ovarian expression of chemokines and their receptors

Journal of Reproductive Immunology63 (2004) 1–9

Ovarian expression of chemokinesand their receptors

Cindy Zhou, Jason Borillo, Jean Wu,Lisa Torres, Ya-Huan Lou∗

Department of Diagnostic Science, Dental Branch, University of Texas Health Science,Center at Houston, Houston, TX 77030, USA

Received in revised form 2 March 2004; accepted 8 March 2004

Abstract

Recent studies suggest involvement of the immune system, including leukocytes and cytokines/chemokines, in various ovarian functions such as ovulation. Using the RT-PCR method, we examinedexpression of various chemokines and their receptors in normal mouse ovaries. Among seventeenexamined chemokines (17 CC types and two CXC types), expressions of CC types MCP-1 andRANTES, and CXC type IP-10 were detected at high levels, while most CC types expressed at variableor low levels. Only five chemokines were not detected in the ovary. We next examined expressionof chemokine receptors. CCR1 and CCR2, which are the receptors for MCP-1 and RANTES, werealso expressed at constitutively high levels while others were not detectable. We further showed thata significant part of expression of both detected chemokines and receptors originated from peripheralblood leukocytes (PBL) circulating in the ovary. However, ovarian tissue was the major contributorof expression. Constitutive expression of several chemokines and their receptors suggests frequentmigrations/movements of leukocytes in the ovary, which may be involved in ovarian functions otherthan ovulation.© 2004 Elsevier Ireland Ltd. All rights reserved.

Keywords: Ovary; Chemokine; Chemokine receptor; RT-PCR

1. Introduction

Immune components, such as leukocytes and cytokines, have been implicated in severalphysiological events in normal ovaries. The presence of different subsets of leukocytes, i.e.

∗ Corresponding author. Tel.:+1-713-500-4059; fax:+1-713-500-4500.E-mail address: [email protected] (Y.-H. Lou).

0165-0378/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.jri.2004.03.002

2 C. Zhou et al. / Journal of Reproductive Immunology 63 (2004) 1–9

neutrophils, macrophages, lymphocytes, and eosinophils at different phases of the ovarianlife cycle in various animal species has been documented (Brannstrom et al., 1993a, 1994;Standaert et al., 1991; Suzuki et al., 1998). A large influx of leukocytes into the ovary, espe-cially during ovulatory processes, has been shown (Brannstrom et al., 1993a). An equallyimportant influx of monocytes into the corpus luteum characterizes the late postovulatoryphase (Brannstrom et al., 1994). Migration of intra-ovarian leukocytes was also well docu-mented. The best example is infiltration of atretic follicles by ovarian macrophages (Gaytanet al., 1998).

Based on these observations, it has long been hypothesized that several physiological pro-cesses of normal ovaries including ovulation, corpus luteum formation, and atresia may havesome similarities to an inflammatory process. For example, there now exists a growing bodyof evidence suggesting that ovulation may indeed constitute a local inflammatory reaction aspostulated earlier (Espey, 1980). Thus, it is possible that leukocytes and cytokines may playkey roles in normal ovarian physiology (Adashi, 1990; Brannstrom and Norman, 1993b).It is reasonable to hypothesize that leukocytes must be attracted away from circulationand into the ovary, presumably by way of local, possibly intra-ovarian, chemoattractants.Chemokines, which are the most likely candidates for local chemoattractants, include afamily of proteins. Chemokines constitute a family of structurally related, small, inducible,secreted proinflammatory proteins involved in a variety of immune responses especiallyas chemoattractants and activators of specific types of leukocytes (Luster, 1998; Wong andFish, 2003). Approximately 40 chemokines, grouped in distinct families, are known to date.Chemokines are secreted by a variety of cells and thus appear to play important roles inmany inflammatory reactions (Horuk, 2001). Each group of chemokines has their distinctreceptors, and so far 15 chemokine receptors have been discovered. Each receptor mayserve as a ligand for multiple chemokines of the same group.

There is a growing body of evidence that chemokines may be directly or indirectly in-volved in follicular development and ovulation, as well as in corpus luteum formation,function, and demise (Garcia-Velasco and Arici, 1998). A recent study demonstrated ele-vated expression of several chemokines during the ovulatory process (Wong et al., 2002).Our objectives were to screen expression of chemokines and their receptors in normalmurine ovaries, which may provide us additional information about potential roles of ovar-ian chemokines in ovarian functions.

