INTRODUCTION
Sea urchins have long been used in embryology and developmental biology as optimal experimental models and many aspects of male and female gamete development are well known, especially from a cellular point of view. Nevertheless, only few data are available on the specific hormonal mechanisms regulating reproductive processes. In this paper we focused on steroids, specifically oestradiol (E2), one of the best known sex hormones in vertebrates. In the last decades several authors have suggested a plausible involvement of steroids in different physiological processes of echinoderms including reproduction, development and regeneration (Watts et al., Reference Watts, Hines, Byrum, McClintock, Marion, David, Guille, Feral and Roux1994; Candia Carnevali, Reference Candia Carnevali and Matranga2005; Sugni et al., Reference Sugni, Mozzi, Barbaglio, Bonasoro and Candia Carnevali2007, Reference Sugni, Barbaglio, Tremolada, Candia Carnevali, Chen and Guô2008). In vertebrates E2 plays a major role in the development and regulation of female reproductive organs and reproductive cycle respectively (Crews, Reference Crews1994; Clinton & Haines, Reference Clinton and Haines1999; Wallace et al., Reference Wallace, Badawy and Wallace1999). When exogenously administered to vertebrate animals, E2 can act as a feminizing agent in male individuals, modifying primary and secondary sexual characters. It may also alter the gonad index (GI: also called gonadosomatic index, GSI) and gonad and/or plasma biochemical components such as proteins, mainly yolk precursors, and lipids (Crews et al., Reference Crews, Cantù, Rhen and Vohra1996; Spearow et al., Reference Spearow, Doemeny, Sera, Leffler and Barkley1999; Wallace et al., Reference Wallace, Badawy and Wallace1999; Omoto et al., 2002; Hecker et al., Reference Hecker, Wan Jong, Park, Murphy, Villeneuve, Coady, Jones, Solomon, Van der Kraak, Carr, Smith, Du Preez, Kendall and Giesy2005). Although far less information is available on the role of E2 in echinoderm reproductive biology, its possible involvement in reproductive cycle regulation of representative echinoderm species was recently suggested (Watts et al., Reference Watts, Hines, Byrum, McClintock, Marion, David, Guille, Feral and Roux1994; LeBlanc et al., Reference LeBlanc, Campbell, den Besten, Brown, Chang, Coats, deFur, Dhadialla, Edwards, Riddiford, Simpson, Snell, Thorndyke, Matsumura, de Fur, Crane, Ingersoll and Tattersfield1999; Barbaglio et al., Reference Barbaglio, Sugni, Di Benedetto, Bonasoro, Schnell, Lavado, Porte and Candia Carnevali2007; Sugni et al., Reference Sugni, Mozzi, Barbaglio, Bonasoro and Candia Carnevali2007, Reference Sugni, Tremolada, Porte, Barbaglio, Bonasoro and Candia Carnevali2010). The presence of this hormone was revealed in different echinoderm classes, mainly in asteroids, in whole body homogenates, in coelomic fluids, or in different specific tissues such as gonads and pyloric caeca (asteroids), and recently even in eggs and embryos (Dieleman & Schöenmakers, Reference Dieleman and Schöenmakers1979; Xu & Barker, Reference Xu and Barker1990; Hines et al., Reference Hines, Watts, Sower and Walker1992; Wasson et al., Reference Wasson, Gower, Hines and Watts2000a; Lavado et al., Reference Lavado, Barbaglio, Candia Carnevali and Porte2006a, Reference Lavado, Sugni, Candia Carnevali and Porteb; Roepke et al., Reference Roepke, Chang and Cherr2006; Barbaglio et al., Reference Barbaglio, Sugni, Di Benedetto, Bonasoro, Schnell, Lavado, Porte and Candia Carnevali2007). In echinoderms, the endogenous synthesis of oestradiol (and oestrogens, in general) was under debate until recently. Several published results demonstrated the presence of vertebrate-like steroid metabolic pathways in different echinoderm tissues and described those involved in androgen metabolism (Schöenmakers & Voogt, Reference Schöenmakers and Voogt1980, Reference Schöenmakers and Voogt1981; Voogt et al., Reference Voogt, den Besten and Jansen1991; Watts et al., Reference Watts, Hines, Byrum, McClintock, Marion and Hopkins1993, Reference Watts, Hines, Byrum, McClintock, Marion, David, Guille, Feral and Roux1994; Wasson et al., Reference Wasson, Gower, Hines and Watts2000a; Janer et al., Reference Janer, LeBlanc and Porte2005a, Reference Janer, Sternberg, LeBlanc and Porteb). However, these papers failed to demonstrate P450-aromatase activity, the enzyme responsible for the conversion of androgens to oestrogens in vertebrates (Hines et al., 1994; Wasson et al., Reference Wasson, Gower, Hines and Watts2000a). A few recent studies measured an endogenous synthesis of oestrogens from androgens in echinoderms, suggesting the presence of an aromatase-like enzyme (Lavado et al., Reference Lavado, Sugni, Candia Carnevali and Porte2006b; Barbaglio et al., Reference Barbaglio, Sugni, Di Benedetto, Bonasoro, Schnell, Lavado, Porte and Candia Carnevali2007).
