17 -Hydroxysteroid Dehydrogenase (17 -HSD) in Scleractinian Corals

17-Hydroxysteroid dehydrogenase (17-HSD) in scleractinian corals and zooxanthellae

Charles H. Blomquista,e,*, P.H. Limaa,e, A.M. Tarrantb, M.J. Atkinsonc and S. Atkinsond

aDepartment of Obstetrics and Gynecology, RegionsHospital, St. Paul, USA

bBiology Department, Woods Hole Oceanographic Institution, Woods Hole, USA

cccUniversitycHawaii Institute of Marine Biology, University of Hawaii at Manoa, Kaneohe, USA

dAlaska SeaLifeCenter, University of Alaska Fairbanks, Seward, USA

eDepartment of Obstetrics, Gynecology and Women’s Health, University of Minnesota, Minneapolis, USA

*Corresponding author. Department of Obstetrics, Gynecology and Women’s Health, University of Minnesota Medical School, Mayo Mail Code 395, 420 Delaware St. SE, Minneapolis, MN 55455, USA. Tel.: +1-612-626-4056; fax: +1-612-626-0665.

E-mail address:

Running title: Steroid dehydrogenase in corals and zooxanthellae

Abstract

Steroid metabolism studies have yielded evidence of 17-hydroxysteroid dehydrogenase (17-HSD) activity in corals. This project was undertaken to clarify whether there are multiple isoforms of 17-HSD, whether activity levels vary seasonally, and if zooxanthellae contribute to activity. 17-HSD activity was characterized in zooxanthellate and azooxanthellate coral fragments collected in summer and winter and in zooxanthellae cultured from M. capitata. More specifically, 17-HSD activity was characterized with regard to steroid substrate and inhibitor specificity, coenzyme specificity, and Michaelis constants for estradiol (E2) and NADP+. Six samples each of M. capitata and T. coccinea (three summer, three winter) were assayed with E2 and NADP+. Specific activity levels (pmol/mg protein) varied 10-fold among M. capitata samples and 6-fold among T. coccinea samples. There was overlap of activity levels between summer and winter samples. NADP+/NAD+ activity ratios varied from 1.6 to 22.2 for M. capatita, 2.3 to 3.8 for T. coccinea and 0.7 to 1.1 for zooxanthellae. Coumestrol was the most inhibitory of the steroids and phytoestrogens tested. Our data confirm that corals and zooxanthellae contain 17-HSD and are consistent with the presence of more than one isoform of the enzyme.

Keywords: 17-hydroxysteroid dehydrogenase; steroid; coral; invertebrate; zooxanthellae; phytoestrogens; estradiol; Scleractinia

1.Introduction

Estrogens, androgens and progestins (vertebrate-type sex steroids, referred to as “sex steroids” hereafter) play a fundamental role in regulating of reproduction and other physiological processes in vertebrates. Sex steroids have also been detected in several invertebrate phyla (Tarrant et al. 1999, Wasson et al., 2000, Verslycke et al., 2002), but ligand-activated orthologs of the sex steroid receptors have not been identified in invertebates (Escriva et al., 1997, Maglich et al., 2001, Thornton et al., 2003). While the mechanism of action of sex steroids in invertebrates is unknown, exogenous sex steroids are biologically active in many invertebrates (Li et al., 1998, Tarrant et al. 2004, Novillo et al., 2005).

Among vertebrates, the levels of sex steroids within tissues are regulated by a variety of steroidogenic enzymes. 17-HSD catalyzes the reversible oxidation and reduction of biologically active 17-hydroxysteroids and relatively inactive 17-ketosteroids and as such plays a gate-keeping role with regard to receptor occupancy (Penning 1997). In vitro incubations have demonstrated 17-HSD activity in diverse invertebrates including echinoderms (Wasson et al., 1998), crustaceans (Swevers et al., 1991), mollusks (Le Curieux-Belfond et al., 2001) and cnidarians (Slattery et al., 1997, Tarrant et al., 2003). Consistent with vertebrates, rates and patterns of 17-HSD activity can vary with tissue, reproductive condition and sex (Matsumoto et al., 1997, Janer et al., 2005).