2. Material and methods

2.1. RT-PCR detection of chemokines

Young female CD1 mice (6–8-weeks-old) were purchased from Harlan (Indianapolis,IN). The mice were maintained in the animal facility at the University of Texas Health Sci-ence Center at Houston, and allowed to acclimate for a minimum of 1 week. Total RNA wasisolated from the ovaries using a commercial kit (Ambion, Austin, TX), and cDNA was syn-thesized using 1 g of total RNA through an RT reaction (Applied Biosystems, Branchburg,NJ). PCR was carried out to detect mRNA for various chemokines and chemokine receptors(Table 1) (PCR Applied Biosystems, Roche, NJ). In general, the cDNA was pre-denatured

C.Z

houetal./JournalofR

eproductiveIm

munology

63(2004)

1–93

Table 1Conditions for PCR to detect expression of chemokines and their receptors in ovaries

Chemokines/receptors Primers (forward/reverse) Anealingtemperature (◦C)

Cycles Expectedsize (bp)

CCRMIP-1� ACTGAAGCCAGCTCTCTCTTCCTC/TTCCTTCTTGGGGTCAGCACAGAC 68 40 416MIP-1� CCTAACCCCGAGCAACACCA/AGGGCTGAGGAGGCCTGTCC 63 40 351MIP-1� TCAACATCATGAAGGTCTCCAC/CTCAGGCAATCAGTTCCAGGTG 59 40 287MIP-3� CTGTACCAAGAGTTTGCTCC/GCACAATATATTTCACCCAAG 60 40 246MIP-3� CACCCTCCATGGCCCTGCTACT/TAACTGCTGCGGCGCTTCATCT 68 40 452MCP-1 ACTGAAGCCAGCTCTCTCTTCCTC/TTCCTTCTTGGGGTCAGCACAGAC 60 35 274MCP-2 TTGAGAGGACGCTAGCCTTCACTCC/AGACAAGGATGAGAAAACACGCAGC 68 40 454MCP-3 GGTACCACTCTCTTTCTCCACCATG/AAGCTTACAGCGGTGAGGAATTTTGC 64 40 381MCP-4 TTGTGAAATCTCCAACTCTTAACC/AAGTATTCCAAAGCATAGAAGAGG 58 40 476HCC 4∗ TAGCCTGTCTCTCCTTGTCCTCATCC/TCCACTAAAGCCTGGTCATCACTGGG 68 40 363RANTES GCCCACGTCAAGGAGTATTTCTAC/AGGACTAGAGCAAGCGATGACAGG 55 35 205MDC ATG GCT ACC CTG CGT GTC CC/CTA GGA CAG TTT ATG GAG TA 60 40 278Eotx-1 ATG CAG AGC TCC ACA GCG CT/TTAT GGTT TTG GAG TTTG GAG 57 40 293Eotx-2 CTG TGC CTG ACC TCC AGA AC/CTA AAC CAC GGT GCT ATT GC 57 40 398CCL26∗ GTCTCATGCTCATAAATAGGGG/AGCTGAGTCACAATTGTTTCGG 55 40 366TARC CAG GAA GTT GGT GAG CTG GTA TA/TTG TGT TCG CCT GTA GTG CAT A 62 40 305TCA3 GAGACAOAAACTTATCACCATG/CGCAAGCTTGGTTAGCAGOGGTT 59 40 308

CXCIP-10 GATGGCTAGTCCTAATTGCCCTTGG/CTGAGTATCTTGATAACCCCTTGGG 63 40 389SDF-1** ACGCCAAGGTCGTCGCCGTOCTGG/GTTAGGGTAATACAATTCCTTAGA 60 40 539

ReceptorsCCR1 TCAGATTTCACAGAAGCCTACCCC/TGTATAAGCCCAGGTAATAAAACC 53 40 339CCR2 ATAAGGGCTCTTOTTTGATCTTTCC/TGGCTATTCCATATACACCTTTCCC 63 40 346CCR6 CTGCAGTTCGAAGTCATC/GTCACACCACCATAATGTTG 59 40 458CCR8 ACCCACAACCTGCTGGACCAGTGG/TGGTCCTGTTGTGGTTCAGGCAGC 70 40 429

House-keepingHPRT CCTGCTGGATTACATTAAAGCACTG/GCTAAGGGCATATCCAACAACAAAC 60 25 352

∗ Primers were designed based on human sequences.


at 94◦C for 4 min, followed by the indicated number of cycles inTable 1: denaturation(94◦C, 1 min) annealing (for the temperature, seeTable 1) synthesis (72◦C, 1 min). PCRwas carried out for 25, 30, 35, and 40 cycles to determine the optimal cycles for each gene(seeTable 1). The products were separated by electrophoresis in a 1.5% agarose gel, stainedwith ethidium-bromide, and visualized under UV light illumination. RT-PCR detection ofexpression of house-keeping gene HPRT was used to determine the quantity and quality ofthe ovarian RNA. The PCR condition for HPRT is also shown inTable 1.

