{"paper_id":"5f1ff75b-4a37-4d88-9c50-8d23a3c68dba","body_text":"DOI 10.1515/pjvs-2016-0076\nOriginal article\nEstradiol reduces connexin43 gap junctions\nin the uterus during adenomyosis in cows\nA.J. Korzekwa, M. Łupicka, B.M. Socha, A.A. Szczepańska\nDepartment of Reproductive Immunology and Pathology,\nInstitute of Animal Reproduction and Food Research,\nPolish Academy of Sciences, 10-748 Olsztyn, Poland\nAbstract\nAdenomyosis is defined as the presence of glandular foci external to the endometrium of the\nuterus, either in the myometrium or/and perimetrium, depending on the progress of this dysfunction.\nTo date, we showed that steroids secretion and prolactin expression and proliferative processes are\ndisturbed during uterine adenomyosis in cows. During endometriosis in eutopic endometrium in\nwomen, gap junctions are down regulated. The transmembrane gap junction protein, connexin (Cx43)\nis necessary for endometrial morphological, biochemical and angiogenic functions. The aim of this\nstudy is recognition of adenomyosis etiology by determination of the role of Cx43 in this process.\nImmunolocalization and comparison of Cx43 mRNA and protein expression in healthy (N=9) and\nadenomyotic uterine tissue (N=9), and Cx43 mRNA expression (real time PCR) in uterine stromal\n– myometrium co-culture under 24-hour stimulation with 17-beta estradiol (10\n-7M) isolated from\nhealthy (N=5) and adenomyotic (N=5) cows were determined.\nCx43 was localized in healthy and adenomyotic uteri. mRNA and protein expression was\ndown-regulated in uterine tissue in adenomyotic compared with healthy cows (p <0.05). Estradiol\nstimulated Cx43 mRNA expression in myometrial cell culture and co-culture of stromal and myomet-\nrial cells in adenomyotic compared with healthy cows (p<0.05). In summary, down-regulation of Cx43\nexpression in the junction zone might play an important role in pathogenesis of adenomyosis. Es-\ntradiol modulates gap junctions during adenomyosis.\nKey words: uterus, gap junction, connexin, adenomyosis, cow\nIntroduction\nAdenomyosis is a uterine dysfunction character-\nized by the presence of endometrial glands with\nstromal elements in the myometrium (Azziz 1989,\nCampo et al. 2012) but the etiology of this disorder is\nstill unclear (Ferenczy 1998, Korzekwa et al. 2014).\nOur previous studies showed that in cows the fre-\nCorrespondence to: A. Korzekwa, e-mail: a.korzekwa@pan.olsztyn.pl, tel.: 48 89 539 31 30\nquency and progression of adenomyosis are positively\ncorrelated with age, what is consistent with studies on\nwomen. The cows older than 5 years had advanced\nstages of adenomyosis (approx. 40% of animals; Ko-\nrzekwa et al. 2013). Moreover, the presence of aden-\nomyosis did not differed based on the sampling loca-\ntion in the uterus. The glandular nests in adenomyotic\ncows were present in myometrial layer of both uterine\nPolish Journal of Veterinary Sciences Vol. 19, No. 3 (2016), 609–617\n\nhorns, ipsilateral and contralateral to ovary with cor-\npus luteum (CL), and near the cervix as well (Ko-\nrzekwa et al. 2013, Łupicka et al. 2015).\nGap junctions are intercellular channels that per-\nmit communication between cells by cytoplasmic ex-\nchange of small molecules including metabolites and\nsecond messengers, as well as ions that transmit electri-\ncal impulses between neighboring cells (Evans and\nMartin 2002). A gap junction channel is composed of\ntwo hemi-channels (connexons), each of which is com-\nposed of six protein subunits (connexins -Cxs) arranged\naround a pore. Connexin43 (Cx43) and Cx32 are the\ntwo most abundant Cxs and are expressed in numerous\ntissues including the ovary, placenta and decidua in\nseveral species, including cattle (Wiesen and Midgley\n1993, 1994, Mayerhofer and Garfield 1995, Wrenzycki\net al. 1995, van Engelen et al. 2009, Xu et al. 2015).\nIn addition, a channel-independent role of Cxs in\nintracellular signaling by interacting with other proteins\nhas been suggested (Saez et al. 2003). Expression of\nt h ed i f f e r e n tC x sm u s tb ep r o p e r l yc o n t r o l l e da st h e i r\nmisregulation may contribute to diseases. As a conse-\nquence, mutations in Cx proteins resulting in modula-\ntion of channel properties are associated with a large\nvariety of specific pathologies (Araya et al. 2005).