An evolutionary approach to endocrine disruption: the mechanisms of reprodutive toxicity of androgenic chemicals in the gastropod Nucella lapillus and the fish Danio rerio.

An evolutionary approach to endocrine disruption: the mechanisms of reproductive toxicity of androgenic chemicals in the gastropod Nucella lapillus and the fish Danio rerio Daniela da Silva Lima Tese de Doutoramento em Ciências Biomédicas 2012 Daniela da Silva Lima An evolutionary approach to endocrine disruption: the mechanisms of reproductive toxicity of androgenic chemicals in the gastropod Nucella lapillus and the fish Danio rerio Tese de Candidatura ao grau de Doutor em Ciências Biomédicas submetida ao Instituto de Ciências Biomédicas Abel Salazar da Universidade do Porto. Orientador: Doutor Miguel Machado Santos Investigador Auxiliar Laboratório de Toxicologia Ambiental, Centro Interdisciplinar de Investigação Marinha e Ambiental Co-orientadores: Professora Doutora Maria Armanda Reis Henriques Professora Catedrática Instituto de Ciências Biomédicas Abel Salazar Doutor Luís Filipe Costa Castro Investigador Auxiliar Laboratório de Estudos Celulares, Moleculares e Analíticos, Centro Interdisciplinar de Investigação Marinha e Ambiental Esta tese foi financiada por uma bolsa de doutoramento da Fundação para a Ciência e a Tecnologia (SFRH/BD/41561/2007) e pelos projectos PTDC/MAR/68106/2006, PTDC/MAR/105199/2008 e PTDC/MAR/115199/2009. À minha família. “Will you walk a little faster?” said a whiting to a snail […] “See how eagerly the lobsters and the turtles all advance! […] Then turn not pale, beloved snail, but come and join the dance.” Alice in Wonderland, the lobster quadrille, Lewis Carroll Nature study will show you how full of beautiful and wonderful things God has made the world for you to enjoy. […] Try and leave this world a little better than you found it and when your turn come to die, you can die happy in feeling that at any rate you have not wasted your time but have done your best. Sir Baden-Powell vi retinoid signaling have never been described in mollusks. These knowledge gaps were addressed in this thesis and resulted in the isolation of four potential candidates to participate in retinoid cascades in N.lapillus: Retinoid X Receptor (NlRXR), Retinoic Acid Receptor (NlRAR), Cytochrome P26 (NlCyp26) and Alcohol Dehydrogenase 3 (NlAdh3). Their molecular characterization was performed and finally modulation by TBT exposure assessed. A pharmacological approach was conducted to clarify the mode of action of TBT focusing in three different targets: interference with neuroendocrine factors, modulation of retinoid and steroid signaling. Overall, the findings of the present work clearly link TBT-induced imposex with the modulation of RXR and retinoid signaling pathways. The involvement of RXR in imposex induction was experimentally demonstrated, given that both its putative endogenous ligand, 9-cis retinoic acid (RA), and the synthetic agonist methoprene acid (MA), were able to induce imposex in N. lapillus. Detailed analysis of RXR transcription in different organs suggests the existence of tissue-specific effects, since NlRXR mRNA levels were specifically and differentially modulated in the CNS and in the penis/PFA, highlighting the pivot role of the central nervous system in imposex development. The opposite pattern of gene transcription observed for NlCyp26 and NlRXR in mature/dormant female gonads and in penis development in TBT-exposed animals points to a coordinated action between NlRXR and NlCyp26 in controlling local retinoid signaling, which may ultimately regulate the development of penis in both male and females and gonad recrudescence in female. The distinct pattern of mRNA transcription observed for NlRXR, NlRAR, NlCyp26 and NlAdh3 suggests their involvement in different biological processes. Furthermore, it provides an indication that N. lapillus does possess, at least partially, the genetic toolkit necessary to metabolize and transduce retinoid signals, although NlAdh3 could not be unmistakably linked with RA signaling. Exploration of other possibly affected pathways by TBT in N. lapillus led to the identification of a novel target, 17-β Hydroxysteroid Dehydrogenase 12 (Hsd17b12), which codes for an enzyme potentially involved in steroid or lipid metabolism. In addition to the isolation and molecular characterization of NlHsd17b12, it was demonstrated here that NlHsd17b12 mRNA levels are impacted by TBT exposure and significantly decreased in digestive glands. Following the hypothesis that endocrine disruption by TBT may involve the interference with conserved retinoid signaling, the modulation of retinoid signaling by TBT and other retinoid substances in the zebrafish model was also investigated here. Exposure of D. rerio to TBT from 5 dpf until adulthood significantly impacted fish fecundity. Analysis of key signaling pathways revealed a significant decrease in the mRNA levels of vii Cytochrome P19a1b (Cyp19a1b) in female brain and in peroxisome proliferator activated receptor gamma (PPARγ), an heterodimeric partner of RXR, in brain of both males and females. Unexpectedly, TBT altered zebrafish sex ratio towards females, contradicting previous results and suggesting a more complex mechanism in the so far reported masculinizing effects in zebrafish. Although additional studies should focus on the detailed biological implications of TBT with the signaling pathways reported here, this work provides relevant insights to the clarification of mollusks retinoid signaling systems. In addition to contribute to the understanding of the possible negative outcomes resulting from TBT exposure in two different groups, this study adds substantial knowledge on retinoid signaling pathway evolution. Here is reported for the first time the cloning of a protostome RAR and Cyp26 orthologues, which were considered to be a chordate novelty. Knowledge on the evolution and functionality of endocrine systems and their molecular pathways (such as retinoid acid signaling) in metazoans is crucial to understand and anticipate the impact of the numerous compounds that are continuously being introduced in the ecosystems. Resumo Nas últimas décadas tem-se verificado um aumento significativo do número de estudos sobre os efeitos adversos da exposição a disruptores endócrinos (EDCs). Sabese que os EDCs podem interferir com a ligação das hormonas aos seus receptores (mimetizando-as ou antagonizando-as) e com o funcionamento normal do sistema endócrino (interferindo com a síntese, o transporte e o metabolismo hormonal). Uma vez que a interacção hormona/receptor é um mecanismo fundamental para a manutenção da homeostasia hormonal, a sua desregulação por EDCs tem sido particularmente estudada. Tradicionalmente, a maioria dos estudos sobre EDCs tem-se focado no impacto de compostos estrogénicos nos ecossistemas aquáticos, enquanto os disruptores que actuam em outras vias, tais como compostos com capacidade androgénica, têm recebido relativamente menos atenção. Outro aspecto importante que tem sido negligenciado por estes estudos prende-se com a influência dos processos evolutivos na funcionalidade dos actuais sistemas endócrinos dos animais. Tendo em conta que na maioria dos casos se usam como modelo espécies de vertebrados, os efeitos dos EDCs em outros grupos filogenéticos tem sido amplamente sub-avaliado. As vias de sinalização dos retinóides são disso exemplo, uma vez que durante muito tempo se pensou serem exclusivas dos animais cordados. Todavia, foi recentemente demonstrado que um dos fenómenos de disrupção endócrina mais conhecidos, o imposex, que se caracteriza pela superimposição de características viii sexuais masculinas (pénis e canal deferente) em fêmeas de gastrópodes marinhos, envolve a activação atípica do Receptor X Retinóico (RXR). Em mamíferos, os retinóides estão envolvidos na regulação de importantes funções biológicas e a sua perturbação pode originar, entre outras consequências, alterações nos processos reprodutivos. Deste contexto surgiu a hipótese de que os efeitos negativos resultantes da exposição ao TBT, observados em várias espécies (incluindo peixes), possam envolver a desregulação das vias de sinalização dos retinóides. Para avaliar esta ideia, foram escolhidas duas espécies filogeneticamente distantes, o peixe teleósteo Danio rerio e o gastrópode Nucella lapillus. A N. lapillus tem sido uma espécie muito utilizada em estudos sobre imposex mas os mecanismos envolvidos neste fenómeno estão ainda por esclarecer. Além disso, a maioria dos genes que se sabe estarem envolvidos nas vias de sinalização dos retinóides em mamíferos nunca foram caracterizados em moluscos. Estas lacunas foram também objecto de estudo desta tese. Como resultado, isolaram-se em N. lapillus quatro genes potencialmente envolvidos nas cascatas de sinalização dos retinóides: o Receptor do Ácido X Retinóico (NlRXR), o Receptor do Ácido Retinóico (NlRAR), o citocromo P26 (NlCyp26) e a álcool desidrogenase 3 (NlAdh3). A sua caracterização molecular foi efectuada e a susceptibilidade a modulação pelo TBT avaliada. Foi utilizada ainda uma abordagem farmacológica para esclarecer o modo de acção do TBT no fenómeno de imposex, nomeadamente, a interferência com factores neuroendócrinos, com as vias de sinalização dos retinóides e dos esteróides. Os resultados do presente trabalho apontam claramente a interferência do TBT com o RXR e com as vias de sinalização do ácido retinóico na indução de imposex. O envolvimento do RXR neste processo foi experimentalmente demonstrado, uma vez que quer o seu presumível ligando natural, o ácido 9-cis retinóico (9-cis RA), quer o seu agonista sintético, o ácido metoprénico (MA), induziram significativamente imposex em N. lapillus. A análise detalhada dos níveis de mRNA do RXR em diferentes órgãos sugere a existência de efeitos específicos em cada tecido e realça o papel fundamental do sistema nervoso central (CNS) no desenvolvimento de imposex. O padrão de transcrição oposto do RXR e da Cyp 26, observado nas gónadas de fêmeas maduras/imaturas e nos pénis em desenvolvimento nos animais expostos a TBT, apontam ainda para uma acção coordenada entre NlRXR e NlCyp26 no controlo local da sinalização por retinóides. Deste modo, estes genes podem estar envolvidos na regulação do desenvolvimento do pénis, tanto em machos como em fêmeas, bem como no recrudescimento das gónadas femininas. ix O padrão de transcrição distinto observado para o NlRXR, NlRAR, NlCyp26 e NlAdh3 sugere o seu envolvimento em diferentes processos biológicos. Mais ainda, a presença destes genes em N. lapillus indica que esta espécie possui, pelo menos em parte, a maquinaria genética necessária para metabolizar retinóides e sinalizar através destes compostos, podendo potencialmente traduzi-los em sinais biológicos. No entanto, não é ainda claro o envolvimento da NlAdh3 nas vias de sinalização dos retinóides. A exploração de outras vias passíveis de serem afectadas por TBT em N. lapillus resultou na identificação de um gene que codifica uma enzima envolvida no metabolismo lipídico ou dos esteróides, a 17-β Hidroxiesteróide desidrogenase tipo 12 (NlHsd17b12). A NlHsd17b12 foi caracterizada e a sua interferência pelo TBT avaliada, tendo-se observado uma diminuição significativa nos seus níveis de expressão nas glândulas digestivas após exposição ao TBT. Para avaliar a hipótese de que o TBT interfere com as vias de sinalização dos retinóides, e estas são conservadas nos metazoários, os efeitos do TBT e de outros retinóides no peixe zebra também foram alvo de estudo. A exposição de D. rerio ao TBT, desde os 5 dias após a fertilização (dpf) até à idade adulta, resultou numa diminuição significativa da sua fecundidade. A avaliação da transcrição de genes possivelmente afectados pelo TBT no peixe-zebra revelou uma diminuição significativa nos níveis de transcritos de citocromo P19a1b (Cyp19a1b) no cérebro das fêmeas, e de proliferadores de peroxissoma gamma (PPARg) no cérebro dos machos e das fêmeas. Contrariamente ao que está descrito na literatura até ao momento, o TBT alterou o rácio entre os sexos a favor das fêmeas, o que parece sugerir que um mecanismo mais complexo está na origem dos efeitos masculinizantes atribuídos ao TBT no peixe-zebra. Estudos futuros deverão centrar-se nas implicações biológicas potencialmente resultantes da interferência do TBT com as vias de sinalização aqui descritas. Para além de contribuir para a compreensão dos efeitos adversos resultantes da exposição ao TBT em organismos de dois grupos taxonómicos diferentes, este trabalho acrescenta um conhecimento substancial sobre a evolução das vias de sinalização dos retinóides. Este estudo descreve pela primeira vez a clonagem de ortólogos do RAR e da Cyp26 em protostómios, genes que até há relativamente pouco tempo eram considerados exclusivos dos cordados. O conhecimento sobre a evolução e a funcionalidade dos sistemas endócrinos e vias de sinalização associadas (como por exemplo, as que envolvem o ácido retinóico) é crucial para compreender e antecipar o impacto dos inúmeros compostos que estão continuamente a ser introduzidos nos ecossistemas. x xi Acronyms List ADHs Alcohol Dehydrogenases AKR Aldo-Keto-Reductase ALDHs Aldehyde Dehydrogenases AR Androgen Receptor ARAT Acyl-coenzyme A: Retinol Acyltransferase ATAT Acyl-coenzyme A: Testosterone Acyltransferase BCO β-Carotene Oxygenase BPA Bisphenol-A CNS Central Nervous System (head ganglia complex) CPA Cyproterone Acetate CRBPs Cellular Retinol Binding Proteins CYP Cytochrome P450 Dax1 Dosage-sensitive sex reversal, adrenal hypoplasia critical region, on chromosome X, gene 1 DBD DNA Binding Domain DGAT1 Diacylgliceracyltransferase 1 DHA Docosahexaenoic Acid DMSO Dimethyl sulfoxide Dpf Days Post Fertilization DR Direct Repeat E1 Estrone E2 Estradiol EDCs Endocrine Disrupting Chemicals EE2 Ethinylestradiol EFSA European Food Safety Authority ER Estrogen receptor xii FABPs Fatty Acid Binding Proteins FBS Fetal Bovine Serum GnRH Gonadotropin Releasing Hormone GnRHR Gonadotropin Releasing Hormone Receptor HEK-293 Human Embryonic Kidney -293 HSD Hydroxysteroid Dehydrogenase HUFA Highly Unsaturated Fatty Acids IMO International Maritime Organization IPCS International Programme on Chemical Safety KAR 3-Ketoacyl-CoA Reductase LBD Ligand Binding Domain LC/MS/MS Liquid Chromatography/Tandem Mass Spectrometry LRAT Lecithin:Retinol Acetyltransferase MA Methoprene Acid MDR Medium-chain Dehydrogenases/Reductases MEHP Mono-(2-ethylhexyl) Phthalate MEPC Marine Environment Protection Committee MS-222 Ethyl 3 - aminobenzoate methenesulforate salt NR Nuclear Receptor ORF Open Reading Frame p38MAPK P38 Mitogen-Activated Protein Kinases PFA Penis Forming Area PMF Penis Morphogenetic Factor PPAR Peroxisome Proliferator Activated Receptor PUFA Polyunsaturated Fatty Acid PVC Polyvinyl Chloride RA Retinoic Acid xiii RALDH Retinal Dehydrogenase RAR Retinoic Acid Receptor RAREs Retinoic Acid Response Elements REs Response Elements RBP Retinol Binding Protein RDH Retinol Dehydrogenase REHs Retynil Esters Hydrolases RF Retrogressive Factor RXR Retinoic X Receptor SDRs Short-chain Dehydrogenases/Reductases TBT Tributyltin TDI Tolerable Daily Intake TPT Triphenyltin TR Thyroid Receptor TTR Transthyretin USP Ultraspiracle UTR Untranslated Region VAD Vitamin A Deficient VDR Vitamin D Receptor VDSI Vas Deference Sequence Index VTG Vitellogenin WHO World Health Organization xiv xv INDEX CHAPTER 1 ...................................................................................................................... 1 1. Introduction .................................................................................................................. 3 1.1 General Introduction ................................................................................................ 3 1.1.1 Imposex in gastropod mollusks: a striking case of endocrine disruption ............ 4 1.1.2 TBT: biological effects ....................................................................................... 6 1.2.3 TBT: legislation and current environmental status ............................................. 8 1.2 The imposex underlying mechanisms: an historical account ................................... 9 1.2.1 Imposex model 1: abnormal release of neurohormones ...................................10 1.2.2 Imposex model 2: interference with steroid pathways ......................................11 1.2.3 Imposex model 3: interference with retinoid signaling pathways ......................15 1.3 The retinoid signaling pathway ...............................................................................17 1.3.1 Retinoid transport and storage .........................................................................19 1.3.2 Retinoid synthesis: canonical or classical pathway ..........................................20 1.3.3 Retinoid synthesis: non-canonical pathway ......................................................21 1.3.4 Retinoid degradation ........................................................................................22 1.3.5 Retinoic acid receptors.....................................................................................23 1.4 Evolution of retinoid signaling .................................................................................27 1.5 Evolution meets toxicology .....................................................................................30 1.6 The animal models .................................................................................................32 1.6.1 The dogwhelk, N. lapillus .................................................................................32 1.6.2 The zebrafish, D. rerio......................................................................................33 1.7 Objectives ..............................................................................................................35 1.8 References ............................................................................................................38 CHAPTER 2 .................................................................................................................... 51 2. Retinoid signaling and imposex in Nucella lapillus ................................................ 51 xxii glands after one and two months of exposure to TBT. 5.1 Schematic representation of the experimental setup used for zebrafish reproductive trials. 182 5.2 Sex ratio observed for each experimental group. 188 5.3 mRNA levels of male and female zebrafish determined in brain (Cyp19a1b and PPARγ) and gonads (Cyp19a1a and PPARγ) after TBT exposure. 189 xxiii List of tables 2.1.1 Oligonucleotide primer sequences used to isolate and characterize NlRXR. 59 2.2.1 Primer sequences used to isolate 18s in N. lapillus. 82 3.1.1 Experimental parameters measured in the exposed N. lapillus at different treatments, three days and two months after the beginning of the experiment. 107 3.3.1 Pairwise % identity of the dogwhelk RAR protein sequence to other RAR genes. 141 3.3.2 Amino acid residues within the ligand-binding pocket of RAR that interact with the ligand. 145 4.1 Imposex parameters in female N. lapillus after one and two months of exposure: VDSI, imposex frequency and penis length. 164 5.1 Primer sequences and annealing temperatures used in qPCR determinations in D. rerio. 184 5.2 Morphometric parameters at the end of the zebrafish experiment. 186 5.3 Zebrafish reproductive parameters from the breeding trials performed at the end of the experimental period. 187 5.4 Summary of qPCR gene transcription results for zebrafish ER2a, ER2b, Vtg1 and RXRβa. 190 xxiv CHAPTER 1 CHAPTER 1 Chapter 1 3 1. Introduction 1.1 General Introduction In the last couple of decades a significant effort has been made to study and acknowledge the adverse effects resulting from environmental contaminant exposure. Some of these compounds have the ability to interfere with the endocrine system, and are generically referred to as endocrine disruptors or endocrine disrupting chemicals (EDCs;WHO / IPCS, 2002). Traditionally, studies on EDCs have focused on identifying these compounds, their effects and the affected species. Nowadays, it has been recognized that such studies should also address other relevant aspects, in particular the mechanisms involved in endocrine disruption phenomena and their ability to induce relevant ecological consequences (Guillette, 2005). Among the most common adverse effects attributed to EDCs are alterations in sexual and reproductive development, changes in immune and nervous system and increased incidence of cancer and thyroid metabolism disorders (WHO / IPCS, 2002). In many cases EDCs interfere with receptormediated mechanisms, mimicking or antagonizing endogenous hormones, or by deregulating their signaling, synthesis, transport and/or metabolism (WHO / IPCS, 2002; Sumpter, 2005). Aquatic ecosystems are traditionally the most affected, since they are the final destination of most of the compounds resulting from human activity. Regarding pollutants, the initial focus of attention was channeled to the effects of heavy metals. However, the emergence of new industrial chemicals and pharmaceutical products, which are often not effectively removed by waste water treatment, also brought more relevance to the study of their effects and the mechanisms by which they act (http://www.epa.gov/bioindicators/aquatic/pollution.html). Since a large proportion of potential EDCs end up in surface waters, aquatic species are particularly vulnerable to their potential adverse effects. Historically, most studies have focused on the impact of estrogenic compounds in aquatic ecosystems, while information on other EDCs, such as compounds with androgenic properties, is much scarcer (Sumpter, 2005). Paradoxically, one of the most ubiquitous phenomena of endocrine disruption in marine organisms is precisely the masculinization of females of several gastropod species. This phenomenon, designated imposex, is known to be caused by exposure to tributyltin (TBT), a compound used in antifouling paints on boats (Smith, 1971, 1981c; Gibbs and Bryan, 1986; Matthiessen and Gibbs, 1998; Barroso et al., 2002; Santos et al., 2005). Other studies have described the deleterious effects of Chapter 1 4 TBT in several Metazoan groups, such as mammals, fish, annelids and crustaceans (Fent, 1996; Janer, 2005). Despite the prominent number of species and the diversity of groups affected by TBT, its exact mode of action is yet to be ascertained. Furthermore, understanding the mechanisms involved in particular endocrine disruption cases, such as those related to TBT, may help to predict, prevent and minimize the impact of other chemicals acting on the same pathways. This may be of pivotal importance, especially if we take into consideration that over 87 000 synthetic substances are currently being commercialized, and during their manufacture many more thousands are produced and released to the environment (Thornton, 2003). It is therefore possible that at least some of these compounds are acting as endocrine disruptors and affecting non-target species, similar to what occurred with TBT and other organotins. 