TOXICOLOGICAL SCIENCES, 141(2), 2014, 423–431 doi: 10.1093/toxsci/kfu145 Advance Access Publication Date: July 24, 2014

Gustavo A. Dominguez*,† , Joseph H. Bisesi, Jr., Kevin J. Kroll‡ , Nancy D. Denslow‡ , and Tara Sabo-Attwood *Department of Environmental Health Sciences, University of South Carolina, Columbia, South Carolina 29208, † Department of Environmental and Global Health, University of Florida, Gainesville, Florida 32610 and ‡ Department of Physiological Sciences, University of Florida, Gainesville, Florida 32611 1

To whom correspondence should be addressed. Fax: (352) 294-5293. E-mail: [email protected].

ABSTRACT The vitellogenin receptor (Vtgr) plays an important role in fish reproduction. This receptor functions to incorporate vitellogenin (Vtg), a macromolecule synthesized and released from the liver in the bloodstream, into oocytes where it is processed into yolk. Although studies have focused on the functional role of Vtgr in fish, the mechanistic control of this gene is still unexplored. Here we report the identification and analysis of the first piscine 5 regulatory region of the vtgr gene which was cloned from largemouth bass (Micropterus salmoides). Using this putative promoter sequence, we investigated a role for hormones, including insulin and 17␤-estradiol (E2 ), in transcriptional regulation through cell-based reporter assays. No effect of insulin was observed, however, E2 was able to repress transcriptional activity of the vtgr promoter through select estrogen receptor subtypes, Esr1 and Esr2a but not Esr2b. Electrophoretic mobility shift assay demonstrated that Esr1 likely interacts with the vtgr promoter region through half ERE and/or SP1 sites, in part. Finally we also show that ethinylestradiol (EE2 ), but not bisphenol-A (BPA), represses promoter activity similarly to E2 . These results reveal for the first time that the Esr1 isoform may play an inhibitory role in the expression of LMB vtgr mRNA under the influence of E2 , and potent estrogens such as EE2 . In addition, this new evidence suggests that vtgr may be a target of select endocrine disrupting compounds through environmental exposures. Key words: vitellogenin receptor; estrogen receptors; largemouth bass; insulin; estrogen; ethinylestradiol; bisphenol-A

Successful reproduction in oviparous vertebrates involves a series of vital events that include providing nutrients to growing embryos. This process involves production of vitellogenin (Vtg), a precursor egg yolk protein that is synthesized in the liver, secreted into the bloodstream, and subsequently taken up into oocytes. Selective incorporation of Vtg into oocytes is mediated by a surface receptor, termed the vitellogenin receptor (Vtgr). In fish, regulation of Vtg synthesis is known to be under control of 17␤-estrogen (E2 ) and is mediated via cognate nuclear receptors. However, the molecular cues and events that control

the transcriptional expression of vtgr have been less well characterized. The majority of studies performed in fish have focused on observing expression profiles during oocyte develop˜ ment (Barucca et al., 2006; Mananos et al., 1997; Marin and Matozzo, 2004; Sumpter and Jobling, 1995). We recently reported a novel function for insulin (INS) in upregulating vtgr expression in previtellogenic stages of largemouth bass (LBM, Micropterus salmoides) ovarian tissue, whereas the steroid hormones, E2 and 11-ketotestosterone (11-KT), repressed expression (Dominguez et al., 2012). These observations suggest a role for hormones in

 C The Author 2014. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved. For permissions, please email: [email protected]

423

Downloaded from http://toxsci.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014

Control of Transcriptional Repression of the Vitellogenin Receptor Gene in Largemouth Bass (Micropterus Salmoides) by Select Estrogen Receptors Isotypes

424

TOXICOLOGICAL SCIENCES, 2014, Vol. 141, No. 2

MATERIALS AND METHODS Partial LMB vtgr promoter isolation. Genomic DNA (gDNA) was extracted from a pool of four previtellogenic ovaries using the Wizard gDNA Purification Kit (Promega). The 5 flanking region adjacent to the LMB vtgr cDNA was isolated using the GenomeWalker Universal Kit (Clontech). Ligation of the gDNA libraries