\nDuring adenomyosis in cows, protein expression\nof the 17-beta estradiol (E2) receptor α (ERα) is in-\ncreased and blood and endometrial E2 concentrations\nare elevated, indicating hormonal abnormalities dur-\ning this condition (Campo et al. 2012). Invasion by\nadenomyotic stromal cells was higher after E2 treat-\nment in an in vitro model of stromal and myometrial\nco-culture of human cells (Taylor et al. 2014). Es-\ntradiol increases Cx43 mRNA expression in rat en-\ndometrium (Gru¨mmer et al. 1994) but also in the my-\nocardium (Chen et al. 2010). Moreover, E2 stimulated\nCx43 mRNA expression in human myometrial uterine\ncells (Di et al. 2001). The existence of estrogen re-\nsponse elements within the Cx43 gene and evidence\nimplicating E2 in the induction of Cx43 transcription\nhas been reported (Petrocelli and Lye 1993, Lye et al.\n1993) providing a molecular basis for the reported\nestrogenic stimulation of Cx43 synthesis (Chung et al.\n2004). Therefore, we designed an in vitro model using\nco-culture of stromal and myometrial uterine cells iso-\nlated from healthy and adenomyotic cows for com-\nparing E2 effect on Cx43 mRNA expression.\nThe aims of the study were: (1) Cx43 immu-\nnolocalization and comparison of Cx43 mRNA and\nprotein expression in healthy and adenomyotic uter-\nine tissues, (2) determination of Cx43 mRNA express-\nion in co-cultured uterine stromal and myometrial\ncells isolated from healthy and adenomyotic cows un-\nder stimulation with E2.\nMaterials and Methods\nMaterial collection\nAll procedures were approved by the Local Ani-\nmal Care and Use Committee, Olsztyn, Poland\n(agreement no. 83/2012/N).\nFrom a total of 30 examined cows, 18 Hol-\nstein/Polish Black and White cows (75%/25%, re-\nspectively), 5-7 years old, were used in the experi-\nments for post mortem collection of uteri (days 8-10 of\nthe oestrous cycle). These days of the estrous cycle\nwere selected for experiments because this study is\ncontinuation of our concerning adenomyosis (Ko-\nrzekwa et al. 2013, 2014, Łupicka et al. 2015) and\nadenomyosis has been observed during the luteal\nstage of the estrous cycle in cattle (Moreira et al.\n2007). Uterine tissues were obtained at the Meat Pro-\ncessing Plant „Warmia” (Biskupiec, Poland) and\ntransported on ice to the laboratory within 40 min.\nDay of the oestrous cycle was evaluated by macro-\nscopic observation of the ovaries and uterus\n(Miyamoto et al. 2000) and confirmed by determina-\ntion of progesterone (P4) levels in peripheral blood\nplasma collected from the jugular vein using radioim-\nmunoassay (RIA). The level of P4 ranged from 15 to\n20 ng/ml. Just before slaughter, each animal was\nexamined by a veterinarian via per rectum ultra-\nsound-guided examination to exclude abnormal ovary\nstructure. The reasons for culling animals from the\nherd were of economic nature and herd renewal, none\nof used for experiments cows was earlier treated for\nendometritis.\nTissue fragments (cross-sections of the uterine\nwall, i.e., endometrium and myometrium) were ob-\ntained from the middle segment of the uterine horn\nipsilateral to the corpus luteum and were divided into\nthree pieces: the first one was fixed in 4% paraformal-\ndehyde (PFA) in 0.1 M PBS (pH 7.4) for histo- and\nimmunohistofluorescence staining, the second was\nfrozen and stored at -86\noC for further mRNA and\nprotein expression determination in uterine tissue\n(endometrium with myometrium), and the third piece\nwas used for immediate isolation and culture of uter-\nine cells.\nThe tissues for experiments were divided into nor-\nmal (uteri without endometrial glands within the my-\nometrial layer, n=9) and adenomyotic (endometrial\nfoci present in the myometrium, n=9), based on\nmicroscopic observation of hematoxylin and eosin-\n-stained uterine cross-section slices.