{"paper_id":"afc1d3db-edac-41c8-818b-26b2d2d84623","body_text":"Glucose transporter expression in\neutopic endometrial tissue and\nectopic endometriotic lesions\nBrett McKinnon 1,2, Dominic Bertschi 1, Carlos Wotzkow 2, Nick A Bersinger 1,2,\nJakob Evers 1 and Michael D Mueller 1,2\n1Department of Obstetrics and Gynecology, Inselspital, Berne University Hospital, Effingerstrasse 102,\nBerne CH-3010, Switzerland\n2Department of Clinical Research, University of Bern, Murtenstrasse 35, Bern CH-3010, Switzerland\nCorrespondence\nshould be addressed\nto B McKinnon\nEmail\nbrett.mckinnon@\ndkf.unibe.ch\nAbstract\nEndometriosis is an extremely prevalent disorder characterized by the growth of\nendometrial tissue at ectopic locations. Glycolysis is an energy-producing mechanism\nthat occurs in almost all cells and requires an adequate uptake of glucose mediated by\nglucose transporter (GLUT) proteins. At present, however, very little is known about\ntheir expression in either the endometrium or the endometriotic lesions. The objective of\nthis study was to examine the expression of SLC2A genes in the endometrium of women\nwith and without endometriosis and in the matching ectopic tissue, and to confirm\nthe presence of the GLUT proteins in ectopic lesions. There was a significantly higher\nexpression of SLC2A3 and a significantly lower expression of SLC2A4 in women with\nendometriosis compared with those without. In women with endometriosis, the\nectopic expression of SLC2A3, SLC2A4 and SLC2A5 was significantly higher than that\nobserved in the matching eutopic tissue. GLUT1 protein expression was present in\nboth epithelial and stromal cells and GLUT3 was confined to CD45-positive leukocytes.\nGLUT4 expression was strong in both ectopic epithelial and stromal cells and localized to\nthe cellular membrane in epithelial cells. These results show that GLUT expression is\naltered between eutopic and ectopic tissue and between women with and without\nendometriosis, and that GLUT4 may represent a significant entry route for glucose into\nthe endometriotic epithelial cells. The inducible nature of GLUT4 and its limited cellular\nexpression may make GLUT4 an attractive target for non-hormone-based treatments\nof endometriosis.\nKey Words\n\" endometriosis\n\" GLUT4\n\" endometrium\n\" glucose\n\" SLC2A\nJournal of Molecular\nEndocrinology\n(2014) 52, 169–179\nIntroduction\nEndometriosis is an extremely prevalent estrogen-\ndependent gynecological disorder affecting at least 10% of\nreproductive-aged women worldwide (Eskenazi & Warner\n1997). It is characterized by the growth of endometrial\ntissue outside the uterine cavity and can result in severe\npelvic pain (Evans et al.2 0 0 7) and subfertility (D’Hooghe\net al. 2003). Currently, the most widely accepted theory of\nendometriosis etiology purports that retrograde menstrua-\ntion allows the implantation of viable endometrial cells at\nectopic sites ( Sampson 1927, Halme et al .1 9 8 4). Ectopic\nJournal of Molecular Endocrinology\nResearch\nB MCKINNON and others GLUT expression in\nendometriosis\n52:2 169–179\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access\n\n\nlesions then secrete chemotactic molecules that recruit\nimmune cells, further stimulating the continued growth of\nthe lesion.\nGlycolysis is an energy-producing mechanism that\nrequires an adequate supply of intracellular glucose\nuptake mediated by the glucose transporter (GLUT)\nproteins encoded by the SLC2A gene family. These\nproteins transport glucose into the cell via a facilitated\ndiffusion down a concentration gradient ( Scheepers et al.\n2004). Cloning of the first gene to encode a GLUT protein\noccurred in 1985 ( Mueckler et al . 1985 ), after which a\nfurther 13 related members of the SLC2A gene family\nwere subsequently identified in human (Uldry & Thorens\n2004).\nLittle information is available on GLUT expression\nin both endometrium and endometriotic lesions. GLUT1\nis a ubiquitously expressed protein, responsible for the\nbasal glucose uptake and is the most prominent GLUT in\nthe endometrium ( Frolova & Moley 2011 b). The high-\naffinity GLUT3 transporter has also been reported in\neutopic endometrial tissue, although its expression is\nconfined to CD45-positive leukocytes ( von Wolff et al .\n2003, Korgun et al . 2005 ). The insulin-sensitive GLUT4\nhas also been identified in endometrial tissue, but with\ncontradictory results. No RNA expression was detected in\nendometrial biopsies (von Wolff et al. 2003) or in isolated\nendometrial stromal cells ( Frolova & Moley 2011 a),\nalthough it has been detected in epithelial endometrial\ncells (Mioni et al. 2004, Mozzanega et al. 2004). The mRNA\nexpression of the GLUT6, GLUT8, GLUT10 and GLUT12\ngenes has also been identified in a single study (Frolova &\nMoley 2011b), although their role in the endometrium has\nnot been examined.\nThe objective of this study was therefore to examine\nboth the eutopic endometrial tissue of women with and\nwithout endometriosis, as well as matching ectopic\nendometriotic lesions to determine the expression of the\nSLC2A genes and their cellular localization.