Identification of a cDNA Encoding a Retinoid X Receptor Homologue from Schistosoma mansoni

Schistosoma mansoni, a multicelluar eukaryotic blood fluke, is a major cause of morbidity worldwide in humans. The study of female parasite growth, development, and gene regulation is important because the eggs produced are responsible for the pathogenesis observed in schistosomiasis. p14, an eggshell precursor gene expressed only in sexually mature females in response to a male stimulus, is a model for female-specific gene regulation. The upstream region of the p14 gene shares sequences present in insect genes known to be regulated in a sex-, temporal-, and tissue-specific manner by members of the steroid receptor superfamily. Herein, we report the identification and characterization of a cDNA that encodes the S. mansoni(Sm) RXR homologue. Sequence analysis predicts and Western blot analysis confirms the synthesis of a 74-kDa protein, the largest member of the RXR family reported to date. We show by electrophoretic mobility shift assay analysis that SmRXR binds tocis–elements of the p14 gene including a direct repeat that follows the “3-4-5” rule of binding elements recognized by members of the steroid receptor superfamily. Furthermore, we demonstrate that SmRXR can act as a transcription activator in the yeast one-hybrid system. Through quantitative reverse transcriptase-polymerase chain reaction, we show that theSmRXR gene is constitutively expressed and thus must play multiple roles throughout the schistosome life cycle.

Schistosoma mansoni, a multicelluar eukaryotic blood fluke, is a major cause of morbidity worldwide in humans. The study of female parasite growth, development, and gene regulation is important because the eggs produced are responsible for the pathogenesis observed in schistosomiasis. p14, an eggshell precursor gene expressed only in sexually mature females in response to a male stimulus, is a model for female-specific gene regulation. The upstream region of the p14 gene shares sequences present in insect genes known to be regulated in a sex-, temporal-, and tissue-specific manner by members of the steroid receptor superfamily. Herein, we report the identification and characterization of a cDNA that encodes the S. mansoni (Sm) RXR homologue. Sequence analysis predicts and Western blot analysis confirms the synthesis of a 74-kDa protein, the largest member of the RXR family reported to date. We show by electrophoretic mobility shift assay analysis that SmRXR binds to cis-elements of the p14 gene including a direct repeat that follows the "3-4-5" rule of binding elements recognized by members of the steroid receptor superfamily. Furthermore, we demonstrate that SmRXR can act as a transcription activator in the yeast onehybrid system. Through quantitative reverse transcriptase-polymerase chain reaction, we show that the SmRXR gene is constitutively expressed and thus must play multiple roles throughout the schistosome life cycle.
Schistosomiasis is a major public health problem afflicting 300 million people in 76 countries (1). Egg production is not only responsible for dissemination of the parasite but also for pathology of the disease in humans. Therefore, studies of female schistosome reproductive development and egg production may define targets useful in disease control. Interestingly, female reproductive development is regulated intimately by a stimulus from the male parasite (2). For many years it has been known that female schistosomes from single sex infections are stunted in size and sexually immature (2). This observation indicates that schistosomes have an intriguing developmental system requiring signaling from the male schistosome leading to direct or indirect activation of a number of female-specific genes (2)(3)(4)(5)(6). One gene, p14, encodes an eggshell precursor expressed only in vitelline cells of mature females in response to male stimulus (3,7,8). The region upstream of transcription initiation of p14 contains elements similar to those found in Drosophila and silkmoths that are known to regulate chorion gene expression in a sex-, tissue-, and temporal-specific manner (2,7,9,10).
Ultraspiracle protein, a member of the RXR 1 subfamily of the steroid receptor superfamily, binds to the s15 chorion gene of Drosophila recognizing an imperfect palindrome containing the core sequence TCACGT (11,12). Members of the steroid receptor superfamily are likely to play a role in female-specific gene regulation of schistosomes. A similar sequence is present as part of an imperfect palindrome in the upstream region of p14 (2) (Fig. 1). Furthermore, a direct repeat that follows the "3-4-5" rule of steroid response elements is present in the p14 upstream region (13,14). This latter sequence consists of two identical half-sites, AACTATCA, spaced by five nucleotides (Fig. 1). Presence of these cis-elements in the p14 gene led to a search for schistosome homologues of steroid hormone receptors. Herein, we report the identification and characterization of Schistosoma mansoni RXR cDNA (SmRXR), and we provide evidence for a possible role in regulation of female-specific gene expression.