Steroidogenic enzyme activities have been reported in alcyonacean soft corals (Slattery et al., 1997) and scleractinian hard corals (Gassman, 1992; Tarrant et al., 2003). We recently reported that scleractinian corals and zooxanthellae (endosymbiotic dinoflagellates) contain 17-HSD activity with estradiol (E2), testosterone (T) and other C18- and C19-hydroxysteroids (Tarrant et al., 2003). Androgens and estrogens have been measured and reported to vary seasonally in several cnidarian species (Slattery et al., 1999; Tarrant et al., 1999; Pernet and Anctil, 2002; Twan et al., 2003). While it has been hypothesized that changes in steroid concentrations help to regulate coral reproductive processes, the mechanism of steroid action has not been elucidated in corals. It is not known whether variation in steroid concentration is caused by variation in endogenous rates of steroidogenesis. Further, the contribution of zooxanthellae contained within the tissues of some corals to steroid metabolism has not been clarified.

It is now well-established that there are multiple isoforms of 17-HSD not only in vertebrates, but in microorganisms and invertebrates as well (Lanišnik Rižner et al., 1999; Breitling et al., 2001). The various isoforms are products of separate genes and differ in substrate and cofactor specificity, subcellular localization and their predominant direction of reaction in intact cells (Peltoketo et al., 1999; Mindnich et al., 2004). Kinetic methods have proven useful in characterizing multiple isoforms of this enzyme in mammalian tissues (Blomquist, 1995). Because little is known of the possible complexity of the 17-HSD content in corals, we used similar kinetic methods to further characterize 17-HSD in two species of corals and zooxanthellae.

2. Materials and Methods

2.1. Preparation of homogenates

Fragments of Montipora capitata and Tubastrea coccinea were collected in December 2002 and July 2003 from the reef flat adjoining CoconutIsland (KaneoheBay, Oahu, HI). Zooxanthellae ("flat plate" strain cultured from M. capitata) were provided by Dr. R. Kinzie (University of Hawaii at Manoa, Honolulu, HI). M. capitata is a hermaphroditic reef-building coral that contains zooxanthellae. In Hawaii M. capitata generally releases bundles of eggs and sperm 1-4 days after the new moon in June, July and August (Hunter, 1988). T. coccinea is a fully heterotrophic coral that does not contain zooxanthellae. Reproduction by T. coccinea is less synchronous; colonies release larvae during from June to January in all lunar phases (Kolinski and Cox, 2003).

Coral homogenates and sonicates of zooxanthellae were prepared as described previously (Tarrant et al., 2003).Coral fragments were weighed and pulverized with a chilled mortar and pestle in ice-cold 0.01 M potassium phosphate buffer, pH 7.0, supplemented with 20% (v/v) glycerol and 1.0 mM EDTA. Homogenates were then prepared in an Ultra-Turrax T25 homogenizer on the lowest setting using three 10-s bursts with 1-min cooling between bursts followed by centrifugation at 1380 x g for 5 min. The supernatants were used for enzyme assays. Suspensions of zooxanthellae were sonicated for two 1.0-min intervals in a bath-type sonicator (Heat Systems, Inc.).

2.2 Enzymatic activity measurements

Kinetic studies were carried out as described previously (Blomquist et al., 1994; Blomquist, 1995). Briefly, 10-l aliquots of homogenate or sonicate were combined with 10 l of reaction buffer containing cofactor, tritiated substrate and additives and incubated at room temperature. Reaction mixtures were transferred in total to the preadsorbent layer of silica gel thin-layer chromatography plates and fractionated into substrate and product by chromatography with benzene:acetone (4:1) as the solvent. Further details are given in the figure legends. Specific activities are expressed as nmol or pmol/mg protein. Protein was measured by the method of Markwell et al. (1981) with bovine serum albumin as the protein standard.

2.3 Data analysis

Michaelis constants (Km) were estimated by the graphical method of Cornish-Bowden and Eisenthal (1978), as described previously (Blomquist, 1995).