The first evidence of E2 involvement in the regulation of an echinoderm reproductive cycle was suggested by the pilot-experiments of Donahue & Jennings (Reference Donahue and Jennings1937), showing that gonadal extracts from the echinoid Lytechinus variegatus could induce a strong oestrogen-like response in both vagina and uterus of ovariectomized rats. This hypothesis was confirmed by several studies showing that oestrogen levels and/or metabolism can vary according to the reproductive cycle and in a sex-specific manner (Schöenmakers & Dieleman, Reference Schöenmakers and Dieleman1981; Voogt & Dieleman, Reference Voogt and Dieleman1984; Xu & Barker, Reference Xu and Barker1990; Hines et al., Reference Hines, Watts, Sower and Walker1992; Wasson et al., Reference Wasson, Gower, Hines and Watts2000a). Particularly, in the sea urchin Paracentrotus lividus higher E2 levels were found in those early stages characterized by the higher activity of nutritive phagocytes, i.e. the somatic cells of the echinoid gonad (Barbaglio et al., Reference Barbaglio, Sugni, Di Benedetto, Bonasoro, Schnell, Lavado, Porte and Candia Carnevali2007). The same authors reported sex-specific differences in the overall E2 titres, females displaying higher mean levels than males (Barbaglio et al., Reference Barbaglio, Sugni, Di Benedetto, Bonasoro, Schnell, Lavado, Porte and Candia Carnevali2007).
A more specific involvement of E2 in echinoderm reproduction is also suggested by a number of experiments using direct hormone administration, in which steroid treatment of individuals or cultured tissues resulted in appreciable physiological effects including increased oocyte diameter/area, ovarian growth, protein synthesis, lipid amount and mitotic activity of male germ cells (Shöenmakers et al., Reference Schöenmakers, Van Bohemen and Dieleman1981; Takahashi & Kanatani, Reference Takahashi and Kanatani1981; Van der Plas et al., Reference Van der Plas, Koenderman, Deibel-van Schijndel and Voogt1982; Harrington & Ozaki, Reference Harrington and Ozaki1986; Watts & Lawrence, Reference Watts and Lawrence1987; Barker & Xu, Reference Barker and Xu1993; Unuma et al., Reference Unuma, Yamamoto and Akiyama1999; Wasson et al., Reference Wasson, Gower and Watts2000b; Roepke et al., Reference Roepke, Chang and Cherr2006). However, most of these experiments were focused on asteroid species and only limited evidence is available on echinoids.
