JOURNAL OF EXPERIMENTAL ZOOLOGY 284:575–585 (1999) Alterations in Serum Steroid Concentrations in the Clearnose Skate, Raja eglanteria: Correlations With Season and Reproductive Status L.E.L. RASMUSSEN,1* D.L. HESS,2 AND C.A. LUER3 Department of Biochemistry and Molecular Biology, Oregon Graduate Institute of Science and Technology, Portland, Oregon 97291-1000 2 Department of Reproductive Science, Oregon Regional Primate Research Center, Beaverton, Oregon 97006 3 Mote Marine Laboratory, Sarasota, Florida 34236 1 ABSTRACT Serum steroid hormones in the peripheral circulation of the clearnose skate, Raja eglanteria, were measured at the time of capture and at various times throughout the year while the animals were maintained as a captive breeding population. These analyses demonstrate interesting correlations between changes in hormone concentrations and annual reproductive events. Animals were sampled once (78 females, 20 males) or multiple times (15 females). For both groups of females, 17β-estradiol was detected throughout the year with significant elevations occurring during October and November when ovarian follicles begin to mature (as determined through necropsy examinations), and January and February when maximum mating activity is observed and egg laying begins. Testosterone and dihydrotestosterone concentrations were significantly elevated in females only during January and February. Testosterone elevations were synchronous with longer-term elevations in 17β-estradiol in females sampled either once or repetitively. Testosterone concentrations in males were significantly elevated during times of maximum breeding activity compared to periods of sexual inactivity. Data from females sampled during five stages of the egg laying process, as defined by the position of palpable egg capsules within the reproductive tract, revealed that 17β-estradiol was highest when egg capsules were forming in the nidamental gland (stage 2) or uterus (stage 3); testosterone and dihydrotestosterone were maximal when eggs were in the uterus (stage 3) or cloaca (stage 4); and progesterone was significantly elevated immediately after oviposition (stage 5), suggesting a possible role for progesterone in the regulation of sequential laying of egg pairs. J. Exp. Zool. 284:575–585, 1999. © 1999 Wiley-Liss, Inc. A series of diverse reproductive strategies has been developed by elasmobranch fish (sharks, skates, and rays) during their long evolutionary history, and the associated patterns of serum steroid hormone concentrations are as heterogeneous as their varying morphologies and life histories. Such complex reproductive adaptations include oviparity (egg laying), aplacental viviparity (lecithotropic and histotropic live bearing), and placental viviparity (Wourms, ’77). The distinctive, accompanying hormonal events are often as complex as hormonal cycles described during mammalian reproduction. This is especially evident in recent endocrinological studies of placental elasmobranchs (Rasmussen and Murru, ’92; Rasmussen and Gruber, ’93; Manire et al., ’95; Manire and Rasmussen, ’97). For example, in the bonnethead shark, Sphyrna tiburo, high serum concentrations of 17β-estradiol and testosterone are © 1999 WILEY-LISS, INC. evident during mating and preovulatory stages, and significantly elevated progesterone is evident during preovulatory, ovulatory, and postovulatory stages; dihydrotestosterone (DHT) is only elevated during the preovulatory period (Manire et al., ’95). In non-placental elasmobranchs, such as the aplacental viviparous electric ray, Torpedo marmorata (Lupo di Prisco et al., ’67), spiny dogfish, Squalus acanthias (Callard et al., ’93), and the oviparous little skate, Raja erinacea (Koob et al., ’86), hormonal patterns are different. In these spe- Grant sponsor: Biosphere Research Corporation; Grant sponsor: Mote Marine Laboratory; Grant sponsor: The Disney Wildlife Conservation Fund. *Correspondence to: L.E.L. Rasmussen, Department of Biochemistry and Molecular Biology, Oregon Graduate Institute of Science and Technology, P.O. Box 91000, Portland, OR 97291-1000. E-mail: email@example.com 576 L.E.L. RASMUSSEN ET AL. cies, evidence implicates gonadal hormones in the direct regulation of the reproductive tract (Sumpter and Dodd, ’79; Dodd and Sumpter, ’84; Callard and Klosterman, ’87; Callard et al., ’89; Koob and Callard, ’91). For example, in R. erinacea, a species in which females undergo two reproductive cycles each year during which egg pairs are laid every five to seven days (Fitz and Daiber, ’63; Richards et al., ’63), clear hormonal patterns establish seasonal changes in reproductive activity in the populations studied. Two periods have been observed wherein a higher proportion (60%) of females is reproductively active, although the population as a whole is reproductively active all year. Elevated 17β-estradiol and testosterone are characteristic of the follicular phase, with the source of hormones identified as the developing ovarian follicles (Koob et al., ’86). Increases in serum progesterone in the little skate occur predominantly in pre- to periovulatory periods in contrast to the longer periods of elevation in mammals and placental sharks (Manire et al., ’95). Recently, Callard et al. (’95) suggested that steroid hormones such as progesterone and 17β-estradiol act as key gating