Induced ovulation and egg deposition in the direct developing anuran Eleutherodactylus coqui

This study investigates ovulation and egg deposition behaviors in the anuran Eleutherodactylus coqui from Puerto Rico in response to stimulation with gonadotropin and gonadotropin releasing hormones. Five hormones were tested by injection over a range of doses, including mammalian LHRH, avian LHRH, fish LHRH, D-Ala6, des-Gly10 ethylamide LHRH and hCG. We report a low level of ovulation and egg deposition in response to all hormones, with the most complete and consistent results from the non-natural D-Ala6, des-Gly10 ethylamide LHRH derivative. To confirm the viability of eggs produced in this manner we performed in vitro fertilization experiments that resulted in the development of normal frogs. Reproductive behaviors in E. coqui are apparently not controlled by a mammalian form of LHRH as reported in other common laboratory anuran species. D-Ala6, des-Gly10 ethylamide LHRH induces ovulation and deposition of mature and fertilizable eggs in E. coqui.


Background
Several amphibian species have been commonly used in studies of reproductive biology. Reasons for this include external fertilization and development in large, easily manipulated eggs. Despite the large amount of information known regarding a few laboratory species (most notably Rana pipiens and Xenopus laevis), the reproductive biology of the majority of amphibian species remains poorly understood. This is unfortunate as amphibians, and especially anurans, show the greatest diversity in reproductive strategies among all of the terrestrial vertebrates, including internal and external fertilization, terrestrial and aquatic breeding, development with a larval stage, direct external development, ovoviviparity, mass seasonal breeding, continuous breeding, and presence or absence of parental care. This diversity in reproductive strategies can be expected to be a result of differences in the physiological control of reproduction, including hormonal control of sexual behaviors.
Our interest has centered on frogs in the neotropical genus Eleutherodactylus. With over 700 described species, this is the largest vertebrate genus and as such is an excellent system for studies of comparative biology [1]. As far as it is known, these species undergo direct development in terrestrial eggs (one species is known to be ovoviviparous [2]), and often exhibit parental care [3]. Many species of these frogs are territorial and continuous or nearly continuous breeders. It has been shown for one species (Eleutherodactylus coqui) that population sizes are limited by the availability of terrestrial retreat and nesting sites as opposed to food availability [4]. These developmental and behavioral adaptations make Eleutherodactylus species quite distinct from other commonly used laboratory frog species.
Eleutherodactylus coqui, the common Puerto Rican coqui, has received attention as a model for acoustic communication and developmental biology (For example: [5][6][7][8][9]). We are interested in understanding the hormonal control of ovulation and egg deposition in this and other Eleutherodactylus species. Reproductive behavior, including ovulation, can often be induced artificially in other species by injection of the pituitary glands of the same or closely related species [10]. This can be difficult for routine applications due to the need for large numbers of animals that must be sacrificed to harvest the pituitaries. Since this would be problematic for E. coqui and most other Eleutherodactylus species because they are difficult to collect and keep in captivity, we have here investigated the ovulatory effect of stimulation with peptide hormones.
Several reproductive hormones have been previously shown to induce ovulation in other anuran species [11][12][13][14][15]. In Xenopus laevis, human chorionic gonadotropin (hCG) is routinely used for this purpose [11,12]. However, Xenopus appears to be unusual in this aspect because it is one of a minority of species of anurans that responds to hCG [10]. In addition to direct stimulation of the gonads with gonadotropins, stimulation with leutinizing hormone releasing hormones (LHRHs) has also been successful in inducing ovulation in some other anurans [14,15]. LHRHs are fairly well conserved among vertebrates and often show considerable cross reactivity between even distantly related species [16]. However, amphibians have several forms of LHRH present in their brains [17]. In Xenopus laevis, the mammalian form of LHRH appears to be the functional form controlling the reproductive pathway and leutinizing hormone (LH) and follicle stimulating hormone (FSH) release [18]. However, it is unknown if this form plays the same role in Eleutherodactylus. We therefore compared the effect of hCG and several different commercially available varieties of LHRH, including a modified form with improved pharmacological stability and enhanced activity in other species [19]. The purpose of this study was to determine which, if any, vertebrate peptide reproductive hormone was able to induce ovulation and egg deposition in the Puerto Rican frog E. coqui as a first step towards elucidating the details of this pathway in Eleutherodactylus frogs. Here we describe the results of trials using several hormones and report that all hormones tested produced ovulation and egg deposition in at least one animal and that D-Ala6, desGly10, ethylamide LHRH most reproducibly induced ovulation and egg deposition in this species.