In the present work we explored echinoderm endocrinology by using P. lividus as the experimental model. Recently E2 and T, as well as their metabolic pathways, were found and investigated in different tissues of P. lividus (Janer et al., Reference Janer, LeBlanc and Porte2005a; Lavado et al., Reference Lavado, Sugni, Candia Carnevali and Porte2006b; Barbaglio et al., Reference Barbaglio, Sugni, Di Benedetto, Bonasoro, Schnell, Lavado, Porte and Candia Carnevali2007). In addition, this species was sensitive to exposure to endocrine disrupter contaminants (EDs) (Janer et al., Reference Janer, Sternberg, LeBlanc and Porte2005b; Lavado et al., Reference Lavado, Sugni, Candia Carnevali and Porte2006b; Sugni et al., Reference Sugni, Mozzi, Barbaglio, Bonasoro and Candia Carnevali2007, Reference Sugni, Tremolada, Porte, Barbaglio, Bonasoro and Candia Carnevali2010), which are compounds well known for interfering with the endocrine system of various vertebrate and invertebrate species.
Paracentrotus lividus is a typical gonochoristic species, although hermaphroditic specimens have been occasionally found (Bacci, Reference Bacci1954; Byrne, Reference Byrne1990; authors, personal observations). In the Mediterranean area it is one of the most common sea urchins and it is a well known edible species (and therefore commercially important), its gonads being highly appreciated in many coastal regions.
Overall, the aim of this investigation was to provide further information on the scarcely known endocrinology of echinoderms, with particular reference to echinoids. Our specific goals were: (1) to analyse thoroughly echinoid reproductive processes and their regulatory mechanisms by focusing on the possible specific role of E2; (2) to provide the necessary background knowledge for developing a valid ecotoxicological model for future studies on endocrine dysfunctions; and (3) to provide new information on possible control mechanisms of gonad development in this edible and commercially important species.
MATERIALS AND METHODS
Experimental animals and maintenance
Adult Paracentrotus lividus specimens were collected from the Ligurian coast of Italy (Tyrrhenian Sea), in a protected marine area (Area Marina Protetta Isola di Bergeggi, 44°14′N 8°26′E). Then the animals were immediately transferred to the laboratory, at the University of Milan. Animal collection took place in April 2005, which, in that year, corresponded to the month of maximum maturation level and spawning of the local sea urchin population. Mean ambital diameter was 35–50 cm. The animals were maintained in 50 l glass aquaria filled with artificial seawater throughout the experimental period (Instant Ocean; salinity about 37‰, as in the Mediterranean Sea) and provided with an internal close-circulation system as well as mechanical, chemical and biological filters. Temperature was set at 16 ± 1°C (close to the environmental temperature in April) and the photoperiod was fixed at 12 hours:12 hours. Except for the first pre-administration period, animals were individually fed with fresh lettuce twice a week and with 2 g mini-blocks of an artificial diet (Table 1) only once a week, in order not to increase excessively nitrite and nitrate levels. Animal conditions as well as temperature and salinity were daily monitored throughout the experimental period (including the pre-administration phase); once a week pH, nitrite and nitrate levels were checked and nitrifying bacteria (Bio-Nitrivec, Sera) were added to the aquarium water (1 ml solution/1l artificial seawater).
Table 1. Composition of the artificial diet.
*, Dictyota dicotoma and Caulerpa prolifica.
Experimental design and oestradiol administration
On their arrival in the laboratory, 20 animals were immediately weighed and sacrificed, and their gonads were removed for GI calculation and histological processing (see below). These animals represented the ‘time 0’ condition. The remaining 30 animals were starved for four weeks (pre-administration phase). According to Spirlet et al. (Reference Spirlet, Grosjean and Jangoux2000) starvation should reset the reproductive cycle to a resting phase, in which gonads are almost devoid of sexual cells. Feeding re-started one week before the first E2 administration. Hormone administration occurred twice a week via intra-coelomic injection at the level of the peristomial membrane (which surrounds the mouth). Twice-weekly injections were preferred to daily administration to reduce animal stress and mortality. Three groups of 10 specimens each were set up in separate aquaria: one control and two different E2 concentrations (5 and 50 ng ml−1). These latter were selected taking into account: (1) E2 levels previously measured in P. lividus coelomic fluid (Lavado et al., Reference Lavado, Sugni, Candia Carnevali and Porte2006b); and (2) previous E2 administration experiments carried out on asteroids (Van der Plas et al., Reference Van der Plas, Koenderman, Deibel-van Schijndel and Voogt1982; Barker & Xu, Reference Barker and Xu1993). Specimens smaller than 40 mm diameter were injected with 100 µl of solution, whereas 200 µl were administered to sea urchins larger than 40 mm. The overall period of hormone administration was 10 weeks, after which the animals were weighed and sacrificed; all the five gonads were removed and weighed for GI calculation: GI= (gonad fresh weight/animal fresh weight) × 100. One gonad was processed for histological analyses and maturation stage evaluation (see below) whereas those remaining were frozen at –20°C until use for lipid and water content measurement.
