Molecular Human Reproduction Vol.8, No.9 pp. 849–854, 2002

Dioxin stimulates RANTES expression in an in-vitro modelof endometriosis

Dong Zhao1, Elizabeth A.Pritts1, Victor A.Chao1, Jean-Francois Savouret2 andRobert N.Taylor1,3

1Center for Reproductive Sciences, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California,San Francisco, HSE 1689, CA 94143–0556, USA and 2INSERM U 135, Hormones, Genes et Reproduction,94270 Le Kremlin-Bicetre, France

3To whom correspondence should be addressed. E-mail: rtaylor@socrates.ucsf.edu

The industrial contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is associated with inflammatory disorders in womenand other mammals. The current studies were performed to investigate the effect of TCDD on RANTES expression in anin-vitro model of endometriosis. The biochemical effects of dioxins are mediated by binding to aryl hydrocarbon receptors(AhR). This study showed that both normal and endometriotic endometrial stromal cells express AhR protein, which wasobserved to be down-regulated by 40–60% after exposure to TCDD. Treatment with TCDD for 24 h increased the luciferaseactivity of the RANTES promoter by 2.5 � 1.0-fold in stromal cells derived from normal endometrium and endometrioticimplants. When AhR were over-expressed in these cells, luciferase activity increased 6.1 � 1.4-fold, and RANTES proteinsecretion increased from undetectable to 31 � 10 pg/100 000 cells. TCDD failed to activate a RANTES construct with a mutateddioxin response element. Other AhR ligands had similar effects to TCDD on RANTES transcription and secretion. Controltransfections using tumour necrosis factor (TNF)-α and nuclear factor (NF)-κB response element reporters indicated that thesepathways are not activated by TCDD in endometrial stromal cells. This study has demonstrated that functional AhR are presentin endometrial and endometriotic stromal cells and that TCDD up-regulates the expression of RANTES, providing a possiblemechanistic link between dioxin exposure and chemokine expression in endometriosis.

Key words: AhR/dioxin/endometrial stromal cells/endometriosis/RANTES

IntroductionEndometriosis is a common gynaecological disorder, affecting5–15% of women of reproductive age (Goldman and Cramer,1990; Haney, 1990). Among theories proposed to explain thisanomaly, Sampson’s hypothesis of retrograde menstruation is mostwidely accepted (Sampson, 1927). This theory, however, falls shortin completely explaining the pathogenesis of this disease. Retrogrademenstruation occurs in 90% of women (Halme et al., 1984), andthe vast majority manifest no pathology. Therefore, other factorsmust be involved in the development of endometriosis.

The aryl hydrocarbon receptor (AhR) is a nuclear receptorwhose natural ligand has not yet been identified. Similar to steroidreceptors, AhR acts as a ligand-dependent transcription factor thatmediates a broad range of responses induced by xenobiotictoxicants including arylhydrocarbons, dioxins, furans and coplanarpolychlorobiphenyls (Fernandez-Salguero, 1996).

The AhR has multiple domains including a basic helix–loop–helix (HLH) domain near the N terminus; the basic regioncontributes to DNA binding and the HLH region to protein–protein dimerization. The C terminal segment contains a complextransactivation domain, consisting of multiple stimulatory andinhibitory subdomains (Whitlock, 1999). The toxic biochemicaleffects of dioxins are mediated by AhR binding. The ligandedAhR associates with its arylhydrocarbon receptor nuclear translocator

© European Society of Human Reproduction and Embryology 849

(ARNT) and moves from the cytoplasm to the nucleus. Theheteromeric complex acts as a signal transducer and transcriptionfactor for target genes, including cytochromes P450 (1A1, 1A2,1B1), detoxifying phase II enzymes, and growth regulatory genesinvolved in cell proliferation, differentiation and inflammation(Whitlock, 1990; Sutter et al., 1991). AhR/ARNT complexesfunction as transcription factors by binding to dioxin responseelements (DRE) within target genes (Whitlock, 1993). The mostpotent AhR ligand is 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD),a well known environmental pollutant. TCDD is an establishedcarcinogen in experimental animals (Chahoud et al., 1989; Huffet al., 1991), but its effects on humans are less well studied(Johnson, 1993; Bertazzi et al., 2001).