to GenomeWalker adaptors was performed following the manufacturer’s instructions. In brief, four ligation reactions were prepared using restriction enzymes (Dra I, EcoR V, Pvu II, and Stu I). Digested and purified gDNA and adaptors were ligated at 16◦ C overnight. Subsequently, two PCR reactions were performed with specific primers designed to exon I of the LMB vtgr cDNA sequence (GenBank HQ32624) (Table 1). PCR cycling conditions were as follows: an initial first stage of seven cycles at 95◦ C for 25 s and 72◦ C for 3 min, a second stage of 32 cycles at 94◦ C for 25 s and 67◦ C for 3 min, and a final stage at 67◦ C for 7 min using a Veriti Thermal Cycler (Applied Biosystems). To continue the nested PCR, primary PCR products were diluted and PCR conditions were set up into three stages: the first stage was seven cycles at 95◦ C for 25 s and 72◦ C for 3 min; the second stage was 32 cycles at 94◦ C for 25 s and 67◦ C for 3 min; and the third stage was 67◦ C for 7 min. All PCR products were separated on 1.5% agarose/EtBr with 1-kb Plus DNA ladder (Invitrogen). Visualized bands were excised and purified using a QIAquick Gel Extraction Kit (Qiagen). Inserts were cloned into a pCR4-TOPO vector (Invitrogen) and sequenced at the University of Florida’s ICBR Core sequencing facility. LMB vtgr promoter analysis. Sequences were analyzed for extension into the coding region of the vtgr gene. Validated sequences upstream of the 5 untranslated region (5’UTR) were analyzed for the presence of transcription factor binding sites using two data bases, TFSearch (http://www.cbrc.jp/research/db/ TFSEARCH.html) and TESS (http://www.cbil.upenn.edu/cgi-bin/ tess/tess) (Table 2). Creation of promoter deletion constructs. The entire fragment obtained (−1,102 bp) was cloned into the enhancer pGL3 luciferase vector (Promega) generating the VtgR-Luc 0 construct. Four deletion constructs of the LMB vtgr promoter were produced by PCR using specific primers with 5 and 3 flanking restriction sites KpnI and HindIII (Promega) (Table 1). The PCR products were ligated into the enhancer pGL3 luciferase vector and designated Vtgr-Luc 1 (−762/+71), Vtgr-Luc 2 (−381/+71), Vtgr-Luc 3 (−267/+71), and Vtgr-Luc 4 (−153/+71). All constructs were verified by sequencing. Promoter activation assays. HEK293 (human embryonic kidney) cells were cultured in Dubelcco’s Modified Eagle Medium (DMEM) supplemented with 10% Fetal bovine serum (FBS), penicillin-streptomycin solution, and L-glutamine. These cells were employed as they are devoid of endogenous Esrs and have served as a model in vitro system for assessing promoter activity of other fish species (Muriach et al., 2014; Rebl et al., 2011; Rebl et al., 2014). HEK cells were seeded at a density of 2 x 105 cells/well in 24-well plates and transfected with 1000 ng of pGL3-empty vector or Vtgr-Luc constructs, and 200 ng of TKrenilla using Fugene 6. After 18 hours, HEK cells were exposed to 10% FBS or 50 ng / ml of porcine insulin (pINS; MP Biologicals). Another set of experiments included 500 ng of pCMV, LMB Esr1, Esr2a, and Esr2b constructs and exposed to 50, 500, 1000nM of E2 (Sigma-Aldrich ) or EE2 (Sigma-Aldrich), and 500 and 1000nM of BPA. All controls received 0.1% ethanol and exposures were performed for 24 h. Luciferase activity was determined as follows: cells were harvested in 100 ␮l of passive lysis buffer, followed by gentle rocking for 20 min at room temperature, centrifuged for 3 min at 12,000 rpm at 4◦ C, and then assayed for transcriptional activation using a Dual Luciferase Reporter Assay System

Downloaded from http://toxsci.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014

regulation of the vtgr gene but the molecular mechanisms controlling these responses, including the involvement of nuclear steroid receptors, have not been defined. To date, information regarding the 5 regulatory region of the vtgr gene and specific transcription factors and upstream signaling pathways that control expression are limited to insects (Cho et al., 2006). In fact, only one putative promoter region of the vtgr gene has been identified in the mosquito (Aedes aegypti), which revealed several transcription factor binding motifs such as AP1, GATA, HNF3, and sites for two ecdysone-responsive early gene proteins, E74 and BR (Cho et al., 2006). In another study investigating virgin alate fire ants (Solenopsis invicta), high levels of vtgr mRNA transcripts correlate with elevated ecdysteroid concentrations, and the insecticide methoprene, an analog to the insect juvenile hormone (JH), was found in vitro to temporally upregulate the levels of vtgr expression (Chen et al., 2004). Although these studies provide valuable information regarding molecular events that are involved in controlling vtgr transcription, the lack of corresponding hormone pathways in fish (i.e., ecdysone and JH) suggests that these regulatory mechanisms are not entirely conserved. Largemouth bass is our model for environmental study because it has a semi-synchronized annual reproductive cycle that allows for monitoring of endocrine biology in discrete windows of the reproductive process (Denslow and Sepulveda, 2007). We have previously cloned the LMB vtgr cDNA and characterized temporal expression levels during oocyte development, revealing maximal expression occurring during primary growth stages (Dominguez et al., 2012), a trend that is consistent with other teleosts (Agulleiro et al., 2007; Davail et al., 1998; Hiramatsu et al., 2004; Luckenbach et al., 2008; Perazzolo et al., 1999). This species has also been extensively employed as a model for studying mechanisms of endocrine disrupting compounds (EDCs), particularly those that have estrogenic properties. These chemicals (xenoestrogens) exert their effects by modulating nuclear estrogen receptors (Esr) and hepatic production of Vtg (Barber et al., 2007; Blum et al., 2008; Garcia-Reyero et al., 2006; Martyniuk et al., 2011). However, the impact these compounds have on vtgr expression has not been examined. Our previous results that show E2 dampens insulin-induced vtgr expression in LMB ovarian tissues support the possibility that control of vtgr expression may be a plausible target of EDCs (Dominguez et al., 2012). The goal of this work was to identify the molecular mechanisms that control vtgr expression in LMB, with a particular focus on Esrs. Here we present the identity and analysis of the first teleost 5 regulatory region of the vtgr gene. Through promoter activation assays we observed a role for select Esr isoforms, Esr1 and Esr2a, in transcriptional repression by the natural ligand E2 . We further revealed the ability of the known xenoestrogen, 17␣ethinylestradiol (EE2 ), to repress vtgr promoter activity via the same receptor subtypes. This response was not observed for an alternate chemical with known estrogenic activity, bisphenol-A (BPA). Finally we provide evidence to suggest this repression may occur through interaction of Esr1 through non-consensus ERE or SP1 DNA binding sites.

DOMINGUEZ ET AL.

425

TABLE 1. Primers Used for GenomeWalker and 5 Deletion Series Primer

Approach

Direction

Sequence 5 →3

GWLMB1 GWLMB2 LMB UP 1 LMB UP 2 LMB UP 3 LMB UP 4 LMB DW 1

GenomeWalking GenomeWalking 5 deletion series 5 deletion series 5 deletion series 5 deletion series 5 deletion series

Reverse Reverse Forward Forward Forward Forward Reverse

ATTAGCATCGGCAGCAGGAGGATCCCCGGT TGCCTGCCTGCCTGCCTTTCTTGCTTGCT CGGGGTACCATCTTCAGGCTGTGT CGTGGTACCGTACAGGGTATTGGGGG CGTGGTACCCAGCAGCACTAACACACC CGTGGTACCGTAGGAAGAGATCAAAAGGCG GCAACCGTACGCTGCGAGAGAAGCTTCG

