Introduction

Imposex is an irreversible pseudohermaphroditic condition in which male genital organs, such as the penis and vas deferens, develop in female gastropods (Bryan et al. 1986; Smith 1971). Imposex is typically induced by very low concentrations (~1 ng/L) of tributyltin (TBT), triphenyltin (TPT), or both, which have been used in antifouling paints for ships and fishing nets since the mid-1960s (Bryan et al. 1986, 1987, 1988; Gibbs et al. 1987; Horiguchi et al. 1994, 1997a). Reproductive failure occurs in the severe stages of imposex, either because of oviduct blockage by the formation of vasa deferentia or because of ovarian spermatogenesis, and eventually results in population decline and/or mass extinction (Gibbs and Bryan 1986; Gibbs et al. 1988, 1990; Horiguchi et al. 2006). Globally, more than 150 species of mesogastropods and neogastropods, including the rock shell Thais clavigera, are affected by imposex (Fioroni et al. 1991; Horiguchi et al. 1997b, 2006; Matthiessen et al. 1999). Imposex among gastropods has been reported to be a clear manifestation of endocrine disruption (Matthiessen and Gibbs 1998; Matthiessen et al. 1999).

Four hypotheses regarding the mechanisms by which organotins induce imposex in gastropods have been proposed: (1) an increase in androgen (e.g., testosterone) levels as a result of TBT-mediated inhibition of aromatase (Bettin et al. 1996); (2) TBT-mediated inhibition of the excretion of androgen sulfate conjugates (Ronis and Mason 1996); (3) TBT interference in the release of penis morphogenetic/retrogressive factor from the pedal/cerebropleural ganglia (Féral and Le Gall 1983); and (4) an increase in the level of alanine-proline-glycine-tryptophan amide neuropeptide in response to TBT (Oberdörster and McClellan-Green 2000). However, the scientific debate continues on the mechanisms by which organotins induce imposex in gastropods because, as noted by Horiguchi et al. (2008) and Horiguchi (2009), each of these hypotheses is insufficient to explain the phenomenon.

Nishikawa et al. (2004) found that T. clavigera has a retinoid X receptor (RXR) that is similar to those in humans and other vertebrates as well as to those in other invertebrates, such as ascidians, insects, pulmonates, jellyfish, and sponges (Bouton et al. 2005; Devine et al. 2002; Freebern et al. 1999; Heyman et al. 1992; Kamimura et al. 2000; Kostrouch et al. 1998; Mangelsdorf and Evans 1995; Mangelsdorf et al. 1992; Nagatomo et al. 2003; Wiens et al. 2003). Nishikawa et al. (2004) also observed that T. clavigera RXR binds to both 9-cis-retinoic acid (9CRA), which is known to be a natural ligand for human RXRs, and organotins. In addition, they demonstrated that a single in vivo injection of 9CRA into female rock shells without morphological signs of imposex induced the development of imposex a month later. Horiguchi et al. (2007) investigated RXR gene expression and measured the RXR protein content in various tissues of wild male and female T. clavigera using quantitative real-time polymerase chain reaction (PCR), Western blotting, and immunohistochemistry with a commercial antibody against human RXR α. Based on the results of this study, Horiguchi et al. (2007) suggested that RXR could be involved in organotin-mediated induction of male-type genitalia (penis and vas deferens) in female rock shells.

In the present study, in the chain of study to further elucidate the role of RXR in the development of imposex caused by organotin compounds in gastropods, we established a polyclonal antibody against RXR of T. clavigera and confirmed that this antibody could detect T. clavigera RXR by immunoblotting. Immunohistochemical staining was applied to examine the localization of RXR protein in tissues of the rock shell, using this antibody.

Materials and methods

Production of a polyclonal antibody against RXR of the rock shell

Full-length T. clavigera RXR (AY704160) was synthesized as a recombinant protein with a tag of glutathione S-transferase (GST) in Escherichia coli (BL21(DE3)LysS) and was injected into two rabbits (SPF Japanese white rabbit). The antibody titer was measured by ELISA 2 months after injection to confirm that the titer was sufficiently increasing. Antibody was affinity purified from the antiserum, using GST-RXR affinity column.

