{"paper_id":"e7734263-b4f6-44b7-ade3-cc0be6a10cde","body_text":"Cycling and early pregnant endometrium as a site of\nregulated expression of the vitamin D system\nP Viganò, D Lattuada, S Mangioni, L Ermellino 1, M Vignali2, E Caporizzo3,\nP Panina-Bordignon3, M Besozzi1 and A M Di Blasio1\nDepartment of Obstetrics, Gynecology and Neonatology, Fondazione Policlinico-Mangiagalli-Regina Elena Hospital, University of Milano, Milano , Italy\n1Pathology Laboratory and Molecular Biology Laboratory, Istituto Auxologico Italiano, Via Zucchi, 18-Cusano Milanino, Milano, Italy\n2Dept of Obstetrics and Gynecology, Clinica Macedonio Melloni, University of Milano, Milano, Italy\n3BioXell S.p.A, Milano, Italy\n(Requests for offprints should be addressed to A M Di Blasio; Email: a.diblasio@auxologico.it)\nAbstract\nIn addition to its calciotropic function, the secosteroid 1,25-dihydroxyvitamin D 3 (1,25(OH)2D3), has potent\nanti-proliferative/immunomodulatory effects on various tissues. Consistently, the enzyme that catalyzes the synthesis\nof 1,25(OH)\n2D3,1 /afii9825-hydroxylase (1/afii9825-OHase) and the vitamin D receptor have a widespread tissue distribution. Among\nsite-specific functions, the hormone has been suggested to be involved in uterine physiology. However, molecular\nanalysis of the vitamin D system in normal endometrium throughout the menstrual cycle as well as its regulation in the\ncontext of endometrial physiological and pathological events have received very limited attention. Thus, we have\nstudied expression, localization and regulation of 1 /afii9825-OHase in human cycling and early pregnant endometrium. The\ncapacity for 1 /afii9825-hydroxylation and the presence of vitamin D receptor in endometrial cells have also been evaluated.\nThe functional significance of these findings has been tested by evaluating gene expression of the catabolic enzyme,\nvitamin D 24-hydroxylase, and of the adhesion protein, osteopontin. Finally, to verify any potential dysfunction of the\nvitamin D system in endometriosis, a reproductive disease characterized by immune-mediated anomalies, we have\nanalyzed expression of 1 /afii9825-OHase in both eutopic and ectopic endometrium of affected patients. Results obtained\nshowed that the active form of the 1 /afii9825-OHase gene was expressed in human endometrial stromal cells independent of\nthe cycle phase but with a significant increase in early pregnant decidua. A similar profile was observed for the protein,\nwhich was abundantly expressed in the cytoplasm of both endometrial stroma and epithelial glands. Both cycling and\nearly pregnant endometrial cells also expressed the vitamin D receptor. In the same cells, 1 /afii9825-OHase mRNA levels\nwere significantly stimulated by the pro-inflammatory cytokine interleukin (IL)-1 /afii9826(50 and 500 pg/ml) while addition of\nthe active form of the hormone could modulate both CYP24 and osteopontin gene expression. The 1 /afii9825-OHase gene\nwas also expressed in ectopic endometrium and its levels were increased in proliferative phase cultures derived from\npatients with endometriosis. Human cycling endometrium may be included among the extrarenal sites able to\nsynthesize vitamin D. The IL-1 /afii9826−mediated induction of 1 /afii9825-OHase gene and the hormonal modulation of osteopontin\nsupport a role for the hormone in the immunological mechanisms underlying uterine function. Abnormalities of this\nsystem are present in endometriosis.\nJournal of Molecular Endocrinology (2006) 36, 415–424\nIntroduction\nVitamin D is a major regulator of mineral ion\nhomeostasis by facilitating calcium and phosphate\nuptake in the gut and by modulating bone cell\ndevelopment and action (Kumar 1980, Jones et al. 1998).\nHowever, the most potent metabolite of vitamin D, the\nsecosteroid 1,25-dihydroxyvitamin D (1,25(OH)\n2D3), has\nalso been demonstrated to a ffect a wide range of\nfunctions not immediately linked to calcium homeosta-\nsis, suggesting a much broader physiological impact of\nthe hormone in the body than originally envisioned\n(Zehnder et al. 2001). Specifically, 1,25(OH)\n2D3 has been\nshown to exert anti-proliferative and immunosuppressive\neffects on several cell types (Peehl et al.1994, Lemire et al.\n1995, Muller & Bendtzen 1996, Long & Santos 1999) .\nThe importance of 1,25(OH) 2D3 as a pleiotropic\nmodulator of tissue functions is strengthened by the\npresence of key components of vitamin D metabolism at\ndifferent sites. The cognate nuclear receptor for\n1,25(OH)\n2D3 (VDR) is ubiquitous in proliferating cells\n(Hewison et al. 2000) and the mitochondrial cytocrome\nP450 enzyme 25-hydroxyvitamin D 3-1/afii9825-hydroxylase\n(1/afii9825-OHase), which catalyzes the synthesis of\n1,25(OH)2D3 from its precursor 25-hydroxyvitamin D 3,\nis expressed in the kidney and in several extrarenal\ntissues (Zehnder et al. 2001, 2002 b). The widespread\nco-expression of VDR and 1 /afii9825-OHase emphasizes a\n415\nJournal of Molecular Endocrinology (2006) 36, 415–424\n0952–5041/06/036–415 © 2006 Society for Endocrinology Printed in Great Britain\nDOI: 10.1677/jme.1.01946\nOnline version via http://www.endocrinology-journals.org\nDownloaded from Bioscientifica.com at 06/11/2026 02:24:47AM\nvia free access\n\n\nputative role for 1,25(OH) 2D3 as an autocrine/paracrine\nagent with diverse physiological functions (Hewison et al.\n2000).\nIn the late 1980s and early 1990s, particular attention\nwas directed towards the potential significance of the\nvitamin D endocrine system in human reproductive\nprocesses. Increased circulating levels of 1,25(OH)\n2D3\nwere observed during estroprogestins treatment as well\nas during human gestation (Kumar 1980, Hartwell et al.\n1990, Salle et al. 2000). Moreover, uterine and placental\ncells were shown to synthesize 1,25(OH)\n2D3 (Acker et al.\n1982, Delvin et al. 1985, Kachkache et al. 1993) and a\nsingle report demonstrated elevated serum levels of the\nhormone in a specific pathological condition of the\nreproductive system, endometriosis (Hartwell et al. 1990),\nwhich is characterized by immune system anomalies\n(Vignali et al. 2002).