2.2. Isolation of peripheral blood leukocytes (PBL) and ovarian cells

Peripheral blood was collected into a test tube with heparin (1 mg) from the arterywhen normal mice were sacrificed. The blood was diluted three-fold with PBS containing0.1 mg/ml heparin, and followed by Ficol (1.119, Sigma, St. Louis) gradient centrifugation.The peripheral blood leukocytes (PBL) were collected from the interface and washed threetimes with PBS. The cells were counted and used either for flow cytometry or for isolationof total RNA.

A well established method with some modifications was applied for isolation of ovariancells (Lou and Borillo, 2003). Briefly, the ovaries were torn open and digested by collagenaseII for 10 min, followed by washing three times with a culture medium containing 10% FCS.The cells were counted and used for flow cytometry or for isolation of total RNA.

2.3. Flow cytometry

Two-color flow cytometry was used to estimate ovarian or peripheral blood CD4+/CD3+T-cell population. Briefly, the isolated normal PBL or ovarian cells (2× 105 cells/sample)were placed on ice in a culture medium containing 1% of FCS. The cells were incubatedwith PE-labeled anti-mouse CD4 monoclonal antibody (GK1.5, Pharmingen, San Diego,CA) and biotin-labeled anti-CD3 monoclonal antibody (145-2C11, Pharmingen, San Diego,CA). After intensive washing, avidin-labeled FITC was added to the cells. The stained cellswere analyzed by a flow cytometer (FACScalibur, DB, San Diego, CA).

3. Results and discussion

3.1. Expression of ovarian chemokines

The investigated chemokines included 17 CC types and two CXC types (Table 1).Chemokine IL-8 was excluded, since its expression in ovaries has been well documented(Change et al., 1998). Expressions of house-keeping gene HPRT were first determined todemonstrate the quantity and quality of the ovarian RNA. The ovarian RNA samples werethen tested by PCR amplification of the above genes without reverse transcription to ruleout potential contamination of genomic DNA. Only these ovarian RNA samples withoutgenomic DNA contamination were used for screening of chemokine expressions. The ex-pression patterns of ovarian chemokines were divided into four categories: (1) expressionat constitutively high levels; (2) expression at constitutively low levels; (3) expression at

C. Zhou et al. / Journal of Reproductive Immunology 63 (2004) 1–9 5

Fig. 1. RT-PCR detection of chemokine expression in the ovary. Each chemokine is indicated at the left; each lanerepresents one individual, and eight individuals are shown. Expression of house-keeping gene HPRT is shown todemonstrate quality and quantity of cDNA; PCR for HPRT without reverse transcription (RT-PCR) is also shownto reveal a lack of genomic DNA contamination. Chemokine expressions are categorized into four groups: (1)constitutively high levels; (2) constitutively low levels; (3) variable levels; and (4) not detectable levels.

variable levels, and (4) no detectable expression (Fig. 1). The first category (expression atconstitutively high levels) includes two CC-type chemokines, MCP-1 and RANTES, andone CXC chemokine IP-10. Interestingly, an increase in the expression levels for those threehas been observed during ovulation (Wong et al., 2002). The chemokines, which were cat-egorized as “expression at constitutively low levels”, included CC chemokines Eotx-2 andTARC. The majority of tested chemokines, including MCP-2, MCP-3, MCP-4, MIP-1�,MIP-1�, HHC-4, and Eotx-1, expressed at variable levels. Expressions of MIP-1�, MIP-3�,MDC, TCA3, and SDF-1 were not detected. The primers for CCL26 were designed basedon human sequence because murine sequence was not available. Although CCL26 was notdetected (data not shown), it is not conclusive.


Together with several previous studies, we conclude that the normal ovary expressed asurprisingly diverse array of chemokines (Wong et al., 2002). The ovaries used in our studywere randomly sampled. Thus, the ovarian chemokines may not be closely associated withtime-stringent events during reproductive cycles.

Expression of MCP-1 has been localized to the corpus luteum, suggesting its potentialroles in corpus luteum formation (Townson et al., 1996). Recently, an elevated level ofMCP-1 expression during ovulatory processes has been described (Wong et al., 2002).Based on this finding, it was postulated that MCP-l might be the chemokine, which mayattract leukocytes into ovulating follicles to mimic an inflammatory response. Our studyshowed a constantly high level of MCP-l expression in normal ovaries. We also observed asimilar expression pattern for RANTES. Although the reproductive stages of the mice werenot determined in our study, it seems that MCP-1 and/or RANTES may also be involved inovarian functions other than ovulation or corpus luteum formation. During ovarian atresia,which occurs more frequently and in a less time-stringent manner, a large number of ovarianmacrophages infiltrate atretic follicles (Gaytan et al., 1998). It is possible that MCP-l mayplay roles in this process.