EXPERIMENTAL PROCEDURES
Identification of SmRXR cDNA-Degenerate oligonucleotides representing the conserved regions of RXR genes with the amino acid sequences, CEGCKGFF (aa 288 -295) and GMKKEAVQEE (aa 335-344), were synthesized for use in PCR (Fig. 2). Using a cDNA pool derived from schistosome worm pair mRNA as a template and the RXR genespecific primers, a 170-base pair fragment was amplified. This PCR product was cloned into TOPO TA vector (Invitrogen), sequenced, and shown to be related to RXRs by NCBI Blast search analysis. The DNA fragment homologous to RXRs was radiolabeled with [␣-32 P]dCTP by random priming and used as a probe to screen a female adult worm ZAP cDNA library. Seven positive clones were identified, the phagemids were excised, and then the cDNA inserts were shown to be identical in sequence. Amino acid comparison of SmRXR to Drosophila USP and human RXR␣ was performed using the Pile-Up program from the Wisconsin Package version 9. Phylogeny trees were constructed using the PAUPsearch and PAUPdisplay program from the Wisconsin Package version 9. The following parameters were used in construction of trees to confirm accuracy of the study: heuristic with parsimony, * This work was supported by National Institutes of Health Grant AI27219. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM  branch-and-bound with parsimony, branch-and-bound with minimal evolution, and neighbor joining.
Expression and Production of Recombinant SmRXR-A fragment of SmRXR encoding amino acids 200 -561 was inserted into pGEX 2T-1 to form pGEX-RXR. TOP 10 competent cells (Invitrogen) were transformed with pGEX-RXR. Transformed cells were grown to 0.45 A 600 at 37°C and then cooled on ice for 20 min. After the addition of 0.5 mM isopropyl-1-thio-␤-D-galactopyranoside, cells were incubated overnight at 20°C. The cells were pelleted by centrifugation at 3000 ϫ g for 20 min. Cells were resuspended in lysis buffer (25 mM Tris, pH 8, 1 mM EDTA, 10% glycerol, 50 mM NaCl, 0.1% sodium deoxycholate, 0.01% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 10 M leupeptin and 75 M aprotinin), sonicated, and the supernatant fraction collected after centrifugation at 18,000 ϫ g for 1 h at 4°C. The GST-SmRXR fusion protein in the S18 fraction was purified by passage over a glutathione-Sepharose column pre-equilibrated with G-S binding buffer (25 mM Tris, pH 8, 1 mM EDTA, 50 mM NaCl, 10% glycerol). GST-SmRXR was eluted by addition of 10 mM reduced glutathione in G-S binding buffer.
Immunological Techniques-Rabbit anti-RXR was purified from serum of a rabbit immunized with recombinant GST-SmRXR by the method of Beal and Mitchell (15). Extracts from adult worm pairs, mature male worms, mature female worms, and 3-h schistosomules were loaded onto a 12.5% SDS-acrylamide gel in triplicate panels. One panel was stained with Coomassie Blue, and the other two panels were transferred onto polyvinylidene difluoride membrane (Bio-Rad) for Western analysis. One panel was probed with preimmunized rabbit serum (1:5000 dilution) and the other panel with affinity purified rabbit anti-RXR antibody (2 g/ml).
