{"paper_id":"15b6f208-89bc-4e0d-8aea-dba85376e43a","body_text":"Endometrial biopsies were collected with an endometrial suction curette (Pipelle,\nLaboratorie CCD, Paris, France) from 91 healthy women of reproductive age, who were\npredominantly White/Caucasian. Written informed consent was obtained, and ethical\napproval was granted from Lothian Research Ethics Committee (LREC/07/S1103/29).\nParticipants were aged 22 to 50 years (median 41; mean 41). All reported regular\nmenstrual cycles (21 to 35 days) and had not taken any exogenous hormones or used an\nintrauterine device for 3 months prior to tissue collection. Women with large\nfibroids (>3 cm) and endometriosis were excluded.\nImmediately after collection, tissue was divided when possible and placed in the\nfollowing: (1) RNA later stabilization solution [Ambion (Europe), Warrington, UK] and\nstored at −70°C for RNA extraction; (2) neutral buffered formalin prior\nto paraffin wax embedding; and (3) phosphate-buffered saline (PBS) for stromal cell\nextraction. If limited tissue was obtained (which often occurred with menstrual phase\ncollection), neutral buffered formalin fixation was prioritized.\nMenstrual stage was carefully categorized according to the following: (1)\nhistological appearance based on the criteria of Noyes  et al.  ( 12 ), assessed by a consultant pathologist; (2)\nthe participant’s reported last menstrual period; and (3) serum progesterone\nand estradiol levels at the time of biopsy (see  Supplemental Methods ). Consistency for all 3\nparameters was necessary before inclusion. Six endometrial tissue samples were\nexcluded due to inconsistent dating and 1 sample due to detection of hyperplasia.\nBiopsies were classified as proliferative, early-mid secretory, late secretory, or\nmenstrual for analysis ( Supplemental Table 1 ).\nA subset of women (n = 78) also had objective measurement of their menstrual blood\nloss using the modified alkaline hematin method, as previously published ( 13, 14 ). In brief, women were given the same\nbrand of tampon and/or pad (Tampax tampons and Always towels, Proctor and Gamble,\nWeybridge, UK), with verbal and written instruction on collection. Used sanitary\nproducts were added to a measured volume of 5% sodium hydroxide. The contents were\nleft for 24 hours to allow conversion of hemoglobin to hematin. During the same time\nperiod, a 1 in 200 dilution of the patient’s venous blood in 5% sodium\nhydroxide was made and stored separately. The optical density (OD) of the samples was\nthen measured using spectrophotometry at 546 nm ( A 546 ).\nMenstrual blood loss (MBL) was calculated using the following equation ( 13 ): MBL = ( OD   of   menstrual   blood   solution × total   volume   of   added   NaOH ) ( OD   of   venous   blood × 200 ) Greater than 80 mL was classified as HMB, and <80\nmL as normal (NMB).\nThe 5-µm tissue sections were deparaffinized in xylene and rehydrated. Slides\nfor TGF- β RI and II were loaded into a Celerus Riptide\ndecloaking chamber (Celerus Diagnostics, Carpinteria, CA). Epitope retrieval was\nperformed using Novocastra Epitope Retrieval solution Ph6 (Leica Microsystems,\nErnst-Leitz-Straße, Wetzlar, Germany). Slides were loaded onto Leica Bond-Max\nautomated immunostainer (Leica Microsystems). Primary antibodies were applied for 2\nhours at 37°C (see  Supplemental Table 2 ), and negative control\ntissues were incubated with isotype-matched IgG at the same concentration as the\nprimary antibody. The presence of antigen was visualized with Bond Polymer refine\ndetection kit (Leica Microsystems). TGF- β 1 detection was\nperformed on the laboratory bench after pH9 antigen retrieval. The ImmPRESS\npolymerized reporter system (Vector Laboratories, Peterborough, UK) was used before\nliquid diaminobenzidine kit (Zymed Laboratories, San Francisco, CA) detection.\nSections were counterstained with hematoxylin, dehydrated, and mounted with Pertex\n(Cellpath, Hemel Hempstead, UK).\nLocalization and intensity of immunostaining were evaluated by two independent,\nmasked observers ( 15 ). The intensity of\nstaining was graded with a 3-point scale (0 = no staining, 1 = mild staining, 2 =\nstrong staining). This was applied to the glands and stromal cells, as well as the\nsurface epithelium and endothelial cells where visualized (note: the latter two\ncellular components were often absent in menstrual phase tissue, accounting for the\nlower n numbers in these groups). The percentage of tissue in each intensity scale\nwas recorded ( 15 ). A value was derived for\neach of the cellular compartments by using the sum of these percentages after\nmultiplication by the intensity of staining.