\nA.J. Korzekwa et al.610\n\nUterine cell isolation and in vitro\nculture\nEndometrial stromal cells were isolated by enzy-\nmatic dissociation as previously described (Łupicka\net al. 2015). After endometrial cell isolation, the my-\nometrial layer of the uterus was dissected with scis-\nsors. About 4 cm-long fragments of muscle tissue\nwere chopped up with scissors into a homogeneous\nmaterial. Approximately 5 g of the chopped tissue\nwas digested in 50 ml of M199 medium (Sigma,\nM2520, St. Louis, MO, USA) containing 0.1% of\nbovine serum albumin (BSA; Sigma, A2058), 20\nμg/ml of gentamicin (Sigma, G1271), 2 mg/ml of col-\nlagenase I (Sigma, C0130), 1 mg/ml of deoxyribonuc-\nlease (Sigma, D5025) and 2 mg/ml of dispase (Life\nTechnologies, 17105-041, Paisley, UK). The enzyme\nsolution with myometrial tissue was held at 37.5\noC\nwith stirring for 30 min. After digestion, the cell sus-\npension was filtered through a mesh to remove un-\ndigested tissue fragments, then the cells were washed\nby centrifugation (10 min at 100 x g, at 4\noC). Cells\nwere resuspended in culture medium (DMEM;\nSigma, D5796) supplemented with 10% of fetal calf\nserum (FCS; Sigma, 12133C) and antibiotics (gen-\ntamicin/amphotericin B; Life Technologies,\n1153727).\nUterine cells were plated in a co-culture system,\nin which myometrial cells were cultured at the bot-\ntom of wells in 6-well culture plates (basal compart-\nment) and stromal cells were cultured on col-\nlagen-coated inserts (Biocoat\n® Cell Culture Inserts\nCollagen Type I, BD Biosciences, Bedford, MA; api-\ncal compartment) at 37.5\noCi nah u m i d i f i e da t m o s -\nphere of 5% CO 2, 95% air. Both cell types were cul-\ntured in a phenol red-free DMEM medium contain-\ning 10% of fetal calf serum (FCS; Sigma, 12133C)\nand antibiotics. The medium (DMEM with 10% FCS\nand antibiotics) was changed every 2 days until 70%\nconfluence was reached (approx. on the 4\nth day of\nculture). For the last 24 h of culture, inserts with\nstromal cells were replaced with myometrial cells.\nAfter changing the medium for DMEM containing\n0.1% BSA and antibiotics, the co-cultures were\nstimulated with E2 (10\n-7M). Total mRNA and cell\nlysates were collected from the cultures. Cell culture\nhomogeneity was confirmed using real-time PCR for\ndetermination of mRNA expression of vimentin and\ndesmin for stromal and myometrial cells, respective-\nly, in separate cultures, before establishing co-cul-\ntures (Zeiler et al. 2007).\nExperimental procedures\nExperiment 1. Localization, mRNA and protein\nexpression of connexin43 in uterine tissues\nof healthy and adenomyotic cows\nCx43 immunolocalization (immunofluorescence)\nand comparison of Cx43 mRNA (real time PCR) and\np r o t e i n( W e s t e r nb l o t t i n g )e x p r e s s i o ni nh e a l t h y\n(N=9) and adenomyotic uterine tissue (N=9) was\nproceeded.\nExperiment 2. The effect of 17-beta estradiol\non Cx43 mRNA expression in uterine stroma\n– myometrium co-culture of healthy\nand adenomyotic cows\nCx43 mRNA expression (real time PCR) was\nevaluated in uterine stromal, myometrial cells and in\nmyometrial cells co-cultured with stromal cells after\n24 h stimulation with E2 in healthy (N=5) and aden-\nomyotic (N=5) cows.\nThe stroma – myometrium co-cultures, after re-\nceiving the 70% of confluence, were stimulated for\n24 h with E2 (10\n-7M). Cells were isolated from\nhealthy (N=5) and adenomyotic (N=5) cows. The\neffective dose of E2 was selected based on the pre-\nliminary study.\nHistochemical staining\nUterine tissue was fixed in 4% PFA and pro-\ncessed for a standard haematoxylin and eosin stain-\ning protocol. Stained cross-sections of the tissue\nwere observed under a light microscope (Nikon\nFXA, Tokyo, Japan). Animals were classified as de-\nscribed previously (Korzekwa et al. 2013, Łupicka et\nal. 2015); briefly, if uterine glands were present only\nin the endometrium, and if the endometrial-myomet-\nrial border was clearly visible, cows were classified as\nnormal/control. Whereas, if the glands penetrated\nthe myometrial layer of the uterus, animals were\nclassified as adenomyotic (according to the classifica-\ntion of Katkiewicz et al. 2005, data not shown).\nImmunofluorescence staining\nImmunohistofluorescence was used to localise\nCx43 in uterine tissues. Cross-sections of uterine\nhorn samples were fixed in 4% PFA in 0.1 M PBS\n(pH 7.4), and cryoprotected in 18% sucrose. Im-\nmunostaining was carried out on consecutive 7 μm\nEstradiol reduces connexin43 gap junctions... 