\nIn the last years, interest in this specific topic has been\ngreatly increased for two reasons: (i) the development of\nmice deficient in VDR or 1 /afii9825-OHase has revealed that\nthese animals show impaired ovarian folliculogenesis\nand uterine hypoplasia (Yoshizawa et al. 1997, Kinuta\net al. 2000, Panda et al. 2001) and (ii) vitamin D seems to\ninfluence the reproductive system not only at the ovarian\nlevel but also at the endometrial level. Indeed, some of\nthe genes recently identified to be uniquely regulated at\nthe site of embryo implantation, for instance calbindin-\nD9k in the mouse and the homeobox ( Hox) A10 gene in\nhumans, are vitamin D-dependent proteins (Rots et al.\n1998, Nie et al. 2000, Salamonsen et al. 2002, Du et al.\n2005).\nTo further investigate the significance of local\n1,25(OH)\n2D3 production in the human uterus, in the\npresent study we have analyzed endometrial expression\nof 1 /afii9825-OHase at both the mRNA and protein levels\nduring the menstrual cycle and in early pregnancy.\nLocal regulation of the enzyme by factors physiologically\npresent at the fetal–maternal interface has also been\naddressed. Furthermore, we have investigated whether\nendometrial 1 /afii9825-OHase expression could e ffectively\nresult in the synthesis of the hormone from its precursor\n25(OH)D\n3, verified the presence of the hormone\nreceptor in endometrium and analyzed the functional\neffects of these results by evaluating the induction of\nspecific target genes. Finally, to confirm the potential\nrole of the vitamin D system in endometriosis, we have\nevaluated 1 /afii9825-OHase expression in both eutopic and\nectopic endometrium of a ffected patients.\nMaterials and methods\nReagents\nProgesterone and 25(OH)D 3 were purchased from\nSigma (Milano, Italy). Human recombinant interleukin\n(IL)-1/afii9826and tumor necrosis factor- /afii9825(TNF-/afii9825) were\nobtained from Amersham Biosciences (Amersham,\nCologno Monzese, Italy). Culture medium consisted of\nHam’s F-10 (Sigma, Italy) supplemented with 2 mM\n-glutamine (Sigma, Italy), antibiotics, fungizone and\n10% heat-inactivated FCS (Sigma, Italy). Collagenase A\nwas purchased from Roche (Milano, Italy) and\nhyaluronidase from Sigma (Mountain View, CA, USA).\nAll reagents for real-time quantitative PCR were from\nApplied Biosystems (Foster City, CA, USA). Primers\nfor 1 /afii9825-OHase and hypoxanthine phosphoribosyl-\ntransferase-1 (HPRT-1) used for qualitative PCR analysis\nwere from Sigma. A conventional RIA kit for the quanti-\ntative measurement of 1,25(OH)\n2D3 was obtained from\nDiaSorin (Stillwater, MN, USA). Monoclonal antibody to\nvitamin D receptor was from Alexis Biochemicals\n(Lausen, Switzerland), antibody to /afii9826-actin was from\nSigma, antiserum to 1 /afii9825-OHase was from The Binding\nSite (Birmingham, UK) and anti-CD45-FITC/CD14-PE\nwas from Becton Dickinson.\nSample collection\nHuman samples were obtained from women who\nattended the endoscopic surgical service of the II\nDepartment of Obstetrics and Gynecology of the\nUniversity of Milan to undergo gynecological laparo-\nscopy for unexplained infertility, pelvic pain or benign\novarian pathology. Women with previous autoimmune,\nneoplastic, hepatic or thyroid disorders were excluded\nfrom the study. All subjects were younger than 40, had\nregular menstrual cycles and none had received\nhormones for at least 3 months.\nSamples of uterine endometrium were obtained from\n77 women. At laparoscopic visualization, 27 had\nevidence of endometriosis that was staged, according to\nthe Revised American Fertility Society Classification\n(1997) (American Society for Reproductive Medicine\n1997), as minimal (stage I) in 4 cases, mild (stage II) in\n5 cases, moderate (stage III) in 11 cases and severe (stage\nIV) in 7 cases. In the 50 patients in whom endometriosis\nwas not diagnosed, laparoscopic examination demon-\nstrated normal pelvic organs in 6 cases, pelvic adhesions\nin 7 cases, benign ovarian pathology in 22 cases and\nbenign uterine pathology in 15 cases. Among patients\nwith endometriosis 14 were in the proliferative and 13 in\nthe secretory phase of the cycle, while among controls 22\nwere in the proliferative and 28 in the secretory phase.\nDating was based on the last menstrual period and on\nhistological examination of the samples.\nEndometriotic samples from peritoneal lesions and\nendometriotic cysts were excised from nine patients with\nsevere endometriosis at operation and immediately\nfrozen at /p180 /p8C.\nDecidual tissues were obtained from 38 healthy\nwomen undergoing elective termination of normal\npregnancies between 8 and 13 weeks of gestation. The\nP VIGANOv and others · Vitamin D and endometrium416\nwww.endocrinology-journals.orgJournal of Molecular Endocrinology (2006) 36, 415–424\nDownloaded from Bioscientifica.com at 06/11/2026 02:24:47AM\nvia free access\n\n\noperative method used was cervical dilatation followed\nby vacuum extraction of the products of conception.\nAfter a careful aspiration of all the trophoblastic\nmaterial, a biopsy curette was used to obtain decidual\ntissue.\nPatients were informed in detail that serum or tissue\nsamples would be used for research purposes and they\ngave a written consent. Approval for this study was\ngranted by the local Human Institutional Investigation\nCommittee.