Background
Amphibians are facing extinction rates that exceed those of birds, mammals, or reptiles [
1]. Since 1980, at least 165 species of amphibians have gone extinct and approximately 30% of the amphibians assessed by the International Union for Conservation of Nature (IUCN) are currently threatened with extinction [
1‐
3]. Assisted reproductive technologies (ART) have been increasingly implemented in captive-breeding settings to augment amphibian populations. Many ART protocols for amphibians combine traditional breeding with laboratory methods (e.g., the use of hormones) for successful breeding and induction of viable oocytes and sperm for cryopreservation and/or in vitro fertilization [
4,
5]. Historically, ART protocols used sperm that were collected from testes macerates, thus sacrificing males [
6‐
8], however, within the past 20 years, exogenous hormones have been used to obtain sperm non-invasively [
9]. To date, at least 12 species of anurans representing seven families have been bred using ART [
10].
One of the most common and effective hormones used in amphibian reproductive technologies is gonadotropin-releasing hormone (GnRH, known in the ART literature as luteinizing-hormone releasing hormone [LHRH]; [
11]). Although LHRH has proven useful in ART in anurans, the dose of hormone required to induce spermiation or ovulation and the latency to spermiation varies by orders of magnitude among frog families [
10]. For example, a dosage of 0.28–0.58 μg/g LHRH is sufficient to promote spermiation in leptodactylids [
12], whereas a 2 μg/g LHRH dose is more effective for myobatrachids [
13]. Further, the dose for closely related species within a genus can vary [
9], as can the optimal time post-hormone administration for collection of gametes [
10].
Currently, hormonal ART protocols are generally developed and optimized for species, not populations, regardless of how genetically or geographically isolated the source populations may be. However, population-level differences may result in variable responses to exogenous hormones that may mask or limit the efficacy of a protocol. To our knowledge, there has been no attempt to understand whether the response to hormones varies among populations within a species. We thus assessed the spermiation response to LHRH in highly divergent intraspecific populations of Red-eyed Treefrogs (Agalychnis callidryas).
The Red-eyed Treefrog is a broadly distributed Neotropical phyllomedusine frog [
14]. This species exhibits genetic and phenotypic divergence across populations [
15‐
17], including strong differentiation in color pattern and body size [
17], skin antimicrobial peptides [
18], and advertisement calls (unpubl. data). Allopatric populations show assortative mating for local males, indicating that some premating behavioral reproductive isolation has evolved [
19]. In addition, offspring of individuals from parapatric or allopatric crosses have lower fitness than parental crosses (unpubl. data). However, genetic analyses using nuclear and mitochondrial DNA indicate that these populations represent a single intraspecific lineage [
15,
16].
Red-eyed Treefrogs are well suited for testing the hypothesis that the extensive phenotypic and genetic divergence among populations is associated with population-level physiological differences. We used exogenous LHRH to test the effects of hormone dosage and time post-injection on sperm viability and count. We predicted that if reproductive physiology has shifted with other divergent traits, ejaculate characteristics would vary among populations and over time.
Discussion
We found that exogenous hormone LHRH was effective in producing viable sperm from all populations of Red-eyed Treefrogs in our study. Collectively, males produced sperm in response to the hormone LHRH in 100% of all dose trials. For successful in vitro fertilization, a concentration of 1 × 10
5 motile sperm/ml is needed to fertilize a clutch of 100–200 eggs [
5]. Female Red-eyed Treefrogs in laboratory settings produce 50 to 200 eggs per clutch (unpubl. data). Therefore, all populations in this study produced sperm count well above this baseline, and most produced over 70% viable sperm (Fig.
2).
We detected variation among populations in sperm viability in response to LHRH. Sperm produced by La Selva frogs showed reduced viability compared to all other populations. In addition, more La Selva individuals failed to produce spermic urine than individuals from any other population: 33% (2/6) of La Selva males injected with 2 μg/g LHRH in dose-dependence trials, and 57% (4/7) of males in time-dependence trials failed to produce any sperm after 3 h, although the differences among populations were not significant (cases of non-production of spermic urine: Bijagual: 0.17% (1/6) in response to 4 μg/g, Pavones: 0.20% (1/5) in response to 2 μg/g, Gandoca: 0% (0/6) in response to 2 or 4 μg/g).