In monkeys, exposure to dioxin correlates directly with anincreased prevalence and severity of endometriosis (Rier et al.,1993; Yang et al., 2000). However, the true effect of TCDD onhuman endometriosis remains a matter of debate (Koninckx et al.,1994; Mayani et al., 1997; Scialli, 2001).

Endometriosis is associated with abnormal immune responses inwomen affected by this disease (Steele et al., 1984; Braun andDmowski, 1998). The peritoneal environment in these womenreveals a predominant monocytic infiltrate (Halme et al., 1987;Oosterlynck et al., 1994). RANTES is a key chemokine expressedby normal endometrium (NE) and endometriotic implant (EI)

D.Zhao et al.

stromal cells (Hornung et al., 1997). The concentration (Khorramet al., 1993) and bioactivity (Hornung et al., 2001a) of RANTESare elevated in the peritoneal fluid of women with endometriosis.We postulated that xenobiotic agents, such as TCDD, mightinfluence the immune response in women with endometriosis. Anin-vitro model of endometriosis was therefore used to investigatethe effect of TCDD on RANTES gene expression.

Materials and methods

MaterialsA total of 10 nmol/l TCDD (Radian Corp., CA, USA) was added in 0.1%ethanol carrier directly to the cells. A total of 1 µmol/l 9,10-dimethyl-1,2-benzanthracene (DMBA) and benzo(a)pyrene (BaP) (Sigma Chemicals, StLouis, MO, USA) were suspended in 0.1% ethanol and mixed beforetreatment (Casper et al., 1999). Vehicle controls (0.1% ethanol) were usedas a comparison.

SubjectsHealthy women with ovulatory menstrual cycles who had not receivedhormones or GnRH agonist therapy for at least 6 months before surgerywere enrolled in this study. Subjects undergoing elective surgery forleiomyomata uteri or other benign uterine conditions were recruited forNE biopsies. Women with laparoscopically confirmed endometriomas wererecruited for EI biopsies. Biopsies were obtained after the patients providedwritten informed consent, under a study protocol approved by the Committeeon Human Research at the University of California, San Francisco (UCSF).Endometrial tissue was obtained by Pipelle® (Cooper Surgical, Shelton,CT, USA) aspiration biopsies. All samples were obtained in the proliferativephase of the cycle and this was confirmed histologically according toestablished criteria (Noyes et al., 1950).

Cell cultureIn brief, dispersed stromal cells were separated from glandular cells anddebris by filtration through narrow gauge sieves with apertures of 40 µm.Stromal cells were plated and subcultured to eliminate contamination bymacrophages or other leukocytes in Modified Eagle’s Medium (MEM)-αsupplemented with 10% fetal calf serum (FCS), antibiotics and nucleosides.All experiments were initiated after 24 h incubation in 2.5% charcoal-stripped FCS. Extensive characterization of cell cultures prepared usingthis protocol previously confirmed �95% purity with cells retainingfunctional steroid receptors and cytoskeletal markers of their endometrialstromal origin (Ryan et al., 1994).

ImmunocytochemistryNE and EI stromal cells were plated onto Lab-Tek (Miles Laboratories,Naperville, IL, USA) 8-chamber slides and stained using the Vectastain EliteABC Kit (Vector Laboratories, Burlingame, CA, USA). Immunoperoxidasestaining was performed using mouse monoclonal IgG antibodies againsthuman vimentin (1:100 dilution; Sigma), rabbit polyclonal IgG antibodiesagainst human cytokeratin (1:200 dilution; Sigma) and AhR (2 µg/ml;Santa Cruz Biotechnology, CA, USA). Non-immune, species-specificantibodies at the same concentrations were used as controls.

