Southern stingrays (Hypanus americanus, formerly Dasyatis americana) are slow to mature and produce small numbers of offspring,1 making them susceptible to population threats that may include perturbation of coastal breeding and pupping grounds and unregulated commercial and recreational fishing. Although the impact of these threats on the long-term survival of the species is unknown,2 aquarium facilities in the United States are striving to develop sustainable breeding populations of elasmobranchs such as stingrays for both education and scientific study. Characterization of the reproductive biology of a species is the first step toward understanding the requirements for conservation and long-term survival under managed care, yet little is known about the natural reproductive biology of H americanus, with most available reproductive data derived from aquarium-housed animals.3
In the authors' experience, reproductive disease has been identified in groups of female southern stingrays housed at multiple aquaria, suggesting that this is a widespread problem and necessitating the development of descriptive criteria and diagnostic tools to distinguish between healthy and abnormal reproductive physiology. This perceived high incidence of disease is in contrast to findings in a postmortem survey4 involving tissues of aquarium-housed elasmobranchs, in which only 3% (2/75) of southern stingrays had reproductive disease.
Information regarding reproduction in southern stingrays is limited, as are descriptions of the histologic and anatomic characteristics of the reproductive tract.5,a Observations of wild and aquarium-housed stingrays indicate that mating can occur immediately after parturition3; therefore, female southern stingrays likely remain in a gravid state for most of their life span. Aquarium-housed stingrays can reproduce biannually.3 To manage the potential overabundance of offspring, this species is often housed in single-sex groups, and as a consequence, females remain in a nonreproductive state for extended periods. Other species, notably mammals that are maintained in a protracted nonreproductive state, develop a range of abnormal reproductive conditions.6,7 Additionally, high circulating concentrations of estrogen have been linked to endometrial hyperplasia,8,9 so it is hypothesized that the prolonged estrogen-dominant environment in non-breeding females may have adverse effects on reproductive physiology. Indeed, in teleosts, reproductive impairment can occur with exposure to exogenous environmental estrogens.10,11
The hormonal control of the reproductive cycle has been described for few elasmobranchs relative to nonmammalian vertebrates.12 Although in vitro research has shown that E2, P4, and T5 are produced by elasmobranchs, these steroid hormones do not always appear to regulate reproduction in classically expected ways.13 For example, serum P4 concentration does not increase during early or midgestation, but serum free P4 (and T5) concentration peaks just prior to parturition in Atlantic stingrays,14 and no P4 is detectable in granulosa tissue throughout gestation in spiny dogfish (Squalus acanthias).15 If P4 is not the functional gestagen in sharks and rays, then another progestational precursor or metabolite may perform that role, if not a different steroid hormone altogether.16,17 Other P4 metabolites have been identified in elasmobranchs as well as the estrogen metabolites E1 and, in limited species, estriol.18,19 One or more of these may be better analytes for characterization of reproductive physiology in elasmobranchs. Whereas E2 is the prime regulatory steroid, E1 has a strong role in reproduction as well20 and is worthy of further investigation. Furthermore, estrogens in general are assigned a more integral function in cyclicity and viviparity over progestogens.21
Radioimmunoassay kits are most commonly used for hormone evaluation owing to their high specificity with respect to primary reproductive steroid hormones (E2, P4, and T5).22 However, EIAs and chemiluminescence immunoassays are also used for monitoring reproductive hormones of various species and have an environmental advantage over radioimmunoassays by eliminating radioactive waste. Although EIAs and chemiluminescence immunoassays have been reported for use in sharks,23,b they have not been developed specifically for or validated in stingrays. The ability to examine potential links between reproductive hormones and reproductive abnormalities in southern stingrays would aid development of treatment protocols and may aid identification of the underlying problem and disease prevention.
The current hypothesis is that reproductive disease in single-sex aquarium-housed southern stingray populations is a function of a persistently high estrogen and low P4 physiologic environment in animals without successive pregnancies.24 The goals of the study reported here were to review and supplement existing information by characterizing normal ovarian and uterine anatomy and reproductive abnormalities in southern stingrays by use of ultrasonography and to measure plasma E2, E1, P4, and T5 concentrations by means of EIAs to examine associations between hormone concentration and reproductive abnormalities.
Materials and Methods
Animals and management
Live animals used in the study were obtained to represent 3 types of populations: aquarium housed, lagoon housed (managed semiwild), and free ranging (wild). Aquarium-housed animals included 48 mature (disk width ≥ 75 cm or > 6 years of age25,26) nonpregnant female southern stingrays that were kept in single- or mixed-sex groups or individually at 9 facilities (Walt Disney World Parks & Resorts, The Florida Aquarium, Georgia Aquarium, The New England Aquarium, North Carolina Aquarium at Fort Fisher, Minnesota Zoo, Ripley's Aquarium of the Smokies, Jacksonville Zoo and Gardens, and SeaWorld Florida Theme Park; Appendix 1). Managed stingrays were fed a variety of fish and shellfish (capelin, clam, shrimp, herring, and smelt) and received an elasmobranch vitamin supplementc 1 to 3 times/wk.
Lagoon-housed animals included 34 hand-caught semiwild but managed female southern stingrays from a group that were housed with males in an enclosed ocean bay (Castaway Cay, Bahamas) under natural environmental conditions. Wild animals included 12 wild adult female southern stingrays that had been hand caught and net corralled off the coast of North Carolina (n = 2) or in Bimini, Bahamas, under protocols from the Bimini Biological Field Station Foundation (10). Both lagoon-housed and wild southern stingrays were offered a daily diet consisting of shrimp, squid, and an aquatic gel product.d No pregnant females were included in either cohort.
In addition to the aforementioned live stingrays, 10 wild mature female specimens (disk width, 87 to 108 cm) that were collected as a part of extant fish surveyse were used to characterize the gross anatomy of the healthy female reproductive system. Three stingrays from 1 facility in which reproductive disease was identified on postmortem examination were used for comparison and description of reproductive disease.
Animal care and use in this project was approved by the animal care and welfare committee at Disney's Animals, Science & Environment. The Department of Fisheries of the Commonwealth of the Bahamas granted a permit to conduct scientific research (No. MAF/LIA/22).