mechanisms, controlling such events as mate selection and mating, follicular development and regression, steroidogenesis, development and maintenance of the reproductive tract, ovulation, vitellogenesis, corpora lutea events, and parturition or oviposition. In contrast to R. erinacea, R. eglanteria is not reproductively active year-round, with its year clearly divided into a reproductive period and a period of reproductive quiescence. With nonoverlapping cycles, the ability to associate hormonal changes with season and reproductive status should be less complicated than in R. erinacea. The reproductive season for R. eglanteria appears to be related to temperature, since breeding activity for R. eglanteria in the Delaware Bay (Fitz and Daiber, ’63) and winter breeding activity for R. eglanteria from the eastern Gulf of Mexico (Luer and Gilbert, ’85) occur at the same water temperature. Although Fitz and Daiber (’63) never observed mating or laying of eggs, their assessment of reproductive activity was based on size–frequency distribution of ovarian eggs. Interestingly, Fitz and Daiber (’63) found two groups of ovarian eggs, a reservoir group and a maturing group. The maturing eggs were assumed to be for the ensuing reproductive season, while the reservoir eggs were thought to make up the maturing group for the following year’s reproductive activity. In addition, they describe the initiation of the egg case formation in the shell gland prior to ovulation (Fitz and Daiber, ’63). Since 1981, the clearnose skate, R. eglanteria, has been successfully bred and maintained in captivity at Mote Marine Laboratory (Luer and Gilbert, ’85; Luer, ’89). Under the appropriate captive conditions, adult female clearnose skates breed readily with adult males and begin to lay fertile eggs (Luer and Gilbert, ’85; Luer, ’89). The resulting embryos and laboratory-raised offspring have been utilized as a unique vertebrate model for a variety of anatomical (Conrad et al., ’94), physiological (Goldstein et al., ’93), biochemical (Bodine et al., ’89), and immunological investigations (Rast et al., ’97). Because considerable information is available about the biology and behavior of reproduction in the clearnose skate, and because the entire reproductive cycle can be monitored in captivity, this species provides an excellent opportunity to decipher some of the sequential and interrelated hormonal changes occurring during the complex reproductive processes of follicular maturation, mating, ovulation, fertilization, oviposition, and sexual quiescence. The reproductive period for R. eglanteria from the eastern Gulf of Mexico is defined as beginning in late fall (October/November) when sexually mature individuals begin to inhabit the near-shore waters. Based on necropsy examinations, this time period appears to be associated with the development and ripening of ovarian follicles for the ensuing reproductive season (Luer, unpublished). Mating behavior occurs from December through late March to early April, with maximal copulatory activity during January and February (Luer and Gilbert, ’85; Luer, ’89). As the temperature of near-shore waters begins to increase, adults move offshore by late February to early March in search of deeper, cooler water and are absent from local waters until the next fall. The captive conditions described here provide the only way to observe the annual reproductive cycle. Following a successful breeding event, sperm make their way up the female’s reproductive tract to the paired nidamental glands, where sperm storage is presumed to occur. This presumption is based on the presence of sperm in nidamental glands examined through tissue imprints and histological sections, with no visible evidence of sperm ever reaching the oviducts (Luer, unpublished). As ova mature, they are ovulated and travel down the oviducts to the nidamental REPRODUCTIVE HORMONE CYCLES IN RAJA EGLANTERIA glands, where fertilization and encapsulation take place (Luer, ’89). Egg laying begins in January and continues through June, occasionally extending into early to mid July if the females are maintained at 20°C (Luer, unpublished). Eggs are laid in pairs, one from each side of the paired reproductive tract, at intervals averaging 4–5 days between laying of successive egg pairs (Luer and Gilbert, ’85). Typically, the last few egg pairs laid at the end of the season are infertile (Luer, unpublished). Time frames for various reproductive activities or events are shown in Fig. 1. The present study describes serum steroid hormone concentrations and hormonal cycles in reproductively active clearnose skates. Specifically, serum 17β-estradiol, testosterone, DHT, and progesterone levels were measured and correlated with the behavior or reproductive event occurring at the time of blood sampling. This study not only increases our knowledge of steroid hormones as they relate to the diversity of reproductive strategies in elasmobranchs, but also provides practical information for rapidly growing interests in captive elasmobranch breeding programs. MATERIALS AND METHODS Research animals Clearnose skates, R. eglanteria, were collected in near-shore waters off Sarasota, Florida. The 577 majority of the animals were collected in the late fall and early winter. Animals were held initially in flow-through outdoor tanks at seasonally dependent