Development of a Nonlethal and Minimally Invasive Protocol to Study Elasmobranch Reproduction

Abstract An understanding of basic reproductive biology is essential for successful species-specific management of elasmobranch fishes (sharks, skates, and rays). Such information is often gained through gross dissection or other lethal techniques, which are not appropriate for threatened and endangered species. Previous work on other vertebrates suggested that sex steroid hormones can be extracted from muscle tissues to identify reproductive status. Collecting for muscle biopsy is quick and minimally invasive and can be done without removing an animal from the water. Thus, the objective of the current study was to determine the efficacy of using muscle steroid hormones to assess the reproductive biology of elasmobranch fishes. The results suggest that concentrations of muscle progesterone, testosterone, and estradiol can be successfully quantified to study reproduction by radioimmunoassay. Additionally, there were significant correlations between the plasma and muscle estradiol concentrations in Spiny Dogfish Squalus acanthias and the progesterone, testosterone, and estradiol concentrations in Atlantic Sharpnose Sharks Rhizoprionodon terraenovae. The present investigation thus demonstrates that skeletal muscle is a nonlethally harvestable tissue that is well suited for studying the reproductive biology of elasmobranchs.


182
PROHASKA ET AL. combined with direct and indirect fishing pressures, make elasmobranchs susceptible to overexploitation worldwide, which can lead to population declines and localized extinction (Dulvy et al. 2003).
Historically, studying the reproductive biology of elasmobranchs entailed sacrificing a relatively large number of specimens and observing the changes in gross reproductive morphology over the reproductive cycle (e.g., Hisaw and Albert 1947;Zeiner and Wolf 1993;Walmsley-Hart et al. 1999;Francis et al. 2001;Ebert 2005;Ruocco et al. 2006). While this approach is still currently appropriate for elasmobranchs with sustainable population sizes (Heupel and Simpfendorfer 2010), it is not suitable for many because of their relatively low or unknown species abundance (IUCN 2011;Hammerschlag and Sulikowski 2011). Because of this limitation, alternative methods for determining size at maturity and/or reproductive cycles in elasmobranchs are being developed and validated (Sulikowski et al. 2007b).
One particular approach that has been utilized to nonlethally study the reproductive biology of elasmobranchs is the analysis of blood steroid hormone concentrations (e.g., Perez and Callard 1993;Sulikowski et al. 2006Sulikowski et al. , 2007aSulikowski et al. , 2007bKneebone et al. 2007;Awruch et al. 2008;Henningsen et al. 2008). Reproductive hormones directly correlate with morphological changes within the reproductive tract, specifically at maturation (e.g., Gelsleichter et al. 2002;Sulikowski et al. 2006) and during reproductive cycles (e.g., Kneebone et al. 2007;Sulikowski et al. 2007aSulikowski et al. , 2007b. However, obtaining blood from large species is impractical and can be problematic because of the difficulties involved with specimen handling. These restrictions also serve to protect the animals from stresses induced by handling procedures on large species (Skomal 2007). This situation is further compounded by the restrictions associated with sample collection from threatened and endangered species. Although plasma steroid hormone analysis is particularly useful for studying the reproductive biology of elasmobranchs, the aforementioned constraints make it necessary for new, less invasive approaches to be developed, especially for studying vulnerable species.
While steroid hormones are easily isolated from blood plasma, they are present in other body depots as well (Hoffmann 1978). For example, steroid hormones have been extracted from milk and/or muscle tissues from cows (Noppe et al. 2008), sheep, and chickens (Sawaya et al. 1998); the urine and feces of free-ranging wildlife (Lasley and Kirkpatrick 1991;Holt and Pickard 1999); blubber from marine mammals (Mansour et al. 2002;Kellar et al. 2006); and the skeletal muscle tissue of a chimaera, the Spotted Ratfish Hydrolagus colliei (Barnett et al. 2009), for various reproductive analyses. Despite the successful extraction of steroid hormones from body depots other than plasma, this approach has never been taken with an elasmobranch. Given the lack of knowledge pertaining to the reproductive biology of elasmobranch fishes, the objectives of the present study were to determine (1) whether reproductive steroid hormones can be quantified from elasmobranch skeletal muscle tissue and (2) whether there is a relationship between the muscle and plasma concentrations of these reproductive hormones.  Figure 1).

