Vitellogenin detection in Caiman latirostris (Crocodylia: Alligatoridae): a tool to assess environmental estrogen exposure in wildlife

J Comp Physiol B (2006) 176: 243–251 DOI 10.1007/s00360-005-0045-8

O R I GI N A L P A P E R

Florencia Rey Æ Jorge G. Ramos Æ Cora Stoker Leonardo E. Bussmann Æ Enrique H. Luque Mo´nica Mun˜oz-de-Toro

Vitellogenin detection in Caiman latirostris (Crocodylia: Alligatoridae): a tool to assess environmental estrogen exposure in wildlife Received: 16 June 2005 / Revised: 30 August 2005 / Accepted: 20 September 2005 / Published online: 15 November 2005
Springer-Verlag 2005

Abstract Environmental pollution with endocrine disrupting compounds (EDCs) has adverse effects on the ecosystem’s health. Caiman latirostris are widely distributed in South American aquatic ecosystems. Caimans have physiological and ecological characteristics that make them particularly vulnerable to EDCs exposure and suitable candidate as a sentinel species. Vitellogenin (Vtg) is a yolk pre-cursor protein synthesized by the liver of non-mammalian vertebrates and induced in response to estrogen. Purified plasma Vtg from caimans injected with estradiol-17b (E2) was used to generate a polyclonal anti-body. Anti-body specificity was assessed using Western blot. The antiserum was also effective in detecting turtle Vtg, exhibiting high cross-reactivity with Vtg from Phrynops hilarii and Trachemys scripta dorbigni. We developed a specific and highly sensitive ELISA for caiman Vtg. This method has a detection limit of 0.1 ng/ml of plasma. The ELISA did not detect Vtg in plasma of non-induced male caimans. Induction of Vtg in male caimans was evaluated in response to one or two (7 days apart) doses of E2. Due to its high sensitivity, ELISA allows to measure the small increases in plasma Vtg after exposure to exogenous estrogen. A priming effect was observed following the second E2 dose, with a tenfold increase in circulating Vtg. Hepatic synthesis was confirmed by immunohistochemistry. The results presented herein suggest that detection of plasma Vtg in male caimans might become a valuable tool in biomonitoring xenoestrogen exposure in a polluted environment. Communicated by G. Heldmaier F. Rey Æ J. G. Ramos Æ C. Stoker Æ E. H. Luque Æ M. Mun˜oz-de-Toro (&) Laboratorio de Endocrinologı´ a y Tumores Hormonodependientes, School of Biochemistry and Biological Sciences, Universidad Nacional del Litoral, C. C. 242, (3000) Santa Fe, Argentina E-mail: [email protected] Fax: +54-342-4575207 L. E. Bussmann Instituto de Biologı´ a y Medicina Experimental, CONICET, Buenos Aires, Argentina

Keywords Vitellogenin Æ Caiman latirostris Æ Xenoestrogen Æ Aquatic ecosystem Æ Reptile Æ Endocrine disruptor Abbreviations BSA: bovine serum albumin Æ DAB: diaminobencidine Æ E2: estradiol-17b Æ EDC: endocrine disrupting compounds Æ NHS: Nhydroxysuccinimide Æ PBS: phosphate buffer saline Æ PBST: PBS-0.05% Tween-20 Æ PMSF: phenylmethylsulphonyl fluoride Æ SDS-PAGE: sodium dodecylsulphate polyacrylamide gel electrophoresis Æ TTBS: tris-buffered saline-Tween Æ TMB: 3,3¢,5,5:-tetra methyl benzidine Æ Vtg: vitellogenin

Introduction Every day, humans and wildlife are exposed to contaminants which have the potential to interfere with their endocrine system by acting as endocrine disrupting compounds (EDCs) (Colborn et al. 1993; McLachlan 2001). The xenobiotic estrogens or xenoestrogens (phytoestrogens, anti-oxidants, plasticizers, detergents, polychlorinated biphenyl congeners and pesticides) are among the EDCs (Sonnenschein and Soto 1998). Reduced fertility, lower hatch rates in oviparous vertebrates, decreased viability of offspring, altered hormone levels and/or sexual behavior have been associated with xenoestrogens exposure (Bergeron et al. 1994; Guillette et al. 1994; Guillette and Crain 1995; Matter et al. 1998; Markey et al. 2001; Ramos et al. 2001; Ramos et al. 2003). To monitor environmental contamination and ecosystem’s health, sensitive and specific tests that would allow the identification and quantitation of markers of xenoestrogen exposure are in great need. Vitellogenin (Vtg) is a large phospholipoglycoprotein normally synthesized in the liver of adult non-mammalian female vertebrates in response to estrogenic stimulation. Circulating Vtg is cleaved to form yolk proteins in the oocytes; these proteins are nutritive material for developing embryos (Tata and Smith 1979). Vtg is