Detection of RXR protein using immunoblotting

Construction of plasmid vectors

The full-coding regions of T. clavigera RXR and human RXR α (NM_002957) were amplified by PCR with KOD Plus DNA polymerase (Toyobo Biochemicals, Osaka, Japan). The PCR product was gel-purified and subcloned into pcDNA3.1(+) vector (Invitrogen, Carlsbad, CA, USA).

Transient expression and immunoblotting assay

To examine the characteristics of anti-T. clavigera RXR antibody, COS-1 cells were seeded in 6-well plates at 5 × 105 cells per well in phenol-red-free DMEM (Sigma–Aldrich, St. Louis, MO, USA) supplemented with 10% charcoal/dextran-treated fetal bovine serum (Hyclone, South Logan, UT, USA). After 24 h, the cells were transfected with 2 μg pcDNA3.1(+) vector fused to T. clavigera RXR or human RXR, and empty pcDNA3.1(+) vector using Fugene 6 transfection reagent (Roche Diagnostics, Basel, Switzerland), according to the manufacturer’s instructions. After 40 h, the cells were washed twice with phosphate-buffered saline (8.1 m mol/L Na2HPO4, 1.47 m mol/L KH2PO4, 2.68 m mol/L KCl, 137 m mol/L NaCl), harvested, lysed in SDS sample buffer, and then analyzed on 10% Laemmli SDS-polyacrylamide gel (Laemmli 1970). The cell lysates were sonicated and then incubated at 70°C for 10 min. Equal volumes of cell lysates were subjected to 10% SDS–PAGE and transferred to Immobilon membranes (Millipore, Bedford, MA, USA) by electroblotting (Towbin et al. 1979). The membrane was blocked with 5% nonfat dry milk in Tris-buffered saline (TBS: 20 m mol/L Tris–HCl, 150 m mol/L NaCl, pH 7.5) and incubated with each primary antibody diluted in TBS (anti-T. clavigera RXR, 1 μg/mL; anti-human RXR Δ N197 [Santa Cruz Biotechnology, Santa Cruz, CA, USA], 1:200; anti-β-actin [clone AC-15: Sigma–Aldrich), 1:1000]. After washing (5 min × 3) with TBS containing 0.1% Tween 20 (TTBS), membranes were incubated with an alkaline phosphatase-conjugated secondary antibody (anti-mouse IgG antibody, 1:1000; anti-rabbit IgG antibody, 1:1000; both from Invitrogen). Following another three washes with TTBS, phosphatase activity was visualized by treating the membrane with 0.2 mM 5-bromo-4-chloro-3-indolylphosphate p-toluidine salt and nitro blue tetrazolium (Sigma–Aldrich) in 100 m mol/L diethanolamine buffer (pH 9.5) containing 5 m mol/L MgCl2. All incubations were performed at room temperature.

Immunohistochemical staining for RXR

In April 2008 both male and female wild rock shells (approximately 300 individuals) were collected at Aikawa on Sado Island in Niigata Prefecture, Japan (38o 01′ 39.75′′ N, 138o 14′ 18.48′′ E; a reference site known to have less than 11 and 8 ng/g wet wt. TBT and TPT, respectively; see Horiguchi et al. 1994). In May 2009 wild imposex-exhibiting female rock shells (approximately 30 individuals) were collected at Hinase in Okayama Prefecture, Japan (34o 43′ 46.54′ N, 134o 16′ 06.01′′ E; a site severely contaminated with TBT and TPT; e.g., tissue concentrations of TBT and TPT in rock shells were 116.4 and 174.2 ng/g wet wt., respectively; see Horiguchi 2004). After collection, the rock shells were immediately dissected and then preserved by immersion in 10% phosphate-buffered formalin at pH 7.4 for 48 h.