\nPrimary human endometrial stromal (HES) cells were isolated from secretory\nendometrial tissue (n = 6) by enzymatic digestion, as previously described ( 16 ). These women met the criteria detailed\npreviously but did not undergo objective measurement of their menstrual blood loss.\nCells were cultured in RPMI 1640 medium supplemented with 10% fetal calf serum, 1%\n200 mM L-glutamine, and 500 mg/mL gentamycin.\nSecretory phase HES cells from 3 patients with a subjective complaint of HMB and not\nusing oral or inhaled corticosteroids were plated at 3 × 10 5 \ncells/well in 6-well plates in 10% RPMI 1640. The next day, cells were washed in PBS\nand incubated in serum-free media overnight. Cells were then treated for 24 hours in\nduplicate with the following: (1) vehicle (1:1000 absolute ethanol); (2) 1 µM\ncortisol ( 17 ); or (3) 1 μM cortisol\nplus 5 μM leucine-serine-lysine-leucine (LSKL), a TSP-1 inhibitor (following a\n2-hour pretreatment with 5 μM LSKL alone). The cell supernatant was collected\nfor enzyme-linked immunosorbent assay (ELISA), and RNA was extracted from cells.\nTotal RNA from cells and endometrial biopsies was extracted using the RNeasy Mini Kit\n(Qiagen, Sussex, UK) with on-column DNase I digestion, according to\nmanufacturer’s instructions. RNA samples were reverse transcribed using the\nSuperscript VILO cDNA synthesis kit (Invitrogen, Paisley, UK), according to\nmanufacturer’s instruction, with appropriate controls. Primers for each gene\nof interest were designed using the Universal Probe Library Assay Design Center\n(Roche Applied Science, Burgess Hill, UK) (see  Supplemental Table 3 ) and purchased from Eurofins\n(MGW Operon, Ebergsberg, Germany). Polymerase chain reaction was carried out using\nABI Prism 7900 (Thermo Fisher Scientific, Loughborough, UK). Samples and controls\nwere analyzed in triplicate using Sequence Detector version 2.3 (Thermo Fisher\nScientific), using the comparative threshold method. Messenger RNA (mRNA) transcripts\nwere normalized relative to the geomean of two appropriate housekeeping genes, 18S\nand ATP5B, as determined by geNorm assay (Primerdesign, Southampton, UK), and\nquantified relative to a positive human liver cDNA sample.\nA TGF- β 1 ELISA was performed using a Human\nTGF- β 1 Quantikine Kit (DB100B; R&D Systems,\nLoughborough, UK), according to the manufacturer's instructions. Samples were\nanalyzed without activation and with latent TGF- β 1 activated\nto the immunoreactive form using 1 m HCl and neutralized with 1.2 m NaOH/0.5 m HEPES\nbuffer. Samples were assayed in duplicate, and after development assays were measured\non a Laboratory Systems Multiscan EX Microplate reader at 450 nm with wavelength\ncorrection at 540 nm. Values were determined by standard curve analysis. Intra-assay\ncoefficient of variability was 2.5%, and the between-batch coefficient of variability\nwas 8.3% for cell culture supernatants.\nSecretory phase HES from three participants (passage <5) were seeded at 2\n× 10 5 /well in 12-well plates in appropriate supplemented media (see\nprevious discussion), and, 16 hours before scratch, medium was changed to serum free.\nEach well of cells was scratched with a sterile 200 μL pipette tip, washed\nwith PBS, and then incubated in serum-free media with vehicle, 1 ng human recombinant\nTGF- β 1 (PeproTech, London, UK), or 10 μg/mL\nTGF- β  type I activin receptor-like kinase receptor\ninhibitor SB 431542 hydrate (Sigma-Aldridge, Dorset, UK) (n = 3 participants,\ntriplicate wells for each). For each well, 4 to 5 images were captured along the\nlength of each wound at 0 and 24 hours using an Axiovert 200 M inverted microscope\n(Carl Zeiss, Jena, Germany). Images were analyzed using AxioVision release 4.72, and\ncalculations of average distance closed for each sample were based on three\nmeasurements at identical positions along each wound image at 0 and 24 hours.