Data collection
Disk width measurements, blood samples for plasma hormone concentration measurement, and ultrasonographic data were collected from the included stingrays at various points over a total period of 1 year (Appendix 1). Examination months were staggered throughout the year among the institutions to allow for data collection during each month of the year to investigate potential seasonal changes. Aquarium-housed stingrays also received a physical examination, blood sample collection, and ultrasonographic examination quarterly.
For all stingrays, disk width was obtained ventrally by measuring the span between the tips of the widest portion of the pectoral fins. Blood samples were collected from the wing or tail vein and placed into tubes containing heparin,f and plasma was separated and stored frozen (−20°C) until analysis. Physical examination, blood sample collection, and ultrasonography were performed with stingrays manually restrained or anesthetized by immersion in salt water with tricaine methane sulfonateg (50 to 55 mg/L) that was buffered 1:1 (wt/wt) with sodium bicarbonate. Females were positioned in dorsal or ventral recumbency, and ultrasonographic images were collected. Although multiple ultrasound units and transducersh–j were used, the staging criteria were not affected by any variances between machines. Video images were recorded in the transverse and sagittal planes for the ovary and in the transverse plane for the uterus for later review.
Ultrasonographic scoring
Ultrasonographic recordings were reviewed to examine the single functional ovary (located dorsolateral to the liver and cranial to the uterus on the left sidea) and uterus. The presence or absence of ovarian disease was scored on a 5-point scale on the basis of the width of the ovary, the number of follicular layers, the presence of cystic structures, and whether the ovary overlapped the uterus (Appendix 2). Care was taken to ensure that, for complete ovarian assessment, the full ovary was viewed and scanned to avoid misinterpretations that can occur with partial or still images. Scoring of the uterus was performed by assessing the degree of intrauterine fluid and debris accumulation as well as the general size of the uterus, which was determined by considering its proximity to midline or the valvular intestine (also known as spiral intestine or spiral valve; Appendix 3). The uterus is a muscular and dynamic structure, compared with the ovary, and therefore rapid changes (eg, expulsion of histotroph on handling) were accounted for and considered during the scoring process.
EIA
Because of the potential for plasma matrix effects to interfere with the EIAs27 and a high affinity of reproductive steroid hormones to elasmobranch steroid-binding proteins,28 plasma samples were first pretreated with a commercially available dissociation reagentk to provide a measure of the total steroid hormone concentration.29 Equal volumes of 1× dissociation reagent were added to plasma samples for P4 and T5 assays, and an equal volume of 50%-strength dissociation reagent in EIA buffer (0.04M NaH2PO4, 0.06M Na2HPO4, 0.15M NaCl, and 0.1% BSA; pH, 7.0) was added for E2 and E1 assays and then vortex mixed for 5 seconds. Serial dilution of these pretreated plasma samples in EIA buffer yielded curves parallel to the standard curves for all 4 steroid hormones tested (all R2 > 0.997) with the following percentage recoveries of spiked hormone: E2, 95%; E1, 100%; P4, 94%; and T5, 108%. For final analyses, samples were further diluted with EIA buffer to a final dilution of 1:50 for P4 and 1:20 for E2, E1, and T5 and assayed within 4 hours after pretreatment with dissociation reagent per the assay manufacturer's instructions.
Pretreated samples were assayed for total E2, E1 (including primary glucuronide and sulfate conjugates), P4, and T5 concentration by use of a competitive double-antibody EIA with polyclonal antisera raised in rabbits against E2 (R0008), E1 (R522-2), P4 (R4859), and T5 (R156/7)30 and corresponding HRP label.l Cross-reactivities for these antisera have been reported.31–34
For all EIAs, polystyrene microtiter plates were precoated with goat anti-rabbit IgGm (10 μg/mL) pretreated with a blocking buffer, blotted dry, and stored at 4°C with desiccant in resealable storage bags. All reagents and plates were brought to room temperature (approx 24°C) prior to use, and 50 μL of standard or sample was added to precoated wells, followed immediately by 50 μL of corresponding HRP conjugate, followed by 50 μL of antibody, excluding 2 blank wells to serve as nonspecific-binding control substances. Plates were incubated for 2 hours at room temperature and then washed 4 times with 0.08% Tween 20n in distilled water. After plates were washed, 100 μL of colorimetric substrateo was added to each well, and the plates were incubated at room temperature until an optical density of 1.0 was reached for the zero-hormone standard substances when read with a microplate reader.p
To account for any residual matrix effects or dissociation-reagent effects on HRP labeling in the EIA, a background correction was determined for each steroid hormone. Plasma was charcoal stripped with heat-activated charcoaln (10 mg/mL), vortex mixed and incubated at 37°C for 2 hours before centrifugation and filtration,35 and treated with dissociation reagentk to generate a background blank for subtraction. Dissociation reagent–treated charcoal-stripped plasma yielded background concentrations of 55, 15, 3, and 20 pg/well at the working dilutions for the E2, E1, P4, and T5 EIAs, respectively. These background concentrations were subtracted from each sample value before calculation of concentration. Respective interassay coefficients of variation for high- and low-concentration controls were 18% and 19% for E2, 16% and 14% for E1, 19% and 18% for P4, and 11% and 13% for T5. Mean ± SD intra-assay coefficients of variation were 2.4 ± 0.1%, 2.2 ± 0.1%, 2.5 ± 0.1%, and 1.8 ± 0.1% for E2, E1, P4, and T5, respectively. Assay sensitivity (the value obtained at 90% to 95% binding) was 4 pg/well for E2 and P4, 0.5 pg/well for E1, and 2 pg/well for T5.
Postmortem assessment of reproductive anatomy
To assess normal reproductive anatomy and to characterize ovarian size, general layers of follicles, and overall appearance, opportunistic postmortem examinations were performed on the 10 wild stingray cadavers. Four cadavers underwent complete pro-section, and the remaining 6 had only the left reproductive tract examined.
Statistical analysis
For females assessed 2 to 4 times within a 1-year period, the mode of the scores for each female was used in calculations of disease prevalence. When an equal number of each of 2 scores was obtained, the higher of the 2 scores was used (ie, individual quarterly scores of 3, 4, 3, and 4 were assigned a mode score 4). Because plasma steroid hormone measurements were not normally distributed, they were rank-transformed before analysis.