Western blotsMedia were removed by aspiration, the cells were washed with 10 mlcold Tris-buffered saline (TBS), and scraped into 500 µl of extractionbuffer (20 mmol/l Tris, pH 8.0, 137 mmol/l NaCl, 10% glycerol, 1%Triton X-100, 2 mmol/l EDTA, 1 mmol/l Pefablock, 2 µmol/l leupeptin,0.14 IU/ml aprotinin, 1 mmol/l vanadate). Cells were lysed by repeatedfreeze–thaw cycles and centrifuged at 15 000 g for 15 min at 4°C. Super-natants of extracts were denatured in sample buffer and 100 µg totalprotein sample was electrophoresed on 8% polyacrylamide sodium dodecylsulphate (SDS) gels at 30 mA for 2 h. Proteins were transferred toImmobilon PVDF membranes (Millipore) for 2 h in a semi-dry transfersystem (Hoefer Scientific Instruments, San Francisco, CA, USA). Membraneswere blocked for 1 h in TBS-Tween plus 5% instant non-fat dry milk,

850

then probed with 1.0 µg/ml anti-AhR polyclonal antibody H-211 (SantaCruz Biotechnology) at 4°C overnight. The membranes were washed inthe same solution before addition of 1:5000 dilution of horseradishperoxidase-conjugated anti-rabbit IgG, and specific bands were visualizedby the ECL (Amersham) method on Kodak X-omat film. Bands werequantified by transmission scanning using NIH Image 1.60 software.

RT–PCRTotal RNA was extracted from stromal cell cultures (NE and EI) usingthe TRIzol reagent kit (Gibco BRL, Gaithersburg, MD, USA). Theconditions for preparing the internal standards, the RT–PCR reaction andthe primers used for semi-quantitative RT–PCR have been describedpreviously (Hornung et al., 1997). Briefly, specific oligonucleotide primerswere designed to amplify sequences from human RANTES mRNA (193bp) and human GAPDH (240 bp) as a positive control (Ercolani et al.,1988) and reaction products were separated on 2% agarose gels.

ELISAA specific sandwich ELISA was used to quantify RANTES in conditionedmedia (Quantikine; R&D Systems, MN, USA). Aliquots of culturesupernatants were each tested in duplicate at several dilutions to ensurelinearity and compared with reference standards of recombinant humanRANTES. In our laboratory, the assay is sensitive to 8 pg/ml, with intra-and inter-assay coefficients of variation of 3 and 7% respectively. Theassay is specific for human RANTES, with no known cross-reactivity withother cytokines or chemokines (R&D Systems Catalogue, 2001).

Gene promoters and luciferase assaysA full length (961 bp) fragment of the human RANTES promoter (Nelsonet al., 1993) was subcloned into the pGL2 vector (Promega, Madison, WI,USA). The DRE of the wild-type RANTES promoter CACGCG at nucleotidepositions –821/–816 was mutated to CTCGAG with QuickChange® site-directed mutagenesis kits (Stratagene) using oligonucleotides containingthe desired mutation. Nuclear factor (NF)-κB and tumour necrosis factor(TNF)-α response element reporter constructs were gifts from Dr DaleLeitman (UCSF). The recombinant human (r)AhR vector [psG5-(hu)AhR]was engineered in the laboratory of Dr Jean-Francois Savouret (INSERMUnite 135, Bicetre, France) from a sequence generously provided by DrChris Bradfield (McArdle Cancer Research Center, University of Wisconsin,USA). All constructs and mutants were sequenced by the UCSF BiomolecularResource Center to verify that the correct sequences and mutationswere present.

Transient transfections were performed in human NE and EI stromalcells grown in MEM-α with 2.5% FCS and antibiotics in 12-well platesat ~50% confluence. A total of 0.45 µg of pGL2-RANTES promoter(firefly luciferase, experimental reporter) was added to each well usingQIAGEN Effectene® reagent (Valencia, CA, USA). The RANTES promotertransfection efficiencies were normalized to an independent control plasmid(0.1 µg renilla luciferase reporter) co-transfected simultaneously. In someexperiments, AhR was over-expressed by co-transfection of an expressionvector (0.25 µg/well). Salmon sperm DNA was added to normalize totaltransfected DNA in each well. The results are presented as a percentageof luciferase activity between treated and untreated cells after correctingfor transfection efficiency. Each reporter vector was assayed in at leastthree independent cultures. The same amount of empty pGL2 vector wasanalysed in pilot experiments as a control and revealed low basal activity.

