{"paper_id":"508276bc-4a4c-4d91-be64-23e13505b46b","body_text":"DOI: 10.1530/JOE-17-0236\nhttp://joe.endocrinology-journals.org © 2017 Society for Endocrinology\nPrinted in Great Britain\nPublished by Bioscientifica Ltd.\nJournal of Endocrinology\n269–278s vallcaneras  and others TNFRp55 deficiency and \nendometriosisResearch\n234:3\n10.1530/JOE-17-0236\nTNFRp55 deficiency promotes the \ndevelopment of ectopic \nendometriotic-like lesions in mice\nSandra Vallcaneras 1, Federica Ghersa 1, Juan Bastón 2, María Belén Delsouc 1, \nGabriela Meresman 2 and Marilina Casais 1\n1Laboratorio de Biología de la Reproducción (LABIR), Facultad de Química, Bioquímica y Farmacia, \nUniversidad Nacional de San Luis, Instituto Multidisciplinario de Investigaciones Biológicas de San Luis \n(IMIBIO-SL-CONICET), San Luis, Argentina\n2Laboratorio de Fisiopatología Endometrial, Instituto de Biología y Medicina Experimental (IBYME-CONICET), \nBuenos Aires, Argentina\nAbstract\nEndometriosis is an inflammatory disease depending on estradiol, with TNF- α being \none of the most representative cytokines involved in its pathogenesis. TNF- α acts \nthrough its bond to the TNFRp55 and TNFRp75 membrane receptors. The aim of this \nstudy was to analyze the effect of the TNFRp55 deficiency on the development of \nectopic endometriotic-like lesions. Endometriosis was induced surgically in mice of the \nC57BL/6 strain, wild type (WT) and TNFRp55−/− (KO). After four weeks, the peritoneal \nfluid was collected and the lesions were counted, measured with a caliper, removed, \nweighed, fixed or kept at −80°C. We evaluated the cell proliferation by proliferating \ncell nuclear antigen (PCNA) immunohistochemistry and apoptosis by TUNEL technique \nin the ectopic lesions. MMP-2 and MMP-9 activities (factors involved in invasiveness) \nwere measured by zymography in the peritoneal fluid; estradiol and progesterone levels \nwere measured by radioimmunoassay in the lesions and in the peritoneal fluid. We \nfound that in KO animals the mean number of lesions established per mouse, the lesion \nvolume, weight and cell proliferation increased and apoptosis decreased. In addition, \nthe activity of MMP-2 and the estradiol level increased, whereas the progesterone level \nwas not significantly modified. In conclusion, the deficiency of TNFRp55 promoted the \nestablishment and development of endometriosis through an increase in the lesion size \nand high levels of estradiol which correlate with an increase in the MMP-2 activity. This \nis evidence of the possible association of the deregulation of the TNFRp55 expression \nand the survival of the endometriotic tissue in ectopic sites.\nIntroduction\nEndometriosis is a chronic disease, depending on \nestrogens, that affects women at reproductive age, \ncausing inflammation, intense pelvic pain and reduced \nfertility. It is characterized by the implantation and \ngrowth of endometrial tissue outside the uterine cavity \n(Greene  et al . 2016). Sampson’s theory, which states that \nepithelial and stromal cells reach the peritoneal cavity \nthrough late menstruation through the fallopian tubes \nand are then spread and implanted in the peritoneal \ncavity, is the mostly accepted mechanism to explain \n234\n3\nCorrespondence \nshould be addressed \nto M Casais \nEmail \nmcasais@unsl.du.ar\nKey Words\n f endometriosis\n f TNFRp55 deficiency\n f estradiol\n f mouse model\nJournal of Endocrinology  \n(2017) 234, 269–278\nDownloaded from Bioscientifica.com at 06/28/2026 10:39:13PM\nvia free access\n\n\nResearch 270\nTNFRp55 deficiency and \nendometriosis\nDOI: 10.1530/JOE-17-0236\nJournal of Endocrinology\ns vallcaneras  and others\nhttp://joe.endocrinology-journals.org © 2017 Society for Endocrinology\nPrinted in Great Britain\nPublished by Bioscientifica Ltd.\n234:3\nits etiology ( Sampson 1927 ). However, this theory does \nnot explain why more than 90% of women with late \nmenstrual bleeding do not develop endometriosis, which \nsuggests that there are genetic, immunological and/or \nbiochemical factors that contribute to the development \nof the disease (Ahn  et al . 2015). A significant component \nin the pathogenesis of endometriosis is the loss of normal \npatterns of communication between the endocrine and \nimmune systems. It is known that endometrial cells \nreduce their capacity of death by apoptosis and increase \ntheir capacity of invasiveness associated to hormonal \nalterations. These changes promote the resistance of \nthe endometrial cells to the normal cleaning of the \nperitoneum by the immune system, responsible for the \nregulation of tissue homeostasis ( Kitawaki  et  al.  2002 , \nAntsiferova & Sotnikova 2012).\nTumor necrosis factor-alpha (TNF-α), has an important \nfunction in the physiology of the proliferation and \ndesquamation of the endometrium during the menstrual \ncycle. In addition, it plays a crucial role in inflammation, \nangiogenesis, cell proliferation and cell death. TNF- α acts \non the target cells via two receptors: TNFRp55 (type 1) and \nTNFRp75 (type 2). TNFRp55 is characterized by a death \ndomain in its intracellular region, whereas type 2 receptor \nlacks such domain. Thus, the activation of TNFRp55 leads \nprimarily to pro-inflammatory and programmed cell death \npathways (Rojas-Cartagena  et al . 2005). In endometriosis, \nthere is evidence that indicates an aberrant function of the \nTNF system. For example, levels of TNFRp55 expression \nin the endometrium of women with endometriosis \nduring the late secretory phase, stage related to the \napoptosis, were lower in relation to the endometrium \nwithout endometriosis, in coincidence with the almost \nabsent apoptosis in the eutopic and ectopic endometrium \n(Boric  et al.  2013). In addition, a recent study that examined \nthe serum levels of the soluble forms of both TNFRp55 and \nTNFRp75, has found that TNFRp55 is significantly higher \nin the serum of endometriosis patients than in that of \ncontrols, and given that these soluble forms can modulate \nthe effects of TNF by acting as antagonists (Othman  et al.  \n2016), this may also be involved in the altered cell death \nobserved in this pathology.\nOn the other hand, the metalloproteinases (MMPs) are \nessential factors in the processes of invasiveness and tissue \nremodeling, which are secreted as latent pro-enzymes and \nactivated by proteolytic cleavage ( Di Carlo  et  al.  2009 ). \nGelatin is an important protein of the extracellular matrix \nand it is sensitive to a high range of tissue proteinases, \nincluding MMP-2 and MMP-9 gelatinases (Jana  et al . 2016,  \nPan  et  al.  2016 ). These enzymes participate in the \nphysiological cyclic changes of the endometrium, and \nseveral studies have shown that their steroid regulation \nis critical for the formation of the fragments of the \nendometrial tissue in ectopic sites ( Bruner  et  al.  1997 , \nPitsos & Kanakas 2009 ). In relation to this, there exists \na correlation between the levels of MMP-2 and estradiol \nin the peritoneal fluid of patients with endometriosis \n(Huang  et al . 2004).\nIn turn, steroid hormones play an important role \nin the development of this pathology. Endometriosis \nseems to be associated to a decrease in the response \ncapacity of the endometrial stromal cells to progesterone, \nwhich may be due to a decrease in the expression of the \nprogesterone receptors (PR). A recent study demonstrates \nthat inflammatory cytokines like TNF α reduce the \nexpression of PR. Thus, this effect may contribute to the \nprogesterone resistance of women with endometriosis, \nwhich in turn is associated with hyperactive action of \nestradiol (Bulun  et  al. 2006, Grandi  et  al. 2016, Li  et  al. \n2016). The proinflammatory and antiapoptotic effects of \nestradiol in endometrial cells appear to be exacerbated in \nwomen with endometriosis (Reis  et al.  2013).\nIn view of all the above, the aim of this study was \nto analyze the effect of the TNFRp55 deficiency on \nthe establishment and growth of endometriotic-like \nlesions in an induced endometriosis model in mouse. In \naddition, we evaluated cell proliferation and apoptosis of \nendometriotic cells as well as factors related to invasiveness \nand endocrine status.\nMaterials and methods\nAnimals\nFemale mice of the C57BL/6 strain, WT and TNFRp55−/− \n(KO) of two months, weighing 19–21 g were used. The \nTNFRp55−/− mice were obtained from the Max von \nPettenkofer-Institute, Munich, Germany. Breeding \ncolonies were established at the Animal Facility of the \nNational University of San Luis (San Luis, Argentina) \nunder rigorous light conditions (12 h light, 07:00–19:00, \nand 12 h darkness), controlled temperature (22 ± 2°C), \nwith water and sterile food ad libitum. The experiments \nwere carried out according to the guidelines for the care \nand use of laboratory animals of the National Institutes of \nHealth (NIH) and the Comité Institucional de Cuidado y Uso \nde Animales de Experimentación  (CICUA) of the National \nUniversity of San Luis, Argentina.\nDownloaded from Bioscientifica.com at 06/28/2026 10:39:13PM\nvia free access\n\n\n271Research\ns vallcaneras  and others TNFRp55 deficiency and \nendometriosis\nDOI: 10.1530/JOE-17-0236\nJournal of Endocrinology\nhttp://joe.endocrinology-journals.org © 2017 Society for Endocrinology\nPrinted in Great Britain\nPublished by Bioscientifica Ltd.\n234:3\nSurgical induction of endometriosis\nThe endometriotic-like lesions were induced \nexperimentally, as reported previously (Bilotas  et al.  2010). \nThe animals were anesthetized via intraperitoneal with \n100 mg/kg of ketamine (Holliday Scott, Buenos Aires, \nArgentina) and 10 mg/kg of xylazine (Richmond, Buenos \nAires, Argentina), and a mid-ventral incision was made to \nexpose the bowels. The right uterine horn is removed from \nthe animal, placed in DMEM-F12 (Gibco) and divided \nlongitudinally. It is later cut in three square pieces of \napproximately 4 mm2, which are sutured to the intestine \nmesentery with only one stitch (supralong 6-0, Ethicon, \nNJ, USA). The area is hydrated with sterile physiological \nsolution supplemented with antibiotic–antimycotic before \nclosing the abdominal wall with the same suture material, \nwith continuous stitches. The mice are monitored daily \nin relation to body weight, food consumption, preening \nbehavior and activity. The mice were killed by cervix \ndislocation after four weeks. Then, a small medioventral \nhole was opened through which 1.5 mL of PBS was injected \nin the peritoneal cavity of each animal, and the peritoneal \nfluid was collected and centrifuged at 10,000 g for 1 min. \nThe supernatant was separated from the precipitate and \nkept at −80°C until the corresponding determinations. \nAfter that, the abdomen was completely opened to have \naccess to the endometriotic-like lesions.\nEvaluation of the ectopic uterine tissue\nThe lesions were identified, counted and measured with \ncaliper in two perpendicular diameters. The volume of \nthe developed lesions was calculated with the following \nequation: V = (4/3) π r12 r2 (r1 and r2 are the radiuses and \nr1 < r2). The lesions were removed, weighed and kept \nin as follows: one lesion per animal was fixed in buffer \nformaldehyde at 4% for 24 h at 4°C. The fixed specimens \nare embedded in paraffin, cut in 5 μm sections and \nstained with hematoxylin–eosin in order to examine \nmicroscopically the presence of histological identity \nsignals of endometriosis (glands and stroma), or prepared \nfor immunohistochemical technique and/or TUNEL. The \nremaining lesions were kept at −80°C for the rest of the \ndeterminations.