3. Results

3.1 Seasonal variation in 17-HSD activity in coral samples

As an approach to the question of whether there is seasonal variation in 17-HSD activity levels, six samples each (3 winter, 3 summer) of M. capitata and T. coccinea were assayed with NADP+ and E2. Rates of estrone (E1) formation were linear for at least 120 minutes (Fig. 1). During the linear phase of E1 formation, rates varied up to 10-fold between the six M. capitata samples and up to 6-fold between the T. coccinea samples. Because of the marked intra-species overlap in activities, a seasonal variation in activity was not apparent.

3.2 E2/T and NADP+/NAD+ activity ratios for coral homogenates and zooxanthellae

The various isoforms of 17-HSD can be differentiated on the basis of their differing E2/T and NADP+/NAD+ activity ratios (Blomquist, 1995; Peltoketo et al., 1999). The E2/T ratio for the M. capitata homogenate with the highest level of activity with E2 was 21.8, whereas the values for the two samples of T. coccinea with the highest activity were lower (Fig. 2). NADP+/NAD+ ratios varied six-fold for M. capitata homogenates. The ratio for the zooxanthellae sonicate was 1.1 (Fig. 3). For T. coccinea homogenates, the ratios were more uniform (Fig. 4).

3.3 Michaelis constants (Km) for E2 and NADP+

Our previous findings (Tarrant et al., 2003) and the results shown in Figures 2-4 were consistent with presence in M. capitata of a form of 17-HSD for which E2 and NADP+ were the preferred substrate and cofactor, respectively. To examine this possibility further, Km-values were estimated for E2 over a 21-fold concentration range and for NADP+ over a 25-fold concentration range. Km-values were 11.3 M for E2 (Fig. 5) and 5.8 M for NADP+ (Fig. 6), and in both cases the results were consistent with the presence of a single enzyme.

3.4 Time-course of E2 formation with NADP+ and NAD+ at a saturating concentration based on Km-values

In our earlier studies we had used NADP+ and NAD+ at 0.5 mM, almost 100-fold greater than the apparent Km-value of 5.8 M for NADP+ (Fig. 6). As an approach to getting a better estimate of a physiologically-relevant NADP+/NAD+ activity ratio for both M. capitata and zooxanthellae, the time-course of E1 formation was assayed with cofactor at a near-saturating level of 50 M, based on a Km of 5.8 M. Activity with NAD+ was barely detectable in M. capitata homogenate, yielding an NADP+/NAD+ ratio of 22.2, whereas that for the zooxanthellae sonicate was 0.7 (Fig. 7).

3.5 Effects of C18- and C19-hydroxysteroids and phytoestrogens on the conversion of E2 to E1 by M. capitata homogenate

The apparent high degree of specificity for E2 and NADP+ of the predominant form of 17-HSD in M. capitata samples and the apparent low affinity for T prompted us to examine a less-polar C18-compound, 17desoxy-E2, and a more polar steroid, estriol (16-hydroxy-E2) as potential inhibitors. As shown in Fig. 8A, neither compound was effective. Two phytochemicals known to be effective inhibitors of human 17-HSD types 1 and 5 (Mäkelä et al., 1998; Krazeisen et al., 2001) were also tested. Zearalanone was somewhat effective and coumestrol at 20 M inhibited activity by 90%.A 50% inhibition was observed with coumestrol at 0.2 M (Fig. 8B), significantly less than the Km for E2 of 11.3 M (Fig. 5). When other phytoestrogens were compared with coumestrol, genistein was almost as effective. Daidzein, chrysin and luteolin were less inhibitory (Fig. 9). When a series of C19-hydroxysteroids was tested, small differences between coral and zooxanthellae were seen. But in general, the C19-steroids were relatively ineffective (Fig. 10), with the possible exception of testosterone versus zooxanthellae 17-HSD. This latter effect differs from what we observed previously with M. capitata homogenate where T had no effect on the conversion of E2 to E1 (Tarrant et al., 2003).

4. Discussion

Our initial findings with regard to 17-HSD in scleractinian corals (Tarrant et al., 2003) demonstrated the presence of activity, particularly with E2, and raised questions with regard to the number of isoforms present, whether there was seasonal variation in activity and the extent to which zooxanthellae contributed to activity in coral. The data in Fig. 1 confirm that 17-HSD activity with E2 is present in both zooxanthellate (M. capitata) and azooxanthellate (T. coccinea) corals and demonstrate further that there is significant overlap in specific activity between summer and winter samples from both species. Because of this wide intra-species variation and the inter-species overlap in specific activities, whether there is seasonal variation in activity remains to be clarified.