temperatures, ranging from 14–22°C. When flow-through water temperatures reached 22°C, animals were moved to recirculating indoor tanks maintained at 20°C. The special conditions during the study comparing hormone levels of female skates housed with and without males included maintaining daily temperature records. Because adult females were maintained as active egg layers for the production of embryos and laboratory-raised offspring for a variety of collaborative research applications, animals were not sacrificed as part of this study. Therefore, information and estimates about ovulation, encapsulation, and egg movement were determined by relative locations of successive egg pairs within the reproductive tracts of egg laying females. Twice each day, the presence of egg capsules was inferred from gentle palpation of the dorsal surface of females, often as part of a routine examination prior to obtaining blood samples. Egg positions were assigned to a stage in the egg laying process: Stage 1 had no eggs present (this group included both immediately pre-ovulatory females and post-ovulatory (greater than 2 days) females. Stage 2 had egg capsules forming in the region of the nidamental gland (in the process of Fig. 1. Time line of events occurring throughout the annual reproductive cycle in the clearnose skate. 578 L.E.L. RASMUSSEN ET AL. encapsulation). The location of the ova is uncertain here because ovulation occurs after the egg capsules are partly formed. Stage 3 had eggs in the uterus (encapsulation complete). Stage 4 had eggs in the cloaca, and stage 5 had eggs laid within 2 days. Because animals were observed several times each day, it was possible to determine the time of oviposition within a few hours. Blood samples were obtained from animals that had been in captivity from 2 weeks to 1 year. Blood samples (1–2 ml) were obtained via caudal venipuncture using a 23-gauge short-bevel needle. Samples were allowed to clot and were centrifuged for 15 min at 1286g. Serum was aspirated and stored at –20°C until thawed for analysis for steroid concentrations by radioimmunoassay (RIA). Single blood samples were obtained from 78 females and 20 males spanning five different breeding and egg laying seasons. Multiple samples were obtained from 15 females with sampling intervals ranging from every 24 hr for 1 week to bimonthly for 8 months. Animals were bled at the time of capture as well as at various times after acclimation to captive conditions. Radioimmunoassays Serum concentrations of 17β-estradiol, testosterone, DHT, and progesterone were determined by RIA after purification by chromatography on Sephadex LH-20 microcolumns. Serum aliquots of 500 µl were extracted by shaking for 5 min with 5 ml of freshly opened or redistilled diethyl ether. After freezing the aqueous phase in an ethanol/ dry ice bath, the organic phase was decanted and brought to dryness under a stream of air. The dried extracts were resolubilized and sequentially chromatographed on two different Sephadex LH columns. On the first column (1.0 g LH-20; elution mixture: hexane:benzene:methanol, 62:20:13 v/v), 17β-estradiol was separated from estrone and all neutral steroids (Resko et al., ’75). The neutral fraction from the first column was applied to the second column (2.5 g LH-20; elution mixture: hexane:benzene:methanol, 85:15:5 v/v), and the appropriate fractions for progesterone, DHT, and testosterone were collected (Resko et al., ’80). The purified steroids were then estimated by RIA as described previously (Resko et al., ’75; Hess et al., ’81). Extraction and chromatographic losses were monitored by adding known amounts of tritiated authentic steroids to independent samples of skate serum and processing these samples in parallel with the unknown samples. Recoveries following the chromatography were 75% for 17βestradiol, 88% for progesterone, and 72 and 70% for DHT and testosterone, respectively. Water blanks were also processed in parallel with the unknown samples to provide estimates of solvent blank methods (pg/ml) for each steroid as follows: 17β-estradiol, 3.2 ± 0.4; progesterone, 22 ± 3.0; DHT, 18.6 ± 9.8; and testosterone, 3.5 ± 1.9. Reported values were corrected for both procedural losses and method blanks before correcting for aliquots assayed. Each sample was diluted with 500 µl of ethanol after chromatography and assayed at different volumes ranging from 5–200 µl. The reported values are the average concentration calculated from aliquots whose values fell between the 5–95% binding limits of the appropriate standard curve following linearization with a logit-log transformation. Intra-assay coefficients of variation did not exceed 12% for these assays, and the sensitivity limits for the steroids determined in this study were 17β-estradiol, 5; progesterone, 10; DHT, 5; and testosterone, 5 pg/tube, respectively. All four steroid hormones demonstrated complete parallelism in the RIAs when compared to the standards, supporting our assumption that the purification procedure eliminated any major interfering substances. Statistics For data from the 78 singly sampled females, standard deviations and standard errors of mean were calculated for sample sizes greater than three when parametric tests were allowed (i.e., data met the criteria for normality and equal variance). Data were indicated as mean ± SEM. Students’ t-tests were utilized to determine significant differences between these data sets at significance levels of P < 0.05 (95%). Data that did not meet the parametric criteria, i.e., data that consistently failed the KolmogorovSmirnov test of normality (non-normal distribution and/or unequal variance), were expressed as medians (25–75 percentile respectively). Data were analyzed by nonparametric tests including the Mann-Whitney Rank Sum test or KruskalWallis one-way ANOVA on ranks (Sokal and Rohlf, ’95). To isolate the group(s) which was significantly different, Dunn’s method for pairwise multiple comparisons was employed (Dunn, ’64). Significant correlations between testosterone and 17β-estradiol concentrations in serially sampled skates were determined by Pearson product moment correlation. REPRODUCTIVE HORMONE CYCLES IN RAJA EGLANTERIA RESULTS Seasonal variation in serum steroid hormone concentrations in female skates The following data are one-time samples obtained from individual animals after at least 2 weeks in captivity. Statistically significant elevations are indicated in comparison to the reproductively inactive months of July and August. Serum 17β-estradiol was detected in all individual adult female skates throughout the entire year. Monthly medians demonstrated significantly elevated 17βestradiol levels during two periods (September through November, and January and February) compared to the reproductively inactive months of July and August (Fig. 2A). The reproductive events that occur during the months demonstrating high serum 17β-estradiol include the development of ovarian follicles (ovarian recrudescence) during October and November and the start of mating activity during January and February. During March through late June, egg laying continues, but copulatory activity gradually decreases through early April (Fig. 1). In contrast, during the summer months female skates are reproductively inactive. Serum testosterone was detectable in all adult female skates assayed (Fig. 2B). Median values (25– 75 percentile) of serum testosterone concentrations were significantly elevated in January through April compared to the period from July through December (Fig. 2B), whereas the apparent elevations during June were not significantly different. DHT concentrations in serum were low (Fig. 2C). At times the amount was below the level of detection, i.e., in May, September and December. Similar significant differences were observed for DHT as noted with testosterone. January and February concentrations were significantly higher than April–May and September–December (Fig. 2C). Progesterone concentrations were generally low (Fig. 2D). However, occasional transitory eleva- Fig. 2. Serum steroid concentrations in female skates throughout their annual reproductive cycle, depicted as monthly medians (25–75 percentiles). Skate sample size = 3– 15/month. (A) Monthly medians of 17β-estradiol (significant elevations September–November and January–February, compared to July and August); (B) monthly medians of testosterone (significant elevations January–April, compared to July–December); (C) monthly medians of dihydrotestosterone (January and February levels significantly higher than April– May and September–December levels); (D) monthly medians of progesterone (February medians significantly higher than March, July–August, and December). Kruskal-Wallis one-way ANOVA on ranks significant difference at P = 0.05. 579 580 L.E.L. RASMUSSEN ET AL. tions from four-fold to 100-fold were observed. Medians for progesterone concentrations during February were significantly higher than during three other periods: March, July–August, and December. July and August are reproductively inactive months; mid-December marks the initiation of copulatory behavior for the season, and egg laying generally has not yet begun; whereas February is characterized by periods of copulation and, in some years, maximum egg laying (Fig. 1). Although there was a 200-fold difference in absolute levels, P and T exhibited similar temporal patterns, i.e., the levels were highest in January, February, and April (Fig. 2B and D). Comparison of serum steroid hormone concentrations in male and female adult skates The comparison of serum steroid hormone concentrations in adult male (n = 20) and female (n = 20) skates during one reproductive season (1995) demonstrated several significant differences between the sexes (Table 1). The two sexes were compared during two (December and January) of the three (December, January, and February) months when observed copulatory activity was greatest. During those 2 months, it was possible to obtain equal numbers of blood samples from males and females. Concentrations of 17β-estradiol were significantly higher in females than males (670 ± 106.8 vs. 76.0 ± 44.9 pg/ml) during December. By January, female values had risen three-fold to 2480 ± 106.8 pg/ml (Table 1), whereas male levels had decreased to 22.8 ± 7.8 pg/ml. In males, the concentrations of progesterone, testosterone, and DHT were significantly higher than in females (Table 1). For both sexes, progesterone levels during January were not significantly different from December levels. For testosterone, a statistically significant difference existed between serum levels in female and male skates, with males being higher during December