Individual and population-level responses to LHRH differed between the two experiments. For example, Bijagual produced higher sperm count in time trials relative to dose-dependence trials, while the La Selva population showed the opposite pattern. Two males per population demonstrated vastly different sperm viability between the dose-dependence trials and the time-dependence trials: for example, one Bijagual male produced 0.7% viable sperm in one trial and 90% viable sperm in a later trial at the same dosage and time point (4 μg/g and 3 h PI). High within-population variability in ejaculate characteristics has been reported across taxa [
25‐
27]. Within amphibians, individual variation in sperm count in response to injections of LHRH was observed to be 44-fold within a population of Peron’s Treefrog (
Litoria peronii) [
22]. Fewer studies have investigated among-population variation in sperm. Hettyey and Roberts [
28] found that sperm quality (longevity and motility), size, and sperm concentration collected from testis macerates varied within and among non-divergent breeding populations of
Crinia georgiana. We are not aware of any other studies investigating interpopulation variation in sperm traits in isolated and potentially divergent populations and/or in response to exogenous hormones. More research is thus called for to reveal how population-level differences may impact the development of ART protocols for rare or threatened species that occur in isolated populations.
As with many ART studies, several caveats apply to our work. Our sample sizes were low due to permit limitations in the number of animals we were able to collect. However, despite this constraint, we detected significant differences in sperm viability in our dose-dependence trials and a significant reduction in viability over a 24 h period in our time-dependence trials. While we used the same males in both experiments, we consider this unlikely to have affected our results based on previous work with amphibians [
11] and based on the fact that sperm count did not drop over the course of the experiment; indeed, one population demonstrated an increase in average sperm viability between experiments. In addition, studies on toads (
Bufo fowlerii, Bufo americanus, & Bufo valliceps) demonstrate that repeated injections continued to be effective with no obvious declines in sperm number, viability or motility [
5,
24].
We did not test for sperm quality among populations, nor did we examine genetic compatibility in an in vitro fertilization setting, important factors in validating ART protocols. For example, within Bufonidae,
Anaxyrus americanus produced viable sperm of good quality in response to the hormone hCG. While a closely related and rare species,
Anaxyrus baxteri, produced similar sperm counts and viability estimates in response to the same dosage of hCG, sperm also had abnormal heads [
12]. In addition, in vitro studies using isolated populations of
Pseudophryne bibronii showed evidence of high embryo mortality in outcrossed clutches indicating genetic incompatibility among populations [
22]. Examining genetic compatibility, sperm quality and in vitro fertilization success are the next steps toward a conservation management plan that involves ART.
Conclusion
To our knowledge, this study presents the first examination of hormone efficacy at inducing spermiation among highly divergent populations of an anuran. The hormone LHRH was effective at inducing spermiation and we recommend for baseline ART protocols a dosage range of 2–4 μg/g LHRH at a time point of 3 h post administration in this species, which may have broader applications for other phyllomedusines. We detected variation both within and among genetically and phenotypically divergent populations.
Documenting temporal variation, peak sperm production and viability in response to hormones are critical components of ART protocols such as cryopreservation of gametes or in vitro fertilization techniques. Sperm count and viability are highly variable over time among species and families in response to hormones [
22] and one study found high intraspecific variation among populations that did not show genetic or phenotypic divergence across their range [
28]. Taken together, these findings indicate that an intraspecific lineage is not enough to justify the assumption that a single ART protocol fits all, but rather that protocols should be viewed as a starting point if populations are highly differentiated and/or isolated. In addition, many ART protocols use non-threatened, closely related, species as a proxy and then test on rare species, but because rare species are often represented by small, isolated populations that may possess unique genotypes, protocol optimization should be adapted to consider and document differences in source populations [
12,
23]. ART has been successfully applied to mammals, cephalopods, reptiles, birds and amphibians [
5,
29‐
33]. Thus, these findings inform conservation and breeding management of threatened or isolated populations spanning a range of diverse taxa.
Acknowledgements
We extend a special thank you to T. Uhlendorf and the CSUN Vivarium team for animal care and husbandry. We thank the Ministerio de Ambiente, Energia y Mares de Costa Rica (MINAE) for research and export permits. We thank A. Vega, M. Akopyan, C. Owen and N. Savant for the collection of
Agalychnis callidryas in the field. We thank R. Espinoza and D. Gray for suggestions and discussion that greatly improved the manuscript. ASTER Costa Rica image with population ranges was retrieved from
https://lpdaac.usgs.gov, maintained by the NASA EOSDIS Land Processes Distributed Active Archive Center (LP DAAC) at the USGS/Earth Resources Observation and Science (EROS) Center, Sioux Falls, South Dakota. The data product for the image was provided by NASA.