Quantification of SmRXR mRNA by RT-PCR-RNA was extracted from the following schistosome developmental stages: 45-day worms, 44-day female (bisex infection), 44-day male (bisex infection), 44-day female (single sex infection), 44-day male (single sex infection), 35-day worms, 32-day worms, 30-day worms, 28-day worms, 21-day, 15-day, 7-day, and 4-day schistosomule. RNA was isolated using the RNA-STAT-60 Kit (Tel-Test) following the manufacturer's instructions. To prepare total cDNA, 5 g of total RNA was used as template for each 50-l reaction. A total reaction volume of 31 l containing RNA, RNasin (Promega), 1 g of oligo(dT), 0.1 g of random hexamers (Life Technologies, Inc.), and water were incubated for 10 min at 70°C. After incubation, 1st strand buffer (Life Technologies, Inc.), 0.01 M dithiothreitol, 0.4 mM dNTPs, and 400 units of Superscriptase (Life Technologies, Inc.) were added to each reaction. After incubation at 37°C for 1 h, 0.5 l of reaction mix was used as template for PCR. SmRXR-specific oligonucleotides were synthesized for PCR: 5Ј AACAGATCTACCTAATTTAA-CATAC-OH (aa 434 -441) as forward primer and 5Ј GCACGTAGTA-AAGCTAATTCAG-OH (aa 581-588) as reverse primer. ␣-Tubulinspecific primers (16) were used in each reaction as a constitutively expressed gene control. To remain in the linear range of PCR, 28 cycles were performed. The PCR products were size-separated on a 1.5% agarose gel, visualized by ethidium bromide staining, and quantified by the Molecular Analyst program, PC version 1.5 (Bio-Rad). The amount of SmRXR PCR product was compared within each reaction to ␣-tubulin product and then this value was compared among the developmental stages.
Construction of Yeast One-hybrid Target Reporter Strains-The Matchmaker One-hybrid System protocol (CLONTECH) was followed with a few exceptions. The target-reporter construct pHisi-154, was made by inserting a 154-base pair DNA fragment, delineating nucleotides Ϫ220 to Ϫ60 of the p14 gene ( Fig. 1), repeated 3 times upstream of the His3 promotor in the pHISi-1 vector. Target-reporter construct, pHisi-EL1, was made by inserting the DNA fragment, 5Ј TTGA-CATTTTTAACTATCACGCTCAACTATCATT-OH ( Fig. 1) repeated six times upstream of the His3 promotor. After preparing competent YM4271 yeast cells as directed by CLONTECH, 10 g of the targetreporter construct was used for transformation. Yeast reporter strains were analyzed by Southern blot to ensure that proper integration of the target-reporter had occurred.
Analysis of SmRXR Activity in Vivo, Yeast One-hybrid Assay-The competent reporter yeast strains were transformed with either pACT-2 activation domain vector (CLONTECH) containing the SmRXR gene downstream of the Gal4-AD (pAC-RXR), pACT-2 vector, pADH1 expression vector containing the SmRXR (pADH-RXR), or pADH1 vector. The pADH1 vector is a 2-m expression vector with a multi-cloning site upstream of a ADH1 promotor. Control transformations without DNA were also performed. In a sterile 15-ml conical tube, 10 g of plasmid DNA, 200 g of sheared salmon sperm DNA, and 330 l of component reporter strain yeast cells were added. After mixing the DNA and yeast, 2 ml of LiAc/polyethylene glycol solution (10 mM Tris, pH 7.5, 1 mM EDTA, pH 7.5, 0.1 M LiAc, and 40% polyethylene glycol) were added, and the suspension was vortexed. The transformation suspension was incubated while shaking at 200 rpm at 30°C for 30 min. After incubation, 230 l of Me 2 SO were added and mixed by inverting the tube gently. Cells were incubated for 15 min at 42°C and then placed on ice for 2 min. Cells were centrifuged at 1000 ϫ g for 10 min, and the supernatant was aspirated. The pellet was resuspended in 2.3 ml of TE buffer. From the control tubes that contained reporter strains lacking DNA, 10 l of cells were diluted 1:100 in TE and 200 l per plate were spread onto SD/Ϫhis media containing 0, 15, 30, 45, and 60 mM 3-amino-1,2,4-triazole (AT, a HIS3 competitor). From each tube containing reporter yeast strains transformed with plasmid DNA, 100 l, were spread onto SD/Ϫleu to test transformation efficiency. All plasmids used contained the selection marker leu2 gene. From each of these tubes, 200 l per plate were spread onto SD/ϪhisϪleu containing AT at concentrations of 0, 15, 30, 45, and 60 mM. The plates were incubated at 30°C and checked each day for growth. On the 7th day, selected colonies were restreaked on SD/ϪhisϪleu plates containing AT.