Statistical analyses were conducted by use of statistical software.q–t Linear mixed-effects modeling was used to determine whether plasma steroid hormone concentrations were associated with ovarian scores. The model included ovarian score as a fixed factor and an intercept representing each stingray as a random effect to account for multiple nonindependent measurements of lagoon-housed (managed semiwild) and aquarium-housed stingrays. Mixed models were fit by restricted maximum likelihood with degrees of freedom for the error term calculated with the Satterthwaite method. Differences between multiple categories for significant factors were compared through post hoc pairwise comparisons. Similar analyses for uterine score were not completed owing to the small number of stingrays with a uterine score of 4 (n = 7) or 5 (2). Seasonality of plasma steroid hormone concentration was tested by the addition of season as a fixed factor to the mixed models. Maximum likelihood fit was compared between hierarchical models by use of the likelihood ratio test and Bayesian information criterion values.
The Pearson correlation (r) between uterine and ovarian scores and the Spearman (ρ) correlation between hormone pairs were calculated. The relationship between stingray size and ovarian score was analyzed via 1-way ANOVA followed by pairwise comparisons with the Tukey honest significant difference test and computation of Benjamini-Hochberg adjusted P values. The relationship between stingray size and habitat (aquarium housed, lagoon housed, and wild) was similarly analyzed. Values of P ≤ 0.05 were considered significant for all analyses.
Results
Postmortem assessment of normal reproductive anatomy
All 10 wild female stingrays that underwent postmortem examination were gravid. All had similar anatomic features (Figure 1); none had overt gross lesions. In all, the ovary abutted the left lateral surface of the epigonal organ. The pink and uniformly textured epigonal organs differed in size among individuals (in several, reaching the caudal coelom) but always overlapped the uterus on the left side. On the left side of the body, multiple ovarian follicles of various stages were present but rarely exceeded 2 consecutive layers; the width of the ovary did not exceed 4 cm. Individual ova rarely exceeded 2 cm in size and then only by 1 or 2 mm in only 1 specimen. No cysts were present. Mature ova were present regardless of stage of pregnancy; follicle numbers and ova sizes were not evaluated. The ovary did not overlap the uterus in any specimen.
Postmortem photographs of a mature female southern stingray (Hypanus americanus) with an anatomically normal liver covering the reproductive organs (A) and an anatomically normal ovary and gravid uterus (liver removed; B) and another female stingray with an abnormally small liver and enlarged ovary, which is overlying an abnormal uterus. E = Epigonal organ. L = Liver. O = Ovary. St = Stomach. Ut = Uterus. Vi = Valvular intestine. Photographs provided by Jill Hendon, Gulf Coast Research Laboratory, University of Southern Mississippi, Ocean Springs, Miss (A and B), and the Georgia Aquarium, Atlanta (C).
Citation: American Journal of Veterinary Research 80, 10; 10.2460/ajvr.80.10.931
The right oviduct and stroma were visibly attenuated with no ova present; the oviduct was associated with epigonal tissue that was more caudal and smaller in size and volume than the contralateral epigonal tissue. Right and left ostia originated ventral and lateral to the esophagus and led dorsally, then laterally, into the oviducts. The ostia were supported by a ligament (presumed falciform) and were connected to each oviduct. The right oviduct widened into a rudimentary and nonfunctional, vestigial right uterus; patency to the right common duct or urogenital pore was not evaluated. The left oviduct expanded into the nidamental gland (also known as the oviducal or shell gland) at the cranial end of the uterus. There was no obvious isthmus.
On the left side of the body, the uterus was attached dorsally to the kidney and widened cranially to the nidamental gland. The luminal surface was lined by a variably thick, pink to white villous mucosa, the trophonemata (uterine villi).36,37,a No cysts were noted in the uteri. The uterus terminated into the urogenital sinus via an area that had dense connective tissue and did not easily distend; this area, analogous to the mammalian cervix, was lateral to the single urinary (urogenital) pore. The sinus then led to the cloaca, but the borders between these were difficult to see. On the right side of the body, uterine tissue was not discernible from the oviduct.
Description of reproductive disease
All 3 stingray cadavers in which reproductive disease was identified on postmortem examination had a swollen coelom manifesting dorsally with unilateral or bilateral bulges prior to death (Figure 2). Other premortem findings included signs of anorexia and lethargy.
Postmortem photographs of a mature female southern stingray with externally visible dorsal coelomic bulging of reproductive organs (arrows; A) and internally visible (dorsal recumbency) cystic vestigial uterus (Vc) and enlarged, histotroph-filled uterus (Ut; B). This coelomic bulging was consistent with severe reproductive disease. Photographs provided by the Georgia Aquarium, Atlanta.
Citation: American Journal of Veterinary Research 80, 10; 10.2460/ajvr.80.10.931
Postmortem examination revealed that the ovary was enlarged, congested, and misshapen with marked amounts of variably sized follicles. Although such findings might be interpreted as normal follicular activity, the number of follicles and the variable, often large amounts of ova suggested advanced disease. All these follicles were friable and easily broken on handling. The ovarian tissue contained cystlike structures. In 1 stingray, oviducts and ostia were bilaterally cystic, with the ostia having evidence of gross glandular proliferation. The epigonal organs were enlarged and congested in 2 stingrays.
In all 3 stingrays, the uterus was enlarged in length and width (Figure 1), and the uterine wall was thickened; in 1 stingray, the uterus had cavitating cysts. In 2 stingrays, the trophonemata were reddened. The uterine horns were filled with thick, opaque, white to tan, and (in 2 stingrays) pink-tinged fluid that was presumed histotroph5 even though none of the 3 were pregnant. In 2 stingrays, the vestigial right uterus had a large or multiseptated coalescing cyst that was considered analogous to cystic endometrial hyperplasia.
Ultrasonographic scoring
In stingrays assigned scores of 2, there was a single layer of uniform follicles and occasional smaller follicles in the periphery of the ovary (Figure 3). In contrast, in stingrays with a score of 3, the ovary appeared wider with multiple layers and various sizes of follicles and occasional cysts. Stingrays with ovarian scores of 4 or 5 had greater numbers of different-sized follicles with heterogeneous echogenicity (presumed to be indicative of varying stages of vitellogenesis) and degeneration and cystic structures. Ovarian scores generally did not change dramatically within individuals over the course of the study.