\nSubjects and methods\nSample collection and patient data\nWomen with idiopathic infertility or pelvic pain were\nrecruited for the study, and prior to laparoscopic surgery,\ninformed consent, patient details and medication history\nwere collected. All laparoscopies were performed in the\nproliferative phase of the menstrual cycle, and during\nsurgery any endometrioti c lesions identified were\nremoved and their location noted. In a subset of patients,\nendometrial biopsies were also collected using a soft\ncurette (Pipelle-de-Cornier, Laboratoire CCD, Paris,\nFrance) and were split into two groups based on the\npresence or absence of histologically-confirmed endo-\nmetriosis. No endometrial biopsies were included if\nthe patients reported hormonal medication use in the\n3 months prior to surgery. All tissue collected was stored\nin RNAlater at K80 8C until further use.\nIn cases were an ectopic endometrial lesion was\nidentified, irrespective of whether a matching biopsy was\ncollected, all ectopic tissue removed during surgery and\ncollected and stored in RNAlater (Invitrogen Life Tech-\nnologies) at K80 8C for subsequent analysis, as was the\nperitoneal fluid. For subsequent analysis of ectopic tissue,\nall samples collected were first split into groups based on\nhormonal treatment including no hormonal medication,\nestrogen- or progesterone-based hormonal contraceptives\n(HCs), or gonadotropin-releasing hormone analogs\n(GnRHa). In addition, within the no hormonal treatment\ncohort, an additional three groups based on lesion\nlocation were established (peritoneal, ovarian and recto-\nvaginal septum (RVS)). As only two lesions with no\nhormonal treatment were available from the RVS group,\nthis region was excluded from the statistical analysis.\nThe presence of endometriosis was confirmed by\nhistological examination, and the menstrual stage of\nwomen not using hormonal medication was confirmed\nby measuring progesterone in the peritoneal fluid with a\nRIA (Coata-count, DPC; Buhlmann Laboratories, Allschwil,\nSwitzerland) ( Supplementary Figure 1 ,s e es e c t i o no n\nsupplementary data given at the end of this article). The\ncutoff value was 27 nmol/l. Institutional Review Board\napproval was obtained from the ethical committee prior\nto the commencement of the study.\nAnalysis of SLC2A gene expression in endometrial and\nendometriotic tissue\nFor both the eutopic endometrial biopsies and ectopic\nlesions, w30 mg tissue was excised and homogenized\nin the FastPrep 120 tissue homogenizer (30 s at 4.0 m/s)\nin cell lysis buffer (Qiagen). The remaining RNA isolation\nwas performed with the RNAeasy Mini Kit (Qiagen)\nand TurboDNase (Ambion, Life Technologies, Zug,\nSwitzerland) for genomic DNase digestion. One microgram\nof total RNA was reverse transcribed in a 25 ml reaction\nwith the Moloney murine leukemia virus reverse tran-\nscriptase (Promega) and random primers. The resulting\ncDNA was diluted 1:20 and the absence of genomic DNA\nconfirmed with a reverse transcriptase-negative control.\nJournal of Molecular Endocrinology\nResearch B MCKINNON and others GLUT expression in\nendometriosis\n52:2 170\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access\n\n\nThe quantitative real-time qPCR was performed\nwith the Rotor-gene SYBR Green PCR Kit (Qiagen) in a\nRotor-gene RG 2000 (Corbett Research, Sydney, NSW,\nAustralia) under the following conditions: 958Cf o r5m i n ,\nfollowed by 40 cycles of 958Cf o r5sa n d6 08C for 10 s. Primers\nwere purchased from Microsynth (Balgach, Switzerland),\nand the sequences are defined in Table 1 . Specificity of\namplification was confirmed by melt curve analysis.\nThe four most stable reference genes among all\nsamples were selected via the geNORM Software program\n(Zwijnaarde, Ghent, Belgium) and the geometric mean\nof all four reference genes was used to normalize the\nexpression of the genes of interest ( Vandesompele et al .\n2002). The reaction efficiency of each assay was\ndetermined via linear regression ( Ruijter et al . 2009) and\nthe relative mRNA expression calculated with the qBASE-\nplus Software (Biogazelle, Zwijnaarde, Belgium).\nThe comparison between SLC2A gene expression in\neutopic endometrium from women with and without\nendometriosis and the comparison between eutopic\nexpression and ectopic expression in matching samples\nfrom women with endometriosis were both performed\nusing the non-parametric Mann–Whitney U test, as was\nthe mRNA comparison between peritoneal and ovarian\nlesions. The comparison between hormonal treatments was\nperformed with a non-parametric one-way ANOVA test with\na post-hoc Dunn’s multiple comparison tests for each group.\nImmunodetection in endometriotic lesions and\nsurrounding tissue\nAfter RNA had been harvested from the ectopic tissue,\nany remaining parts of the samples were fixed in formalin\nand embedded into paraffin (FFPE) blocks. Previous studies\nhave confirmed the effectiveness of using tissue stored in\nRNAlater for immunohistochemistry ( Florell et al . 2001).\nFor imaging studies, serial sections of 4mM thickness were\ncut from each FFPE block and mounted onto glass slides\n(SuperFrost, Braunschweig, Germany). Prior to analysis,\nthe sections were dewaxed in xylene and rehydrated\nthrough a series of ethanol. Epitope retrieval for all\nantibodies was performed via heating slides in 10 mM\ncitrate buffer at pH 5.5 for 5 min in a 450 W microwave\noven. Endogenous peroxidase activity was blocked by\ntreating sections with 3% hydrogen peroxide (H\n2O2) for\n10 min and non-specific binding blocked by incubation\nwith 3% BSA in Tris-buffered saline (TBS; Tris 100 mM,\nNaCl 0.15 M; pH 7.4) for 30 min.