1.1.1 Imposex in gastropod mollusks: a striking case of endocrine disruption In 1960, in a routine examination of a batch of Nucella lapillus (Figure 1. 1A) collected in Plymouth Sound, England, Blaber (Blaber, 1970) did a rather intriguing discovery: he noticed the presence of an “outgrowth behind the right cephalic tentacle” in females, a structure resembling a male penis but slightly smaller (Figure 1.1 B, C). While this would not have been surprising for other species of hermaphrodite mollusks, N. lapillus was known to be a gonochoristic species. However, given that he observed this condition (with varying incidence) in the following sampling campaigns along different sites, he concluded that it could be a natural phenomenon, at least for a percentage of the population close to the breeding season. This was the first description of the phenomenon designated “imposex” one year later by Smith, after observing a similar condition affecting females of the American mud-snail, Ilyanassa obsoleta. Smith then defined imposex as the “superimposition of male characters on to parasitized and unparasitized females” of marine gastropods (Smith, 1971). It took Smith approximately ten years to demonstrate a correlation between imposex frequency and the distance to shipping harbors and marinas (Smith, 1981b), and to establish that this was an abnormal phenomenon caused by exposure to components of antifouling paints used in ships’ hulls (Smith, 1981a), more specifically, to the biocide TBT (Smith, 1981c). But this story had started way before, in the 1940s, when the industry discovered TBT as a potent biocide agent and its cost-effective properties soon turned it out very popular. TBT’s toxicity at low concentrations towards animals and plants in combination with its low mammalian toxicity (in comparison with mercury, lead and arsenic, used Chapter 1 5 previously) was considered a true revolution for the market of antifouling paints in the 1950s (Santillo, 2001). Growth of barnacles, clams and other organisms in hulls diminished ships’ performance, increased fuel consumption, and its mechanical removal was costly and time consuming. TBT, first used in “contact leaching” antifouling paints, and later coupled with a soluble matrix, which allowed it to dissolve gradually in water, not only reduced the costs caused by fouling, but also the need to frequent repaint boats, as its effects could last up to 4 years (Hall et al., 1987). TBT is commonly referred to as an organotin, which is a group of compounds, typically of anthropogenic origin, that encloses molecules consisting of an atom of tin (Sn) bound to an organic group (Figure 1.1 D). From all the derivatives of tin, organotins are commercially the most relevant, being used as Polyvinyl Chloride (PVC) stabilizers, catalysts for polyurethane and silicon elastomers, and pesticides (Fent, 1996). However, the toxicity of organotins increases with the number of organic groups bound to the tin atom, reaching its maximum in trialkylated compounds such as tributyl-, triphenyland tricyclohexyltin, which is precisely the case of TBT (Fent, 1996). Thus, when the first reports of serious deleterious effects in aquatic animals appeared, TBT, ubiquitously present in aquatic ecosystems, with elevated and broad spectrum toxicity, became one of the main suspects. In addition to the cases of imposex and severe diminution of N. lapillus populations in British coastal areas, another early alarming sign came from the Arcachon Bay, in the French Atlantic coast, an important location of oyster production. Here, imposex was observed in Ocenebra erinacea, leaving local population close to extinction, while oyster production decreased dramatically (Evans and Nicholson, 2000; Santillo, 2001). Although normal spawning seasons were reported at the time, oyster larvae displayed high mortality, difficulty in settlement and deformed shells in adulthood (Alzieu et al., 1989). The reproductive failure and shell deformities reported here matched those found in British coasts and coincided with an increase use of organotin containing paints in ships hulls (Santillo, 2001). Meanwhile, across the Atlantic, Smith demonstrated that I. obsoleta was indeed a gonochoristic gastropod, and that the induction of male characteristics in females was an aberrant phenomenon caused by exposure to TBT (Smith, 1981c). On the European coast, analytical data showed a clear correlation between tissue organotin levels and imposex severity, again pinpointing TBT as the causative agent (Bryan et al., 1986). The final proof was presented by Gibbs and Bryan (Gibbs and Bryan, 1986), who demonstrated experimentally that exposure to TBT at concentrations as low as 0.5 ng TBT Sn/L could induce imposex in N. lapillus (Figure 1.1 E, F) and cause, at advanced stages, female reproductive failure. Chapter 1 6 Figure 1.1 Adult N. lapillus (A). Developing penis in an imposex affected N. lapillus female (B). Male N. lapillus penis (C). Chemical structure of tributyltin chloride (D). Schematic representation of a normal (E) and an imposex affected female whelk (F; adapted from Bright and Ellis, 1990). 1.1.2 TBT: biological effects As a result of widespread contamination of TBT throughout aquatic ecosystems, imposex affected species have been detected worldwide. Imposex is not, however, the sole example of the adverse effects of organotins in aquatic animals, though it is certainly the most impressive and well-studied. Other mollusks, for yet unknown reasons, do not develop imposex when exposed to TBT. Instead, another phenomenon has been observed in some species, such as the periwinkle Littorina litorea: the transformation of female pallial organs into male morphological structures (Bauer et al., 1997; Ketata et al., 2008). This has been designated as intersex, and, at advanced stages, a prostate gland and a sperm groove develops, compromising their reproductive ability (Oehlmann et al., 1998). Still, the growth of a penis is a rare event in this species (Matthiessen and Gibbs, 1998). In mollusks, other effects that have been attributed to TBT exposure are: reduced rates of fertilization and development (Fent, 1996); shell thickening (Fent, 1996); alterations in steroid and lipid metabolism (Janer et al., 2006; Santos et al., 2011); interference with the immune system (causing decrease in total hemocyte count, phagocytosis, membrane stability and lysozyme activity, for instance; Gopalakrishnan et al., 2011). Mollusks are not the only phylum affected by TBT exposure. In fact, organotins are known today to have diverse consequences on species that belong to different Chapter 1 7 phylogenetic groups, from mammals to algae and bacteria (Fent, 1996; Janer, 2005). Since aquatic ecosystems are traditionally the most affected by pollution, studies on the effects of TBT have also been performed in fish, specially addressing consequences at the reproductive level. For instance, exposure of zebrafish (Danio rerio) to TBT during early developmental stages has resulted in altered sex ratio towards males (McAllister and Kime, 2003; Shimasaki et al., 2003; Santos et al., 2006a) and low quality sperm (McAllister and Kime, 2003). Diminished fertility (Nakayama et al., 2004, 2005), alterations in follicular development (Zhang et al., 2007), decrease in total and viable hatchability in Sillago japonica (Shimasaki et al., 2006; Nakayama et al., 2005) and accumulation of TBT in medaka (Oryzias latipes) eggs (Nakayama et al., 2005) are some of the consequences observed in TBT-exposed females. Male fish are also affected, showing altered sexual behavior (Nakayama et al., 2004), disrupted spermatogenesis (Zhang et al., 2009a), reduced gonadal development, decreased sperm levels (Haubruge et al., 2000; Zhang et al., 2009b) and histological damage in the testes (Zhang et al., 2009b). Decreased gonadossomatic index is a common consequence in both sexes (Zhang et al., 2009b). TBT was also shown to increase swimup failure and larval eyes defects in medaka (Nakayama et al., 2005), and morphological abnormalities such as dorsal curvature, twisted tails and pericardial edema in S. marmoratus (Zhang et al., 2011). Furthermore, TBT exposure resulted in increased levels of testosterone and lower estradiol levels in ovaries and testes of S. marmoratus. The latter was accompanied by a decrease in Sertoli cells marker and increase in lipid droplets (Zhang et al., 2007, 2009a). Finally, it has also been demonstrated that TBT interferes with the expression of genes believed to be involved in fish reproduction and/or sex determination, such as Sox9, Dax1, Cyp19a, SF1, ER, PPARγ, RXR α, β and γ (Pavlikova et al., 2010; McGinnis and Crivello, 2011; Zhang et al., 2009a, 2011). Although earlier studies indicated a low toxicity of TBT in mammalian cells (Santillo, 2001), subsequent evidence indicates that mammals are also sensitive to TBT exposure. Some of the described effects are similar to those observed in fish: interference with steroid metabolism (Janer, 2005), decreased estradiol levels and ERα and ERβ expression (Chen et al., 2008), imunotoxicity (Chen et al., 2011). Moreover, recent data suggested the involvement of organotins in the development of obesity (Grün et al., 2006). Indeed, TBT has recently been considered an “environmental obsesogen” (Grün and Blumberg, 2009) as it causes the induction of adipogenesis in both cell culture models and in vivo: increases adipose mass in frogs and mice (Grün et al., 2006), induces the differentiation of preadipocyte 3T3-L1 cells into adipocytes and expression of adipogenic gene marker (Kanayama et al., 2005; Li et al., 2011), increases triglyceride storage (Li et Chapter 1 14 Figure 1.3 Diagram illustrating imposex model 2. It assumes that in mollusks, testosterone induces male reproductive tract development. TBT would block testosterone synthesis by inhibiting aromatase (A), sulfotransferase (B) or ATAT (C). Despite the different attempts to explain the higher levels of testosterone in imposex affected females, this imposex mechanism theory has a major drawback: the absence of evidence to prove that sexual differentiation in mollusks is driven by vertebrate-like steroids such as testosterone and/or estradiol, which act through their respective androgen and estrogen receptors. Although activation of androgen receptormediated transcription and cell proliferation by TBT has been described in human cell lines (Yamabe et al., 2000), the existence of a functional androgen signaling pathway in mollusks, which is also activated by TBT, remains controversial (Santos et al., 2005; Sternberg et al., 2008; Markov et al., 2009). Moreover, the involvement of steroids in the reproduction of mollusks remains unclear, as no sex specific differences on testosterone levels throughout the reproductive cycle were found in the mud-snail I. obsoleta (Sternberg et al., 2008). Therefore, although different studies confirm that steroids in mollusks may have biological effects (such as oogenesis stimulation, maturation of Chapter 1 15 testicular elements and precocious spermatogenesis (Siah, et al., 2003; Oehlmann et al., 2006), there is no evidence to indicate that functional vertebrate-like steroid receptors exist in this group. These puzzling observations suggest that steroid signaling, if present in mollusks, might operate in a distinct way to that of vertebrates (Markov et al., 2009) and emphasize the need for deeper studies on the endocrinology of mollusks. 1.2.3 Imposex model 3: interference with retinoid signaling pathways A rather different proposal on the mechanism of imposex induction by TBT resulted from an attempt to identify the affinity of known EDCs towards human nuclear receptors (Nishikawa et al. al., 1999; Nishikawa et al., 2004; Kanayama et al., 2005; Figure 1.4). Receptor-mediated mechanisms are one of the possible means by which EDCs exert their effects, by mimicking or antagonizing endogenous hormones. Nuclear receptors are a superfamily of transcription factors activated by ligands of diverse chemical nature (such as lipophilic hormones, vitamins, lipids or other intracellular signals) that function as an interface between the cellular environment / body and the genome; they integrate upstream signals into coordinated gene expression and adequate cellular responses (Sonoda et al., 2008). Nishikawa and co-workers (2004) showed that both TBT and TPT were able to increase the interaction between human retinoic X receptor (RXR) and the co-activator TIF2. This induction of ligand-dependent interaction was somewhat higher than the one caused by RXR’s putative natural ligand 9-cis retinoic acid (9-cis RA). Since this interaction was known to be correlated with transcriptional activity, soon after, Nishikawa and co-authors hypothesized that this could be one of the molecular pathways involved in imposex induction by organotins (Nishikawa et al., 2004). Therefore, in pursuit of this hypothesis, these researchers successfully cloned the RXR orthologue from the mollusk T. clavigera (TcRXR). TcRXR displayed a basic structure similar to that of other NR, consisting of six modular units, two of which highly conserved: the DNA binding domain (DBD) and the ligand binding domain (LBD). The DBD and LBD of TcRXR presented 89.6 % and 83.9 % of amino acid identity with hRXRa, respectively. Nishikawa and co-authors further demonstrated that, similar to the human orthologue, TcRXR LBD could bind in vitro to 9-cis RA and to the organotins TBT and TPT. Finally, they successfully demonstrated that exposure to 9-cis RA significantly induced imposex in T. clavigera. Chapter 1 16 Figure 1.4 Imposex model 3. Activation of RXR would induce male genitalia development. This pathway would be repressed in normal females, but exposure to TBT would abnormally activate RXR signaling and induce the growth of male reproductive tract. These findings provided a completely distinct perspective on the imposex phenomenon, but they also raised a series of other pertinent questions. Is the development of masculine structures in females caused by 9-cis RA induction reproducible in other species of mollusks? Since it is known that in vertebrates 9-cis RA can also bind retinoic acid receptor (RAR), is this phenomenon specifically mediated by RXR? Do retinoids exert biological functions in mollusks? If so, do mollusks possess the enzymatic machinery required to synthesise, store, signal and metabolize retinoids? Addressing these questions is essential to fully corroborate this hypothesis, and comprehend the imposex mechanism. The absence of a full understanding of the mechanisms of action of TBT is the real drawback to prevent similar imposex-like scenarios. It is therefore crucial to characterize the basic mechanisms like retinoid or steroid signaling in animals other than mammals in order to recognize their biological significance and sensitivity to endocrine disrupting chemicals. Chapter 1 17 1.3 The retinoid signaling pathway “Retinoid” is a term used by IUPAC-IUB (1982) to designate ‘‘compounds consisting of four isoprenoid units joined in a head-to-tail manner; all retinoids may be formally derived from a monocyclic parent compound containing five carbon-carbon double bonds and a functional terminal group at the terminus of the acyclic portion.’’ This early definition included all compounds, biologically active or inactive, structurally related to vitamin A, but not synthetic compounds that displayed retinoid-like activity. Some authors use a broader definition of “retinoid” that also accommodates compounds that do not resemble retinol but elicit retinoid-like activity (Theodosiou et al., 2010). In the context of this thesis, the term “retinoid” is used to designate a group of natural and synthetic compounds structurally related to vitamin A (retinol), which regulate important cellular functions including morphogenesis and embryogenesis in vertebrates, cell proliferation, differentiation and apoptosis and homeostasis processes (MacLean et al., 2007; Albalat, 2009). Retinoid unbalance has been associated with clinical conditions such as obesity, diabetes, cardiovascular disease, leukemia and skin disorders (Kane et al., 2008; Ziouzenkova and Plutzky, 2008). Studies involving disruption of retinoic acid (RA) signaling have demonstrated that vitamin A is fundamental for proper embryonic development, namely for correct patterning and neural differentiation, for neural tube, heart, eye, kidney and urogenital tract development, and for the initiation of differentiation of the anterior region of the presomitic mesoderm that originates new somites (ClagettDame and Knutson, 2011). In addition to retinol, the retinoid family also includes retinaldehydes (retinals), retinyl esters and retinoic acids (RAs; Figure 1.5). Retinyl esters are the most abundant in animal tissues, being found mainly as retinyl palmitate (though retinyl oleate and retinyl stearate have also been found; Theodosiou et al., 2010). Retinol, retinal and retinyl esters are precursors of the most active form of vitamin A, RA, which includes all three stereoisomers: all-trans, 9-cis and 13-cis RA (Campo-Paysaa et al., 2008). The presence of the isomers all-trans and 13-cis has been unequivocally detected in vivo, being all-trans the most abundant in mice and humans (Thatcher and Isoherranen, 2009). As for 9-cis RA, until very recently, its presence in biological samples had not been detected and its in vivo role was questioned. However, 9-cis RA has been detected in mouse pancreas (Kane et al., 2010), in rainbow trout tissues (Gesto et al., 2012a), in fiddler crabs (Albalat, 2009), in the mollusk thick top shell (Gesto et al., 2012b) and in insect embryos (Nowickyj et al., 2008). Chapter 1 18 Figure 1.5 Structures of some natural retinoids and β-carotene (adapted from Theodosiu et al., 2010 and Ross et al., 2000). Retinoids’ biological functions and mode of action resembles that of endogenous hormones. However, since animals cannot synthesize retinol de novo and need to obtain it from the diet, retinoids are sometimes called “dietary hormones” (Novák et al., 2008). Animals obtain these necessary compounds either by directly ingesting retinol and retynil esters stored in other animal tissues or indirectly from carotenoids existent in plants (Theodosiou et al., 2010). Plants, as well as some microorganisms, have the ability to synthesize carotenoids (α-carotene, β-carotene, and β-cryptoxanthin). β-carotene can then be cleaved symmetrically into two molecules of retinaldehyde (which can be either reduced to retinol or oxidized to RA) or asymmetrically into variable chain length apocarotenals, whose function has not been deciphered so far (although its presence has been detected in tissues; Simões-Costa et al., 2008). Retinoid homeostasis is maintained by a delicate balance involving the synthesis and storage of retinoid precursors, their transformation into the biologically active RA and its precise degradation to ensure specific responses (Figure 1.6). In addition to this complex network of enzymes, retinoid signaling also requires the existence of functional nuclear receptors, able to integrate retinoid signals into adequate gene expression patterns (Figure 1.6). Therefore, retinoid actions in vertebrates can be regulated at two levels: metabolism and signaling. The enzymatic machinery responsible for the coordination of spatial and temporal levels of retinoids includes enzymes of synthesis, degradation and possibly of storage. As for signaling, retinoid functions are mediated by Chapter 1 19 the binding of RA to homodimers or heterodimers formed by members of two families of nuclear receptors, RAR and RXR (Marlétaz et al., 2006). This in turn regulates the expression of genes involved in their own signaling pathways (RA synthesis, metabolism and signaling) and in others pathways as well (eg, homeobox genes; Schubert et al., 2005; Marlétaz et al., 2006). 