\nImmunohistochemistry\nProliferating cell nuclear antigen (PCNA), also called \ncyclin, is a 36-KD auxiliary protein of DNA polymerase \ndelta that has been found to be a useful marker in \nimmunocytochemical studies of cell proliferation because \nits expression correlates with the proliferative state of the \ncell ( Bravo  et al . 1987 ). All sections were deparaffinized \nin xylene and rehydrated through graded alcohols. \nEndogenous peroxidase was blocked by treatment \nwith 3% H 2O2 for 30 min, followed by microwaving in \n0.01 M sodium citrate buffer for antigen retrieval. All \nthe sections were then blocked with 4% BSA in PBS for \n2 h at room temperature and incubated overnight at 4°C \nwith the anti-mouse PCNA rabbit polyclonal antibody \n(1:200, FL-261, Santa Cruz Biotechnology) in PBS with \n1% BSA at room temperature. After that, sections were \nincubated for 1 h at room temperature with 1:200 goat \nbiotinylated anti-rabbit IgG antibody (Sigma-Aldrich) \nfollowed by incubation with a streptavidin–peroxidise \nconjugate (VectorLabs, Burlingame, CA, USA) for 30 min \nat room temperature. The signal was developed with \ndiaminobenzidine (DAB) as substrate (Cell Marque, CA, \nUSA), and finally, the sections were counterstaining with \nGill’s hematoxylin, dehydrated through grades alcohols, \nclarified in xylene and properly mounted. As a negative \ncontrol, one section of each slide was assayed without the \nprimary antibody. PCNA-positive cells were identified by \nthe presence of brown nuclear reactivity. The percentage \nof PCNA-positive cells was established using a standard \nlight microscope at 400×. Cell proliferation was quantified \nby counting a minimum of 100–150 randomly selected \nepithelial and stromal cells per field and independently \nof the number of cells, 4–6 random fields per section \nwere counted. The percentage of proliferating cells was \ncalculated on the total, and these percentages were then \nused to obtain the mean of each experimental group.\nTUNEL assay\nFragmented DNA of apoptotic cells was detected using In \nSitu Cell Death Detection Kit POD TUNEL assay (Cat No. \n11684817910 Roche), according to the manufacturer’s \ninstructions. The detection was achieved using the \nperoxidase substrate, hydrogen peroxide and the stable \nchromogen DAB. Using this procedure, apoptotic \nnuclei are stained dark brown. Finally, sections were \ncounterstained with hematoxylin, mounted and analyzed \nin a light microscope. The number of apoptotic nuclei \nrelative to total cells was determined by counting 150 \nrandomly selected epithelial and stromal cells per field, \nand 5 random fields per section were counted. The \npercentage of apoptotic cells was calculated on the total, \nand these percentages were then used to obtain the mean \nof each experimental group.\nDownloaded from Bioscientifica.com at 06/28/2026 10:39:13PM\nvia free access\n\n\nResearch 272\nTNFRp55 deficiency and \nendometriosis\nDOI: 10.1530/JOE-17-0236\nJournal of Endocrinology\ns vallcaneras  and others\nhttp://joe.endocrinology-journals.org © 2017 Society for Endocrinology\nPrinted in Great Britain\nPublished by Bioscientifica Ltd.\n234:3\nGelatin zymography\nMMP-2 and MMP-9 enzymatic activities were determined \nby SDS-PAGE gelatin zymography. The samples, normalized \nin the same quantity of proteins, were mixed with the \nsame volume of the sample buffer (0.3 M Tris–HCl pH 6.8, \n2% SDS, 40% glycerol and 0.1% bromophenol blue) and \nincubated for approximately 30 min at 37°C. Then they \nwere subjected to electrophoresis in polyacrylamide gels \nat 10% (SDS-PAGE) containing 0.2% of gelatin (Merck), \nunder non-reducing denaturalizing conditions at 4°C. \nOnce electrophoresis was completed, the gels were washed \nwith 2.5% of TritonX-100 (v/v) in buffer TNC (50 mM Tris–\nHCl pH 7.5, 0.5 M NaCl, 10 mM CaCl2 and 0.02% NaN3), \ntwice for 15 min to remove the SDS and thus allow the \nproteinases to recover their activity. After washing, the \ngels were incubated in TNC buffer with 1% TritónX-100 \nfor 24 h at 37°C. Finally, the coloration with a 0.5% of \nCoomassie brilliant blue R-250 solution was carried out, \nwith the subsequent discoloration. The MMPs activity was \nevaluated observing the lysis zones. The bands intensity \nwas determined by densitometry using ImageJ software \nand was expressed in arbitrary units.\nRadioimmunoassay (RIA)\nSamples of peritoneal fluid supernatant and homogenates \nof endometriotic lesions were used to measure progesterone \nand estradiol levels using a RIA kit (Beckman Coulter and \nDIAsource, respectively, Diagnos Med SRL, Buenos Aires, \nArgentina) following the manufacturer’s instructions. \nThe assays sensitivity was 0.05 ng progesterone/mL and \n<2.7 pg estradiol/ml. The inter- and intra-assay coefficients \nof variation in all the assays were <10.0%.\nWestern blot\nProtein extracts were obtained using TRIzol reagent, \nfollowing the manufacturer’s indications (Invitrogen Life \nTechnologies). Protein concentration was determined by \nLowry method. Aliquots containing 40 μg of total protein \nwere subjected to electrophoresis in 10% SDS-PAGE gels \nand then electrotransferred to PVDF membrane (Millipore \nCorporation) at 100 V for 1 h in a transfer buffer (25 mM \nTris, 192 mM glycine and 20% v/v methanol, pH 8.3). The \nmembrane was immersed in 5% non-fat milk in a PBST \nsolution (KH2PO4 0.015 M, NaH2PO4 0.017 M, KCl 0.076 M, \nNaCl 0.14 M (pH 7.4), 0.5% Tween 20) for 1 h at room \ntemperature, followed by an overnight incubation at 4°C \nwith either rabbit anti-P450aromatase (SC-30086) or goat \nanti ACTB (SC-1615, Santa Cruz Biotechnology), 1:1000 \ndilution in 1% solution of non-fat powdered milk in PBST. \nAfter incubation with primary antibody, membranes \nwere washed in PBST and incubated with donkey anti-\ngoat IgG peroxidase-linked antibody (Cat sc-2020, Santa \nCruz Biotechnology) 1:5000 dilution in 1% milk for 1 h \nat room temperature and goat anti-rabbit IgG peroxidase-\nlinked antibody (Cat: sc-2004, Santa Cruz Biotechnology) \n1:5000 dilution in 1% milk for 3 h at room temperature, \nrespectively. Following washing in PBST, blots were \ndeveloped using an enhanced chemiluminescence Western \nblotting detection system, Thermo Scientific SuperSignal \nWest Pico Chemiluminescent (Pierce Biotechnology) and \nexposed to X-ray films, Thermo Scientific CL-XPosure \nFilm (Pierce Biotechnology). The mean of intensity of each \nband was measured using the NIH ImageJ software (Image \nProcessing and Analysis in Java fromhttp://rsb.info.nih.