\nPrimary antibody incubations were performed over-\nnight in TBS containing 3% BSA with the following\ndilutions: 1:50 for rabbit anti-human GLUT1 and rabbit\nanti-human GLUT4, 1:25 for rabbit anti-human GLUT3,\nand 1:75 for rabbit anti-human CD45 (all from Abcam,\nCambridge, UK). After incubation, slides were washed with\nTBS containing 0.1% Tween 20 (TBST) and subsequently\nincubated with affinity purified, biotin-conjugated goat\nanti-rabbit IgG (Dako, Glostrup, Denmark) for 90 min at\nroom temperature. After washing, the sections were\nincubated with avidin–biotin–HRP complex (Vectastain\nABC Kit, Vector Laboratories, Burlingame, CA, USA) for\n45 min. Detection of bound antibody was via 1–5 min\nincubation at room temperature with 3,3\n0-diaminobenzidine\nsubstrate. Slides were counterstained with hematoxylin\nand mounted in Aquatex (Merck). Negative controls were\nperformed in the absence of primary antibody. Immuno-\nstained slides were photographed with a Nikon Eclipse E800\nmicroscope (Nikon, Japan) at a magnification of 20! and\n60!, and the endometriotic lesions, large nerve fiber\ntrunks and blood vessels were identified via the hemato-\nxylin staining. Individual leukocytes were identified via\nCD45 reactivity.\nFor immunofluorescence, the sample was prepared\nexactly as described for the immunohistochemistry and\nco-staining was performed by incubating the sections\nT able 1 Primer sequences used for real-time PCR\nGene GenBank no. Forward Reverse Product size (bp)\nGAPDH NM_002046 TGCACCACCAACTGCTTAGC GGCATGGACTGTGGTCATGAG 86\nACTB NM_001101 CTGGAACGGTGAAGGTGACA AAGGGACTTCCTGTAACAATGCA 96\nYWHAZ NM_003406 ACTTTTGGTACATTGTGGCTTCAA CGCCAGGACAAACCAGTAT 92\nRPL13A NM_012423 CCTGGAGGAGAAGAGGAAAGAGA TTGAGGACCTCTGTGTATTTGTCAA 125\nSLC2A1 NM_006516 AAGGTGATCGAGGAGTTCTACA ATGCCCCCAACAGAAAAGATG 119\nSLC2A2 NM_000340 TTGCTGGAAGAAGCATATCAGG TGACTAATAAGAATGCCCGTGAC 148\nSLC2A3 NM_006931 GTTCCCCTCACTGGATGAAA TATTTGGATGGCTCTCCCAC 95\nSLC2A4 NM_001042.2 CGTCGGGCTTCCAACAGATA CGCAGAGAACACAGCAAGGA 89\nSLC2A5 NM_003039 CAAGAAAGCCCTACAGACGC AACAGCTTCAGCACGGAGAT 117\nSLC2A7 NM_207420 TGTGCAGGCATCTCCTACAG CGAAAACCTCGGTCATTGTT 97\nSLC2A8 NM_014580.3 GTCCTCACCAACTGGCTCAT CAAGCCAGAAGGCTCCATAG 91\nJournal of Molecular Endocrinology\nResearch B MCKINNON and others GLUT expression in\nendometriosis\n52:2 171\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access\n\n\nwith both the rabbit anti-human GLUT4 (1:25) and the\ngoat anti-human cytokeratin 19 (CK19; 1:500 dilution,\nSanta Cruz Biotechnology) antibody in 3% BSA solution\nof TBS. Secondary antibodies were anti-goat IgG Cy5\n(Millipore, Zug,Switzerland)and anti-rabbitIgG Alexa-Fluor\n488 (Invitrogen). Nuclei were stained with 406-diamidino-\n2-phenylindole (DAPI) and slides mounted with Prolong\nGold Antifade (Invitrogen). Images were photographed with\nthe Zeiss LSM 710 Confocal Microscope. Negative controls\nwere performed in the absence of the primary antibodies.\nResults\nSample collection\nEndometrial biopsies were collected from 13 women who\nwere confirmed to have no visible endometriotic lesions\nand 15 women with endometriosis. There was no signi-\nficant difference in either the age or the BMI between the\ntwo groups (Table 2). We also collected a total of 18 ectopic\nlesions from the 15 endometriotic women, as multiple\nectopic samples were collected from two women. One\nwomen had three lesions, two of which were peritoneal\nand one in the RVS; whereas the second women had\ntwo lesions, one each from the peritoneal cavity and\nthe ovary.\nAdditional ectopic tissue without matching eutopic\nbiopsy was collected; 48 samples in total. Of these samples,\n24 were from women with no hormonal treatment and\n12 of these lesions were located in the peritoneal region,\nten on the ovaries, and only two in the RVS. To determine\nthe effect of hormonal treatment on SLC2A gene\nexpression, a further 14 samples were available from\nwomen taking estrogen- or progesterone-based HC and\nten women using GnRHa.\nFor the immunoanalysis of ectopic lesions, sufficient\nremaining tissue was available from 19 patients. These\nsamples included eight from women who were not using\nhormonal treatments, six using HC, and five using GnRHa.\nThe tissue collected for this study is listed inTable 2.\nSLC2A gene expression in eutopic and ectopic\nendometrial tissue\nThere was a significantly increased expression of SLC2A3\nin the endometrium of women with endometriosis\ncompared with those without. By contrast, there was a\nsignificantly decreased expression of SLC2A4 in the\nendometrial tissue of women with endometriosis com-\npared with women without endometriosis ( Fig. 1A). No\nsignificant difference wa s observed between the two\ngroups for any of the other genes examined. No expression\nof SLC2A2 was observed in either group.\nIn women with endometriosis, there was a signi-\nficant increase in SLC2A3, SLC2A4 and SLC2A5 observed\nin the ectopic lesions compared with their matching\neutopic endometrium ( Fig. 1B ). No significant variation\nwas observed in SLC2A1 , SLC2A 7a n d SLC2A 8. No\nexpression was observed for SLC2A2 in any of the ectopic\nor eutopic samples.\nSLC2A gene expression in endometriotic tissue from\ndifferent locations and under hormonal treatments\nThe comparison of mRNA expression of the SLC2A family\nin different regions was only performed in samples\nT able 2 Clinical data and sample parameters\nEutopic tissue Ectopic tissue\nn Age BMI Total P O RVS\nRNA analysis\nHealthy eutopic endometrium 13 30.2 G1.7 24.4 G1.7\nMatching eutopic and ectopic tissue 15 33.3 G1.8 23.7 G1.3 18 a 97 2\nP value 0.2339 0.7522\nEctopic samples – 48 15 24 9\nNo hormonal treatment – 24 12 10 2\nEstrogen/progesterone – 14 1 9 4\nGnRHa – 10 2 5 3\nProtein analysis\nEctopic samples – 19 4 9 6\nNo hormonal treatment – 8 3 3 2\nEstrogen/progesterone – 6 1 3 2\nGnRHa – 5 0 3 2\nP , peritoneal; O, ovarian; RVS, recto-vaginal septum.