1.3.1 Retinoid transport and storage Retinoid levels are kept constant by the equilibrium between their availability in target tissues, circulation in the blood and storage in liver and adipose tissue. This balance is kept by the action of proteins involved in its transportation and by enzymes that esterify or deesterify retinol, mobilizing it as needed. All dietary retinoids are enzymatically converted into retinol in the intestine lumen before uptake by the enterocytes (Blomhoff and Blomhoff, 2006). There, retinol binds to specific proteins from the family of fatty acid binding proteins (FABPs) called cellular retinol binding proteins (CRBPs). CRBPs influence retinol availability and physiological function, facilitating cellular uptake and storage as retinyl esters. In most tissues, the retinol-CRBPII complex, also called holoCRBP, delivers retinol to lecithin:retinol acetyltransferase (LRAT) to be esterified with long-chain fatty acids, mainly palmitate (Blomhoff and Blomhoff, 2006). In other tissues, such as the mammary gland and skin, this reaction is performed by acyl-CoA:retinol acyltransferase (ARAT) or diacylgliceracyltransferase (DGAT1), respectively (Novák et al., 2008; Napoli, 2012). It has been suggested that the ratio apo-CRBP1/holo-CRBP1 regulates LRAT conversion of retinol into retinyl-esters. This reaction, performed by LRAT, typically (but not exclusively) occurs in intestinal cells, after which most retinol esters are incorporated into chylomicrons and transported to target tissues or to the liver, where they are stored. When necessary, retinol esters can be metabolized back to retinol by retinyl esters hydrolases (REHs; Theodosiou et al., 2010). In this case, retinol binds to a different protein, the retinol binding protein (RBP), that appears to keep endogenous retinoid levels stable, for holo-RBP presents a concentration of 2 µM regardless of fluctuations in retinol intake (Theodosiou et al., 2010). In a normal vitamin A sufficiency status, most of the retinol is transported to specific liver cells, called stellate cell, where it is stored in the form of retinyl esters in cytoplasmic lipid droplets (Blomhoff and Blomhoff, 2006; Novák et al., 2008). Total retinol (retinyl esters plus retinol) in the stellate cells accounts for 50-80 % of the overall amount present in the whole body, and about 95% of the retinol is in the form of retinyl esters in the abovementioned lipid droplets (Blomhoff and Blomhoff, 2006). Chapter 1 20 When retinol is required in peripheral tissues, Holo-RBP enters the blood circulation to be distributed to tissues and associates to transthyretin (TTR), a protein that prevents retinol from being degraded by the kidney. In the target cells, a membrane receptor (Stra6) recognizes RBP, allowing the complex to enter the cell and deliver retinol to be processed into biologically active retinoids (Blomhoff and Blomhoff, 2006; Theodosiou et al., 2010). 1.3.2 Retinoid synthesis: canonical or classical pathway In target tissues, the transformation of retinol into its active form, RA, involves two successive dehydrogenation steps: first, the conversion of retinol into retinaldehyde, and then, the conversion of the latter into retinoic acid. The first reaction is catabolized by alcohol dehydrogenases (ADHs), cytosolic enzymes classified as medium-chain dehydrogenases/reductases (MDRs), and by retinol dehydrogenases (RDHs), microsomal enzymes of the short chain dehydrogenases/reductase (SDRs) families. From a biochemical point of view, this is considered the rate-limiting step in RA synthesis (Simões-Costa et al., 2008). The second reaction, which is in part responsible for the regulation of RA levels in target cells, requires the action of enzymes that belong to the aldehyde dehydrogenases (ALDHs). ADHs/RDHs In vitro, the oxidation of retinol into retinaldehyde is a reversible reaction that can be performed by ADHs or by RDHs, the direction of the reaction depending on the ratio of the co-factors NAD/NADH. However, the main responsible for retinal synthesis in vivo is still a subject of intense debate. For a long time, it seemed that ADHs, especially the ubiquitously expressed ADH3, would be the most relevant in processing retinol (SimõesCosta et al., 2008). Disruption of mice Adh3 gene caused reduced viability and growth, but these effects could be rescued by retinol supplementation, indicating that in the absence of ADH3 other enzymes could compensate its function in retinol metabolism. However, these Adh3 null mice died when kept in a vitamin A deficient diet (VAD), thus suggesting that when retinol quantity is restricted ADH3 function cannot be compensated by other enzymes (Theodosiou et al., 2010). The apparent crucial involvement of Adh3 gene in retinal synthesis was recently challenged by the finding that an enzyme of the microsomal SDR family, RDH10, might play a more relevant role (Sandell et al., 2012). An Rdh10 null mouse model presented embryos with shorter anteroposterior axes, dilated and un-looped hearts, smaller somites, defects in embryo turning and forelimbs growth, Chapter 1 21 dying before growth (Sandell et al., 2012). More studies are therefore needed to address the in vivo potential of these and other candidate enzymes in modulating retinol/retinal interconversion. ALDHs/RALDHs ALDHs are a superfamily of enzymes with the ability to metabolize aldehyde substrates, including retinaldehyde, both in the all-trans and 9-cis conformation (CampoPaysaa et al., 2008). In vertebrates, members of the ALDH1A are involved in the irreversible conversion of retinal in RA, a reaction that coincides with the signaling activation by RA. Vertebrates possess several ALDH genes, the best studied being ALDH1A1, ALDH1A2, ALDH1A3 and ALDH8, coding for RALDH1, RALDH2, RALDH3, RALDH4, respectively (Theodosiou et al., 2010). Rodents have an additional ALDH1A, called ALDH1A4 in rat and ALDH1A7 in mice. Knockout of Raldh2 in mice suggests that its main function is to provide RA during development, as raldh2 -/- die before birth but can be rescued by RA supplementation. All the other RALDHs are involved in RA synthesis, although Knockout experiments suggest that their role is less important than Raldh2 (Theodosiou et al., 2010). 1.3.3 Retinoid synthesis: non-canonical pathway An alternative pathway for RA synthesis has been proposed after the discovery of a gene encoding the enzyme β-carotene-15,15’-oxygenase, BCO-I, that can cleave βcarotene at the central double bond, producing two retinal molecules in the all-trans configuration (Von Lintig and Vogt, 2000; Wyss et al, 2000). Retinaldehyde can then be transformed into retinol or into RA and this is the reason why β-carotene is sometimes designated as provitamin A (Ziouzenkova and Plutzky, 2008). It is believed that BCO-I may play an important role in the visual cycle, perhaps in the synthesis of retinaldehyde, which is required for photoreception. BCO-I KO in zebrafish lead to malformations in the retina but also caused several morphogenetic malformations during embryonic development, thus pointing to a major role of BCO-I as a source of retinoid precursors during vertebrate development (Simões-Costa et al., 2008). Another enzyme, named BCO-II, is also a candidate to have a function in RA synthesis, as it is able to asymmetrically cleave β-carotene into a β-carotenal molecule plus one of β-ionone (Kiefer et al., 2001), although its exact role is yet to be characterized. Chapter 1 22 β-apocarotenal is converted first to β-apocarotenoic acid and subsequently to RA, but the enzymes involved in this process are still unknown (Theodosiou et al., 2010). 1.3.4 Retinoid degradation Fine-tuned regulation of retinoic acid biological effects is achieved by enzymes of the cytochrome P450 family and belonging to the CYP26 subfamily. These enzymes have been implicated in the catabolism of retinoic acid into various metabolites with minor biological activity (such as 4-oxo-RA, 4-OH-RA or 18-OH-RA): perhaps by hydroxylation on C4 or C18 of the β-ionone ring of RA (Theodosiou et al., 2010). For a while, some controversy surrounded the role of CYP26: would it generate these metabolites, assuming they could play a role, or restrict RA effects by degrading it. Studies using Cyp26 KO animals have confirmed the latter hypothesis, as disruption of the RA synthetic enzyme RALDH2 in mice rescued the severe phenotypes observed in Cyp26 null animals (Thatcher and Isoherranen, 2009). It is now widely accepted that Cyp26 and RA synthesis enzymes act in opposite spatial distribution, allowing the creation of RA gradients responsible for specific cellular functions (Simões-Costa et al., 2008; Theodosiou et al., 2010). Figure 1.6 Overview of the retinoids and the enzymatic machinery involved in vertebrate retinoid synthesis, storage, metabolism and signaling pathways. In the amniotes canonical synthesis pathway retinol is reversibly oxidized to retinal by ADHs/RDHs. The retinal is irreversibly converted to RA by ALDHs. RA in the 9-cis conformation binds RXR and RAR, while all-trans RA binds RAR only and activation of the RAR-RXR heterodimer regulates gene transcription. Retinoids are stored mainly in the form of retinyl esters, which can be converted back to retinol retinyl ester hydrolases (REHS). Circulating retinol binds to Rbp and Ttr, and enters the cells by binding to Stra6. Within the cell, retinol is usually bound to Crbp. The acenstral RA synthetic route involves the degradation of β-carotene to retinal by BCO enzymes. RA is inactivated by CYP26 enzymatic metabolization, yielding other metabolites (adapted from Albalat, 2009 and Simões-Costa et al., 2008). Chapter 1 23 1.3.5 Retinoic acid receptors What was for a longtime a subject of intense speculation is nowadays consensual: most RA biological activity in vertebrates is mediated by the heterodimer RAR/RXR (Mark et al., 2009). Both RAR and RXR share a similar structure to that of other nuclear receptors consisting of five to six modular units: A / B, C, D, E and F (Germain et al., 2003; Figure 1.7). The AB region contains the ligand independent activation function 1 (AF1), to which coactivator binds (Germain et al., 2003). Next, there is the C region or DBD that recognizes response elements half sites in the promoters of the genes. This recognition is mediated through two structures, named zinc fingers, consisting of a complex of 4 cysteines surrounding a zinc ion (ZnII;Dawson and Xia, 2012). In the junction between the DBD and the LBD (region E), is located the D region, an area that is believed to behave as a hinge, conferring flexibility to the receptor and enabling it to bind to different NRs. The LBD is a very conserved structure composed of 12 α helixes and a small β-sheet between H5 and H6, that, in the absence of a ligand, forms what is generally called an “anti-parallel helical sandwich” (Pérez et al., 2011; Samarut and Rochette-Egly, 2011). This central hydrophobic cavity is also known as the "ligandpocket”; the specificity of the ligand towards the ligand pocket determines a particular physiological response (Samarut and Rochette-Egly, 2011). Upon ligand binding, α helixes suffer a conformational shift that allows the release of co-repressors, binding of co-activators and consequent recruitment of other regulatory proteins (de Groot et al., 2005). Besides interacting with the ligand and the coactivators, the LBD also functions as a dimerization interface between NRs. Finally, in the C-terminal end of the receptor, is the F region, not very conserved and still of unknown function. Chapter 1 30 1.5 Evolution meets toxicology The implications of the existence of functional retinoid signaling outside deuterostomes are important not only for the understanding of the evolution of the species, but also from a toxicological perspective. In particular, comprehension of the evolutionary process responsible for the interand intra-specific genomic diversification of NR is of major importance and thus the full range of metazoan phyla must be considered (Figure 1.10). As aforementioned, various classes of receptors have been found in diploblastic (Cnidaria) and in triploblastic animals, indicating that considerable variation already existed in this super-family before the appearance of the bilateral symmetry in metazoans (Thornton, 2003). This ancestry makes the vast majority of animal species potential targets of NR-mediated endocrine disruption. However, it is the same evolutionary diversity that mines the extrapolation of the functions of receptors (and compounds capable of disrupting them) from vertebrates to other groups. Hence, when the model studies are vertebrate-specific and their effects are not known in invertebrates, it is abusive to assume that a particular compound is a universal endocrine disruptor. (Thornton, 2003). Conversely, a compound designed to bind to a specific vertebrate NR may end up activating unknown signaling pathways in non-target invertebrates. In vertebrates, it is known that RA is involved in numerous physiological processes, including the establishment of anterior-posterior patterning in embryos (White et al., 2007), the regulation of the immune system, reproduction and vision (Simões-Costa et al., 2008). Direct or indirect disturbance of this signaling pathway results in birth defects, fertility problems, vision constraints, tumors and neurodegenerative diseases (Niederreither and Dollé, 2008). As a corollary, one may hypothesize that the higher ancestry of the retinoid signaling pathway turns a greater number of metazoan species than previously anticipated susceptible to EDCs, acting via this retinoid signaling pathway. The TBT case is the perfect example of such situations. Even if laboratory studies had been performed to detect its binding abilities towards vertebrate’s nuclear receptors, one could never predict such deleterious effects in aquatic mollusks, for retinoid signaling in this group had not been discovered. Two species that belong to phylogenetically distant groups (mollusks and vertebrates, Figure 1.10) have been selected to study the proposed hypothesis: endocrine disruption caused by TBT involves modulation of conserved retinoid signaling pathways. The selected species, N. lapillus and D.rerio have long been used in toxicological studies and biological impact has been reported upon TBT exposure. Apart from RXR, no gene involved in retinoid signaling has been cloned in mollusks. Therefore, Chapter 1 31 the study of this signaling pathway is a pre-requisite to be fulfilled, before the assessment of the impact of TBT exposure. As such, this study will also contribute to the understanding of retinoid signaling evolution throughout metazoans. Figure 1.10 Phylogenetic relationships between Metazoan phyla. Blue boxes indicate animal lineages where NR mediated endocrine disruption has been suggested. Note: the cephalochordate phylogenetic position within chordates has been reviewed in the past years. They are now considered basal chordates and the urochordates the sister clade of vertebrates. Modified from Holland (1999). Chapter 1 32 1.6 The animal models 1.6.1 The dogwhelk, N. lapillus N. lapillus (Linnaeus, 1758) is a mollusk that belongs to the Gastropoda (order Neogastropoda and family Muricidae), the largest and more diverse class of the phylum Mollusca, with over 62 000 described species. N. lapillus has a wide geographical distribution and can be found all along the Atlantic coast of Europe (from the 73oN in the Barents Sea to 37oN in southern Portugal) and on the Atlantic coast of North America (since 50oN Notre Dame Bay to 41oN Long Island; Crothers, 1985). This species is very common in the intertidal zone of rocky shores, both in exposed and sheltered areas, which, according to several studies, influences the shape of the shell (Crothers, 1975, 1977, 1983, 1985). N. lapillus is a very common active predator in the intertidal zone and therefore of great importance in controlling the major occupiers of that environment (Menge, 1976; Burrows and Hughes, 1991). It feeds mostly on mussels (Mytilus sp) and barnacles (Semibalanus balanoides, Balanus sp, Elminius modestus, Cthamalus sp), but also on limpets and other mollusks, though in less quantities (eg, Patella sp, Gibulla sp, Monodontas lineata, Ostrea edulis and Littorina sp; Crothers, 1985). In these gastropods, the sexes are separated (gonochoric) and fertilization is internal (Fretter, 1953). In the breeding season, they form aggregates of up to 30 adults in crevices and pools. Reproduction seems to occur throughout the year, although in the British shores a peak in spring and winter is evident (Moore, 1936; Hughes and Burrows, 1993). After mating, females lay eggs in protective capsules which can contain up to 600 eggs. Only 6% the eggs are viable and develop into embryos which feed on the others: the so-called “nurse eggs”. Embryonic development can last up to five months in temperate zones (Feare, 1970), and comprises a trochophore stage, followed by the veliger, after which the juveniles hatch. These are very similar to adults, differing in size and reproductive capability (Feare, 1970). Their ecology has been thoroughly studied and they have been used for several decades as a sentinel species in monitoring studies, mostly associated with TBT contamination. In addition, the various stages of imposex development have been thoroughly described in the literature and their severity can be classified according to the general scheme proposed by Bettin et al., (1996), displayed in Figure 1.11. Hence, the large body of available literature favors the use of this species as a model organism in the present work. Chapter 1 33 Figure 1.11 General scheme of imposex evolution in prosobranchs. Abbreviations: ac, aborted capsules; cg, capsule gland; gp, genital papilla; obc, open bursa copulatrix; ocg, open capsule gland; ocv, occlusion of the vulva; p, penis; pd, pems duct; pr, prostate; te, tentacle; vd, vas deferens; vdp, vas deferens passage into capsule gland; vds, vas deferens section (Bettin et al., 1996). 1.6.2 The zebrafish, D. rerio D. rerio, also known as zebrafish, is a freshwater fish of the Cyprinidae family, native to northeastern India and adjacent regions (Engeszer et al., 2007). This species has been widely used as model in many studies due to the large number of advantageous features, both from a technical and conceptual perspective. From a practical point of view, it stands out from other animal models for its easy and low maintenance costs, high fecundity, short life-cycle (completed in 10-12 weeks), short and fully described embryonic development (Kimmel et al., 1995; Dooley and Zon, 2000; Hill et al., 2005; Spence et al., 2008). Furthermore, numerous tools and information Chapter 1 34 are available for zebrafish studies, including the completely sequenced genome (http://www.ensembl.org/index.html) and a histological atlas of the gonads (http://www.rivm.nl/fishtoxpat). In the environment, zebrafish has a seasonal reproduction, but under optimal laboratory conditions, females can deposit between 50-200 eggs, 3-4 times per week. Fertilization is external; eggs are transparent and develop rapidly: 24 hours after fertilization (hpf) the main body plan is established and 96 hpf most organs are already fully developed (Spence et al., 2008). Zebrafish is considered an undifferentiated gonochoric species, meaning that all individuals undergo an initial phase in which their gonads are nonfunctional ovaries with only immature oocytes I (Örn et al., 2003). By the 23th day, 50% of the animals develop mature and functional ovaries, while the remaining ovarian tissues regress and produce functional testes (Hill and Janz, 2003). Exposure to estrogens has been reported to alter zebrafish sexual differentiation, but the mechanisms underlying this phenomenon are unknown (Van den Belt et al., 2003; Soares et al., 2009). Similarly, masculinizing effects have also been described, again without explanation (McAllister and Kime, 2003; Santos et al., 2006a). The understanding of such effects is delayed by the lack of knowledge on the molecular factors regulating its normal sexual determination and differentiation. It has been suggested that autosomal genes as Sox9a and Sox9b (this species has no identified sex chromosomes) or antiMuellerian hormone (Segner, 2009) are involved, but their role in sexual differentiation and in mediating the impact of EDCs is still unclear. From a conceptual standpoint, and despite the additional genome duplication that occurred in teleost fish (Hill et al., 2005), the use of fish species in the study of EDCs effects are relevant, since teleost and mammal’s close phylogenetic relationship facilitates the extrapolation of knowledge on diseases, biology, genetic processes and toxicology studies (Dooley and Zon, 2000; Spence et al., 2008). However, the need to further characterize the molecular and genetic mechanisms that regulate the sexual differentiation in zebrafish remains to be elucidated. Chapter 1 35 1.7 Objectives The central aim of this thesis is to address the mechanisms of endocrine disruption of a priority androgenic chemical, TBT. Given that the major drawback to address the mode of action of EDCs in invertebrates lies on the lack of knowledge of their endocrinology and molecular signaling pathways, it is an objective of this thesis to elucidate the presence of molecular signaling pathway involved in retinoid and, to a lesser extent, steroid metabolism. These two pathways were selected since they have been suggested to be major targets of TBT. Understanding the mode of action of endocrine disruptors such as TBT is essential not only to improve risk assessment of the chemical under study, but also to anticipate the impact of other chemicals acting through the same signaling pathways. The initial sections address the molecular target of TBT in N. lapillus, while also dissecting some of the molecular components of the retinoid and steroid signaling pathway in gastropod mollusks. The second section analyses the impacts of TBT in RA signaling in the teleost D. rerio. Given that both species belong to distinct phyla, their study is expected to provide insights into the evolution of this signaling pathway, and its exploitation by the ubiquitous contaminant TBT. In order to accomplish these generic aims, four specific objectives were defined: 1. To address the possible interference of TBT with retinoid signaling related genes in mollusks, in particular with RXR. N. lapillus RXR was first isolated and a pharmacological approach was used to characterize its involvement in the imposex development. In order to get further insights into the mechanisms of imposex induction by TBT, the transcription levels of RXR in different tissues after TBT exposure were quantified (Chapter 2); 2. To identify genes hypothetically involved in retinoid signaling in mollusks, as this pathway is yet to be described outside chordates, and gain more insights on their possible biological function. Two key genes hypothetically involved in retinoid metabolism in vertebrates (Adh3 and Cyp26) were isolated, its basal transcription patterns determined and possible modulation after TBT exposure assessed (Chapter 3). Since the RAR/RXR is the heterodimer mediating retinoid signaling in vertebrates, the first RAR Chapter 1 36 orthologue in a protostome species was isolated and its hypothetical function discussed (Chapter 3); 3. To identify other possible signaling pathways affected by TBT exposure, such as genes involved in steroid metabolism in mollusks. The 17β hydroxysteroid dehydrogenase type 12 (Hsd17b12) was cloned and its functional and toxicological aspects were evaluated (Chapter 4). 