611\n\nTable 1. Oligonucleotide sequences used for real-time PCR.\nGene Oligonucleotide sequences Product size (bp) GeneBank\nvimentin FWD 5’-GACCTGGAGCGTAAAGTGG-3’\nREV 5’-GACATGCTGTTCTTGAATCTGG-3’\n108 BC118269\ndesmin FWD 5’-GACCCAGGCAGCCAACAAG-3’\nREV 5’-GTCGATCTCGCAGGTGTAGG-3’\n100 BC133410\ncx43 FWD 5’-TGAGTGCCGTTTACACTTGC-3’\nREV 5’-GGCAAGAGACACCAATGACA-3’\n125 J05535\nGAPDH FWD 5’-CACCCTCAAGATTGTCAGCA-3’\nREV 5’-GGTCATAAGTCCCTCCACGA-3’\n103 BC102589\ncryostat sections. To block endogenous peroxidase,\nthe sections were treated with hydrogen peroxide in\nmethanol and washed in 0.1 M PBS. The sections\nwere blocked with 10% normal donkey serum\n(Sigma, G9023) for 1 h at room temperature (ap-\nprox. 23\noC; RT), and incubated overnight at RT with\na 1:500 dilution of anti-Cx43 (Sigma, C8093), Next,\nthe cells were washed 3x with PBS and incubated\n1 h at RT with secondary antibodies conjugated with\ncyanine 3 (CY\n3; Jackson ImmunoResearch, West\nGrove, PA, 715-165-150). Connexin43 was visualized\nwith confocal imaging using a Nikon C1 confocal\nmicroscope.\nTotal RNA isolation\nTotal RNA was extracted from uterine tissues (ap-\nprox. 30 mg) and from cultured cells using TRI-Re-\nagent (Sigma, T9424) according to the manufacturer’s\ninstructions. The content and purity of RNA was as-\nsessed on a NanoDrop 1000 (Thermo Fisher Scien-\ntific, ND-1000, Wilmington, DE, USA). One micro-\ngram of each sample of total RNA was reverse-tran-\nscribed to cDNA with the QuantiTect Reverse Tran-\nscription kit (Life Technologies, 205313), as described\nin the supplier’s protocol. The cDNA obtained was\nstored at -20\noC until real-time PCR was applied.\nReal-time PCR quantification\nmRNA expression for Cx43 in tissues and cells\nwas determined by quantitative real-time PCR. The\nexperiments were performed using the Applied Bio-\nsystems 7900 (Applied Biosystems, Foster City, CA,\nUSA) with SensiFAST SYBR Hi-ROX Kit (Bioline\nReagents, BIO-92002, London, UK) according to the\nmanufacturer’s instructions. The real-time PCR re-\naction mix (20 μl) contained 19 μlo fS e n s i F A S T\nSYBR Hi-ROX Master Mix, 0.5 μMo fs e n s ea n d\nantisense primers, and 1 μl of reverse-transcribed\ncDNA (50 ng). Primer sequences used for determi-\nnation of Cx43 and glyceraldehyde 3-phosphate de-\nhydrogenase ( GAPDH) mRNA expression are de-\ntailed in Table 1. Standard curves consisting of serial\ndilutions of the appropriate cDNA were plotted for\nefficiency evaluation. Amplification was initiated by\nan initial enzyme activation step (2 min, 95\noC). The\nPCR steps were as follows: 40 cycles of denaturation\n(5 sec, 95\noC), then annealing and extending (20 sec,\n60oC). After amplification, melting curves were ac-\nquired by stepwise increases of temperature from 50\nto 95oC to ensure that a single product was amplified\nand no primer-dimer structures were formed. Con-\ntrol reactions in the absence of the template or\nprimers were performed to confirm that products\nwere free from genomic DNA contamination. Disso-\nciation curves analysis was carried out after each re-\naltime experiment to confirm the presence of only\none amplification product. Data were normalized us-\ning the ΔΔ C\nt method. Samples were amplified in du-\nplicates. Data are shown as the average fold increase,\nwith S.E.M., and are expressed relative to the house-\nkeeping gene GAPDH.\nWestern blotting\nProtein expression for Cx43 in the tissues and\ncells was determined by Western blotting. Proteins\nfrom homogenized tissues and in vitro cultured cells\nwere released with lysis buffer containing 50 mM\nTris-HCl (pH 8.0), 150 mM NaCl, 5 mM EDTA,\n0.1% SDS, 1% TritonX-100, 0.5% sodium\ndeoxycholate and protease inhibitors (Sigma, P8340).\nThe lysates were stored at -86\noC until further analy-\nsis. Protein concentrations were measured by the\nBradford’s method.