The E2/T activity ratio of 21.8 for the M. capitata sample with the highest specific activity (Fig. 2) is consistent with the presence of a 17-HSD isoform highly specific for E2, among the C18- and C19-steroids tested as substrates or inhibitors. NADP+/NAD+ ratios of 10.3 at 0.5 mM cofactor (Fig. 3) and 22.2 at 50 M (Fig. 7) suggest further that NADP+ is the preferred cofactor for this enzyme.

As in our previous studies (Tarrant et al., 2003), the highest 17-HSD specific activity was observed in sonicates of zooxanthellae (Fig. 7). NADP+/NAD+ activity ratios of 0.7-1.1 for zooxanthellae 17-HSD when compared with values of 1.6-22.2 for M. capitata suggest a variable contribution of the enzyme in zooxanthellae to activity in M. capitata, in particular, a minimal contribution to the M. capitata sample with the highest level of activity. It is noteworthy that the levels of activity with E2 and NADP+ in extracts of the azooxanthellate coral, T. coccinea, were comparable to those in some M. capitata samples but with NADP+/NAD+ activity ratios of 2.2 to 3.8. This suggests there may be another 17-HSD isoform present in T. coccinea different from both the predominant form in M. capitata and that in zooxanthellae.

The apparent complexity of the 17-HSD enzymology of coral and zooxanthellae raises interesting questions as to where these enzymes, particularly the predominant E2-specific, NADP+-dependent activity, fit in either the short chain alcohol dehydrogenase or aldo-ketosteroid reductase superfamilies (Peltoketo et al., 1999; Mindnich et al., 2004). In addition, the role of these isoforms in the physiology of coral and zooxanthellae remains unclear. There is growing evidence of a role of vertebrate-type sex steroids in coral spawning based both on variations in steroid content as well as enzymatic activities, such as aromatase and glucuronyl transferase (Slattery et al., 1999; Tarrant et al., 1999; Twan et al., 2003).

17-HSDs are widespread in microorganisms, invertebrates and vertebrates (Lanišnik Rižner et al., 1999; Baker, 2001; Breitling et al., 2001). While the various 17-HSD isoforms share overlapping substrate affinities, the sequence identity among isoforms is relatively low (Peltoketo et al, 1999). The diversity of enzymes with 17-HSD activity has apparently resulted from a combination of gene duplications and convergent evolution (Baker 2001). While the phylogenetic relationship of cnidarian enzymes to other metazoan 17-HSDs is currently unknown, identification of candidate cnidarian 17-HSDs will be facilitated by the sea anemone (Nematostella vectensis) genome sequencing project (Darling et al., 2005) .

Recent studies have demonstrated a remarkable multifunctionality among 17-HSDs with various isoforms able to recognize not only steroids but fatty acids and bile acids, as well (Mindnich et al., 2004). This active site plasticity may have relevance for coral 17-HSD. In our study, the specific binding at the active site of E2 but not other C18-steroids, which have been shown to be comparable to E2 in their binding to mammalian 17-HSD isoforms (Blomquist, 1995), suggests the natural substrate for the predominant NADP+-dependent coral enzyme may not be E2 but rather a compound more closely related structurally to coumestrol, which showed a markedly higher affinity with an IC50 of 0.2 M (Fig. 8B) compared with a Km for E2 of 11.3 M (Fig. 5). Both zooxanthellate and azooxanthellate coral can generate a variety of complex sterols by enzymatic pathways involving a number of oxidoreduction reactions (Higgs and Faulkner, 1977; Kokke et al., 1983; Ahmed et al., 2003; Duh et al., 2004); thus a sterol or other terpenoid may be the natural substrate for the apparent 17-HSD in the present study.