Binding Site Name

Symbol

Sequence

Position

Scorea

POU class 3 homeobox 2 POU class 1 homeobox 1 Estrogen response element Ccaat enhancer binding protein beta Retinoic X receptor alpha and beta POU class 3 homeobox 2 Activator protein 1 Hepatocyte nuclear factor 1 homeobox 1 Forkhead box A GATA-binding factor 1 Specific protein 1 Nuclear factor of kappa Estrogen response element Specific protein 1 Specific protein 1 Specific protein 1 Hepatocyte nuclear factor 3 alpha GATA-binding factor 1

POU3F2 POU1F1a ERE C/EBPb RXR-a/b POU3F2 AP-1 HNF-1 HNF-3 GATA1 Sp1 NFkappaB ERE Sp1 Sp1 Sp1 HNF-3a GATA1

CATACAAAAT ATCGTC TGACC CTGGGAA GGGGTCA ATTTTCATT TGATTGG GTTAAT GTAAATA GGATAG GGAGGGG GGAATATCC GGTCA GGGGCC ACGCCCC CCGCCT TGTTTGT ACCGGATA

−1165 −1082 −1048 −937 −888 −779 −724 −683 −530 −530 −490 −484 −433 −426 −359 −299 −171 −136

12 14 10 13 14 12 14 12 14 12 14 16 10 12 14 12 12 16

a

La score was calculated using Transcription Element Search System (TESS) web tool.

(Promega) on Synergy H1 microplate reader (Biotek Instruments, Winooski, VT). Electrophoretic mobility shift assay. Electrophoretic mobility shift assay (EMSA) was performed using the Gelshift Chemiluminescent EMSA Kit (Active Motif) following the manufacture’s protocol. In brief, a DNA target probe was designed to include a region of the promoter that contained a half ERE and SP1 site (ACACCGGTCACCGGGGCCGCTCG). The target was labeled with biotin and annealed with its untagged compliment to produce double-stranded DNA probe. Probes were annealed by incubating the complimentary strands at 65o C for 5 min followed by slow cooling. To perform the EMSA, 10 pM of the annealed labeled probes were incubated with supplied buffer, 50 ng/␮l poly D (I-C) and nuclear extract isolated from HEK cells that overexpressed LMB Ers1. Control protein consisted of cells transfected with the corresponding empty expression vector. Some treatments also received unlabeled competitor probe. After 20 min at room temperature, the samples were separated on a 6% acrylamide gel at 100 V for 1 h, transferred to a nylon membrane, and cross-linked. Migration of the biotin labeled probe was detected by chemiluminescence on a Bio-Rad Imager using Quantity One software. Statistical analysis. SigmaPlot version 12.0 (Systat Software Inc., San Jose, CA) software for Windows was used for all statistical

analysis. For the transcriptional regulation of LMB vtgr promoter region, data were first normalized to renilla (luciferase/renilla ratio) and then plotted as fold change compared with the empty vector control. One-way ANOVA with Tukey’s multiple comparison test was used to analyze differences in fold change of luciferase activity between treated and untreated cells after testing for normality and variance homogeneity (p ⬍ 0.05).

RESULTS Cloning of the LMB vtgr Partial Promoter To investigate the transcriptional control of the LMB vtgr gene, we cloned a partial putative promoter region from gDNA using a genome walking approach. Using the previously obtained LMB vtgr cDNA (Dominguez et al., 2012) to design a series of primers (Table 1) we were able to PCR amplify a 1102-bp fragment. Sequencing revealed an overlapping region with the 5 portion of the vtgr cDNA verifying extension upstream of the transcriptional start site and into the regulatory region of the gene (Figs. 1 A and B). Analysis of the LMB vtgr Promoter Analysis of the promoter sequence for transcription factor binding motifs was performed by TESS web tool service. The comprehensive analysis revealed several potential sites described in Table 2. Two ERE (estrogen response element) half-sites were lo-

Downloaded from http://toxsci.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014

TABLE 2. Predicted Transcriptional Factor Binding Sites for the Promoter Region of Vitellogenin Receptor Gene of LMB

426

TOXICOLOGICAL SCIENCES, 2014, Vol. 141, No. 2

cated at −1052 to −1048 and −433 to −429-bp positions from the predicted transcriptional start site. Also, a number of other elements were recognized that have been shown to interact directly and indirectly (through tethering) with Esrs such as GATA1 (−530 to −525; −136 to 129 bp), Sp1 (−490 to −485; −426 to −421 bp; −359 to −353; −299 to −294; −145 to −140 bp), and AP-1 (−724 to −718 bp) (Li et al., 2001; Webb et al., 1999). Additional putative binding motifs that were identified include: POU3F2 (−1165 to 1156 and 779 to 771 bp), C/EBPb (−937 to −931 bp), RXR-a/b (−888 to 882 bp), HNF-1 (−683 to −678 bp), NFkappaB (−484 to −476 bp), and HNF-3 (−530 to −525 and −171 to −165 bp) (Figs. 1 A and B).

Endogenous Activity of the LMB Vtgr Promoter in HEK Cells To investigate the activity of the LMB 5 regulatory region of the vtgr gene, we subcloned the entire promoter fragment (1102 bp) into the enhancer pGL3 luciferase vector (Vtgr-Luc). This construct was transfected into HEK cells and luciferase activity was measured following 24 h of exposure to 10% FBS. Initial results showed no change in luciferase activity (data not shown). We performed similar experiments with a 5 deletion series of the LMB vtgr promoter (Fig. 2A) in order to determine if loss of certain DNA motifs altered activity. We tested the functionality of four constructs in response to 10% FBS. Our results determined that luciferase activity was significantly increased for all four constructs. VTG-Luc-1 showed the highest response to 10% FBS