Yeast Two-hybrid Assay-Yeast two-hybrid assay, originally developed by Fields and Song (17), was performed with the HybriZAP II Two-hybrid System (Strategene) according to the directions supplied from the manufacturer. SmRXR cDNA was ligated into the binding domain vector, pAS2-1 (CLONTECH) forming pAS-RXR. Control DNA included pGal4, p53/pSV40, and pLaminC/pSV40 (Stratagene). Yeast cells transformed with pAS-RXR, p53/pSV40, and pLaminC/pSV40 were plated on SD/Ϫtrp plates. Yeast transformed with pGal4 were spread on SD/Ϫleu plates. After 1 week incubation at 30°C, colonies were assayed for ␤-galactosidase activity using the CLONTECH protocol.
In Vitro Transcription-Translation of SmRXR-SmRXR cDNA was inserted into the pCITE vector (Novagen) forming pCITE-RXR and transcribed and translated using the Single Tube Protein System (Novagen). The manufacturer's protocol for [ 35 S]methionine incorporation was followed except an additional 1.25 mM cold methionine was added to each 50-l reaction. Negative control reactions with circular pCITE DNA were also performed. Samples were examined by SDS-polyacrylamide gel electrophoresis, followed by autoradiography and then used immediately.
Analysis of SmRXR Activity in Vitro, EMSA-A 250-bp sequence of the p14 upstream region, nucleotides Ϫ300 to Ϫ50 (Fig. 1), was radio-FIG. 1. p14 upstream region. The nucleotides in gray are similar to sequences found in insect genes known to be important in regulation of sex-, tissue-, and temporal-specific chorion gene expression. Underlined in gray are cis-elements important for SmRXR interaction as shown by the one-hybrid assay and gel shift assays. The double underlined sequences represent half-sites of a direct repeat or imperfect palindrome, respectively. The sequence of the probe used in EMSA analysis extended from nucleotides Ϫ300 to Ϫ50. The 154-base pair fragment inserted into the target reporter plasmid mapped from nucleotides Ϫ220 to Ϫ60. Underlined in dots are consensus CAAAT and TATA boxes, respectively. Nucleotide 1 designates the initiation site for transcription. The start ATG codon is designated by MET (10).

Identification of a Schistosome RXR Homologue cDNA-A
170-base pair DNA fragment that showed homology to RXRs was amplified using degenerate oligonucleotides representing conserved regions of RXR genes as primers and adult schistosome cDNA as template in PCR (Fig. 2). The sequence of the 5Ј primer used for PCR encodes for an amino acid sequence found in the second zinc finger of all members of retinoid receptors, RARs and RXRs (19 -22). The sequence of the 3Ј primer encodes for an amino acid sequence of the hinge region that is identical in all identified RXRs and is not found in other members of the steroid receptor superfamily. After screening 400,000 plaques of a female adult worm ZAP cDNA library with the RXR homologous DNA fragment, seven positive clones were obtained. Restriction digest and sequencing demonstrated that all seven clones were identical. The nucleotide sequence of SmRXR translates into a protein of 743 amino acids with an approximate mass of 74 kDa. SmRXR is the largest reported RXR to date. Sizes of RXRs range from 410 amino acids for mRXR␤ to 508 amino acids for dUSP (11,23,24). Comparison of the deduced amino acid sequence of SmRXR to other RXRs reveals that this protein exhibits significant homology to the DNA binding domain and the dimerization domain of other members of the RXR family (Fig. 2). For example, comparison of the amino acids of the DNA binding domain of SmRXR (aa 271-336) to those amino acids in the same domain of hRXR␣ (aa 135-200) and dUSP (aa 104 -169) shows identities of 69 and 68%, respectively (11,22,24,25). The first zinc finger within the DNA binding domain of SmRXR has only four amino acid changes from that of hRXR␣ (25). The second zinc finger exhibits lower homology. Within the dimerization and ligand binding domain are heptad repeats (Fig. 2) containing hydrophobic amino acids at positions 1, 5, and 8 (21). These heptad repeats form a dimerization motif (21). The amino acids of the dimerization motif of SmRXR (aa 531-680) share 48 and 35% identity to those of hRXR␣ (aa 325-463) and dUSP (aa 357-506), respectively (11,21,24,25). Adjacent to the heptad repeats at the amino and carboxyl ends of the dimerization and ligand binding domain are two sub-domains, i and