\nWestern blot analysis was performed as previous-\nly described (Korzekwa et al. 2011). Equal amounts\nof protein were dissolved in SDS gel-loading buffer,\nheated to 95\noC for 4 min and separated in 10%\nSDS-PAGE gel. Separated proteins were electrob-\nA.J. Korzekwa et al.612\n\nA) B)\nC) D)\n20 µm\nFig. 1. Immunfluorescence detection of Cx43 in uterine tissues from control cows and from cows with adenomyosis. A – nega-\ntive control, B – Cx43 in adenomyotic tissue, C – in healthy tissue. Arrows indicate the most intense histochemical reactions.\nScale bars: 20 μm.\nlotted onto 0.2 μm nitrocellulose membranes in\ntransfer buffer. After blocking in 5% non-fat dry\nmilk in TBS-T buffer for 1.5 h at RT, the membranes\nwere incubated overnight with a 1:500 dilution of\nanti-Cx43 (Sigma, C8093) antibodies; GAPDH\n(Sigma, G8795; monoclonal anti-glyceraldehyde-3-\n-phosphate dehydrogenase antibody produced in\nm o u s e )e x p r e s s i o nw a su s e da sar e f e r e n c e .P r o t e i n s\nwere detected by incubating the membranes with\na 1:20,000 dilution of secondary polyclonal\nanti-mouse alkaline phosphatase-conjugated anti-\nbody (Sigma, A 3562) for 1.5 h at RT. Western blots\nwere quantitated using the Kodak 1 D program\n(Eastman Kodak, Rochester, NY, USA).\nHormone determination\nMeasurements of P4 in blood plasma were per-\nformed using a direct radioimmunoassay (RIA; DI-\nASource ImmunoAssays S.A., Nivelles, Belgium). The\nstandard curve ranged from 0.12 to 36 ng/ml and the\neffective dose for 50% inhibition (ED 50) of the\nassay was 0.05 ng/ml. The intra- and inter-assay coeffi-\ncients of variation (CV) were 6.5% and 8.6%,\nrespectively.\nStatistical analysis\nIn Experiment 1, statistically significant differen-\nces in mRNA and protein expression between healthy\nand adenomyotic uterine tissue were detected using\nStudent’s t-test (GraphPad PRISM Version 5.00, San\nDiego, CA, USA). The data are shown as the mean\n± SEM of values obtained in nine separate experi-\nments. The results were considered significantly dif-\nferent when p<0.05.\nIn Experiment 2, statistically significant differen-\nces in mRNA expression between healthy and aden-\nomyotic uterine cells and between control and E2\nstimulated cells were performed using two-way\nANOVA followed by the Bonferroni multiple com-\nparison test (GraphPad). All data were expressed as\nmeans ± SEM. Differences were considered signifi-\ncant at p<0.05.\nResults\nPreliminary division of the material\nBased on microscopic examination of hematoxylin\nand eosin stained tissue slices, bovine uteri were\ndivided into two groups: control and adenomyotic.\nEstradiol reduces connexin43 gap junctions... 613\n\n0.0\n0.5\n1.0\n1.5\n2.0\n2.5\n*\n0.0\n0.2\n0.4\n0.6\n0.8\nhealthy adenomyotic\ncows\n*\nGAPDH (37 kDa)\nCx43 (43 kDa)\na)\nb)\nCx43/GAPDH protein expression\n(arbitary units)\nCx43/GAPDH mRNA expression\n(arbitary units)\nFig. 2. mRNA (a) and protein (b) expression of Cx43 in\nuterine tissues obtained from control cows and from cows\nwith adenomyosis. Data were normalized against glyceral-\ndehyde-3-phosphate dehydrogenase ( GAPDH). Bars repre-\nsent the mean ± SEM. Asterisks indicate statistical differ-\nence between uterine normal and adenomyotic tissues\n(p<0.05), as determined by Student’s t-test. Representative\nblots for Cx43 and GAPDH are shown below the graphs.\nExperiment 1. Localization, mRNA and protein\nexpression of connexin43 in uterine tissues\nof healthy and adenomyotic cows\nImmunofluorescence revealed the localization of\nCx43 in uteri of healthy and adenomyotic cows. The\nprotein was localized in the endometrium as well in\nthe myometrium layer of uteri. Figure 1A represents\ncontrol tissue with Cx43 antibody omitted. In the case\nof adenomyotic tissues, immunoreactivity was ob-\nserved near the glands located in endometrium\nand myometrium, characteristic for adenomyosis\n0.0\n0.2\n0.4\n0.6\n0.8\n1.0\n2\n4\n6\nestradiol (10\n-7 M)\ncontrol\n0\n1\n2\n3\n3\n4\n5\n6\n*\na\nb\n0\n1\n2\n3\n4\n5\n6\n*\na\nb\na\nb\ncontrol E2 control E2\nhealthy adenomyosis\na) STROMAL CELLS\nb) MYOMETRIAL CELLS\nc) COCULTURE CELLS\nCx43/GAPDH mRNA expression (arbitrary units)\nFig. 3. mRNA expression of Cx43 in stromal (a), myometrial\n(b) and co-cultured, stromal and myometrial (c) uterine cells\nisolated from control cows and from cows with adenomyosis.