244

normally undetectable in the plasma of immature female and male of oviparous animals, pre-sumably because they lack enough circulating estrogen to trigger Vtg gene expression. Nevertheless, Vtg can be induced by exposure to exogenous estrogens or to xenoestrogens. Thus, the presence of Vtg in the blood of immature female and/or male caimans would be a suitable biomarker of exposure to exogenous estrogenic compounds. The potential use of Vtg as a biomarker of estrogenicity has already been explored in fishes, turtles and amphibians for which in vivo and in vitro assays have been developed (Sumpter and Jobling 1995; Palmer et al. 1998; Kime et al. 1999; Selcer et al. 2001; Hutchinson and Pickford 2002; Cheek et al. 2001; Cheek et al. 2004; Marin and Matozzo 2004; Tada et al. 2004). Broad-snouted caimans (Caiman latirostris Daudin 1802—Crocodylia: Alligatoridae) are widely distributed in South American aquatic ecosystems (Yanosky 1990; Verdade 1995). Caimans spend a large portion of their lives in the water, they are long-lived animals, and they are at the top of the food web. All these make them particularly susceptible to EDCs exposure. The abovementioned characteristics led us to hypothesize that caimans, specifically C. latirostris, may be a suitable species to be used as sentinel for monitoring xenoestrogen contamination in the environment. This study reports the development of a highly specific and sensitive immunoassay for the detection of the estrogen-induced Vtg in C. latirostris.

Material and methods Animals and samples All the experiments, both in the laboratory and in the field work, were conducted in full compliance with the Universidad Nacional del Litoral Institutional Animal Care and Use Committee. The following animals were used: (1) juvenile broadsnouted caimans (C. latirostris) with an average body weight of 2.08±0.20 kg, (2) juvenile slider turtles (Trachemys scripta dorbigni) weighing 0.35±0.05 kg, (3) juvenile side neck turtles (Phrynops hilarii) (0.84± 0.11 kg), and (4) sexually mature P. hilarii (2.85± 1.25 kg). Juvenile animals were raised under controlled environmental conditions in the facilities of the ranching program ‘‘Proyecto Yacare´’’ (Santa Fe, Argentina). Food was provided ad libitum. In caimans, sex was determined by examination of external genitalia, as previously described (Stoker et al. 2003). Juvenile turtle sex was evaluated by histological examination of the gonads. Adult P. hilarii were captured in a pond located in an urban recreational area. Adult turtles were identified, weighed, sexed, and blood samples were collected. Immediately after that, the turtles were released. Blood was drawn from the occipital sinus of caimans and from the caudal vein of turtles. The samples were collected in pre-cooled (0C) heparinized tubes containing the pro-

tease inhibitor aprotinin (Sigma, St. Louis, MO, USA) at 0.0175 trypsin-inhibiting units (TIU)/ml and the serine protease inhibitor PMSF (Sigma) (0.5 mM). Plasma was separated by centrifugation at 4C and stored at 80C. Caiman liver samples were fixed in 10% PBSformalin for 6 h at room temperature and embedded in paraffin. Induction, characterization and purification of caiman vitellogenin For 7 days, four immature female caimans were daily injected subcutaneously with 1 mg/kg body weight of estradiol-17b (E2) (Sigma) dissolved in ethanol and diluted in sesame oil (ethanol-sesame oil 1:9). Blood samples were collected immediately before treatment (basal level) and 7 days after the last injection. The nontreated controls were four immature female caimans injected with vehicle (sesame oil). To identify Vtg four different approaches were used: (1) plasma proteins of controls and E2-treated animals were separated by native and SDS-PAGE and stained with Coomassie blue (Sigma), (2) phosphorilated proteins were recognized in gels by the cationic carbocyanine dye, Stains All (Sigma) (Green et al. 1973), (3) molecular weights of E2-induced proteins were compared with the molecular weights reported for Vtg of other species, and (4) protein reaction with the primary polyclonal antibody no. 498 from Dr. KW Selcer (Department of Biological Sciences, Duquesne University, Pittsburgh, PA, USA) was tested. This antiserum was generated against a portion of the Xenopus laevis Vtg molecule, a peptide representing a highly conserved region of the Vtg molecule (Selcer et al. 2001). In every electrophoresis a comparison between control and E2-treated female caiman plasmas was done. The molecular mass of Vtg was calculated as the average from three gels. Caiman Vtg was isolated following the protocol described by Wiley et al. (1979), with modifications. Before Vtg precipitation, additional aprotinin and PMSF were added to plasma samples. All the solutions used were pre-cooled. The precipitation was performed by mixing aliquots of 500 ll of E2-treated female plasma with 2 ml of 20 mM Na2EDTA (pH 7.7) and 160 ll of 0.5 M MgCl2, and the mixture centrifuged at 2,500g for 15 min at 4C. The supernatant was discarded, and the pellet containing Vtg was resuspended in 200 ll of 1 M NaCl, 50 mM Tris–HCl (pH 7.5) and centrifuged at 2,500g for 30 min at 4C to separate impurities. The supernatant containing the dissolved Vtg was collected, and Vtg was precipitated again by decreasing the ionic force with 2.5 ml of distilled H2O and centrifuged at 2,500g for 15 min at 4C. The supernatant was stored at 20C while the pellet was redissolved in 100 mM NaCl, 50 mM Tris–HCl (pH 7.5). This fraction was centrifuged at 18,400g for 10 min at 4C. The lipid fraction formed at the top of the solution was discarded. This new fraction was pooled with the last supernatant and used