METHODS
Sampling.-Immediately after the capture of an individual, an 8-mL aliquot of blood was collected by caudal venipuncture using a heparinized needle and Vacutainer. Blood samples were placed in a cooler (4 • C) for up to 24 h before hematocrit analysis and centrifugation at 1,242 × g for 5 min. The separated plasma was stored at -20 • C in the laboratory until extraction. Gross morphological parameters were recorded, including mass (kg), FL, and natural TL, which were measured to the nearest centimeter over a straight line along the axis of the body from the tip of the snout to the posterior notch of the caudal fin and the posterior tip of the upper lobe of the caudal fin while in its natural position, respectively. Afterwards, all sharks were euthanized by severing the cervical vertebrae. Sharks captured in the Gulf of Maine and off the coast of Rhode Island were kept in a shipboard cooler at 4 • C before transport on ice to the University of New England's Marine Science Center and dissected approximately 24 h after capture. All sharks captured in the Gulf of Mexico were dissected approximately 1 h after capture. During dissections, the following were recorded: ovary and oviducal gland mass (to the nearest g); oviducal gland and the five maximum follicle diameters (to the nearest mm); and, if present, pup sex and stretch total length (STL), which was measured to the nearest millimeter over a straight line along the axis of the body from the tip of the snout to the posterior tip of the upper lobe of the caudal fin while fully extended along the axis of the body. In addition, a 5-g white skeletal muscle tissue sample from behind the second dorsal fin was collected and immediately stored at −20 • C until extraction.
Plasma steroid hormone extraction.-Each plasma sample was extracted for 17β-estradiol ([E 2 ]), testosterone ([T]), and progesterone ([P 4 ]) following the protocols from Tsang and Callard (1987) and Sulikowski et al. (2004). A 500-µL aliquot NONLETHAL PROTOCOL TO STUDY ELASMOBRANCH REPRODUCTION 183 FIGURE 1. Centers of sampling areas for Spiny Dogfish (1 and 2) and Atlantic Sharpnose Sharks (3). of each plasma sample was extracted twice for each hormone with 10 volumes of ethyl ether (ACS grade), and the liquid phase was evaporated at 37 • C in a heat block under a stream of nitrogen. To account for procedural loss, each sample was spiked with 1,000 counts/min of tritiated [E 2 ], [T], or [P 4 ] (Perkin-Elmer, Waltham, Massachusetts) prior to extraction. Extracts were reconstituted in phosphate-buffered saline with 0.1% gelatin (PBSG).
Skeletal muscle tissue steroid hormone extraction.-Two grams of white skeletal muscle tissue from each shark were homogenized with 8 mL of cold phosphate-buffered saline (PBS) for 30 s using a Kinematica PT 10-35 polytron (Bohemia, New York). Then, 500-µL aliquots of homogenate, in quadruplicate, were extracted for each hormone. Nonradiolabeled [E 2 ], [T], and [P 4 ] (Steraloids, Inc., Newport, Rhode Island) were used to make stocks (6.4 µg/mL for [E 2 ], and 80 µg/mL for [T] and [P 4 ]: all in ethanol, ACS grade) to "cold-spike" samples to calculate recoveries following extraction and generate the standard curves for radioimmunoassay. A 100-µL aliquot of each stock was evaporated under a stream of nitrogen, reconstituted in PBSG, and serially diluted to make working stocks with final concentrations of 200 pg/100 µL for [T] and [P 4 ] and 320 pg/100 µL for [E 2 ]. Working stocks were utilized to cold-spike muscle homogenates for calculating sample recovery. Of the four homogenate replicates, two were cold-spiked with a 50-µL aliquot of the corresponding working stock (100 pg of [T] or [P 4 ], 160 pg of [E 2 ]) to track sample recovery. All samples were heated in a 50 • C water bath for 15 min, briefly vortexed, and then cooled to room temperature. Samples were extracted with 10 volumes (5 mL) of 2:1 chloroform : methanol (ACS