BIOLOGY OF REPRODUCTION 79, 518–524 (2008)Published online before print 21 May 2008.DOI 10.1095/biolreprod.108.069468

Receptor Transporter Protein 4 (RTP4) in Endometrium, Ovary, and Peripheral BloodLeukocytes of Pregnant and Cyclic Ewes1

C.A. Gifford,3 A.M. Assiri,4 M.C. Satterfield,5 T.E. Spencer,5 and T.L. Ott2,6

Department of Animal and Veterinary Science,3 University of Idaho, Moscow, Idaho 83843School of Molecular Biosciences,4 Washington State University, Pullman, Washington 99164Center for Animal Biotechnology and Genomics,5 Department of Animal Science, Texas A&M University,College Station, Texas 77843Department of Dairy and Animal Science,6 Pennsylvania State University, University Park, Pennsylvania 16802

ABSTRACT

Interferon-tau (IFNT) is secreted by the conceptus trophoblastand signals pregnancy recognition in ruminants. IFNT regulatesexpression of genes in the endometrium, peripheral bloodleukocytes (PBLs), and corpus luteum (CL). Microarray analysisidentified that expression of (chemosensory) receptor transport-er protein 4 (RTP4) increased in PBLs during early pregnancy incows. In the present study, we cloned and characterized RTP4transcription during early pregnancy in ewes. Endometrium,PBLs, and CL were collected on Days 11, 13, and 15 of the cycleand on Days 11, 13, 15, 17, and 19 of pregnancy. Northern blotanalysis revealed an expected 1.6-kb mRNA and an unexpected2.6-kb mRNA. In endometria, RTP4 mRNA levels in cyclic ewesremained low, whereas RTP4 mRNA increased from Day 11 toDay 17 in pregnant ewes. Levels of RTP4 mRNA also increasedfrom Day 15 to Day 19 in CL and PBL samples from pregnantewes only. The RTP4 mRNA was located in the glandularepithelium, stratum compactum, and caruncular stroma. Ovineglandular epithelial cells were treated with IFNT to determine ifIFNT alone could induce RTP4. IFNT increased RTP4 more than70-fold at 1.5 h after treatment, with maximal induction ofnearly 300-fold above values observed in nontreated controls at6 h after treatment. These results indicate that RTP4 mRNAlevels are induced in the ovine endometrium, PBLs, and CL byIFNT during early pregnancy and in cell culture in response toIFNT. If RTP4 expression affects G protein-coupled receptorfunction, it may be important for establishment of pregnancy indomestic ruminants.

interferon, pregnancy, RTP4, sheep, uterus

INTRODUCTION

Interferon-tau (IFNT) is produced by trophoblast cells of theblastocyst and is the signal for maternal recognition ofpregnancy in ruminants [1]. IFNT is a member of the type Iinterferon (IFN) family, which includes IFNA, IFNB, andIFNW (alpha, beta, and omega, respectively) [2], and itpossesses many functional features shared by other type I IFNs.Ruminant conceptuses produce IFNT during early pregnancy,with maximal production from Day 14 to Day 16 in sheep [3]

and from Day 16 to Day 19 in cattle [4]. IFNT acts in aparacrine manner on the luminal epithelium and superficialglandular epithelium to inhibit expression of oxytocin receptors[5–7], thereby blocking luteolytic pulses of prostaglandin F2a(PGF2a) and maintaining a functional corpus luteum (CL) [1].IFNT also induces expression of a large number of IFN-stimulated genes (ISGs) in the endometrium [8] including20,50-oligoadenylate synthetase (OAS) [9], beta2-microglobulin[10], interferon-stimulated gene 15 (ISG15) [11], myxovirusresistance 1/interferon-inducible protein p78 (MX1), andmyxovirus resistance 2 (MX2) [12–14]. We and otherspostulate that these genes are involved in preparing the uterusfor establishment and maintenance of pregnancy; however,only recently has it become clear that ISGs in peripheral bloodimmune cells also are increased by pregnancy.

Yankey et al. [15] first reported systemic induction of ISGsin peripheral blood leukocytes (PBLs) in response topregnancy, even though IFNT had not previously beendetected in the circulation [16]. The level of MX1 mRNAwas elevated in PBLs from Day 15 through Day 30 afterinsemination in pregnant compared to bred, nonpregnant ewes[15]. Subsequently, we found that PBLs from pregnant cowsexhibited increased expression of MX1 and ISG15 on Days 18and 20 of gestation and of MX2 on Days 16, 18, and 20 ofgestation compared with bred, nonpregnant cows [14].Likewise, Han et al. [17] found that overall mean bloodISG15 mRNA levels were higher from Day 15 to Day 32 ofgestation, with maximal levels on Day 20 in pregnantcompared with inseminated, nonpregnant cows. These dataare consistent with work by Spencer et al. [18], who showedintrauterine infusions of recombinant ovine IFNT and systemictreatment with IFNT both increased levels of MX1 and ISG15mRNA in the CL. Moreover, ISG15 and OAS are increased inthe CL of early pregnant ewes [19].

It is becoming increasingly clear that many ISGs induced bypregnancy and IFNT in the endometrium also are induced inPBL. These results support the hypothesis that one couldidentify genes regulated during early pregnancy in theendometrium by noninvasively sampling PBLs and subjectingthem to gene expression analysis. Taking this approach, weidentified a large number of known and novel ISGs that aredifferentially expressed in PBLs from pregnant as compared tocyclic cattle (Ott and Gifford, unpublished observations). Oneof the novel ISGs, chemosensory receptor transporter protein 4(RTP4, also known as IFRG28), was upregulated in PBLs frompregnant compared to nonpregnant cows 19 days afterinsemination.