It is interesting to mention ovarian expression of IP-10. IP-10, a CXC chemokine,has been known to target activated T-lymphocytes (Xanthou et al., 2003). Unlike ovar-ian macrophages, T-cells were scarce within the ovarian tissue (Brannstrom et al., 1994;Standaert et al., 1991; Suzuki et al., 1998). IP-10 may not be necessary as a chemoattractantfor T-lymphocytess in the ovary. IP-10 has been shown to play a critical role during earlygestation by regulation of blastocyst migration and adhesion (Nagaoka et al., 2003). It ispossible that IP-10 may play a role by direct regulation of migration of certain types ofovarian cells.

3.2. Expression of chemokine receptors

Based on the chemokine expression data, we decided to measure expression of theirreceptors. We first focused on two CC chemokine receptors, CCR1 and CCR2 (Table 1),which are ligands for RANTES and MCP-1. Constitutive expression of CCR1 and CCR2was detected at high levels (Fig. 2). Although it remains unknown which type of cellsexpress CCR1 and CCR2, it is reasonable to predict that ovarian macrophages may be the

Fig. 2. RT-PCR detection of ovarian expression of chemokine receptors, which are indicated at the left. Eachlane represents an ovarian sample from one individual. Expression of house-keeping gene HPRT is shown todemonstrate quality and quantity of cDNA.


source of the expression. Constitutive expression of these receptors suggests the frequentoccurrence of leukocyte migration in a normal ovary. It remains unknown, however, whetherthe migrating leukocytes originated from the circulation or are residential.

We next observed expression of CCR6 and CCR8, which are ligands for the chemokinesthat were not expressed or expressed at variable levels. These two were either not detectableor at very low levels (Fig. 2). The results were well expected, since these three receptorscorrespond to chemokines that were expressed at low levels or not expressed.

3.3. Estimation of chemokine and their receptors by ovarianperipheral blood leukocytes

Although a few studies including this study have shown the expression of variouschemokines and their receptors in normal ovaries, it remained unknown whether the ex-pression, especially the receptors, might originate from periphery blood leukocytes (PBL)which are circulating through normal ovarian tissues (Standaert et al., 1991). Thus, it wasnecessary to estimate possible expression of chemokines and their receptors by PBLs. Wefirst calculated the number of PBLs within an ovary. CD4+ cells were chosen as an index,because only a minute number of CD4+ cells have been detected in a normal ovary, which isbelieved to be in periphery blood circulation. The number of ovarian CD4+ T-cells was de-termined by flow cytometry on ovarian cells, and it was then used to estimate PBLs whichwere circulating through the ovary. Flow cytometry showed approximately 4900 CD4+T-cells in an ovary (0.44% of 1.12× 106 cells per ovary). The CD4+ T-cell population inperipheral blood from the same individuals was 2.4%. Thus, an un-fractionated 1.7 × 105

peripheral blood leukocytes× 2.4% = 4900 cells contained an equal number of CD4+cells in one ovary and were used for isolation of total RNA. The same fraction (1%) of totalRNA from either ovary or PBL was used for the RT-PCR to determine expression of CCR1,CCR2, as well as MCP-1, RANTES, and IP-10 expressions. Both CCR1 and CCR2 expres-sion were detected at significant levels (Fig. 3). However, comparison between ovarian andPBL RNA samples suggested that ovarian tissue/leukocytes were still the main sources forCCR1 and CCR2 expression, as an equivalent number of PBLs showed much lower levelsof CCR1 and CCR2 expression.

Expression of RANTES and MCP-1 in PBLs was also detected at significant levels inPBLs (Fig. 3). This observation on estimation of chemokine expressions by intra-ovarian

Fig. 3. RT-PCR detection of chemokines MCP-1, RANTES and IP-10, and receptors CCR1 and CCR2 expressionin peripheral blood leukocytes, the number of which is equivalent to these in an ovary. Expressing of HPRT is alsoshown.


PBLs argues the necessity of taking into consideration the influence of ovarian PBLs whenwe investigate immune molecules in the ovary. On the other hand, expression of IP-10 wasnot detectable in PBLs. As we discussed earlier, constant expression of IP-10 by ovariantissue suggests a role of IP-10 in certain ovarian functions.