(Fig. 2B) resulting in 18-fold increase in activity compared with control. Vtgr-Luc 2 had a significant (28%) loss in luciferase activity compared with Vtgr-Luc 1. Further deletions represented by Vtgr-Luc 3 and Vtgr-Luc 4 did not result in any additional loss of activity compared with Vtgr-Luc 2. Transcriptional Activation of the LMB Vtgr Promoter by Insulin and FBS Because results from our previous study determined that INS induced the expression of vtgr mRNA in ovarian tissues ex vivo (Dominguez et al., 2012), we explored the influence of this hormone on promoter activity. HEK cells were transfected with the VTGR-Luc 1, 2, 3, and 4 constructs and exposed to 50-ng/ml INS. Results from these experiments failed to show enhanced activity over basal levels for any of the constructs or doses tested. We also performed a set of experiments with 10% FBS for each construct which also failed to elicit any statistical modulation in promoter activity compared with each respective construct control (Supplementary fig. 1). LMB Vtgr Promoter Activity Is Suppressed by E2 Via Esr1 and Esr2a To determine whether E2 would alter promoter activity, experiments were performed with HEK cells transfected with the various VTGR-Luc constructs and LMB Esr isotypes (Esr1, Esr2a, and Esr2b) and exposed to 500nM of E2 for 24 h. Data showed that a significant repression in luciferase activity was observed in E2 exposed cells containing Esr1 or Esr2a (Fig. 3) but not Esr2b. This

Downloaded from http://toxsci.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014

FIG. 1. Analysis of the largemouth bass vtgr gene promoter region. (A) The 5 UTR sequence of the cloned vtgr promoter is numbered relative to the transcription start site (+1). Several putative DNA response elements are shown in color or underlined. (B) Illustration of the cloned LMB vtgr promoter region labeled with identified putative transcription factor binding sites.

DOMINGUEZ ET AL.

427

FIG. 3. Analysis of LMB Vtgr-Luc activity in HEK cells exposed to E2 . HEK cells were transfected with (A) VtgR-Luc 1, (B) VtgR-Luc 2, (C) VtgR-Luc 3, or (D) VtgR-Luc4 constructs in combination with LMB Esr1, Esr2a, or Esr2b and exposed to E2. All data are graphed as mean ± SEM of fold change in luciferase activity compared with vehicle control. Significant differences among treatments compared with control are represented by letters (p ⬍ 0.05).

decrease in activity was most dramatic for Vtgr-Luc 1 in the presence of Esr1 which resulted in a 77.2% loss in activity compared with unexposed control cells. Esr2a caused a 41.6% decrease in activity. The presence of Esrb2 had no impact on promoter activity for any of the VTGR-Luc constructs tested.

LMB vtgr Promoter Activity is Suppressed by EE2 but not BPA via Esr1 and Esr2a To assess the impact of xenoestrogens on vtgr promoter activity, HEK cells were transfected with the distinct Esrs and Vtgr-Luc 1 and exposed to various doses of E2 , EE2, and BPA. Results from these experiments reveal that EE2 caused a similar decrease in Vtgr Luc 1 activity compared with E2 that was dependent primar-

ily on Ers1, although a significant but smaller amount of repression was observed for Esr2a. In contrast, both doses of BPA failed to modulate promoter activation in any of the experiments (Figs. 4 A–C).

Electrophoretic Mobility Shift Assay To begin to hone in on potential DNA binding sites that ESR1 may interact with in suppressing vtgr promoter activity, we designed a probe containing a half ERE and SP1 site. Using nuclear cell extract that contained an overexpressed LMB Esr1 protein, we were able to observe a shift compared with cells that were incubated with nuclear extract that did not contain the LMB Esr1 (Fig. 5). In addition, we were able to negate this shift with an unlabeled

Downloaded from http://toxsci.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014

FIG. 2. Construction and activation of 5 vtgr promoter deletion series. (A) Representation of different deletion luciferase constructs showing each fragment size with respect to the transcriptional start site. (B) HEK cells were transfected with different Vtgr-Luc constructs or pGL3 (empty vector control ) in combination with TK-renilla and exposed to 10% FBS for 24 h. All data are graphed as mean ± SEM of fold change in luciferase activity compared with control. Significant differences among treatments compared with control are represented by letters (p ⬍ 0.05).

428

TOXICOLOGICAL SCIENCES, 2014, Vol. 141, No. 2

FIG. 4. Analysis of LMB Vtgr-Luc activity in HEK cells exposed to EE2 and BPA. HEK cells were transfected with VtgR-Luc 1 construct in combination with Esr1 (A), Esr2a (B), and Esr2b (C), and exposed to different doses of E2 , EE2 , and BPA. Control VtgR-Luc 1 was exposed to ethanol. All data are graphed as mean ± SEM of control luciferase activity. Significant differences (p ⬍ 0.05) among treatments from control are represented by letters.

competitor. These results imply the LMB Esr1 is interacting with the DNA probe containing the ERE and SP1 sites.

DISCUSSION The goal of this research was to investigate hormonal signals and downstream molecular events that contribute to the transcriptional regulation of the vtgr gene in LMB. Parallel trends in expression of esr subtypes and vtgr during early follicle growth combined with the observed E2 -mediated repression of vtgr mRNA in ovarian tissues previously reported by our group led us to investigate the regulatory link between sex steroids and vtgr expression (Dominguez et al., 2012; Sabo-Attwood et al., 2007). To clarify the molecular events controlling vtgr expression, a portion of the LMB 5 regulatory sequence was obtained from the vtgr gene. This is the first vertebrate vtgr promoter region identified to date. Analysis of this sequence for DNA binding motifs revealed the presence of putative ERE half-sites (−1052 to −1048 and −433 to −429 bp) and response elements for GATA1 (−530 to −525; −136 to 129 bp), Sp1 (−490 to −485; −426 to −421 bp; −359 to −353; −299 to −294; −145 to −140 bp), AP-1 (−724 to −718 bp), POU3F2 (−1165 to 1156 and 779 to 771 bp), C/EBPb (−937 to −931 bp), RXR-a/b (−888 to 882 bp), HNF-1 (−683 to −678 bp), NFkappaB (−484 to −476 bp), and HNF-3 (−530 to