Certain G protein-coupled receptors (GPCRs) requireaccessory proteins for transport from the trans-Golgi networkto the cell surface, allowing GPCRs to interact with appropriate

1GenBank accession no. EF682065.2Correspondence: Troy L. Ott, Department of Dairy and AnimalScience, 324 Henning Bldg. Pennsylvania State University, UniversityPark, PA 16802. FAX: 814 863 6042; e-mail: tlo12@psu.edu

Received: 27 March 2008.First decision: 16 April 2008.Accepted: 30 April 2008.� 2008 by the Society for the Study of Reproduction, Inc.ISSN: 0006-3363. http://www.biolreprod.org

518

ligands [20]. The RTP4 belongs to a gene family consisting ofRTP1, RTP2, RTP3, and RTP4 [21]. Both RTP1 and RTP2appear to be specifically expressed in olfactory neurons and tofacilitate functional expression of odorant receptors [21]. TheRTP4 is expressed in a broad range of tissues and recently wasidentified as a cofactor for functional expression of some bittertaste receptors [22]. Based on our previous observations thatISGs, including RTP4, are regulated in PBLs during earlypregnancy [14, 15] (Ott and Gifford, unpublished observa-tions), we postulate that RTP4 will be upregulated inendometrium by the conceptus and IFNT during earlypregnancy in ruminants. Therefore, the objectives of thepresent experiments were to clone and sequence the ovineRTP4 mRNA and to characterize RTP4 levels in the uterus,PBLs, and ovary of cyclic and early pregnant ewes.

MATERIALS AND METHODS

Animals

Mature, crossbred ewes were housed with a vasectomized ram andobserved for estrus. At estrus, ewes were randomly assigned to be bred to eithera vasectomized ram (cyclic) or an intact ram (pregnant). Ewes also wererandomly assigned to collection day, and blood and tissue samples werecollected on Days 11, 13, and 15 of the estrous cycle and on Days 11, 13, 15,17, and 19 of pregnancy (n¼ 4 samples per day per status). Tissues were snap-frozen, and PBLs were isolated from blood samples as described previously[14]. An additional group of ewes underwent hysterectomy on Days 10, 12, 14,and 16 of the estrous cycle and on Days 10, 12, 14, 16, 18, and 20 of pregnancy(n¼ 5 ewes per day per status). Pregnancy was confirmed by identification of aconceptus at the time of tissue collection. All procedures were in compliancewith the Guide for the Care and Use of Agriculture Animals in Research andTeaching and were approved by the Institutional Animal Care and UseCommittees at the University of Idaho or Texas A&M University.

Cell Culture

Immortalized ovine glandular epithelial (oGE) cells [23] were plated at 80%confluency in six-well plates (Costar 3516; Corning, Inc.) in Dulbeccomodified Eagle medium (15.63 g/L; Sigma) with 10% fetal bovine serum (FBS;Invitrogen) under 5% CO

2at 38.58C for 24 h before treatment. Cells were

treated with either medium alone or 10 000 antiviral units/ml of IFNT (kindlyprovided by Dr. Fuller Bazer, Texas A&M University) for 0, 1.5, 3, 6, 12, or 24hours (n ¼ 3 wells per treatment per time). Cells were approximately 90%confluent when harvested using Trizol reagent (Invitrogen). The entireexperiment was conducted twice.

RNA Isolation and cDNA Synthesis

Total RNA was extracted from the frozen endometrium and CL samplesfrom pregnant and nonpregnant ewes using Trizol reagent according to themanufacturer’s recommendations. After extraction, the RNA was furtherpurified using RNeasy (Qiagen) spin columns according to the manufacturer’srecommendations. The PBLs were isolated and the RNA extracted usingmethods described previously [14]. Total RNA (1 lg) was treated with DNase(RQ1; Promega) and used for cDNA synthesis with Superscript III (Invitrogen)according to the manufacturer’s recommendations.

Cloning and Sequencing

The SMART RACE (i.e., rapid amplification of cDNA ends) cDNAAmplification kit (BD Biosciences) was used to synthesize SMART first-strand50-RACE-Ready cDNA and 30-RACE-Ready cDNA from endometrial RNAisolated from a Day 17 pregnant ewe. The SMART first-strand cDNA was usedto amplify the RTP4 50-RACE and 30-RACE by using BD Advantage 2 PCRkit (BD Biosciences). The PCR was carried out using the Universal primer Amix, which is provided with the SMART RACE kit, and the RTP4 cDNAspecific primers, CACATGTACCTGGAGAACCAGA and AGGTAGCTCT-GAAACCTTCCTG, to amplify RTP4 50-RACE and 30-RACE, respectively.This experiment generated only one 5 0 and one 3 0 sequence, withapproximately 200 bp of overlap between the sequences. These sequenceswere used to design another set of RTP4 gene-specific primers (forward,GGGACTGTCAGCATGGACGCCAAGCCT; reverse, GACACCA-GAAAGGTAAACAGAGCGAAA) to amplify the full coding sequence of

RTP4 from Day 17 pregnant sheep endometrial RNA. The PCR reactions wereperformed as five cycles at 948C for 5 sec and 728C for 3 min; then five cyclesat 948C for 5 sec, 708C for 10 sec, and 728C for 3 min; and then 35 cycles at948C for 5 sec, 688C (708C for 50-RACE) for 10 sec, and 728C for 3 min. ThePCR products were gel purified by using GenElute Spin Columns (Sigma-Aldrich) and subsequently cloned into PCR II-TOPO TA vector (Invitrogen).Plasmid sequencing was conducted by the Nucleic Acids Core Facility, HuckInstitutes of Life Sciences, at Penn State. Because sequence results from RACEexperiments were identical to the sequence obtained from full-length PCRamplification, sequencing results are presented from PCR amplification.