245

to perform ion-exchange chromatography. A Hi-Trap Q Sepharose XL column (16·25 mm; 5 ml) (Amersham Bioscience AB, Uppsala, Sweden) connected to a Minipuls3 peristaltic pump (Gilson Inc., Villiers le Bel, France) was used. Absorbance at 280 nm was measured with a Beckman DU 530 spectrophometer (Beckman Instruments, Inc. Fullerton, CA, USA). The sample was diluted 1:9 with Tris–HCl (25 mM, pH 9) and with additional PMSF (0.5 mM) and loaded into the column with a flow rate of 0.5 ml/min. The column was eluted with 50 ml of a linear gradient from 0 to 1 M NaCl, 25 mM Tris–HCl (pH 9) with a flow rate of 1 ml/min. Fractions of 1 ml were collected and stored at 80C until analyzed by SDS-PAGE. Total protein concentration in plasma and in each fraction obtained during purification was measured using the method described by Bradford (1976), using BSA fraction V (Sigma) as standard. Production of caiman vitellogenin antiserum Polyclonal anti-bodies against purified caiman Vtg were generated according to the protocol described by Diano and Le Bivic (1996), with minor modifications. Briefly, caiman Vtg purified from ion-exchange chromatography was loaded in native gels and electroblotted onto a nitrocellulose membrane (0.45 lm, Protean, Schleicher and Schuell BioScience Inc., Keene, NH, USA). The membrane was cut into small pieces and macerated by sonication in an ice bath (High Intensity Ultrasonic Processor, Vibra-Cell, Sonics and Materials Inc., Newtown, CT, USA). Two rabbits were immunized with approximately 100 lg of Vtg according to Vaitukaitis et al. (1971). Two weeks later, rabbits were boosted with other 100 lg of Vtg and bled from the auricular vein. Subsequent boosts (three in total), and bleeds were performed monthly. Serum titration was performed by dot and Western blot (see below) using purified Vtg and plasma of E2-treated female caimans. Samples with titers higher than 1:8,000 were stored at 20C. Rabbit polyclonal anti-Vtg was purified using a 1 ml affinity column (Hi Trap rProtein A FF, Amersham, USA) following manufacturer’s instructions. A second round of chromatography was assessed from the IgGcontaining fraction using a N-hydroxysuccinimide (NHS) agarose activated column. Briefly, a total of 5 mg Vtg was dialyzed against a solution containing 200 mM NaHCO3 and 500 mM NaCl (pH 8.3) and concentrated in 1 ml. Vtg solution was added to a 1 ml Hi-Trap NHS activated HP column (Amersham) and incubated 1 h at room temperature. After deactivation of the remaining active groups, the column was washed with PBS, then with elution buffer (glycine 100 mM, pH 3) and finally equilibrated with PBS. The IgG-enriched fraction was loaded into the column and washed with PBS. Elution was performed with 100 mM glycine buffer (pH 3) and fractions of 1 ml were collected with 60 ll of 1.5 M Tris–HCl (pH 8.6). Absorbance was measured at

280 nm and fractions containing double purified antiVtg were pooled and used in the immunoassays. Resultant anti-body was named LETHW 03-07. Western blot To characterize anti-caiman Vtg serum, proteins of whole plasma from E2-treated and untreated animals or purified Vtg were electrophoretically separated under native or denaturing conditions and electroblotted. Nitrocellulose membranes were blocked with 5% w/v non-fat dry milk in Tris-buffered saline-Tween (TTBS; 25 mM Tris, 0.14 M NaCl, 0.05% v/v Tween-20, pH 7.4) overnight at 4C to prevent non-specific binding. Membranes were rinsed with TTBS (0.5 M NaCl) (two short washes followed by three washes of 10 min each with agitation) and incubated 1.5 h with anti-caiman Vtg LETHW 03-07, 1:1,000 or 1:1,600 (diluted in 2% non-fat dry milk in TTBS) at 25C in an orbital shaker. The membrane was rinsed again with TTBS (0.5 M NaCl) and incubated in orbital shaker for 1 h at 25C with peroxidase-labeled goat anti-rabbit (Amersham) (diluted 1:2,500 in 2% non-fat dry milk in TTBS). The membrane was rinsed again in TTBS (0.5 M NaCl) and developed using 0.5 mg/ml diaminobencidine (DAB; Sigma) for 10 min. Additionally, the same protocol was followed using Dr. Selcer’s primary anti-Vtg anti-body no. 498 (diluted 1:100 in 2% non-fat dry milk in TTBS). Dot blot The dot blot assay was optimized as a Vtg screening method. For this purpose, 0.3–80 ng of purified Vtg were mixed with a constant amount of male plasma (basal condition) and directly added to a nitrocellulose membrane. Blots were dried and the immunodetection of Vtg was performed following the protocol described for Western blot. The primary anti-body LETHW 03-07 was tested at 1:500–1:8,000 dilutions and the secondary anti-body (Amersham) was diluted 1:2,500. Immunohistochemistry LETHW 03-07 Vtg antiserum was tested in an immunohistochemical protocol using liver samples of E2treated and non-treated caimans. Sections (5 lm in thickness) of paraffin embedded tissue were mounted on 3-aminopropyl triethoxysilane (Sigma)-coated slides. A standard immunohistochemical technique (avidin-biotin-peroxidase) with a microwave pre-treatment was used (Mun˜oz-de-Toro et al. 1998). Endogenous peroxidase activity and non-specific binding sites were blocked. Primary anti-body (LETHW 03-07), diluted 1:100, was incubated overnight at 4C. The reaction was developed using DAB. Samples were counterstained with Mayer’s hematoxylin and mounted with permanent mounting

246

medium (PMyR, Buenos Aires, Argentina). For negative controls, primary anti-body was replaced with nonimmune rabbit serum (Sigma).

the negative controls, and the ELISA sensitivity was calculated as the slope of the standard curve. Induction of Vitellogenin in juvenile male caimans

Cross-reactivity of the anti-body with turtle vitellogenin Plasma proteins from juvenile T. s. dorbigni and P. hilarii treated with E2 following the Palmer and Palmer (1995) protocol (1 mg/kg bw sc twice; 7 days apart), were separated by electrophoresis and immunoblotted. Reactivity of LETHW 03-07 with turtle Vtg was tested by Western blot. Plasma samples from sexually mature P. hilarii specimens were screened for Vtg throughout their reproductive season.