4. To evaluate the reproductive effects caused by abnormal exposure of zebrafish to retinoid agonists, with special emphasis on the consequences of TBT exposure in vivo, namely, in understanding the molecular pathways affected. In order to achieve this, a zebrafish life-cycle exposure to retinoid agonists was performed, their effects in reproduction assessed, and the possible impacted molecular pathways investigated (Chapter 5). The objectives of this thesis are explored through the following chapters and partly published in international journals, as indicated next: Chapter 1: Introduction Chapter 2: Retinoid signaling and imposex in N. lapillus 2.1 Imposex induction is mediated through the Retinoid X Receptor signaling pathway in the neogastropod Nucella lapillus, Castro, L. Filipe C., Lima, D., Machado, A., Melo, C., Hiromori, Y., Nishikawa, J., Nakanishi, T., Reis-Henriques, M. A., Santos, M. M., 2007. Aquat. Toxicol. 85, 57–66. (adapted from) 2.2 Tributyltin-induced imposex in marine gastropods involves tissue-specific modulation of the Retinoid X Receptor, Lima, D., Reis-Henriques, M.A, Silva, R., Santos, A.I., Castro, L. Filipe C., Santos, M.M., 2011. Aquat. Toxicol. 101, 221–227. Chapter 3: Retinoid signaling in N. lapillus: Adh3, Cyp26 and RAR Chapter 1 37 3.1 Molecular characterization of Adh3 from the mollusk Nucella lapillus: tissue gene expression after tributyltin and retinol exposure, Lima, D., Coelho, I., André, A., Melo, C., Ruivo, R., Reis-Henriques M.A., Santos, M.M., Castro, L.F.C (accepted by the Journal of Molluscan Studies) 3.2 Isolation of the first protostome cytochrome P26 orthologue in the imposexsensitive gastropod Nucella lapillus: molecular and toxicological insights Daniela Lima et al. (in preparation) 3.3 Isolation and basal characterization of the Retinoic Acid Receptor (RAR) in the gastropod mollusk Nucella lapillus Daniela Lima et al. (in preparation) Chapter 4: The 17β hydroxysteroid dehydrogenase type 12 in the neogastropod Nucella lapillus: functional and toxicological insights Lima, D., Machado, A., Reis-Henriques, M. A., Rocha, E., Santos, M. M., Castro, L. F. C (acceptable pending revisions by the Journal of Steroid Biochemistry) Chapter 5: Evaluation of the reproductive impact of TBT and other retinoid receptors agonists in the zebrafish, Danio rerio Daniela Lima et al. (in preparation) Chapter 6: General discussion, Final considerations and conclusions, Future perspectives, References Chapter 1 38 1.8 References Albalat, R. 2009. The retinoic acid machinery in invertebrates: ancestral elements and vertebrate innovations. Mol. Cell. Endocrinol. 313, 23–35. Albanis, T., Allera, A., Bachmann, J., Barbaglio, A. B. P., Dittmann, N., Carnevali, D. C., Ciceri, F., Dagnac, T. D. M., Falandysz, J., Galassi, S., Hala, D., Janer, G., Jeannot, R., J., S, King, I., Klingmuller, D., Kloas, W., Kusk, K.O., Lavado, R. L. S, Lutz, I., Oehlmann, J., Oredsson, S., Porte, C., R.-W., M, Sakkas, V., Schmitt, C., Schulte-Oehlmann, U., S. M., and Tyler X, van Aerle, R., van Ballegoy, C. W., 2006. Executive summary on the project results of COMPRENDO - Comparative Research on Endocrine Disrupters. Alzieu, C. l., Sanjuan, J., Michel, P., Borel, M., Dreno J. P., 1989. Monitoring and assessment of butyltins in Atlantic coastal waters, Marine Pollution Bulletin. Marine Pollution Bulletin 20. Barroso, C. M., Moreira, M.H., 2002. Spatial and temporal changes of TBT pollution along the Portuguese coast: inefficacy of the EEC directive 89/677. Mar. Pollut. Bull. 44, 480-486 Barroso, C. M., Moreira, M. H., Bebianno, M. J., 2002. Imposex, female sterility and organotin contamination of the prosobranch Nassarius reticulatus (L.) from the Portuguese coast. Mar. Ecol. Prog. Ser. 230, 127-135. Bauer, B., Fioroni, P., Schulte-Oehlmann, U., Oehlmann, J., Kalbfus, W. 1997. The use of Littorina littorea for tributyltin (TBT) effect monitoring - Results from the German TBT survey 1994/1995 and laboratory experiments. Environ. Pollut. 96, 299–309. Berto, D., Giani, M., Boscolo, R., Covelli, S., Giovanardi, O., Massironi, M., Grassia, L., 2007. Organotins (TBT and DBT) in water, sediments, and gastropods of the southern Venice lagoon (Italy). Mar. Pollut. Bull. 55, 425-435. Bettin, C., Oehlmann, J., Stroben, E., 1996. TBT-induced imposex in marine neogastropods is mediated by an increasing androgen level. Helgol. Mar. Res. 50, 299-317. Biddie, S. C., John, S., Hager, G. L., 2010. Genome-wide mechanisms of nuclear receptor action. Trends Endocrin. Met. 21, 3-9. Blaber, S. J. M., 1970. The occurence of a penis-like outgrowth behing the right tentacle in spent females of Nucella lapillus (L.) Proc. Malacol. Soc. London 39, 231-233. Blomhoff, R., Blomhoff, H. K., 2006. Overview of retinoid metabolism and function. J. Neurobiol, 606–630. Chapter 1 39 Bright, D. A., Ellis, D. V., 1990. A comparative survey of imposex in northeast Pacific neogastropods (Prosobranchia) related to tributyltin contamination, and choice of a suitable bioindicator. Can. J. Zool. 68: 1915-1924. Bruck, N., Vitoux, D., Ferry, C., Duong, V., Bauer, A., de Thé, H., Rochette-Egly, C., 2009. A coordinated phosphorylation cascade initiated by p38MAPK/MSK1 directs RAR alpha to target promoters. Embo. J. 28, 34-47. Bryan G. W., Gibbs P. E., Hummerstone L. G., Burt. G. R., 1986. The decline of the gastropod Nucella lapillus around south-west England: Evidence for the effect of tributyltin from antifouling paints. J. Mar. Biol Assoc. UK. 66, 611–640. Burrows, M. T., Hughes, R. N., 1991. Optimal foraging decisions by dogwhelks, Nucella lapillus (L.): Influences of mortality risk and rate-constrained digestion. Funct. Ecol. 5, 461. Campo-Paysaa, F., Marlétaz, F., Laudet, V., Schubert, M., 2008. Retinoic acid signaling in development: tissue-specific functions and evolutionary origins. Genesis. 46, 640656. Cañestro, C., Postlethwait, J. H., Gonzàlez-Duarte, R., Albalat, R., 2006. Is retinoic acid genetic machinery a chordate innovation? Evol. Dev. 8, 394-406. Chen, Y., Zuo, Z., Chen, S., Yan, F., Chen, Y., Yang, Z., Wang, C., 2008. Reduction of spermatogenesis in mice after tributyltin administration. Toxicology, 251, 21-27. Chen, Q.; Zhang, Z.; Zhang, R.; Niu, Y.; Bian, X., Zhang, Q., 2011. Tributyltin chlorideinduced immunotoxicity and thymocyte apoptosis are related to abnormal Fas expression. Int. J. Hyg. Envir. Heal., 214, 145-50. Clagett-Dame, M., Knutson, D., 2011. Vitamin A in reproduction and development. Nutrients. 3, 385-428. Cooke, G. M., 2002. Effect of organotins on human aromatase activity in vitro. Toxicol. Lett. 25, 121-130. Coray, C., Bard, S., 2007. Persistence of tributyltin-induced imposex in dogwhelks (Nucella lapillus) and intersex in periwinkles (Littorina littorea) in Atlantic Canada. Water Qual. Res. J. Can. 42, 111-122. Crothers, J. H., 1975. On variation in Nucella lapillus (L): shell shape in populations from the south coast of England. Proc. Malacol. Soc. London 41, 489-498. Crothers, J. H., 1977. On variation in Nucella lapillus (L): shell shape in populations towards the Southern limit of its European range. J. Mollus. Stud. 43, 181-188. Crothers, J. H., 1983. Variation in dog-whelk shells in relation to wave action and crab predation. Biol. J. Linn. Soc. 20, 85-102. Chapter 1 46 tissues by tributyltin and the second messenger activator, forskolin. Chem-Biol Interact. 185, 119-127. Pérez, E., Bourguet, W., Gronemeyer, H., Lera, A. R. D., 2012. Modulation of RXR function through ligand design. BBALipid Lipid Met. 1821, 57-69. Ronis, M. J. J., Mason, A.Z., 1996. The metabolism of testosterone by periwinkle (Littorina littorea) in vitro and in vivo: effects of tributyltin. Mar. Environ. Res. 42, 161-166. Saitoh, M., Yanase, T., Morinaga, T., Tanabe, M., Mu, Y-M., Nishi, Y., Nomura, M., Okabe, T., Goto, K., Takayanagi, R., 2001. Tributyltin or triphenyltin inhibits aromatase activity in the human granulosa-like tumor cell line KGN. Biochem. Biophys. Res. Commun. 289, 198-204. Samarut, E., Rochette-Egly, C., 2011. Nuclear retinoic acid receptors: conductors of the retinoic acid symphony during development. Mol. Cell. Endocrinol. 1-13. Sandell, L. L., Lynn, M. L., Inman, K. E., McDowell, W., Trainor, P. A., 2012. RDH10 Oxidation of vitamin A is a critical control step in synthesis of retinoic acid during mouse embryogenesis. PloS one 7, e30698. doi:10.1371/journal.pone.0030698. Santos, M. M., Hallers-Tjabbes, T. C., Santos, A., Vieira, N., 2002. Imposex in Nucella lapillus, a bioindicator for TBT contamination: re-survey along the Portuguese coast to monitor the effectiveness of EU regulation. J. Sea Res. 48, 217-223. Santos, M. M., Castro, L. F. C., Vieira, M. N., Micael, J., Morabito, R., Massanisso, P., Reis-Henriques, M. A., 2005. New insights into the mechanism of imposex induction in the dogwhelk Nucella lapillus. Comp. Biochem. Physiol. C. 141, 101109. Santos, M. M., Micael, J., Carvalho, A. P., Morabito, R., Booy, P., Massanisso, P., Lamoree, M., Reis-Henriques, M. A., 2006a. Estrogens counteract the masculinizing effect of tributyltin in zebrafish. Comp. Biochem. Physiol. Toxicol. Appl. Pharm. 142, 151-155. Santos, M. M., Reis-Henriques, M. A., Vieira, M. N., Solé, M., 2006b. Triphenyltin and tributyltin, single and in combination, promote imposex in the gastropod Bolinus brandaris. Ecotox. Environ. Safe. 64, 155-162. Santos, M. M., Enes, P., Reis-Henriques, M. A., Kuballa, J., Castro, L. F. C., Vieira, M. N., 2009. Organotin levels in seafood from Portuguese markets and the risk for consumers. Chemosphere. 75, 661-666. Santos, M. M., Reis-Henriques, M. A., Castro, L. F. C., 2011. Lipid homeostasis perturbation by organotins: effects on vertebrates and invertebrates. In Biochemical and biological effects of organotins, Bentham Science Publishers, 115-134. Chapter 1 47 Santillo, P. J., 2001. Chapter 13 - Tributyltin (TBT) antifoulants: a tale of ships, snails and imposex. In Late lessons from early warnings: the precautionary principle 18962000, EEA, Environmental issue report No 22/2001 135–148. Schubert, M., Yu, J.-K., Holland, N. D., Escriva, H., Laudet, V., Holland, L. Z., 2005. Retinoic acid signaling acts via Hox1 to establish the posterior limit of the pharynx in the chordate amphioxus. Development 132, 61-73. Segner, H., 2009. Zebrafish (Danio rerio) as a model organism for investigating endocrine disruption. Comp. Biochem. Physiol.. Toxicol. Appl. Pharm. 149, 187–195. Shimasaki, Y., Kitano, T., Oshima, Y., Inoue, S., Imada, N., Honjo, T., 2003. Tributyltin causes masculinization in fish. Environ. Toxicol. Chem. 22, 141-144. Shimasaki, Y., Oshima, Y., Inoue, S., Inoue, Y., Kang, I. J., Nakayama, K., Imoto, H., Honjo, T., 2006. Effect of tributyltin on reproduction in Japanese whiting, Sillago japonica. Mar. Environ. Res. 62, Suppl, S245–8. Siah, A., Pellerin, J., Amiard, J.-C., Pelletier, E., Viglino, L., 2003. Delayed gametogenesis and progesterone levels in soft-shell clams (Mya arenaria) in relation to in situ contamination to organotins and heavy metals in the St. Lawrence River (Canada). Comp. Biochem. Phys. C. 135, 145-156. Simões-Costa, M., Azambuja, A. P., Xavier-Neto, J., 2008. The search for non‐chordate retinoic acid signaling: lessons from chordates. J. Exp. Zool. 310B, 54-72. Smith, B. S., 1971. Sexuality in the american mud snail, Nassarius obsoletus Say. J. Molluscan Stud. 39, 377-378. Smith, B. S., 1981a. Male characteristics on female mud snails caused by antifouling bottom paints. J. Appl. Toxicol. 1, 22-25. Smith, B. S., 1981b. Reproductive anomalies in stenoglossan snails related to pollution from marinas. J. Appl. Toxicol. 1, 15-21. Smith, B. S., 1981c. Tributyltin compounds induce male characteristics on female mud snails Nassarius obsoletus (Ilyanassa obsoleta). J. Appl. Toxicol. 141-144. Soares, J., Coimbra, A. M., Reis-Henriques, M. A., Monteiro, N. M., Vieira, M. N., Oliveira, J. M. A., Guedes-Dias, P., Fontaínhas-Fernandes, A., Parra, S. S., Carvalho, A. P., Castro, L. F. C., Santos, M. M., 2009. Disruption of zebrafish (Danio rerio) embryonic development after full life-cycle parental exposure to low levels of ethinylestradiol. Aquat. Toxicol. 95, 330-338. Sonoda, J., Pei, L., Evans, R. M., 2008. Nuclear receptors: decoding metabolic disease. FEBS letters. 582, 2–9. Spence, R., Gerlach, G., Lawrence, C., Smith, C., 2008. The behaviour and ecology of the zebrafish, Danio rerio. Biol. Rev. 83, 13–34. Chapter 1 48 Spooner, N., Gibbs, P. E., Bryan, G. W., Goad, L. J., 1991. The effect of tributyltin upon steroid titers in the female dogwhelk, Nucella lapillus, and the development of imposex. Mar. Environ. Res. 2, 37-49. Sternberg, R. M., LeBlanc, G., 2006. Kinetic characterization of the inhibition of acyl coenzyme A: steroid acyltransferases by tributyltin in the eastern mud snail (Ilyanassa obsoleta). Aquat. toxicol. 78, 233-242. Sternberg, R.M., Hotchkiss, A.K., LeBlanc, G.A., 2008. The contribution of steroidal androgens and estrogens to reproductive maturation of the eastern mud snail Ilyanassa obsoleta. Gen. Comp. Endocrinol. 156, 15-26. Sternberg, R., Gooding, M., Hotchkiss, A., LeBlanc, G., 2010. Environmental-endocrine control of reproductive maturation in gastropods: implications for the mechanism of tributyltin-induced imposex in prosobranchs. Ecotoxicology 19, 4-23. Sternberg, R. M., 2012. Organotins as endocrine disruptors: an examination of tributyltininduced imposex in Neogastropods. In Biochemical and biological effects of organotins, Bentham Science Publishers, 75-82. Stroben, E., Oehlmann, J, B. C., 1991. TBT-induced imposex and the role of steroids in marine snails. Proceedings of the 10th World Meeting of the ORTEP Association, 68-73. Sumpter, J. P., 2005. Endocrine Disrupters in the Aquatic Environment: An Overview. Acta Hydrochim. Hydrobiol. 33, 9-16. Thatcher, J. E., Isoherranen, N., 2009. The role of CYP26 enzymes in retinoic acid clearance. Expert Opinion Drug. Metab.. Toxico., 1-21. Theodosiou, M., Laudet, V., Schubert, M., 2010. From carrot to clinic: an overview of the retinoic acid signaling pathway. Cell. Mol. Life Sci. 67, 1423-1445. Thornton, J. W., 2003. Nonmammalian nuclear receptors: evolution and endocrine disruption. Pure Appl. Chem. 75, 1827-1839. Tillmann, M., Schulte-Oehlmann, U., Duft, M., Markert, B., Oehlmann, J., 2001. Effects of endocrine disruptors on prosobranch snails (Mollusca: Gastropoda) in the laboratory. Part III: Cyproterone acetate and vinclozolin as antiandrogens. Ecotoxicology. 10, 373–388. Urquiza, A. M., Liu, S., Sjöberg, M., Zetterström, R. H., Griffiths, W., Sjövall, J., Perlmann, T. P., 2000. Docosahexaenoic acid, a ligand for the retinoid X receptor in mouse brain. Science. 290, 2140-2144. Van den Belt, K., Verheyen, R., Witters, H., 2003. Effects of 17alpha-ethynylestradiol in a partial life-cycle test with zebrafish (Danio rerio): effects on growth, gonads and female reproductive success. Sci. Total Environ. 309, 127-137. Chapter 1 49 Vasiliou, V., Pappa, A., Estey, T., 2004. Role of human aldehyde dehydrogenases in endobiotic and xenobiotic metabolism. Drug. Metab. Rev. 36:279-299. Viglino, L., Pelletier, E., St-Louis, R., 2004. Highly persistent butyltins in northern marine sediments: a long-term threat for the Saguenay Fjord (Canada). Environ. Toxicol. Chem. 23, 2673-2681. Von Lintig, J., Vogt, K., 2000. Filling the gap in vitamin A research. molecular identification of an enzyme cleaving beta-carotene to retinal. The J. Biol. Chem. 275, 1191511920. White, R. J., Nie, Q., Lander, A. D., and Schilling, T. F., 2007. Complex regulation of cyp26a1 creates a robust retinoic acid gradient in the zebrafish embryo. PLoS Biology. 5, 12. Wyss, A, Wirtz, G, Woggon, W, Brugger, R, Wyss, M, Friedlein, A, Bachmann, H, Hunziker, W., 2000. Cloning and expression of beta,beta-carotene 15,150dioxygenase. Biochem. Bioph. Res. Commun. 271, 334-336. Yamabe, Y., Hoshino, A., Imura, N., Suzuki, T., Himeno, S., 2000. Enhancement of androgen-dependent transcription and cell proliferation by tributyltin and triphenyltin in human prostate cancer cells. Toxicol. Appl. Pharm. 169, 177–184. Zhang, J., Zuo, Z., Chen, Y., Zhao, Y., Hu, S., Wang, C., 2007. Effect of tributyltin on the development of ovary in female cuvier (Sebastiscus marmoratus). Aquat. Toxicol. 83, 174-179. Zhang, J., Zuo, Z., He, C., Cai, J., Wang, Y., Chen, Y., Wang, C., 2009a. Effect of tributyltin on testicular development in Sebastiscus marmoratus and the mechanism involved. Environ. Toxicol. Chem. 28, 1528-1535. Zhang, J., Zuo, Z., He, C., Wu, D., Chen, Y., Wang, C., 2009b. Inhibition of thyroidal status related to depression of testicular development in Sebastiscus marmoratus exposed to tributyltin. Aquat. Toxicol. 94, 62-67. Zhang, J., Zuo, Z., Wang, Y., Yu, A., Chen, Y., Wang, C., 2011. Tributyltin chloride results in dorsal curvature in embryo development of Sebastiscus marmoratus via apoptosis pathway. Chemosphere. 82, 437-442. Zheng, R., Wang, C., Zhao, Y., Zuo, Z., Chen, Y., 2005. Effect of tributyltin, benzo(a)pyrene and their mixture exposure on the sex hormone levels in gonads of cuvier (Sebastiscus marmoratus). Environ. Toxicol. Pharm. 20, 361-367. Ziouzenkova, O., Plutzky, J., 2008. Retinoid metabolism and nuclear receptor responses: New insights into coordinated regulation of the PPAR-RXR complex. FEBS Letters. 582, 32-38. Chapter 1 50 Anonymous, 1990. Report of the Meeting MEPC (Marine Environment Protection Committee) 30/24, IMO, London. MEPC., 2001. International Convention on the Control of Harmful Anti-fouling Systems on Ships, IMO, London, 5 October 2001 WHO/IPCS report (International Program on Chemical Safety), 2002. Global Assessment of the state-of-science of endocrine disruptors. CHAPTER 2 CHAPTER 2 2. Retinoid signaling and imposex in Nucella lapillus 2.1 Imposex induction is mediated through the Retinoid X Receptor signaling pathway in the neogastropod Nucella lapillus Castro, L. Filipe C., Lima, D., Machado, A., Melo, C., Hiromori, Y., Nishikawa, J., Nakanishi, T., Reis-Henriques, M. A., Santos, M. M., 2007. Aquat. Toxicol. 85, 57–66. (adapted from) 2.2 Tributyltin-induced imposex in marine gastropods involves tissue-specific modulation of the Retinoid X Receptor Lima, D., Reis-Henriques, M.A, Silva, R., Santos, A.I., Castro, L. Filipe C., Santos, M.M., 2011. Aquat. Toxicol. 101, 221–227. Chapter 2 52 Chapter 2 53 2.1 Imposex induction is mediated through the Retinoid X Receptor signaling pathway in the neogastropod Nucella lapillus 2.1.1 Abstract The imposex phenomenon in female prosobranch gastropods provides one of the best documented examples of endocrine disruption in wildlife. While many field studies have demonstrated the negative impact of tributyltin (TBT) upon female gastropods, the mechanism(s) underlying imposex development has not yet been fully clarified. Over the years several hypotheses have been raised to determine the biochemical and molecular determinants of this process. Nevertheless, the interplay between the different suggested pathways (neuroendocrine, steroid and retinoid) is still unknown. Hence, through a combination of exposure experiments, we show that the 9-cis retinoic acid (9-cis RA), the proposed natural ligand of the retinoic X receptor (RXR), induces imposex in females Nucella lapillus to the same degree of tributyltin, when administered at similar concentrations (1µg/g body weight). Methoprene acid, a selective ligand for RXR, also induces imposex, albeit to lower degree than that of the positive control. In contrast, testosterone significantly induced imposex, but had no effect on female penis induction, while the neuropeptide APGWamide had no effect on imposex development. These results clearly demonstrate that imposex induction in N. lapillus is mediated through the modulation of the retinoic acid signaling pathways. In addition to the effects reported in female dogwhelks, both TBT and RA significantly increased male penis length, thus suggesting that TBT may also impact male secondary sex organs through the retinoic acid signaling pathways. As a step for future studies, we have cloned the orthologue of N. lapillus RXR and provide experimental evidence that it binds 9-cis RA. Finally, the basal expression level of RXR in several tissues of N. lapillus was determined through Real Time PCR, thus showing that RXR is ubiquitously expressed in mollusk tissues, with the highest expression levels being recorded in female and male gonads. The mechanistic impacts of the overall findings to the imposex process are discussed. Chapter 2 54 2.1.2 Introduction The development of male accessory sex organs in female prosobranch gastropods, a phenomenon termed imposex, provides a striking example of endocrine disruption. It is now well established that exposure to organotin compounds (e.g. tributyltin – TBT and for some species triphenyltin - TPT) is the proximal cause for this phenomenon. While the ecological and population impact of these compounds is well studied, the mechanism through which they induce and promote the development of a penis-like structure and a vas deferens in female snails is still to be deciphered. In fact, the molecular pathways controlling the development of secondary sexual organs in mollusks are poorly understood. Historically, several hypotheses have been raised to explain the chain of events and molecular factors leading to imposex development in gastropods. The original work of Féral and Le Gall (1983) with transplantation experiments suggested the hierarchic involvement of two illusive molecules, named the retrogressive factor (RF) and the penis morphogenetic factor (PMF). This study also suggested the fundamental role of two anatomical structures, the pedal and cerebropleural ganglia. In an experiment reminiscent of the work of Féral and Le Gall (1983), Oberdörster and McClellan-Green (2000) proposed that the neuropeptide APGWamide is the PMF since it induces imposex in Ilyanassa obsoleta. Despite these observations, APGWamide failed to promote imposex in the prosobranch gastropod Bolinus brandaris (Santos et al., 2006). The disruption of steroid signaling and physiological balance has also been proposed as a potential driver for imposex development. Some studies indicate that the androgenic effects of TBT