−525 and −171 to −165 bp) transcription factors. A number of studies confirm that Esrs can interact with both ERE half sites and non-consensus sites present in a number of gene promoters such as insulin receptor (Garc´ıa-Arencibia et al., 2005), prolactin (Anderson and Gorski, 2000), and progesterone receptor A (Petz and Nardulli, 2000) and that these sites are also present in a number of gene promoters cloned from fish species (Cheshenko et al., 2008; Wang et al., 2013). The only other species for which a 5 regulatory region of the Vtgr gene has been identified belongs to the mosquito (Aedes aegypti) (Cho et al., 2006). Despite divergent endocrine systems between fish and insects (i.e., ecdysone), a number of motifs were conserved including GATA1, AP-1, and HNF-3. Furthermore, the placement of these sites was similar for the two promoters. Although activation of the mosquito promoter specifically by select transcription factors has not been assessed, recombinant human GATA and HNF3 proteins were shown to bind regions of the sequence through electromobility shift assay (EMSA). This group further revealed motifs specific to ecdysone signaling and highlighted their importance in vtgr gene regulation. Although ecdysone signaling does not exist in vertebrates, it has been theorized that vertebrate Esrs evolved from ecdysone receptors present in invertebrates (Thornton, 2001; Thornton et al., 2003). Our interest was to test the activation of the promoter in response to hormones and sex steroids because our previous work showed INS increased vtgr mRNA and E2 repressed this upregulation in LMB ex vivo ovarian tissues. The presence of multiple DNA binding motifs known to be targets of downstream INS and E2 signaling, such as GATA and ERE half sites, respectively, also

Downloaded from http://toxsci.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014

FIG. 5. Binding of ESR1 to an ERE half site and/or SP1 site of the LMB vtgr promoter region. (A) DNA probe containing a half ERE and SP1 site. (B) EMSA was resolved in 6% acrylamide gel. Lanes: (1) probe only. (2) Probe + control NE. (3) Probe + control NE + ULC. (4) Probe + ESR1 NE. (5) Probe + ESR1 NE + ULC. (6) Probe + ULC. Abbreviations: ESR1: estrogen receptor 1; NE: nuclear extract; ULC: unlabeled competitor probe.

DOMINGUEZ ET AL.

been shown to function directly or indirectly as regulatory binding sites for ESRs (Webb et al., 1999). Furthermore, ESRs can be inhibitory when acting through AP-1 or via cross-talk with GATA pathways (Cherlet and Murphy, 2007; Holth et al., 1997; Qi et al., 2004). The additional repression of promoter activity by Esr2a for the deletion constructs suggests that interactions with Sp-1 sites (present in Vtgr-Luc 2, Vtgr-Luc 3, and Vtgr-Luc 4) may be preferential targets for this receptor subtype. Our evaluation of Esrs direct binding to the promoter fragment through EMSA suggests that Esr1 participates in the downregulation of vtgr by interacting with SP1 and ERE half sites. This result is consistent with the downregulation of vtgr mRNA by E2 observed in ovarian tissues ex vivo (Dominguez et al., 2012). These data emphasize a functional and valuable role for Esrs in early stages of reproduction in teleosts, highlighting their involvement in controlling vtgr expression. Numerous chemicals that bind Esrs have been shown to impair fish reproduction in females through ovarian malformations and production of poorly developed eggs. Little is known regarding the specific functions and downstream targets of Esrs in the ovary; however, based on our collective observations that E2 downregulates vtgr mRNA in ovarian tissues and that Esrs repress vtgr promoter activity supported investigations to determine whether the select xenoestogens, EE2 , and BPA could modulate vtgr promoter activity in a manner similar to E2 . Overall, repression of vtgr promoter activity was similarly influenced by E2 and EE2 , showing this response was modulated primarily by Esr1. The fact that the repression profiles were similar is consistent with the relatively high binding affinity that each compound has for the Esrs. Furthermore, both E2 and EE2 have slightly greater binding affinity for the Esr1 subtype compared with Esr2a and Esr2b as described in previous fish studies (Gale et al., 2004; Passos et al., 2009). However, Esr binding studies performed in fish and other species generally agree that the relative binding affinity (RBA) for EE2 can be greater (range of 89–647) than E2 (100)(Denny et al., 2005; Gale et al., 2004; Passos et al., 2009; Shyu et al., 2011); therefore, it is somewhat surprising that EE2 does not produce a greater response (repression of vtgr-Luc activity) based on binding affinity alone. In addition to RBA, there are a number of other factors in play including saturation of the Esrs, differential recruitment of co-regulatory proteins, and cross-talk between the Esrs and other pathways that likely contribute to the observed results. In contrast, BPA failed to modulate vtgr promoter activity with any of the Esrs. Unlike EE2 , BPA binds the Esrs with weak RBA comparably and may explain the lack of response observed (Kuiper et al., 1997; Molina-Molina et al., 2013; Passos et al., 2009). Although we have not characterized exact binding affinities for the compounds employed in these studies, the calculated RBA for BPA is approximately 1.3 for the Esr1 of Gilthead seabream (Passos et al., 2009). In this case, the weak RBA of BPA for the Esrs likely contributes to the observed results in the current study where E2 and EE2 repress activity of the vtgr promoter and BPA does not. In addition, it is known that ligands can induce conformational changes upon binding the Esrs that lead to differential transcriptional complex formation. It is proposed that these complexes may be directed to different suites of DNA response elements in a ligand-dependent manner. BPA differentially recruits co-regulatory proteins to the Esrs and it has recently been shown to be the case for SRC-3 by our group (a known co-accessory protein, Sabo-Attwood, unpublished data). The precise role that the Esrs play in the ovary in response to E2 and xenoestrogens is not well known. Based on previous literature it is well established that long-term exposure of