Real-Time RT-PCR

Steady-state levels of RTP4 mRNA were quantified in samples ofendometrium, PBLs, and oGE cells as described previously [14] except thatthe 18S ribosomal RNA was used as an internal control. The annealingtemperature for RTP4 primers (forward, CACATGTACCTGGAGAACCAGA;reverse, AGGTAGCTCTGAAACCTTCCTG) and 18S primers (forward,AAACGGCTACCACATCCAAG; reverse, CGCTCCAAGATCCAACTA)was 56.68C. Steady-state mRNA levels in the CL were quantified as describedpreviously [14] with minor modifications. Power SYBR Green Master Mix(Applied Biosystems) was used, and quantification was accomplished using theApplied Biosystems 7500 fast real-time RT-PCR machine with 18S ribosomalRNA as an internal control.

Northern Blot Analysis

Northern blot analysis was conducted as described previously [13]. TotalRNA (10 lg) from samples of CL, endometrium, and oGE cells was used forgel electrophoresis and transfer, and blots were incubated with a biotinylated,513-bp ovine RTP4 antisense cRNA probe. All other steps were completed asdescribed previously [13] with the exception of RNase A treatment.

In Situ Hybridization Analysis

The location of RTP4 mRNA in the uterus and ovary was analyzed using a513-bp, radiolabeled, antisense cRNA probe for RTP4 using methods describedpreviously [24]. The cRNA probe used for in situ analysis was the samesequence as the probe used for Northern blot analysis and, therefore, likelyrecognizes both forms of RTP4 mRNA (see Results).

Data Analysis

Data for RTP4 mRNA levels in the CL, PBLs, and endometrium arepresented as least squares means with the standard error of the relative fold-change from the mean Day 11 cyclic corrected critical threshold value (DCt) orfrom time-point zero samples (oGE cell culture experiment). Fold-change wasthen calculated by the DDCt method [25], with 18S rRNA serving as theinternal control. Using this conservative approach, we do not arbitrarilyeliminate variation from the data set by setting all Day 11 cyclic values to one.Data were analyzed using the MIXED procedure in SAS (Ver 9.1; SASInstitute). Animal was the experimental unit, and fold-change in the dependentvariable was tested against status, day, and status 3 day. In cell cultureexperiments, well was the experimental unit, and fold-change was testedagainst time, treatment, and time 3 treatment.

RESULTS

Ovine RTP4 DNA Sequence Analysis

A full-length RTP4 mRNA transcript was cloned from theendometrium of Day 17 pregnant ewes and sequenced(GenBank accession no. EF682065), yielding a predictedmRNA size of 1.672 kb. As illustrated in Figure 1, the inferredRTP4 protein is 454 residues, with a predicted molecularweight of approximately 57.5 kDa (Swiss Institute ofBioinformatics: ExPASy translate tool; http://ca.expasy.org/tools/) [26]. Consistent with RTP4 full-length sequence,Northern blot analysis revealed an mRNA transcript ofapproximately 1.6 kb in the endometrium, CL, and oGE cells(Fig. 2). Interestingly, an additional mRNA transcript ofapproximately 2.6 kb in size was observed in these samples(Fig. 2).

RTP4 IN PREGNANT AND CYCLIC EWES 519

RTP4 Levels

A status 3 day interaction was detected (P , 0.05) for RTP4mRNA levels in endometrial samples from pregnant comparedto cyclic ewes. Therefore, effects were analyzed within status.As depicted in Figure 3, RTP4 levels increased (P , 0.01,quadratic) by 17-fold on Day 15 and by 34-fold on Day 17 inpregnant ewes, whereas RTP4 mRNA in the endometrium ofcyclic ewes did not change (P . 0.30) from Day 11 to Day 15.No status 3 day interaction was detected (P . 0.10) for levelsof RTP4 in PBLs from pregnant and cyclic ewes when Days11, 13, and 15 were analyzed. The RTP4 levels, however,increased (P , 0.05, quadratic) by nearly 20-fold on Days 17and 19 in PBLs from pregnant ewes (Fig. 3). Likewise, nostatus 3 day interaction was detected (P . 0.10) for RTP4levels in CL from pregnant and cyclic ewes when Days 11, 13,and 15 were analyzed. Steady-state RTP4 mRNA levels,however, increased (P , 0.05, quadratic) in the CL frompregnant ewes by more than 7-fold on Day 15 and 14-fold onDay 17 (Fig. 3). In oGE cells, IFNT treatment increased levelsof RTP4 (P , 0.01) by more than 70-fold as early as 1.5 h aftertreatment, with maximal induction of more than 300-fold at 6 hafter IFNT treatment.

RTP4 Localization

In the endometrium, RTP4 mRNA was localized to theglandular epithelium, stratum compactum, and caruncularstroma (Fig. 4). Although RTP4 mRNA was more abundantin pregnant animals, variation in RTP4 levels was observedwithin day and status (Fig. 4). In addition, some endometrialsections exhibited scattered cells that hybridized strongly with

FIG. 1. Amino acid alignment of predictedhuman, ovine, and bovine RTP4. Shadedareas represent homologous amino acids,and dashes represent gaps in the proteinsequence. The molecular weights for hu-man, ovine, and bovine RTP4 are 28, 57.5,and 60 kDa, respectively. Ovine RTP4amino acid sequence shares approximately61% and 82% homology with human RTP4and bovine RTP4 sequences, respectively.