Male caimans were used to evaluate their ability to synthesize Vtg in response to E2. Animals were injected with 1 mg/kg bw of E2 sc twice; 7 days apart. Blood was collected before treatment (basal level), at the time of the second injection, and 7 days after the last injection. Plasma Vtg was screened by dot blot and circulating levels were measured using ELISA. Vtg expression in liver samples was evaluated by immunohistochemistry.

Competitive ELISA

Results

A competitive ELISA was developed to measure the ability of soluble Vtg to inhibit the binding of anti-Vtg anti-body to solid-phase immobilized antigen. Assay conditions, such as antigen coating mass, reagents dilutions and incubation times, were optimized using a checkerboard binding approach and purified C. latirostris Vtg as a reference standard. All samples were run in duplicate on the same plate. Polystyrene microtiter plates (Greiner Bio-One GmbH, Frickenhausen, Germany) were coated, overnight at 4C, with 100 ll of Vtg standard solution (50 ng/well) using sodium carbonate buffer (50 mM, pH 9.6) as diluent. Wells were washed three times with PBS and blocked 1.5 h at 37C with 4% w/v non-fat dry milk in PBS-0.05% Tween-20 (PBST). LETHW 03-07 dilutions ranged from 1/500 up to 1/ 50,000 were tested, and 1/1,000 was selected as the final optimal dilution to perform competitive ELISA. Purified Vtg was diluted in non-treated male caiman plasma at starting concentration of 25 lg/ml, serial twofold dilutions in ELISA buffer were mixed with an equal volume of 1/500 (final dilution 1/1,000) anti-Vtg, and incubated overnight at 4C. The pre-incubated mixtures were transferred to the plates (100 ll per well) and incubated for 30 min at 37C, before being washed with PBST. The plates were incubated for 1 h at 37C with peroxidasecoupled goat anti-rabbit (Amersham) at 1:1,500. Following a final washing step, plates were developed with 3,3¢,5,5-TetraMethylBenzidine (TMB) (Zymed, San Francisco, CA, USA) and the reaction was stopped with 12% H2SO4. The absorbance of each well was measured at 450 nm using a plate reader (Multiskan EX, Thermo Electro Co., Finland, EU). Negative controls were prepared using non-treated male caiman plasma or ELISA buffer. Dilutions of E2-treated caiman plasma were also tested. The standard curve was constructed by plotting the absorbance values versus the log values of the Vtg concentration. The detection limit was established as the titer value that differs in three standard deviations from

Vitellogenin induction and characterization As shown in Fig. 1a, E2 treatment induced the expression of many proteins. A pair of them with molecular masses of 205 and 225 kDa in SDS-PAGE met the properties described for Vtg in other oviparous species. They were detected in all E2-treated caiman plasma, while they were absent in non-treated males and juvenile animals; and they are phosphorilated proteins, as evidenced by their blue stain developed with Stains All (Fig. 1b). Further confirmation was achieved by Western blot (Fig. 1c) using the primary anti-body no. 498 provided by Dr. Selcer, which proved to be effective in detecting alligator Vtg (Selcer et al. 2001). Purified caiman Vtg was eluted during the second peak with 550 mM NaCl in Tris–HCl buffer using Hi-Trap Q Sepharose XL column (Fig. 2). Anti-body characterization The polyclonal anti-body generated against purified Vtg (LETHW 03-07) was successfully used in Western blot, dot blot, immunohistochemistry and ELISA. Western and dot blot Figure 3 shows the E2-induced proteins detected using the newly developed LETHW 03-07 anti-body. The immunoblotting showed two strong bands with molecular weights of 205 and 225 kDa (Fig. 3). Performing PAGE in native conditions and immunoblotting with LETHW 03-07, two high molecular weight proteins were also detected (MW 395 and 415 kDa) (data no shown). The minimum amount of Vtg detected by Western blot was 0.01 lg (0.3 mg/ml). Under the same conditions, dot blot analysis was even more sensitive, detecting amounts of plasma Vtg as low as 0.6 ng.