Chemicals and solution preparation
All chemicals were of reagent grade. 17-ß oestradiol was purchased from Sigma.
The oestradiol solution was prepared as follows: 50 mg E2 were dissolved in 10 ml acetone (Merck) then 1 ml of this solution was dissolved in 10 ml acetone. Finally, 20 µl of this latter solution was dissolved in 200 ml of autoclaved and filtered (0.2 µm filters) artificial seawater (ASW) so that a final concentration of 50 ng ml−1 E2 was obtained (higher tested concentration). The lower tested concentration was obtained by a further dilution (1:10) in ASW. Maximum used acetone concentration was 0.01% (in the higher tested E2 solution). Control solution was prepared by adding 20 µl acetone in 200 ml ASW (0.01%). In order to further reduce acetone concentration, which is a highly volatile compound all the three solutions were left to evaporate for half an hour on a stirring magnet: this should allow acetone evaporation without markedly changing ASW salinity. Seawater was used as final carrier-solution since it has an osmolarity very similar to echinoderm body fluids. The obtained stock solutions were subdivided in aliquots and maintained at 4°C in the dark, in order to prevent oestradiol degradation. New stock solutions were prepared every 2 weeks.
Maturation stage evaluation
Maturation stages were determined by histological analysis. Standard methods for light microscopy were employed, as described by Barbaglio et al. (Reference Barbaglio, Sugni, Di Benedetto, Bonasoro, Schnell, Lavado, Porte and Candia Carnevali2007). Briefly, gonads were pre-fixed with 2% glutaraldehyde in 0.1 M cacodylate buffer for 2 hours, then, after overnight washing in the same buffer, post-fixed with 1% osmium tetroxide in the same buffer (2 hours). After standard dehydration in an ethanol series, the samples were embedded in Epon-Araldite 812. The semithin sections, cut with a Reichert Ultracut E were stained by conventional methods (crystal violet-basic fuchsin) and then observed with a Jenaval light microscope.
Five reproductive stages were considered following Unuma (Reference Unuma, Yokota, Matranga and Smolenicka2002): Spent (immediately after the spawning event), Recovery (phagocytosis and nutrient accumulation phase), Growing, Premature and Mature (progressive stages of active gametogenesis).
Lipid and water content analyses
Frozen gonads were freeze-dried for 24 hours, weighed and minced with a pestle. Lipid extraction with hexane (Fluka) was carried out by the Soxhlet method for about eight hours. Samples were then concentrated with Rotavapor, dried under nitrogen flow and weighed to obtain the total amount of lipids. Water content was calculated as the difference between wet and dry weights of each sample.
Statistical analyses
Statistical analyses of the GI, lipid percentage and water content data were performed by using the one-way analysis of variance test (whenever normality and homogeneity of variance were verified; Dunnet's test for multiple comparisons) or Kruskal–Wallis test (Dunn's test for multiple comparisons). Sex-ratio was analysed by χ2 test. Statistical analyses were performed by the computer program GraphPad Prism 4 or Excel 2000.
RESULTS
Animal health
Only one animal of the 50 ng ml−1 group died during the whole experimental period. This animal died three days after the first E2 injection. The remaining specimens appeared to be healthy throughout the experiment.