FIG. 2. Top left) Northern blot analysis of receptor transporter protein 4(RTP4) in total RNA from corpus luteum in cyclic ewes 15 days afterbreeding (15C) and in pregnant ewes 15 days (15P) and 17 days (17P) afterbreeding. Top right) Northern blot analysis of RTP4 in total RNA fromendometrium of pregnant (15P) and cyclic ewes (15C) 15 days afterbreeding. Bottom) Northern blot analysis of RTP4 in total RNA fromimmortalized oGE cells treated with 10 000 antiviral units/ml ofinterferon-tau for 12 h (12h IFN) or 6 h (6h IFN) and control cells thatwere not treated for 12 h (12h NT) or 6 h (6h NT).

520 GIFFORD ET AL.

the RTP4 probe and exhibited a phenotype similar to that ofimmune cells (cell size and nuclear morphology); however, cellidentity was not confirmed. In the ovary, RTP4 mRNA waslocalized primarily to the CL and increased with pregnancy.

DISCUSSION

Our working hypothesis is that changes in uterine geneexpression associated with early pregnancy could be detectedby examining gene expression in PBLs during early pregnancy.Using microarray analysis of PBLs from pregnant and bred,nonpregnant cows, RTP4 was identified as a gene upregulatedin response to early pregnancy. We then cloned and sequencedan endometrial RTP4 mRNA and measured steady-state levelsof RTP4 mRNA in the endometrium, CL, and PBLs duringearly pregnancy in ewes. In endometrium and CL, Northernblot analysis detected two RTP4 mRNA transcripts. The 1.6-kbmRNA was expected based on the size of the endometrialRTP4 mRNA cloned by PCR, but a larger, 2.6-kb mRNAtranscript was identified in endometrium, CL, and oGE cells.Because of limited amounts of mRNA, we were unable toconduct Northern blot analysis on mRNA extracted fromPBLs. The RTP4 levels increased during early pregnancy in allthree tissues studied compared to levels in cyclic ewes and inresponse to IFNT treatment in oGE cells.

Some GPCRs require proteins to transport them from thetrans-Golgi to the cell membrane [20, 27, 28]. RTP4 belongs toa family of GPCR transporters that is necessary for functionalexpression of some bitter taste receptors [21, 22]. The predictedRPT4 transcript in cattle (BC105539) (Fig. 1) and from the

endometrium of sheep (Fig. 1) is larger than human RTP4(NM_022147) (Fig. 1). National Center for BiotechnologyInformation Basic Local Alignment Search Tool (BLAST)results, however, showed that the predicted ovine RTP4 aminoacid sequence shared approximately 61% and 82% homologywith human RTP4 and bovine RTP4 sequences, respectively(Fig. 1). The RTP4 also is referred to as 28-kDa interferonresponsive protein (IFRG28; NP_071430). Although humanRTP4 has a molecular weight of approximately 28 kDa,predicted translated sequences of bovine (60.6 kDa) and ovine(57.5 kDa) RTP4 are much larger; thus, the IFRG28nomenclature may not be appropriate for domestic ruminants.Further studies are needed to determine if this discrepancy intranscript and predicted protein size leads to differentfunctional properties of RTP4 in domestic ruminants. More-over, an alternate, 2.6-kb transcript was detected in the CL andendometrium of sheep and in oGE cells (Fig. 2). Our RACEanalysis using endometrial RNA did not provide any evidencefor the larger transcript. It is possible, however, that sequencedifferences exist in the primer binding regions that aresufficient to prevent detection of the larger transcript byRACE. Although the full sequence for the 2.6-kb form of RTP4has not yet been determined, the 513-bp probe used in thepresent study was complementary to the 50 end of the 1.6-kbovine RTP4 transcript and also hybridized with the 2.6-kb formof RTP4, indicating the two forms share homology over a largeregion. The identity of this larger hybridizing band is currentlyunder investigation.

FIG. 3. Steady-state mRNA levels for RTP4 from cyclic (open bars; n ¼ 4 per day) or pregnant (closed bars; n ¼ 4 per day) ewes in CL (upper left),endometrium (upper right), PBLs (lower left), and immortalized oGE cells (lower right) that were either control (open bars; n ¼ 6 wells/time point) ortreated with interferon-tau (closed bars; n¼6 wells/time point). Endometrium, CL, and PBL results are presented as relative fold-change compared to Day11 cyclic; error bars represent the SEM. Levels of RTP4 in oGE cells are represented as relative fold-change compared to time-point zero; error barsrepresent the SEM. In the CL, a status 3 day interaction was not detected between Day 11 and Day 15 (P . 0.10), but levels of RTP4 increased (P , 0.05,quadratic) by more than 14-fold in pregnant ewes on Day 17 and by more than 4-fold on Day 19 of pregnancy. In the endometrium, RTP4 levels increased(P , 0.01, quadratic) by 17-fold on Day 15 of gestation to maximal levels of 34-fold on Day 17 in pregnant ewes, whereas cyclic ewes showed no change(P . 0.30) in RTP4 levels. In PBLs, a status 3 day interaction was not detected between Day 11 and Day 15 (P . 0.10), but levels of RTP4 increased (P ,0.05, quadratic) by nearly 20-fold on Days 17 and 19 of pregnancy. Interferon-tau treatment increased (treatment 3 time; P , 0.01) RTP4 mRNA in oGEcells by more than 70-fold at 1.5 h after treatment, with maximal induction of 300-fold by 6 h after treatment.