Methods
Eleutherodactylus coqui were collected near El Verde Field Station in El Yunque National Forest, Puerto Rico. The frogs were housed in the laboratory in 38 l glass aquaria separately or in pairs as described [20]. The aquaria each had approximately four cm of moist peat moss as a substrate with ten-centimeter long, 2.5 cm diameter poly vinyl chloride pipe sections as retreat sites. A shifted, 12 hour day photoperiod was maintained so that night began at 12:00 PM (noon). Twice a week the frogs were fed three-week old crickets (Fluker's Cricket Farm, Port Allen, LA) dusted with vitamin powder (Blair's Super Preen Nutritional Supplement, Neon Pet Products, La Mirada, CA). After a period of about two weeks the frogs typically became gravid, which was determined by gently applying pressure to the abdomen to examine for the presence of large, white egg masses.
Snout vent length and weight of gravid frogs was 43.5 +/-5.0 mm (sd) and 8.9 +/-1.0 g, respectively. Hormones were purchased from Sigma-Aldrich company and prepared by dilution in phosphate buffered saline solution (PBS) (138 mM NaCl, 2.7 mM KCl, 1.5 mM KH 2 PO 4 , 8.1 mM Na 2 HPO 4 , pH 7.2) to a concentration of 1 mg/ml for LHRHs and 5 mg/ml for hCG. Stock solutions were stored at -80°C until use. For injections, the stock solutions were further diluted with PBS to a total volume of 100 µl. Gravid females were placed into the corner of a plastic bag to restrain the frog and a 1 ml tuberculin syringe was used to deliver a sub-cutaneous injection into the anterior dorsum. The frogs were then returned to the aquarium in the plastic bag for observation. Injections were done at roughly 12:00 PM so that females would ovulate during the dark phase of the photoperiod. The effect of the hormone administration was assessed the following morning at approximately 8:00 AM. Except as described below, all frogs were injected only once. In vitro fertilization experiments were carried out by mincing the testes from a single frog in sperm dilution buffer (10 mM NaCl, 0.2 mM KCl, 0.1 mM CaCl 2 , 0.1 mM MgCl 2 , 0.5 mM Hepes pH 7.5) and adding this solution dropwise over the tops of the eggs. All animals were handled and experiments performed in accordance with the standards outlined in the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Results and Discussion
Five different peptide hormones were tested for their ability to induce ovulation and egg deposition in E. coqui: mammalian LHRH (Glu1, His2, Trp3, Ser4, Tyr5, Gly6, Leu7, Arg8, Pro9, Gly10), avian LHRH (Gln8), fish LHRH (Trp7, Leu8), D-Ala6, desGly10, ethylamide LHRH, and hCG. Increasing doses of each hormone were used to establish a dose response curve (see table 1). Ovulation was observed in at least one trial with each of the hormones. No ovulation was observed following injection with PBS alone in six gravid animals. Ovulation was observed in two cases using 7 and 33 µg of mammalian LHRH, in which case the frogs deposited four and three eggs, respectively. No ovulation was observed in thirteen other trials using from 3 to 33 µg of mammalian LHRH. Ovulation was observed on one occasion using 28 µg of avian LHRH, in which case the frog deposited six eggs. No ovulation was observed in twelve other trials using from 3 to 33 µg of avian LHRH. Ovulation was also observed in two cases using 7 and 20 µg of fish LHRH, in which case the frogs deposited one and five eggs, respectively. No ovulation was observed in fourteen other trials using from 3 to 33 µg of fish LHRH.
In two trials using D-Ala6, desGly10, ethylamide LHRH, the viability of deposited eggs was tested by in vitro fertilization. In the first trial with thirteen eggs, one embryo developed normally and in the second trial using twenty one eggs, three embryos developed and normal froglets were obtained. This indicates that after hormonal stimulation the ovulated oocytes underwent nuclear maturation and acquired a functional jelly coat after passage through the oviducts.

Conclusions
Although all of the unmodified LHRHs induced some ovulatory activity in E. coqui, none was particularly more effective compared to the others either in terms of percentage of animals that laid eggs or the numbers of eggs deposited. Previous work has indicated that a form indistinguishable from mammalian LHRH appears to control reproductive behaviors in Xenopus [18], and mammalian LHRH has also been shown to induce ovulation in Rana catesbeiana and Rana temporaria [14,15]. From the results of this study, it is not obvious that mammalian LHRH is the important form controlling ovulation in E. coqui. It is therefore possible that E. coqui utilizes a similar, but distinct LHRH to control reproduction, but what this form might be is not clear. Although it is most similar to mammalian LHRH, the D-Ala6, desGly10, ethylamide LHRH derivative possesses several modifications that have been shown to both increase the receptor binding affinity and the pharmacological half-life of the compound [19]. This results in high activity in ovulation assays in mammals and fish that correlates with our observations of high activity in E. coqui [19,21]. The number of eggs deposited in response to the D-Ala6, desGly10, ethylamide LHRH derivative is comparable to our previous observations of an average of 23 eggs per clutch laid during natural mating events in this species [20]. The lack of consistent function of hCG in E. coqui is not altogether surprising. Although hCG functions well to induce ovulation in Xenopus, it does not consistently induce ovulation in many other amphibian species [10] and although hCG did stimulate ovulation in E. coqui in a number of cases, it also produced severe hemorrhaging and other side effects at high doses. These results provide a method for induction of ovulation and egg deposition in E. coqui that can be used for further studies of the reproductive biology in this species. It will be of interest to see if other Eleutherodactylus species respond to these reproductive hormones in a similar fashion.

Author's Contributions
SFM conceived of the study, participated in the design and coordination of the study and drafted the manuscript. CB, ET, ARE, and SV carried out the collection of the frogs, husbandry and hormone injections. All authors read and approved the final manuscript.