. Mol Cell Endocrinol 148:87–94 15. Toyoda Y, Chang MC 1974 Fertilization of rat eggs in vitro by epididymal spermatozoa and the development of eggs following transfer. J Reprod Fertil 36:9 –22 16. Berger T, Miller G, Horner CM 2000 In vitro fertilization after in vivo treatment of rats with three reproductive toxicants. Reprod Toxicol 14:45–53 17. Heaslip RJ, Lombardo LJ, Golankiewicz JM, Ilsemann BA, Evans DY, Sickles BD, Mudrick JK, Bagli J, Weichman BM 1994 Phosphodiesterase-IV inhibition, respiratory muscle relaxation and bronchodilation by WAY-PDA-641. J Pharmacol Exp Ther 268:888 – 896 18. Burnouf C, Auclair E, Avenel N, Bertin B, Bigot C, Calvet A, Chan K, Durand C, Fasquelle V, Feru F, Gilbertsen R, Jacobelli H, Kebsi A, Lallier E, Maignel J, Martin B, Milano S, Ouagued M, Pascal Y, Pruniaux MP, Puaud J, Rocher MN, Terrasse C, Wrigglesworth R, Doherty AM 2000 Synthesis, structureactivity relationships, and pharmacological profile of 9-amino-4-oxo-1-phenyl3,4,6,7-tetrahydro[1,4]diazepino[6,7,1-hi]indoles: discovery of potent, selective phosphodiesterase type 4 inhibitors. J Med Chem 43:4850 – 4867 19. Newman SP, Rivero X, Luria X, Hooper G, Pitcairn GR 1997 A scintigraphic study to evaluate the deposition patterns of a novel anti-asthma drug inhaled from the Cyclohaler dry powder inhaler. Adv Drug Deliv Rev 26:59 – 67 20. Damber JE, Cajander S, Gafvels M, Selstam G 1987 Blood flow changes and

214

21.

22. 23. 24. 25. 26.

27. 28.

Endocrinology, January 2005, 146(1):208 –214

vascular appearance in preovulatory follicles and corpora lutea in immature, pregnant mare’s serum gonadotropin-treated rats. Biol Reprod 37:651– 658 Souness JE, Griffin M, Maslen C, Ebsworth K, Scott LC, Pollock K, Palfreyman MN, Karlsson JA 1996 Evidence that cyclic AMP phosphodiesterase inhibitors suppress TNF␣ generation from human monocytes by interacting with a ‘low-affinity’ phosphodiesterase conformer. Br J Pharmacol 118:649 – 658 Conti M, Kasson BG, Hsueh AJW 1984 Hormonal regulation of a 3⬘,5⬘adenosine monophosphate phosphodiesterase in cultured rat granulosa cells. Endocrinology 114:2361–2368 Schmidtke J, Meyer H, Epplen JT 1980 Cyclic AMP phosphodiesterases of the rat ovary: oestrous cycle dependent activity change of high affinity form. Acta Endocrinol (Copenh) 95:404 – 411 Jin S-LC, Richard F, Kuo W-P, D’Ericole AJ, Conti M 1999 Impaired growth and fertility of cAMP-specific phosphodiesterase PDE4D-deficient mice. Proc Natl Acad Sci USA 96:11998 –12003 Bukovsky A, Chen TT, Wimalasena J, Caudle MR 1993 Cellular localization of luteinizing hormone receptor immunoreactivity in the ovaries of immature, gonadotropin-primed and normal cycling rats. Biol Reprod 48:1367–1382 MacKenzie SJ, Houslay MD 2000 Actions of rolipram on specific PDE4 cAMP phosphodiesterase isoforms and on the phosphorylation of cAMP-responseelement-binding protein (CREB) and p38 mitogen-activated protein (MAP) kinase in U937 monocytic cells. Biochem J 347:571–578 Cohan VL, Showell HJ, Fisher DA, Pazoles CJ, Watson JW, Turner CR, Cheng JB 1996 In vitro pharmacology of the novel phosphodiesterase type 4 inhibitor, CP-80633. J Pharmacol Exp Ther 278:1356 –1361 Ukita T, Sugahara M, Terakawa Y, Kuroda T, Wada K, Nakata A, Ohmachi Y, Kikkawa H, Ikezawa K, Naito K 1999 Novel, potent, and selective phos-

McKenna et al. • PDE4 Inhibitors Induce Ovulation

29. 30. 31. 32. 33. 34.

35.

36.

phodiesterase-4 inhibitors as antiasthmatic agents: synthesis and biological activities of a series of 1-pyridylnaphthalene derivatives. J Med Chem 42: 1088 –1099 Masciarelli S, Horner K, Liu C, Park SH, Hinckley M, Hockman S, Nedachi T, Jin C, Conti M, Manganiello V 2004 Cyclic nucleotide phosphodiesterase 3A-deficient mice as a model of female infertility. J Clin Invest 114:196 –205 Lin C 2004 Tissue expression, distribution, and regulation of PDE5. Int J Impot Res 16(Suppl 1):S8 –S10 Sakakabara H, Conti M, Weiner RI 1998 Role for phosphodiesterases in the regulation of gonadotropin-releasing hormone in GT1 cells. Neuroendocrinology 68:365–373 Kumari M, Cover PO, Poyser RH, Buckingham JC 1997 Stimulation of the hypothalamo-pituitary-adrenal axis in the rat by three type-4 phosphodiesterase inhibitors: in vitro and in vivo studies. Br J Pharmacol 121:459 – 468 Park J-Y, Su Y-Q, Arigo M, Law E, Jin S-L C, Conti M 2004 EGF-like growth factors as mediators of LH action in the ovulatory follicle. Science 303:682– 684 European Recombinant LH Study Group 2001 Human recombinant luteinizing hormone is as effective as, but safer than, urinary human gonadotropin in inducing final follicular maturation and ovulation in in vitro fertilization procedures: results of a multicenter double-blind study. J Clin Endocrinol Metab 86:2607–2618 Mattheij JA, de Boer P, Dekker B, Rijnders E, Swarts JJ 1986 Administration of human chorionic gonadotropin but not of luteinizing hormone-releasing hormone at pro-oestrus or late di-oestrus has a deleterious effect on pregnancy in the rat. Gynecol Obstet Invest 22:84 –90 Zhang M, Shi H, Segaloff DL, Van Voorhis BJ, Zheng M 2001 Expression and localization of luteinizing hormone receptor in the female mouse reproductive tract. Biol Reprod 64:179 –187

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.