RTP4 IN PREGNANT AND CYCLIC EWES 521

The type I IFN receptor is expressed in all cell types in theendometrium, with highest expression in luminal epithelium[29]. Induction of many ISGs by IFNT is limited to theendometrial stroma, immune cells, and glandular epithelialcells [1]. Levels of RTP4 mRNA increased in the endometriumin response to pregnancy and in oGE cells treated with IFNT(Fig. 3). Results from the current experiment show that RTP4mRNA is localized to the glandular epithelium, stratumcompactum, and caruncular stroma (Fig. 4); however, ani-mal-to-animal variation in the spatial pattern of RTP4localization in the endometrium was observed (Fig. 4). Itmay be that RTP4 is upregulated for a short period of time,from Day 15 to Day 19 of pregnancy, and that individualanimals vary in levels of RTP4 between these days. We are inthe process of examining samples from later in gestation todetermine if this is the case. Alternatively, RTP4 levels mayexhibit variable spatial regulation in the endometrium duringearly pregnancy. Therefore, it is possible that the cross-sectionsused in the current study were from areas of the endometriumwith lower/higher levels of RTP4 mRNA. Also, RTP4 mRNAwas upregulated in unidentified cells that resembled immunecells based on their size, nuclear morphology and scattereddistribution in some, but not all, of the tissue sections examined(Fig. 4). Again, this variation likely results from variation in thespatial pattern of expression of RTP4 mRNA. We haveobserved a similar spatial pattern for MX2 localization during

early pregnancy in endometrium (Ott et al., unpublishedobservations).

In CL from pregnant ewes, RTP4 increased by more than 7-fold on Day 15 and 14-fold on Day 17 (Fig. 3). Spencer et al.[18] showed that either subcutaneous or intrauterine injectionsof IFNT increased levels of ISG15 and MX1 mRNA in the CL.Likewise, recent evidence suggests that ISG15 and OAS areincreased in the CL during early pregnancy [19]. Results fromthe current experiment localized RTP4 transcripts to luteal cellsspecifically (data not shown). Oliveira et al. [19] found thatISG15 protein localized predominately to large luteal cells withvariable expression in small luteal cells and that ISG15 was notdetected in the ovarian stroma. This is interesting, becausemost tissues and cell types express the type I IFN receptor [30]yet RTP4 transcripts were not detected in the ovarian stroma.Likewise, unpublished work in our laboratory found that MX2was spatially and temporally regulated in the ovary duringearly pregnancy, with transcripts localized primarily to lutealcells. Collectively, results from the current and previous studies[18, 19] (Ott et al., unpublished observations) indicate thatISGs are regulated in a cell-specific manner in the ovary duringearly pregnancy.

The function(s) of ISGs in the CL during early pregnancy isunclear. It is possible, however, that IFNT/pregnancy affectsCL function through regulation of ISGs, including RTP4 inluteal cells, resulting in changes supporting the establishmentof pregnancy. Because exogenous progesterone can support

FIG. 4. In situ hybridization analyses of receptor transporting protein 4 (RTP4) mRNA in cyclic and pregnant ewes. Cross-sections of the uterine wallfrom cyclic (C) and pregnant (P) ewes were hybridized with radiolabeled antisense or sense ovine RTP4 cRNA probes. The RTP4 mRNA was detected inglandular epithelium (GE), stratum compactum (SC), and caruncular stroma (S) and increased in response to pregnancy. Photomicrographs are shown at203 magnification which represents a field width of 315 lm.

522 GIFFORD ET AL.

pregnancy following luteectomy or PGF2a-induced luteal

regression [31, 32], it is unlikely that the CL secretes anythingother than progesterone, which is essential for pregnancymaintenance. Interestingly, the CL on Day 13 of pregnancy insheep is exposed to higher basal levels of PGF

2a comparedwith the CL on Day 13 of the cycle [33]. Recent evidencesuggests that reduction of PGF

2a with flunixin meglumine 14days after breeding increases pregnancy rates in beef cows[34], possibly indicating that elevated levels of PGF

2a duringearly pregnancy can be detrimental to establishment ofpregnancy. Some evidence, however, indicates the CLbecomes more resistant to the luteolytic effects of PGF2a asearly as Day 13 of pregnancy, because higher doses of PGF2aare required to induce luteolysis in pregnant compared withnonpregnant ewes during this time [35–37]. The mechanism(s)resulting in luteal resistance is not known. Because PGF2areceptor expression [38] and mRNA encoding the receptor [39]are not downregulated, resistance likely is conferred bydisruption of second-messenger activation by PGF2a. Silva etal. [40] showed that the CL is capable of metabolizing PGF

2ato 13,14-dihydro-15-keto PGF2a metabolite (PGFM) and thatPGFM production is greater in early pregnant compared tononpregnant ewes. Moreover, levels of 15-hydroxyprostaglan-din dehydrogenase mRNA are increased at the time of lutealresistance to PGF2a [40]. Both IFNT and early pregnancyincrease ISGs in the CL [18, 19] (Ott et al., unpublishedobservations) around the time that the CL becomes resistant toPGF2a. It may be that elevated ISGs contribute to reducing thesensitivity of the CL to PGF

2a, thus promoting establishment ofpregnancy.