One week after the beginning of the E2 administration period (five weeks from the field collection and after 2 injections) some animals (both males and females) in each of the three experimental aquaria simultaneously released a considerable amount of gametes. Random sampling in each of the three aquaria revealed a high percentage of fertilization events and the presence of viable embryos (gastrula stage).
Sex-ratio
The sex-ratio values measured at the end of the experimental period were different between control and treated groups (Figure 1), even though this difference was not statistically significant (P > 0.05). In the control group a 1:1 ratio was found, whereas in the 5 ng ml−1 group females were more numerous than males (70% the former). This tendency was more evident in the animals treated with the higher E2 concentration where the female percentage slightly increased reaching about 80%.
Fig. 1. Sex-ratio calculated in the different experimental groups. CTL, control.
Similarly to the control group, in the ‘time 0’ group the sex-ratio was again close to 1:1 (55% females–45% males; Figure 1).
Gonad index
At the end of the hormonal treatment period, mean animal weight was very similar among the three experimental groups (P > 0.05), with mean values of 42.36 ± 7.72 for the control, 39.30 ± 2.97 for the 5 ng ml−1 group and 40.27 ± 4.68 for the 50 ng ml−1 group. Although not statistically significant, a GI reduction could be observed in both E2-treated groups and in both sexes, the higher tested concentration being more effective (Figure 2): female specimens appeared to be more sensitive and displayed a 33% (5 ng ml−1 group) and a 42% (50 ng ml−1) GI decrease compared with the control, with mean values of 1.57 ± 0.59 and 1.38 ± 0.52 respectively (control group: 2.34 ± 1.04); males displayed a 24% (5 ng ml−1 group) and 30% (50 ng ml−1) reduction when compared with the controls, with mean values of 1.82 ± 1.05 and 1.69 ± 1.19 respectively (control group: 2.40 ± 1.07).
Fig. 2. Gonad index (GI) values (mean ± SEM) in females and males of the different experimental groups. CTLs, controls.
Mean GI in the ‘time 0’ group was 6.21: 6.89 for the females and 5.54 for males (data not shown).
Lipid and water content
Lipid content (expressed as percentage of the total gonad weight) was calculated on the basis of both fresh and dry weight.
Overall, mean lipid content was higher in the E2 treated groups, considering both fresh (P < 0.05 in the 50 ng ml−1 group, males and females pooled together; Figure 3A) and dry weight. Particularly, considering the latter, sex-specific tendencies could be observed (Figure 3B): although not statistically significant, female specimens displayed a dose-dependent increase, whereas in males of the lower E2 tested concentration (5 ng ml−1) slightly higher lipid content was measured if compared with the 50 ng ml−1 group. In females mean dry weight increased by about 20% and 30% (when compared to controls) in the 5 ng ml−1 and 50 ng ml−1 respectively. Mean values registered in the different groups are reported in Table 2.
Fig. 3. Gonad lipid content: (A) mean lipid percentages (±SEM), expressed as wet weight, in the different experimental groups. Males and females pooled together; (B) mean lipid percentages (±SEM), expressed as dry weight, in the different experimental groups. *, P < 0.05; CTLs, controls.
Table 2. Lipid content (expressed as percentage of the total gonad weight) in male and female specimens of the three experimental groups (mean ± SEM). N = 2–5.
On the contrary, water content (%) was decreased by the E2 treatment, since in the control group a mean value of 73.33 ± 2.42 was measured whereas 7.5% and 16.5% reductions were found in the 5 ng ml−1 and 50 ng ml−1 (P < 0.05, males and females pooled together) respectively.
Maturation stage
No difference (P > 0.05) was detected between the control and E2-treated group in the relative frequency of the gonad maturation stages (Figure 4). Almost all the samples were in the Recovery stage, i.e. most of the gonad lumen was usually occupied by well-developed nutritive phagocytes, rich in different granule types, whereas germinal cells were often present in low number and were situated at the periphery of each tubule section (Figure 5A). In the only two male samples of the 50 ng ml−1 group, germinal cells were so rare and scarcely differentiated to make sex identification very difficult, and only at high magnification some reabsorbing spermatozoa could be observed within the phagocytes. A similar but less marked situation was found in one male of the 5 ng ml−1 group. Only one female specimen, displaying ova and oocytes at different developmental stages (Premature), was found in the 50 ng ml−1 group (Figure 5B).