Downloaded from http://toxsci.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014

suggests that these factors may influence vtgr transcription at the level of the promoter. To validate that the promoter was functional, the activity of the various deletion constructs in response to 10% FBS was tested. Surprisingly, there was no induction of activity with the longest fragment (Vtgr-Luc 0 −1020 bp) but there was a significant modulation in activity observed for the remaining shorter constructs. Perhaps the longer sequence contains DNA elements that provide a mechanism for robust inhibitory effects on transcription by INS or other factors present in FBS. Also, this 5 region (−829 bp to −1020 bp) may hold valuable regulatory information which requires further study. For the remaining constructs, in general, the shorter sequences resulted in less overall activity compared with the longer sequence contained in the Vtgr-Luc 1 construct. By comparing the loss in activity with the deletion constructs with the DNA binding motifs present in those regions, it is not unreasonable to speculate which transcription factors are involved in modulating activity. For example, removal of binding sites present in Vtgr-Luc 1 for AP-1, Sp1, ERE, GATA-1, and NFkappaB resulted in a significant loss of luciferase activity for Vtgr-Luc 2, Vtgr-Luc 3, and Vtgr-Luc 4 in HEK cells. Because HEK cells were stimulated with serum, it is difficult to decipher which factors are precisely responsible for driving this activity but it is possible to consider downstream pathways that may be involved by the motifs identified. The inactivation of the largest sequence is perplexing and further studies are required to examine the factors contributing to this response in greater detail. A number of the DNA binding motifs identified are known to interact with transcription factors that are targets of INS and E2 signaling pathways. To determine whether these agents had the capacity to modulate vtgr promoter activity we first examined whether INS could enhance activation of the LMB vtgr promoter. It was surprising that INS had no effect on promoter activity; particularly because an upregulation of vtgr mRNA by INS in ex vivo ovarian tissues of LMB was previously observed. Due to the complexity of promoter regulation, there are numerous reasons that could explain these results such as the requirement of additional response elements, repression of select motifs, or the need for multiple hormones and growth factors to work in concert. Extending the promoter sequence may offer additional detail pertaining to INS regulation at the level of the vtgr promoter in LMB. As previously revealed, a role for E2 in repressing vtgr expression ex vivo suggests that Esrs are involved in the transcriptional control of this gene. Using the same Vtgr-Luc constructs the influence of each specific Esr isoform on promoter activity in response to the natural ligand E2 was assessed. One of the most novel observations of these experiments is the differential suppression of promoter activity by the Esr subtypes. Both Esr1 and Esr2a caused significant repression of Vtgr-Luc 1; however, the response was greatest for the former receptor. Esr2b had neither an effect on promoter activity nor an influence on Esr2a activity in co-transfection studies. Based on our knowledge of motifs present in each promoter construct, it is apparent that both Esr1 and Esr2a are likely repressing activity through sites present in the 5’ region of the Vtgr-Luc 1 sequence. This region does indeed contain ERE half sites that may be direct targets of Esrs as has been previously shown in other studies (Anderson and Gorski, 2000; Petz and Nardulli, 2000); however, it is now realized that these receptors can indirectly associate with DNA elements through tethering with other transcription factors. A number of identified motifs including GATA-1, AP-1, Sp1, C/EBP, NFkappaB, and HNF-1 have

429

430

TOXICOLOGICAL SCIENCES, 2014, Vol. 141, No. 2

SUPPLEMENTARY DATA Supplementary data are available online at http://toxsci. oxfordjournals.org/.

FUNDING National Institute of Environmental Health Sciences (RO1 ES015449 to N.D.); University of Florida (Project # 45838 to T.S.) and South Carolina Research Foundations (Project # 53002 to T.S.) .

ACKNOWLEDGMENTS The authors thank Dr Melinda Prucha for help in analyzing the promoter region. We also thank Cody Smith for assisting with cell culture care for this study.

REFERENCES ´ M., Morais, S., Cerda, ` J. and Babin, P. J. Agulleiro, M. J., Andre, (2007). High transcript level of fatty acid-binding protein 11 but not of very low-density lipoprotein receptor is correlated to ovarian follicle atresia in a teleost fish (Solea senegalensis). Biol. Reprod. 77, 504–516. Anderson, I. and Gorski, J. (2000). Estrogen receptor ␣ interaction with estrogen response element half-sites from the rat prolactin gene. Biochemistry 39, 3842–3847. Barber, D. S., McNally, A. J., Garcia-Reyero, N. and Denslow, N. D. (2007). Exposure to p,p -DDE or dieldrin during the repro-