\ngov/ij/). P450aromatase (P450AROM) protein levels were \nnormalized against ACTB (endogenous control).\nRNA isolation and RT-PCR analysis\nTotal RNA was isolated from endometriotic lesions using \nTRIzol reagent (Invitrogen Life Technologies), according to \nthe manufacturer’s instructions. Purified total RNAs were \nthen quantified and assessed for purity by measurement \nof the 260/280 ratio using an UV spectrophotometer \nBeckman DU-640 B (CA, USA). Only samples with \n260/280 ratio of 1.8–2.0 were used. The integrity was \nconfirmed by running 2 µg RNA on a 0.8% agarose gel. \nAfter GelRed (Biotium, Hayward, CA, USA) staining, \nRNA bands were visualized with a UV transilluminator, \nand 28S and 18S rRNA band patterns were analyzed. \nTwo micrograms of total RNA were reverse transcribed \nat 37°C using random primers and M-MLV Reverse \nTranscriptase (Promega) in a 26 µL reaction mixture. For \namplification of the reverse transcription (RT) products, \nthe reaction mixture consisted of 1 × Green GoTaq \nreaction buffer, 0.2 mM deoxynucleoside triphosphates, \n0.5 µM specific oligonucleotide primers and 1.25 U Go Taq \nDNA polymerase (Promega) in a final volume of 50 µL. \nThe PCR primers were designed using Primer Express 3.0 \nsoftware (Applied Biosystems). The primers information \nis shown in Table  1. The cDNA was amplified using a \nthermal cycler (My Cycler, BioRad). Reaction products \nwere electrophoresed on 2% agarose gels, visualized with \nGelRed and examined by ultraviolet transillumination. \nBand intensities of RT-PCR products were quantified \nusing ImageJ (Image Processing and Analysis in Java from \nDownloaded from Bioscientifica.com at 06/28/2026 10:39:13PM\nvia free access\n\n\n273Research\ns vallcaneras  and others TNFRp55 deficiency and \nendometriosis\nDOI: 10.1530/JOE-17-0236\nJournal of Endocrinology\nhttp://joe.endocrinology-journals.org © 2017 Society for Endocrinology\nPrinted in Great Britain\nPublished by Bioscientifica Ltd.\n234:3\nhttp://rsb.info.nih.gov/ij/). Relative levels of mRNA were \nexpressed as the ratio of signal intensity for the target \ngenes relative to that for the housekeeping gene Actb.\nStatistical analysis\nStatistical analysis was performed using GraphPad Prism \n(Version 5, GraphPad Software). Values are presented \nas the mean ± s.e.m . Differences between the means of \neach group were analyzed using the Student’s t-test. \nPearson’s correlation coefficient was used to evaluate the \nrelationship between estradiol levels and MMP-2 activity \nin peritoneal fluid. Differences were considered to be \nstatistically significant when P < 0.05.\nResults\nEffect of the TNFRp55 deficiency on the endometriotic-\nlike lesions establishment and growth\nThe macroscopic evaluation of the ectopic uterine tissue \nrevealed that the number of established lesions increased \nin animals with receptor deficiency (P < 0.05) (Fig. 1A ). In \naddition, we observed that the lesions volume and weight \nwere higher in KO animals than in WT animals ( P < 0.05 \nand P < 0.001, respectively) (Fig. 1B  and C). These results \nshow the relevance of TNFRp55 in the development of the \nectopic endometrial tissue.\nEffect of the TNFRp55 deficiency on the cell proliferation \nand apoptosis in endometriotic-like lesions\nThe TNFRp55 deficiency caused an increase in the \npercentage of cell proliferation ( P < 0.01) (Fig. 2A ) and a \ndecrease in the percentage of apoptosis cells ( P < 0.001) \n(Fig.  2B ). There were no significant differences between \nepithelial and stromal cell. These results suggest that \nthe increase in the volume and weight of the lesions \nin the KO animals might be associated to the low  \nelimination rate and high proliferation rate of the \nendometriotic cells.\nEffect of TNFRp55 deficiency on the factors  \nregulating invasiveness\nWe analyzed the enzymatic activity of MMP-2 and MMP-9 \nin homogenates of endometriotic lesions and peritoneal \nfluid, and although no significant changes were observed \nin the lesion in both experimental groups ( Fig.  3A ), in \nperitoneal fluid, the activity of MMP-2 ( P < 0.001) and \nits zymogene increased significantly in the KO animals \n(P < 0.01) (Fig. 3B ). These results suggest that the TNFRp55 \ndeficiency modifies the peritoneal environment, which \ncontributes to the remodeling and establishment of \nlesions.\nEffect of the TNFRp55 deficiency on the steroid \nhormones levels\nIn the KO animals, the levels of estradiol increased in the \nperitoneal fluid and in lesion ( P < 0.05) (Fig.  4A and B), \nwhereas the expression, analyzed in ectopic uterine tissue, \n(mRNA and protein) of P450arom, synthesis enzyme, \nwas not significantly modified ( Fig.  4C  and D). These \nresults indicate the close relationship between estradiol \nand TNFRp55. Pearson’s test was applied to determine \nthe possible correlation between estradiol levels and \nMMP-2 activity in peritoneal fluid samples. For the entire  \ngroup of mice ( n = 16), positive correlation was found \n(coefficient (r) Pearson: 0.67;  P < 0.01). While, there was \nno correlation in the wild-type mice ( n = 8) and deficient \nmice (n = 8) (Fig. 5 ).