247

Fig. 1 Gel electrophoresis of plasma samples from E2-treated and non-treated juvenile female caimans. Samples were separated on a 5–8% SDS-PAGE. Coomassie blue staining shows two bands (a), and the cationic carbocyanine dye, Stains All, identified the proteins as being phosphorilated (b). c Western blot analysis of plasma using a polyclonal anti-body against a synthetic peptide from X. laevis Vtg provided by Dr. KW Selcer (dilution 1/100). Vtg protein bands of 205 and 225 kDa (arrows) are present in plasma of E2-treated caimans, but not in controls

Fig. 3 Western blot analysis of plasma from E2-treated and nontreated juvenile female caimans. Samples were separated in SDSPAGE gels and processed for immunoblotting analysis using LETHW 03-07 antiserum as described in Material and methods. Negative controls: 20 lg of plasma proteins of E2-treated female were incubated with rabbit pre-immune serum. To measure the basal levels of Vtg in non-treated female caimans, 20 lg of plasma proteins were incubated with LETHW 03-07. To measure proteins induced following E2 treatment, plasma aliquots containing 0.3 lg of proteins were incubated with LETHW 03-07. Only E2-treated juvenile female samples exhibited reaction with LETHW 03-07 showing two bands of 205 and 225 kDa (arrows)

Fig. 2 Ion exchange elution profile from purified plasma of E2treated female caimans. Vtg: Vitellogenin-containing peak eluted with 550 mM NaCl in Tris–HCl buffer, collected in fractions of 1 ml. Chromatography was performed using a Hi-Trap Q Sepharose XL column

Immunohistochemistry LETHW 03-07 anti-body was suitable for detecting Vtg expression in caiman paraffin embedded tissue. Hepatocytes from E2-treated female caimans showed a

well-localized immunostaining. Cytoplasmic aggregates near the apical membrane opposite to the nucleus were observed (Fig. 4a, b). Cross-reaction with turtle Vtg LETHW 03-07 exhibited cross-reactivity when plasma samples from both turtle species (P. hylarii and T. s. dorbigni) were tested (Fig. 5). The protein recognized by LETHW 03-07 was induced after E2 treatment in juvenile turtles and was present in plasma of adult female turtles throughout the reproductive season. Turtle Vtg has a mass of approximately 215 kDa.

248 Fig. 4 Photomicrographs of caiman liver sections immunostained with LETHW 03-07 anti-body. Hepatocytes from control juvenile female (a) and male (c), lacking Vtg expression. E2-treated juvenile female (b) and male (d) samples showing Vtg granules in the apical region of hepatocytes (arrows)

Competitive ELISA As shown in Fig. 6, when the competitive ELISA was performed in the optimized conditions (LETHW 03-07, 1:1,000 and anti-rabbit peroxidase-labeled, 1:1,500) a broad working range of 0.006 to 1.56 lg/ml of caiman Vtg is observed. The detection limit of the assay was

Fig. 5 Vtg induction in juvenile turtles (P. hilarii and T. s. dorbigni) following E2 treatment. a P. hilarii plasma proteins stained with Coomassie blue separated on a 5% SDS-PAGE. b Western blot of plasma samples from both turtle species. Arrows show that a single protein induced by E2 treatment (~215 kDa) is recognized by the LETHW 03-07 anti-body

0.1 ng/ml and the sensitivity, represented by the slope of the standard curve, was 0.20 log ml/lg. Vitellogenin induction in male juvenile caimans After treatment, the total protein concentration in male caiman plasma increased from 53.5±3.1 mg/ml to 58.0±2.9 mg/ml and 65.5±2.8 mg/ml after the first and second E2 doses, respectively. This increase was primarily due to induction of proteins identified as Vtg by SDS-PAGE and Western blots. ELISA was used to determine levels of Vtg in plasma samples. Vtg was not detectable at basal levels or after vehicle administration in all male caiman studied. Seven days after a single dose of E2 (1 mg/kg) all males (n=7) exhibited detectable Vtg levels (1.18±1.11 mg/ml), and on day 14 (7 days after the second E2 dose) Vtg levels in the same animals were significantly higher (10.15±2.40 mg/ml). Interestingly, mean Vtg plasma levels increased about tenfold higher after the second E2 dose compared to mean values found after the first dose. However, a wide range of variation in individual responses was observed, ranging from 3 up to 24-fold. In hepatic tissue of male caiman treated with E2, Vtg expression was clearly detected by immunohistochemistry (Fig. 4c, d).

Discussion In the present study a specific and highly sensitive ELISA method for C. latirostris Vtg was developed. ELISA yielded quantitative measurements of plasma

249

Fig. 6 Calibration curve for C. latirostris Vtg quantification using a competitive ELISA as described in Material and methods. The lowest Vtg concentration detected was 1 ng/ml and the sensitivity, calculated as the slope of the calibration curve, was 0.20 log ml/ lg. Data were obtained from five identical assays, bars represent standard errors of the mean

Vtg in caimans and allowed assessment of exposure to exogenous estrogen. Estradiol-17b has been previously used to induce Vtg production in other oviparous vertebrate species including fishes, amphibians and reptiles (Palmer and Palmer 1995; Selcer et al. 2001; Herbst et al. 2003). By treating male and female juvenile C. latirostris with E2 we induced the production of two plasma proteins with an approximate mass of 205 and 225 kDa in SDSPAGE. This is consistent with the reported molecular weight of Vtg in Alligator mississipiensis (Selcer et al. 2001). Other features exhibited by these two plasma proteins support their identification as Vtg. They were also recognized by the polyclonal anti-body generated against a portion of the X. laevis Vtg molecule (Selcer et al. 2001). Moreover, its phosphorilated nature was demonstrated using the cationic carbocyanine dye, which differentially stains phosphoproteins on polyacrylamide gels (Green et al. 1973). It is likely that the two Vtg plasma proteins in C. latirostris may represent the product of two different genes, since multiple Vtg genes has been described in fish, Xenopus and chicken (Bowman et al. 2000). In Western blot assays, our LETHW 03-07 anti-Vtg serum, generated against the purified whole protein from C. latirostris, exhibited a stronger reaction with caiman Vtg than the anti-body no. 498 provided by Dr. Selcer and used as positive control. The rationality for this difference might be that our antiserum raised against the whole molecule recognized larger number of epitopes than Selcer’s antiserum. Our antiserum was also effective in detecting turtle Vtg, exhibiting high cross-reactivity with Vtg from both turtle species assessed (P. hylarii and T. s. dorbigni). This cross-reactivity increases the number of target species in which LETHW 03-07 anti-body could be used.