appear to be caused by interference with steroid biosynthesis. Elevation of testosterone and/or testosterone/estradiol ratio has been reported in TBT laboratory exposed snails (Spooner et al., 1991; Schulte-Oehlmann et al., 1995; Bettin et al.,1996; Santos et al., 2005) and clams (Morcillo et al., 1998). This imbalance would be caused by TBT inhibition of aromatase, an enzyme responsible for the aromatization of androgens to estrogens (Bettin et al., 1996). This was further supported by the observations that an aromatase inhibitor was able to induce imposex under laboratory conditions (Bettin et al., 1996), and by the depressed aromatase activity in wild populations of Buccinum undatum affected by imposex (Santos et al., 2002). However, in female Nucella lapillus a selective aromatase inhibitor was also shown to induce imposex but to a significant lower extent than TBT, which suggests that aromatase inhibition may not be the primary mechanism involved in the development of imposex (Santos et al., 2005). Other hypotheses have been put forward to explain the steroid imbalance induced Chapter 2 55 by TBT. Hence, Ronis and Mason (1996) have demonstrated that TBT decreased the amount of testosterone sulphur-conjugates in the periwinkle (Littorina littorea), leading to an increase in the levels of free testosterone in the tissues. As this experiment has been performed under extremely high TBT levels, it may not reflect realistic exposure conditions in the field. More recently, Gooding et al. (2003) have demonstrated that TBT decreases the esterification of testosterone with fatty acids in the mud snail Ilyanassa obsoleta, thus leading to an increase in free testosterone which could then induce imposex. This observation has been partially corroborated by Santos et al. (2005) for N. lapillus. Recently, Nishikawa et al. (2004) demonstrated that TPT efficiently binds the Thais clavigera orthologue of the nuclear receptor Retinoid X Receptor (RXR). Moreover, an injection experiment with 9-cis RA, the suggested natural ligand of RXR, promoted the development of imposex in 50% of the injected females. In this context, organotins would mimic the endogenous ligand of RXR, and thus activate the signaling cascades which are retinoic acid dependent. This hypothesis has been contradictorily tested in N. lapillus, when compared to T. clavigera, since no imposex outgrowth was observed following 9-cis RA injection (Oehlmann et al., 2007). The interplay between the different suggested pathways (neuroendocrine, steroid and retinoid) is still unknown. Furthermore, most of the molecular targets proposed to be involved in this mechanism to date have not been described in mollusks. For example, the CYP19 gene has not been found outside chordates (Callard et al., 1984; Castro et al., 2005; Mizuta and Kubokawa, 2007). Finally, the different responses between species to 9cis RA and APGWamide exposure highlight the need for a more comprehensive reassessment of the mechanisms underlying imposex induction. Through a combination of exposure experiments, we show that the proposed natural ligand of RXR (9-cis RA) induces imposex in N. lapillus to the same degree of the positive control (TBT), when administered at similar concentrations (1µg/g body weight). Methoprene acid, a selective ligand for RXR, also induces imposex, albeit to lower degree than that observed for retinoic acid and TBT, while the neuropeptide APGWamide had no effect with respect to imposex induction. Testosterone significantly induced imposex, but had no effect on female penis length increment. As a step for future studies, we have cloned the orthologue of N. lapillus RXR and provide experimental evidence that it binds 9-cis RA. Finally, we determined the basal expression level of RXR in several tissues of N. lapillus through Real Time PCR. The mechanistic impacts of the overall findings to the imposex process are discussed. Chapter 2 62 Figure 2.1.2 Imposex frequency (A), VDSI (B), male penis length (C) and female penis length (imposex; D) in N. lapillus after 2 month-exposure to the different treatments (2 replicates per treatment). Values are mean ±S.E. (n males= 14-25, n females= 19-28). * p<0.05; ** p<0.01; ***p<0.001, significantly different from solvent control. 2.1.4.2 N. lapillus RXR (NlRXR) Through a combination of PCR strategies we isolated the orthologue of RXR in the gastropod N. lapillus (Figure 2.1.3). Our 5’RACE PCR uncovered two sequence variants with different sizes. The analysis of one of the PCR fragments with approximately 1Kb suggests that this is probably an incompletely spliced transcript. The sequence contains a significant number of a repetitive element typical of intron sequences (not shown), and thus will not be analysed further. To demonstrate the integrity of the cDNA assembled contig, a PCR with specific primers flanking the coding region was performed (Figure 2.1.1). The entire ORF encodes a potential protein with 441 amino acids. Unexpectedly, the sequencing of several clones demonstrated the existence of yet another variant. An in-frame 15 bp insertion in the T-box region is responsible for this variation (Figure 2.1.3A). This difference is most probably due to alternative splicing, although we have not confirmed this. The sequences were named NlRXRa (441 amino acids) and NlRXRb (446 amino acids) and have been deposited in GenBank (Accession numbers EU024473, EU024474). The sequence now retrieved has all the main features typical of the RXR nuclear receptor (Figure 2.1.3B and 2.1.3C). The degree of conservation is particularly evident in the DNA binding domain (DBD) and the ligand binding domain (LBD) with the previously described T. clavigera RXR (Figure 2.1.3). We next undertook phylogenetics to fully determine the orthology of the isolated sequence. The Maximum Likelihood analysis shows that the N. lapillus RXR sequence robustly groups with other described mollusk RXR sequences (Figure 2.1.4). Chapter 2 63 Figure 2.1.3 Alignment of the T-box region of the two variants found in N. lapillus (A), the DNA-binding domain (B), and the ligand-binding domain (C). The T-box insertion is shown in bold; box delimitates the Pbox and the D-box; AF2 region is underlined; black circles above sequence indicate residues known to interact with 9-cis RA; dots indicate insertion. Accession numbers: T. clavigera (TcRXR) AAU12572; B. glabrata (BgRXR) AAL86461; and Homo sapiens RXR alpha (HsRXRa) NP_002948. Chapter 2 64 Figure 2.1.4 Molecular Phylogenetic analysis of RXR by Maximum Likelihood method. The percentage of trees in which the associated taxa clustered together is shown next to the branches. 2.1.4.3 Ligand Binding Assay The LBD of NlRXR protein was expressed in E.coli as a fusion with GST and tested the binding ability to 9-cis RA. As results, we found that NlRXR efficiently binds to 9-cis RA. Scatchard analysis of the binding of [3H]9-cis RA to NlRXR yielded Kd values of 12.9 nM (Figure 2.1.5), that is similar to the value of T. clavigera RXR (15.2 nM; Nishikawa et al., 2004). The crystal structure of the human RXRa LBD bound to 9-cis RA has been reported (Egea et al., 2000). In this paper, Egea et al. (2000) demonstrated that 9-cis RA is buried in a hydrophobic pocket formed by residues located on helices H3, H5, H7 and H11, and the β-turn. 9-cis RA contacts to amino acids of hRXR including I268, C269, A271, A272, Q275, W305, N306, L309, F313, R316, L326, A327, V342, I345, V349, R371, C432 and H435. Recently, a similar finding has been reported for the RXR LBD 9-cis RA-bound of the mollusk B. glabrata (de Groot et al., 2005). Because these Chapter 2 65 residues are completely conserved in NlRXR (Figure 2.1.3C), it is reasonable that NlRXR binds to 9-cis RA. Figure 2.1.5 The LBD of N. lapillus RXR expressed in E. coli was incubated with increasing concentrations of 3H-labeled 9-cis RA in the absence (total binding) or presence of 400-fold nonlabeled 9-cis RA (nonspecific binding) – upper panel. Nonspecific binding was subtracted from total binding and plotted as specific binding. Scatchard analysis, specific 9-cis RA binding to dogwhelk RXR was transformed by Scatchard analysis and plotted – lower panel. Linear regression yielded Kd = 12.9 nM. 2.1.4.4 Tissue expression The expression of NlRXR was determined in several adult N. lapillus tissues through Real Time PCR (Figure 2.1.6). Given the very small degree of sequence variation between N. lapillus RXR isoforms, we designed primers which determine the combined expression of both NlRXRa and NlRXRb. Preliminary semi-quantitative PCR with primers flanking the reported T-box variation site showed that they are both expressed in all the tested tissues, although isoform a has a higher expression level than b (not shown). This approach shows that RXR is ubiquitously expressed in all the tested tissues. However, clear differences were observed between tissues, with the highest levels localized to the ovary and testis (Figure 2.1.6). Chapter 2 66 Figure 2.1.6 NlRXR tissue expression through Real Time PCR (see text for details). P – penis, IFA – imposex forming area, ApsF – female mantle and accessory pallial sex glands, GdM – male mantle and pallial portion of genital ducts, DgF – female digestive gland, DgM – male digestive gland, GF - female cerebral ganglia, GM – male cerebral ganglia, T – testis, and Oovary. 2.1.5 Discussion The imposex mechanism is still poorly understood. Although many hypotheses have been put forward, the key molecular determinants have remained elusive. Historically, the most important proposals have been the aromatase inhibition, which leads to testosterone tissue concentrations imbalance and the abnormal release of the neuropeptide APGWamide in response to elevate levels of TBT (Bettin et al., 1996; Oberdörster and McClellan-Green, 2000). More recently, Nishikawa et al. (2004) clearly pointed to the unique role of the nuclear receptor RXR in the initiation of the imposex development. In particular, organotins apparently mimic the role of the natural ligand, binding RXR with high affinity (Nishikawa et al., 2004). Despite these findings, Oehlmann et al. (2007) contradictorily showed that in N. lapillus 9-cis RA (a suggested natural RXR ligand) had no significant effects on imposex parameters after almost two months upon injection. This conceptual framework (interplay between retinoids, steroids and neuroendocrine factors), leads us to test the in vivo impact on imposex parameters of a number of compounds: (a) testosterone, (b) APGWamide, (c) two concentrations of 9-cis RA, (d) methoprene acid, and (e) TBT. The percentage of imposex-bearing females reached 100% in the positive control (TBT) and in the groups injected with two 9-cis RA Chapter 2 67 concentrations. Significant imposex induction was also observed in the methoprene acid injected animals (63%). To a lesser extend testosterone elevated as well the number of imposex females (30%). In a previous study with T. clavigera, a 9-cis RA (1µg/g body weight) injection clearly induced imposex (50% of imposex females, 1 month after injection), but this compound was slightly less effective than the positive control triphenyltin (Nishikawa et al., 2004). In the present study, not only all N. lapillus females injected with 9-cis RA developed imposex, as the severity of imposex development was identical to those females injected with 1µg/g body weight of TBTCl. These results confirm the early findings of Nishikawa et al. (2004) for T. clavigera, thus suggesting that the imposex induction effects of TBT in female gastropods is mediated through RXR. The selected TBT concentration (1µg/g body weight) is of environmental relevance and was used because it had previously been demonstrated to induce maximum female penis growth in N. lapillus over a two month period after injection (Stroben et al., 1992). The lowest 9-cis RA dose administered in the present study was identical to the TBT concentration used, and renders the same severity of imposex development, further suggesting that both compounds act at the same signaling pathway. Moreover, since in the previous study with T. clavigera not all females developed imposex after being injected with 1 µg/g body weight of 9-cis RA, we have selected two concentrations levels for the present study, 1 and 2 µg/g body weight. The lack of differences in the imposex parameters between the two 9-cis RA tested dosed is most likely associated with the fact that in N. lapillus 1µg/g body weight of 9-cis RA is enough to elicit maximum imposex development over the duration of the experiment. On the contrary, a recently reported experiment with similar conditions with the dogwhelk retrieved non-significant imposex induction (Oehlmann et al., 2007). This apparent contradiction is puzzling, given the degree of similarity in the experimental setup (e.g. 9-cis RA concentration). We propose that this difference is most likely due to the use of different carriers, i.e. vitamin A free peanut oil in Oehlmann et al. (2007) study, and FBS as carrier in ours and Nishikawa et al. (2004) studies. Perhaps dosing 9-cis RA in peanut oil affects its biodisponibility during the course of the experiment, rendering it less bioactive. The induction of imposex via RXR signaling is further confirmed by the injection with methoprene acid. It has been showed that methoprene acid directly binds to RXR and is a transcriptional activator in both insect and mammalian cells (Harmon et al., 1995). Furthermore, this activity is RXR-specific unlike the 9-cis RA, which binds weakly to RAR nuclear receptors (Harmon et al., 1995). Animals injected with methoprene significantly showed imposex induction. However, VDSI and female penis length did not attain the severity observed in animals exposed to 9-cis RA Chapter 2 68 and TBT, which suggests a suboptimal RXR binding and activation. This scenario is supported by binding assays with the mollusk B. glabrata RXR (Bouton et al., 2005). Through proteolysis protection assay, methoprene acid binds the receptor but only at higher concentrations suggesting that it can enter the ligand pocket but the binding is not with high affinity (Bouton et al., 2005). Thus, the loosely binding of methoprene acid most likely impacts the ability to fully promote imposex development. Despite these caveats, these results indicate that RXR is the primary target for TBT -mediated endocrine disruption in gastropods. A growing body of evidences indicates that testosterone may also have a physiological role in gastropod reproduction similar to that in vertebrates. Based on this assumption, two early studies on the mechanisms of TBT-induced imposex were able to demonstrate elevated levels of testosterone in N. lapillus imposex bearing females (Spooner et al., 1991; Bettin et al., 1996), and the ability of testosterone (either administered through injection or water) to promote female penis growth. Yet, both studies have used animals that already had imposex at the start of exposure, and thus imposex induction by testosterone could not be tested. Because female N. lapillus at Praia da Apúlia show a very low percentage of imposex (approximately 15%), it was possible to test this hypothesis. The experimental procedure including the testosterone concentrations selected were similar to those reported by Spooner et al. (1991). Our data corroborates previous studies showing that testosterone is able to induce imposex in N. lapillus, but it was far less effective than TBT and 9-cis RA in the increasing of imposex severity. In the Spooner et al. (1991) study, 0.1 and 10 µg testosterone administered through injection were able to promote penis growth after 42 days, which is not supported by the present study, where 1 µg testosterone administered twice did not show the capability to increase female penis length. As the dogwhelk females’ used in the Spooner et al. (1991) experiment already had a penis at the start of the experiment, one explanation for the observed differences may lie on the fact that penis promotion may require lower concentrations of testosterone than induction. Recent studies have demonstrated that free testosterone levels in mollusks are mainly controlled through esterification with fatty acids, which is mediated by a microsomal acyl-coenzyme A: testosterone acyltransferase (ATAT; Gooding and LeBlanc, 2001; Janer et al., 2005). Similar to other mollusks, testosterone in female N. lapillus tissues is mainly stored in the esterified form, which suggests that ATAT is equally active in dogwhelks (Santos et al., 2005). This process was shown to be one of the targets of TBT, as in the mud snail (I. obsoleta), TBT seems to interfere with testosterone esterification leading to a decrease of testosterone-fatty acid esters and to a concomitant increase amount of free testosterone Chapter 2 69 (Gooding et al., 2003). In a follow-up study, Gooding and Leblanc (2004) have recently demonstrated that normal I. obsoleta injected with testosterone (0.5 µg) can rapidly control excess free testosterone levels to concentrations similar of non-exposed animals. This may be one of the explanations why in our study only 35% of testosterone-injected females developed early imposex stages. Interestingly, in mammals, retinoids are mainly stored as retinyl esters by the action of lecithin: retinol acyltransferase (LRAT). More recently, it has been demonstrated the existence of an acyl-coenzyme A: retinol acyltransferase (ARAT) which is equally involved in retinyl esters synthesis (Kaschula et al., 2006). While the metabolism of retinoids in mollusks has never been studied, one cannot exclude the possibility that ARAT can also be present in mollusks, and may also be TBT sensitive. As has been shown in our study, potential alterations in retinoic acid metabolism could lead to imposex induction. Additionally, in mammals, steroids hormones such as progesterone have been demonstrated to inhibit ARAT (Kaschula et al., 2006). Furthermore, it has been demonstrated that retinoic acid is involved in the regulation of testicular functions in rodents, such as interfering with testosterone production. Whether ARAT is in fact present in mollusks, and whether any cross-talk between steroid hormones and retinoids do exist in gastropods should be addressed in future studies. The APGWamide injections had no effect on imposex parameters. This result follows similar findings reported by Santos et al. (2006) for imposex promotion with B. brandaris, which together do not support those obtained in I. obsoleta (Oberdörster and McClellan-Green, 2000). Although this outcome might be interpreted as conflicting results, they could be explained by species-specific response differences. In fact, the injected APGWamide (here and in other experiments) represents a synthetic peptide from Lymnaea stagnalis and not the endogenous peptide. Recently, it was suggested that mollusk APGWamide family members could signal through the gonadotropin releasing hormone receptor (GnRHR) along with the gonadotropin releasing hormone (GnRH; Rodet et al., 2005). This possibility implies that the APGWamide described imposex induction (Oberdörster and McClellan-Green, 2000), could result from GnRhR activation. In this context, the potential involvement of a receptor could also explain the different reported sensitivities given the expected differences in receptor protein sequence between species. Interestingly, the Octopus GnRH-like peptide has been shown to participate in steroidogenesis via GnRHR in the male and female reproductive tissues (Kanda et al., 2006). This observation is even more relevant when we take into account the fact that TBT alters steroid titters in mollusks (Spooner et al., 1991). Finally, 9-cis RA is a regulator (though negative) of GnRH gene expression in immortalized GnRH neurons (Cho et al., 2001). Likewise, in the invertebrate chordate Ciona intestinalis embryo Chapter 2 70 exposure to all-trans RA induces GnRH-II up-regulation (Ishibashi et al., 2005). The interception of these lines of evidence represents a scenario that integrates the interaction between 3 cascades (retinoic, neuroendocrine and steroid) on imposex induction. We propose that the endogenous action of APGWamide and GnRH (both protein and mRNA) should be characterized in the context of imposex development to fully determine their role in the outgrowth of male genitalia in female gastropods. The length of male penises was also found to be significantly elevated in the specimens injected with either TBT or 9-cis RA. This result highlights the process similarity between imposex formation and accessory sex organ development in male gastropods. Taking together these findings and the fact that retinoids are also known to play an important role in mammalian male reproductive organ development (Ogino et al., 2001; Livera et al., 2002), it may be suggested that male reproductive control mechanisms involving retinoid signaling pathways are ancient and have been conserved throughout Metazoa evolution. This hypothesis is currently being addressed in more detail at our laboratory. To determine whether NlRXR binds 9-cis RA (as suggested in vivo) we performed an in vitro analysis to test this possibility. Our results indicate that indeed that is the case since the LBD of NlRXR binds with strong affinity to the suggested endogenous ligand. This pattern is similar to that reported for other mollusk species like T. clavigera and B. glabrata (Nishikawa et al., 2004; Bouton et al., 2005). We should note however that the role of 9-cis RA as RXR’s natural ligand has been questioned. In mouse keratinocytes, a critical experiment demonstrated that 9-cis RA is not the ligand for RXR (Callejá et al., 2006). Thus, future studies should also approach whether 9-cis RA in mollusks and in the context of imposex represents the physiological RXR ligand or not. The nuclear receptor RXR has been characterized in several invertebrate species, but these are mostly arthropods. In mollusks, only for T. clavigera and B. glabrata has the cDNA been isolated, but without a precise documentation on basal tissue expression. We have isolated the orthologue of this gene family in N. lapillus. Our PCR strategy uncovered two sequence variants, which differ in an insertion/deletion of 5 amino acids in the T-box (region adjacent to the DBD, Figure 2.1.4A), probably due to alternative splicing. A similar finding has been reported for the RXR sequence of crustaceans (Durica et al., 2002; Wu et al., 2004; Kim et al., 2005). The T-box, which is located in the Nterminal part of the D domain, is also important for DNA binding (Orlowski et al., 2004). It plays an important role mediating hormone response element binding interactions with RXR homodimers (Zhao et al., 2000). In crustaceans this difference in protein sequence impacts DNA-binding kinetics (Wu et al., 2004). We found the sequence carrying the Chapter 2 71 inserted 5 amino acids to have a lower expression level, when assayed through semiquantitative PCR (not shown). The biological relevance of this finding in the context of imposex will be investigated in the future, in particular if they involve differences on downstream RXR responsive element selection or heterodimer partner recruitment. As a framework for future studies, we analysed the basal tissue expression profile of NlRXR through Real Time PCR. Although RXR mRNA was found in every tested tissue, a strong differential expression was detected in the gonads, in particular the ovary. While comparable data is absent regarding mollusks, in arthropods the gonads are similarly major expression organs (e.g. Durica et al., 2004). In vertebrates, RXR has been implicated in reproduction, with male RXRB null mice being sterile (e.g. Kastner et al., 1996). Apart from imposex, organotins have been shown to elicit abnormal ovarian spermatogenesis in several mollusk species, like the abalone and the dogwhelk (Horiguchi et al., 2002; Horiguchi et al., 2006; Gibbs et al., 1988). Interestingly, organotin concentration in the gonads is positively correlated with imposex length in the females of Babylonia japonica (Horiguchi et al., 2006). Whether organotin accumulation in the gonads impacts cell differentiation via RXR or the precise role of the gonads in the imposex induction is still to be addressed. In summary, the findings reported here clearly support the premise that RXR is a molecular target of organotins in N. lapillus. In this context, imposex development begins with the activation of a signaling cascade which is dependent of the RXR activation/inhibition. Furthermore, the fact that male penises are also affected either by TBT or 9-cis RA, suggests that the normal process of accessory sex organ development in gastropods is retinoic dependent. To determine the temporal RXR expression pattern on precise anatomical structures upon organotin stimulation, RXR gene targets (e.g. gonadotropin releasing hormone), as well as investigating potential RXR heterodimeric partners remains a future challenge. 