Representative ultrasonographic images of the ovary (A, C, E, G, I) and uterus (B, D, F, H, J; midline indicated [arrow]) in southern stingrays (positioned in dorsal recumbency) with ultrasonographic ovarian or uterine scores of 1 (A and B), 2 (C and D), 3 (E and F), 4 (G and H), and 5 (I and J). The ovary is outlined in panels A and B, and 2 ovarian cysts are visible in panels G and I (asterisks). E = Epigonal organ. St = Stomach. T = Trophonemata. Uf = Uterine fluid. Vi = Valvular intestine.
Citation: American Journal of Veterinary Research 80, 10; 10.2460/ajvr.80.10.931
No semiwild or wild stingrays were identified with ovarian scores of 4 or 5, and only 1 of the 12 had an ovarian score of 3. In contrast, 83% of aquarium-housed stingrays had an ovarian score ≥ 3 (Table 1), and 66% had advanced or severe disease (score of 4 or 5). Many stingrays with ovarian scores of 4 or 5 had other abnormal structures identified, namely a cystic vestigial right ovary and numerous cysts within the oviduct. For female stingrays in breeding situations, an ovarian score of 3 that included some cystic structures did not necessarily indicate that the female was infertile. One female with an ovarian score of 4 also became pregnant during the study. Given these findings, it was determined that although retained follicles and cystic or degenerate structures may be present, a score of 3 may represent an intermediate pathological phase that might be resolved if the cause of the underlying disease were better understood. In contrast, ovarian scores of 4 and 5 were deemed indicative of a health problem that warranted veterinary intervention. Stingrays with an ovarian score of 5 required regular veterinary intervention; nevertheless, some died of complications from the disease.
Number (%) of southern stingrays (Hypanus americanus) from aquarium-housed, lagoon-housed (managed semiwild), and wild habitats with various ultrasonographic ovarian scores.
| Population | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| Aquarium housed (n = 48) | 2 (4) | 7 (15) | 8 (17) | 15 (32) | 16 (34) |
| Managed semiwild (n = 34) | 12 (35) | 22 (65) | 0 (0) | 0 (0) | 0 (0) |
| Wild (n = 12) | 0 (0) | 11 (92) | 1 (8) | 0 (0) | 0 (0) |
See Appendix 1 for numbers of measurements and Appendix 2 for score descriptions.
In contrast to the ovary, the uterus was a more dynamic organ with variation in scores due to the stingray's ability to produce rapid changes in uterine fluid volume. The degree of abnormal uterine anatomy was correlated with ovarian score (r = 0.77; P < 0.001), and females with advanced ovarian disease also had an abnormal uterus, often with exuberant production of histotroph causing a grossly enlarged uterus that engulfed the coelomic cavity (Figure 3). None of the wild or semiwild stingrays had an abnormal uterus (score of 4 or 5) on ultrasonography but 3 of 12 wild stingrays had an intermediate score of 3 (Table 2). In contrast, 20% of aquarium-housed stingrays had an abnormal uterus, whereas 58% had an intermediate score of 3.
Disk width
Disk width differed significantly (P < 0.001) among aquarium-housed (mean ± SD, 111.7 ± 19.8 cm), semiwild (83.1 ± 9.7 cm), and wild (83.1 ± 9.7 cm) stingray groups. Specifically, semiwild and wild stingrays were smaller than aquarium-housed stingrays (P < 0.001) but did not differ from each other in size (P = 0.28). As disk width increased, so did the severity of ovarian disease (P < 0.001). Disk widths for stingrays with ultrasonographic ovarian scores of 3 (mean ± SD, 99.3 ± 14.9 cm), 4 (112.5 ± 15.8 cm), and 5 (124.9 ± 12.6 cm) differed significantly from each other as well as those for stingrays with scores of 1 (80.6 ± 12.6 cm) and 2 (85.6 ± 8.9 cm). However, no significant (P = 0.20) difference in disk widths was identified between stingrays with ovarian scores of 1 or 2.
Plasma steroid hormone concentrations
Overall, plasma total E2 and P4 concentrations were below the EIA detection limit for 49% (n = 93) and 22% (41) of samples (n = 188), respectively. Considering samples from the same population (aquarium housed [n = 135], managed semiwild [41], and wild [12]), most (68% [28/41]) managed semiwild stingrays had E2 concentrations below the detection limit, compared with 46% (62/135) of aquarium-housed and 25% (3/12) of wild stingrays. Plasma P4 concentration was below the detection limit for most (9/12) wild stingray samples, compared with 44% (18/41) of semiwild stingray samples and 10% (14/135) of aquarium-housed stingray samples. Conversely, plasma total E1 (or its primary conjugates) and total T5 concentrations were detectable in ≥ 96% of samples.
Plasma E2, E1, and P4 concentrations increased with increasing ovarian scores. Stingrays with ovarian scores of 4 or 5 had significantly higher concentrations of these steroid hormones than those with lower scores (Figure 4). Plasma E1 concentration also differed significantly among scores 1, 2, and 3. For stingrays that had a detectable E2 concentration and a paired E1 value (n = 85), plasma E1 concentration was a mean of 2.5 times as high as the E2 concentration and was equal to or higher than the E2 concentration in 88% (75/85) of those samples. Generally, higher E1 values were indicative of ovarian disease, but there were several exceptions to this, whereby approximately 20% (5/24) of ovarian scores of 5 corresponded with plasma E1 concentrations < 10 ng/mL and approximately 20% (18/87) of scores from 1 to 3 corresponded with concentrations > 10 ng/mL. Conversely, no differences in plasma T5 concentration were observed among ovarian scores.
Box-and-whisker plots of plasma E2 (A), E1 (B), P4 (C), and T5 (D) concentrations versus ultrasonographic ovarian score in female southern stingrays (48 aquarium housed, 34 semiwild, and 12 wild). Boxes represent the interquartile (25th to 75th percentile) range, the central horizontal line represents the median, whiskers represent the data points farthest from the median that are not outliers (ie, that are within 1.5 times the interquartile range of the first and third quartiles), and dots represent outliers. Note that the scale of the y-axis differs among hormones.a,b Scores with different letters differ significantly (P ≤ 0.05). See Appendix 1 for numbers of measurements and Appendix 2 for score descriptions.