grade; histology grade), and vortexed for 1 min before snap-freezing in a dry ice acetone bath. The liquid phase was removed by piercing the frozen homogenate layer with a pasture pipette and transferring it to a new test tube. The liquid phase was then evaporated at 37 • C in a heat block under a stream of nitrogen. One mL of 70% methanol (histology grade) was added to the dry extract, and the samples were stored at -20 • C for 24 h. Samples were centrifuged at 962 × g for 10 min at 4 • C, and the liquid phase was decanted and evaporated at 37 • C in a heat block under a stream of nitrogen. Extracts were reconstituted in 200 µL of PBSG.
Individual muscle sample recoveries were calculated by subtracting the mean concentration of hormone found in the two nonspiked replicates from the mean concentration of hormone found in the two cold-spiked replicates. The resulting concentration of hormone was then divided by the concentration of hormone that was initially added to the two cold-spiked replicates (100 pg of [T] or [P 4 ], 160 pg of [E 2 ]). To calculate the final concentration of hormone in each skeletal muscle sample (in pg/g), the mean concentration of hormone found in the two nonspiked replicates was calculated, multiplied by a dilution factor of 8 (to convert the units to pg/g), and finally divided by the individual sample recovery.
Radioimmunoassay.-Steroid hormone concentrations from both plasma and muscle extracts were determined following a modified radioimmunoassay procedure from Tsang and Callard (1987 , were diluted to final concentrations of 1:18,000, 1:10,000, and 1:2,500, respectively. Tritiated hormones and antibodies were added to the reconstituted plasma samples (100 and 50 µL in duplicate) and to muscle samples (200 µL) using PBSG to bring the total volume to 400 µL. Assays were incubated at 4 • C for 24 h before separating free from bound hormone by the addition of a carbon (0.2%; Acros Organics, Fairlawn, New Jersey) and dextran 70 (0.02%; Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) suspension followed by centrifugation at 1,242 × g for 10 min at 4 • C. The resulting supernatant was then decanted into scintillation vials and 3.5 mL of Ecolume (MPO Biomedicals, Solon, Ohio) was added. The radioactivity level was then measured by a Perkin-Elmer Tri-Carb 2900TR liquid scintillation analyzer (Waltham, Massachusetts). The mean intra-assay coefficients of variation were 8, 10, and 9% for [E 2 ], [T], and [P 4 ], respectively, in the Spiny Dogfish plasma assays and 7, 7, and 6% in the Atlantic Sharpnose Shark plasma assays. The mean interassay coefficients of variation were 10, 11, and 12% for [E 2 ], [T], and [P 4 ], respectively, in the Spiny Dogfish plasma assays and 8, 8, and 9% in the Atlantic Sharpnose Shark plasma assays. The mean intra-assay coefficients of variation were 8, 10, and 8% for [E 2 ], [T], and [P 4 ], respectively, in the Spiny Dogfish muscle assays and 10, 8, and 9% in the Atlantic Sharpnose Shark muscle assays. The mean interassay coefficients of variation were 10, 11, and 12% for [E 2 ], [T] and [P 4 ], respectively, in the Spiny Dogfish muscle assays. In calculating the means ± SEs for the concentrations of each steroid hormone per stage, any value that was nondetectable was assigned the lowest possible concentration that the assay would have been able to detect in the aliquots utilized.
Statistical analysis.-Linear regressions were performed for [E 2 ], [T], and [P 4 ] quantified from plasma and skeletal muscle. Data were transformed if the variables failed tests of normality or homogeneity of variance. If transformed variables still violated the assumptions the nonparametric Kendall's tau rank correlation was conducted. All data were analyzed using R 2.13.1 (R-Core Development, 2011), and means and SEs were graphed in Sigmaplot 12 (Systat Software, San Jose, California). All tests were considered significant at α = 0.05.