\naSome eutopic samples have multiple matching ectopic samples.\nJournal of Molecular Endocrinology\nResearch B MCKINNON and others GLUT expression in\nendometriosis\n52:2 172\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access\n\n\nfrom women who had not taken hormonal medication.\nResults show a significantly higher expression of both\nSLC2A1 and SLC2A8 in the peritoneal lesions compared\nwith ovarian lesions ( Fig. 2A ). No significant difference\nwas observed between any of the other genes examined.\nAn analysis on the effect of hormonal treatment on\nSLC2A gene expression found no significant variation\nbetween any of the genes examined (Fig. 2B).\nGLUT1, GLUT3 and GLUT4 protein expression and\ntheir localization in endometriotic lesions and\nsurrounding tissue\nIn the endometriotic lesions, there was strong immuno-\nreactivity of GLUT1 in both the epithelial and the stromal\ncells (Fig. 3A). In the stromal cells, immunoreactivity was\npredominantly nuclear, whereas epithelial cells showed\nboth nuclear and membranous localization ( Fig. 3B ).\nGLUT3 showed only sparse immunoreactivity in scattered\ncells throughout the endometriotic lesions ( Fig. 3C\nand D ). The GLUT3 immunoreactivty coincided with\nthe immunoreactivity observed for CD45 ( Fig. 3E and F ,\nblack arrows). GLUT4 showed strong immunoreactivity\nin both the epithelial and the stromal cells of the endo-\nmetriotic lesions ( Fig. 3G ). In the stromal cells this\nimmunoreactivity appear ed predominantly nuclear\nwhereas the epithelial immunoreactivity was both\nnuclear, and membranous (Fig. 3H). No background signal\nwas observed in the negative controls (Fig. 3I and J).\nAs dissected, endometriotic lesions will include\nstructures other than endometrial tissue. Non-endometrial\ntissue, including blood vessels ( Fig. 4 ) and large nerve\nfiber trunks (Fig. 5) were also examined in these samples.\nStrong GLUT1 immunoreactivity was observed in the\nendothelial cells lining the blood vessels surrounding the\nendometriotic lesions ( Fig. 4A ). This immunoreactivity\nappeared to be predominantly nuclear, with some mem-\nbranous localization (Fig. 4B). GLUT3 showed very limited\nimmunoreactivity in the cells surrounding the blood\nvessels ( Fig. 4C and D ). GLUT4 immunoreactivity was\n80\n60\n40\n16\n12\n8\n4\n0.8\n0.4\n0.0\n8\n6\n4\n2\n0\nRelative mRNA expressionRelative mRNA expression\nSLC2A1 SLC2A3 SLC2A4 SLC2A5 SLC2A7 SLC2A8\nSLC2A1\nA\nB\nSLC2A3\nNon-endometriosis\nEndometriosis\nSLC2A4 SLC2A5 SLC2A7 SLC2A8\nTissue source\nEutopic\nEctopic\n****\n****\n*\n****\n*\nFigure 1\nSLC2A gene expression in the eutopic and ectopic endometrial tissue of\nwomen with and without endometriosis. (A) A significantly higher\nexpression of SLC2A3 was observed in women with endometriosis\n(0.38G0.11, nZ15) compared with those without endometriosis\n(0.13G0.04, nZ13, PZ0.0448). A significantly lower expression of SLC2A4\nwas observed in women with endometriosis (0.13 G0.02, nZ15) compared\nwith those without endometriosis (53.87 G21.07, nZ13, P!0.0001). No\nsignificant difference for SLC2A1, SLC2A5, SLC2A7 or SLC2A8 was\nobserved between women with or without endometriosis. No expression\nin either group was observed in SLC2A2. (B) The eutopic expression\nof SLC2A3 (EuZ0.38G0.11, nZ15 vs Ec Z1.98G0.51, nZ18; P!0.0001),\nSLC2A4 (EuZ0.13G0.02, nZ15 vs Ec Z2.88G1.03, nZ18; P!0.0001), and\nSLC2A5 (EuZ0.15G0.07, nZ15 vs Ec Z0.27G0.10, nZ18; PZ0.0464) was\nsignificantly lower than the paired ectopic expression. No significant\nchange was observed in SLC2A1, SLC2A7 and SLC2A8. All values are\nexpressed as mean G\nS.E.M.* P!0.05 and **** P!0.0001.\nRelative mRNA expression\n8A\nB\n6\n4\n2\n0\nRelative mRNA expression\n6\n4\n2\n0\nSLC2A1\nSLC2A1 SLC2A3\nNo hormone\nHC\nOvarian\nPeritoneal\n*\n*\nGnRHa\nSLC2A4 SLC2A5 SLC2A7 SLC2A8\nSLC2A3 SLC2A4 SLC2A5 SLC2A7 SLC2A8\nFigure 2\nSLC2A gene expression in endometriotic lesions from different locations\nand under hormonal treatment. (A) A significantly stronger expression of\nSLC2A1 (peritonalZ2.93G0.69, nZ12 vs ovarian Z1.23G0.49, nZ10;\nP!0.011) and SLC2A8 (peritonealZ4.71G2.03, nZ12 vs ovarian Z1.71G\n1.37, nZ10; P!0.019) was observed in peritoneal lesions compared with\novarian lesions. No significant variation was observed in SLC2A3, SLC2A4,\nSLC2A5 and SLC2A7. (B) A comparison of SLC2A gene expression in the\nectopic tissue of women using hormonal treatment found no significant\nvariation in all genes examined. All values represent mean G\nS.E.M.* P!0.05.\nJournal of Molecular Endocrinology\nResearch B MCKINNON and others GLUT expression in\nendometriosis\n52:2 173\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access\n\n\nalso observed in the endothelial cells lining the blood\nvessels (Fig. 4E and F ). The negative control showed no\nbackground signal in the endothelial cells (Fig. 4G and H).\nT h el a r g en e r v efi b e rt r u n k ss h o w e dap o s i t i v e\nimmunoreactivity to the GLUT1 antibody ( Fig. 5A ),\nwhich appeared to be predominantly nuclear, although\nwith some immunoreactivity also in the cytosol and the\nmembrane ( Fig. 5B ). GLUT3 showed some weak to\nmoderate immunoreactivity in these large nerve fiber\ntrunks ( Fig. 5C ), which was predominantly nuclear\n(Fig. 5D). GLUT4 immunoreactivity was also moderate to\nstrong in the large nerve fiber trunks ( Fig. 5E), although\nthis immunoreactivity appeared to be almost exclusively\nnuclear (Fig. 5F). Negative controls showed no background\nsignal (Fig. 5G and H).