2.1.6 Acknowledgments We acknowledge Hugo Santos for his help with the aquariums maintenance. The authors thank two anonymous referees for their insightful comments. This research has been supported by project POCI/MAR/59462/2004 (Portugal). Daniela Lima was funded by Fundação para a Ciência e a Tecnologia (SFRH / BD / 41561 / 2007). Chapter 2 78 2.2.2 Introduction Imposex is one of the best documented examples of endocrine disruption in wildlife. It is characterized by the superimposition of male characteristics, such as a penis and a vas deferens, onto females of marine gastropods. Ever since it was first described in Nucella lapillus (Blaber, 1970), numerous studies have been published on the subject. A clear association between exposure to tributyltin (TBT), the active ingredient in antifouling paints, and imposex has been demonstrated for several species. Currently, at least 195 species of prosobranch gastropods are known to be affected, albeit the mechanisms responsible for it are yet to be fully elucidated (Sternberg et al., 2010). A basic understanding on mollusks’ endocrinology is still today far from achieved, which hinders our comprehension of the imposex process. Several hypotheses have been raised over the years to explain this condition. A series of in vitro transplantation experiments using the prosobranchs Ocenebra erinacea and Crepidula fornicata are still today particularly informative (Féral and Le Gall, 1983a). This early study indicated that TBT, at environmentally relevant levels, acts on the cerebropleural ganglia, through the action of a “retrogressive factor” (RF), leading to the abnormal release of a “penis morphogenetic factor” (PMF) by the female pedal ganglia. This work highlighted the pivotal role that the central nervous system (CNS) has in female gastropod masculinization by TBT. Other tissues did not intervene directly in imposex induction. Similarly, available evidences indicate that male penis formation in gastropods is under the control of the CNS (Féral and Le Gall, 1983b). In a study reminiscent of the work of Féral and Le Gall (1983a), Oberdörster and McClellan-Green (2000) proposed that the PMF could be the neuropeptide APGWamide, as it was able to induce imposex in Ilyanassa obsoleta and APGWamide immunoreactive neurons were detected in the CNS of several gastropods’ species (de Lange and van Minnen, 1998). However, experiments conducted by (Santos et al., 2006) and (Castro et al., 2007) showed that APGWamide failed to induce imposex in Bolinus brandaris and N. lapillus. Since testosterone itself was shown to induce imposex in several gastropod species, some alternative hypothesis postulated that TBT may impact testosterone metabolism (Spooner et al., 1991). It has also been suggested that TBT would competitively inhibit P450 aromatase activity, thereby preventing the conversion of androgens to estrogens (and consequently increasing testosterone levels; Bettin et al., 1996; Santos et al., 2002); alternatively, TBT would inhibit testosterone excretion (Ronis and Mason, 1996). Recently, the interference of TBT in steroid balance was proposed to be due to a decrease in the esterification of testosterone, thus leading to an increase in free testosterone which could then induce imposex (Gooding et al., 2003). Despite these Chapter 2 79 observations, the existence of a functional androgen signaling pathway in mollusks remains controversial (Castro et al., 2005; Markov et al., 2009; Sternberg et al., 2008a). The discovery of the intriguing ability of TBT to bind and activate the human retinoid X receptor (RXR) at the same levels of its natural ligand 9-cis retinoic acid (9-cis RA) has expanded our understanding of the process (Nishikawa et al., 2004). In the wild rock shell Thais clavigera, injection of 9-cis RA into females was able to induce the development of imposex (Nishikawa et al., 2004). This finding was further confirmed with the dogwhelk N. lapillus, where injections of 9-cis RA and a selective RXR agonist, methoprene acid, induced imposex (Castro et al., 2007), thus reinforcing the hypothesis of an RXR-mediated induction of imposex by TBT. In vertebrates, the retinoic acid signaling pathways regulate genes involved in many biological processes, such as cell proliferation, differentiation and apoptosis during embryonic development and other physiological processes in adults (Albalat and Cañestro, 2009; MacLean et al., 2007). The male reproductive differentiation also seems to be under control of retinoid action, as 50% of RXRβ disrupted mice die before or at birth, and those that survive become sterile (Kastner et al., 1996). RXR isoforms are also differentially expressed during mice external genitalia formation (Ogino et al., 2001). In invertebrates, retinoic acid signaling is less known, although RXR has been identified in a wide range of metazoans, i.e., sponges (Wiens et al., 2003), arthropods, nematodes, platyhelminthes and mollusks (Simões-Costa et al., 2008). So far, RXR has been characterized in various mollusks’ species, namely, T. clavigera, Biomphalaria glabrata, N. lapillus and I. obsoleta (Nishikawa et al., 2004; Bouton et al., 2005; Castro et al., 2007; Sternberg et al., 2008b). Previously, we have isolated the orthologue of RXR in N. lapillus and demonstrated that NlRXR effectively binds to 9-cis RA in vitro (Castro et al., 2007). NlRXR gene expression levels were determined through qPCR, showing RXR to be expressed ubiquitously. In addition, RXRs proposed ligand 9-cis RA and an RXR agonist metophrene acid induced imposex in female N. lapillus and impacted male penis outgrowth (Castro et al., 2007). Here, we test whether TBT exposure would alter RXR levels in various target tissues. We hypothesize that the patterns of RXR expression upon TBT exposure may provide significant insights into the involvement of RXR in normal male penis development and imposex induction, elucidating namely the timings and tissues targeted by organotins. Thus, in the present study, N. lapillus specimens were exposed to an environmentally relevant concentration of TBT (100 ng Sn/L TBT; Berto, 2007), and RXR expression determined before and after imposex initiation. RXR gene expression was evaluated in potential target tissues: the CNS, penis/penis forming area (PFA), gonads, a tissue which previously indicated a role Chapter 2 80 of RXR in reproductive recrudescence (Sternberg et al., 2008b) and digestive gland, a metabolic tissue. The biological relevance of the findings is discussed with respect to imposex and male penis development in gastropod mollusks. 2.2.3 Material and Methods 2.2.3.1 Chemicals TBTCl (96%) and DMSO were purchased from Sigma-Aldrich. 2.2.3.2 Experimental conditions and dosing Adult N. lapillus were collected in January 2009 at praia da Apúlia, an area in the North of Portugal known to display an imposex incidence below 5% (Santos et al., unpublished data). Animals were allowed to acclimatize to laboratory conditions for one week before the onset of the experiment. After this period, about 35 specimens per replicate (three replicate per treatment) were placed in 30 L aquaria filled with artificial salt water (salinity 35 ppm, pH 8.3, conductivity = 48 ms/cm, redox potential =−76 mV), and maintained at 16.5 oC ± 1 in an acclimatized room under a photoperiod of 12 hours light: 12 hours dark, with artificial airing. Salt water was prepared with sera premium salt and carbon activated filtrated tap water, and changed three times a week. Animals were fed with mussels from their origin site once a week. Two treatment groups were set: solvent control (DMSO) and TBT Cl at 100 ng Sn/L TBT. The percentage of solvent in the aquaria did not exceed 0.0002%, a concentration known to have no effect in imposex development (Santos et al., 2005). The kinetics of imposex induction after TBT exposure has previously been study in N. lapillus specimens from Apúlia (Santos et al., unpublished data; Castro et al., 2007). At the TBT concentrations and water temperature used in the present study, the first signs of imposex appear around week five post-exposure. Hence, we selected sacrificing animals at one month exposure (before the first signs of imposex) and after two months, where the majority of females is expected to have developed imposex. After one month of exposure, 24 animals (twelve males and twelve females) per treatment were sacrificed. Sex, maturation status and penis size/imposex were evaluated under a binocular microscope. The selected tissues (i.e., CNS, gonad, digestive gland, penis/PFA) were dissected with the help of a surgical scissor under the binocular microscope (80 x amplification), and stored separately in RNAlater (Sigma) at -80 oC. The dissection started with the collection Chapter 2 81 of penis or PFA which are located behind the right tentacle. To collect the CNS, a cut between both tentacles was performed in the dorsal area allowing the collection of the CNS. Finally, approximately 50 mg of gonad and digestive gland were dissected. In order to avoid RNA degradation or contamination among tissues, the dissection material was cleaned with water and RNase away. The CNS and the female PFA weighted approximately 2-3 mg, whereas male and female penis weight varied between 3-7 mg. Two months after the initiation of exposure, the remaining animals were sacrificed and the same parameters evaluated. The severity of imposex (measured as Vas Deference Sequence Index-VDSI) was determined using the imposex scheme of (Bettin et al., 1996). No differences in mortality rates were observed between treatment groups. 2.2.3.3 Tissue RNA extraction RNA extraction in gonads, digestive gland and CNS was performed using the Kit illustra RNAspin Mini RNA Isolation, GE Healthcare (animal tissues protocol), with on column DNAseI digestion. Given the low amount of tissue and RNA concentration in penis and PFA, RNA was extracted using the kit Qiagen RNeasy® Plus Micro, which is adequate for samples with less than 5 mg of tissue. The cDNA synthesis was performed with the iScript™cDNA Synthesis Kit (Bio-Rad) according to the manufacturer’s instructions, using 500 ng of total RNA for gonad, digestive gland and CNS and 50 ng of total RNA for penis and PFA. 2.2.3.4 qPCR assays Tissue expression levels of NlRXR were determined by qPCR (Bio-Rad, iQ5). Briefly, 0.8 µL cDNA was added to a reaction mixture containing 1x iQ SYBR Green supermix (Bio-Rad) 2 µM of each primer, in a final volume of 20 µL. In each plate, a “no template control” was included, and samples were run in duplicate. qPCR profile and primers used were previously described (see Castro et al., 2007). 18s was used as reference gene, and primers sequences and annealing temperature used in the different PCR reactions are given in Table 2.2.1. Real time PCR primers were designed using the Software Beacon Designer 5. PCR profile had the following conditions: 95 oC of initial denaturation during 10 min; 95 oC 15s, 56 oC annealing 30 s, and 72 oC extension 30 s (data collection) for forty cycles. A melting curve was generated for every run to confirm the specificity of the assays. Relative gene expression was calculated using the 2-∆∆Ct formula (Livak and Schmittgen, 2001). NlRXR gene expression in each organ was normalized to 18s, except for gonads, for which we could not find a stable reference gene, and therefore data is presented without normalization (2-∆Ct). Chapter 2 82 Table 2.2.1 Primer sequences used to isolate 18s in N. lapillus. Primer name Sequence Annealing temperature (oC) 18s F 5’-CTATTGGAGGGCAAGTCTGG-3’ 50 18s R 5’-GGTGAGTTTTCCCGTGTTGA-3’ 18s real time F 5’-ATTCGCTGGTGTTGCTTCATC-3’ 56 18s real time R 5’-TCCTGGTGGTGCCCTTCC-3’ 2.2.3.5 Statistical analysis Statistical analyses were performed using the software Statistica 7.0. MannWhitney U-test was used to compare, for each time-point and sex, statistical differences between solvent control and TBT exposed groups. One-way ANOVA, followed by a Student-Newman-Keuls multiple comparison test was used to compare RXR gene expression in female PFA at the end of exposure and male penis lengths. Chi-square test was used for comparing imposex frequency between groups, using imposex frequency of the control group as expected values. 2.2.4 Results 2.2.4.1 Imposex and experimental parameters Figure 2.2.1 (A, B and C) displays imposex indexes (VDSI, imposex frequency and female penis length, respectively) in N. lapillus during the course of the experiment. No differences were observed in any of the analyses indexes after one month. However, exposure to TBT significantly induced imposex in female N. lapillus after two months, if compared with the control group. Similarly, male penis lengths from the TBT exposed groups were significantly increased at the end of the experiment in comparison with the control group (Fig. 2.2.2). Chapter 2 83 Figure 2.2.1 Imposex parameters in N. lapillus after one and two months of exposure: VDSI (A), Imposex frequency (B) and female penis length (C). Values are mean ± S.E. (n= 1246). *** p ≤ 0.001, significantly different from control. Chapter 2 84 Figure 2.2.2 Temporal change of male average penis length throughout the experiment. Values are mean ± S.E. (n= 1227). Different letters indicate significant differences, p ≤ 0.05. 2.2.4.2 NlRXR expression Figures 2.2.3-5 display N. lapillus RXR gene expression during the course of the experiment. TBT-exposed animals displayed a decrease in RXR gene expression in the CNS, either after one or two months, which was statistically significant in females (P ≤ 0.01 and P ≤ 0.05, respectively) and a similar trend in RXR mRNA levels was observed in males, though differences were not significant (P = 0.059; Fig. 2.2.3). Regarding RXR gene expression in penis and PFA, significant differences between control and TBT exposed snails were not observed after one month exposure for females (Fig. 2.2.4A) and during the entire experiment for males (Fig. 2.2.4). In contrast, after two months of exposure to TBT a significant (P ≤ 0.001) increase in RXR gene expression was observed in females with imposex VDS 3-4 in comparison with both control and imposex females with VDS 1-2 (Fig. 2.2.4B). The absolute levels of normalized RXR gene expression in the penis of females displaying imposex VDS levels of 3-4 were comparable to those of male penis from the TBT treatment group. In contrast to the CNS and PFA, RXR gene expression in gonads and digestive gland was not affected by TBT, at any of the analyzed time-points (Fig. 2.2.5). Chapter 2 85 Figure 2.2.3 Normalized RXR gene expression in male and female CNS after one (A) and two (B) months of exposure. Values are mean ± S.E. (n= 6-9). * p ≤ 0.05; ** p ≤ 0.01, significantly different from solvent control. Chapter 2 86 Figure 2.2.4 Normalized RXR gene expression in male penis and female penis/PFA after one (A) and two (B) months of exposure. Values are mean ± S.E. (n= 5-12). ***p ≤ 0.001 indicates significant differences. Chapter 2 87 Figure 2.2.5 RXR gene transcription in male and female gonad (A) and normalized RXR gene transcription in male and female digestive gland (B). Values are mean ± S.E. (gonad: n= 7-8; digestive gland: n= 5-8). 2.2.5 Discussion It has recently been shown that TBT causes ligand dependent transactivation of the human RXRα and that it binds with high affinity to both human and rock shell RXR (Kanayama et al., 2005). Injection of females with its suspected natural ligand, 9-cis RA, demonstrated that 9-cis RA induces imposex both in T. clavigera (Nishikawa et al., 2004) and N. lapillus (Castro et al., 2007), to the same extent of TBT and within the same range of concentrations (Castro et al., 2007). Hence, compiling evidence indicates that the primary target of TBT in gastropods, in the context of imposex induction, involves an inappropriate regulation of RXR signaling pathways. To determine the cascade of molecular and physiological events leading to imposex and male penis development, we must first understand (a) which organs are targeted by TBT mediated-disruption involving Chapter 2 94 Durica, D. S., Wu, X., Anilkumar, G., Hopkins, P.M., Chung, A. C. K., 2002. Characterization of crab EcR and RXR homologs and expression during limb regeneration and oocyte maturation. Mol. Cell. Endocrinol. 189, 59-76. Féral, C., Le Gall, S., 1983a. The influence of a pollutant factor (tributyltin) on the neuroendocrine mechanism responsible for the occurrence of a penis in the females of Ocenebra erinacea. In: Lever, J., Boer, H.H. (Eds.), Molluscan Neuroendocrinology, North Holland, Amsterdam, 173-175. Féral, C., Le Gall, S., 1983b. The neuroendocrine mechanism responsible for penis differentiation in Crepidula fornicata. In: Lever, J., Boer, H.H. (Eds.), Molluscan Neuroendocrinology. North Holland, Amsterdam. 169-173. Gooding, M. P., Wilson, V. S., Folmar, L. C., Marcovich, D. T., LeBlanc, G. A., 2003 The biocide tributyltin reduces the retention of testosteroneas fatty acid esters in the mud snail (Ilyanassa obsoleta). Environ. Health Perspect. 111, 426-30. Grun, F., Watanabe, H., Zamanian, Z., Maeda, L., Arima, K., Cubacha, R., Gardiner, D. M., Kanno, J., Iguchi, T., Blumberg, B., 2006. Endocrine-disrupting organotin compounds are potent inducers of adipogenesis in vertebrates. Mol. Endocrinol. 20, 2141-2155. Horiguchi, T., Nishikawa, T., Ohta, Y., Shiraishi, H., Morita, M., 2010. Time course of expression of the retinoid X receptor gene and induction of imposex in the rock shell, Thais clavigera, exposed to triphenyltin chloride. Anal. Bioanal. Chem. 396, 597-607. Kanayama, T., Kobayashi, N., Mamiya, S., Nakanishi, T., Nishikawa, J-I., 2005. Organotin Compounds Promote Adipocyte Differentiation as Agonists of the Peroxisome Proliferator-Activated Receptor γ/Retinoid X Receptor Pathway. Mol. Pharmacol. 67, 766-774. Kastner, P., Mark, M., Leid, M., Gansmuller, A., Chin, W., Grondona, J. M., Décimo, D., Krezel, W., Dierich, A., Chambon, P., 1996. Abnormal spermatogenesis in RXR beta mutant mice. Genes Dev. 10, 80-92. Le Gall, S., Feral, C., Lengronne, C., Porchet, M., 1987. Partial purification of the neuroendocrine mitogenic factor in the mollusc Crepidula fornicata L. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 86, 393-396. Lin, Y. Q., Zhuang, H. L., Yang, G. S., 2007. The role of 9-cis RA in the regulation of pig preadipocyte differentiation. Fen Zi Xi Bao Sheng Wu Xue Bao 40, 211-6. Livak, K. J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using realtime quantitative PCR and the 2-[Delta][Delta]CT method. Methods 25, 402-408. Chapter 2 95 MacLean, G., Li, H., Metzger, D., Chambon, P., Petkovich, M., 2007. Apoptotic extinction of germ cells in testes of Cyp26b1 knockout mice. Endocrinology 148, 4560-4567. Maden, M., Corcoran, J., 1996. Role of thyroid hormone and retinoid receptors in the homeotic transformation of tails into limbs in frogs. Dev. Genet. 19, 85-93. Maden, M., Hind, M., 2003. Retinoic acid, a regeneration-inducing molecule. Dev. Dyn. 226, 237-244. Markov, G. V., Tavares, R., Dauphin-Villemant, C., Demeneix, B. A., Baker, M. E., Laudet, V., 2009. Independent elaboration of steroid hormone signaling pathways in metazoans. Proc. Natl. Acad. Sci. U.S.A. 106, 11913-11918. Marletaz, F., Holland, L. Z., Laudet, V., Schubert, M., 2006. Retinoic acid signaling and the evolution of chordates. Int. J. Biol. Sci. 2, 38-47. Nishikawa, J-I., Mamiya, S., Kanayama, T., Nishikawa, T., Shiraishi, F., Horiguchi, T., 2004. Involvement of the retinoid X receptor in the development of imposex caused by organotins in gastropods. Environ. Sci. Technol. 