\nTable 1 Primer used for semi-quantitative RT-PCR amplification.\nGen Sequences (sense above, antisense below; 5′–3′) GenBank accession # Amplicon length (pb) No. of cycles\nHsd3b1 GTCTTCAGACCAGAAACCAAG M58567 213 35\n CCTTAAGGCACAAGTATGCAG    \nAkr1c18 TTCGAGCAGAACTCATGGCTA NM_134066 141 35\n CAACCAGGTAGAATGCCATCT    \nPgr AB CTGTGCCTTACCATGTGGCA NM_008829.2 389 35\n TTCACCATGCCCGCCAGGAT    \nPgr B GGTCCCCCTTGCTTGCA NM_008829 121 35\n CAGGACCGAGGAAAAAGCAG    \nP450arom CCCGAGCCTTTGGAGAACAA NM_007810.3 161 40\n TGAGGGTCAACACATCCACG    \nActb CGGAACCGCTCATTGCC NM_007393.5 289 35\n ACCCACACTGTGCCCATCTA    \nDownloaded from Bioscientifica.com at 06/28/2026 10:39:13PM\nvia free access\n\n\nResearch 274\nTNFRp55 deficiency and \nendometriosis\nDOI: 10.1530/JOE-17-0236\nJournal of Endocrinology\ns vallcaneras  and others\nhttp://joe.endocrinology-journals.org © 2017 Society for Endocrinology\nPrinted in Great Britain\nPublished by Bioscientifica Ltd.\n234:3\nWe also analyzed the effect of TNFRp55 deficiency \non the progesterone levels in peritoneal fluid and \nhomogenates of endometriotic lesions, the expression \nof the metabolic enzymes and PR. In KO animals, the \nlevels of progesterone were not significantly modified \n(Fig.  6A  and B), whereas the expression (mRNA) of  \nHsd3b1, synthesis enzyme, and of Akr1c18, degradation \nenzyme, decreased in endometriotic tissue ( P < 0.01 and \nP < 0.005, respectively) ( Fig. 6C ); therefore, the hormone \nlevels were not modified. To determine if the TNFRp55 \ndeficiency modifies the action sites of progesterone, we \nanalyzed the expression (mRNA) of the progesterone \nreceptors Pgr A and Pgr B, but no changes were observed \n(Fig. 6D ).\nDiscussion\nExtensive cell proliferation, tissue remodeling and aberrant \napoptosis occur at the ectopic sites where endometrial \ntissue deposits develop into endometriotic lesions. In \naddition, there is evidence of an aberrant function of the \nTNF system in this pathology. In the endometrium of \ncontrol women, the highest levels of TNF and TNFRp55 \nare produced during the late secretory phase. This phase is \nrelated to the apoptosis, which is the physiological process \ninvolved in the menstrual shedding in a non-conceptional \ncycle. In contrast, the levels of TNFRp55 expression in the \nendometrium of women with endometriosis during the late \nsecretory phase are lower in relation to the endometrium \nFigure 1\nEstablishment and growth of endometriotic \nlesions. The number of established endometriotic \nlesions (A), volume (B) and weight (C) were \nassessed in mice wild type (WT) n = 12 year \nTNFRp55−/− (KO) n = 11 after 4 week of induced \nendometriosis. Statistical comparisons were \nperformed by Student’s t-test. *P < 0.05, \n■ P < 0.001.\nFigure 2\nCell proliferation and apoptosis in endometriotic \nlesions. (A) The percentage of proliferating cells \nwas assessed by immunohistochemistry for PCNA \nin endometriotic lesions. Micrographs show \nrepresentative histological sections of \nendometriotic lesions of WT (n = 11) (i) and KO \n(n = 7) (ii). As a negative control, one section of \neach slide was assayed without the primary \nantibody (iii). Statistical comparisons were \nperformed by Student’s t-test. ● P < 0.01. (B) The \npercentage of apoptosis cells was assessed by \nTUNEL in endometriotic lesions. Micrographs \nshow representative histological sections of \nendometriotic lesions of WT (n = 8) (iv) and KO \n(n = 8) (v). As a negative control, one section of \neach slide was assayed without the primary \nantibody (vi). Statistical comparisons were \nperformed by Student’s t-test. ■ P < 0.001.\nDownloaded from Bioscientifica.com at 06/28/2026 10:39:13PM\nvia free access\n\n\n275Research\ns vallcaneras  and others TNFRp55 deficiency and \nendometriosis\nDOI: 10.1530/JOE-17-0236\nJournal of Endocrinology\nhttp://joe.endocrinology-journals.org © 2017 Society for Endocrinology\nPrinted in Great Britain\nPublished by Bioscientifica Ltd.\n234:3\nof women without endometriosis. This is in coincidence \nwith the almost absent apoptosis in the eutopic and \nectopic endometrium, which might favor the survival and \ngrowth of the menstrual debris outside the uterus ( Rojas-\nCartagena  et al . 2005, Boric  et al . 2013). Considering that \nthe abnormal survival of endometrial cells may result \nin their continuing growth into ectopic locations and \nthat TNFRp55 plays a role in triggering apoptosis, in this \nstudy, we selected TNFRp55 as molecular target and we \ninvestigated the effect of TNFRp55 deficiency on the \ndevelopment of endometriotic-like lesion in a murine \nmodel. One of the major limitations of mouse model is \nthe lack of menstruation and subsequent spontaneous \nendometriosis. However, this model has many similarities \nto the disease in humans, such as the growth of the ectopic \nendometrial tissue that has been shown to be estrogen \ndependent. In addition, it is a versatile model that has \nbeen used to investigate the mechanisms involved in \nthe peritoneal attachment of endometrial cells, how \nthe immune system and hormones affect endometriosis \nas well as the effects of drugs and therapeutic products \n(Grümmer 2006).\nFigure 3\nEnzyme activities of MMP-2 and MMP-9. MMP-2 and MMP-9 (proenzyme and active forms) enzymatic activities were determined by SDS-PAGE gelatin \nzymography in endometriotic lesions (A) and peritoneal fluid (B). The gel photographs were quantified using ImageJ and expressed as in relative units. \nResults are expressed as mean ± s.e.m . of eight animals per experimental group. Student’s t-test was used. ● P < 0.01, ■ P < 0.001.