c , respectively, known to be important for ligand-dependent transregulation in other members of the steroid hormone receptor superfamily (21, 26 -32). The i sub-domain, thought to function in hormone-relieved repression, has 20 -45% conservation among steroid receptors of vertebrates and arthropods ( (11,(21)(22)(23)(24)(25). The c sub-domain, a small amphipathic helical structure located at the carboxyl end of the dimerization and ligand binding domain, is conserved only among steroid hormone receptors that share the same ligand (22,23,(25)(26)(27)(28)(29)(30)(31)(32). For example, the c sub-domains of mRXR␤, hRXR␤, and hRXR␣, which bind 9-cis-retinoic acid, are identical to one another except for a single amino acid change of the last carboxyl terminus amino acid (23,25,27,35). The c sub-domains of RARs, which bind all-trans-retinoic acid, are also identical to one another but are dissimilar to the c sub-domains of the vertebrate RXRs (22, 23, 25-27, 36 -41). The c domain of dUSP is not homologous to vertebrate RXR c domains and does not bind to 9-cis-retinoic acid (24). Nine-cis-retinoic acid is not likely to be the ligand of SmRXR due to the variant sequence of the c domain. Immunological Identification of SmRXR in Different Worm Stages-Using an affinity purified antibody produced to a truncated form of recombinant GST-SmRXR, native SmRXR was identified in adult worm pair, mature male, mature female, and 3 h shistosomule extracts (Fig. 3). Native SmRXR has an approximate molecular mass of 74 kDa as observed in the Western blot (panel B). For a control, adult worm pair extract was probed with preimmune rabbit serum (panel C). There are irrelevant bands that appear on the Western blot, but none the 74-kDa size. These bands are not present in panel B because the serum was affinity purified.
Developmental Expression of the SmRXR Gene-RT-PCR was performed on total RNA isolated from various developmental stages to evaluate developmental expression of the SmRXR gene. In order to control for the expression of multiple schistosome RXR homologues, primers to non-conserved regions of SmRXR were used in the RT-PCR reactions. ␣-Tubulin-specific primers that represent a constitutive gene were added to each reaction. A 465-bp SmRXR cDNA fragment and a 500-bp ␣-tubulin product were amplified (Fig. 4). By comparing the amount of SmRXR PCR product to that of ␣-tubulin within each reaction and then comparing this value among the different life stages, we demonstrate that SmRXR is constitutively expressed throughout development (Fig. 4).
SmRXR Binds to the p14 Gene Upstream cis-Elements and Activates Transcription-To determine if SmRXR binds to ciselements of the p14 upstream region and drives transcription of a female-specific gene, we employed the yeast one-hybrid system. The reporter yeast strains, YTR-154, YTR-EL1, and YTR-pHisi, were transformed with pAC-RXR and pAC-2. YTR-154 transformed with salmon sperm DNA alone only grew on SD/Ϫhis media. This is expected because the His3 promotor is leaky. Therefore, growth of YTR-154 transformed with pAC-RXR on SD/ϪhisϪleu media with even low concentrations of aminotriazole would be a result of SmRXR-AD fusion protein binding to cis-elements of p14 and driving transcription of the HIS3 gene. The transformed YTR-154 cells were spread on SD/ϪhisϪleu plates with increasing amounts of AT. After 7 days of incubation, two classes of colonies grew that ranged from 0.25 to 0.5 mm in size and from 1.5 to 2.5 mm in size (data not shown). Colonies of both sizes were picked and streaked on SD/ϪhisϪleu plates containing up to 60 mM aminotriazole. The smaller colonies did not grow. The larger colonies grew on SD/ϪhisϪleu containing 60 mM AT (Fig. 5A). YTR-154 cells transformed with pAC-2 grew as small colonies (0.25-0.5 mm) in the original spreads. These colonies were picked and streaked on SD/ϪhisϪleu containing a range of AT concentrations. At 15 mM AT these colonies failed to grow on SD/ ϪhisϪleu (Fig. 5A). The negative control strain YTR-pHisi, transformed with either pAC-RXR or pAC-2, grew as small colonies in the original spread but failed to grow when restreaked on media containing AT (data not shown). Therefore, it is evident that SmRXR fused to the Gal4 activation domain binds to cis-elements of p14 and drives transcription of the HIS3 reporter gene in the one-hybrid system.