Here, we showed that RTP4 was increased in PBLs ofpregnant ewes. We initially identified RTP4 as a candidategene from microarray analysis of cDNA from PBLs collectedfrom pregnant dairy cattle. We postulated that many ISGs thatare regulated in the uterus during early pregnancy likewise areregulated in PBLs. If this is the case, it could provide anoninvasive method for discovering uterine genes associatedwith establishment of pregnancy or for diagnosing fertilityproblems in humans or animals. Systemic upregulation of ISGsin PBLs during early pregnancy was observed in cattle [14, 17]and sheep [15]. Previously, IFNT was not detected in thesystemic circulation or uterine venous or lymphatic drainage inpregnant ewes [16]. Recent evidence [19], however, suggeststhat a type I IFN is released from the uterus and has anendocrine action. Oliveira et al. [19] showed that bioactive IFNwas greater in the uterine vein compared to the uterine artery inpregnant ewes. Thus, IFN likely activates ISGs in the CL, andthe significance of this activation is currently under investiga-tion.

Because GPCRs are the largest superfamily of proteins thathave been identified [41], it is difficult to speculate about thefunction(s) of RTP4 in the uterus. Higher basal levels of PGF

2aexperienced during early pregnancy [33], however, indicatethat the uterine PGF

2a biosynthetic pathway is still capable ofproducing PGF2a. Oxytocin binds a GPCR and drives PGF2abiosynthesis [42]. Results from the current experiment clearlyindicate that RTP4 is induced during early pregnancy. It isintriguing to speculate that RTP4 may act in concert with otherISGs to regulate PGF2a production during early pregnancy,although to our knowledge, no evidence currently supports thisspeculation. Alternatively, RTP4 is thought to transportchemosensory GPCRs [21, 22], and it is becoming increasinglyclear that chemokines may be involved in the establishment ofpregnancy. Most chemokine receptors are GPCRs [43], andselectivity of these receptors results largely from specificityand receptor expression patterns [44]. Many chemokines and

chemokine receptors have been identified in the endometriumof pregnant and nonpregnant women [45–47] and are thoughtto orchestrate recruitment of lymphocytes to the endometrium[48]. It also has been proposed that chemokines may aid introphoblast attachment and invasion [48]. In sheep, thechemokine CXCL10 was upregulated in the endometrium ofpregnant ewes, and the receptor (CXCR3) was localized to thetrophectoderm [49]. Moreover, chemotaxis assays suggestedthat CXCL10 regulates migration and/or distribution ofperipheral blood mononuclear cells in the uterus during earlypregnancy [49]. In the current experiment, RTP4 was inducedin the endometrium and PBLs of pregnant ewes. It is possiblethat RTP4 affects chemokine receptors during early pregnancyto aid in maintaining the delicate chemokine balance. Furtherstudies are needed to determine the role(s) of RTP4 in theendometrium during early pregnancy.

The role of RTP4 or other ISGs outside the uterus duringearly pregnancy also is not clear. Recently, Kosaka et al. [50]showed peripheral blood mononuclear cells promoted attach-ment of BeWo-cell spheroids (an in vitro model for studyingattachment mechanisms to endometrial epithelial cells) toendometrial cells derived from human uteri in the lateproliferative and early secretory phases. Those authorspostulated that peripheral blood mononuclear cells may beable to induce endometrial cells to become receptive to anembryo. Furthermore, it could be postulated that systemicactivation of ISGs might be a compensatory response tocounteract progesterone-induced uterine immunosuppression/immunomodulation [51]. This area is largely unexplored,however, and further studies are needed to determine the roleof systemic activation of ISGs during early pregnancy inruminants.

In conclusion, we used global gene expression profiling inPBLs from blood samples collected during early pregnancy toidentify RTP4, a GPCR transporter, as a pregnancy-regulatedgene in ruminants. We showed that RTP4 mRNA increased inthe endometrium during early pregnancy and in endometrialcell lines treated with IFNT. In addition, we demonstrated thatRTP4 levels increased in luteal cells during early pregnancy.These studies contribute to the growing body of evidenceindicating that conceptus signaling in the uterus during earlypregnancy results in an increased expression of a large numberof ISGs in tissues outside the reproductive tract. Further studiesare needed to determine the function(s) of RTP4 duringestablishment of pregnancy in domestic ruminants.

REFERENCES

1. Spencer TE, Bazer FW. Conceptus signals for establishment andmaintenance of pregnancy. Reprod Biol Endocrinol 2004; 2:49–63.

2. Roberts RM, Liu L, Alexenko A. New and atypical families of type Iinterferons in mammals: comparative functions, structures, and evolution-ary relationships. Prog Nucleic Acid Res Mol Biol 1997; 56:287–325.

3. Farin CE, Imakawa K, Roberts RM. In situ localization of mRNA for theinterferon, ovine trophoblast protein-1, during early embryonic develop-ment of the sheep. Mol Endocrinol 1989; 3:1099–1107.

4. Helmer SD, Hansen PJ, Anthony RV, Thatcher WW, Bazer FW, RobertsRM. Identification of bovine trophoblast protein-1, a secretory proteinimmunologically related to ovine trophoblast protein-1. J Reprod Fertil1987; 79:83–91.

5. Choi Y, Johnson GA, Burghardt RC, Berghman LC, Joyce MM, TaylorKM, Stewart MD, Bazer FW, Spencer TE. Interferon regulatory factor-tworestricts expression of interferon-stimulated genes in the endometrialstroma and glandular eptithelium of the ovine uterus. Biol Reprod 2001;65:1038–1049.

6. Fleming JW, Choi Y, Johnson GA, Spencer TE, Bazer FW. Cloning of theovine estrogen receptor-alpha promoter and functional regulation by ovineinterferon-tau. Endocrinology 2001; 142:2879–2887.

7. Spencer TE, Mirando MA, Mayes JS, Watson GH, Ott TL, Bazer FW.

RTP4 IN PREGNANT AND CYCLIC EWES 523

Effects of interferon-tau and progesterone on estrogen-stimulated expres-sion of receptors for estrogen, progesterone and oxytocin in theendometrium of ovariectomized ewes. Reprod Fertil Dev 1996; 8:843–853.