Fig. 4. Relative frequency of the maturation stages in the different experimental groups. CTLs, controls.
Fig. 5. Semithin sections of a tubule: (A) control sample in Recovery stage: the tubule contains a meshwork of nutritive phagocytes (p) filled by different granule types. w, acinal wall. Scale bar = 50 µm; (B) 50 ng ml−1 sample in Premature stage: large developing oocytes (o) can be found at the periphery of the tubule section whereas the lumen is still filled by nutritive phagocytes. n, nucleus; scale bar = 50 µm.
No mature samples were found in any of the experimental groups.
DISCUSSION
There was almost no mortality throughout the experimental period and the single death in the 50 ng ml−1 group, occurring after the first injection, was probably not caused by oestradiol administration: if so, further deaths would have been expected in the following nine weeks of treatment. Therefore, the tested E2 concentrations did not cause acute toxicity effects, as confirmed by the relatively good health conditions displayed by the hormonally treated animals.
Despite starvation for 40 days, some animals did not totally resorb the gametogenic portion of the gonads and, at the beginning of the injection period, still had a considerable number of viable gametes (most of which could give rise to embryos: authors, personal observations). According to Spirlet et al. (Reference Spirlet, Grosjean and Jangoux2000) starvation should force the gonads into a maturation stage characterized by few (or no) relict ova or spermatozoa that will be eventually resorbed by the somatic cells (nutritive phagocytes). The difference between their results and ours is probably due to different starvation periods, that of Spirlet et al. (Reference Spirlet, Grosjean and Jangoux2000) being slightly longer (two months instead of 40 days). Field specimens often display gonads with a well developed meshwork of phagocytes surrounding a central mass of stocked and apparently viable gametes (authors, personal observations). Accumulated and ‘saved’ ova and spermatozoa might be subsequently released in response to a specific stimulus. In many animals spawning is hormonally triggered and most often steroids, E2 in particular, are involved. Nevertheless, in the present study the observed release of gametes was probably not induced by E2 administration as it took place also in the control aquaria. The simultaneous spawning event that occurred independently in all the three aquaria might be better related to a lunar influence since there was a full moon on the night of the gamete release. The lunar-phase control of spawning has also been suggested by Holland (1991) for another echinoderm, the crinoid Oxycomanthus japonica.
The hormone apparently influenced the sex-ratio calculated at the end of the administration period: a higher percentage of females were found in the E2 treated samples and this effect was more evident at the highest concentrations (Figure 1). However, it is unlikely that the hormone caused complete sex reversal in adult sea urchins. According to the literature E2-induced sex reversal of adult animals is very rare and is reported for only those species (mainly fish) that are naturally ‘programmed’ to change sex during their life (hermaphroditic species; Chang et al., Reference Chang, Lau and Lin1995; Pandian & Sheela, Reference Pandian and Sheela1995). E2 is well known as a feminizing agent of vertebrate embryos: even in those groups with a genetic sex determination mechanism, exposure/administration to this hormone during early life stages (i.e. during sexual differentiation) induces the conversion of genotypically male specimens to phenotypically females (sex-reversion; Baroiller et al., Reference Baroiller, Guiguen and Fostier1999; Clinton & Hines, 1999; Sheehan et al., Reference Sheehan, Willingham, Gaylor, Bergeron and Crews1999; Wallace et al., Reference Wallace, Badawy and Wallace1999; Coveney et al., Reference Coveney, Shaw and Renfree2001). The specimens employed in this research were adults far from the sexual differentiation period, therefore unlikely to be influenced by E2 administration. This conclusion is further suggested by the lack of any ‘intermediate’ situation of sex reversal (i.e. simultaneous hermaphroditic individuals) in E2-treated samples. The strongly unbalanced sex-ratio observed in the group treated with the higher dose is probably due to an unbalanced distribution of male and female individuals in the aquaria at the beginning of the experiment. As with most sea urchins, P. lividus lacks marked signs of sexual dimorphism and therefore it is difficult to recognize the sex a priori, thus preventing an homogeneous distribution of the sexes in the different experimental group. Although this species is usually expected to have a sex-ratio close to 1:1 (due to its genetic sex determination mechanism; Lipani et al., Reference Lipani, Vitturini, Sconzo and Barbata1996), in our past field samplings of the P. lividus population from which these same experimental animals are collected (see Materials & Methods), we often observed a female-unbalanced sex-ratio (authors, personal observations) and this tendency was also visible in the ‘time 0’ samples (Figure 1).