\nFigure 4\nEstradiol levels and P450aromatase expression. \nEstradiol levels were measured by RIA in \nperitoneal fluid (A) and endometriotic lesions (B). \nMeasurement by RT-PCR of expression (mRNA) of \nP450arom and Actb as housekeeping gene (C). \nMeasurement by Western blot of expression \n(protein) of P450arom and Actb as load control \n(D). The gel photographs were quantified using \nImageJ and expressed in relative units. Results are \nexpressed as mean ± s.e.m . of eight animals per \nexperimental group. Student’s t-test was used. \n*P < 0.05.\nDownloaded from Bioscientifica.com at 06/28/2026 10:39:13PM\nvia free access\n\n\nResearch 276\nTNFRp55 deficiency and \nendometriosis\nDOI: 10.1530/JOE-17-0236\nJournal of Endocrinology\ns vallcaneras  and others\nhttp://joe.endocrinology-journals.org © 2017 Society for Endocrinology\nPrinted in Great Britain\nPublished by Bioscientifica Ltd.\n234:3\nThe obtained results show that the TNFRp55 \ndeficiency promoted the establishment and the growth \nof endometriotic lesions associated to a high rate of cell \nproliferation and a low rate of apoptosis. Altogether, these \ndata constitute strong evidence of the association of the \nderegulation of the TNFRp55 expression and the survival \nof the endometriotic tissue in ectopic sites.\nThe MMPs have a key role in the pathogenesis of \nendometriosis. Thus, in a model of induced endometriosis \nin mouse, it has been demonstrated that the activity \nsuppression of these enzymes leads to inhibition in the \nendometriosis progression with the subsequent decrease in \nthe weight of the endometriotic-like lesions (Bruner  et al.  \n1997). In addition, increased levels of MMP-2 and MMP-9 \nhave been detected in peritoneal fluid of patients with \nendometriosis ( Amălinei  et  al.  2010 ). In our study, we \ndetected an increase of the enzymatic activity of MMP-2 \nin the peritoneal fluid of the KO mice, which indicates \nthat the TNFRp55 deficiency modifies the peritoneal \nenvironment contributing to the lesions remodeling and \nestablishment. In turn, several studies have shown that \nthe steroid regulation of MMP is critical for the formation \nof the fragments of the endometrial tissue in ectopic \nsites. In relation to this, there exists a correlation between \nthe levels of MMP-2 and estradiol in the peritoneal fluid \nof patients with endometriosis ( Huang  et  al.  2004 ). In \nagreement with this, we also observed the presence of \nhigh levels of estradiol that correlate with an increase in \nMMP-2 activity, which might favor the establishment and \ndevelopment of endometriotic lesions. It is important to \nemphasize that the TNFRp55−/− mice lack expression of \nTNFRp55 but display normal numbers of high affinity \nTNFRp75 molecules ( Pfeffer  et  al.  1993 ). In surgically \ninduced endometriosis in rat and mouse, the treatment \nwith etanercept, a fusion protein consisting of human \nrecombinant soluble TNFRp75 conjugated to a human \nFc antibody subunit, was associated with negative \nimmunohistochemical staining for TNFRp75 and reduced \nendometriosis development (Islimye  et al.  2011, Liu  et al . \n2016). These data indicate that the lack of activity of the \nTNFRp75, a receptor associated mainly with proliferation, \ncell survival and angiogenesis ( Haider  et al . 2009, Cabal-\nHierro & Lazo 2012 ), suppresses the development of \nendometriosis and that the effects observed in TNFRp55 \ndeficient animals of the present study may be due to \nactivation of TNFRp75-dependent pathways.\nFigure 5\nPearson’s correlation between estradiol levels and MMP-2 activity in \nperitoneal fluid samples. For the entire group of mice (n = 16), coefficient \n(r) Pearson: 0.67.\nFigure 6\nProgesterone levels, expression of metabolic enzymes and progesterone receptors. Progesterone levels were measured by RIA in peritoneal fluid (A) and \nendometriotic lesions (B) Measurement by RT-PCR of expression (mRNA) of Hsd3b1, Akr1c18 (C), Pgr (D). Actb was used as housekeeping gene. The gel \nphotographs were quantified using ImageJ and expressed in relative units. Results are expressed as mean ± s.e.m . of eight animals per experimental \ngroup. Student’s t-test was used. ● P < 0.01, ♦P < 0.005.\nDownloaded from Bioscientifica.com at 06/28/2026 10:39:13PM\nvia free access\n\n\n277Research\ns vallcaneras  and others TNFRp55 deficiency and \nendometriosis\nDOI: 10.1530/JOE-17-0236\nJournal of Endocrinology\nhttp://joe.endocrinology-journals.org © 2017 Society for Endocrinology\nPrinted in Great Britain\nPublished by Bioscientifica Ltd.\n234:3\nSpecifically in relation to the steroid hormones \nand endometriosis, estradiol is a factor favoring its \ndevelopment, while the role of progesterone seems to be \nattenuating. The TNFRp55 deficiency did not modify either \nthe levels of progesterone or its action sites. On the other \nhand, we observed that the TNFRp55 deficiency caused a \nstrong effect on the estradiol levels, which suggests the \nclose relationship between estradiol and TNFRp55. It is \nwell known that endometriotic lesions have the capacity \nto produce estradiol since they express the complete \nset of steroidogenic enzymes ( Kianpour  et  al . 2015 ). \nAmong these enzymes, P450arom plays an important \nrole in endometriosis ( Bulun  et  al . 2000 , Bilotas  et  al . \n2010). However, in our experimental model, this enzyme \nwas not modified by the TNFRp55 deficiency. It is also \nimportant to take into account that the beginning and \ndevelopment of this disorder might be promoted not only \nby the estrogen synthesized at local level but also by the \nsystemic estrogen (Rizner 2009). Thus, the estradiol from \nthe ovary and from the conversion of androstenedione \ncirculating in the adipose tissue and skin reaches the \nlesion via circulation. Therefore, the previous mechanism \nmight explain the increase observed in the estradiol level \nin peritoneal fluid and endometriotic lesion.