38, 6271-6276. Nowickyj, S. M., Chithalen, J. V., Cameron, D., Tyshenko, M. G., Petkovich, M., Wyatt, G. R., Jones, G., Walker, V. K., 2008. Locust retinoid X receptors: 9-Cis-retinoic acid in embryos from a primitive insect. Proc. Natl. Acad. Sci. U.S.A. 105, 9540-9545. Ogino, Y., Suzuki, K., HaraguchiI, R., Satoh, Y., Dolle, P., Yamada, G., 2001. External genitalia formation. Ann. N. Y. Acad. Sci. 948, 13-31. Ronis, M. J. J., Mason, A. Z., 1996. The metabolism of testosterone by the periwinkle (Littorina littorea) in vitro and in vivo: Effects of tributyl tin. Mar. Environ. Res. 42, 161-166. Santos, M. M., Ten Hallers-Tjabbes, C.C., Vieira, N, Boon, J.P., Porte, C., 2002. Cytochrome P450 differences in normal and imposex-affected female whelk Buccinum undatum from the open North Sea. Mar. Environ. Res. 54, 661-665. Santos, M. M., Castro, L. F. C., Vieira, M. N., Micael, J., Morabito, R., Massanisso, P., Reis-Henriques, M.A., 2005. New insights into the mechanism of imposex induction in the dogwhelk Nucella lapillus. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 141, 101-109. Santos, M. M., Reis-Henriques, M. M., Vieira, M. N., Solé, M., 2006. Triphenyltin and tributyltin, single and in combination, promote imposex in the gastropod Bolinus brandaris. Ecotoxicol. Environ. Saf. 64, 155-162. Simões-Costa, M. S., Azambuja, A. P., Xavier-Neto, J., 2008. The search for nonchordate retinoic acid signaling: lessons from chordates. J. Exp. Zool. B Mol. Dev. Evol. 310B, 54-72. Chapter 2 96 Spooner, N., Gibbs, P. E., Bryan, G. W., Goad, L. J., 1991. The effect of tributyltin upon steroid titres in the female dogwhelk, Nucella lapillus, and the development of imposex. Mar. Environ. Res. 32, 37-49. Sternberg, R. M., Hotchkiss, A. K., LeBlanc, G. A., 2008a. The contribution of steroidal androgens and estrogens to reproductive maturation of the eastern mud snail Ilyanassa obsoleta. Gen. Comp. Endocrinol. 156, 15-26. Sternberg, R. M., Hotchkiss, A. K., LeBlanc, G.A., 2008b. Synchronized Expression of Retinoid X Receptor mRNA with Reproductive Tract Recrudescence in an Imposex-Susceptible Mollusc. Environ. Sci. Technol. 42, 1345-1351. Sternberg, R., Gooding, M., Hotchkiss, A., LeBlanc, G., 2010. Environmental-endocrine control of reproductive maturation in gastropods: implications for the mechanism of tributyltin-induced imposex in prosobranchs. Ecotoxicology 19, 4-23. Wan, Y-J. Y., Cai, Y., Magee, T. R., 1998. Retinoic acid differentially regulates Retinoic Acid Receptor-mediated pathways in the HEP3B cell line. Exp. Cell Res. 238, 241247. Wan, Y-J. Y., Wang, L., Wu, T-C. J., 1994. The expression of retinoid X receptor genes is regulated by all-trans-and 9-cis-retinoic acid in F9 Teratocarcinoma Cells. Exp. Cell Res. 210, 56-61. Wiens, M., Batel, R., Korzhev, M., Muller, W. E. G., 2003. Retinoid X receptor and retinoic acid response in the marine sponge Suberites domuncula. J. Exp. Biol. 206, 32613271. CHAPTER 3 3. Retinoid signaling in Nucella lapillus: Adh3, Cyp26 and RAR CHAPTER 3 3.1 Molecular characterization of Adh3 from the mollusk Nucella lapillus: tissue gene expression after tributyltin and retinol exposure Lima, D.*, Coelho, I.*, André, A., Melo, C., Ruivo, R., Reis-Henriques M.A., Santos, M.M., Castro, L.F.C (accepted by the Journal of Molluscan Studies) *equally contributed 3.2 Isolation of the first protostome cytochrome P26 orthologue in the imposexsensitive gastropod Nucella lapillus: molecular and toxicological insights Daniela Lima et al. (in preparation) 3.3 Isolation and basal characterization of the Retinoic Acid Receptor (RAR) in the gastropod mollusk Nucella lapillus Daniela Lima et al., (in preparation) Chapter 3 98 Chapter 3 99 3.1 Molecular characterization of Adh3 from the mollusk Nucella lapillus: tissue gene expression after tributyltin and retinol exposure 3.1.1 Abstract Organotin interference with animal endocrine systems represents a notable case of physiological disruption. Neogastropod mollusks are particularly sensitive to the exposure to these compounds, developing a condition termed imposex, the superimposition of male sexual secondary features onto females. Recently, various studies have shown that the Retinoic X Receptor (RXR) is a high affinity ligand to tributyltin (TBT), while simultaneously hampering the receptor gene expression. Curiously, TBT has been shown to down regulate Adh3 in ascidians, an enzyme controversially linked with retinol oxidation. Here we isolated the Adh3 orthologue in Nucella lapillus, characterized its basal tissue expression profile and addressed the gene expression dynamics in gonads and digestive gland upon TBT and retinol exposure. We find that TBT does not affect Adh3 expression in N. lapillus tested tissues. However, the exposure to retinol, the precursor of retinoic acid in vertebrates, caused a significant down-regulation of Adh3 levels in female gonads. 3.1.2 Introduction The last decades have witnessed a significant and hazardous increase in the contamination of aquatic environments by endocrine disrupting chemicals (EDCs). Although the determination of the biological impact of EDCs in marine ecosystems is now an established approach for androgenic and estrogenic chemicals, a full comprehension of the effects on marine life is far from achieved (Blystone et al., 2008; Sumpter et al., 2005). However, sound and irrefutable evidence indicate that EDCs, affect the normal endocrine system of several groups of organisms, thus leading to impairment of important physiological functions. In fact, reproductive problems, carcinogenesis and other toxic effects have been described in wildlife fauna (Sumpter, 2005). Due to its widespread use in agriculture, industry and as antifouling paints for ships, boats and fishing nets, organotin compounds represent one of the most hazardous pollutants in marine environments and Chapter 3 100 are of particular concern due to their bioaccumulation ability and persistence (Fent, 1996; Santos et al., 2009). The effects of organotin compounds in reproductive and non-reproductive relatedparameters have been described in both invertebrates and vertebrates (Antizar-Ladislao, 2008). The most striking example of endocrine disruption in wildlife is the phenomenon of imposex in neogastropods characterized by the development of male secondary sexual organs in females, namely the penis and a vas deferens, due to exposure to tributyltin (TBT; Gibbs and Bryan, 1986). In advanced stages of imposex the vas deferens might block the oviduct, compromising normal breeding activity which may lead to population decline (Sternberg et al., 2010). Thus, the understanding of the molecular pathways underlying the action of these compounds is of crucial importance. It has been demonstrated that TBT is a high affinity ligand for both retinoid X receptors (RXRs) and peroxisome proliferator-activated receptor gamma (PPARγ). In mammals, PPARγ plays an important role in lipid homeostasis, promotes adipocyte differentiation and regulates adipogenesis (Kanayama et al., 2005). Recently, RXR has been suggested to mediate the imposex development in prosobranch gastropods (Nishikawa, 2006; Castro et al., 2007; Nakanishi, 2008). Indeed, TBT seems to bind both human and rock shell (Thais clavigera) RXR with the same affinity as its endogenous ligand, 9-cis retinoic acid (9-cis RA), which induces imposex in T. clavigera and Nucella lapillus, and causes ligand dependent transactivation of the human RXRα (Nishikawa et al., 2004). Hence, the retinoid signaling cascades might constitute one of the primary biological targets for organotin compounds. The impact of organotins on overall gene expression is, nonetheless, far from understood. In mollusk gastropods TBT has been shown to alter RXR tissue gene expression (Lima et al., 2011). However, downstream effectors of RXR mediated signaling have not been fully disclosed. Interestingly, in TBT-exposed ascidian Ciona intestinalis, a cDNA microarray technique has revealed a strong differential expression in over 200 genes (though without a full description of the impacted gene families; Azumi et al., 2004). Among those, the alcohol dehydrogenase class III (Adh3) gene, the suggested ancestral form of the medium-chain dehydrogenase-reductase family, was strongly down-regulated. Contrasting with vertebrates, which have several ADHs, ADH3 is the typical and unique ADH form observed in invertebrates (Godoy et al., 2007). Adh3 shows contrasting gene expression patterns in distinct animal lineages. While invertebrate ADH3 seems to present a specific expression pattern, generally in the digestive tract, vertebrate Adh3 is ubiquitously expressed (Cañestro et al., 2003). In agreement with the latter, Adh3, whose biological activity has been preserved from protostomes to deuterostomes, has been Chapter 3 101 suggested to perform housekeeping functions namely roles in nitric acid homeostasis and regulation of formaldehyde levels (Cañestro et al., 2003). However, in mammals, this basic function was challenged. Indeed, the disruption of the Adh3 gene in mice led to negative (but reversible through retinol supplementation) effects on RA metabolism and growth suggesting a role for ADH3 in RA synthesis as the sought retinol dehydrogenase, which converts retinol to retinal later supplying the retinaldehyde dehydrogenases (RALDHs; Molotkov et al., 2002). In the current work we provide the characterization of a mollusk Adh3 and addressed the gene expression dynamics, using real-time PCR, in the gastropod N. lapillus upon exposure to TBT and retinol. Although TBT was successful at inducing imposex after 2 months exposure, ADH3 expression was not significantly altered. In contrast, retinol, the precursor of RA in vertebrates, caused a significant down-regulation of ADH3 levels in female gonads. 3.1.3 Methods 3.1.3.1 Experimental exposure conditions Adult N. Lapillus were collected in February 2008 at Praia de Apúlia, Portugal, where imposex incidence among the population is below 5% (Santos et al., unpublished data). The animals were brought to laboratory and were allowed to acclimate for 1 week prior to the onset of the experiment. The experiment was carried out with about 27 animals per replicate, placed in 15L aquaria filled with artificial salt water maintained at 15 ◦C ± 1 in an acclimatized room under a photoperiod of 12 hours light: 12 hours dark and artificial airing. Salt water (salinity 35ppm, pH 8.3, conductivity = 48 ms/cm, redox potential =−76 mV) was prepared using sera premium salt and carbon activated filtrated tap water. Water was changed twice a week, and animals were fed with mussels from their origin site once a week. The following experimental conditions were tested in duplicate: All-trans retinol (sigma) 1 µg/g body weight (b.w.), TBTSn (Aldrich) 1 µg/g b.w. and control fetal bovine serum (FBS; sigma), which was used as a carrier for both retinol and TBT. The concentrations used were selected based in previous studies (Castro et al., 2007). After the acclimatization period, animal were anesthetized for 30 min in 7% MgCl2 and were injected into the foot with one of the experimental compounds, retinol, TBT and FBS. Animals were sacrificed 3 days after the beginning of the experiment and after two months (n=6/12). Shell weight was measured prior to cracking and removing of snails. Chapter 3 102 Other parameters such as sex, maturation status and penis size were evaluated under a binocular microscope. Imposex in TBT-exposed females was determined as a positive control, to validate the injection procedure. The severity of imposex (measured as Vas Deference Sequence Index-VDSI) was determined using the imposex scheme of Bettin et al. (1996). Samples were collected and stored in RNA later (Sigma) at -80ºC. The remaining animals were sacrificed after two months of exposure and the same parameters were evaluated. The mortality rate ranged from 11% (Control FBS) to 18% (TBT exposed). 3.1.3.2 Field population sampling For the characterization of basal tissue expression levels of Adh3, samples were collected from Praia da Apúlia. Specimens were collected in April 2008, taken to laboratory and analysed under a binocular stereoscope for the determination of sex and presence of imposex in females (again using the imposex scheme of Bettin et al., 1996). As expected, imposex frequency in these animals was below 5%. Organs common to both sexes (digestive gland, gonads, head ganglia complex, tentacles, kidney, and gills), male- (prostate and penis) and female-specific tissues (albumen gland, capsule gland, sperm ingesting gland and penis forming area) were removed and stored in RNAlater at -80o C until further analysis. 3.1.3.3 Tissue RNA extraction Total RNA extraction from the different studied tissues was performed using the “illustra RNAspin Mini RNA Isolation Kit”, GE Healthcare (animal tissues protocol), with on-column DNase I digestion. RNA quality was assessed in a 1% agarose gel and its concentration was determined by fluorescence (Fluoroskan Ascent, Labsystems) using the “Quant-iT RiboGreen RNA Assay Kit” (Invitrogen). First strand cDNA synthesis was performed using the iScript™ cDNA Synthesis Kit (Bio-Rad), according to the manufacturer’s instructions, using five hundred nanograms of total RNA of each sample. 3.1.3.4 ADH3 isolation and characterization We used a combination of PCR primers designed in conserved regions of the Adh3 from various species to isolate the Adh3 orthologue from N. lapillus. Initially, we used primers ADH3F 5’ GGCTACTGGRGTGTGCCA 3’ and ADH3R 5’ AGTTGAGGGCAGCWCCGT 3’. The PCR profile was as follows: 95°C 5 minutes, with 35 cycles at 94°C 30 seconds, 50°C 30 seconds and 72°C 30 seconds. A band of approximately 400bp was isolated from the agarose gel and cloned using the pGEM-T Chapter 3 103 easy vector system (Promega). Sequencing of the isolated clones was performed with the M13Foward and M13 reverse primers (Stabvida). To extend the initial sequence we prepared 5’ and 3’ RACE cDNA from a pool of various N. lapillus RNAs using the Clontech kit and following the manufacturer instructions (Takara Bio USA). Race primers were designed from the initial sequence (ADHRACER 5’ GGAATTCGATTAGTTGAGGGCAGCTC 3’ and ADHRACEF 5’ AAGACCAACCTGTGCCAGAAGATCAG 3’). However, while 3’RACE PCR was successful, 5’RACE PCR was ineffective. 3.1.3.5 Real time PCR Assays The expression of Adh3 in the various tissues was determined by real-time PCR. An initial dilution of the cDNA (1:4) was done, of which 5µL were added to a reaction mixture containing 1x iQ SYBR Green supermix (Bio-Rad) and 2 µM of forward and reverse primer in a final volume of 25µL. A “no template control” was included in each reaction 96-well plate, and all samples were run in duplicate. Quantitative PCR primers were designed using the Software Beacon Designer 5: ADHRTF 5’ GGCAGGTCAGGAAATATC 3’ and ADHRTR 5’ CTCGTTGATCTTCTCCAG 3’. The Real Time PCR profile was initiated at 95ºC for 5 min, followed by forty cycles of denaturation at 95 oC for 10s, annealing at 58 oC for 30s, and extension at 72 oC for 30s (data collection). A melting curve was generated for every run to confirm the specificity of the assays. The PCR efficiency for the gene of interest was determined through a standard curve, using six 5-fold serial dilutions (efficiency of 94.8%). Relative gene expression was calculated with using the 2-∆∆Ct formula and. Rpl8 was used as reference gene. Primer sequences as follows: NlRpl8RTF 5’ GCATCATCATCAGCCACAAC 3’ and NlRpl8RTR 5’ ACCACCACCAGCAACAATG 3’. Real time amplification efficiency was calculated using seven 5-fold serial dilutions (efficiency 95.1%) with the following PCR profile: 95°C 10 minutes, with 40 cycles 95°C 10 s, 60°C 30 seconds and 72°C 30 seconds. 3.1.3.6 Statistical Analysis Statistical analysis of the results was performed using the software Statistica 7.0. Imposex frequency was tested by a Chi-Square test using the results of the solvent control treatment as the expected results. After testing for ANOVA assumptions (homogeneity of variances and normality of data), statistical differences in Adh3 gene expression in the three treatment groups were evaluated through a one-way ANOVA, Chapter 3 110 causing imposex (Nishikawa et al., 2004; Castro et al., 2007). Significantly, reproductive recrudescence in gastropods has also been suggested to be mediated by RXR and impacted by TBT (Sternberg et al., 2008). Thus, the disruption of the retinoid acid pathway in mollusks is crucial for the development of imposex. Until recently, the molecular components of the retinoic acid signaling pathway were thought to be chordate innovations (Albalat and Cañestro, 2009). However, representatives of important components of this pathway such as the retinoic acid receptor (RAR) or the retinoic acid degradating enzyme CYP26 have now been found in pre-chordate phyla (Albalat and Cañestro, 2009). In vertebrates the pathway leading to production of active retinoids (9-cis RA and all-trans RA) involves the initial oxidation of retinol to retinaldehyde. This enzymatic step is performed by SDR-Rdh gene family members and Adh class genes. In invertebrates though, the role of ADH3 in retinol oxidation has been questioned (see Cañestro et al., 2010). In contrast to vertebrates, the single copy invertebrate Adh3 is largely expressed in digestive tissues and not ubiquitously as would be expected if a role in retinol metabolism was present (Godoy et al., 2007). In fact, functionally the enzyme has been shown to participate in formaldehyde and nitric oxide metabolism (GonzàlezDuarte and Albalat, 2005). In this study we set out to isolate and characterize Adh3 in the TBT-sensitive gastropod mollusk N. lapillus. Since TBT modulates RXR (a crucial component of the retinoic acid signaling pathway) and Adh3 has been shown to be down regulated by TBT in ascidians (Azumi et al., 2004), we hypothesized whether a similar finding might be observed in N. lapillus. However, no such effect was detected in the tested tissues. Also, as part of an ongoing study aimed at dissecting the retinoid molecular pathway in N. lapillus, we analyzed the impact of retinol exposure in the gene expression of Adh3. Unexpectedly, Adh3 was clearly downregulated in female gonads (a trend in testis is also observed but did not reach statistical significance). Since retinoic acid has been suggested to play a role in gonad recrudescence in gastropods (Sternberg et al., 2008), the reported finding could hypothetically indicate a participation of ADH3 in retinol oxidation. Alternatively, a toxic effect of retinol in mollusk physiology could have led to the down play of Adh3 expression. As research on this topic progresses, a crucial question relates to the role and gene portfolio of the SDR-Rdh family in invertebrates namely mollusks. In conclusion, we provide here the molecular characterization of Adh3 in N. lapillus. Our data indicates that molluskan Adh3 gene expression in the gonads and digestive gland is not affected by TBT exposure. In contrast, we show that retinol down regulates Adh3 in female gonads. Whether this results from a direct role of ADH3 in retinoid acid metabolism should be investigated in the future. Chapter 3 111 3.1.6 Acknowledgements We acknowledge the financial support of FCT (Portugal) Project PTDC/MAR/68106/2006. Daniela Lima is a recipient of an FCT fellowship (SFRH/BD/41561/2007). 3.1.7 References Albalat, R., Cañestro, C., 2009. Identification of Aldh1a, Cyp26 and RAR orthologs in protostomes pushes back the retinoic acid genetic machinery in evolutionary time to the bilaterian ancestor. Chem. Biol. Interact. 178, 188-96. Antizar-Ladislao, B., 2008. Environmental levels, toxicity and human exposure to tributyltin (TBT)-contaminated marine environment. A review. Environ. Int. 34, 292-308. Azumi, K., Fujie, M., Usami, T., Miki, Y., Satoh, N., 2004. A cDNA microarray technique applied for analysis of global gene expression profiles in tributyltin-exposed ascidians. Marine Environ. Res. 58, 543-546. Bettin, C., Oehlmann, J., Stroben, E., 1996. TBT-induced imposex in marine neogastropods is mediated by an increasing androgen level. Helgolander Meeresun. 50, 299-317. Blystone, C. R., Wilson, V. S., Hartig, P. C., Ankley, G. T., Foster, P. M., Gray, C. l., Gray, L. E., 2008. Fifteen years after "Wingspread" environmental endocrine disrupters and human and wildlife health: where we are today and where we need to go. Toxicol. Sci. 105, 235-59. Cañestro, C., Godoy, l., Gonzàlez-Duarte, R., Albalat, R., 2003. Comparative expression analysis of Adh3 during arthropod, urochordate, cephalochordate, and vertebrate development challenges its predicted housekeeping role. Evol. Dev. 5, 157-62. Cañestro, C., Albalat, R., Postlethwait, J.H., 2010. Oikopleura dioica alcohol dehydrogenase class 3 provides new insights into the evolution of retinoic acid synthesis in chordates. Zool. Sci. 27, 128-33. Castro, L. F., Lima, D., Machado, A., Melo, C., Hiromori, Y., Nishikawa, J., Nakanishi, T., Reis-Henriques, M. A., Santos, M. M., 2007. Imposex induction is mediated through the Retinoid X Receptor signaling pathway in the neogastropod Nucella lapillus. Aquat. Toxicol. 85, 57-66. Chapter 3 112 Créton, R., Zwaan, G., Dohmen, R., 1993. Specific developmental defects in molluscs after treatment with retinoic acid during gastrulation. Dev. Growth Differ. 35, 357364. deFur, P. L., Crane, M., Ingersoll, C., Tattersfield, l. (eds) 1999. Endocrine disruption in invertebrates: endocrinology, testing, and assessment. In: Proceedings of the workshops on endocrine disruption in invertebrates, Noordwijkerhout, SETAC Press, Pensacola, The Netherlands. Fent, K., 1996. Ecotoxicology of organotin compounds. Crit. Rev. Toxicol. 26, 1-117. Gibbs, P. E., Bryan, G. W., 1986. Reproductive failure in populations of the dog-whelk, Nucella lapillus, caused by imposex induced by tributyltin from antifouling paints. J. Mar. Biol. Assoc. U.K. 66, 767-777. Godoy, l., Gonzàlez-Duarte, R., Albalat, R., 2007 Analysis of planarian Adh3 supports an intron-rich architecture and tissue-specific expression for the urbilaterian ancestral form. Comp. Biochem. Physiol. B 146, 489-95. Gonzàlez-Duarte, R., Albalat, R., 2005. Merging protein, gene and genomic data: the evolution of the MDR-ADH family. Heredity 95, 184-97. Kanayama, T., Kobayashi, N., Mamiya, S., Nakanishi, T., Nishikawa, J., 2005. Organotin compounds promote adipocyte differentiation as agonists of the peroxisome proliferator-activated receptor gamma/retinoid X receptor pathway. Mol. Pharmacol. 