Citation: American Journal of Veterinary Research 80, 10; 10.2460/ajvr.80.10.931
The addition of season as a fixed factor in the mixed model to predict ovarian score resulted in a significant likelihood ratio test result for E2 and E1, but the Bayesian information criterion value increased for all steroid hormones, and therefore the most parsimonious model (without season) was retained. No statistical testing was performed for plasma hormone concentration by uterine score because only 2 stingrays had high uterine scores; however, a similar pattern of increasing plasma concentration was observed for E2 and E1, but not P4, and a uterine score of 5 corresponded with a higher T5 concentration than observed for the other scores (Figure 5).
Box-and-whisker plots of plasma steroid hormone concentrations versus ultrasonographic uterine score in the stingrays represented in Figure 4. See Figure 4 for remainder of key.
Citation: American Journal of Veterinary Research 80, 10; 10.2460/ajvr.80.10.931
Discussion
The present study provided the first description of the normal ultrasonographic anatomy of the reproductive system and of reproductive disease in mature female southern stingrays. Among aquarium-housed stingrays, 64% had an enlarged ovary (scores of 4 or 5) as assessed via ultrasonography. In contrast, only 1 of the wild and semiwild stingrays received an ovarian score of 3, and none had a higher score. Severity of reproductive disease was positively correlated with plasma total E2, E1, and P4 concentration, and plasma total T5 concentration was high in stingrays assigned the highest uterine score. Although severity of ovarian disease increased significantly with disk width in the study, additional data are needed regarding wild (presumed normal) stingrays of similar (larger) size for a better assessment.
To date, stingray reproductive hormones have predominantly been measured by means of radioimmunoassay with antisera targeting the parent steroid hormones E2, P4, and T522 following solvent extraction.38 The presence of an elasmobranch steroid-binding protein previously identified in dogfish (Scyliorhinus canicula) with a high serum protein affinity for E2, P4, T5, and corticosterone,28 which has been described as having an “almost unlimited capacity,”39 may make the measurement of free versus total reproductive steroid hormones more difficult in southern stingrays.40 Measurement of plasma steroid hormone (E2, E1, P4, and T5) concentrations by EIA was possible following sample pretreatment with a dissociation reagent; therefore, it was likely that the southern stingrays had a similar protein that may impact free steroid EIA measurement. Liberating hormone from carrier globulins results in the measurement of total hormone rather than the free or bioavailable circulating hormone.41 Still, there is justification that measurement of plasma total steroid concentrations may be a biologically relevant predictor in animals with reproductive disorders and disease states.40–42 Further optimization of the E2 assay used in the present study is warranted. The investigation of plasma steroid hormone concentration as a potential diagnostic test and linkage with the cause of underlying disease could aid in the determination of appropriate hormone treatments for stingrays with reproductive disease.
Regular pregnancy cycles with normal hormone fluctuations are presumed to prevent accumulation of ova in southern stingrays, and unbred females are presumed to be at higher risk of ovarian disease than those that have had a recent pregnancy. The result can be sustained vitellogenesis43,44 and lack of ovulation. A persistent estrogen-dominant environment has been linked to cystic ovaries and endometrial hyperplasia in terrestrial mammals.6,7,24 In teleosts exposed to E2 in 1 study,45 vitellogenesis was prolonged and intermittent doses resulted in sustained plasma vitellogenin concentrations. Additionally, ova produce estrogens in elasmobranchs,46–48 so an increase in the number of ova may create a more estrogen-rich environment. Vitellogenin production is stimulated by E2 in the aplacental viviparous spiny dogfish,49 oviparous little skate (Raja erinacea43), and aplacental viviparous marbled electric ray (Torpedo marmorata50); therefore, it is likely that the mechanism is conserved across elasmobranchs and that southern stingrays respond likewise. Consequently, in the aquarium-housed stingrays of the present study, estrogens may be posited to have a vitellogenic role, as evidenced by the observed increase in plasma E2 and E1 concentrations with increasing ovarian score. Although EIAs are available to measure serum vitellogenin concentration in some teleost species,51 these assays have not been validated for use in elasmobranchs and may be considered for future research.
In Atlantic stingrays (Hypanus sabinus, formerly Dasyatis sabina), E2 secretion increases dramatically during the period when histotroph is produced.14,52 In the present study, plasma E2 concentration was high only for stingrays with the highest uterine score, whereas plasma E1 concentration increased progressively with increasing uterine score. Interestingly, plasma T5 (a precursor hormone to E2 and E1) concentration was high only in stingrays with the highest uterine score. It is possible that T5 in southern stingrays acts as a precursor to a hormone involved with histotroph production, but in the present study, plasma E2 concentration was correlated with plasma T5 concentration, whereas plasma E1 concentration was not. Additionally, in H sabinus, secretion of T5 increases just prior to parturition,14 so a high circulating T5 concentration may play a role in the abnormally large histotrophfilled uterus. Aside from this finding, plasma T5 concentration provided no additional information about reproductive disease in the assessed southern stingrays. Biosynthesis research involving small-spotted catsharks (Scyliorhinus canicula L) and spiny dogfish revealed that T5-glucuronide, E2, and E1 were produced in ovarian stroma and follicular membranes of maturing ova following incubation with radiolabeled androstenedione,53 but no estrogens were detected following incubation of ova with radiolabeled T5.54 Additional investigation is warranted using immunoassays that target reproductive steroid metabolites in elasmobranchs and teleosts (eg, 20α-hydroxyprogesterone found in Squalus spp19).
To the authors' knowledge, the study reported here represented the first in which an E1 EIA was used in elasmobranchs. Findings indicated that plasma concentrations of total E1 in female southern stingrays were higher than those of total E2 and, furthermore, that this steroid hormone may be a more consistent steroid signal to monitor for indication of ovarian and uterine disease. Conversely, plasma E2 concentrations (as measured via the reported method) were mostly low or undetectable in female stingrays, regardless of whether they had a healthy or diseased reproductive system. Additionally, E1 reportedly stimulates vitellogenesis in teleosts, sometimes in synergy with E2 with both contributing to the total estrogen synthetic capacity of an organism.20,55 Thus, it seemed prudent to examine both hormones and their primary circulating metabolites simultaneously.56 In the present study, both E2 and E1 were present in plasma samples from female southern stingrays and concentrations increased with increasing severity of reproductive disease.