\nThere is strong evidence of the bidirectional \ncommunication between the endocrine and immune \nsystem. For example, it has been described that estrogens \nmay contribute to tumor development, blocking the \nability of immune cells to induce apoptosis of target \ncancer cells (Jiang  et al.  2008). In endometriosis, the high \nlevel of estradiol can also play an important role in a local \ndecrease of immunosurveillance (Vetvicka  et al.  2016). In \naddition, in vitro studies suggest that endometriotic cells \nrespond to estrogen-induced antiapoptotic signaling more \nintensely than normal cells (Reis  et al.  2013). We postulate \nthat hormones, especially estradiol, may regulate the \nexpression of the gene that codifies TNFRp55 and that \nthis might be a mechanism involved in the survival of \nthe endometriotic cells in ectopic sites. In support of this \nhypothesis, we have mapped 1800 bp upstream from \nthe initial site of the Tnfrsf1a gene transcription using \nMatInspector software ( Quandt  et  al.  1995 ), and we \nidentified four putative estrogen response elements and \none putative progesterone response element (results not \nshown). In addition, a study through in situ hybridization \ndemonstrated that the uterus of adult mice expresses \nmRNA of TNFRp55 and after the ovariectomy and 72 h of \nestradiol administration, the mRNA that codifies TNFRp55 \nin uterine cells decreased ( Roby  et  al. 1996). These date \nindicate that estradiol may modulate the expression of \nthe gene encoding TNFRp55. Whether this mechanism is \ninvolved in the decreased susceptibility of endometriotic \ncells to apoptosis associated to estradiol remains to be \ninvestigated.\nIn conclusion, the deficiency of TNFRp55 promoted the \nestablishment and development of endometriosis through an \nincrease in the lesion size and high levels of estradiol, which \ncorrelate with an increase in the MMP-2 activity. Further \nstudies on the deregulation of the TNFRp55 expression and \nthe survival of the endometrial cells in ectopic sites might \ncontribute to improve the knowledge about pathophysiology \nand discover possible therapies or biomarkers of endometriosis.\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 work was supported by Grant PROICO 2-1812 CyT-UNSL and Fundación \nFlorencio Fiorini (2012–2013).\nAcknowledgements\nThe authors gratefully thank Laboratorio de Inmunopatología (IMIBIO-\nSL CONICET, UNSL) for providing KO mice. They also thank PhD Carlos \nTelleria, Co-Director Postdoctoral Scholarship-CONICET. This work is part \nof the Postdoctoral Scholarship-CONICET of PhD Sandra Vallcaneras and \nDoctoral thesis of Federica Ghersa.\nReferences\nAhn SH, Monsanto SP, Miller C, Singh SS, Thomas R & Tayade C 2015 \nPathophysiology and immune dysfunction in endometriosis. \nBioMed Research International 2015 article ID 795976. \n(doi:10.1155/2015/795976)\nAmălinei C, Căruntu ID, Giuşcă SE & Bălan RA 2010 Matrix \nmetalloproteinases involvement in pathologic conditions. Romanian \nJournal of Morphology and Embryology 51 215–228.\nAntsiferova Y & Sotnikova N 2012 The local immune mechanisms \ninvolved in the formation of endometriotic lesions. In Endometriosis: \nBasic Concepts and Current Research Trends, pp 211–44. Eds K Chaudhury \n& BN Chakravarty. Rijeka, Croatia: Intech. (doi:10.5772/1193)\nBilotas M, Meresman G, Stella I, Sueldo C & Barañao RI 2010 Effect of \naromatase inhibitors on ectopic endometrial growth and peritoneal \nenvironment in a mouse model of endometriosis. Fertility and Sterility \n93 2513–2518. (doi:10.1016/j.fertnstert.2009.08.058)\nBoric MA, Torres M, Pinto C, Pino M, Hidalgo P, Gabler F, Fuentes A & \nJohnson MC 2013 TNF system in eutopic endometrium from women \nwith endometriosis. Open Journal of Obstetrics and Gynecology 3 \n271–278. (doi:10.4236/ojog.2013.32051)\nBravo R & Macdonald-Bravo H 1987 Existence of two populations \nof cyclin/proliferating cell nuclear antigen during the cell cycle: \nassociation with DNA replication sites. Journal of Cell Biology 105 \n1549–1554. (doi:10.1083/jcb.105.4.1549)\nDownloaded from Bioscientifica.com at 06/28/2026 10:39:13PM\nvia free access\n\n\nResearch 278\nTNFRp55 deficiency and \nendometriosis\nDOI: 10.1530/JOE-17-0236\nJournal of Endocrinology\ns vallcaneras  and others\nhttp://joe.endocrinology-journals.org © 2017 Society for Endocrinology\nPrinted in Great Britain\nPublished by Bioscientifica Ltd.\n234:3\nBruner KL, Matrisian LM, Rodgers WH, Gorstein F & Osteen KG  \n1997 Suppression of matrix metalloproteinases inhibits \nestablishment of ectopic lesions by human endometrium in nude \nmice. Journal of Clinical Investigation 99 2851–2857. (doi:10.1172/\nJCI119478)\nBulun SE, Zeitoun KM, Takayama K & Sasano H 2000 Molecular basis for \ntreating endometriosis with aromatase inhibitors. Human Reproduction \nUpdate 6 413–418. (doi:10.1093/humupd/6.5.413)\nBulun SE, Cheng YH, Yin P, Imir G, Utsunomiya H, Attar E, Innes J & \nJulie Kim J 2006 Progesterone resistance in endometriosis: link to \nfailure to metabolize estradiol. Molecular and Cellular Endocrinology \n248 94–103. (doi:10.1016/j.mce.2005.11.041)\nCabal-Hierro L & Lazo PS 2012 Signal transduction by tumor necrosis \nfactor receptors. Cellular Signalling 24 1297–1305. (doi:10.1016/j.\ncellsig.2012.02.006)\nDi Carlo C, Bonifacio M, Tommaselli GA, Bifulco G, Guerra G &  \nNappi C 2009 Metalloproteinases, vascular endothelial growth \nfactor, and angiopoietin 1 and 2 in eutopic and ectopic \nendometrium. Fertility and Sterility  91 2315–2323. (doi:10.1016/j.