If in fact E2 is not the natural substrate for the apparent 17-HSD in coral and zooxanthellae extracts, then there is the possibility that exogenous estrogens or other environmental contaminants could interfere with the synthesis of and thereby theaction of the natural substrate. As a result, one or more aspects of coral physiology could be affected by a non-receptor mediated mechanism. Environmental chemicals have been shown to disrupt steroid metabolism in vertebrates (Kirk et al., 2003). While some studies have shown effects of environmental chemicals on steroid metabolism in invertebrates, results have been less consistent and require further study (Janer et al. 2005). Further testing of various coral-derived steroids, sterols and terpenoids as inhibitors of NADP+-dependent E2 metabolism in coral extracts may yield clues as to the physiological role of this enzyme.

Acknowledgments

We thank Dr. R. Kinzie for samples of zooxanthellae, and Linda Sackett-Lundeen, B.S.M.T., and Heather Gray, for doing the figures. Support for this work was provided by the EPA STAR fellowship program and the University of Hawaii Sea Grant College Program.

References

Ahmed, A.F., Dai, C-F., Kuo, Y-H, Sheu, J-H., 2003. 1,3,5-Trihydroxy-24-methylencholestan-6-one: a novel steroid from a soft coral Sinularia Gibberosa. Steroids 68, 377-381.

Baker, M.E. 2001. Evolution of 17-hydroxysteroid dehydrogenases and their role in androgen,

estrogen and retinoid action. Mol. Cell. Endocrinol. 171, 211-215.

Blomquist, C., Bealka, D., Hensleigh, H., Tagatz, G., 1994. A comparison of 17-hydroxysteroidoxidoreductase type 1 and type 2 activity of cytosol and microsomes from human term placenta, ovarian stroma and granulosa-luteal cells. J. Steroid Biochem. Mol. Biol. 49, 183-189.

Blomquist, C., 1995. Kinetic analysis of enzymatic activities: prediction of multiple forms of 17-hydroxysteroid dehydrogenase. J. Steroid Biochem. Mol. Biol. 55, 515-524.

Breitling, R., Laubner, D., Adamski, J., 2001. Structure-based phylogenetic analysis of short-chain alcohol dehydrogenases and reclassification of the 17-hydroxysteroid dehydrogenasefamily. Mol. Biol. Evol. 18, 2154-2161.

Cornish-Bowden, A., Eisenthal, R., 1978. Estimation of Michaelis constant and maximumvelocity from the direct linear plot. Biochim. Biophys. Acta 523, 268-272.

Darling, J.A., Reitzel, A.R., Burton, P.M., Mazza, M.E., Ryan, J.F., Sullivan, J.C., Finnerty, J.R.

2005. Rising starlet: the starlet sea anemone, Nematostella vectensis.Bioessays 27: 211-221.

Duh, C-Y., El-Gamal, A.A.H., Song, P-Y., Wang, S-K., Dai, C-F., 2004. Steroids andsesquiterpenoids from the soft corals Dendronephthya gigante and Lemnalia cervicorni. J. Nat. Prod. 67, 1650-1653.

Escriva, H., Safi, R., Hanni, C., Langlois, M.C., Saumitou-Laprade, P., Stehelin, D., Capron, A., Pierce, R., Laudet, V. 1997. Ligand binding was acquired during evolution of nuclear receptors. Proc. Natl. Acad. Sci. USA 94, 6803-6808.

Gassman, N.J., 1992. Aspects of the steroid biochemistry of the scleractinian coral, Favia fragum. Ph.D. Dissertation, Miami: University of Miami, 158 pp.

Higgs, M.D., Faulkner, D.J., 1977. 5-Pregna-1,20-dien-3-one and related compounds from a soft coral. Steroids 30, 379-388.

Hunter, C.L., 1988. Environmental cues controlling spawning in two Hawaiian corals, Montipora verrucosa and M. dilatata. Sixth International Coral Reef Symposium, Townsville, Australia, 727-732.

Janer, G., LeBlanc, G.A., Porte, C., 2005. A comparative study on androgen metabolism in three invertebrate species. Gen. Comp. Endocrinol. 143: 211-221.

Kirk, C.J., Bottomley, L., Minican, N., Carpenter, H., Shaw, S., Kohli, N., Winter, M., Taylor, E.W., Waring, R.H., Michelangeli, F., Harris, R.M. 2003. Environmental endocrine disrupters dysregulate estrogen metabolism and Ca2+ homeostasis in fish and mammals via receptor-independent mechanisms. Comp. Biochem. Physiol. A 135, 1-8.