and January. However, at times other than these 2 months, the observed ranges overlapped. In females, the testosterone levels rose significantly from December to January (Table 1) and were highest during February (Fig. 2C). In males, testosterone levels increased from December to January, with the trend continuing into February (Table 1). The levels in males remained significantly higher than those in females. A similar male/female pattern was observed for DHT, with serum levels in males again significantly higher than those in females. During January, values in both female and male skates increased. In the males, the concentrations reached 6005.4 ± 1166.7 pg/ml. Effect of the presence or absence of male skates on serum steroid hormone concentrations in female skates Serum steroid hormone concentrations in three groups of adult female skates were compared during one reproductive season. All females were collected one month prior to the start of the study. Group 1 females were maintained in indoor tanks in the absence of males. Group 2 females were maintained in indoor tanks in the presence of males. Group 3 females were maintained outdoors with males. Because of space limitations, it was TABLE 1. Serum steroid hormone differences between male and female skates during December and January† Hormone December 17β-Estradiol Testosterone DHT Progesterone January 17β-Estradiol Testosterone2 DHT Progesterone Male (n = 20) Female (n = 20) Difference between male and female 76.0 35,390.0 3,716.7 150.0 ± ± ± ± 44.9 7,971.2 281.7 7.4 670.3 2,096.5 90.0 14.7 ± ± ± ± 106.8 510.2 90.0 7.41 F > M* M > F*,** M > F* M > F* 22.8 42,706.0 6,005.4 150.0 ± ± ± ± 7.8 16,787.0 1,166.7 7.4 2,480.0 14,000.0 550.5 42.7 ± ± ± ± 106.8 6,010.3 400.0 47.41 F > M* M > F* M > F* M>F †Units = pg/ml. *Significant difference. Statistical information: degrees of freedom = 24; P = 0.001–0.047; t = 4.5 to –5.0. **Mann-Whitney rank sum test, P = 0.001. 1 Because the brief, well-circumscribed rises in progesterone occur after each egg-laying event, the monthly median data do not accurately portray reproductive events. They are not really seasonal and are best described in individual animals. 2 Testosterone concentrations in males were highest in December through February, varying from 35,000 to 42,000 pg/ml. November and April levels were 9,000–10,000 pg/ml. REPRODUCTIVE HORMONE CYCLES IN RAJA EGLANTERIA not possible to maintain females outside without the presence of males. Photoperiods for all groups were similar but not identical, since the indoor environment was maintained at 12 hr light, 12 hr dark, and the outdoor environment averages 11 hr light, 13 hr dark during the months of January and February. The temperature was maintained at 20°C for the two indoor groups (groups 1 and 2), and fluctuations between 18 and 20°C occurred for group 3. All populations began to lay eggs, but eggs laid by the indoor animals without males were infertile. In all groups, January and February testosterone levels were high, and short-duration progesterone elevations were observed (Table 2). However, no significant differences were observed between groups. Concentrations of two serum hormones, 17β-estradiol and DHT, demonstrated distinct differences between group 1 (indoors, with no males) and group 2 (indoors, with males) as compared to group 3 (outdoors, with males) (Table 2). Significantly higher concentrations of 17β-estradiol and DHT were observed during January–February in females maintained with males (both indoors and outside) compared to the indoor group without males (Table 2). Serum hormone concentrations during sequential movement of eggs within the reproductive tract Sixty observations of egg positions were obtained prior to blood sampling. The egg locations were categorized into five groups as defined earlier. A number of significant differences was established in serum concentrations of steroid hormones among these five groupings of egg positions (Fig. 3). When no egg capsules were detected, 17β-estradiol levels (stage 1) were significantly 581 lower than when eggs were found either in the area of the nidamental gland (stage 2) or partially down the uterus (stage 3) (Fig. 3A). Similarly, testosterone and DHT levels were significantly lower in stage 1 than when eggs were detected in stages 2 and 3 or in the cloaca (stage 4) (Fig. 3B and D). In contrast, progesterone concentrations were significantly higher at stage 5 (immediately after egg laying and up to 48 hr post-egg laying) than at all other stages (Fig. 3C). Hormone patterns in serially sampled female skates Fifteen serially sampled female skates (each sampled on multiple occasions) demonstrated consistent patterns of reproductive hormones during the sampling periods from October to June, which spanned both reproductive and nonreproductive periods. Figure 4 depicts the mean serum 17βestradiol and testosterone (±SEM) values of these fifteen serially sampled females. 