Data are presented as mean � SD. Comparisons between treatmentswere evaluated using Wilcoxon non-parametric tests with StatView software.Two-tailed tests with P � 0.05 were considered significant.

ResultsImmunocytochemistry revealed that NE and EI stromal cellsexpress AhR. The receptor protein was localized in both thecytoplasm and nuclei of endometrial stromal cells (Figure 1).Vimentin and cytokeratin served as positive and negative controlsrespectively for the immunocytochemistry analyses.

The specificity of the AhR immunolocalization was confirmed

AhR activation induces RANTES expression

Figure 1. Immunocytochemisty of endometrial stromal cells. (A,B) Normal endometrium (NE) and endometriotic implant (EI) stromal cells stained withanti-vimentin antibody as a positive control. (C,D) NE and EI stromal cells stained with anti-cytokeratin antibody as a negative control. (E,F) NE andEI stromal cells stained with anti-AhR antibody; AhR is present in both the cytoplasm and nucleus.

Figure 2. TCDD exposure decreases the amount of AhR protein.Western blot of normal endometrium (NE) and endometriotic implant(EI) stromal cells lysates, probed for human AhR (110 kDa). Cells weretreated with or without TCDD (10 nmol/l) for 2 days. Stromal cellstransfected with rAhR vectors were included as positive controls.

by Western immunoblots. Western blot analysis of samples fromisolated endometrial stromal cells revealed a dominant bandcorresponding to 110 kDa as reported for the human AhR (Probstet al., 1993; Wang et al., 1995). Treatment with TCDD caused a40–60% decrease in AhR protein in NE and EI stromal cells(Figure 2).

To assess the expression of RANTES mRNA in culturedendometrial stromal cells treated with and without TCDD for 48 h,

851

Figure 3. TCDD up-regulates RANTES mRNA. Semi-quantitativeRT–PCR of normal endometrium (NE) and endometriotic implant (EI)stromal cells treated with or without TCDD. TCDD treatment increasedRANTES mRNA over the basal amount.

we used semi-quantitative RT–PCR assays. TCDD treatmentincreased the expression of RANTES mRNA in NE and EI stromalcells by ~2-fold (Figure 3). Intron-spanning primers used to amplifytranscripts of a constitutive gene, GAPDH, indicated that the RNA

D.Zhao et al.

Figure 4. TCDD treatment increases RANTES protein secretion.Supernatants of normal endometrium (NE) and endometriotic implant(EI) stromal cells treated with TCDD were analysed for RANTESproduction by ELISA. TCDD treatment increased the secretion ofRANTES protein by stromal cells transfected with rAhR, but not innon-transfected cells. *P � 0.05 compared with basal RANTESsecretion.

Figure 5. TCDD activates RANTES gene expression. Luciferase reportergene assay in endometrial stromal cells using the wild-type RANTESpromoter reporter construct (–961 bp, 0.3 µg/well). Transfectionefficiency was normalized by renilla luciferase. In the absence ofrAhR vectors, TCDD increased luciferase activity by 2.5 � 1.0-fold,*P � 0.05 compared with control RANTES expression. In the presenceof rAhR vectors, TCDD treatment increased luciferase activity by 6.1 �1.4-fold, *P � 0.05 compared with controls. Even in the presence ofrAhR, TCDD did not increase luciferase activity from a RANTESconstruct with a mutated DRE site.

preparations were of good quality and not contaminated bygenomic DNA.

With over-expression of AhR (Figure 2, lanes 3 and 6),enhanced production of RANTES protein was detected in endo-metrial stromal cells treated with TCDD for 48 h; RANTESsecretion by NE and EI stromal cells was increased fromundetectable to 31 � 10 pg/100 000 cells (n � 3, P � 0.01,Figure 4).

To explore the role of specific domains within the RANTESgene promoter, endometrial stromal cells were transiently transfectedwith wild-type RANTES promoter constructs cloned upstream ofthe firefly luciferase reporter gene.