\nCell culture\nEstablishment of stromal cell monolayers from normal\nendometrial tissue has been described in detail in pre-\nvious studies (Viganò et al. 1993). Di ffuse and strong\ncytoplasmatic immunostaining for vimentin was demon-\nstrated in nearly all (90%) cultured endometrial stromal\ncells. Cytofluorimetric analysis showed that leukocyte\ncontamination (CD45-positive cells) of our cultures was\nless than 2%. Briefly, tissue was minced and digested in\n10 ml Ham’s F-10 containing 0·2% collagenase. Single\nstromal cells were separated from epithelium by di fferen-\ntial sedimentation at unity gravity and selective adhesion\nto tissue culture dishes. Decidual tissue was minced\nthoroughly between two scalpels and digested for 1 h in\nHAM’S F-10 with 0·1% collagenase and 0·2% hyaluron-\nidase. Decidual cells were then separated from dead cells\nand red cells by Ficoll-Hypaque density gradient. The\ncells at the interface were plated and left at 37 /p8C over-\nnight, then washed several times to remove non-adherent\ncells and debris. Endometrial or decidual cells were\ncultured to subconfluence in Ham’s F-10 with 10% FCS\nand antibiotics in a humidified atmosphere of 95% air\nand 5% CO\n2 at 37 /p8C. Subconfluence was reached after a\nminimum of 8 days during which culture medium was\nchanged every other day. To evaluate the e ffects of\ndifferent compounds on endometrial 1 /afii9825-OHase mRNA\nexpression, endometrial stromal cells were cultured with\n0·1% bovine serum albumin when stimulated for 24 h\nwith and without IL-1 /afii9826or TNF- /afii9825and with charcoal-\nstripped calf serum when stimulated for up to 9 days with\nand without progesterone. To evaluate the e ffects of\n1,25(OH)\n2D3 on endometrial vitamin D-24-hydroylase\ngene ( CYP24) and osteopontin mRNA expression, cells\nwere cultured with 0·1% bovine serum albumin with and\nwithout different concentrations of the hormone for four\nhours.\nImmunohistochemistry\nImmunohistochemical analysis of 1 /afii9825-OHase expression\nwas performed using a previously described method\n(Zehnder et al. 2002b). Briefly, tissues were formalin-\nfixed, para ffi n-embedded, cut into 4-µm sections and\nplaced on pre-treated slides. Sections were dewaxed and\nrehydrated through graded alcohol using standard\nprocedures. They were processed in 0·01 M sodium\ncitrate bu ffer (pH 6·0) in a microwave for 15 min at the\nmaximum power for antigen retrieval. Slides were then\nincubated with 1% hydrogen peroxide in Tris–bu ffered\nsaline, pH 7·6 to block endogenous peroxidase activity,\nand then washed in Tris. The slides were then incubated\nwith 1 /afii9825-OHase antiserum (1:150) in 10% normal swine\nserum overnight at 4 /p8C. After rinsing with Tris–\nbuffered saline for 15 min, donkey anti-sheep IgG\nperoxidase conjugated (1:200) was added to sections\nfor 45 min. Staining was developed using 3,3 /p9-\ndiaminobenzidine (2·5 mg/ml) followed by counter-\nstaining with Mayer’s hematoxylin. Negative control\nsections included the omission of primary antibody, the\nuse of primary antibody preadsorbed with a 100-fold\nexcess of immunizing peptide or the analysis of the liver\nas a negative tissue (Zehnder et al. 2001).\nExtraction of mRNA, qualitative RT-PCR analysis for\n1/afii9825-OHase and sequence analysis\nTotal RNA was extracted from cells and tissues using\nTrizol (Invitrogen Life Technology, Carlsbad, CA,\nUSA). One microgram total RNA was transcribed into\ncDNA using 1 mM of each dNTP, 0·75 µg random\nhexamer primers, 25 units RNase inhibitor, and 200\nunits reverse transcriptase (RT) (Promega Corporation,\nMadison, WI, USA) in a total volume of 25 µl. cDNA\nwas amplified with 500 nM 1 /afii9825-OHase-specific primers\n(upstream: 5 /p9ATG ACC CAG ACC CTC AAG 3 /p9;\ndownstream: 5 /p9GTC GCA GAC TAC GTT GTT\nCAG 3 /p9) using RedTaq DNA polymerase (Sigma)\naccording to the following protocol: 94 /p8C for 5 min (1\ncycle), 94 /p8C for 30 s, 59 /p8C for 30 s, 72 /p8C for 30 s (38\ncycles), 72 /p8C for 5 min (1 cycle). The primers were\ndesigned to generate a fragment of 492 bp that spans\ntwo introns. In each experiment, a negative control was\nprepared using all reagents and substituting 1 µl water\nfor the reverse transcriptase. Integrity of RNA and\nabsence of genomic contamination was assessed by\namplification of HPRT-1 gene as previously described\n(Viganò et al. 2002). The specificity of the products\ngenerated by the indicated primers was verified by\nsequence analysis.\nEvaluation of gene expression by real-time\nRT-quantitative PCR (RT-qPCR) analysis\nOne microgram total RNA was reverse transcribed for\n2 h at 37 /p8C using the high-capacity cDNA archive kit.\nThe ABI Prism 7900 sequence detection system (Applied\nBiosystems) was used for real-time RT-qPCR analysis\nusing HPRT-1 as an endogenous control. Real-time\nPCR was performed using specific primers and probes\nfor 1 /afii9825-OHase, osteopontin and CYP24 target genes\nVitamin D and endometrium · P VIGANOv and others 417\nwww.endocrinology-journals.org Journal of Molecular Endocrinology (2006) 36, 415–424\nDownloaded from Bioscientifica.com at 06/11/2026 02:24:47AM\nvia free access\n\n\n(Assays-on-Demand Gene Expression Products, Applied\nBiosystems). Validation experiments were performed\nusing the 1:2 diluted templates. Reaction conditions\nincluded 10 µl 2 /p2TaqMan Universal PCR Master\nMix, 1 µl primers and probes mixture, 50 ng template\ncDNA and nuclease-free water on a 96-well reaction\nplate. The total reaction volume was 20 µl. The cycling\nconditions were as follows: 2 min at 50 /p8C, 10 min at\n95 /p8C, and 40 cycles of 15 s at 95 /p8C followed by 1 min\nat 60 /p8C. The data were analyzed using the comparative\nCt method, where Ct is the cycle number at which\nfluorescence first exceeds the threshold. The /afii9797cycle\nthreshold ( /afii9797Ct) values from each sample were obtained\nby subtracting the values for the reference gene from the\nsample Ct. For each experimental sample the 2\n/p1/afii9797Ct has\nbeen calculated and data have been graphically\nindicated as relative expression.\nWestern blot for VDR and 1 /afii9825-OHase\nSamples of endometrial stromal and decidual cell\ncultures were treated with a lysis bu ffer containing\n150 mM NaCl, 10 mM Tris–HCl, 1 mM EDTA, 1%\nTriton X-100, 10% glycerol, 1 mM PMSF, 10 µg/ml\nleupeptin and 10 µg/ml aprotinin. Lysates were\nsubsequently centrifuged at 13 000 g for 30 min and the\nsupernatant was collected for protein analysis. Sample\nprotein concentration was determined using a commer-\ncial protein assay kit (BCA Protein Assay Kit, Pierce\nBiotechnology, Rockford, IL, USA). Proteins resolved by\nSDS-PAGE were transferred to Hybond ECL nitrocel-\nlulose membranes (Amersham, Italy). For VDR\ndetection, after brief washing in TBST (25 mM\nTris–HCl (pH 7·5), 50 mM NaCl, 0·1% Tween 20), the\nmembrane was blocked with 5% skim milk plus 5%\nBSA/TBST overnight at 4 /p8C. All subsequent steps were\nperformed at room temperature. The membrane was\nincubated for 3 h with 11 µg anti-VDR antibody diluted\nin 25 ml 5% skim milk plus 5% BSA/TBST. After one\nhour washing with TBST, membranes were incubated\nfor 1 h with peroxidase-conjugated anti-rat IgG (Sigma).