The induction of Vtg in oviparous species is used as a standard biomarker of exposure to estrogenic compounds (Palmer and Palmer 1995; Sumpter and Jobling 1995; Irwin et al. 2001; Cheek et al. 2004). Many environmental contaminants are considered to be xenoestrogens since they show an estrogenic response in different in vivo and/or in vitro assays (Soto et al. 1994; Nakada et al. 2004). The use of physiological markers, such as E2-induced proteins, as indicators of early response to environmental contamination with xenoestrogens may provide important information regarding the potential effects of these contaminants on ecosystem’s health. Assessing exposure of wildlife to environmental contaminants often involves killing animals for collection of biologic samples. The ELISA developed in this study offers certain advantages such as involves a non-aggressive sampling technique, is highly sensitive, specific and cost effective, allowing the screening of large number of samples for low levels of Vtg. Plasma levels of Vtg in males and juvenile females of caiman would be a useful biomarker of xenobiotic estrogen exposure. Our results support this hypothesis and showed that caimans are very sensitive to E2. Moreover, when two doses of E2 were given, a priming effect was induced by the first dose in terms of Vtg induction, suggesting that some kind of molecular ‘‘memory’’ mechanisms may be working at different levels (i.e. genomic, epigenetic, stability of the RNA, recruitment of steroid receptors co-activators, protein synthesis, processing, stored pools, export etc.). The significance of this priming effect is currently unknown; however, as large variations in individual caiman responses were observed we could speculate that differences in xenoestrogens exposure during embryo development could be the mechanism underlying E2 sensitivity enhancement (Foran et al. 2002). The priming effect described here, in caimans, could be a useful model to test the hypothesis that an epigenetic imprinting mechanism is responsible to transmit the ‘‘memory’’ of prior estrogen/xenoestrogens exposure. Furthermore, Vtg assay can be used to study female caiman reproductive biology in free-living animals. Female alligators and caimans exhibit distinct seasonal reproductive cycle beginning in spring with the synthesis of Vtg, previous to the ovulation (Guillette et al. 1997; Lance 2003; Stoker 2004). In American alligators, only a portion of the female population exhibits reproductive activity in a given season mainly because of different quality of habitat conditions (Guillette et al. 1997; Lance 2003). Measuring the blood levels of Vtg might be a useful tool to identify female caimans that are reproductively active and also to identify the proportion of the adult population that is preparing to nest in a particular year. Therefore, vitellogenesis can be a useful indicator of both the reproductive physiology and the endocrine disruption by xenobiotics (Kime et al. 1999). Both, inhibition and stimulation of vitellogenesis can have direct consequences on reproductive capacity (Murphy et al. 2005). Altered vitellogenesis due to

250

reduced hepatic synthesis of Vtg and/or its decreased serum levels as a result of environmental contamination have been reported in mature female fishes (Thomas 1990; Kirubagaran and Joy 1995; Hwang et al. 2000). On the other hand, additive, antagonistic and synergistic effects have been reported for xenoestrogens present in the environment (Rice et al. 2003). Therefore, in order to have a more complete assessment of the conditions of a particular environment it would be appropriate to test for the presence of chemical compounds as well as to use a biological assay to evaluate the impact on wildlife. The selection of sentinel species and biomarkers is critical to develop a feasible monitor system. Previous studies showed that turtles and fishes are useful species to monitor environmental contamination by xenoestrogens (Bergeron et al. 1994; Crews et al. 1995). Recently, we confirmed that sex differentiation in C. latirostris is temperature-dependent, and we showed sex reversal effect and/or altered gonadal architecture after in ovum exposure to a low dose of the xenoestrogen bisphenol A (Stoker et al. 2003). While sex determination, gonadal histoarchitecture and sex hormones profile are potential biomarkers of pre-natal exposure to xenoestrogens, the induction of Vtg in oviparous male vertebrates represents a more sensitive biochemical endpoint of post-natal exposure to estrogenic compounds (Palmer and Palmer 1995; Sumpter and Jobling 1995; Cheek et al. 2004). In the present study, we provide new tools that will allow us to better characterize C. latirostris as sentinel species for monitoring xenoestrogen contamination in aquatic ecosystem. Ongoing studies will provide new data regarding Vtg levels in caimans in their natural habitats. These results will be useful to identify wetland areas with high and low xenoestrogen pollution and significant to preserve caiman wild populations and ecosystem’s health. Acknowledgments The authors thank Dr. MV Maffini (Tufts University School of Medicine, Boston, MA, USA) for her critical reading of the manuscript and Dr. KW Selcer (Duquesne University, Pittsburgh, PA, USA) for providing the polyclonal anti-body (no. 498) generated against X. laevis Vtg. We are also grateful to the team of ‘‘Proyecto Yacare’’ (Santa Fe, Argentina) for providing with the caimans and for their proficiency in animal care, and Fundacio´n Vida Silvestre Argentina (Chaco, Argentina) for assistance during the field work. This work was supported by grants from Morris Animal Foundation D04ZO-108 (Colorado, USA), the Argentine National Agency for the Promotion of Science and Technology (ANPCyT) (PICT-99 N 13-7002), and the Universidad Nacional del Litoral (Santa Fe, Argentina). F.R. is Fellow and J.G.R., L.E.B. and E.H.L. are Career Investigators of the Argentine National Council for Science and Technology (CONICET).