8. Gray CA, Abbey CA, Beremand PD, Choi Y, Farmer JL, Adelson DL,Thomas TL, Bazer FW, Spencer TE. Identification of endometrial genesregulated by early pregnancy, progesterone, and interferon tau in the ovineuterus. Biol Reprod 2006; 74:383–394.

9. Johnson GA, Stewart MD, Gray CA, Choi Y, Burghardt RC, Yu-Lee LY,Bazer FW, Spencer TE. Effects of the estrous cycle, pregnancy, andinterferon tau on 20,50-oligoadenylate synthetase expression in the ovineuterus. Biol Reprod 2001; 59:784–794.

10. Vallet JL, Barker PJ, Lamming GE, Skinner N, Huskisson NS. A low-molecular-weight protein which is increased by ovine trophoblast protein-1 is a beta

2-microglobulin-like protein. J Endocrinol 1991; 130:R1–R4.

11. Johnson GA, Spencer TE, Burghardt RC, Joyce MM, Bazer FW.Interferon-tau and progesterone regulate ubiquitin cross-reactive proteinexpression in the ovine uterus. Biol Reprod 2000; 62:622–627.

12. Ott TL, Yin J, Wiley AA, Kim H, Gerami-Naini B, Spencer TE, Bartol FF,Burghardt RC, Bazer FW. Effects of the estrous cycle and early pregnancyon uterine expression of Mx protein in sheep (Ovis aries). Biol Reprod1998; 59:784–794.

13. Hicks BA, Etter SJ, Carnahan KG, Joyce MM, Assiri AM, Carling SJ,Kodali K, Johnson GA, Hansen TR, Mirando MA, Woods GL,Vanderwall DK, Ott TL. Expression of the uterine Mx protein in cyclicand pregnant cows, gilts, and mares. J Anim Sci 2003; 81:1552–1561.

14. Gifford CA, Racicot K, Clark DS, Austin KJ, Hansen TR, Lucy MC,Davies CJ, Ott TL. Regulation of interferon-stimulated genes in peripheralblood leukocytes in pregnant and bred, nonpregnant dairy cows. J DairySci 2007; 90:274–280.

15. Yankeey SJ, Hicks BA, Carnahan KG, Assiri AM, Sinor SJ, Kodali K,Stellflug JN, Ott TL. Expression of the antiviral protein Mx in peripheralblood mononuclear cells of pregnant and bred, nonpregnant ewes. JEndocrinol 2001; 170:R7–R11.

16. Bazer FW, Spencer TE, Ott TL. Interferon tau: a novel pregnancyrecognition signal. Am J Reprod Immunol 1996; 37:412–420.

17. Han H, Austin KJ, Rempel LA, Hansen TR. Low blood ISG15 mRNA andprogesterone levels are predictive of nonpregnant dairy cows. J Endocrinol2006; 191:505–512.

18. Spencer TE, Stagg AG, Ott TL, Johnson GA, Ramsey WS, Bazer FW.Differential effects of intrauterine and subcutaneous administration ofrecombinant ovine interferon tau on the endometrium of cyclic ewes. BiolReprod 1999; 61:464–470.

19. Oliveira JF, Henkes LE, Ashley RL, Purcell SH, Smirnova NP,Veeramachaneni DNR, Anthony RV, Hansen TR. Expression of ISGs inextrauterine tissues during early pregnancy in sheep is the consequence ofendocrine IFN-s release from the uterine vein. Endocrinology 2008; 149:1252–1259.

20. Brady AE, Limbird LE. G protein-coupled receptor interacting proteins:emerging roles in localization and signal transduction. Cell Signal 2002;14:297–309.

21. Saito H, Kubota M, Roberts RW, Chi Q, Matsunami H. RTP familymembers induce functional expression of mammalian odorant receptors.Cell 2004; 119:679–691.

22. Behrens M, Bartelt J, Reichling C, Winnig M, Kuhn C, Meyerhof W.Members of RTP and REEP gene families influence functional bitter tastereceptor expression. J Biol Chem 2006; 281:20650–20659.

23. Johnson GA, Burghardt RC, Newton GR, Bazer FW, Spencer TE.Development and characterization of immortalized ovine endometrial celllines. Biol Reprod 1999; 61:1324–1330.

24. Satterfield MC, Bazer FW, Spencer TE. Progesterone regulation ofpreimplantation conceptus growth and galectin 15 (LGALS15) in theovine uterus. Biol Reprod 2006; 75:289–296.

25. Kubista M, Andrade JM, Bengtsson M, Forootan A, Jonak J, Lind K,Sindelka R, Sjoback R, Sjogreen B, Strombom L, Stahlberg A, Zoric N.The real-time polymerase chain reaction. Mol Aspects Med 2006; 27:95–125.

26. Gasteiger E, Gattiker A, Hoogland C, Ivanyi I, Appel RD, Bairoch A.ExPASy: the proteomics server for in-depth protein knowledge andanalysis. Nucleic Acids Res 2003; 31:3784–3788.

27. Fredriksson R, Lagerstrom MC, Lundin L, Schioth HB. The G-protein-coupled receptors in the human genome form five main families:phylogenetic analysis, paralogon groups, and fingerprints. Mol Pharmacol2003; 63:1256–1272.

28. Robas N, O’Reilly M, Katugampola S, Fidock M. Maximizingserendipity: strategies for identifying ligands for orphan G-protein-coupled receptors. Curr Opin Pharmacol 2003; 3:121–126.