\nFor 1 /afii9825-OHase detection, the membrane was blocked\nwith 5% skim milk plus 5% BSA in TBST for 2 h at\nroom temperature. The membrane was incubated\novernight with anti-1 /afii9825-OHase antibody diluted 1:500 in\n5% skim milk plus 5% BSA/TBST. After 1 h washing\nwith TBST, membranes were incubated for 1 h with\nperoxidase-conjugated anti-sheep IgG (Sigma). For\nprotein load control, anti- /afii9826-actin mouse monoclonal\nantibodies were used. Anti-mouse IgG secondary\nantibody was used at 1:5000 dilution. Bound antibodies\nwere visualized by chemiluminescence. Control experi-\nments were included where primary antibody was\nomitted or primary antibody was preabsorbed with a\n100-fold excess of immunizing peptide.\nMeasurement of 1,25(OH)2D3 production\nEndometrial cells were seeded (5 /p2105cells/well/ml) in a\n6-well plate. After washing with serum-free medium,\ncells were incubated with di fferent concentrations of\n25(OH)D\n3 solubilized in absolute ethanol (0·1% final\nconcentration) in 1% FCS culture medium for 24 h. The\nconditioned medium and cell monolayers were har-\nvested. The quantitative detection of 1,25(OH)\n2D3 levels\nwas performed using a commercially available RIA kit\nwith an intra-assay coe ffi cient of variation of 6·8–11·3%,\nan interassay coe ffi cient of variation of 11·2–14·6%\nand a sensitivity of <2 pg/ml. The concentration of\n1,25(OH)\n2D3 levels was expressed as pg/ml.\nStatistical analysis\nDifferences between groups were compared, as appro-\npriate, by unpaired Student’s t-test, analysis of variance\n(ANOVA) and Fisher protected least significant\ndifference-test as post-test. Probability <0·05 was\nconsidered to be statistically significant.\nResults\nExpression of 1 /afii9825-OHase and vitamin D receptor in\nnormal endometrium\nNormal endometrial stromal cells obtained from samples\nin di fferent phases of the menstrual cycle and early\npregnant decidual cells were first evaluated for 1/afii9825-OHase\nmRNA expression by qualitative RT-PCR. During cell\nisolation and culture, a great e ffort was directed towards\nthe complete elimination of immune cells as potential\ncontaminants. Only cell populations 98% free of CD45-\nand CD14-positive cells as evaluated by phenotypic\nanalysis were included in the study. PCR products were\nconsistently detected in all samples analyzed (Fig. 1A).\nEstimated and actual size of the PCR products was\n492 bp. The identity of the amplified products with the\nprimer-defined 1/afii9825-OHase DNA sequence was confirmed\nby sequence analysis (data not shown). Quantification of\n1/afii9825-OHase mRNA levels in early pregnant decidual cells\nand in endometrial cells in di fferent phases of the cycle\nwas performed by real-time RT-qPCR analysis. Results\nof the experiments performed in normal endometrial\nstromal cells derived from n=28 di fferent tissues ( n=13\nand n=15 in the proliferative and secretory phases\nrespectively) and decidual cells obtained from n=25 dif-\nferent samples indicated that levels of 1 /afii9825-OHase mRNA\nwere similar in endometrial stromal cells independent of\nthe phase of the cycle but were significantly increased in\ndecidual cells ( P<0·05) (Fig. 1B). Western blot analysis\nshowed that the observed di fferences in 1 /afii9825-OHase\nmRNA levels were maintained at the protein level. A\nsingle western blot species (56 kDa) that corresponded to\nP VIGANOv and others · Vitamin D and endometrium418\nwww.endocrinology-journals.orgJournal of Molecular Endocrinology (2006) 36, 415–424\nDownloaded from Bioscientifica.com at 06/11/2026 02:24:47AM\nvia free access\n\n\nthe reported renal 1 /afii9825-OHase (Fig. 1C) was detected in\nboth proliferative ( n=3) and secretory ( n=3) phase\nendometrial stromal cells and in first trimester decidual\ncells ( n=4). Densitometric analysis of western blots\nshowed an increase of about 40% in 1 /afii9825-OHase expres-\nsion in decidua versus normal cycling endometria of both\nphases.\nTo determine the cellular localization of 1 /afii9825-OHase, its\nexpression was also evaluated by immunohistochemistry.\nComparable immunostaining for 1 /afii9825-OHase was present\nin the endometrium of both proliferative and secretory\nphases (Fig. 2C,D). The protein was abundantly\nexpressed in the cytoplasm of both stromal and epithelial\nlayers with a di ffuse distribution. As relative high levels of\n1/afii9825-OHase protein have previously been shown in first\ntrimester decidua (Zehnder et al. 2002b) this tissue was\nused as the positive control (Fig. 2A,B). Immunopositive\ncells were di ffusely distributed throughout the decidua\nand had no special relationship with the blood vessels.\nStaining was absent in the negative controls (Fig. 2E,F).\nTo evaluate whether the cycling endometrium could\nalso be a target of 1,25(OH)\n2D3 action, the presence of\nnuclear receptor VDR protein was investigated in\nproliferative ( n=3) and secretory ( n=3) phase endo-\nmetrial stromal cells by western blot analysis. Decidual\ncells ( n=4) have previously been shown to express the\nvitamin D receptor and were used as a positive control\n(Zehnder et al. 2002b). Rat antibody to human VDR\nreacted with both endometrial and decidual samples\nresulting in a band of 54 kDa (Fig. 3).