\nLocalization of GLUT4 transporter protein to the\nmembrane of endometriotic epithelial cells\nThe nuclei of all cells in both the positive ( Fig. 6A) and\nthe negative control ( Fig. 6B) were identified with DAPI.\nImmunofluorescence of GLUT4 ( Fig. 6C ), above the\nbackground and autofluorescence observed in the\nnegative control (Fig. 6D), was identified in the epithelial\ncell membranes. Immunofluorescence of CK19 ( Fig. 6E),\nabove the background and autofluorescence observed\nin the negative control ( Fig. 6F ), was also observed in\nthe epithelial cell membranes. An overlay of the images\nshowed a significant co-localization (arrow) of the CK19\n(red staining) and the GLUT4 (green staining) immuno-\nfluorescence in the epithelial membranes ( Fig. 6G ).\nFigure 3\nGLUT protein localization in endometriotic lesions. (A) GLUT1 immuno-\nreactivity was detected in endometriotic lesions. (B) A higher magnification\nshowed that moderate GLUT1 expression was present in both epithelial\n(black arrow) and stromal cells (red arrow). (C) Sparse GLUT3 immuno-\nreactivity was observed in endometriotic lesions. (D) A higher magni-\nfication identified cells with GLUT3 immunoreactivity (black arrows).\n(E) CD45 immunoreactivity was observed within endometriotic lesions.\n(F) A higher magnification identified immunoreactivity of CD45 in cells\nsimilar to that of GLUT3 (black arrows). (G) Strong GLUT4 immunoreactivity\nwas observed in endometriotic lesions. (H) A higher magnification shows\nimmunoreactivity is predominantly nuclear in the stromal cells (red arrow)\nand both nuclear and membranous in the epithelial cells (black arrow).\n(I and J) Negative control showed limited background staining in either\nepithelial or stromal cells. (A, C, E, G and I) Scale bars represent 50 mM and\nthe areas outlined in black represent the region of higher magnification.\n(B, D, F , H and J) Scale bars represent 10 mM.\nJournal of Molecular Endocrinology\nResearch B MCKINNON and others GLUT expression in\nendometriosis\n52:2 174\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access\n\n\nThe absence of significant immunofluorescence in the\nnegative control sample in the same regions confirmed the\nspecificity of both the GLUT4 and CK19 immunofluores-\ncence in the endometriotic epithelial cell membranes\n(Fig. 6H).\nDiscussion\nGLUT proteins are the principle route for glucose entry\ninto the cells. Very little is known, however, about the\nexpression of these transporters in endometrial tissue and\neven less in endometriotic lesions. The results of this study\nshow a significant variation in the expression of both the\nSLC2A3 and SLC2A4 genes in the eutopic endometrial\ntissue of women with and without endometriosis. It also\nreports an upregulation ofSLC2A3, SLC2A4 and SLC2A5 in\nthe ectopic tissue compared with the matching eutopic\ntissue. Further analysis of the GLUT1, GLUT3 and GLUT4\nprotein expression indicates that GLUT1 expression was\npresent in endometriotic lesions, GLUT3 expression was\nrestricted mainly to CD45-positive leukocytes and large\nnerve fiber trunks, and GLUT4 was expressed extensively\nin the ectopic endometrial epithelial and stromal cells\nwith a strong expression in the membrane of the epithelial\ncells. The increased expression of SLC2A4 and the\nlocalization of GLUT4 to the epithelial cell membranes\ncould allow GLUT4 to provide a significant route of\nglucose entry into endometriotic epithelial cells.\nIn eutopic endometrial tissue, we found a significant\nvariation in the gene expression of SLC2A3 and SLC2A4\nbetween women with and without endometriosis.\nSLC2A3, which was found to be significantly increased in\nwomen with endometriosis, is expressed in endometrial\nmacrophages (von Wolff et al. 2003) and it is possible that\nthe increase in GLUT3 gene expression in the endome-\ntrium of women with endometriosis observed in this study\nwas due to the increased number of macrophages found\nin the endometrium of these women (Ota et al. 1996). By\ncontrast, GLUT4 was significantly reduced. The variation\nin SLC2A4 expression between women with and without\nendometriosis is difficult to interpret in the context of\nolder studies as it has not previously been examined in\nwomen with endometriosis, but in women without\nendometriosis, both the presence ( Mioni et al . 2004 ,\nMozzanega et al . 2004 ) and absence ( von Wolff et al .\n2003) of GLUT4 has been reported. It is also possible that\nthese large variations and contradicting results for GLUT4\nexpression may be due to factors not controlled in this\nor other studies, such as serum insulin or glucose levels.\nThe large difference observed, however, represents an\ninteresting finding and a further investigation of the\ndifference inSLC2A4 expression between women with and\nFigure 4\nGLUT expression in surrounding blood vessels. (A) Strong GLUT1 immuno-\nreactivity was observed in the endothelial cells surrounding endometriotic\nlesions. (B) A higher magnification showed that immunoreactivity was\nnuclear with some membranous staining (black arrows). (C and D) Very\nlittle GLUT3 expression was observed in endothelial cells. (E) Moderate\nGLUT4 immunoreactivity was observed in endothelial cells surrounding the\nblood vessel. (F) A higher magnification showed a predominantly nuclear\nbut with some membrane localization (black arrows). (G and H) Negative\ncontrols showed limited background immunoreactivity in either epithelial\nor stromal cells. (A, C, E and G) Scale bars represent 50 mM and the areas\noutlined in black represent the region of higher magnification. (B, D, F\nand H) Scale bars represent 10 mM.