The tissues were dehydrated, embedded in paraffin, and sectioned to 4 μm. To enhance their antigenicity, deparaffinized and rehydrated sections were immersed in 0.8 M urea and autoclaved at 121°C for 15 min. Sections were then treated with 0.3% hydrogen peroxide in methanol for 30 min at room temperature and blocked in serum-free blocking solution (DakoCytomation, Dako, Glostrup, Denmark) for 30 min at room temperature. After incubation with anti-T. clavigera RXR antibody (dilution 1:1000) for 16 h at 4°C, sections were visualized with a labeled avidin–biotin system, using conjugated horseradish peroxidase (LSAB kit, Dako) and 0.01% 3,3-diaminobenzidine tetrahydrochloride (Dojindo Laboratories, Kumamoto, Japan) in 50 mM Tris–HCl (pH 7.6) containing 0.068% imidazole (Sigma–Aldrich) and 0.02% hydrogen peroxide. Control sections were incubated with normal rabbit serum (Dako) instead of the anti-RXR antibody. All sections were counterstained with Mayer’s hematoxylin.

Results

Production of a polyclonal antibody against RXR of the rock shell

Two sets of approximately 90 mL of affinity-purified antiserum (89 and 86 mL for rabbits no. 1 and 2, respectively) were obtained. The antibody titer measured by ELISA is shown in Fig. 1. The values of OD450 were over 1.1 in the antiserums at a dilution of 1:12,500, and the values were around 0.9 even in the antiserums at a dilution of 1:62,500 (Fig. 1). Affinity-purified antiserum of rabbit no. 2 was selected to use in the following experiments.

Fig. 1
figure 1

The antibody titers in antiserums of two rabbits against RXR of the rock shell Thais clavigera measured by ELISA. Antiserums were diluted in 100–62,500 folds. Before, before antigen injections; After, approximately 2 months after antigen injection

Full size image

Detection of RXR protein using immunoblotting

The reaction was represented as a major band of approximately 50 kDa (Fig. 2a, b). Because the molecular weight of T. clavigera RXR is approximately 49 kDa, immunoblotting demonstrated that this antiserum against T. clavigera RXR could detect T. clavigera RXR (Fig. 2a, b). Anti-T. clavigera RXR antibody strongly recognized RXR in extracts from T. clavigera RXR-expressed culture cells but not in those from human RXR α-expressed cells (Fig. 2a). Using anti-human RXR α antibody, which recognized a range of domains from the DNA-binding domain to the F domain of human RXRs, we found expression of both T. clavigera and human RXRs (Fig. 2c). Thus, this antiserum against RXR of the rock shell was able to specifically detect RXR protein in tissues of T. clavigera.

Fig. 2
figure 2

Specification of anti-Thais clavigera RXR antibody. COS-1 cells were transiently expressing (1) T. clavigera RXR, (2) human RXR α and (3) pcDNA3.1(+). Equal volumes of cell lysates were analyzed by immunoblotting using (a, b) anti-T. clavigera RXR antibody, (c) anti-human RXR α antibody, and (d) anti-β-actin antibody

Full size image

Immunohistochemical staining for RXR: localization of RXR in penis and vas deferens of male and imposex-exhibiting female rock shells

Immunohistochemical staining with anti-serum against T. clavigera RXR revealed the localization of RXR in the penis of wild males and imposex-exhibiting females (Fig. 3). In the vas deferens and penis in wild males, RXR was invariably detected in the nuclei of epithelial cells and in the cells of the smooth muscle layer surrounding the epithelium, as well as in cells in the interstitial tissue (Fig. 3b). Nuclear staining was also seen in the epidermis. No immunoreactivity was detected in any tissues treated with normal rabbit serum as negative controls (Fig. 3a), confirming binding specificity. In females exhibiting severe imposex, RXR expression was essentially the same as that in males, appearing in the nuclei of the epithelial cells lining the vas deferens and the surrounding smooth muscle cells (Fig. 3c). RXR was also evident in some of the nerve cells in the head ganglia of normal males (Fig. 3d), normal females, and in those of females showing imposex. Localization of RXR in penis and vas deferens of male and imposex-exhibiting female rock shells as well as localization of RXR in nerve cells in the head ganglia of rock shells were similar to findings previously reported by Horiguchi et al. (2007).