\nData were normalized against glyceraldehyde-3-phosphate\ndehydrogenase (GAPDH). Statistical differences were ana-\nlysed by two-way ANOVA followed by the Bonferroni\nmultiple comparison test (GraphPad). All data were ex-\npressed as means ± SEM. Differences were considered sig-\nnificant at p <0.05. Asterisks indicate statistical differences\nbetween uterine normal and adenomyotic groups and differ-\nent letters (a, b) between control and E2 treatment\n(* p<0.05).\n(Fig. 1B). In healthy cows Cx43 immunoreactivity\nwas observed in endometrium, and near blood vessels\nin myometrium (Fig. 1C,D).\nmRNA and protein expression of Cx43 was de-\ncreased in adenomyotic uterine tissue compared with\nnormal uteri (p<0.05, Fig. 2A,B).\nA.J. Korzekwa et al.614\n\nExperiment 2. The effect of 17-beta estradiol\non Cx43 mRNA expression in uterine stroma\n– myometrium co-culture of healthy\nand adenomyotic cows\nThe expression of Cx43 was not different between\nhealthy and adenomyotic cows and E2 did not influ-\nence Cx43 expression in stromal cells (p >0.05;\nFig. 3A).\nIn myometrial cells, expression of Cx43 increased\nafter E2 stimulation in the adenomyotic group and the\neffect of E2 was higher in adenomyotic myometrial\ncells compared with healthy cells (p<0.05; Fig. 3B).\nIn co-culture of stromal and myometrial cells,\nmRNA expression of Cx43 was elevated after E2\nstimulation (p <0.05). Moreover the increase in ex-\npression after E2 treatment was higher in aden-\nomyotic compared with healthy cell co-cultures\n(p<0.05; Fig. 3C).\nDiscussion\nThe present study demonstrated, for the first time,\nCx43 mRNA and protein expression and localization\nin uterine tissue and cells during the estrous cycle in\ncows. mRNA and protein expression of Cx43 was de-\ncreased in adenomyotic uterine tissue compared with\nnormal uteri. Moreover differences in Cx43 express-\nion during adenomyosis including the effect of E2\nwere examined. In uterine stroma – myometrium\nco-culture of adenomyotic cows, E2 caused increase\nof Cx43 mRNA expression with healthy cell co-cul-\ntures.\nSo far, Cx43 function have been described in con-\nnection with pregnancy and labor in human (Laws et\nal. 2008), porcine (Romek and Karasiński 2011) and\nmouse (Doring et al. 2006) uteri. All these studies\nreported an increase in Cx43 in the myometrium dur-\ning term and preterm delivery as essential for the de-\nvelopment of uterine contractions. Xu et al. (2015)\nhave showed lately that in human uterine myometrial\ncells prostaglandin F2alpha directly stimulated Cx43\nprotein expression. Thus, prostaglandins as main fac-\ntors influencing on uterine function regulate Cx43 and\nthis regulation may also differ during adenomyosis but\nit demands further study.\nDuring pathological conditions, gap junctions in\nthe uterus are supposed to be dysregulated. Connex-\nin43 immunohistostaining and protein expression\nwere inhibited in uterus with diagnosed endometriosis\ncomparing with uteri of healthy women (Yu et al.\n2014). In our study, we demonstrated decreased Cx43\nmRNA and protein expression in adenomyotic uteri,\nwhich is in accordance with the results concerning\nCx43 expression during endometriosis in women.\nNevertheless, in our in vitro experiment, Cx43 mRNA\nexpression was similar in stromal, myometrial and\nco-cultured cells of healthy and adenomyotic cows.\nThe discrepancies between results obtained from\nuterine tissue and cells may arise because Cx express-\nion is regulated by other factors that are not present\nduring culture of pure cells.\nThe pathogenesis and aetiology of both endomet-\nriosis and adenomyosis are still unclear but during\nboth disorders E2 overproduction has been observed\n(Takahashi et al. 1989, Chen et al. 2010, Korzekwa et\nal. 2013). In our study, E2 stimulated Cx43 mRNA\nexpression in myometrial and co-cultured both my-\nometrial and stromal cells collected from adenomyotic\ncows as well as in co-cultures of cells derived from\nhealthy animals.