Although not statistically significant, the observed decrease in the female GI (Figure 2) perfectly matches with other recent findings: in physiological conditions higher E2 levels correspond to low ovary index values in P. lividus (Barbaglio et al., Reference Barbaglio, Sugni, Di Benedetto, Bonasoro, Schnell, Lavado, Porte and Candia Carnevali2007), suggesting that this hormone could be involved in the first phase of the gonad cycle. A similar relationship was found during the annual reproductive cycle of the sea urchin L. variegatus (Wasson et al., Reference Wasson, Gower, Hines and Watts2000a) and of the asteroid Sclerasterias mollis (Xu & Barker, Reference Xu and Barker1990). However, different results were reported when E2 was exogenously administered to the same echinoid species and to the starfish Asterias rubens (increased ovarian growth; Schöenmakers et al., Reference Schöenmakers, Van Bohemen and Dieleman1981; Van der Plas et al., Reference Van der Plas, Koenderman, Deibel-van Schijndel and Voogt1982; Wasson et al., Reference Wasson, Gower and Watts2000b) and S. mollis (no effect on the GI; Barker & Xu, Reference Barker and Xu1993). The differences observed could be related to a species-specific hormonal mechanism or to different experimental conditions (E2 administration type, treatment length, etc). In most vertebrates E2 injection or exposure causes a decrease in male GI (Spearow et al., Reference Spearow, Doemeny, Sera, Leffler and Barkley1999; Kinberg et al., 2000; Sorenson et al., Reference Sorensen, Shoenfuss, Adelman and Swackhamer2001; Hecker et al., Reference Hecker, Wan Jong, Park, Murphy, Villeneuve, Coady, Jones, Solomon, Van der Kraak, Carr, Smith, Du Preez, Kendall and Giesy2005) although contrasting results (i.e. male GI increase) can be found even within this well studied animal group (Lerner et al., Reference Lerner, Björnsson and McCormick2007).
In order to understand more deeply the observed GI reduction, we decided to analyse the gonad biochemical components, focusing particularly on lipids. In vertebrates the latter have long been known to be important storage materials. In our experiment, despite the GI decrease previously described, E2 treatment appeared to increase the mean lipid content of the gonads, suggesting its involvement in the incorporation and rearrangement of lipid material within the gonad (Figure 3). In physiological conditions higher E2 levels were reported during those reproductive stages characterized by nutrient accumulation and processing (Barbaglio et al., Reference Barbaglio, Sugni, Di Benedetto, Bonasoro, Schnell, Lavado, Porte and Candia Carnevali2007). This hypothesis is further suggested by studies on other echinoderm species. Indeed, in the sea urchin L. variegatus E2 administration (in combination with progesterone) increases the lipid content of the gonads (Wasson et al., Reference Wasson, Gower and Watts2000b). In the asteroid A. rubens E2 treatment increases the lipid content in the pyloric ceaca, the nutrient storage site of starfish (Van der Plas et al., Reference Van der Plas, Koenderman, Deibel-van Schijndel and Voogt1982). In many vertebrate species lipids are accumulated before breeding (Bullough, 1962) and their endogenous dynamics can be altered by both natural (E2) and synthetic oestrogens (Aftergood & Alfin-Slater, Reference Aftergood and Alfin-Slater1965; Bertolotti & Spady, Reference Bertolotti and Spady1996; Sharpe & MacLatchy, Reference Sharpe and MacLatchy2007). In the catfish Heteropneustes fossilis hepatic lipid concentrations closely follow the reproductive cycle (Singh & Singh, Reference Singh and Singh1990). In echinoderms, the hypothesis that nutrient accumulation within the gonads is hormonally controlled is supported by previous studies by Shyu et al. (Reference Shyu, Blumenthal and Raff1987), according to whom, the expression of the mayor yolk protein (MYP), the most abundant protein of sea urchin yolk plates, might be under oestrogen control, as suggested by the presence of a vertebrate-like oestrogen responsive element within the encoding MYP gene.