17β-Estradiol levels rose gradually in all of these females beginning in November/December, reaching highest levels in mid-February. Testosterone levels also increased steadily until January, then again rose until March. Testosterone concentrations remained high during the egg laying process. There was a statistically significant difference in 17βestradiol levels during February and March compared to the nonreproductive months of June and October and in testosterone levels during January, February, and March compared to May, June, and October (Mann-Whitney rank sum, P = 0.001). The high levels of testosterone observed during January and February in these serially sampled females were consistent with the high levels obtained in January–February from individually sampled females (Fig. 2B). In contrast, elevations in progesterone were only randomly TABLE 2. Serum steroid hormone concentrations (pg/ml) during January–February in three groups of female skates1 Group 1: indoors (females only) mean ± SE Group 2: indoors (with males) mean ± SE Group 3: outdoors (with males) mean ± SE 17β-Estradiol 1,441.8 ± 295.2 3,567.1 ± 71.9 4,023.1 ± 511.9 Testosterone DHT 9,220.9 ± 63.9 362.3 ± 74.8 11,330.8 ± 79.9 662.3 ± 64.7 14,530.0 ± 101.9 761.0 ± 141.9 Hormone Progesterone 1 67.0 (median) 92.0 (median) Statistical criteria: Kruskal-Wallis one-way ANOVA on ranks and Dunn’s test, P < 0.05. 67.0 (median) Significant difference Yes: Groups 3 and 2 greater than group 1 Q = 6.7, 4.1 No Yes: Groups 3 and 2 greater than group 1 Q = 3.6 No 582 L.E.L. RASMUSSEN ET AL. Fig. 3. Serum steroid levels during five stages of egg laying in the female skate. Code for stages: 1 = pre-ovulatory, no egg capsule present; 2 = egg capsule present in vicinity of nidamental gland; 3 = encapsulated egg present partially down uterine tract; 4 = encapsulated eggs present in cloaca; 5 = eggs laid within 0–48 hr. Statistical analyses using Kruskal-Wallis one-way ANOVA on ranks: (A) 17β-estradiol. Stage 1 significantly less than stages 2 (Q = 3.2) and 3 (Q = 3.6), and stages 2 and 3 significantly greater than stages 1, 4, and 5. (B) Testosterone. Stage 1 significantly less than stages 2 (Q = 3.3), 3 (Q = 5.3), and 4 (Q = 5.6); in addition, stages 3 and 4 greater than stages 1, 2 and 5. (C) Progesterone. Stage 5 significantly greater than all other stages: 1 (Q = 4.2), 2 (Q = 3.5), 3 (Q = 2.9), and 4 (Q = 2.9). (D) Dihydrotestosterone. Stage 1 significantly less than stages 2 (Q = 3.2), 3 (Q = 4.6), 4 (Q = 4.6), and 5 (Q = 3.1). detected, primarily in January through April. Interestingly, there was a significant correlation (Pearson product moment) P = 0.69 between 17βestradiol and testosterone concentrations for the months studied. haps be related to the continual ovulations that occur throughout the egg laying process. Other species of elasmobranchs also demonstrate increased 17β-estradiol during critical female reproductive events. For example, in the Atlantic stingray, Dasyatis sabina, two distinct elevations of 17β-estradiol were observed in females, one prior to ovulation and the other prior to initiation of histotroph production by the uterus (Snelson et al., ’97). These results contrast with 17β-estradiol elevations observed during the preovulatory period in the placental bonnethead shark (Manire et al., ’95) and the depression of 17β-estradiol in the aplacental spiny dogfish during pregnancy (Callard et al., ’89). Especially elevated levels of 17β-estradiol are consistent with pre-ovulatory folliculogenesis. Beginning with the early work on the aplacental viviparous ray, Torpedo marmorata, of Lupo di Prisco et al. (’67) and continuing with other researchers (Fasano et al., ’92), it was demonstrated that 17β-estradiol and androgen levels increased DISCUSSION This study provides a quantitative description of circulating steroid hormone concentrations in female clearnose skates throughout their annual reproductive cycle with special attention to fluctuations during the active egg laying period. 17βEstradiol concentrations are detectable all year, with circulating levels increasing during ovarian growth and prior to ovulation and onset of observed breeding within the captive population (September, October, and November). Levels are especially high during the early period of egg laying (January and February). With the exception of December, the extended period of elevated levels of 17β-estradiol during diverse reproductive events may result from receptor stability or per- REPRODUCTIVE HORMONE CYCLES IN RAJA EGLANTERIA Fig. 4. Serum hormone concentrations in 15 serially sampled female skates, from October through June. Upper graph: 17β-estradiol; lower graph: testosterone. Standard errors are indicated by bars. *Indicates significant elevations during several months in comparison to June and October levels (P < 0.001, Mann-Whitney rank sum test). as females reached the ultimate maturational stage before ovulation. In the oviparous lesser spotted dogfish, high levels of 17β-estradiol prior to ovulation suggest a functional role in yolk production or oocyte maturation (Craik, ’79). In the small spotted catshark, Scyliorhinus canicula, Sumpter and Dodd (’79) clearly demonstrated a ten-fold seasonal variation in 17β-estradiol and testosterone, with hormonal levels rising with the recrudescence of the ovary. The elegant work by Koob et al. (’86) demonstrated a “correlation of rising estradiol levels with follicle development during recrudescence and the preovulatory elevation of estradiol during egg laying cycles in the little skate, Raja erinacea.” This study reveals an interesting pattern of serum progesterone elevation, clearly seen in the data (Fig. 3C). An elevation in progesterone oc- 583 curred after each oviposition of