To define further the interaction between the p14 gene and SmRXR-AD, YTR-EL1 that contained the direct repeat of the upstream region of p14 (Fig. 1) was used. As with the YTR-154 strain, after transformation with pAC-RXR there were larger colonies at higher concentrations of AT. YTR-EL1 cells grew best on 45 mM AT but did grow at 60 mM AT, at a slower rate.
On the restreak plates it only took 3 days for YTR-154 growth to appear on SD/ϪhisϪleu containing 60 mM AT, whereas it took 5-6 days for growth of YTR-EL1 to appear on plates containing 60 mM AT (Fig. 5B). Growth of YTR-EL1 appeared on day 4 on the plates with 45 mm AT. Fig. 5B shows growth of five individual colonies of YTR-EL1 on 60 mm AT. The negative control included in this experiment was YTR-EL1 transformed with pAC-2 alone. Again there were small background colonies on the original spread that did not grow on the restreak (Fig.  5B). Therefore SmRXR-AD binds to the sequence including the direct repeat and enhances the transcription of the HIS3 gene.
Because RXR forms heterodimers in many species (42)(43)(44)(45)(46)(47)(48)(49)(50), we attempted to employ the yeast two-hybrid system to obtain schistosome clones that express proteins that interact with SmRXR. As a control, we transformed Y190 yeast cells with pAS-RXR, the bait plasmid containing the GAL4 DNA binding domain. SmRXR alone activated ␤-galactosidase expression (data not shown). This result indicated that SmRXR by itself may activate gene transcription, perhaps as a homodimer, or by interacting with yeast proteins. In order to determine if SmRXR by itself, or perhaps with the help of a yeast protein, could drive transcription when recognizing cis-elements of p14, SmRXR was cloned into an expression vector that lacked the Gal4 activation domain and DNA binding domain (pADH-RXR). When YTR-154 was transformed with pADH-RXR, the cells as in the case of transformation with pAC-RXR (above) grew on SD/ϪhisϪleu media. There were very small colonies, 0.25-0.05 mm in size, and larger colonies, 1.5-2.5 mm in size. Again the smaller colonies did not grow after being restreaked on SD/ϪhisϪleu plates containing AT, but the larger colonies grew on SD/ϪhisϪleu containing 60 mM AT (Fig. 5C). Growth of these colonies appeared at day 3 on the restreak plates containing 60 mM AT. As in the above experiment, YTR-154 transformed with salmon sperm alone grew only on SD/Ϫhis The lower bands are PCR products amplified using SmRXR specific primers. media without AT. Therefore, SmRXR is binding to cis-elements of the p14 upstream region and driving transcription of the HIS3 gene without the Gal4 activation domain. Controls of this experiment included transforming YTR-154 with pADH1. Again small colonies grew on all the plates of the orginal spread but did not grow on SD/ϪhisϪleu media containing aminotriazole after the restreak (Fig. 5C). Similar to the above experiment YTR-pHisi, transformed with both pADH-RXR and pADH1, grew on all plates of the original streak but not after the restreak (data not shown).