29. Rosenfeld CS, Han CS, Alexenko AP, Spencer TE, Roberts RM.Expression of interferon receptor subunits, IFNAR1 and IFNAR2, in theovine uterus. Biol Reprod 2002; 67:847–853.

30. Constantinescu SN, Croze E, Murti A, Wang C, Basu L, Hollander D,Russell-Harde D, Betts M, Garcia-Martinez V, Mullersman JE, PfefferLM. Expression and signaling specificity of the IFNAR chain of the type Iinterferon receptor complex. Proc Natl Acad Sci U S A 1995; 92:10487–10491.

31. Seals RC, Lemaster JW, Hopkins FM, Schrick FN. Effects of elevatedconcentrations of prostaglandin F

2alphaon pregnancy rates in progestogen-

supplemented cattle. Prostaglandins Other Lipid Mediat 1998; 56:377–389.

32. Bridges PJ, Wright DJ, Buford WI, Ahmad N, Hernandez-Fonseca H,McCormick ML, Schrick FN, Dailey RA, Lewis PE, Inskeep EK. Abilityof induced corpora lutea to maintain pregnancy in beef cows. J Anim Sci2000; 78:2942–2949.

33. Silvia WJ, Lewis GS, McCracken JA, Thatcher WW, Wilson L Jr.Hormonal regulation of uterine secretion of prostaglandin F

2alphaduring

luteolysis in ruminants. Biol Reprod 1991; 45:655–663.34. Merrill ML, Ansotegui RP, Burns PD, MacNeil MD, Geary TW. Effects

of flunixin meglumine and transportation on establishment of pregnancy inbeef cows. J Anim Sci 2007; 85:1547–1554.

35. Silvia WJ, Niswender GD. Maintenance of the corpus luteum of earlypregnancy in the ewe. III. Differences between pregnant and nonpregnantewes in luteal responsiveness to prostaglandin F

2alpha. J Anim Sci 1984;

59:746–753.36. Silvia WJ, Niswender GD. Maintenance of the corpus luteum of early

pregnancy in the ewe. IV. Changes in luteal sensitivity to prostaglandinF

2alphathroughout early pregnancy. J Anim Sci 1986; 63:1201–1207.

37. Lacroix MC, Kann G. Aspects of the antiluteolytic activity of theconceptus during early pregnancy in ewes. J Anim Sci 1986; 63:1449–1458.

38. Wiepz GJ, Wiltbank MC, Nett TM, Niswender GD, Sawyer HR.Receptors for prostaglandins F

2alphaand E

2in ovine corpora lutea during

maternal recognition of pregnancy. Biol Reprod 1992; 47:984–991.39. Juengel JL, Wiltbank MC, Meberg BM, Niswender GD. Regulation of

steady-state concentrations of messenger ribonucleic acid encodingprostaglandin F2alpha receptor in ovine corpus luteum. Biol Reprod1996; 54:1096–1102.

40. Silva PJ, Juengel JL, Rollyson MK, Niswender GD. Prostaglandinmetabolism in the ovine corpus luteum: catabolism of prostaglandin F

2alpha

coincides with resistance of the corpus luteum to PGF2alpha

. Biol Reprod2000; 63:1229–1236.

41. Vassilatis DK, Hohmann JG, Zeng H, Li F, Ranchalis JE, Mortrud MT,Brown A, Rodriguez SS, Weller JR, Wright AC, Bergmann JE, GaitanarisGA. The G protein-coupled receptor repertoires of human and mouse. ProcNatl Acad Sci U S A 2003; 100:4903–4908.

42. Gimpl G, Fahrenholz F. The oxytocin receptor system: structure, function,and regulation. Physiol Rev 2001; 81:629–683.

43. Murdoch C, Finn A. Chemokine receptors and their role in inflammationand infectious diseases. Blood 2000; 95:3032–3043.

44. Baggiolini M, Dewald B, Moser B. Human chemokines: an update. AnnuRev Immunol 1997; 15:675–705.

45. Red-Horse K, Drake PM, Gunn MD, Fisher SJ. Chemokine ligand andreceptor expression in the pregnant uterus: reciprocal patterns incomplementary cell subsets suggest functional roles. Am J Pathol 2001;159:2199–2213.

46. Kai K, Nasu K, Nakamura S, Fukuda J, Nishida M, Miyakawa I.Expression of interferon-c-inducible protein-10 in human endometrialstromal cells. Mol Hum Reprod 2002; 8:176–180.

47. Kitaya K, Nakayama T, Daikoku N, Fushiki S, Honjo H. Spatial andtemporal expression of ligands for CXCR3 and CXCR4 in humanendometrium. J Clin Endocrinol Metab 2004; 89:2470–2476.

48. Lea RG, Sandra O. Immunoendocrine aspects of endometrial function andimplantation. Reproduction 2007; 134:389–404.

49. Nagaoka K, Sakai A, Nojima H, Suda Y, Yokomizo Y, Imakawa K, SakaiS, Christenson RK. A chemokine, interferon (IFN)-c-inducible protein 10kDa, is stimulated by IFN-s and recruits immune cells in the ovineendometrium. Biol Reprod 2003; 68:1413–1421.

50. Kosaka K, Fujiwara H, Tatsumi K, Yoshioka S, Higuchi T, Sato Y,Nakayama T, Fujii S. Human peripheral blood mononuclear cells enhancecell-cell interaction between human endometrial epithelial cells andBeWo-cell spheroids. Hum Reprod 2003; 18:19–25.

51. Robertson SA. Control of the immunological environment of the uterus.Rev Reprod 2000; 5:164–174.

524 GIFFORD ET AL.