The finding that reproductive disease was correlated with disk width in the southern stingrays was not surprising given that stingray size is correlated with age57 and age may contribute to reproductive disease. Additional research is warranted to examine body condition and size of stingrays as factors that may contribute to disease as well. It should be noted that the wild and semiwild stingrays in the present study were smaller than aquarium-housed stingrays. Additionally, no effect of season was identified on findings for the aquarium-housed stingrays, and this may have reflected the time it takes for reproductive disease to progress, which in the authors' experience is several years. Although geographic location and environment were not examined as part of the present study, those factors warrant further evaluation as well.
Reproductive disease appeared to be a serious problem with a high prevalence in the aquarium-housed female southern stingrays of the present study. The combination of a large ovary, large uterus, and other cystic structures as previously described presumably result in anorexia and lethargy owing to prevented expansion of the stomach, induced discomfort, or other factors. Additional research is suggested, including thyroid hormone testing and evaluation of the role of iodine (a key nutritional component), body condition, size, and age in the development of reproductive disease. Although we identified some normal features of the reproductive system in mature female southern stingrays, a further understanding of the normal, seasonal reproductive patterns of estrogens, P4, and T5 and their circulating metabolites in all age and size classes would be important to elucidate mechanisms of reproductive disease in this species. Overall, data obtained through the present study may assist in the development of treatment strategies and, ideally, prevention of reproductive disease in southern stingrays (eg, via stimulation of ovulation or inhibition of vitellogenesis and its associated estrogen production).
Acknowledgments
Supported in part by the Bahamas Department of Marine Resources, National Institute of General Medical Sciences (grant No. P20GM104932), Chemistry Research Core of the Center of Research Excellence in Natural Products Neuroscience, and Morris Animal Foundation (grant No. D14ZO-804).
The authors declare that there were no conflicts of interest.
Presented in part in abstract form at the American Association of Zoo Veterinarians Conference, Orlando, Fla, October 2014, and Atlanta, July 2016, as well as the 46th Annual International Association for Aquatic Animal Medicine Conference, Chicago, April 2015.
This report represents a SeaWorld technical contribution (No. 2018-21).
The authors thank Charlene Burns, Tonya Clauss, Alexa McDermott Delaune, Michelle Davis, Charles Innes, and Robert George for assistance with the study; Cayman Adams for assay development and sample analysis; Carlos Rodriguez for assistance with pathological descriptions; Mandi Schook for statistical guidance; and Jill Hendon for help with collection of and access to wild animals.
ABBREVIATIONS
| E1 | Estrone |
| E2 | 17β-estradiol |
| EIA | Enzyme immunoassay |
| HRP | Horseradish peroxidase |
| P4 | Progesterone |
| T5 | Testosterone |
Footnotes
Schwert HL. The reproductive system of the female sting ray, Dasyatis americana. MSc thesis, Zoology Department, Cornell University, Ithaca, NY, 1967.
Anderson BC. Non-lethal methods for assessing reproductive status of bonnethead sharks (Sphyrna tiburo). MS thesis, Department of Biology, University of North Florida, Jacksonville, Fla, 2015. Available at: digitalcommons.unf.edu/etd/605. Accessed Apr 4, 2019.
Mazuri Vita-Zu Shark/Ray, Mazuri Exotic Animal Nutrition, St Louis, Mo.
Aquatic gel 57W9, Mazuri Exotic Animal Nutrition, St Louis, Mo.
Provided by Jill Hendon, Gulf Coast Research Laboratory, University of Southern Mississippi, Ocean Springs, Miss.
BD Canada, Mississauga, ON, Canada.
MS222, Argent Chemical Laboratories Inc, Redmond, Wash.
Titan or Sonosite 180 Plus with C60 (2 to 5 MHz) and L52 (5 to 10 MHz) linear transducers, Sonosite Inc, Bothell, Wash.
Ibex Pro with CLi3.8 (2.5 to 5 MHz) curvilinear transducer and L6.2 (5 to 8 MHz) linear transducer, EI Medical Imaging, Loveland, Colo.
GE Logiqbook with 4C-RS (2 to 5.5 MHz) and 8C-RS (4 to 10 MHz) curvilinear transducers, GE Healthcare, Waukesha, Wisc.
Dissociation reagent No. X017, Arbor Assays, Ann Arbor, Mich.
Provided by CJ Munro, University of California-Davis, Davis, Calif.
Arbor Assays, Ann Arbor, Mich.
Sigma Aldrich Chemicals, St Louis, Mo.
ABTS, Sigma Aldrich Chemicals, St Louis, Mo.
Dynex MRX Revelation, Dynex Technologies, Chantilly, Va.
R: a language and environment for statistical computing, version 3.3.1, R Foundation for Statistical Computing, Vienna, Austria.
lme4: Linear mixed-effects models using ‘Eigen’ and S4, version 1.1–14, Bates D, Maechler M, Bolker B, et al. Available at: CRAN.R-project.org/package=lme4. Accessed Oct 29, 2017.
lmerTest: Tests in linear mixed effects models, R package version 2.0–33, Kuznetsova A, Brockhoff PB, Christensen RHB. Available at: CRAN.R-project.org/package=lmerTest. Accessed Oct 29, 2017.
ggplot2: Create elegant data visualisations using the grammar of graphics, version 2.2.1, Wickham H, Chang W. Available at: cloud.r-project.org/web/packages/ggplot2/index.html. Accessed Oct 29, 2017.
References
1. Cortés E. Life history patterns, demography, and population dynamics In: Carrier JC, Musick JA, Heithaus MR, eds. Biology of sharks and their relatives. Boca Raton, Fla: CRC Press Inc, 2004;449–469.
2. Grubbs RD, Carlson JK, Romine JG, et al. Critical assessment and ramifications of a purported marine trophic cascade. Sci Rep 2016;6:20970.
3. Henningsen AD. Notes on reproduction in the southern stingray, Dasyatis americana (Chondrichthyes: Dasyatidae), in a captive environment. Copeia 2000;2000:826–828.
4. Garner MM. A retrospective study of disease in elasmobranchs. Vet Pathol 2013;50:377–389.
5. Hamlett WC. Reproductive biology and phylogeny of Chondrichthyes: Sharks, batoids, and chimaeras. Enfield, NH: Science Publishers Inc, 2005.