YTR-EL1 cells were also transformed with pADH-RXR. Yeast colonies grew on SD/ϪhisϪleu containing 60 mM AT (Fig.  5D). Growth of these colonies on media containing 60 mM AT was slower than YTR-154 transformed with pADH-RXR and YTR-EL1 transformed with pAC-RXR. Growth of YTR-EL1 transformed with pADH-RXR on the restreak plates containing 60 mM AT appeared on days 6 and 7 depending on the clone. Small colonies grew on plates that contained YTR-EL1 transformed with pADH1, but these colonies did not grow when restreaked on SD/ϪhisϪleu containing AT. Indeed, SmRXR binds to the sequence containing the direct repeat (Fig. 1) and drives transcription of HIS3 (Fig. 5D).
EMSA Analysis of SmRXR-DNA Interactions-EMSA were performed to study the in vitro interaction between p14 DNA and SmRXR. A 250-bp sequence, nucleotides Ϫ300 to Ϫ50, of the p14 upstream region (Fig. 1) was used as a probe in gel shift assays to determine whether or not SmRXR, synthesized by in vitro coupled transcription/translation, binds to RXR recognition sites. Shown in lane 1 of Fig. 6 is the probe without lysate from the transcription-translation mix. In lane 2, the binding reaction contains lysate from a transcription-translation reaction with pCITE vector used as template. There are a few irrelevant shifts of the probe caused by contents of the lysate binding to the p14 probe. In lane 3 it is apparent that SmRXR causes unique shifts of the radiolabeled p14 fragment. These shifts are denoted by the two arrows (Fig. 6). When cold p14 DNA is added in 100 ϫ molar amount (lanes 4 and 5), the upper shift is competed, whereas the lower shift identified by the arrow is competed to a much lesser extent. Nonspecific DNA (lanes 6 -8) does not interfere with the changes in electrophoretic mobility. Therefore, the shifts observed in lane 3 are specific for SmRXR-p14 interaction. The direct repeat competes the shifts to a great extent (lanes 9 -11). At 250 ϫ, the competitor completely competes the two shifts denoted by both arrows. The DNA fragment that includes the sequence GTAGAATCA, competes the upper shift slightly at 250 ϫ molar amounts (lane 14). The sequence that includes the imperfect palindrome (lanes [15][16][17] competes the upper shift at 100 ϫ and at 250 ϫ competes both shifts. In vitro studies demonstrate that the direct repeat sequence of the p14 gene is a preferential binding site for SmRXR.

DISCUSSION
Schistosomes are the only trematodes with separate sexes. This fact has driven the evolution of an interesting biological interplay between the sexes, such that the male through an unknown stimulus is able to regulate female-specific gene expression (2). A number of female-specific genes have been isolated, but as of yet no factors directly involved in the regulation of female-specific genes have been identified. In this paper we report the identification of SmRXR cDNA and provide evidence for its involvement in female-specific gene expression.
Members of the RXR subfamily in other species are important in the regulation of sex-, tissue-, and temporal-specific genes. For example, USP of Drosophila is employed in the regulation of chorion genes, embryogenesis, larval development, and pupation (11,12,45). This pleiotrophy of function is representative of RXR subfamily members. RXRs are known to form heterodimers with various partners that are essential for the regulation of a number of genes involved in homeostasis and development (45)(46)(47)(48)(49)(50). Examples of RXR partners include the ecdysone receptor of Drosophila, vitamin D receptor, thyroid hormone receptor, and retinoic acid receptors (RAR) (42)(43)(44)(45)(46)(47)(48)(49)(50). When RXR forms heterodimers, such as RXR/RAR and USP/ecdysone receptor, the conformation of its binding partner changes allowing a greater affinity not only for the DNA response element but also for the ligand (43)(44)(45)(46)(47)(48)(49)(50)(51). RT-PCR results demonstrated that SmRXR mRNA is constitutively expressed throughout schistosome development, not only in mature females as are the female-specific genes (2,3,7,9,10). Furthermore, Western blot analysis showed that the 74-kDa SmRXR protein is present in both sexes, mature and immature. This demonstrates that SmRXR is not female-specific and suggests that, like other members of the RXR subfamily, SmRXR is important in the regulation of a number of genes. Future studies will entail identifying SmRXR-binding partners involved in the regulation of female-specific genes and defining the putative role(s) SmRXR plays in the regulation of other genes.