a pair of eggs and was detected in both individually and serially sampled female skates. These progesterone elevations are observable; they occur just after ovipositioning and thus may be occurring prior to ovulation. Progesterone quite likely could be the maturation-inducing steroid of the clearnose skate. Even the monthly data demonstrate that the greatest frequency of progesterone elevations occurs early in the egg laying period when the maximum number of eggs are laid; as a few eggs are laid in June, these elevations may be detectable in monthly sampling (Fig. 2D). Preovulatory elevations of progesterone in the little skate have been correlated with encapsulation of eggs prior to ovulation (Koob et al., ’86). The correlation of hormone levels with egg laying stages suggests that during the period after a pair of eggs is laid, short-term elevations in progesterone affect encapsulation or the timing of the next ovulation. This may not be an absolute synchrony, but rather the continuation of a sequence since intervals between laying of egg pairs can be as short as 2 days, with a mean interval of 4.5 days. In contrast, serum testosterone in female clearnose skates assessed on a monthly basis (Fig. 2B) does not rise during ovarian recrudescence or prior to initial ovulatory events. However, testosterone concentrations increase dramatically with the onset of breeding activity and remain high during egg laying. We observed that serum testosterone and DHT concentrations were only significantly elevated during January and February, the period of maximum copulations and initiation of egg laying. In these months, there was a synchrony of 17β-estradiol and testosterone elevations, with testosterone elevations often slightly preceding 17βestradiol increases. Testosterone elevations have been observed during mating and gestation in lemon sharks (Rasmussen and Murru, ’92; Rasmussen and Gruber, ’93), and, interestingly, in reproducing female Atlantic stingrays, testosterone elevations precede 17β-estradiol elevations (Snelson et al., ’97). The elevation in testosterone prior to 17β-estradiol elevation revealed in the individual skates (Fig. 4) are not apparent in the egg laying data; rather, 17β-estradiol elevations slightly precede testosterone elevation during the egg laying process (Fig. 3). Koob et al. (’86) and Tsang and Callard (’87) demonstrated in vivo in R. erinacea that elevated serum titers of 17β-estradiol and testosterone were characteristic of the follicular phase, and by in vitro studies they 584 L.E.L. RASMUSSEN ET AL. showed that testosterone is the predominant steroid product of the theca. In their 1986 in vivo study of R. erinacea, Koob et al. demonstrated that the average daily titer of testosterone during egg production was less than 4 ng/ml, whereas in the current study of R. eglanteria, testosterone concentrations were of this low order of magnitude only during stage 1 (preovulatory, no egg capsule present, but no recent oviposition). During the subsequent egg laying processes (stages 2–5, from capsule formation in the nidamental gland to ovopositing), testosterone levels were greater than 10 ng/ml (Fig. 3). Serum testosterone concentrations were significantly higher when encapsulated eggs were in the uterus (stage 3) and in the cloaca (stage 4), stages equivalent to the “egg retention” period reported by Koob et al. (’86). Thus, these results in R. eglanteria contrast with the findings in R. erinacea (Koob et al., ’86) that testosterone levels reached their lowest values at the time of capsule formation. Instead, sustained testosterone elevations in the clearnose skate during egg laying stages suggest that this hormone has a functional role in egg laying. The comparison in single-time sampled females and serially sampled individual females for serum testosterone and 17β-estradiol during a ninemonth period demonstrated considerable similarity. Females from both data sets demonstrated significantly higher testosterone from January through April. 17β-Estradiol levels were significantly higher during January and February in both groups. The patterns of elevations and declines for testosterone and 17β-estradiol were similar when data from the serially sampled animals were compared (Fig. 4). Comparison of hormonal data between female skates, with and without the presence of males, allows for the separation of one variable affecting the reproductive process of egg laying, specifically the laying of fertilized versus nonfertilized eggs. Female skates maintained indoors without males and under conditions of constant temperature and photoperiod nevertheless demonstrated high serum progesterone during egg laying. Egg laying continued although the eggs were infertile, suggesting that ovulatory events, egg production, and oviposition are at least somewhat affected by progesterone concentrations. Testosterone was similar in both groups, suggesting testosterone may also have a role in the egg laying process. 