\nRegulation of 1 /afii9825-OHase gene expression in normal\nendometrial stromal cells\nWe also investigated transcriptional regulation of\n1/afii9825-OHase expression in endometrial stromal cells by\nIL-1/afii9826(n=6), TNF- /afii9825(n=6) or progesterone ( n=6) using\nreal time RT-qPCR. IL-1 /afii9826, at concentrations of 50 and\n500 pg/ml, significantly increased 1 /afii9825-OHase expression\nby 44% and 130% respectively, while TNF- /afii9825at\nconcentrations of 0·1 ng/ml and 1 ng/ml could not elicit\na significant increase (Fig. 4). Incubation for up to 9 days\nwith progesterone had no e ffect on endometrial\n1/afii9825-OHase mRNA levels (data not shown). Similar\nexperiments could not be performed in epithelial\nendometrial cells given the di ffi culty in sub-culturing this\nparticular cell population (Viganò et al. 1993). Thus, no\nconclusions can be inferred for the regulation of\n1/afii9825-OHase expression in glandular cells.\nProduction of 1,25(OH)2D3 and effect of the\nhormone on specific target genes in endometrial\nand decidual cells\nMeasurement of 1,25(OH) 2D3 levels in the supernatant\nof endometrial stromal cells treated with 25(OH)D 3\nFigure 1 Expression of 1 /afii9825-OHase in normal cycling\nendometrial stromal cells and early pregnant decidual cells.\n(A) Qualitative RT-PCR of representative cases of proliferative\nphase endometrial stromal cells (lane 1), secretory phase\nendometrial stromal cells (lane 2) and early pregnant decidual\ncells (lane 3). Size marker is shown in the first left lane of the\ngel (MW). Lane 4 shows the negative control performed by\nsubstituting 1 µl water for the RNA. (B) Real time-qPCR\nanalysis of 1 /afii9825-OHase mRNA in cultured cells derived from\nn=13 proliferative phase endometrial samples (PR), n=15\nsecretory phase endometrial samples (SE) and n=25 decidual\nsamples. Data are presented as mean±\nS.E.M. 1/afii9825-OHase\nrelative expression. * P,0·05 versus decidual cells.\n(C) Western blot profile of total homogenate of endometrial\nand decidual cells. Three representative cases of proliferative\nphase endometrial stromal cells (lane 1), secretory phase\nendometrial stromal cells (lane 2) and early pregnant decidual\ncells (lane 3) are shown. Lane 4 represents a negative control\nin which primary antibody was preabsorbed with a 100-fold\nexcess of immunizing peptide. Detection of /afii9826-actin was used\nfor protein load control.\nVitamin D and endometrium ·\nP VIGANOv and others 419\nwww.endocrinology-journals.org Journal of Molecular Endocrinology (2006) 36, 415–424\nDownloaded from Bioscientifica.com at 06/11/2026 02:24:47AM\nvia free access\n\n\nresulted in a constitutive and dose-dependent production\nof the hormone, thus indicating that endometrium\nrepresents a site of local conversion from the precursor\nto the active form (Fig. 5A). Levels of produced\n1,25(OH)\n2D3 were similar to those detected in other\nrecognized extrarenal sites of production (Fritsche et al.\n2003).\nThe functional consequences of 1,25(OH) 2D3 produc-\ntion and the presence of VDR were tested by evaluating\nthe expression level of two target genes, CYP24 and\nosteopontin. CYP24 is one of the most potent\n1,25(OH)\n2D3-responding genes and its protein product\nis responsible for the hydroxylation reaction that\ndeactivates the biologically active vitamin D sterol.\nOsteopontin is an adhesion molecule with roles in\nimplantation and decidualization (Johnson et al. 2003)\nand is regulated by vitamin D in di fferent tissues\n(Christakos et al. 2003). Levels of CYP24 mRNA were\nnegligible in unstimulated endometrial cells and were\ndetectable but still very low in unstimulated decidual\nFigure 2 Cellular localization of 1 /afii9825-OHase in human cycling endometrium and early pregnant decidua as evaluated by\nimmunohistochemistry. The presence of immunoreactive 1 /afii9825-OHase is indicated by the brown staining in the cytoplasm of both\nepithelial and stromal cells of endometrial biopsies collected during the proliferative (C) and the secretory (D) phase of the\nmenstrual cycle. Similarly, early pregnant decidua show 1 /afii9825-OHase immunostaining (A, low magnification; B, higher magnification).\nThe negative controls for specificity of 1 /afii9825-OHase expression were performed by preabsorbing the antiserum with 100-fold excess\nof immunizing peptide (E) or testing a negative tissue such as the liver (F). (Original magnification: ×100 for A and E and ×200 for\nB,C,D and F .)\nP VIGANOv and others · Vitamin D and endometrium420\nwww.endocrinology-journals.orgJournal of Molecular Endocrinology (2006) 36, 415–424\nDownloaded from Bioscientifica.com at 06/11/2026 02:24:47AM\nvia free access\n\n\ncells. A significant and strong increase was observed in\nboth types of culture after addition of 1,25(OH) 2D3 for\n4 h (Fig. 5B). Similar to that reported for other\nfibroblast-like cells (Tashiro et al. 2004), in endometrial\ncells transcription levels went up to about 2000-fold for a\n1,25(OH)\n2D3 concentration of 1000 nM; in decidual\ncells this increase was still very strong. Cells were also\ntested for osteopontin expression, which in the basal\ncondition, as previously reported by von Wol ff et al.\n(2004), was significantly increased in decidual cells when\ncompared with endometrial stroma. Addition of\n1,25(OH)\n2D3 for 4 h at a concentration of 1000 nM\nsignificantly stimulated osteopontin expression levels by\nabout 60–70% in both types of culture (Fig. 5C).\nExpression of 1 /afii9825-OHase in eutopic and ectopic\nendometrium from patients with endometriosis\nComparison of mRNA levels for 1 /afii9825-OHase in eutopic\nendometrial samples of patients with ( n=14 and n=13 in\nthe proliferative and secretory phases respectively) and\nwithout endometriosis by real time RT-qPCR demon-\nstrated a significant increase for proliferative phase\ncultures derived from patients a ffected by the disease\n(P<0·05) (Fig. 6A). Western blot analysis confirmed that\nthese di fferences were maintained at the protein level\n(Fig. 6B).\nThree biopsies of endometriotic peritoneal lesions and\nnine biopsies of endometriotic cysts were tested for\n1/afii9825-OHase mRNA expression by qualitative RT-PCR.\nAmplified DNA products were consistently detected in\nall samples analyzed (Fig. 6C).\nDiscussion\nRecent findings have emphasized the potential role of\n1,25(OH)\n2D3 in decidual physiology (Zehnder et al.