To date, a genetic system to study the functions of schistosome genes in vivo does not exist. Therefore, evidence for the involvement of SmRXR in female-specific gene regulation was obtained by the yeast one-hybrid system and EMSA analysis. In these studies, we showed that SmRXR binds to cis-elements of the p14 gene and drives transcription of a reporter gene in yeast. Members of the steroid receptor superfamily commonly bind to cis-elements that consist of two half-sites spaced by 3, 4, or 5 nucleotides (13,14). EMSA analysis demonstrates that the preferential binding site for SmRXR, in vitro, is a sequence containing the half-site AACTATCA spaced by five nucleotides (Fig. 1). In the upstream region of s15, a Drosophila chorion gene, the sequence TCACGT is an essential part of an imperfect palindrome known to be a USP response element (12). Interestingly, a similar sequence, TCACGCT, occupies the 3Ј end of the first half-site and part of the spacer region of the direct repeat in the p14 gene upstream region (Fig. 1). To define the region of the p14 gene involved in SmRXR binding interaction and trans-activation, the direct repeat sequence was used as a target element in the one-hybrid assay. SmRXR most likely acts as a homodimer to drive transcription of this chimeric reporter gene. It is unlikely that SmRXR interacts with a yeast protein. RXRs have been shown to form heterodimers with only members of the steroid receptor family, and to date, steroid receptors have not been identified in yeast (42)(43)(44)(45)(46)(47)(48)(49)(50).
An evolutionary study was performed to better understand the function of SmRXR. Knowing the rate at which SmRXR has evolved as compared with other RXRs and RARs may answer questions about how it functions in regulating gene expression. Steroid receptors have recently been identified in early metozoans, such as cnidarians and schistosomes (52). Interestingly, in a nuclear receptor evolutionary study, the DNA binding domain of a schistosome RXR homologue, which is not identical to the SmRXR cDNA identified in our lab, was described (52). The divergence of subfamilies of the steroid superfamily was previously determined to have occurred before the divergence of the arthropods and vertebrates (22,53) (Fig. 7). From the Escriva et al. (52) study and the phylogenetic tree designed using full-length cDNA sequences of RXRs and RARs shown in this report (Fig. 7), the above statement can be extended to note that the divergence of subfamilies occurred before the divergence of cniderians and platyhelminthes (52,53). Because SmRXR is deeply rooted in evolution (Fig. 7), comparison of this gene with other steroid receptors may lead to the identification of the original ancestor of the steroid receptor family.
The trans-activating function of steroid receptors is often dependent on dimerization and ligand binding (22,(42)(43)(44)(45)(46)(47)(48)(49)(50). Ligand binding is thought to change the conformation of the steroid receptor into an active state (51). The ability of members of the RXR subfamily to bind specific ligands appears to have evolved after the vertebrate and arthopod lineages diverged, as of yet ligands have not been found for arthropod FIG. 6. Demonstration of SmRXR DNA binding activity by EMSA. Lane 1, DNA alone. Lane 2, lysate from the control transcription-translation reaction. Lane 3, lysate containing in vitro translated SmRXR. Lanes 4 and 5, binding reactions contained cold p14 competitor in 10 and 100 ϫ amounts, respectively. Lanes 6 -8, binding reactions contained a 250-bp fragment of a irrelevant DNA competitor in 10, 100, and 250 ϫ amounts, respectively. Lanes 9 -17, binding reactions contained 10, 100, and 250 ϫ amounts of three different DNA sequences, the direct repeat, a sequence containing GTAGAATCA, and the imperfect palindrome sequence, respectively (Fig. 1). Arrows denote unique shifts caused by SmRXR-p14 interaction.
This report provides the first evidence that RXRs play a role in gene regulation in primitive eukaryotic organisms such as schistosomes. Not only does this study add to an increasing amount of information on the evolution of steroid receptors but also has initiated the study of steroid receptors involved in a very intriguing interplay between the separate sexes of schistosomes and the regulation of female-specific genes. The study of steroid receptors, such as SmRXR, involved in female-specific gene regulation may lead to pharmacological compounds that prevent female maturation and egg production, thereby controlling schistosomiasis.