For the full length RANTES construct (961 bp of 5� flankingDNA), containing a single DRE consensus motif, 24 h treatmentwith TCDD increased luciferase activity by 2.5 � 1.0-fold (P �0.05) in stromal cells from both NE (n � 3) and EI (n � 3)sources. When rAhR was over-expressed in these cells, theluciferase activity increased 6.1 � 1.4-fold (P � 0.05, Figure 5).To further investigate the role of this DRE site, we used site-directed mutagenesis to disrupt DRE in the RANTES promoterconstruct (Figure 5). With the mutated DRE construct (Figure 6),

852

Figure 6. Cartoon of RANTES reporter construct. –961 wild-type (wt)construct reveals one Dioxin Responsive Element (DRE) at –821 bp.The mutations introduced into the DRE site of the mutant construct areshown.

Figure 7. TCDD fails to activate NF-κB and TNF response elements.(A) A total of 0.3 µg NF-κB responsive element reporter construct wereco-transfected into endometrial stromal cells. TCDD treatment had noeffect on NF-κB signalling. (B) A total of 0.3 µg TNF-responsiveelement reporter constructs were co-transfected into endometrial stromalcells. TCDD treatment had no effect on TNF-RE activation.

TCDD failed to increase luciferase activity even when rAhR wasover-expressed (n � 3, Figure 5).

TCDD exposure had no effect on endometrial stromal cellstransiently transfected with NF-κB or TNF-α response elementconstructs in the presence of rAhR. The relative luciferase activitieswere 1.0 � 0.2-fold (n � 5, Figure 7A) and 1.2 � 0.1-fold (n �5, Figure 7B) respectively. Neither of the results were statisticallysignificant. By contrast, TNF-α stimulated the TNF-α responseelement luciferase construct �3-fold (data not shown).

Cells treated with a combination of DMBA/BaP (alternativeAhR ligands) showed similar effects on RANTES as determinedby both luciferase activity and protein secretion (data not shown).

DiscussionDioxins are a family of halogenated heterocyclic hydrocarbon by-products of industrial processes such as paper bleaching andherbicide synthesis. Dioxins directly alter the expression of target

AhR activation induces RANTES expression

genes (Whitlock, 1990) or may act as anti-estrogens or weakestrogens through interference with the estrogen receptor (Safeet al., 1991). The increased incidence of endometriosis inpopulations from industrialized countries has suggested a possiblelink with exposure to dioxins and other AhR ligands (Rier et al.,1993; Yang et al., 2000). However, social as well as environmentalfactors could be responsible for this association. Later age at firstpregnancy, shorter duration of breast-feeding and smaller familysize may contribute to the risk of endometriosis. While a studyby Pauwels et al. showed an increased odds ratio of 4.3 for theassociation of serum dioxin equivalents and endometriosis, thisfailed to reach statistical significance (Pauwels et al., 2001).Whether endometriosis is associated with high exposure to dioxin-like compounds remains to be proven.

Our study examined the presence and activation of AhR incultures of human NE and EI stromal cells. We found AhR proteinto be localized in both cytoplasm and nuclei of endometrial stromalcells. Western blotting indicated that AhR protein concentrationswere decreased by TCDD treatment, reportedly through theubiquitin–proteasome pathway (Pollenz, 1996; Ma and Baldwin,2000). Our findings are of interest with respect to a recent study,which reported high levels of AhR mRNA in endometriotic ovariancysts (Khorram et al., 2002). From these and our results,we suggest that the relationship between dioxin exposure andendometriosis is not direct. The co-expression of other factors,such as cellular AhR and ARNT are necessary to activate thetranscription of target genes. The potential AhR modulator7-ketocholesterol (Savouret et al., 2001) might also influence theeffect of dioxin exposure in this pathway.

At least two phenomena appear to limit RANTES productionin our in-vitro model. One is the down-regulation of AhR byTCDD (Ma and Baldwin, 2000) and the other is the spontaneousloss of nuclear receptors in cultured cells (Sadovsky et al., 1992).To restore TCDD responsiveness in our cells, rAhR expressionvectors were constructed and transfected. Under this condition,TCDD treatment for 48 h dramatically increased RANTES secretion.A combination of DMBA and BaP, also ligands for AhR, causeda similar stimulation in RANTES secretion. This finding supportsa generalized effect of AhR ligands on this inflammatory pathway.