\n2002b). To the best of our knowledge, this is the first\nstudy that has investigated the vitamin D system in\nhuman cycling endometrium. The presence of the\nenzyme that catalyzes the synthesis of the active form of\nvitamin D in cycling endometrium and its up-regulation\nin first trimester decidua, support the possibility that this\nhormone might be involved in some mechanisms of\npregnancy establishment or maintenance.\nEven more intriguing in this context is the observation\nthat IL-1/afii9826caused a strong increase in 1 /afii9825-OHase mRNA\nexpression in endometrial stromal cells. IL-1 /afii9826is thought\nto actively participate in the synchronized cooperation\nbetween the endometrium and the preimplanting\nembryo under the influence of steroid hormones, both in\nmice and humans (Lindhard et al. 2002). In humans,\nsecretion of embryonic IL-1 /afii9826seems to be the first\nresponse of the blastocyst to the receptive endometrium,\nstimulating a second wave of cytokines essential for\nattachment of the blastocyst (Lindhard et al. 2002). We\nhave herein demonstrated that 1 /afii9825-OHase is among\nthose genes transcriptionally modulated by IL-1 /afii9826at the\nendometrial level. Interestingly, in contrast to what has\nbeen observed for macrophages and endothelial cells\n(Pryke et al. 1990, Zehnder et al. 2002a), the mechanism\nof cytokine-induced 1 /afii9825-OHase activity at the endome-\ntrial level seems to rely mostly on IL-1 /afii9826. Indeed, from\nthe results of this study, the TNF- /afii9825-mediated e ffect on\nendometrial 1 /afii9825-OHase expression is not as potent as\nthat obtained with very low concentrations of IL-1 /afii9826.\nThe e ffect of vitamin D in the uterus may be exerted\neither via the regulation of specific target genes (Du et al.\n2005) or through the well established immunomodula-\ntory e ffects of the hormone (Lemire et al. 1995). Among\nthe target genes there are calbindins and the HoxA10\nFigure 3 Expression of VDR in endometrial stromal cells and\nearly pregnant decidual cells as evaluated by Western blot.\nThree representative cases of proliferative phase endometrial\nstromal cells (lane 1), secretory phase endometrial stromal\ncells (lane 2) and early pregnant decidual cells (lane 3) are\nshown. Lane 4 represents a negative control in which primary\nantibody was preabsorbed with a 100-fold excess of\nimmunizing peptide.\nFigure 4 Effect of IL-1 /afii9826and TNF- /afii9825treatment on 1 /afii9825-OHase\ngene expression as evaluated by real-time RT-qPCR\nanalysis. Messenger RNA levels were evaluated in\nendometrial stromal cells treated with and without different\nconcentrations of the cytokines for 24 h. Data from n=6\nexperiments were analyzed by real time qPCR and\npresented as mean±\nS.E.M. relative expression. * P,0·05 vs\ncorresponding unstimulated controls.\nVitamin D and endometrium · P VIGANOv and others 421\nwww.endocrinology-journals.org Journal of Molecular Endocrinology (2006) 36, 415–424\nDownloaded from Bioscientifica.com at 06/11/2026 02:24:47AM\nvia free access\n\n\ngene which are critical for implantation in mice and in\nhumans respectively (Benson et al. 1996, Taylor et al.\n1998, Nie et al. 2000, Salamonsen et al. 2002, Du et al.\n2005) and are known to be transcriptionally regulated by\nvitamin D in di fferent cell types (Rots et al. 1998, Nie\net al. 2000, Du et al. 2005). We have herein identified\nosteopontin as another possible target gene regulated by\nvitamin D. Osteopontin is an acidic member of the small\nintegrin-binding ligand N-linked glycoprotein family of\nextracellular matrix proteins/cytokines. This versatile\nprotein is a major constituent of the uterine-placental\nmicroenvironment and exerts its influence as a\ncomponent of the endometrial gland secretion and as a\ngene product expressed by uterine stroma contributing\nto a decidualization-like transformation that correlates\nwith the degree of conceptus invasiveness (Johnson et al.\n2003). While 1,25(OH)\n2D3 has been shown to induce\nthe gene in some tissues, this is the first demonstration of\nthis e ffect in endometrial and decidual cells, giving\nfurther support for a role of the hormone as a local\nparacrine signal.\nIf we consider a possible action of 1,25(OH)\n2D3 as a\nnatural regulator of the immune system, it is important\nto note that these e ffects can be viewed in the context of\nthe local immune responses supposed to be critical for\nthe development of a normal pregnancy. Indeed, in a\nsuccessful pregnancy, the cytokine profile is thought to\nbe shifted away from cell-mediated (T helper (Th)1-type)\nresponses towards humoral immunity (Th2-type) (Hill\net al. 1995, Viganò et al. 2002). Vitamin D promotes the\nshift away from Th1-type responses and favours a\nTh2-type immunity by inhibiting the secretion of IL-12,\nIL-2, TNF and /afii9828-interferon by T cells, macrophages and\ndendritic cells (Limire et al. 1995, Muller & Bendtzen\n1996, D’Ambrosio et al. 1998, Long & Santos 1999).\nIn the present study, we have also demonstrated\nexpression of 1 /afii9825-OHase by ectopic endometriotic\nimplants and the presence of higher levels in eutopic\ncells of women a ffected by the disease. As for many\nother molecules found to be aberrantly present in\nFigure 5 Synthesis of 1,25(OH) 2D3 in endometrial cells and\ntranscriptional effects on specific target genes. (A) 1,25(OH) 2D3\nlocal synthesis from 25(OH)D 3 was measured by an RIA assay.\nEndometrial stromal cells were cultured for 24 h in 1% FCS\nculture medium in the presence or absence of 25(OH)D\n3.( B )\nReal time RT-qPCR analysis of Cyp24 mRNA in cultures\nderived from n=3 endometrial samples and n=5 decidual\nsamples. Cells were treated with and without different\nconcentrations of the hormone for 4 h. Data are presented as\nmean±\nS.E.M. Cyp24 relative expression. * P,0·05 versus\ncorresponding unstimulated samples. (C) Real time RT-qPCR\nanalysis of osteopontin mRNA in cultures derived from n=4\nendometrial samples and n=4 decidual samples. Cells were\ntreated with and without different concentrations of the\nhormone for 4 h. Data are presented as mean±\nS.E.M.\nosteopontin relative expression. * P,0·05 versus corresponding\nunstimulated or indicated samples.