6. Agnew DW, Munson L, Ramsay EC. Cystic endometrial hyperplasia in elephants. Vet Pathol 2004;41:179–183.
7. Munson L. Diseases of captive cheetahs (Acinonyx jubatus): results of the cheetah research council pathology survey, 1989–1992. Zoo Biol 1993;12:105–124.
8. Antunes CMF, Stolley PD, Rosenshein NB, et al. Endometrial cancer and estrogen use. N Engl J Med 1979;300:9–13.
9. Meissner WA, Sommers SC, Sherman G. Endometrial hyperplasia, endometrial carcinoma, and endometriosis produced experimentally by estrogen. Cancer 1957;10:500–509.
10. Nash JP, Kime DE, Van der Ven LT, et al. Long-term exposure to environmental concentrations of the pharmaceutical ethynylestradiol causes reproductive failure in fish. Environ Health Perspect 2004;112:1725–1733.
11. Ye RR, Peterson DR, Kitamura S-I, et al. Sex-specific immunomodulatory action of the environmental estrogen 17α-ethynylestradiol alongside with reproductive impairment in fish. Aquat Toxicol 2018;203:95–106.
12. Chester Jones I, Bellamy D, Chan D, et al. Biological actions of steroid hormones in nonmammalian vertebrates. In: Idler DR, ed. Steroids in nonmammalian vertebrates. New York: Academic Press, 1972;414–480.
13. Callard IP, Koob TJ. Endocrine regulation of the elasmobranch reproductive tract. J Exp Zool 1993;266:368–377.
14. Snelson FF Jr, Rasmussen L, Johnson MR, et al. Serum concentrations of steroid hormones during reproduction in the Atlantic stingray, Dasyatis sabina. Gen Comp Endocrinol 1997;108:67–79.
15. Tsang PC, Callard IP. Regulation of ovarian steroidogenesis in vitro in the viviparous shark, Squalus acanthias. J Exp Zool 1992;261:97–104.
16. Ozon R. Estrogens in fishes, amphibians, reptiles and birds. In: Idler DR, ed. Steroids in nonmammalian vertebrates. New York: Academic Press, 1972;390–413.
17. Ozon R. Androgens in fishes, amphibians, reptiles and birds. In: Idler DR, ed. Steroids in nonmammalian vertebrates. New York: Academic Press, 1972;328–389.
18. Di Prisco CL, Vellano C, Chieffi G. Steroid hormones in the plasma of the elasmobranch Torpedo marmorata at various stages of the sexual cycle. Gen Comp Endocrinol 1967;8:325–331.
19. Wotiz HH, Botticelli CR, Hisaw FL, et al. Estradiol-17β, estrone, and progesterone in the ovaries of dogfish (Squalus Suckleyi). Proc Natl Acad Sci U S A 1960;46:580–583.
20. van Bohemen CG, Lambert JGD. Estrogen synthesis in relation to estrone, estradiol, and vitellogenin plasma levels during the reproductive cycle of the female rainbow trout, Salmo gairdneri. Gen Comp Endocrinol 1981;45:105–114.
21. Browning HC. Role of prolactin in regulation of reproductive cycles. Gen Comp Endocrinol 1969;2(suppl 2):42–54.
22. Pratt JJ. Steroid immunoassay in clinical chemistry. Clin Chem 1978;24:1869–1890.
23. Mills SC, Mourier J, Galzin R. Plasma cortisol and 11-ketotes-tosterone enzyme immunoassay (EIA) kit validation for three fish species: the orange clownfish Amphiprion percula, the orangefin anemonefish Amphiprion chrysopterus and the blacktip reef shark Carcharhinus melanopterus. J Fish Biol 2010;77:769–777.
24. Penfold LM, Powell D, Traylor-Holzer K, et al. “Use it or lose it”: characterization, implications, and mitigation of female infertility in captive wildlife. Zoo Biol 2014;33:20–28.
25. Bigelow HB, Schroeder WC. Fishes of the western north Atlantic. No. 1. Part 2. Sawfishes, guitarfishes, skates and rays. New Haven, Conn: Sears Foundation for Marine Research, Yale University, 1953.
26. Ramírez-Mosqueda E, Pérez-Jiménez JC, Mendoza-Carranza M. Reproductive parameters of the southern stingray Dasyatis americana in southern Gulf of Mexico. Lat Am J Aquat Res 2012;40:335–344.
27. Selby C. Interference in immunoassay. Ann Clin Biochem 1999;36:704–721.
28. Martin B. Steroid-protein interactions in nonmammalian vertebrates: I. Elasmobranch steroids-binding protein (E. SBP) in dogfish serum (Scyliorhinus canicula). Gen Comp Endocrinol 1975;25:42–51.
29. Anderson DC. Sex-hormone-binding globulin. Clin Endocrinol (Oxf) 1974;3:69–96.
30. Munro CJ, Lasley BL. Non-radiometric methods for immunoassay of steroid hormones. Prog Clin Biol Res 1988;285:289–329.
31. Barnes SA, Andrew Teare J, Staaden S, et al. Characterization and manipulation of reproductive cycles in the jaguar (Panthera onca). Gen Comp Endocrinol 2016;225:95–103.
32. Graham KM, Mylniczenko ND, Burns CM, et al. Examining factors that may influence accurate measurement of testosterone in sea turtles. J Vet Diagn Invest 2016;28:12–19.
33. Munro C, Stabenfeldt G. Development of a microtitre plate enzyme immunoassay for the determination of progesterone. J Endocrinol 1984;101:41–49.
34. Munro CJ, Stabenfeldt GH, Cragun JR, et al. Relationship of serum estradiol and progesterone concentrations to the excretion profiles of their major urinary metabolites as measured by enzyme immunoassay and radioimmunoassay. Clin Chem 1991;37:838–844.
35. Lean JM, Davies JT, Fuller K, et al. A crucial role for thiol antioxidants in estrogen-deficiency bone loss. J Clin Invest 2003;112:915–923.
36. Hamlett WC, Eulitt AM, Jarrell RL, et al. Uterogestation and placentation in elasmobranchs. J Exp Zool 1993;266:347–367.
37. Hamlett WC, Musick JA, Eulitt AM, et al. Ultrastructure of uterine trophonemata, accommodation for uterolactation, and gas exchange in the southern stingray, Dasyatis americana. Can J Zool 1996;74:1417–1430.