DISCUSSION
Concentrations of reproductive hormones were successfully detected in the muscle and plasma of female Spiny Dogfish and Atlantic Sharpnose Sharks during specific chronological events in the reproductive cycle. In addition, significant relationships were found between plasma and skeletal muscle concentrations of [E 2 ] in Spiny Dogfish and [E 2 ], [P 4 ], and [T] in Atlantic Sharpnose Sharks, suggesting that the concentrations of these hormones in the skeletal muscle of elasmobranchs have the potential to be an indicator of gestational stage.

Spiny Dogfish
In the current study, plasma and muscle [P 4 ] in Spiny Dogfish exhibited patterns during distinct gestational stages similar to those previously observed by Tsang and Callard (1987). For example, in both studies [P 4 ] concentrations were elevated in the first half of gestation and decreased from mid gestation through parturition (Tsang and Callard 1987). Although not measured by Tsang and Callard (1987), the preovulatory levels of [P 4 ] concentrations determined in the present study were relatively elevated in both plasma and muscle when compared to the other gestational stages examined. Based on the morphological data, this increase in [P 4 ] could inhibit further vitellogenesis (Paolucci and Callard 1998;Koob and Callard 1999), as no substantial increase in follicle diameter was noted between near-parturition and preovulatory females (Table 1). In this species follicles develop concurrently with gestation, with mating and ovulation occurring shortly after birth; thus, there is presumably little follicular growth after parturition (Castro 2009). There was an observed disparity between the patterns of plasma and muscle [P 4 ] concentrations during the candle stage of gestation. While the reasons behind this disparity cannot be determined from the current study, the differences could be attributed to [P 4 ] uptake/clearance (D'Ercole et al. 1984) or metabolism (Payne and Hales 2004) in the plasma and skeletal muscle.
Testosterone concentrations in plasma and muscle displayed similar patterns, remaining low as gestation progressed from the candle stage through mid gestation. However, this hormone began to increase in the plasma near the end of gestation and peaked in both tissues during preovulation. While there was no significant relationship between [T] concentrations in the plasma and muscle of this species, the patterns exhibited by plasma and skeletal muscle are consistent with the values observed by Tsang and Callard (1987). The functional significance of the peak in [T] during preovulation has yet to be studied. It is likely not related to folliculogenesis, since follicles develop concurrently with gestation in Spiny Dogfish, with theoretically little follicular growth from parturition to preovulation (Castro 2009). It is possible, however, that the peak in [T] plays a role in the storage and maintenance of sperm viability within the oviducal gland, which has been proposed in other elasmobranch species (Manire et al. 1995;Henningsen 1998).
Estradiol concentrations in the plasma and muscle of Spiny Dogfish remained relatively low from preovulation to mid gestation. However, substantial increases in this hormone began after mid gestation and the level remained elevated until just prior to parturition. These observed [E 2 ] patterns in the current study parallel those found by Tsang and Callard (1987 low concentrations in the plasma and muscle, which were expected because of the presumed minimal follicular growth after parturition (Castro 2009).

Atlantic Sharpnose Sharks
The patterns of plasma and muscle [P 4 ] and [T] concentrations were similar in that they were relatively low from preovulation through early gestation before reaching a distinct peak in early-mid-gestation females. The pronounced elevation of plasma and muscle [P 4 ] and [T] concentrations coincided with the gestational stage at which uterine compartmentalization and implantation are suggested to arise in Atlantic Sharpnose Sharks (Castro and Wourms 1993). Interestingly, similar findings were reported by Manire et al. (1995), who observed that [P 4 ] and [T] concentrations were heightened in the Bonnethead Sphyrna tiburo during early gestation prior to implantation, which was proposed to be a result of uterine compartmentalization (Schlernitzauer and Gilbert 1966;Callard et al. 1992). Notably, muscle [P 4 ] and [T] concentrations were found to be significant predictors of [P 4 ] and [T] in the plasma, suggesting that these two hormones in the skeletal muscle tissue have the potential to be indicators of gestational stage in this species.
The patterns of plasma and muscle [E 2 ] concentrations in Atlantic Sharpnose Sharks also concurrently varied, being elevated during preovulation before decreasing through mid gestation. This same pattern was also observed by Manire et al. (1995), who noted that in the Bonnethead, [E 2 ] peaks in the preovulatory stage during the final phase of folliculogenesis and decreases from early to mid gestation. In addition, muscle [E 2 ] concentrations in Atlantic Sharpnose Sharks were significant predictors of plasma [E 2 ] concentrations, indicating that skeletal muscle [E 2 ], just like [P 4 ] and [T], has the potential to be an indicator of gestational stage in this species.

CONCLUSIONS
In the present study, we developed a method of quantifying steroid hormones in skeletal muscle and tested their efficacy, relative to plasma hormone concentrations, in assessing the reproductive status of sharks. An advantage of using skeletal muscle rather than plasma is that muscle tissue can be obtained nonlethally and with minimal invasion. The results suggest that white skeletal muscle tissue of Spiny Dogfish and Atlantic Sharpnose Sharks is a depot for the reproductive hormones [P 4 ], [T], and [E 2 ], which can be reliably measured by radioimmunoassay. Additionally, the patterns of muscle reproductive hormones, which are coincident with gestational stage, paralleled those in plasma despite the different reproductive modes exhibited by these two sharks. Our results suggest that skeletal muscle steroid hormone content correlates well with gestational stage in yolk sac placental species. As such, future Atlantic Sharpnose Shark sampling efforts will focus on collecting individuals in the later stages of gestation to determine whether the parallel patterns observed in plasma and muscle steroid hormone concentrations are consistent throughout the reproductive cycle. While there were no significant relationships between plasma and muscle [P 4 ] and [T] concentrations in Spiny Dogfish, the similar patterns exhibited by these two hormones in plasma and skeletal muscle suggest that they could still be used to broadly assess gestational stage in that species. Certainly, because of the stronger relationship between plasma and skeletal muscle [E 2 ] concentrations, this hormone may be a better indicator of gestational stage in Spiny Dogfish.