\nJournal of Molecular Endocrinology\nResearch B MCKINNON and others GLUT expression in\nendometriosis\n52:2 175\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access\n\n\nwithout endometriosis, taking into account factors such as\nthe insulin and glucose serum concentrations, should be\nperformed on a larger sample size.\nAn analysis of gene expression in ectopic tissue found\na significantly higher expression of SLC2A1 and SLC2A8\nin the peritoneal lesions compared with the ovarian\nlesions. The difference supports the notion that endo-\nmetriotic lesions may be a collection of related but\nindividual diseases ( Nisolle & Donnez 1997 ) and that\novarian and peritoneal lesions may acquire different\namounts of glucose or may even utilize different\nmechanisms to acquire sufficient glucose. In addition, as\nprevious studies have shown that both ectopic lesions\nand the peritoneal microenvironment can be significantly\naffected by hormonal treatments (Nirgianakis et al. 2013),\nwe also examined whether these had an effect on SLC2A\nexpression. No significant variation, however, was\nobserved in the expression of any genes after either HC\ntreatment or GnRHa use, indicating that the effectiveness\nof these drugs is not involved with a modulation of the\nmetabolic capabilities of the lesions.\nIn a comparison between the SLC2A gene expre-\nssion in eutopic and matching ectopic tissue, we found a\nsignificant upregulation of expression in SLCA3, SLC2A4\nand SLC2A5. Of these three genes, there was a large\nexpression of both SLC2A3 and SLC2A4 in the ectopic\ntissue. SLC2A5, although significantly higher in ectopic\ntissue, still showed a very low expression and thus was\nnot considered for further investigation. By contrast,\nSLC2A1 surprisingly showed no significant change in\nexpression event although it is significantly increased by\nthe hypoxic environment created during malignant\nlesion growth ( Goldman et al . 2006 , Szablewski 2013 ).\nWe therefore chose to further analyze the presence of\nGLUT1, GLUT3 and GLUT4 at the protein level in the\nectopic tissue.\nSLC2A1 gene expression in endometriotic lesions\nhas not previously been reported; however, the GLUT1\nprotein expression in benign endometriotic lesions was\nfound in 16–19% of ovarian endometrioma, which\nincreased to 95% in endometriosis-associated clear cell\ncarcinoma ( Kato et al . 2012 ). In the ectopic tissue, we\nfound no significant variation in the SLC2A1 gene\nexpression compared with the matching eutopic tissue,\nbut did find GLUT1 expression in most of the cell types\nexamined with immunohistochemistry. These results in\ncombination suggest that GLUT1 is present in ectopic\nendometriotic lesions but it is not upregulated, at least\nat the genetic level, by the change to benign but\npathological ectopic endometriotic lesions.\nFigure 5\nGLUT expression in surrounding nerves. (A) Moderate GLUT1\nimmunoreactivity was observed in large nerve fiber trunks surrounding\nendometriotic lesions. (B) A higher magnification showed that immuno-\nreactivity was both nuclear and membranous (black arrows). (C) Weak to\nmoderate GLUT3 immunoreactivity was observed in large nerve fiber\ntrunks. (D) A higher magnification showed that immunoreactivity was\npredominantly nuclear (black arrows). (E) Moderate to strong GLUT4\nimmunoreactivity was observed in the large nerve fiber trunks. (F) A higher\nmagnification showed almost exclusive nuclear localization (black arrows).\n(G and H) Negative control showed limited background staining in either\nepithelial or stromal cells. (A, C, E and G) Scale bars represent 50 mM and the\nareas outlined in black represent the region of higher magnification. (B, D,\nF and H) Scale bars represent 10 mM.\nJournal of Molecular Endocrinology\nResearch B MCKINNON and others GLUT expression in\nendometriosis\n52:2 176\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access\n\n\nGLUT3 is a high-affinity transporter and is usually\nexpressed in tissue with high metabolic requirements,\nsuch as the brain ( Shepherd et al . 1992). It has also been\nidentified in normal endometrium; however, in these\nstudies it was confined to the CD45-positive cells ( von\nWolff et al . 2003). The immunohistochemistry results of\nthis study indicate that GLUT3 expression was also\nprimarily in the CD45-positive leukocytes that infiltrate\nthe endometriotic lesions, as well as the surrounding large\nnerve fiber trunks in the ectopic tissue. Increased numbers\nof immune cells have been detected in endometriotic\nlesions ( Halme et al . 1983 ), and nerve fibers are found\nproximal to lesions, particularly in the peritoneal cavity\nand the RVS (Mechsner et al. 2007, McKinnon et al. 2012).\nIt is hence possible that the increase inSLC2A3 expression\nobserved in this study is derived from the increased\ninfiltration of leukocytes and nerve fibers and therefore\nthat GLUT3 may not contribute substantially to the\nglucose requirements of the ectopic tissue.