\nP VIGANOv and others · Vitamin D and endometrium422\nwww.endocrinology-journals.orgJournal of Molecular Endocrinology (2006) 36, 415–424\nDownloaded from Bioscientifica.com at 06/11/2026 02:24:47AM\nvia free access\n\n\nendometriosis, it is di ffi cult to clarify whether the\nincreased endometrial 1 /afii9825-OHase mRNA levels in the\nproliferative phase represent a primary event or a\nconsequence of the disease. It might be hypothesized\nthat a secondary endometrial response to the peritoneal\ninflammatory reactions, immediately following men-\nstruation, is responsible for this increase. Alternatively,\nendometrium from women a ffected might be constitu-\ntively more able to produce vitamin D. In both cases, the\nhormone, mostly for its ability to modulate cytokine\nproduction and inflammatory mediators (D’Ambrosio\net al. 1998) may influence the functional activities of\nspecific immune populations, such as natural killer cells\nthat, in these women, are characterized by peculiar\nfeatures (Vignali et al. 2002).\nIn conclusion, the results of this study support the\nfollowing observations: (i) human cycling endometrium\nmay be included among those sites capable of extrarenal\nsynthesis and action of vitamin D; (ii) the enzyme\n1/afii9825-OHase is expressed in human endometrial stromal\ncells independently of the phase of the menstrual cycle\nbut its expression is up-regulated in early pregnant\nversus cycling endometrium; (iii) IL-1 /afii9826is a potent\ninducer of endometrial 1 /afii9825-OHase mRNA expression;\n(iv) endometrium also expresses the vitamin D receptor;\n(v) in endometrial and decidual cells, the Cyp24 gene,\nwhose product catalyzes the first step of the degradation\npathway of 1,25(OH)\n2D3, is strongly induced transcrip-\ntionally by the hormone; (vi) the osteopontin gene is a\ntarget of vitamin D action in both cycling and early\npregnant endometrium; (vii) in endometriosis patients,\nthe gene coding for 1 /afii9825-OHase is expressed also in\nectopic endometrium and its expression is enhanced in\neutopic endometrium during the proliferative phase.\nTaken together, these results confirm the necessity to\nfurther investigate the functional role of the vitamin D\nsystem at the endometrial level.\nAcknowledgements\nThe authors declare that there is no conflict of interest\nthat would prejudice the impartiality of this scientific\nwork.\nReferences\nAcker GM, Garabedian M, Guillozo H, Pecquinot MA & Balsan S\n1982 In vitro formation of 25-hydroxyvitamin D 3 metabolites in\nendometrium: dependence on hormonal status of the rat.\nEndocrinology 111 2103–2109.\nAmerican Society for Reproductive Medicine 1997 Revised\nAmerican Society for Reproductive Medicine classification of\nendometriosis: 1996. Fertility and Sterility 67 817–821.\nBenson GV, Lim H, Paria BC, Satokata I, Dey SK & Maas RL\n1996 Mechanisms of reduced fertility in Hoxa-10 mutant mice:\nuterine homeosis and loss of maternal HOXA-10 expression.\nDevelopment 122 2687–2696.\nChristakos S, Dhawan P, Liu Y, Pen gX&P orta A 2003 New\ninsights into the mechanisms of vitamin D action. Journal of\nCellular Biochemistry 88 695–705.\nD’Ambrosio D, Cippitelli M, Cocciolo MG, Mazzeo D, Di Lucia P,\nLang R, Sinigagli aF&P anina-Bordignon P 1998 Inhibition of\nIL-12 production by 1,25-dihydroxyvitamin D\n3: involvement of\nFigure 6 Expression of 1 /afii9825-OHase in eutopic endometrial stromal cells from women with and without endometriosis and in ectopic\ntissues. (A) Real time RT-qPCR analysis of 1 /afii9825-OHase mRNA in cultured cells derived from n=28 eutopic endometrial samples\nfrom disease-free women and n=27 eutopic samples from affected women. Data are presented as mean± S.E.M. 1/afii9825-OHase relative\nexpression. * P,0·05 versus corresponding secretory phase samples. PR, proliferative phase; SE, secretory phase. (B) Western\nblot profile of total homogenate of eutopic endometrial cells. Representative cases of proliferative (lane 1) and secretory (lane 2)\nphase eutopic endometrial cells from disease-free women and proliferative (lane 3) and secretory (lane 4) phase eutopic\nendometrial cells from affected women are shown. Lane 5 represents a negative control in which primary antibody was\npreabsorbed with a 100-fold excess of immunizing peptide. Detection of /afii9826-actin was used for protein load control. (C) Analysis of\n1/afii9825-OHase mRNA by qualitative RT-PCR in eutopic endometrial stromal cells of women with endometriosis (lanes 1–2, proliferative\nand secretory phases of the cycle respectively), in two biopsies of endometriotic peritoneal lesions (lanes 3–4) and in one biopsy\nof endometriotic cysts (lane 5). Lane 6 is the negative control prepared using all reagents and substituting 1 µl water for the\nreverse transcriptase. Size marker is shown on the left side of the gel (MW).\nVitamin D and endometrium ·\nP VIGANOv and others 423\nwww.endocrinology-journals.org Journal of Molecular Endocrinology (2006) 36, 415–424\nDownloaded from Bioscientifica.com at 06/11/2026 02:24:47AM\nvia free access\n\n\nNF-/afii9837B downregulation repression of the p40 gene. Journal of\nClinical Investigation 101 252–262.\nDelvin EE, Arabian A, Glorieux FH & Mamer OA 1985 In vitro\nmetabolism of 25-hydroxycholecalciferol by isolated cells from\nhuman decidua. Journal of Clinical Endocrinology and Metabolism\n60 880–885.\nDu H, Daftary GS, Lalwani SI & Taylor HS 2005 Direct regulation\nof HoxA10 by 1,25-(OH)\n2D3 in human myelomonocytic cells and\nhuman endometrial stromal cells. Molecular Endocrinology\n19 2222–2233.\nFritsche J, Mondal K, Ehrnsperger A, Andreese nR&K r e u t zM\n2003 Regulation of 25-hydroxyvitamin D 3-1/afii9825-hydroxylase and\nproduction of 1 /afii9825,25-dihydroxyvitamin D 3 by human dendritic\ncells. Immunobiology 102 3314–3316.\nHartwell D, Riis BJ & Christiansen C 1990 Changes in vitamin D\nmetabolism during natural and medical menopause. Journal of\nClinical Endocrinology and Metabolism 71 127–132.\nHewison M, Zehnder D, Blan dR&S t e w a r tP M 2000\n1/afii9825-hydroxylase and the action of vitamin D. Journal of Molecular\nEndocrinology 25 141–148.