To investigate whether the induction of RANTES secretion istranscriptionally regulated, we established a model system oftransiently transfected human RANTES promoter-luciferase reporterconstructs in primary human NE and EI stromal cells. Treatmentwith TCDD increased luciferase activity up to 6-fold. A similarstimulation was noted using DMBA and BaP. However, when wemutated the DRE site, TCDD treatment no longer increasedRANTES expression. These data indicate that TCDD up-regulatesRANTES gene expression, and that this effect is both AhR- andDRE-dependent.

A previous study from our group showed that NF-κB activationis responsible for interleukin-1β stimulation of RANTES geneexpression in endometriotic stromal cells (Lebovic et al., 2001).It was also reported that TNF-α treatment can increase RANTESmRNA and protein production in endometrial cells (Altman et al.,1999; Hornung et al. 2001b). To test the hypothesis that TCDDcould directly activate NF-κB or TNF-α and thereby stimulateRANTES gene expression, we transiently transfected NF-κB orTNF-α response element luciferase constructs into endometrialstromal cells. Exposure to TCDD failed to activate NF-κB orTNF-α response elements in the absence or presence of over-expressed AhR. These findings suggest that the increase inRANTES expression by AhR complexes is not mediated throughNF-κB or TNF-α pathways.

853

In conclusion, the combination of TCDD and AhR up-regulatesthe expression of RANTES via a pathway that appears to beDRE-dependent and independent of TNF-α and NF-κB activation.Other AhR ligands have similar effects. These findings provide amechanistic link between xenobiotic exposure, macrophage recruit-ment and inflammation in endometriosis.

AcknowledgementsThe immunocytochemistry expertise of Ms Emily Ricke is gratefullyappreciated. This work was supported by a National Institute of Healthgrant from the NICHD, through co-operative agreement U54-HD37321, aspart of the Specialized Cooperative Centers Program in ReproductionResearch.

ReferencesAltman, G.B., Gown, A.M., Luchtel, D.L. and Baker, C. (1999) RANTES

production by cultured primate endometrial epithelial cells. Am. J.Reprod. Immunol., 42, 168–174.

Bertazzi, P.A., Consonni, D., Bachetti, S., Rubagotti, M., Baccarelli, A.,Zocchetti, C. and Pesatori, A.C. (2001) Health effects of dioxin exposure:a 20-year mortality study. Am. J. Epidemiol., 153, 1031–1044.

Braun, D.P. and Dmowski, W.P. (1998) Endometriosis: abnormalendometrium and dysfunctional immune response. Curr. Opin. Obstet.Gynecol., 10, 365–369.

Casper, R.F., Quesne, M., Rogers, I.M., Shirota, T., Jolivet, A., Milgrom,E. and Savouret, J.F. (1999) Resveratrol has antagonist activity on thearyl hydrocarbon receptor: implications for prevention of dioxin toxicity.Mol. Pharmacol., 56, 784–790.

Chahoud, I., Krowke, R., Schimmel, A., Merker, H.J. and Neubert,D. (1989) Reproductive toxicity and pharmakokinetics of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Arch. Toxicol., 63, 432–439.

Ercolani, L., Florence, B., Denaro, M. and Alexander, M. (1988) Isolationand complete sequence of a functional human glyceraldehyde-3-phosphatedehydrogenase gene. J. Biol. Chem., 263, 15335–15341.

Fernandez-Salguero, P.M., Hilbert, D.M., Rudikoff, S., Ward, J.M. andGonzalez, F.J. (1996) Aryl-hydrocarbon receptor-deficient mice areresistant to 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced toxicity. Toxicol.Appl. Pharmacol., 140, 173–179.

Goldman, M.B. and Cramer, D.W. (1990) The epidemiology of endometriosis.Prog. Clin. Biol. Res., 323, 15–31.

Halme, J., Hammond, M.G., Hulka, J.F., Raj, S.G. and Talbert, L.M.(1984) Retrograde menstruation in healthy women and in patients withendometriosis. Obstet. Gynecol., 64, 151–154.