\nGLUT4 is an inducible, insulin-sensitive trans-\nporter predominantly expressed in skeletal muscle and\nadipose tissue. In response to insulin, GLUT4 trans-\nlocates from intracellular compartments to the plasma\nmembrane, resulting in a rapid increase in glucose uptake\n(Bryant et al . 2002). Once at the membrane, GLUT4 can\nincrease glucose uptake into the cell by 10- to 40-fold\n(Shepherd & Kahn 1999 ). The immunohistochemical\nand immunofluorescence images in this study confirm\nthat the GLUT4 protein is found in significant concen-\ntrations in the endometriotic epithelial and stromal cells,\nand in particular at the epithelial plasma membrane.\nThis expression and membrane localization indicates that\nGLUT4 has the potential to provide glucose to the lesions\nand thus may represent an important transporter for\nglucose entry into ectopic tissue.\nIn summary, the results of this study show that there\nis an altered expression of the SLC2A genes between\nwomen with and without endometriosis, as well as\nbetween the eutopic endometrial tissue of women with\nendometriosis and their matching ectopic lesions. More-\nover, we were able to show that there was a significant\nexpression of the GLUT4 transporter at the membrane of\nFigure 6\nGLUT4 membrane localization in endometriotic epithelial cells. (A, B, E and\nG) Endometriotic lesions were incubated with both GLUT4 and CK19\nprimary antibodies and fluorescent-conjugated secondary antibodies.\n(B, D, F and H) Negative controls were performed in the absence of primary\nantibodies. (A and B) Nuclei in both the positive and the negative controls\nwere visualized with DAPI. (C) Strong GLUT4 immunofluorescence (green)\nabove the background and autofluorescence observed in (D) the negative\ncontrol was observed predominantly in apical and basal regions of the\nglandular structures of the endometriotic lesions (white arrows).\n(E) CK19 immunofluorescence above the background and auto-\nfluorescence observed in (F) the negative control was also observed in the\nepithelial cell membranes (white arrows). (G) An overlay of the GLUT4 and\nCK19 immunofluorescence showed a significantly stronger signal of both\nGLUT4 and CK19 in the epithelial cell membranes above the background\nand autofluorescence observed in (H) the negative control. Scale bars\nrepresent 50 mM.\nJournal of Molecular Endocrinology\nResearch B MCKINNON and others GLUT expression in\nendometriosis\n52:2 177\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access\n\n\nendometriotic epithelial cells. The inducible nature of\nGLUT4 and its limited cellular expression may make\nGLUT4 an attractive target for non-hormone-based\npharmaceuticals.\nSupplementary data\nThis is linked to the online version of the paper at http://dx.doi.org/10.1530/\nJME-13-0194.\nDeclaration of interest\nThe authors declare that there is no conflict of interest that could be\nperceived as prejudicing the impartiality of the research reported.\nFunding\nThis research did not receive any specific grant from any funding agency in\nthe public, commercial or not-for-profit sector.\nAuthor contribution statement\nB M conceived the experimental design, performed experiments and\nprepared the manuscript. C W and D B contributed to experimental\nprocedures. N A B contributed to the collection of tissue and manuscript\nediting. J E contributed to tissue collection. M D M supervised the project\nand edited the manuscript.\nAcknowledgements\nThe authors would like to thank the theatre staff and Anne Vaucher in the\nlaboratory for tissue collection and the progesterone measurements.\nReferences\nBryant NJ, Govers R & James DE 2002 Regulated transport of the glucose\ntransporter GLUT4. Nature Reviews. Molecular Cell Biology 3 267–277.\n(doi:10.1038/nrm782)\nD’Hooghe TM, Debrock S, Hill JA & Meuleman C 2003 Endometriosis\nand subfertility: is the relationship resolved? Seminars in Reproductive\nMedicine 21 243–254. (doi:10.1055/s-2003-41330)\nEskenazi B & Warner ML 1997 Epidemiology of endometriosis. Obstetrics\nand Gynecology Clinics of North America 24 235–258. (doi:10.1016/\nS0889-8545(05)70302-8)\nEvans S, Moalem-Taylor G & Tracey DJ 2007 Pain and endometriosis.\nPain 132 (Suppl 1) S22–S25. ( doi:10.1016/j.pain.2007.07.006)\nFlorell SR, Coffin CM, Holden JA, Zimmermann JW, Gerwels JW,\nSummers BK, Jones DA & Leachman SA 2001 Preservation of RNA for\nfunctional genomic studies: a multidisciplinary tumor bank protocol.\nModern Pathology 14 116–128. (doi:10.1038/modpathol.3880267)\nFrolova AI & Moley KH 2011a Glucose transporters in the uterus: an\nanalysis of tissue distribution and proposed physiological roles.\nReproduction 142 211–220. (doi:10.1530/REP-11-0114)\nFrolova AI & Moley KH 2011b Quantitative analysis of glucose transporter\nmRNAs in endometrial stromal cells reveals critical role of GLUT1 in\nuterine receptivity. Endocrinology 152 2123–2128. (doi:10.1210/\nen.2010-1266)\nGoldman NA, Katz EB, Glenn AS, Weldon RH, Jones JG, Lynch U,\nFezzari MJ, Runowicz CD, Goldberg GL & Charron MJ 2006 GLUT1\nand GLUT8 in endometrium and endometrial adenocarcinoma.\nModern Pathology 19 1429–1436. ( doi:10.1038/modpathol.3800656)\nHalme J, Becker S, Hammond MG, Raj MH & Raj S 1983 Increased\nactivation of pelvic macrophages in infertile women with mild\nendometriosis. American Journal of Obstetrics and Gynecology 145\n333–337.\nHalme J, Hammond MG, Hulka JF, Raj SG & Talbert LM 1984 Retrograde\nmenstruation in healthy women and in patients with endometriosis.\nObstetrics and Gynecology 64 151–154.