ductive season alters hepatic CYP expression in largemouth bass (Micropterus salmoides). Aquat. Toxicol. 81, 27–35. Barucca, M., Canapa, A., Olmo, E. and Regoli, F. (2006). Analysis of vitellogenin gene induction as a valuable biomarker of estrogenic exposure in various Mediterranean fish species. Environ. Res. 101, 68–73. ´ Blazquez, M., Zanuy, S., Carrillo, M. and Piferrer, F. (1998). Structural and functional effects of early exposure to estradiol-17␤ and 17␣-ethynylestradiol on the gonads of the gonochoristic teleost Dicentrarchus labrax. Fish Physiol. Biochem. 18, 37–47. Blum, J. L., Nyagode, B. A., James, M. O. and Denslow, N. D. (2008). Effects of the pesticide methoxychlor on gene expression in the liver and testes of the male largemouth bass (Micropterus salmoides). Aquat. Toxicol. 86, 459–469. Chen, M. E., Lewis, D. K., Keeley, L. L. and Pietrantonio, P. V. (2004). cDNA cloning and transcriptional regulation of the vitellogenin receptor from the imported fire ant, Solenopsis invicta Buren (Hymenoptera: Formicidae). Insect Mol. Biol. 13, 195–204. Cherlet, T. and Murphy, L. C. (2007). Estrogen receptors inhibit Smad3 transcriptional activity through Ap-1 transcription factors. Mol. Cell. Biochem. 306, 33–42. Cheshenko, K., Pakdel, F., Segner, H., Kah, O. and Eggen, R. I. L. (2008). Interference of endocrine disrupting chemicals with aromatase CYP19 expression or activity, and consequences for reproduction of teleost fish. Gen. Comp. Endocrinol. 155, 31– 62. Cho, K.-H., Cheon, H.-M., Kokoza, V. and Raikhel, A. S. (2006). Regulatory region of the vitellogenin receptor gene sufficient for high-level, germ line cell-specific ovarian expression in transgenic Aedes aegypti mosquitoes. Insect Biochem. Mol. Biol. 36, 273–281. Davail, B., Pakdel, F., Bujo, H., Perazzolo, L. M., Waclawek, M., Schneider, W. J. and Le Menn, F. (1998). Evolution of oogenesis: The receptor for vitellogenin from the rainbow trout. J. Lipid Res. 39, 1929–1937. Denny, J. S., Tapper, M. A., Schmieder, P. K., Hornung, M. W., Jensen, K. M., Ankley, G. T. and Henry, T. R. (2005). Comparison of relative binding affinities of endocrine active compounds to fathead minnow and rainbow trout estrogen receptors. Environ. Toxicol. Chem. 24, 2948–2953. Denslow, N. D. and Sepulveda, M. (2007). Ecotoxicological effects on fish reproduction. In The Fish Oocyte: From Basic Studies to Biotechnological Applications (P.J. Babin, J. Cerda and E. Lubzens, Eds.), Springer, Dordrecht, The Netherlands, 255– 322. Dominguez, G. A., Quattro, J. M., Denslow, N. D., Kroll, K. J., Prucha, M. S., Porak, W. F., Grier, H. J. and Sabo-Attwood, T. L. (2012). Identification and transcriptional modulation of the largemouth bass, Micropterus salmoides, vitellogenin receptor during oocyte development by insulin and sex steroids. Biol. Reprod. 87, 1–12. ˜ R. and Maule, A. G. (2004). Interaction of xenoGale, W. L., Patino, biotics with estrogen receptors ␣ and ␤ and a putative plasma sex hormone-binding globulin from channel catfish (Ictalurus punctatus). Gen. Comp. Endocrinol. 136, 338–345. ´ ´ J., Carmen Carranza, N., Campion, Garc´ıa-Arencibia, M., Davila, M. and Calle, C. (2005). Identification of two functional estrogen response elements complexed with AP-1-like sites in the human insulin receptor gene promoter. J. Steroid Biochem. Mol. Biol. 94, 1–14. Garcia-Reyero, N., Barber, D. S., Gross, T. S., Johnson, K. G., Sepulveda, M. S., Szabo, N. J. and Denslow, N. D. (2006). Di´ etary exposure of largemouth bass to OCPs changes expres-

Downloaded from http://toxsci.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014

male fish to E2 or xenoestrogens can result in the formation of ovotestis/intersex. However, the majority of these oocytes are of primary growth stage (possible cortical alveoli) and do not contain Vtg, despite the increases in circulating plasma Vtg in these ´ exposed fish (Blazquez et al., 1998; Lange et al., 2011; Lange et al., 2009). Although estrogens may control hepatic Vtg synthesis and perhaps early oocyte growth, the mechanisms controlling Vtg uptake are poorly understood. In addition to active uptake of Vtg by the Vtgr, there are also issues of cell patency which seem to be required before uptake can occur (in normal females). As we have shown previously in ex vivo studies (Dominguez et al., 2012), there is likely complex cross-talk that involves insulin and estrogens in regulating, in part, the acquisition of Vtg into developing oocytes. Recently, microarray analysis studies performed by our group have also highlighted alternate gene networks likely involved in distinct stages of oocyte development in LMB, including B-cell and T-cell receptor-mediated signaling cascades and fibronectin regulation (Martyniuk et al., 2013). These latter studies reveal the series of complex signaling pathways involved in normal oocyte development. The outcomes of this study reveal a significant and novel role for fish Esr subtypes in regulation of the vtgr gene. Through cloning of the 5 regulatory region of the LMB vtgr gene, this is the first report showing the ability of Esr1 and Esr2a to repress activity of this region, highlighting DNA motifs for future study that potentially include half ERE and SP1 sites. Furthermore the vtgr gene is exposed as a potential target of EDCs, such as EE2 , that are associated with adverse effects on reproduction and development in fish and other species. These observations have importance in elucidating significant aspects of endocrine biology in teleosts and expose the impact that EDCs can have on molecular targets in reproduction.

DOMINGUEZ ET AL.