\nEstradiol is necessary to provoke Cx43 expression\nif the putative estrogen response elements are located\nin the Cx43 gene region (Lefebvre et al. 1995). The\nconnection between Cx43 and endometriosis of uterus\nwere described by Regidor et al. (1997). These\nauthors observed Cx43 by immunohistostaining,\nwhich was correlated with a high E2 serum level in\nwomen. Our results indicate that during adenomyosis,\nespecially in myometrial cells, E2 which is a potential\ninducer of this disease increases gap junction forma-\ntion in the bovine uterus, which may support the mi-\ngration of glands from the stroma to the myometrium.\nIn conclusion, our results revealed that during ad-\nenomyosis in the bovine uterus, gap junctions are ex-\npressed at a lower level than in healthy cows in the\nmid-luteal stage. Moreover, Cx43 expression is stimu-\nlated by E2 in adenomyotic myometrial and co-cul-\ntured stromal and myometrial uterine cells. These\ndata suggest the involvement of gap junctions in de-\nvelopment of adenomyosis and a correlation between\nE2 influence and gap junction increase. However, elu-\ncidating the connections between steroids and gap\njunctions and the molecular mechanism of E2 action\nduring adenomyosis of the uterus in cattle will require\nfurther study.\nAcknowledgements\nThe authors wish to thank W. Krzywiec for techni-\ncal support and tissue collection and M. Domin – the\nowner of the slaughterhouse (Meat Processing Plant\n„Warmia”, Biskupiec, Poland) for permitting collec-\ntion of the material. We are grateful to dr. G. Bodek\n(Laboratory in vitro, Institute of Animal Reproduc-\ntion and Food Research of PAS, Olsztyn, Poland) for\nmaking pictures of uterine Cx43 histostaining ob-\nserved under fluorescence confocal microscope.\nEstradiol reduces connexin43 gap junctions... 615\n\nReferences\nAraya R, Eckardt D, Maxeiner S, Kru¨ger O, Theis M, Will-\necke K, Sa´ez JC ( 2005) Expression of connexins during\ndifferentiation and regeneration of skeletal muscle: func-\ntional relevance of connexin43. J Cell Sci 118: 27-37.\nAzziz R ( 1989) Adenomyosis: current perspectives. Obstet\nGynecol Clin North Am 16: 221-235.\nCampo S, Campo V, Benagiano G ( 2012) Infertility and\nadenomyosis. Obstet Gynecol Int 2012: 786132.\nChen CC, Lien HY, Hsu YJ, Lin CC, Shih CM, Lee TM\n(2010) Effect of pravastatin on ventricular arrhythmias in\ninfarcted rats: role of connexin43. J Appl Physiol\n109: 541-552.\nChen YJ, Li HY, Hunag CH, Twu NF, Yen MS, Wang PH,\nChou TY, Liu YN, Chao KC, Yang MH ( 2010) Oestro-\ngen-induced epithelial-mesenchymal transition of en-\ndometrial epithelial cells contributes to the development\nof adenomyosis. J Pathol 222: 261-270.\nChung TH, Wang SM, Wu JC ( 2004) 17beta-estradiol re-\nduces the effect of metabolic inhibition on gap junction\nintercellular communication in rat cardiomyocytes via the\nestrogen receptor. J Mol Cell Cardiol 37: 1013-1022.\nDi WL, Lachelin GC, McGarrigle HH, Thomas NS, Becker\nDL (2001) Oestriol and oestradiol increase cell to cell\ncommunication and connexin43 protein expression in hu-\nman myometrium. Mol Hum Reprod 7: 671-679.\nDoring B, Shynlova O, Tsui P, Eckardt D, Janssen-Bienhold\nU, Hofmann F, Feil S, Feil R, Lye SJ, Willecke K (2006)\nAblation of connexin43 in uterine smooth muscle cells of\nthe mouse causes delayed parturition. J Cell Sci\n119: 1715-1722.\nEvans WH, Martin PE. Gap junctions: structure and func-\ntion (review) (2002) Mol. Membr. Biol 19: 121-136.\nFerenczy A ( 1998) Pathophysiology of adenomyosis. Hum\nReprod 4: 312-322.\nGrummer R, Chwalisz K, Mulholland J, Traub O, Winter-\nhager E ( 1994) Regulation of connexin26 and connex-\nin43 expression in rat endometrium by ovarian steroid\nhormones. Biol Reprod 51: 1109-1116.\nKatkiewicz M, Wierzchoń M, Boryczko Z ( 2005) Aden-\nomyosis of cows; uterus – latent cause of infertility? Med\nWet 61: 1378-1381.\nKorzekwa AJ, Bah MM, Gęstwicka M, Socha B, Skarżyński\nDJ (2013) Adenomyosis in the bovine uterus: correlation\nbetween frequency, age, and 17β-estradiol – progesterone\nequilibrium. Theriogenology 79: 165-172.\nKorzekwa AJ, Bodek G, Bukowska J, Blitek A, Skarzynski\nDJ ( 2011) Characterization of bovine immortalized\nluteal endothelial cells: action of cytokines on production\nand content of arachidonic acid metabolites. Reprod Biol\nEndocrinol 9: 1-9.\nKorzekwa A, Łupicka M, Socha B, Mannelli Ch, Skarzynski\nDJ (2014) Is adenomyosis a problem in reproduction and\nfertility? Pol J Vet Sci 17: 187-194.