Halme, J., Becker, S. and Haskill, S. (1987) Altered maturation andfunction of peritoneal macrophages: possible role in pathogenesis ofendometriosis. Am. J. Obstet. Gynecol., 156, 783–789.

Haney, A.F. (1990) Etiology and histogenesis of endometriosis. Prog. Clin.Biol. Res., 323, 1–14.

Hornung, D., Ryan, I.P., Chao, V.A., Vigne, J.L., Schriock, E.D. andTaylor, R.N. (1997) Immunolocalization and regulation of the chemokineRANTES in human endometrial and endometriosis tissues and cells. J.Clin. Endocrinol. Metab., 82, 1621–1628.

Hornung, D., Bentzien, F., Wallwiener, D., Kiesel, L. and Taylor, R.N.(2001a) Chemokine bioactivity of RANTES in endometriotic and normalendometrial stromal cells and peritoneal fluid. Mol. Hum. Reprod., 7,163–168.

Hornung, D., Klingel, K., Dohrn, K., Kandolf, R., Wallwiener, D. andTaylor, R.N. (2001b) Regulated on Activation, Normal T-cell-Expressedand -Secreted mRNA expression in normal endometrium and endometrioticimplants: assessment of autocrine/paracrine regulation by in situhybridization. Am. J. Pathol., 158, 1949–1954.

Huff, J.E., Salmon, A.G., Hooper, N.K. and Zeise, L. (1991) Long-term carcinogenesis studies on 2,3,7,8-tetrachlorodibenzo-p-dioxin andhexachlorodibenzo-p-dioxins. Cell Biol. Toxicol., 7, 67–94.

Johnson, E.S. (1993) Important aspects of the evidence for TCDDcarcinogenicity in man. Environ. Health Perspect., 99, 383–390.

Khorram, O., Taylor, R.N., Ryan, I.P., Schall, T.J. and Landers, D.V.(1993) Peritoneal fluid concentrations of the cytokine RANTES correlatewith the severity of endometriosis. Am. J. Obstet. Gynecol., 169,1545–1549.

D.Zhao et al.

Khorram, O., Garthwaite, M. and Golos, T. (2002) Uterine and ovarianaryl hydrocarbon receptor (AHR) and aryl hydrocarbon receptor nucleartranslocator (ARNT) mRNA expression in benign and malignantgynaecological conditions. Mol. Hum. Reprod., 8, 75–80.

Koninckx, P.R., Braet, P., Kennedy, S.H. and Barlow, D.H. (1994) Dioxinpollution and endometriosis in Belgium. Hum. Reprod., 9, 1001–1002.

Lebovic, D.I., Chao, V.A., Martini, J.F. and Taylor, R.N. (2001) Interleukin-1 beta induction of RANTES chemokine gene expression in endometrioticstromal cells depends on an NF-κB site in the proximal promoter. J.Clin. Endocrinol. Metab., 86, 4759–4764.

Ma, Q. and Baldwin, K.T. (2000) 2,3,7,8-Tetrachlorodizenzo-p-dioxin-induced degradation of Aryl Hydrocarbon Receptor (AhR) by theubiquitin-proteasome pathway. J. Biol. Chem., 275, 8432–8438.

Mayani, A., Barel, S., Soback, S. and Almagor, M. (1997) Dioxinconcentrations in women with endometriosis. Hum. Reprod., 12, 373–375.

Nelson, P.J., Kim, H.T., Manning, W.C., Goralski, T.J. and Krensky, A.M.(1993) Genomic organization and transcriptional regulation of theRANTES chemokine gene. J. Immunol., 151, 2601–2612.

Noyes, R.W., Hertig, A.T. and Rock, J. (1950) Dating the endometrialbiopsy. Fertil. Steril., 1, 3–25.

Oosterlynck, D.J., Meuleman, C., Lacquet, F.A., Waer, M. and Koninckx,P.R. (1994) Flow cytometry analysis of lymphocyte subpopulations inperitoneal fluid of women with endometriosis. Am. J. Reprod. Immunol.,31, 25–31.