17β-Estradiol, however, was elevated in both the indoor and outdoor animals in the presence of males, but was 2–3-fold lowered in females without males, implicating its role in the events associated with mating and fertilization. ACKNOWLEDGMENTS Sincere appreciation is expressed to C.J. Walsh, P.C. Blum, C. Harris, J. Kaiser, and C. Manire for their assistance at the Mote Marine Laboratory. This research was supported by Biosphere Research Corporation (L.E.L.R. and D.L.H.), the Mote Marine Laboratory (C.A.L.), and The Disney Wildlife Conservation Fund (C.A.L.). LITERATURE CITED Bodine AB. Luer CA, Gangjee SA, Walsh CJ. 1989. In vitro metabolism of the pro-carcinogen aflatoxin B1 by liver preparations of the calf, nurse shark, and clearnose skate. Comp Biochem Physiol 94C:447–453. Callard IP, Klosterman L. 1987. Reproductive physiology. In: Shuttlesworth T, editor. The physiology of elasmobranch fishes. Berlin: Springer-Verlag. p 277–291. Callard IP, Klosterman L, Sorbera LA, Fileti LA, Reese JC. 1989. Endocrine regulation of reproduction in elasmobranchs: archetype for terrestrial vertebrates. J Exp Zool Suppl 2:12–22. Callard IP, Fileti LA, Koob TJ. 1993. Ovarian steroid synthesis and the hormonal control of the elasmobranch reproductive tract. Environ Biol Fishes 38:175–185. Callard IP, Putz O, Paolucci M, Koob TJ. 1995. Elasmobranch reproductive life-histories: endocrine correlates and evolution. In: Goetz FW, Thomas P, editors. Proceedings of the fifth international symposium on the reproductive physiology of fish. Austin, TX: Fish Symp 95. p 204–208. Conrad GW, Paulsen AQ, Luer CA. 1994. Embryonic development of the cornea in the eye of the clearnose skate, Raja eglanteria: I. Stromal development in the absence of an endothelium. J Exp Zool 269:263–276. Craik JCA. 1979. Simultaneous measurements of rates of vitellogenin synthesis and plasma levels of oestradiol in an elasmobranch. Gen Comp Endocrinol 38:264–266. Dodd JM, Sumpter PJ. 1984. Fishes. In: Lamming GE, editor. Marshall’s physiology of reproduction. Edinburgh: Churchill Livingstone. p 1–126. Dunn OJ. 1964. Multiple comparisons using rank sums. Technometrics 6:241–252. Fitz ES Jr, Daiber FC. 1963. An introduction to the biology of Raja eglanteria Bose 1802 and Raja erinacea Mitchell 1925 as they occur in Delaware Bay. Bull Bingham Oceanogr Coll 18:69–97. Goldstein L, Luer CA, Blum PC. 1993. Taurine transport characteristics of the embryonic skate (Raja eglanteria) heart. J Exp Biol 182:291–295. Hess DL, Spies HG, Hendrickx AG. 1981. Diurnal steroid patterns during gestation in the rhesus macaque: onset, daily variation and the effects of dexamethasone treatment. Biol Reprod 24:609–616. Koob TJ, Callard IP. 1991. Reproduction in female elasmobranchs. In: Kinns R, editor. Comparative physiology, oocytes and reproduction, volume 4. New York: Karger. p 155–210. Koob TJ, Tsang P, Callard IP. 1986. Plasma estradiol, testosterone and progesterone levels during the ovulatory cycle of the skate (Raja erinacea). Biol Reprod 35:267–275. REPRODUCTIVE HORMONE CYCLES IN RAJA EGLANTERIA Luer CA. 1989. Elasmobranchs (sharks, skates and rays) as animal models for biomedical research. In: Woodhead AD, editor. Nonmammalian animal models for biomedical research. Boca Raton, FL: CRC Press. p 122–141. Luer CA, Gilbert PW. 1985. Mating behavior, egg deposition, incubation and hatching in the clearnose skate, Raja eglanteria. Environ Biol Fishes 13:161–171. Lupo di Prisco C, Vellano C, Chieffi G. 1967. Steroid hormones in the plasma of the elasmobranch Torpedo marmorata at various stages of the sexual cycle. Gen Comp Endocrinol 8:325–331. Manire CA, Rasmussen LEL. 1997. Serum concentrations of steroid hormones in the mature male bonnethead shark, Sphyrna tiburo. Gen Comp Endocrinol 107:414–420. Manire CA, Rasmussen LEL, Hess DL, Hueter RE. 1995. Serum steroid hormones and the reproductive cycle of the female bonnethead shark, Sphyrna tiburo. Gen Comp Endocrinol 89:107–118. Rasmussen LEL, Murru FL. 1992. Long-term studies of serum concentrations of reproductively related steroid hormones in individual captive carcharhinids. Aust J Mar Freshwater Res 43:273–281. Rasmussen LEL, Gruber SH. 1993. Serum concentrations of reproductively related circulating steroid hormones in the free-ranging lemon shark, Negaprion brevirostris. Environ Biol Fishes 38:167–174. Rast JP, Anderson MK, Strong S, Luer C, Litman RT, Litman 585 GW. 1997. α, β, γ and δ T cell antigen receptor genes arose early in vertebrate phylogeny. Immunity 6:1–11. Resko J, Ellinwood WE, Pasztor LM, Buhl AE. 1980. Sex steroids in the umbilical circulation of fetal rhesus monkeys from the time of gonadal differentiation. J Clin Endocrinol Metab 50:900–905. Resko JA, Ploem JG, Stadelmen HL. 1975. Estrogens in fetal and maternal plasma of the rhesus monkey. Endocrinology 97:425–430. Richards S, Merriman D, Calhoun LH. 1963. Studies on the marine resources of southern New England. IX. The biology of the little skate, Raja erinacea Mitchell. Bull Bingham Oceanogr Coll 18:5–67. Snelson FF Jr, Rasmussen LEL, Johnson MR, Hess DL. 1997. Serum concentrations of steroid hormones during reproduction in the Atlantic stingray, Dasyatis sabina. Gen Comp Endocrinol 108:67–79. Sokal RR, Rohlf FJ. 1995. Biometry. New York: Freeman and Company. Sumpter JP, Dodd JM. 1979. The annual reproductive cycle of the female lesser spotted dogfish (Scyliorhinus canicula) and its endocrine control. J Fish Biol 15:687–695. Tsang P, Callard IP. 1987. Morphological and endocrine correlates of the reproductive cycle of the aplacental viviparous dogfish, Squalus acanthias. Gen Comp Endocrinol 66:182–189. Wourms JP. 1977. Reproduction and development in chondrichthyan fishes. Am Zool 32:276–293.