The last parameter taken into account in the present work was the maturation stage. This was not influenced by the E2 administration, as a similar relative percentage of stages was found among the three experimental groups, the most frequent stage being the Recovery stage (Figure 4). Therefore the present study suggetst that E2 does not induce oocyte development/maturation as observed in starfish (Takahashi & Kanatami, 1981; Schöenmakers et al., Reference Schöenmakers, Van Bohemen and Dieleman1981; Barker & Xu, Reference Barker and Xu1993) and in vertebrates. Several explanations could be considered for these different results: (1) class-specific mechanism: echinoids and asteroids could simply have a different hormonal regulation of oocytes development; this hypothesis is supported by Wasson et al. (Reference Wasson, Gower and Watts2000b) who reported the inhibited growth of individual oocytes after dietary administration of E2 in the echinoid Lytechinus variegatus; and (2) specific reproductive windows of sensitivity: according to Schöenmakers et al. (Reference Schöenmakers, Van Bohemen and Dieleman1981), there is a threshold size for E2 effectiveness, as only already developed oocytes appeared to be positively affected by the hormone whereas no effects were observed on small size oocytes. Unlike all previous works on both asteroids and echinoids, in the present experiment E2 was administered to previously starved animals, theoretically deprived of developing gametes (the specimens still having ripe gametes spawned them at the very beginning of the administration period; mentioned in the latter). Therefore, the lack of effects on oocyte development could be due to the absence of the proper E2 sensitive targets (i.e. oocytes over a certain threshold size).
Although preliminary, overall the present research improves knowledge of basic echinoderm endocrinology, providing a new perspective about the role of E2 in echinoderm reproductive processes. Our past and present results suggest that in sea urchins E2 could have a function different from that reported for vertebrates and also suggested for asteroid echinoderms. As previously mentioned, especially in the latter the hormone appears to influence oocyte development more or less directly; on the contrary our results suggest that, at least in echinoid females, E2 could be involved in the regulation of nutritive phagocytes activities, i.e. phagocytosis processes and/or nutrient (lipid?) accumulation, gamete maturation being possibly controlled directly by other factors. Many echinoderm processes (including reproduction) are regulated by neuroendocrine factors (LeBlanc et al., Reference LeBlanc, Campbell, den Besten, Brown, Chang, Coats, deFur, Dhadialla, Edwards, Riddiford, Simpson, Snell, Thorndyke, Matsumura, de Fur, Crane, Ingersoll and Tattersfield1999; Thorndyke & Candia Carnevali, Reference Thorndyke and Candia Carnevali2001). For example, in starfish spawning can be stimulated by a specific nervous factor (gonad stimulating substance, or radial nerve factor) released by the radial nerve cords (Caine & Burke, Reference Caine, Burke, Keegan and O'Connor1985) and targeting specifically the follicular cells surrounding the oocyte (Smiley, Reference Smiley1990; Mita, Reference Mita2000). Thus E2 may have a broad spectrum role in the reproductive biology of echinoderms. Further experiments including different types of analyses (e.g. hormone levels and enzymatic activities), different concentrations and a higher number of specimens will help to clarify this hypothesis.
ACKNOWLEDGEMENT
This research work was supported by the MIUR COFIN 2003 Research Project (Coordinator: Professor M.D. Candia Carnevali).