Pauwels, A., Schepens, P.J., D’Hooghe, T., Delbeke, L., Dhont, M.,Brouwer, A. and Weyler, J. (2001) The risk of endometriosis andexposure to dioxins and polychlorinated biphenyls: a case–control studyof infertile women. Hum. Reprod., 16, 2050–2055.

Pollenz, R.S. (1996) The aryl-hydrocarbon receptor, but not the aryl-hydrocarbon receptor nuclear translocator protein, is rapidly depleted inhepatic and non hepatic culture cells exposed to 2,3,7,8-tetrachlorodibenzp-p-dioxin. Mol. Pharmacol., 49, 391–398.

Probst, M.R., Reisz-Porszasz, S., Agbunag, R.V., Ong, M.S. and Hankinson,O. (1993) Role of the aryl hydrocarbon receptor nuclear translocatorprotein in aryl hydrocarbon (dioxin) receptor action. Mol. Pharmacol.,44, 511–518.

Rier, S.E., Martin, D.C., Bowman, R.E., Dmowski, W.P. and Becker, J.L.(1993) Endometriosis in rhesus monkeys (Macaca mulatta) followingchronic exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Fund. Appl.Toxicol., 21, 433–441.

854

Ryan, I., Schriock, E., Taylor, R.N. (1994) Isolation, characterization andcomparison of human endometrial and endometriosis cells in vitro. J.Clin. Endocrinol. Metab., 78, 642–649.

Sadovsky, Y., Riemer, R.K. and Roberts, J.M. (1992) The concentrationof estrogen receptors in rabbit uterine myocytes decreases in culture.Am. J. Obstet. Gynecol., 167, 1631–1635.

Safe, S., Astroff, B., Harris, M., Zacharewski, T., Dickerson, R., Romkes,M. and Biegel, L. (1991) 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD)and related compounds as antioestrogens: characterization and mechanismof action. Pharmacol. Toxicol., 69, 400–409.

Sampson, J. (1927) Peritoneal endometriosis due to menstruationdissemination of endometrial tissue into the peritoneal cavity. Am. J.Obstet. Gynecol., 14, 422–429.

Savouret, J.F., Antenos, M., Quesne, M., Xu, J., Milgrom, E. and Casper,R.F. (2001) 7-ketocholesterol is an endogenous modulator for thearylhydrocarbon receptor. J. Biol. Chem., 276, 3054–3059.

Scialli, A.R. (2001) Tampons, dioxins and endometriosis. Reprod. Toxicol.,15, 231–238.

Steele, R.W., Dmowski, W.P. and Marmer, D.J. (1984) Immunologic aspectsof human endometriosis. Am. J. Reprod. Immunol., 6, 33–36.

Sutter, T.R., Guzman, K., Dold, K.M. and Greenlee, W.F. (1991) Targetsfor dioxin: genes for plasminogen activator inhibitor-2 and interleukin-1 beta. Science, 254, 415–418.

Wang, X., Thomsen, J.S., Santostefano, M., Rosengren, R., Safe, S.,Perdew, G.H. (1995) Comparative properties of the nuclear arylhydrocarbon (Ah) receptor complex from several human cell lines. Eur.J. Pharmacol., 293, 191–205.

Whitlock, J.P. Jr (1990) Genetic and molecular aspects of 2,3,7,8-tetrachlorodibenzo-p-dioxin action. Annu. Rev. Pharmacol. Toxicol., 30,251–277.

Whitlock, J.P. Jr (1993) Mechanistic aspects of dioxin action. Chem. Res.Toxicol., 6, 754–763.

Whitlock, J.P. Jr (1999) Induction of cytochrome P4501A1. Annu. Rev.Pharmacol. Toxicol., 39, 103–125.

Yang, J.Z., Agarwal, S.K. and Foster, W.G. (2000) Subchronic exposureto 2,3,7,8-tetrachlorodibenzo-p-dioxin modulates the pathophysiology ofendometriosis in the cynomolgus monkey. Toxicol. Sci., 56, 374–381.

Submitted on February 8, 2002; accepted on June 20, 2002