Fig. 3
figure 3

Immunohistochemical expression of RXR in male and imposex-exhibiting female T. clavigera. Males and imposex-exhibiting females were collected at Aikawa (reference site, April 2008) and Hinase (contaminated site, May 2009), respectively. a Penis of a male stained with normal rabbit serum without antibody, showing no staining. b Penis of a male stained with anti-T. clavigera RXR antibody, showing positive nuclear staining in the epithelial cells of the vas deferens, smooth muscle cells surrounding the epithelium, and interstitial cells. c Penis of a severely affected imposex-exhibiting female, showing similar RXR expression to that in the male of (b). d Head ganglion (pedal ganglion) of a normal male, showing positive staining in the nerve cells. Sections were counterstained with hematoxylin. Scale bars indicate 50 μm. hg, head ganglion; i, interstitial tissue (or interstitial cell); m, muscle layer (or smooth muscle cell); vd, vas deferens

Full size image

Discussion

In this study, we established a polyclonal antibody against RXR of T. clavigera as a specific antibody that recognizes T. clavigera RXR protein. This antibody could be very useful in analyses of the mode of action of organotin compounds, such as TBT and TPT, during their induction of imposex in gastropods. Horiguchi et al. (2007) reported localization of RXR in several tissues of T. clavigera on the basis of immunohistochemical staining with anti-human RXR α polyclonal antibody (D-20, Santa Cruz Biotechnology). The present study has confirmed the localization of RXR in penis and vas deferens of male and imposex-exhibiting female rock shells as well as in nerve cells in the head ganglia, results that are similar to those of Horiguchi et al. (2007). Thus, our hypothesis that RXR is involved in the differentiation of male-type genitalia in normal male rock shells and organotin-exposed females is further supported by these results.

RXR gene expression was observed previously in various tissues of the males, normal females, and imposex-exhibiting females examined (Horiguchi et al. 2007). RXR protein was also detected in penis and vas deferens, as well as the head ganglia (Fig. 3), suggesting that the tissue distribution of RXR in T. clavigera is extensive. This may indicate that RXR has multiple physiological functions or roles in various tissues of the rock shell, similar to vertebrates, where retinoic acids (RAs) play key roles in embryo patterning and organogenesis (Morris-Kay 1997; Redfern 1997).

To further analyze the induction mechanism of imposex caused by organotins in gastropods, the natural ligand to T. clavigera RXR and the possibility that T. clavigera RXR forms a heterodimer by coupling with other nuclear receptors (e.g., retinoic acid receptor (RAR), thyroid hormone receptor, vitamin D receptor, and peroxisome-proliferator-activated receptor (PPAR)) needs to be identified (Horiguchi et al. 2007).

Dmetrichuk et al. (2008) recently detected all-trans retinoic acid (ATRA) and 9CRA in the central nervous system of adults of the pulmonate gastropod Lymnaea stagnalis by high-performance liquid chromatography/mass spectrometry. Because ATRA and 9CRA were detected in tissue of L. stagnalis, the species likely also has metabolic enzymes to synthesize or transform RAs. The presence of 9CRA, a possible candidate as a natural ligand for RXR, should be investigated in the prosobranch gastropod T. clavigera. It is also necessary to examine whether the rock shell can inherently synthesize 9CRA and/or ATRA from β carotene; this may not be the case in light of the fact that no report has described the genes encoding RAR or the enzymes involved in the synthesis and degradation/metabolism of RAs (e.g., Raldh2, Cyp26) in invertebrates, except in the Prochordata (e.g., ascidians and amphioxus; Escriva et al. 1997; Nagatomo and Fujiwara 2003).

The activation of RXR–PPAR heterodimers by organotin compounds is considered to promote adipocyte differentiation (Grün and Blumberg 2006; Grün et al. 2006; Kanayama et al. 2005), and the binding and activation properties of various organotins with RXR–PPAR heterodimers were analyzed by le Maire et al. (2009). Gastropod imposex is induced by very low concentrations (~1 ng/L) of TBT and/or TPT (Bryan et al. 1986, 1987, 1988; Gibbs et al. 1987; Horiguchi et al. 1994, 1997a), but the mechanism of how such nanomolar levels of TBT and/or TPT can activate T. clavigera RXR remains to be clarified.

We also must examine the physiological and biochemical responses downstream of the RXR signaling pathway in order to elucidate the mechanisms of RXR-mediated transcription regulation. Such knowledge would improve our understanding of the entire mechanism of induction of imposex caused by organotin compounds, such as TBT and TPT, in gastropods, including the differentiation, growth and formation of male-type genitalia in normal male and organotin-exposed female rock shells.