67, 766-74. Lima, D., Reis-Henriques, M. A., Silva, R., Santos, A. I., Castro, L. F. C., Santos, M. M., 2011. Tributyltin-induced imposex in marine gastropods involves tissue-specific modulation of the retinoid X receptor. Aquat. Toxicol. 101, 221-7. Molotkov, A., Fan, X. , Deltour, l., Foglio, M. H., Martras, S., Farrés, J., Pares, X., Duester, G., 2002. Stimulation of retinoic acid production and growth by ubiquitously expressed alcohol dehydrogenase Adh3. Proc. Natl. Acad. Sci. U. S. A. 99, 5337-42. Nakanishi, T., 2008. Endocrine disruption induced by organotin compounds; organotins function as a powerful agonist for nuclear receptors rather than an aromatase inhibitor. J. Toxicol. Sci. 33, 269-76. Nishikawa, J., Mamiya, S., Kanayama, T., Nishikawa, T., Shiraishi, F., Horiguchi, T., 2004. Involvement of the retinoid X receptor in the development of imposex caused by organotins in gastropods. Environ. Sci. Technol. 38, 6271-6. Nishikawa, J., 2006. Imposex in marine gastropods may be caused by binding of organotins to retinoid X receptor. Mar. Biol. 149, 117-124. Chapter 3 113 Santos, M. M., Enes, P., Reis-Henriques, M. A., Kuballa, J., Castro, L. F. C., Vieira, M. N., 2009. Organotin levels in seafood from Portuguese markets and the risk for consumers. Chemosphere 75, 661-666. Sternberg, R. M., Hotchkiss, A. K., Leblanc, G. A., 2008. Synchronized expression of retinoid X receptor mRNA with reproductive tract recrudescence in an imposexsusceptible mollusc. Environ. Sci. Technol. 42, 1345-51. Sternberg, R. M., Gooding, M. P., Hotchkiss, A. K., Leblanc, G.A., 2010. Environmentalendocrine control of reproductive maturation in gastropods: implications for the mechanism of tributyltin-induced imposex in prosobranchs. Ecotoxicology 19, 4-23. Sumpter, J. P., 2005. Endocrine disrupters in the aquatic environment: an overview. Acta Hydrochim. Hydrobiol. 33, 9-16. Zgombić-Knight, M., Ang, H. L., Foglio, M. H., Duester, G., 1995. Cloning of the mouse class IV alcohol dehydrogenase (retinol dehydrogenase) cDNA and tissue-specific expression patterns of the murine ADH gene family. J. Biol. Chem. 270, 10868-77. Chapter 3 114 Chapter 3 115 3.2 Isolation of the first protostome cytochrome P26 orthologue in the imposex-sensitive gastropod Nucella lapillus: molecular and toxicological insights 3.2.1 Abstract We report here the isolation of the first protostome cytochrome P26 gene in the lophotrochozoan gastropod Nucella lapillus. The phylogenetic analysis confirms that the isolated sequence is part of CYP26 evolutionary clade, a clear indication on the bilaterian ancestry of the RA signaling pathway molecular components. Given the recently reported pivot role of mammal CYP26 in the control of meiosis initiation, we provide here the first protostome CYP26 tissue expression profile by qPCR, in several female and male tissues, as well as tissue expression dynamics in gonads between mature and dormant animals. Additionally, considering that the endocrine disruptor chemical tributyltin (TBT) has been reported to induce imposex in female gastropods through the modulation of retinoid signaling pathways, the interference of TBT with the transcription levels of CYP26 in a set of N. lapillus tissues was evaluated. The results of the present study show that, similar to vertebrates, NlCYP26 is ubiquitously expressed in a broad range of tissues. Interestingly, the expression dynamic of CYP26 in female gonads follows an opposite pattern to that of RXR, with a decrease transcription in mature and increased transcription in dormant animals. These results suggest the coordinated action of NlRXR and NlCYP26 in female gonad development, and that NlCYP26 could be involved in the degradation of endogenous RA. Exposure to TBT significantly decreased the transcription levels in penis of females at advanced stages of imposex, which may suggest a novel role of NlCYP26 in imposex development. Overall, these results extend the role of retinoid pathways in gonad development and imposex induction in gastropods, supporting the hypothesis of a phylogenetic conserved nature of retinoid signaling within Bilateria. 3.2.2 Introduction Retinoids (vitamin A and its analogs) have a key role in several biological functions in vertebrates such as regulation of development, immune function, vision or reproduction (Malik, 2000; Blomhoff and Blomhoff, 2006; Theodosiou et al., 2010). Alterations in retinoid signaling pathways cause congenital malformations, fertility problems, vision Chapter 3 116 defects, tumor and neurodegenerative disorders (Novák et al., 2008). Retinoids are taken up from the diet as carotenoids, retinol or retinyl-esters. Whereas vertebrates have an elaborate system involving retinoid uptake, transport, storage, mobilization, activation and catabolism, much less is known in invertebrates (Theodosiou et al., 2010). Yet, the presence of bioactive retinoids has been reported in a few invertebrate species and the presence of retinoid metabolism and storage in the gastropod Osilinus lineatus has recently been demonstrated (Gesto et al., 2012). Retinoid signal transduction occurs essentially through two types of nuclear receptors: the retinoid X receptors (RXRs) and the retinoic acid receptors (RARs). Whereas RXR is highly conserved throughout metazoans, RAR was recently identified in a lophotrochozoan genome, although a functional characterization is still lacking (Albalat and Cañestro, 2009). The active metabolite of vitamin A, retinoic acid (all-trans RA), is a key regulator of many biological processes such as embryonic development and reproduction in mammals (Niederreither and Dollé, 2008). However, the homeostatic control of RA levels within a physiological range is essential as both depressed and excess RA levels have been associated in vertebrates with several pathologies (Niederreither and Dollé, 2008). The maintenance of tissue’s RA levels is achieved mostly by the action of enzymes from the cytochrome p450 superfamily, the CYP26 family. In mammals, the CYP26 family has three members, CYP26A1, CYP26B1 and CYP26C1 (Ross and Zolfaghari, 2011). Although the CYP26 genes display different expression profile, they are able to catalyze the oxidation of all-trans RA and 9-cis RA (CYP26C1) to less active metabolites, i.e., 4hydroxy and 4 –oxo and 18 – hydroxyl-RA. While the biological activity of these metabolites is still not fully characterized, the available data suggest that their activity may be context specific (Ross and Zolfaghari, 2011). In mice, null mutations of genes for CYP26C1 and CYP26B1 are lethal, thus showing the central role during embryonic development (Niederreither and Dollé, 2008). In germ cell development, the sex-specific time of meiosis entry is under the control of RA (Bowles et al., 2006; Koubova et al., 2006). During embryogenesis, whereas germ cells in the ovary enter meiosis, in the testis germ cells do not enter meiosis until puberty. CYP26 enzymes have been suggested to have a central role in this mechanism, as they seem to prevent an increase in RA levels in testis, thus delaying the entry into meiosis. Similarly, CYP26 seems to be determinant for ovary meiosis (Bowles et al., 2006; Koubova et al., 2006). Compiling evidences indicate that mollusks have a fully elaborated retinoid system (Gesto et al., 2012). This is supported by the presence of active retinoid metabolism and the presence of biological effects of retinoids in several species (Nowickyj et al., 2008). In Lymnaea stagnalis RA has been suggested to be involved in neuronal regeneration and Chapter 3 117 axon pathfinding and embryo exposure to retinol, retinal and all-trans RA (in the range of 10-5 to 10-7M) leads to developmental defects characterized by increased rates of embryos with arrested development and abnormal shell and eyes (Créton et al., 1993; Dmetrichuk et al., 2008). A link between the dynamics of RXR transcription in female and male gonads of the mud snail Ilyanassa obsoleta and the maturation status has been established by Sternberg et al. (2008), suggesting a role of RXR in reproduction of mollusks. Similarly, Gesto et al. (2012) have recently reported a sex-specific pattern of retinoid storage and metabolism is gonad/digestive gland of the gastropod O. lineatus. Furthermore, administration of 9-cis RA leads to an increase in male penis length in the prosobranch gastropod N. lapillus as well as penis development in females (imposex), thus implying that, similar to vertebrates, RA may be involved in mollusks genitalia formation (Castro et al., 2007). In fact, the development of male penis in female prosobranch gastropods has been the best characterized example of endocrine disruption in aquatic ecosystems. Today, it is well established that imposex induction by the antifoulant tributyltin is mediated through RXR signaling pathways (Castro et al., 2007; Sternberg et al., 2010). Hence, present data suggest that the control of retinoid levels within a physiological range may be crucial to both vertebrates and invertebrates. Recently, the search in the unpublished Lophotrochozoans genomes (e.g., mollusks and annelids) has hinted at the presence of key players of the RA machinery in protostomes, including a CYP26 orthologue (Albalat and Cañestro, 2009). Hence, considering the central role of CYP26 in vertebrates, in the present study we isolated the first protostome CYP26 orthologue in the prosobranch gastropod N. lapillus and provide a tissue expression profile by qPCR in several female and male tissues. Furthermore, we investigated the tissue expression dynamics in gonads at mature and dormant stages in both sexes. Additionally, since the antifoulant agent TBT has been shown to alter retinoid signaling in gastropods, a toxicological study was performed to evaluate the potential interference of TBT with the transcription levels of CYP26 in a set of N. lapillus tissues. 3.2.3 Material and Methods 3.2.3.1 Field population sampling For the characterization of basal tissue expression levels, adult N. lapillus were collected at Praia da Apúlia, a site located in the North of Portugal known to display an imposex frequency below 5% (Santos et al., unpublished data). Specimens were collected at a single time point, taken to laboratory and analyzed under a binocular stereoscope for Chapter 3 118 the determination of sex and presence of imposex in females (using the imposex scheme of Bettin et al., 1996). As expected, imposex frequency in these animals was below 5%. Organs common to both sexes (digestive gland, gonads, head ganglia complex, tentacles, kidney, and gills), male- (prostate and penis) and female-specific tissues (albumen gland, capsule gland, sperm ingesting gland and penis forming area) were removed and stored in RNAlater at -80 oC until further analysis. For the characterization of mature and dormant CYP26 gene transcription levels, adults were sampled between June and July 2008 at Praia da Apúlia. Gene transcription was determined in male and female gonads (4-6 samples per sex and reproductive status). The maturation status of the animals was determined based on external morphological characteristics. Mature females displayed a very developed gonad and capsule gland, while dormant presented these organs underdeveloped and with an orange/brownish color. Males were classified as mature when displayed very developed gonad and sperm was present in the seminal vesicle, and as dormant in the absence of visible sperm and gonad not developed. RNA extraction and cDNA synthesis were performed as described above. Since RXR gene transcription in gonads of the mud snail, Ilyanassa obsoleta (Sternberg et al., 2008) has been suggested to vary according to the maturation stage, we also determined RXR gene transcription in gonads of mature and dormant animals, and contrasted the gene transcription dynamic with that of CYP26. 3.2.3.2 Experimental exposure to TBT The experimental conditions are similar to those described in detail in Lima et al. (2011). Briefly, adult N. lapillus were collected in Praia de Apúlia and brought to laboratory, where they were allowed to acclimate for 1 week prior to the onset of the experiment. About 35 specimens per replicate (three replicate per treatment) were placed in 30 L aquaria filled with artificial salt water and two treatment groups were set: solvent control (DMSO at 0.0002%) and TBT Cl at 100 ng Sn/L TBT. After one and two months of exposure, 16 animals (eight males and eight females) per treatment were sacrificed. Sex, maturation status and penis size/imposex were evaluated under a binocular microscope. The severity of imposex (measured as Vas Deference Sequence Index-VDSI) was determined using the imposex scheme of Bettin et al. (1996). No differences in mortality rates were observed between treatment groups. 3.2.3.3 Tissue RNA extraction RNA extraction in gonads, digestive gland and CNS was performed using the Kit illustra RNAspin Mini RNA Isolation, GE Healthcare (animal tissues protocol), with on Chapter 3 119 column DNAseI digestion. Given the low amount of tissue and RNA concentration in penis and PFA, RNA was extracted using the kit Qiagen RNeasy® Plus Micro, which is adequate for samples with less than 5 mg of tissue. The cDNA synthesis was performed with the iScript™cDNA Synthesis Kit (Bio-Rad) according to the manufacturer’s instructions, using 500 ng of total RNA for gonad and CNS and 50 ng of total RNA for penis and PFA. 3.2.3.4 NlCyp26 isolation and characterization We designed various sets of degenerate primers to isolate Cyp26 sequences from N. lapillus through RT-PCR. Initially, we were able to isolate a fragment with approximately 120 nucleotides with similarity to Cyp26 sequences with primers CYP26F3 (5’ TGGGCTGGCCCTTCdtnggngarac 3’) and CYP26R3 (5’gctgccgaacaggtgngtywtrwa 3’). The retrieved sequence was further expanded using a specific forward (5’ TGGAGTTTGCTAGAAAGGGGGCAGA 3’) and a reverse degenerated primer (5’ CGCACGCCGCcnccraanggna 3’) located downstream of the first fragment. This strategy allowed the isolation of a contig overlapping sequence with 1079 nucleotides. We made various attempts to clone the full ORF using RACE cDNA and PCR, but these were unsuccessful. PCR bands were direct sequenced after DNA band isolation (GE illustra GFX). 3.2.3.5 Sequence analysis and phylogenetics The retrieved cDNA sequence was assembled and the similarity to other known Cyp26 determined with the BLAST program (http://www.ncbi.nlm.nih.gov/blast). The predicted amino acid sequence was produced using the translate tool of the EXPASy Bioinformatics resource portal (http://web.expasy.org/translate/). For the calculation of the phylogenetic tree sequences were collected from Ensembl, GenBank and JGI databases. Multiple sequence alignment was implemented in MEGA5 (Tamura et al., 2011) with the ClustalW function. The evolutionary history was inferred using the Maximum-likelihood method. Statistical support was determined with the bootstrap test (1000 replicates). 3.2.3.5 qPCR assays Tissue expression levels of NlCyp26 were determined by qPCR (Bio-Rad, iQ5). Briefly, 0.8 µL cDNA was added to a reaction mixture containing 1x iQ SYBR Green supermix (Bio-Rad) 0.2 µM of each primer, in a final volume of 20 µL. In each plate, a “no template control” was included, and samples were run in duplicate. qPCR profile and primers used were previously described (see Castro et al., 2007). Primers sequences and Chapter 6 222 compounds are potent inducers of adipogenesis in vertebrates. Mol. Endocrinol. 20, 2141-2155. Horiguchi, T., Nishikawa, T., Ohta, Y., Shiraishi, H., Morita, M., 2010. Time course of expression of the retinoid X receptor gene and induction of imposex in the rock shell, Thais clavigera, exposed to triphenyltin chloride. Anal. Bioanal. Chem. 396, 597-607. Janer, G., Lyssimachou, A., Bachmann, J., Oehlmann, J., Schulte-Oehlmann, U., Porte, C., 2006. Sexual dimorphism in esterified steroid levels in the gastropod Marisa cornuarietis: the effect of xenoandrogenic compounds. Steroids. 71, 435-444. Janer, G., Porte, C., 2007. Sex steroids and potential mechanisms of non-genomic endocrine disruption in invertebrates. Ecotoxicology. 16, 145-60. Janer, G., Navarro, J. C., Porte, C., 2007. Exposure to TBT increases accumulation of lipids and alters fatty acid homeostasis in the ramshorn snail Marisa cornuarietis. Comp. Biochem. Phys. C. 146, 368-74. le Maire, A., Grimaldi, M., Roecklin, D., Dagnino, S., Vivat-Hannah, V., Balaguer, P., Bourguet, W., 2009. Activation of RXR–PPAR heterodimers by organotin environmental endocrine disruptors EMBO reports 10, 367-373. Lovekamp-Swan, T., Jetten, A. M., Davis, B. J., 2003. Dual activation of PPARalpha and PPARgamma by mono-(2-ethylhexyl) phthalate in rat ovarian granulosa cells. Mol. Cell. Endocrinol. 201, 133-141. Lubzens, E., Young, G., Bobe, J., Cerdà, J., 2010. Oogenesis in teleosts: How fish eggs are formed Gen. Comp. Endocrinol. 165, 367-389. Luu-The, V., Tremblay, P., Labrie, F., 2006. Characterization of type 12 17betahydroxysteroid dehydrogenase, an isoform of type 3 17beta-hydroxysteroid dehydrogenase responsible for estradiol formation in women. Mol. Endocrinol. 20, 437-43. Lyssimachou, A., Jenssen, B. M., Arukwe, A., 2006. Brain cytochrome P450 aromatase gene isoforms and activity levels in Atlantic salmon after waterborne exposure to nominal environmental concentrations of the pharmaceutical ethynylestradiol and antifoulant tributyltin. Toxicol. Sci. 91, 82-92. Lyssimachou, A., 2008. An investigation on different mechanisms of endocrine disruption in fish and molluscs. Chapter 6 223 Lyssimachou, A., Navarro, J. C., Bachmann, J., Porte, C., 2009. Triphenyltin alters lipid homeostasis in females of the ramshorn snail Marisa cornuarietis. Environ. Pollut. 157, 1714-20. Markov, G. V., Tavares, R., Dauphin-Villemant, C., Demeneix, B. A., Baker, M. E., Laudet, V., 2009. Independent elaboration of steroid hormone signaling pathways in metazoans. P. Natl. Acad. Sci. USA. 106, 11913–11918. McAllister, B. G., Kime, D. E., 2003. Early life exposure to environmental levels of the aromatase inhibitor tributyltin causes masculinisation and irreversible sperm damage in zebrafish (Danio rerio). Aquat. Toxicol. 65, 309–316. McGinnis, C. L., Crivello, J. F., 2011. Elucidating the mechanism of action of tributyltin (TBT) in zebrafish. Aquat. toxicol. 103, 25-31. Nagasaki, S., Suzuki, T., Miki, Y., Akahira, J., Kitada, K., Ishida, T., Handa, H., Ohuchi, N., Sasano, H, 2009. 17Beta-hydroxysteroid dehydrogenase type 12 in human breast carcinoma: a prognostic factor via potential regulation of fatty acid synthesis. Cancer Res. 69, 1392-9. Nakayama, K., Oshima, Y., Yamaguchi, T., Tsuruda, Y., Kang, I. J., Kobayashi, M., Imada, N., Honjo, T., 2004. Fertilization success and sexual behavior in male medaka, Oryzias latipes, exposed to tributyltin. Chemosphere. 55, 1331-1337. Nishikawa, J.-I., Mamiya, S., Kanayama, T., Nishikawa, T., Shiraishi, F., Horiguchi, T., 2004. Involvement of the retinoid X receptor in the development of imposex caused by organotins in gastropods. Environ. Sci. Technol. 38, 6271-6276. Oberdörster, E., McClellan-Green, P., 2000. The neuropeptide APGWamide induces imposex in the mud snail, Ilyanassa obsoleta. Peptides 21, 1323–1330. Ogino, Y., Suzuki, K., Haraguchi, R., Satoh, Y., Dolle, P., Yamada, G., 2001. External genitalia formation: role of fibroblast growth factor, retinoic acid signaling, and distal urethral epithelium. Ann. N. Y. Acad. Sci., 13-31. Pavlikova, N., Kortner, T. M., Arukwe, A., 2010. Modulation of acute steroidogenesis, peroxisome proliferator-activated receptors and CYP3A/PXR in salmon interrenal tissues by tributyltin and the second messenger activator, forskolin. Chem-Biol Interact. 185, 119-127. Prehn, C., Möller, G., Adamski, J., 2009. Recent advances in 17beta-hydroxysteroid dehydrogenases, J. Steroid. Biochem. Mol. Biol. 114, 72-77. Samarut, E., Rochette-Egly, C., 2011. Nuclear retinoic acid receptors: Conductors of the retinoic acid symphony during development. Mol. Cell. Endocrinol. 1-13. Chapter 6 224 Sandell, L. L., Lynn, M. L., Inman, K. E., McDowell, W., Trainor, P. A., 2012. RDH10 oxidation of vitamin A is a critical control step in synthesis of retinoic acid during mouse embryogenesis. PloS one 7, e30698. doi:10.1371/journal.pone.0030698. Santos, M. M., Castro, L. F. C., Vieira, M. N., Micael, J., Morabito, R., Massanisso, P., Reis-Henriques, M. A., 2005. New insights into the mechanism of imposex induction in the dogwhelk Nucella lapillus. Comp. Biochem. Physiol. C. 141, 101109. Santos, M. M., Micael, J., Carvalho, A. P., Morabito, R., Booy, P., Massanisso, P., Lamoree, M., Reis-Henriques, M. A., 2006. Estrogens counteract the masculinizing effect of tributyltin in zebrafish. Comp. Biochem. Phys. C. 142, 151-155. Simões-Costa, M., Azambuja, A., Xavier-Neto, J., 2008. The search for non‐chordate retinoic acid signaling: lessons from chordates. J. Exp. Zool. 310B, 54-72. Spooner, N., Gibbs, P. E., Bryan, G. W., Goad, L. J., 1991. The effect of tributyltin upon steroid titers in the female dogwhelk, Nucella lapillus, and the development of imposex. Mar. Environ. Res. 2, 37-49. Sternberg, R.M., Hotchkiss, A.K., LeBlanc, G.A., 2008. The contribution of steroidal androgens and estrogens to reproductive maturation of the eastern mud snail Ilyanassa obsoleta. Gen. Comp. Endocrinol. 156, 15-26. Thatcher, J. E., Isoherranen, N., 2009. The role of CYP26 enzymes in retinoic acid clearance. Expert. Opinion Drug. Metab. Toxico., 1-21. Theodosiou, M., Laudet, V., Schubert, M., 2010. From carrot to clinic: an overview of the retinoic acid signaling pathway. Cell. Mol. Life Sci. 67, 1423–1445. Thornton, J. W., 2003. Nonmammalian nuclear receptors: Evolution and endocrine disruption. Pure Appl. Chem. 75, 1827-1839. Waxman, J. S., Yelon, D., 2011 Zebrafish retinoic acid receptors function as contextdependent transcriptional activators. Developmental biology 352, 128-40.