References Bergeron JM, Crews D, McLachlan JA (1994) PCBs as environmental estrogens: turtle sex determination as a biomarker of environmental contamination. Environ Health Perspect 102:780–781 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254

Bowman CJ, Kroll KJ, Hemmer MJ, Folmar LC, Denslow ND (2000) Estrogen-induced vitellogenin mRNA and protein in sheepshead minnow (Cyprinodon variegates). Gen Comp Endocrinol 120:300–313 Cheek AO, Brouwer TH, Carroll S, Manning S, McLachlan JA, Brouwer M (2001) Experimental evaluation of vitellogenin as a predictive biomarker for reproductive disruption. Environ Health Perspect 109:681–690 Cheek AO, King VW, Burse JR, Borton DL, Sullivan CV (2004) Bluegill (Lepomis macrochirus) vitellogenin: purification and enzyme-linked immunosorbent assay for detection of endocrine disruption by papermill effluent. Comp Biochem Physiol C Toxicol Pharmacol 137:249–260 Colborn T, vom Saal FS, Soto AM (1993) Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ Health Perspect 101:378–384 Crews D, Cantu AR, Bergeron JM, Rhen T (1995) The relative effectiveness of androstenedione, testosterone, and estrone, precursors to estradiol, in sex reversal in the red-eared slider (Trachemys scripta), a turtle with temperature-dependent sex determination. Gen Comp Endocrinol 100:119–127 Diano M, Le Bivic A (1996) Production of highly specific polyclonal antibodies using a combination of 2D electrophoresis and nitrocellulose-bound antigen. In: Walker J (ed) The protein protocols handbook. Humana Press, Totowa, NJ, USA, 122:703–710 Foran CM, Peterson BN, Benson WH (2002) Transgenerational and developmental exposure of Japanese medaka (Oryzias latipes) to ethinylestradiol results in endocrine and reproductive differences in the response to ethinylestradiol as adults. Toxicol Sci 68:389–402 Guillette LJ Jr, Crain DA (1995) Endocrine-disrupting contaminants and reproductive abnormalities in reptiles. Comments Toxicol 5:381–399 Guillette LJ Jr, Gross TS, Masson GR, Matter JM, Percival HF, Woodward AR (1994) Developmental abnormalities of the gonad and abnormal sex hormone concentrations in juvenile alligators from contaminated and control lakes in Florida. Environ Health Perspect 102:680–688 Guillette LJ Jr, Woodward AR, Crain DA, Masson GR, Palmer BD, Cox MC, You-Xiang Q, Orlando EF (1997) The reproductive cycle of the female American alligator (Alligator mississippiensis). Gen Comp Endocrinol 108:87–101 Green MR, Pastewka JV, Peacock AC (1973) Differential staining of phosphoproteins on polyacrylamide gels with a cationic cabocyanine dye. Anal Biochem 56:43–51 Herbst LH, Siconolfi-Baez L, Torelli JH, Klein PA, Kerben MJ, Schumacher IM (2003) Induction of vitellogenesis by estradiol17b and development of enzyme-linked immunosorbant assays to quantify plasma vitellogenin levels in green turtles (Chelonia mydas). Comp Biochem Physiol B 135:551–563 Hutchinson TH, Pickford DB (2002) Ecological risk assessment and testing for endocrine disruption in the aquatic environment. Toxicology 182:383–387 Hwang UG, Kagawa N, Mugiya Y (2000) Aluminium and cadmium inhibit vitellogenin and its mRNA induction by estradiol-17 beta in the primary culture of hepatocytes in the rainbow trout Oncorhynchus mykiss. Gen Comp Endocrinol 119:69–76 Irwin LK, Gray S, Oberdo¨rster E (2001) Vitellogenin induction in painted turtle, Chrysemys picta, as a biomarker of exposure to environmental levels of estradiol. Aquat Toxicol 55:49–60 Kime DE, Nash JP, Scott AP (1999) Vitellogenesis as a biomarker of reproductive disruption by xenobiotics. Aquaculture 177:345–352 Kirubagaran R, Joy KP (1995) Changes in lipid profiles and 32Puptake into phosphoprotein (vitellogenin) content of the ovary and liver in the female catfish, Clarias batrachus, exposed to mercury. Biomed Environ Sci 8:35–44 Lance VA (2003) Alligator physiology and life history: the importance of temperature. Exp Gerontol 38:801–805