\nfertnstert.2008.03.079)\nGrandi G, Mueller MD, Papadia A, Kocbek V , Bersinger NA, Petraglia F, \nCagnacci A & McKinnon B 2016 Inflammation influences steroid \nhormone receptors targeted by progestins in endometrial stromal cells \nfrom women with endometriosis. Journal of Reproductive Immunology \n117 30–38. (doi:10.1016/j.jri.2016.06.004)\nGreene AD, Lang SA, Kendziorski JA, Sroga-Rios JM, Herzog TJ & Burns \nKA 2016 Endometriosis: where are we and where are we going? \nReproduction 152 R63–R78. (doi:10.1530/REP-16-0052)\nGrümmer R 2006 Animal models in endometriosis research. Human \nReproduction Update 12 641–649. (doi:10.1093/humupd/dml026)\nHaider S & Knöfler M 2009 Human tumour necrosis factor: physiological \nand pathological roles in placenta and endometrium. Placenta 30 \n111–123. (doi:10.1016/j.placenta.2008.10.012)\nHuang HF, Hong LH, Tan Y & Sheng JZ 2004 Matrix metalloproteinase \n2 is associated with changes in steroid hormones in the sera and \nperitoneal fluid of patients with endometriosis. Fertility and Sterility 81 \n1235–1239. (doi:10.1016/j.fertnstert.2003.10.027)\nIslimye M, Kilic S, Zulfikaroglu E, Topcu O, Zergeroglu S & Batioglu \nS 2011 Regression of endometrial autografts in a rat model of \nendometriosis treated with etanercept. European Journal of Obstetrics \nand Gynecology and Reproductive Biology 159 184–189. (doi:10.1016/j.\nejogrb.2011.06.029)\nJana S, Chatterjee K, Ray AK, DasMahapatra P & Swarnakar S 2016 \nRegulation of matrix metalloproteinase-2 activity by COX-2-PGE2-\npAKT axis promotes angiogenesis in endometriosis. PLoS ONE 11 \ne0163540. (doi:10.1371/journal.pone.0163540)\nJiang X, Patterson NM, Ling Y, Xie J, Helferich WG & Shapiro DJ 2008 Low \nconcentrations of the soy phytoestrogen genistein induce proteinase \ninhibitor 9 and block killing of breast cancer cells by immune cells. \nEndocrinology 149 5366–5373. (doi:10.1210/en.2008-0857)\nKianpour M, Nematbakhsh M & Ahmadi SM 2015 Asymmetric \ndimethylarginine (ADMA), nitric oxide metabolite, and estradiol \nlevels in serum and peritoneal fluid in women with endometriosis. \nIranian Journal of Nursing and Midwifery Research 20 48–49. \n(doi:10.4103/1735-9066.160997)\nKitawaki J, Kado N, Ishihara H, Koshiba H, Kitaoka Y & Honjo H 2002 \nEndometriosis: the pathophysiology as an estrogen-dependent \ndisease. Journal of Steroid Biochemistry Molecular Biology 83 149–155. \n(doi:10.1016/S0960-0760(02)00260-1)\nLi Y, Adur MK, Kannan A, Davila J, Zhao Y, Nowak RA, Bagchi MK, Bagchi \nIC & Li Q 2016 Progesterone alleviates endometriosis via inhibition \nof uterine cell proliferation, inflammation and angiogenesis in \nan immunocompetent mouse model. PLoS ONE 11 e0165347. \n(doi:10.1371/journal.pone.0165347)\nLiu Y, Sun L, Hou Z, Mao Y, Cui Y & Liu J 2016 rhTNFR: Fc suppresses the \ndevelopment of endometriosis in a mouse model by downregulating \ncell proliferation and invasiveness. Reproductive Science 23 847–857. \n(doi:10.1177/1933719115620495)\nOthman ER, Hornung D, Hussein M, Abdelaal II, Sayed AA, Fetih AN & \nAl-Hendy A 2016 Soluble tumor necrosis factor-alpha receptors in \nthe serum of endometriosis patients. European Journal of Obstetrics \nand Gynecology and Reproductive Biology 200 1–5. (doi:10.1016/j.\nejogrb.2016.02.025)\nPan H, Zhang P, Li JR, Wang H, Jin MF, Feng C & Huang HF 2016 c-Fos-\nregulated matrix metalloproteinase-9 expression is involved in \n17β-estradiol-promoted invasion of human endometrial stromal cell. \nCurrent Molecular Medicine 16 266–275. (doi:10.2174/ \n1566524016666160225153454)\nPfeffer K, Matsuyama T, Kündig TM, Wakeham A, Kishihara K,  \nShahinian A, Wiegmann K, Ohashi PS, Krönke M & Mak TW \n1993 Mice deficient for the 55 Kd tumor necrosis factor receptor \nare resistant to endotoxic shock, yet succumb to L. monocytogenes \ninfection. Cell 73 457–467. (doi:10.1016/0092-8674(93)90134-C)\nPitsos M & Kanakas N 2009 The role of matrix metalloproteinases in \nthe pathogenesis of endometriosis. Reproductive Science 16 717–726. \n(doi:10.1177/1933719109333661)\nQuandt K, Frech K, Karas H, Wingender E & Werner T 1995 MatInd and \nMatInspector: new fast and versatile tools for detection of consensus \nmatches in nucleotide sequence data. Nucleic Acids Research 23 \n4878–4884. (doi:10.1093/nar/23.23.4878)\nReis FM, Petraglia F & Taylor RN 2013 Endometriosis: hormone \nregulation and clinical consequences of chemotaxis and apoptosis. \nHuman Reproduction Update 19 406–418. (doi:10.1093/humupd/\ndmt010)\nRizner TL 2009 Estrogen metabolism and action in endometriosis. \nMolecular and Cellular Endocrinology 307 8–18. (doi:10.1016/j.\nmce.2009.03.022)\nRoby KF, Laham N & Hunt JS 1996 Cellular localization and steroid \nhormone regulation of mRNA encoding tumour necrosis factor \nreceptor I in mouse uterus. Journal of Reproduction and Fertility 106 \n285–290. (doi:10.1530/jrf.0.1060285)\nRojas-Cartagena C, Appleyard CB, Santiago OI & Flores I 2005 \nExperimental intestinal endometriosis is characterized by increased \nlevels of soluble TNFRSF1B and downregulation of Tnfrsf1a and \nTnfrsf1b gene expression. Biology of Reproduction 73 1211–1218. \n(doi:10.1095/biolreprod.105.044131)\nSampson JA 1927 Metastatic or embolic endometriosis, due to the \nmenstrual dissemination of endometrial tissue into the venous \ncirculation. American Journal of Pathology 3 93–110.\nVetvicka V , Laganà AS, Salmeri FM, Triolo O, Palmara VI, Vitale SG, Sofo V  \n& Králíčková M 2016 Regulation of apoptotic pathways during \nendometriosis: from the molecular basis to the future perspectives. \nArchives of Gynecology and Obstetrics 294 897–904. (doi:10.1007/\ns00404-016-4195-6)\nReceived in final form 23 June 2017\nAccepted 3 July 2017\nAccepted Preprint published online 4 July 2017\nDownloaded from Bioscientifica.com at 06/28/2026 10:39:13PM\nvia free access","source_license":"CC0","license_restricted":false}