\nKato M, Yamamoto S, Takano M, Matsubara O & Furuya K 2012\nAberrant expression of the mammalian target of rapamycin,\nhypoxia-inducible factor-1a, and glucose transporter 1 in the\ndevelopment of ovarian clear-cell adenocarcinoma. International\nJournal of Gynecological Pathology 31 254–63. (doi:10.1097/PGP.\n0b013e318237d66c)\nKorgun ET, Celik-Ozenci C, Seval Y, Desoye G & Demir R 2005 Do glucose\ntransporters have other roles in addition to placental glucose transport\nduring early pregnancy? Histochemistry and Cell Biology 123 621–629.\n(doi:10.1007/s00418-005-0792-3)\nMcKinnon B, Bersinger NA, Wotzkow C & Mueller MD 2012 Endo-\nmetriosis-associated nerve fibers, peritoneal fluid cytokine concen-\ntrations, and pain in endometriotic lesions from different locations.\nFertility and Sterility 97 373–380. (doi:10.1016/j.fertnstert.2011.11.011)\nMechsner S, Schwarz J, Thode J, Loddenkemper C, Salomon DS & Ebert AD\n2007 Growth-associated protein 43-positive sensory nerve fibers\naccompanied by immature vessels are located in or near peritoneal\nendometriotic lesions. Fertility and Sterility 88 581–587. (doi:10.1016/\nj.fertnstert.2006.12.087)\nMioni R, Chiarelli S, Xamin N, Zuliani L, Granzotto M, Mozzanega B,\nMaffei P, Martini C, Blandamura S, Sicolo N et al. 2004 Evidence for\nthe presence of glucose transporter 4 in the endometrium and its\nregulation in polycystic ovary syndrome patients. Journal of\nClinical Endocrinology and Metabolism 89 4089–4096. (doi:10.1210/\njc.2003-032028)\nMozzanega B, Mioni R, Granzotto M, Chiarelli S, Xamin N, Zuliani L,\nSicolo N, Marchesoni D & Vettor R 2004 Obesity reduces the\nexpression of GLUT4 in the endometrium of normoinsulinemic\nwomen affected by the polycystic ovary syndrome. Annals of the New\nYork Academy of Sciences 1034 364–374. (doi:10.1196/annals.1335.038)\nMueckler M, Caruso C, Baldwin SA, Panico M, Blench I, Morris HR,\nAllard WJ, Lienhard GE & Lodish HF 1985 Sequence and structure of\na human glucose transporter. Science 229 941–945. (doi:10.1126/\nscience.3839598)\nNirgianakis K, Bersinger NA, McKinnon B, Kostov P, Imboden S &\nMueller MD 2013 Regression of the inflammatory microenvironment\nof the peritoneal cavity in women with endometriosis by GnRHa\ntreatment. European Journal of Obstetrics, Gynecology, and Reproductive\nBiology 170 550–554. (doi:10.1016/j.ejogrb.2013.08.010)\nNisolle M & Donnez J 1997 Peritoneal endometriosis, ovarian endo-\nmetriosis, and adenomyotic nodules of the rectovaginal septum are\nthree different entities. Fertility and Sterility 68 585–596. (doi:10.1016/\nS0015-0282(97)00191-X)\nOta H, Igarashi S, Hayakawa M, Matsui T, Tanaka H & Tanaka T 1996\nEffect of danazol on the immunocompetent cells in the eutopic\nendometrium in patients with endometriosis: a multicenter\ncooperative study. Fertility and Sterility 65 545–551.\nRuijter JM, Ramakers C, Hoogaars WM, Karlen Y, Bakker O, van den Hoff MJ\n& Moorman AF 2009 Amplification efficiency: linking baseline and bias\nin the analysis of quantitative PCR data. Nucleic Acids Research 37 e45.\n(doi:10.1093/nar/gkp045)\nSampson JA 1927 Peritoneal endometriosis due to the menstrual\ndissemination of endometrial tissue into the peritoneal cavity.\nAmerican Journal of Obstetrics and Gynecology 14 422–469.\nScheepers A, Joost HG & Schurmann A 2004 The glucose transporter\nfamilies SGLT and GLUT: molecular basis of normal and aberrant\nJournal of Molecular Endocrinology\nResearch B MCKINNON and others GLUT expression in\nendometriosis\n52:2 178\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access\n\n\nfunction. Journal of Parenteral and Enteral Nutrition 28 364–371. (doi:10.\n1177/0148607104028005364)\nShepherd PR & Kahn BB 1999 Glucose transporters and insulin action –\nimplications for insulin resistance and diabetes mellitus. New England\nJournal of Medicine 341 248–257. (doi:10.1056/\nNEJM199907223410406)\nShepherd PR, Gould GW, Colville CA, McCoid SC, Gibbs EM &\nKahn BB 1992 Distribution of GLUT3 glucose transporter protein\nin human tissues. Biochemical and Biophysical Research\nCommunications 188 149–154. (doi:10.1016/0006-291X(92)\n92362-2)\nSzablewski L 2013 Expression of glucose transporters in cancers.Biochimica\net Biophysica Acta 1835 164–169. (doi:10.1016/j.bbcan.2012)\nUldry M & Thorens B 2004 The SLC2 family of facilitated hexose and polyol\ntransporters. Pflu¨gers Archiv: European Journal of Physiology 447 480–489.\n(doi:10.1007/s00424-003-1085-0)\nVandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A &\nSpeleman F 2002 Accurate normalization of real-time quantitative\nRT-PCR data by geometric averaging of multiple internal control genes.\nGenome Biology 3. RESEARCH0034. (doi:10.1186/gb-2002-3-7-\nresearch0034)\nvon Wolff M, Ursel S, Hahn U, Steldinger R & Strowitzki T 2003 Glucose\ntransporter proteins (GLUT) in human endometrium: expression,\nregulation, and function throughout the menstrual cycle and in\nearly pregnancy. Journal of Clinical Endocrinology and Metabolism 88\n3885–3892. (doi:10.1210/jc.2002-021890)\nReceived in final form 16 December 2013\nAccepted 8 January 2014\nAccepted Preprint published online 10 January 2014\nJournal of Molecular Endocrinology\nResearch B MCKINNON and others GLUT expression in\nendometriosis\n52:2 179\nhttp://jme.endocrinology-journals.org /C2092014 Society for Endocrinology\nDOI: 10.1530/JME-13-0194 Printed in Great Britain\nPublished by Bioscientifica Ltd.\nDownloaded from Bioscientifica.com at 06/13/2026 01:26:08PM\nvia free access","source_license":"CC0","license_restricted":false}