(2009). A yeast assay based on the gilthead sea bream (teleost fish) estrogen receptor ␤ for monitoring estrogen mimics. Ecotoxicol. Environ. Saf. 72, 1529–1537. Perazzolo, L. M., Coward, K., Davail, B., Normand, E., Tyler, C. R., Pakdel, F., Schneider, W. J. and Le Menn, F. (1999). Expression and localization of messenger ribonucleic acid for the vitellogenin receptor in ovarian follicles hroughout oogenesis in the rainbow trout, Oncorhynchus mykiss. Biol. Reprod. 60, 1057– 1068. Petz, L. and Nardulli, A. (2000). Sp1 binding sites and an estrogen response element half-site are involved in regulation of the human progesterone receptor A promoter. Mol. Endocrinol. 14, 972–985. Qi, X. M., Borowicz, S., Pramanik, R., Schultz, R. M., Han, J. H. and Chen, G. (2004). Estrogen receptor inhibits c-Jun-dependent stress-induced cell death by binding and modifying c-Jun activity in human breast cancer cells. J. Biol. Chem. 279, 6769– 6777. ´ r, T., Goldammer, T. and Seyfert, H.Rebl, A., Rebl, H., Korytaˇ M. (2014). The proximal promoter of a novel interleukin-8encoding gene in rainbow trout (Oncorhynchus mykiss) is strongly induced by CEBPA, but not NF-␬B p65. Dev. Comp. Immunol. 46, 155–164. Rebl, A., Rebl, H., Liu, S., Goldammer, T. and Seyfert, H.-M. (2011). Salmonid Tollip and MyD88 factors can functionally replace their mammalian orthologues in TLR-mediated trout SAA promoter activation. Dev. Comp. Immunol. 35, 81–87. Sabo-Attwood, T., Blum, J. L., Kroll, K. J., Patel, V., Birkholz, D., Szabo, N. J., Fisher, S. Z., McKenna, R., Campbell-Thompson, M. and Denslow, N. D. (2007). Distinct expression and activity profiles of largemouth bass (Micropterus salmoides) estrogen receptors in response to estradiol and nonylphenol. J. Mol. Endocrinol. 39, 223–237. Shyu, C., Cavileer, T. D., Nagler, J. J. and Ytreberg, F. M. (2011). Computational estimation of rainbow trout estrogen receptor binding affinities for environmental estrogens. Toxicol. Appl. Pharmacol. 250, 322–326. Sumpter, J. P. and Jobling, S. (1995). Vitellogenesis as a biomarker for estrogenic contamination of the aquatic environment. Environ. Health Perspect. 103, 173–178. Thornton, J. W. (2001). Evolution of vertebrate steroid receptors from an ancestral estrogen receptor by ligand exploitation and serial genome expansions. Proc. Natl Acad. Sci. U.S.A. 98, 5671–5676. Thornton, J. W., Need, E. and Crews, D. (2003). Resurrecting the ancestral steroid receptor: ancient origin of estrogen signaling. Science 301, 1714–1717. Wang, Q., Sham, K. W. Y., Ogawa, S., Li, S., Parhar, I. S., Cheng, C. H. K., Liu, X. and Lin, H. (2013). Regulation of the two kiss promoters in goldfish (Carassius auratus) by estrogen via different ER␣ pathways. Mol. Cell. Endocrinol. 375, 130–139. Webb, P., Nguyen, P., Valentine, C., Lopez, G. N., Kwok, G. R., McInerney, E., Katzenellenbogen, B. S., Enmark, E., Gustafsson, J.˚A., Nilsson, S., et al. (1999). The estrogen receptor enhances AP-1 activity by two distinct mechanisms with different requirements for receptor transactivation functions. Mol. Endocrinol. 13, 1672–1685.

Downloaded from http://toxsci.oxfordjournals.org/ at University of California, San Francisco on December 19, 2014

sion of genes important for reproduction. Aquat. Toxicol. 78, 358–369. Hiramatsu, N., Chapman, R. W., Lindzey, J. K., Haynes, M. R. and Sullivan, C. V. (2004). Molecular characterization and expression of vitellogenin receptor from white perch (Morone americana). Biol. Reprod. 70, 1720–1730. Holth, L. T., Sun, J. M., Coutts, A. S., Murphy, L. C. and Davie, J. R. (1997). Estrogen receptor diminishes DNA-binding activities of chicken GATA-1 and CACCC-binding proteins. DNA Cell Biol. 16, 1477–1482. Kuiper, G. G. J. M., Carlsson, B., Grandien, K., Enmark, E., Haggblad, J., Nilsson, S. and Gustafsson, J.-a. (1997). Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors ␣ and ␤. Endocrinology 138, 863–870. Lange, A., Paull, G. C., Coe, T. S., Katsu, Y., Urushitani, H., Iguchi, T. and Tyler, C. R. (2009). Sexual reprogramming and estrogenic sensitization in wild fish exposed to ethinylestradiol. Environ. Sci. Technol. 43, 1219–1225. Lange, A., Paull, G. C., Hamilton, P. B., Iguchi, T. and Tyler, C. R. (2011). Implications of persistent exposure to treated wastewater effluent for breeding in wild roach (Rutilus rutilus) populations. Environ. Sci. Technol. 45, 1673–1679. Li, C., Briggs, M. R., Ahlborn, T. E., Kraemer, F. B. and Liu, J. (2001) Requirement of Sp1 and estrogen receptor alpha interaction in 17beta-estradiol-mediated transcriptional activation of the low density lipoprotein receptor gene expression. Endocrinology 142, 1546–1553. Luckenbach, J., Iliev, D., Goetz, F. and Swanson, P. (2008). Identification of differentially expressed ovarian genes during primary and early secondary oocyte growth in coho salmon, Oncorhynchus kisutch. Reprod. Biol. Endocrinol. 6, 2. ˜ ˜ Mananos, E. L., Nu´ nez Rodr´ıguez, J., Le Menn, F., Zanuy, S. and Carrillo, M. (1997). Identification of vitellogenin receptors in the ovary of a teleost fish, the Mediterranean sea bass (Dicentrarchus labrax). Reprod. Nutr. Dev. 37, 51-61. Marin, M. G. and Matozzo, V. (2004). Vitellogenin induction as a biomarker of exposure to estrogenic compounds in aquatic environments. Mar. Pollut. Bull. 48, 835–839. Martyniuk, C. J., Prucha, M. S., Doperalski, N. J., Antczak, P., Kroll, K. J., Falciani, F., Barber, D. S. and Denslow, N. D. (2013). Gene expression networks underlying ovarian development in wild largemouth bass (Micropterus salmoides). PLoS ONE 8, e59093. Martyniuk, C. J., Spade, D. J., Blum, J. L., Kroll, K. J. and Denslow, N. D. (2011). Methoxychlor affects multiple hormone signaling pathways in the largemouth bass (Micropterus salmoides) liver. Aquat. Toxicol. 101, 483–492. ´ Molina-Molina, J.-M., Amaya, E., Grimaldi, M., Saenz, J.-M., Real, ´ M., Fernandez, M. F., Balaguer, P. and Olea, N. (2013). In vitro study on the agonistic and antagonistic activities of bisphenol-S and other bisphenol-A congeners and derivatives via nuclear receptors. Toxicol. Appl. Pharmacol. 272, 127– 136. ´ Muriach, B., Carrillo, M., Zanuy, S. and Cerda-Reverter, J. (2014). Characterization of sea bass FSH␤ 5 flanking region: Transcriptional control by 17␤-estradiol. Fish Physiol. Biochem. 40, 849–864. Passos, A. L. S., Pinto, P. I. S., Power, D. M. and Canario, A. V. M.

431