\nHill JA, Polgar K & Anderson DJ 1995 T-helper 1-type immunity to\ntrophoblast in women with recurrent spontaneous abortion.\nJournal of the American Medical Association 273 1933–1936.\nJohnson GA, Burghardt RC, Bazer FW & Spencer TE 2003\nOsteopontin: roles in implantation and placentation. Biology of\nReproduction 69 1458–1471.\nJones G, Strugnell SA & De Luca HF 1998 Current understanding\nof the molecular actions of vitamin D. Physiological Review\n78 1193–1231.\nKachkache M, Rebut-Bonneton C, Demignon J, Cynober E &\nGarabedian M 1993 Uterine cells other than stromal decidual\ncells are required for 1,25-dihydroxyvitamin D\n3 production during\nearly human pregnancy. FEBS Letters 333 83–88.\nKinuta K, Tanaka H, Moriwake T, Aya K, Kat oS&S e i n oY2 0 0 0\nVitamin D is an important factor in estrogen biosynthesis of both\nfemale and male gonads. Endocrinology 141 1317–1324.\nKumar R 1980 The metabolism of 1,25-dihydroxyvitamin D\n3.\nEndocrine Reviews 1 258–267.\nLemire JM, Archer DC, Beck L & Spiegelberg HL 1995\nImmunosuppressive actions of 1,25-dihydroxyvitamin D 3:\npreferential inhibition of Th1 functions. Journal of Nutrition\n125 1704s–1708s.\nLindhard A, Bentin-Ley U, Ravn V, Islin H, Hviid T, Rex S,\nBangsboll S & Sorensen S 2002 Biochemical evaluation of\nendometrial function at the time of implantation. Fertility and\nSterility 78 221–233.\nLong KZ & Santos JI 1999 Vitamins and the regulation of the\nimmune response. Pediatric Infectious Disease Journal 18 283–290.\nM u l l e rK&B e n dtzen K 1996 1,25-Dihydroxyvitamin D\n3 as a\nnatural regulator of human immune functions . Journal of\nInvestigative Dermatology Symposium Proceedings 1 68–71.\nNie GY, Li Y, Wang J, Minoura H, Findlay JK & Salamonsen L\n2000 Complex regulation of calcium-binding protein D9k\n(calbindin-D\n9k) in the mouse uterus during early pregnancy and at\nthe site of embryo implantation. Biology of Reproduction 62 27–36.\nPanda DK, Miao D, Tremblay ML, Sirois J, Farookhi R,\nHendy GN & Goltzman D 2001 Targeted ablation of the\n25-hydroxyvitamin D 1 /afii9825-hydroxylase enzyme: evidence\nfor skeletal, reproductive, and immune dysfunction. PNAS\n98 7498–7503.\nPeehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey T &\nFeldman D 1994 Antiproliferative e ffects of 1,25-dihydroxyvitamin\nD\n3 on primary cultures of human prostastic cells. Cancer Research\n54 805–810.\nPryke AM, Duggan C, White CP, Pose nS&M ason RS 1990\nTumor necrosis factor-alpha induces vitamin D-1-hydroxylase\nactivity in normal human alveolar macrophages. Journal of Cellular\nPhysiology 142 652–656.\nRots NY, Liu M, Anderson EC & Freedmann LP 1998 A\ndifferential screen for ligand-regulated genes: identification of\nHoxA10 as a target of vitamin D\n3 induction in myeloid leukemic\ncells. Molecular and Cellular Biology 18 1911–1918.\nSalamonsen LA, Nie GY & Findlay JK 2002 Newly identified\nendometrial genes of importance for implantation. Journal of\nReproductive Immunology 53 215–225.\nSalle BL, Delvin EE, Lapilonne A, Bishop NJ & Glorieux F 2000\nPerinatal metabolism of vitamin D 1–3. American Journal of Clinical\nNutrition 71 1317S–1324S.\nTashiro K, Abe T, Oue N, Yasui W & Ryoji M 2004\nCharacterization of vitamin D-mediated induction of the CYP 24\ntranscription. Molecular and Cellular Endocrinology 226 27–32.\nTaylor HS, Arici A, Oliv e D & Igrashi P 1998 HOXA 10 is expressed\nin response to sex steroids at the time of implantation in the human\nendometrium. Journal of Clinical Investigation 101 1379–1384.\nViganò P, Di Blasio AM, Dell’Antoni oG&V ignali M 1993 Culture\nof endometrial cells: a new simple technique to completely\nseparate epithelial glands. Acta Obstetrica et Gynecologica Scandinavica\n7 87–92.\nViganò P, Ga ffuri B, Somigliana E, Busacca Mi, Di Blasio AM &\nVignali M 1998 Expression of intercellular adhesion molecule\n(ICAM)-1 mRNA and protein is enhanced in endometriosis versus\nendometrial stromal cells in culture. Molecular Human Reproduction\n4 1150–1156.\nViganò P, Somigliana E, Mangioni S, Vignali M, Vignali M &\nDi Blasio AM 2002 Expression of interleukin-10 and its receptor\nis up-regulated in early pregnant versus cycling human\nendometrium. Journal of Clinical Endocrinology and Metabolism\n87 5730–5736.\nVignali M, Infantino M, Matrone R, Chiodo I, Somigliana E,\nBusaccaM&V iganò P 2002 Endometriosis: novel\netiopathogenetic concepts and clinical perspectives. Fertility and\nSterility 78 665–678.\nvon Wol ff M, Bohlmann MK, Fiedler C, Urse lS&S trowitzki T\n2004 Osteopontin is up-regulated in human decidual stromal cells.\nFertility and Sterility 81 741–748.\nYoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K,\nYoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y et al .\n1997 Mice lacking the vitamin D receptor exhibit impaired bone\nformation, uterine hypoplasia and growth retardation after\nweaning. Nature Genetics 16 391–396.\nZehnder D, Bland R, Williams MC, McNinch RW, Howie AJ,\nStewart PM & Hewison M 2001 Extrarenal expression of\n25-hydroxyvitamin D\n3-1/afii9825-hydroxylase. Journal of Clinical\nEndocrinology and Metabolism 86 888–894.\nZehnder D, Bland R, Chana RS, Wheeler DC, Howie AJ,\nWilliams MC, Stewart PM & Hewison M 2002 a Synthesis of\n1,25-dihydroxyvitamin D 3 by human endothelial cells is regulated\nby inflammatory cytokines: a novel autocrine determinant of\nvascular cell adhesion. Journal of the American Society of Nephrology\n13 621–629.\nZehnder D, Evans KN, Kilby MD, Bulmer JN, Innes BA, Stewart PM\n& Hewison M 2002 b The ontogeny of 25-hydroxyvitamin D\n3\n1/afii9825-hydroxylase expression in human placenta and decidua.\nAmerican Journal of Pathology 161 105–114.\nReceived in final form 7 February 2006\nAccepted 20 March 2006\nMade available online as an Accepted Preprint 22 March 2006\nP VIGANOv and others · Vitamin D and endometrium424\nwww.endocrinology-journals.orgJournal of Molecular Endocrinology (2006) 36, 415–424\nDownloaded from Bioscientifica.com at 06/11/2026 02:24:47AM\nvia free access","source_license":"CC0","license_restricted":false}