It is crucial to determine chemokine expressions at each stage of estrous cycles, sothat we would be able to predict their potential roles in ovarian functions. Nevertheless, ourstudy demonstrated that some chemokines expressed at variable levels in randomly sampledovaries. Although it is difficult to reach any conclusions at the present time merely based onour observations as described above, those data suggest that expression of some chemokinesat variable levels may be related with the reproductive cycle. Based on our primary screenof ovarian chemokine expressions, we have been investigating the expression of severalselected chemokines at each stage of estrous cycles.

Acknowledgements

This study was supported by the NIH grant RO1 HD35993 (YHL) and internal researchsupport from the Dental Branch, University of Texas Houston Health Science Center (YHL).C. Zhou is a post doctoral trainee supported by NIH T32 HD07324.

References

Adashi, E.Y., 1990. The potential relevance of cytokines to ovarian physiology: the emerging role of residentovarian cells of the white blood cell series. Endocrinol. Rev. 11, 444–454.

Brannstrom, M., Mayrhofer, G., Robertson, S.A., 1993a. Localization of leukocyte subsets in the rat ovary duringthe periovulatory period. Biol. Reprod. 48, 277–286.

Brannstrom, M., Norman, R.J., 1993b. Involvement of leukocytes and cytokines in the ovulatory process andcorpus luteum function. Hum. Reprod. 8, 1762–1775.

Brannstrom, M., Pascoe, V., Norman, R.J., McClure, N., 1994. Localization of leukocyte subsets in the folliclewall and in the corpus luteum throughout the human menstrual cycle. Fertil. Steril. 61, 488–495.

Change, R.J., Gougeon, A., Erickson, G.F., 1998. Evidence for a neutrophil-interleukin-8 system in human fol-liculogenesis. Am. J. Obstet. Gynecol. 178, 650–657.

Espey, L.L., 1980. Ovulation as an inflammatory reaction a hypothesis. Biol. Reprod. 22, 73–106.Garcia-Velasco, J.A., Arici, A., 1998. Chemokines and human reproduction. Fertil. Steril. 71, 983–993.Gaytan, F., Morales, C., Bellido, C., Aguilar, E., Sanchez-Criado, J.E., 1998. Ovarian follicle macrophages: is

follicular atresia in the immature rat a macrophage-mediated event? Biol. Reprod. 58, 52–59.Horuk, R., 2001. Chemokine receptors. Cytokine Growth Factor Rev. 12, 313–335.Lou, Y.H., Borillo, J., 2003. Migration of T-cells from nearby inflammatory foci into antibody bound tissue: a

relay of T-cell and antibody actions in targeting native autoantigen. J. Autoimmun. 21, 27–35.Luster, A.D., 1998. Chemokines–chemotactic cytokines that mediate inflammation. N. Engl. J. Med. 338, 436–445.Nagaoka, K., Nojima, H., Watanabe, F., Chang, K.T., Christenson, R.K., Sakai, S., Imakawa, K., 2003. Regulation

of blastocyte migration, apposition, and initial adhesion by a chemokine, interferon gamma-inducible protein10 kDa (IP-10), during early gestation. J. Biol. Chem. 278, 29048–29056.

Standaert, F.E., Zamora, C.S., Chew, B.P., 1991. Quantitative and qualitative changes in blood leukocytes in theporcine ovary. Am. J. Reprod. Immunol. 25, 163–168.

Suzuki, T., Sasano, H., Takaya, R., Fukaya, T., Yajima, A., Date, F., Nagura, H., 1998. Leukocytes in normal-cyclinghuman ovaries: immunohistochemical distribution and characterization. Hum. Reprod. 13, 2186–2191.


Townson, D.H., Warren, J.S., Flory, C.M., Naftalin, D.M., Keyes, P.L., 1996. Expression of monocyte chemoat-tractant protein-1 in the corpus luteum of the rat. Biol. Reprod. 54, 513–520.

Wong, K.H.H., Negishi, H., Adashi, E.Y., 2002. Expression, hormonal regulation, and cyclic variation ofchemokines in the rat ovary: key determinants of the intraovarian residence of representatives of the whiteblood cell series. Endocrinology 143, 784–791.

Wong, M.M., Fish, E.N., 2003. Chemokines: attractive mediators of the immune response. Semin. Immunol. 15,5–14.

Xanthou, G., Williams, T.J., Pease, J.E., 2003. Molecular characterization of the chemokine receptor CXCR3:evidence for the involvement of distinct extracellular domains in a multi-step model of ligand binding andreceptor activation. Eur. J. Immunol. 33, 2927–2936.

ovarian expression of chemokines and their receptors

Documents