38. Tricas TC, Maruska KP, Rasmussen L. Annual cycles of steroid hormone production, gonad development, and reproductive behavior in the Atlantic stingray. Gen Comp Endocrinol 2000;118:209–225.
39. Idler DR, Freeman HC. Sex hormone binding proteins I. Binding of steroids by serum of an elasmobranch (Raja radiata). Gen Comp Endocrinol 1969;13:75–82.
40. Faix JD. Principles and pitfalls of free hormone measurements. Best Pract Res Clin Endocrinol Metab 2013;27:631–645.
41. Ekins R. Measurement of free hormones in blood. Endocr Rev 1990;11:5–46.
42. Clapauch R, Mattos TM, Silva P, et al. Total estradiol, rather than testosterone levels, predicts osteoporosis in aging men. Arq Bras Endocrinol Metabol 2009;53:1020–1025.
43. Perez LE, Callard IP. Regulation of hepatic vitellogenin synthesis in the little skate (Raja erinacea): use of a homologous enzyme-linked immunosorbent assay. J Exp Zool 1993;266:31–39.
44. Polzonetti-Magni AM, Mosconi G, Soverchia L, et al. Multihormonal control of vitellogenesis in lower vertebrates. Int Rev Cytol 2004;239:1–46.
45. Panter GH, Thompson RS, Sumpter JP. Intermittent exposure of fish to estradiol. Environ Sci Technol 2000;34:2756–2760.
46. Fasano S, D'Antonio M, Pierantoni R, et al. Plasma and follicular tissue steroid levels in the elasmobranch fish, Torpedo marmorata. Gen Comp Endocrinol 1992;85:327–333.
47. Koob TJ, Tsang P, Callard IP. Plasma estradiol, testosterone, and progesterone levels during the ovulatory cycle of the skate (Raja erinacea). Biol Reprod 1986;35:267–275.
48. Tsang PC, Callard IP. Morphological and endocrine correlates of the reproductive cycle of the aplacental viviparous dogfish, Squalus acanthias. Gen Comp Endocrinol 1987; 66:182–189.
49. Ho S, Wulczyn G, Callard I. Induction of vitellogenin synthesis in the spiny dogfish, Squalus acanthias. Bull Mt Desert Isl Biol Lab Salisb Cove Maine 1980;19:37–38.
50. Prisco M, Valiante S, Maddalena Di Fiore M, et al. Effect of 17β-estradiol and progesterone on vitellogenesis in the spotted ray Torpedo marmorata Risso 1810 (Elasmobranchii: Torpediniformes): studies on females and on estrogen-treated males. Gen Comp Endocrinol 2008;157:125–132.
51. Heppell SA, Denslow ND, Folmar LC, et al. Universal assay of vitellogenin as a biomarker for environmental estrogens. Environ Health Perspect 1995;103:9–15.
52. Snelson FF Jr, Williams-Hooper SE, Schmid TH. Reproduction and ecology of the Atlantic stingray, Dasyatis sabina, in Florida coastal lagoons. Copeia 1988;1988:729–739.
53. Simpson TH, Wright RS, Hunt SV. Steroid biosynthesis in vitro by the component tissues of the ovary of dogfish (Scyliorhinus caniculus L.). J Endocrinol 1968;42:519–527.
54. Callard I, Leathem J. In vitro steroid synthesis by ovaries of elasmobranchs and snakes. Arch Anat Microsc Morphol Exp 1965;54:35–48.
55. Urist MR, Schjeide AO. The partition of calcium and protein in the blood of oviparous vertebrates during estrus. J Gen Physiol 1961;44:743–756.
56. Baird DT, Fraser IS. Blood production and ovarian secretion rates of estradiol-17β and estrone in women throughout the menstrual cycle. J Clin Endocrinol Metab 1974;38:1009–1017.
57. Henningsen AD, Leaf RT. Observations on the captive biology of the southern stingray. Trans Am Fish Soc 2010;139:783–791.
Appendix
Sources of captive southern stingrays (Hypanus americanus) and timing of blood sample collection and ultrasonographic examination in a study of reproductive health and disease in southern stingrays.
| Population | No. of stingrays | Total No. of samples | Collection frequency |
|---|---|---|---|
| Aquarium housed | |||
| A | 9 | 21 | Quarterly (4 seasons/y) |
| B | 4 | 16 | Quarterly |
| C | 1 | 1 | Once (summer) |
| D | 4 | 11 | Quarterly (4 seasons/y) |
| E | 4 | 8 | Twice (fall and spring) |
| F | 2 | 4 | Twice (spring and summer) |
| G | 12 | 45 | Quarterly (4 seasons/y) |
| H | 3 | 9 | Quarterly (4 seasons/y) |
| I | 9 | 20 | Twice (winter and spring or summer) |
| Managed semiwild (lagoon-housed) Castaway Cay, Bahamas | 34 | 59 | Twice (summer and winter) |
| Wild | |||
| North Carolina | 2 | 2 | Once (summer) |
| Bimini Biological Field Station, Bimini, Bahamas | 10 | 10 | Once (fall) |
Appendix 2
Scoring system used in the ultrasonographic assessment of the ovaries of female southern stingrays.
| Score | Status | Follicles | Cysts | Width* (cm) | Uterine overlap |
|---|---|---|---|---|---|
| 1 | Inactive | Small (< 1 cm) or no follicles | No | — | No |
| 2 | Active | Single layer | No | — | No |
| 3 | Transitional or active | 1–2 layers | Sometimes | — | Little or no |
| 4 | Overactive and cystic | 2–4 layers | Yes | 4–6 | Yes |
| 5 | Markedly overactive and cystic | > 4 layers | Yes | > 7 | Yes |
Cross-sectional width from sagittal view.
— = Not applicable.
Appendix 3
Scoring system used in the ultrasonographic assessment of the uterus of female southern stingrays.
| Score | Status | Contents | Position |
|---|---|---|---|
| 1 | Inactive | Flat or none | Within left coelom |
| 2 | Active | Minor fluid or debris | Within left coelom |
| 3 | Transitional or active | Moderate fluid or debris | On midline or within left coelom |
| 4 | Enlarged | Copious fluid or debris | Lumen extends over midline |
| 5 | Grossly enlarged | Copious fluid or debris | Apposes spiral valve |