\nLaws MJ, Taylor RN, Sidell N, DeMayo FJ, Lydon JP, Gut-\nstein DE, Bagchi MK, Bagchi IC ( 2008) Gap junction\ncommunication between uterine stromal cells plays\na critical role in pregnancy-associated neovascularization\nand embryo survival. Development 135: 2659-2668.\nLefebvre DL, Piersanti M, Bai XH, Chen ZQ, Lye SJ (1995)\nMyometrial transcriptional regulation of the gap junction\ngene, connexin-43. Reprod Fertil Dev 7: 603-611.\nŁupicka M, Socha B, Szczepańska A, Korzekwa A ( 2015)\nExpression of pluripotency markers in the bovine uterus\nwith adenomyosis. Reprod Biol Endocrinol 13: 110.\nLye SJ, Nicholson BJ, Mascarenhas M, MacKenzie L, Pet-\nrocelli T ( 1993) Increased expression of connexin-43 in\nthe rat myometrium during labor is associated with an\nincrease in the plasma estrogen:progesterone ratio. En-\ndocrinology 132: 2380-2386.\nMayerhofer A, Garfield RE ( 1995) Immunocytochemical\nanalysis of the expression of gap junction protein connex-\nin 43 in the rat ovary. Mol Reprod Dev 41: 331-338.\nMiyamoto Y, Skarzynski DJ, Okuda K (2000) Is tumor nec-\nrosis factor α a trigger for the initiation of endometrial\nprostaglandin F\n2α release at luteolysis in cattle? Biol Rep-\nrod 62: 1109-1115.\nMoreira L, Carvalho ECQ, Caldas-Bussiere MC (2007) His-\ntopathological aspects of adenomyosis in bovine uteri in\ndifferent phases of the estrous cycle. Arq Brasi Med Vet\nZootec 59: 1097-1102.\nPetrocelli T, Lye SJ (1993) Regulation of transcripts encod-\ning the myometrial gap junction protein, connexin-43, by\nestrogen and progesterone. Endocrinology 133: 284-290.\nRegidor PA, Regidor M, Schindler AE, Winterhager\nE( 1997) Aberrant expression pattern of gap junction\nconnexins in endometriotic tissues. Mol Hum Reprod\n3: 375-381.\nRomek M, Karasinski J (2011) Quantification of connexin43\ngap junctions in porcine myometrium by confocal micro-\nscopy and stereology. Reprod Domest Anim 46: 29-38.\nSaez JC, Berthoud VM, Branes MC, Martinez AD, Beyer EC\n(2003) Plasma membrane channels formed by connexins:\ntheir regulation and functions. Physiol. Rev 83: 1359-1400.\nTakahashi K, Nagata H, Kitao M ( 1989) Clinical usefulness\nof determination of estradiol levels in the menstrual\nblood for patients with endometriosis. Nippon Sanka\nFujinka Gakkai Zasshi 41: 1849-1850.\nTaylor AH, Kalathy V, Habiba M ( 2014) Estradiol and\ntamoxifen enhance invasion of endometrial stromal cells\nin a three-dimensional coculture model of adenomyosis.\nFertil Steril 101: 288-293.\nvan Engelen E, Breeveld-Dwarkasing VN, Everts ME, van\nder Weyden GC, Taverne MA, Rutten VP ( 2009)\nSmooth muscle cells of the bovine cervical stroma may\nhave a secretory, rather than a contractile function during\nparturition. Reprod Domest Anim 44: 303-311.\nWiesen JF, Midgley AR Jr ( 1993) Changes in expression of\nconnexin 43 gap junction messenger ribonucleic acid and\nprotein during ovarian follicular growth. Endocrinology\n133: 741-746.\nWiesen JF, Midgley AR Jr (1994) Expression of connexin 43\ngap junction messenger ribonucleic acid and protein dur-\ning follicular atresia. Biol Reprod 50: 336-348.\nWrenzycki C, Herrmann D, Carnwath JW, Niemann\nH( 1996) Expression of the gap junction gene connexin43\n(Cx43) in preimplantation bovine embryos derived in\nvitro or in vivo. J Reprod Fertil 108: 17-24.\nXu C, You X, Liu W, Sun Q, Ding X, Huang Y, Ni X (2015)\nProstaglandin F\n2α regulates the expression of uterine acti-\nvation proteins via multiple signaling pathways. Repro-\nduction 149: 139-146.\nA.J. Korzekwa et al.616\n\nYu J, Berga SL, Zou W, Sun HY, Johnston-MacAnanny E,\nYalcinkaya T, Sidell N, Bagchi IC, Bagchi MK, Taylor\nRN (2014) Gap junction blockade induces apoptosis in\nhuman endometrial stromal cells. Mol Reprod Dev.\n81(7): 666-75.\nZeiler M, Leiser R, Johnson GA, Tinneberg HR,\nPfarrer C ( 2007) Development of an in vitro model for\nbovine placentation: a comparison of the in vivo and in\nvitro expression of integrins and components of extracellu-\nlar matrix in bovine placental cells. Cells Tissues Organs\n186: 229-42.\nEstradiol reduces connexin43 gap junctions... 617","source_license":"CC0","license_restricted":false}