251 Marin MG, Matozzo V (2004) Vitellogenin induction as a biomarker of exposure to estrogenic compounds in aquatic environments. Mar Pollut Bull 48:835–839 Markey CM, Luque EH, Mun˜oz-de-Toro M, Sonnenschein C, Soto AM (2001) In utero exposure to bisphenol A alters the development and tissue organization of the mouse mammary gland. Biol Reprod 65:1215–1223 Matter JM, Crain DA, Sills-McMurry C, Pickford DB, Rainwater TR, Reynolds KD, Rooney AA, Dikerson RL, Guillette LJ Jr (1998) Effects of endocrine disrupting contaminants in reptiles: alligators. In: Kendal RJ, Dickerson RL, Giesy JP, Suk WA (eds) Principles and processess for evaluating endocrine disruption in wildlife. SETAC Press, Pensacola, FL, pp 267–289 McLachlan JA (2001) Environmental signaling: what embryos and evolution teach us about endocrine disrupting chemicals. Endocr Rev 22:319–341 Mun˜oz-de-Toro MM, Maffini MV, Kass L, Luque EH (1998) Proliferative activity and steroid hormone receptor status in male breast carcinoma. J Steroid Biochem Mol Biol 67:333–339 Murphy CA, Rose KA, Thomas P (2005) Modeling vitellogenesis in female fish exposed to environmental stressors: predicting the effects of endocrine disturbance due to exposure to a PCB mixture and cadmium. Reprod Toxicol 19:395–409 Nakada N, Nyunoya H, Nakamura M, Hara A, Iguchi T, Takada H (2004) Identification of estrogenic compounds in wastewater effluent. Environ Toxicol Chem 23:2807–2815 Palmer BD, Palmer SK (1995) Vitellogenin induction by xenobiotic estrogens in the red-eard turtle and African clawed frog. Environ Health Perspect 103:19–25 Palmer BD, Huth LK, Pieto DL, Selcer KW (1998) Vitellogenin as a biomarker for xenobiotic estrogens in an amphibian model system. Environ Toxicol Chem 17:30–36 Ramos JG, Varayoud J, Sonnenschein C, Soto AM, Mun˜oz-deToro M, Luque EH (2001) Prenatal exposure to low doses of bisphenol A alters the periductal stroma and glandular cell function in the rat ventral prostate. Biol Reprod 65:1271–1277 Ramos JG, Varayoud J, Kass L, Rodrı´ guez H, Costabel L, Mun˜ozde-Toro M, Luque EH (2003) Bisphenol A induces both transient and permanent histofunctional alterations of the hypothalamic-pituitary-gonadal axis in prenatally exposed male rats. Endocrinology 144:3206–3215 Rice C, Birnbaum LS, Cogliano J, Mahaffey K, Needham L, Rogan WJ, vom Saal FS (2003) Exposure assessment for endocrine disruptors: some considerations in the design of studies. Environ Health Perspect 111:1683–1690 Selcer KW, Nagaraja S, Foret P, Wagner D, Williams L, Palmer BD (2001) Vitellogenin as a biomarker for estrogenic chemicals:

development of antibodies and primers with broad species applications. In: Robertson RL, Hansen LG (eds) PCBs: recent advances in the environmental toxicology and health effects. The University Press of Kentucky, Lexington, KY, pp 285–292 Sonnenschein C, Soto AM (1998) An updated review of environmental estrogen and androgen mimics and antagonists. J Steroid Biochem Molec Biol 65:143–150 Stoker C, Rey F, Rodriguez H, Ramos JG, Sirosky P, Larriera A, Luque EH, Mun˜oz-de-Toro M (2003) Sex reversal effects on Caiman latirostris exposed to environmentally relevant doses of the xenoestrogen bisphenol A. Gen Comp Endocrinol 133:287– 296 Stoker C (2004) Caiman latirostris como monitor biolo´gico de contaminacio´n ambiental por xenoestro´genos. Universidad Nacional del Litoral. Santa Fe, Argentina, Thesis Soto AM, Chung KL, Sonnenschein C (1994) The pesticides endosulfan, toxaphene, and dieldrin have estrogenic effects on human estrogen sensitive cells. Environ Health Perspect 102:380–383 Sumpter JP, Jobling S (1995) Vitellogenesis as a biomarker for estrogenic contamination of the aquatic environment. Environ Health Perspect 103:173–178 Tada N, Saka M, Ueda Y, Hoshi H, Uemura T, Kamata Y (2004) Comparative analyses of serum vitellogenin levels in male and female Reeves´ pond turtles (Chinemys reevesii) by immunological assay. J Comp Physiol B 174:13–20 Tata JR, Smith DF (1979) Vitellogenesis: a versatile model for hormonal regulation of gene expression. Rec Prog Horm Res 35:47–95 Thomas P (1990) Effect of Aroclor 1254 and cadmiun on reproductive endocrine function and ovarian growth in Atlantic Croaker. Mar Environ Res 28:499–503 Vaitukaitis J, Robbins J, Nieschlag E, Ross GT (1971) A method for producing specific antisera with small doses of immunogen. J Clin Endocrinol Metab 33:988–991 Verdade LM (1995) La conservacio´n y el manejo de los caimanes y cocodrilos en Ame´rica Latina. In: Larriera A, Verdade L (eds) vol. 1. Biologia reprodutiva do jacare´-de-papo-amarelo (Caiman latirostris) em Sao Paulo, Brasil. Santa Fe, Argentina, pp 57–79 Wiley HS, Opresko L, Wallace RA (1979) New methods for the purification of vertebrate vitellogenin. Anal Biochem 97:145– 152 Yanosky AA (1990) Histoire naturelle du Caı¨ man a` museau large (Caiman latirostris), un Alligatorine´ mal connu. Rev Fr Aquariol Herpetol 17:19–31