\nAnalysis was carried out using GraphPad Prism Software (San Diego, CA). For\ncomparison of multiple data sets with two grouping variables ( i.e. ,\nHMB  versus  NMB and stage of menstrual cycle, mRNA, and\nimmunohistochemistry data), a two-way analysis of variance was used, with\nBonferroni’s multiple comparisons test. A paired one-way analysis of variance\nwith Tukey’s multiple comparisons test was used to compare cell culture\ntreatments. Tissue and cell endometrial mRNA results were expressed as the quantity\nrelative to a comparator sample of RNA from human liver. A value of\n P  < 0.05 was considered significant.\n\nTGF- β 1 mRNA was examined by quantitative reverse\ntranscription polymerase chain reaction in whole endometrial biopsies from women\nsampled at various stages of the menstrual cycle who had objectively determined\nmenstrual blood loss. Overall, the stage of the menstrual cycle had a significant\nimpact on  TGFB1  expression ( P  = 0.0025,\n F  = 5.339), with the late secretory phase resulting in\nsignificantly higher levels of  TGFB1  than endometrium from the\nproliferative ( P  < 0.001) or early-mid secretory\n( P  < 0.01) phases ( Fig.\n1 ). There was no significant difference between the late secretory and\nmenstrual phase. The increased transcription of TGFB1 in the late secretory phase did\nnot continue into the menstrual phase.\nTGFβ1  in the human endometrium.\nTGF- β 1 mRNA concentrations in endometrium from\nacross the menstrual cycle in women with HMB (blood loss >80 mL) and NMB\n(blood loss <80 mL). E/MS, early-mid secretory; LS, late secretory; M,\nmenstrual; P, proliferative. *** P \n< 0.001; ** P  < 0.01.\nWe compared the expression of  TGFB1  in women with NMB and HMB ( Fig. 1 ) and found no significant difference in\n TGFB1  expression between the two groups at any cycle stage. In\naddition, the two major TGF β 1 receptors, type I and type II,\nwere examined in the late secretory and menstrual endometrial samples. Neither\n TGFBR1  nor  TGFBR2  expression was significantly\ndifferent in endometrium from women with HMB  versus  NMB [ Fig. 2(A)  and  2(B) ]. Immunohistochemical staining revealed maximal staining of TGFBR1 in\nsurface and glandular epithelial cells, with lower intensity staining in the stromal\ncompartment [ Fig. 2(C) ]. TGFBR2 showed a similar\npattern, with highest immunostaining in epithelial and endothelial cells [ Fig. 2(D) ]. Semiquantitative histoscoring by two\nmasked independent observers confirmed no differences in either receptor when\ncomparing women with HMB and NMB throughout the perimenstrual phase [ Fig. 2(E)  and  2(F) ].\nTGF- β RI and TGF- β RII in the\nhuman endometrium before and during menstruation. (A)\nTGF- β RI mRNA concentrations in endometrium from\nwomen with normal (NMB; <80 mL) and heavy (HMB; >80 mL) menstrual\nbleeding during the late secretory (LS) and menstrual (M) phases. (B)\nTGF- β RII mRNA concentrations. (C)\nImmunohistochemical staining of TGF- β RI in endometrium\nfrom the late secretory phase. Arrow indicates endothelial cells. (D)\nImmunohistochemical staining of TGF- β RII in endometrium\nfrom the late secretory phase; inset: negative control. (E) Immunohistochemical\nhistoscore of TGF- β RI in human endometrium from women\nwith heavy and normal bleeding during the late secretory and menstrual phases.\n(F) Immunohistochemical histoscore of TGF- β RII in human\nendometrium from women with heavy and normal bleeding during the late secretory\nand menstrual phases. (Note: lower n numbers appear in surface epithelium and\nendothelial cell scoring due to the inability to identify these cells in some\ntissues.) GE, glandular epithelium; SE, surface epithelium; St, stromal cell\ncompartment.\nAs the numerous cell types in the endometrium expressed\nTGF- β 1 receptors, we examined the localization of TGFB1 by\nimmunohistochemistry. TGFB1 could be immunolocalized to the cytoplasm of the surface\nepithelium, glandular epithelium, stromal cells, and endothelial cells throughout the\nperimenstrual phase of the cycle in women with NMB (<80 mL) and HMB\n(>80 mL) [ Fig. 3(A) ]. Semiquantitative\nhistoscoring revealed that protein in the menstrual phase was similar to late\nsecretory phase. There was significantly reduced TGFB1 staining in the stromal cell\ncompartment of endometrium from women with HMB  versus  those with NMB\n[ Fig. 3(B) ]. This suggests some\nposttranscriptional regulation of TGF- β 1 in stromal\ncells.\nImmunohistochemistry for TGF- β 1 in human endometrium\nfrom the perimenstrual phase. (A) Staining of late secretory (LS) and menstrual\n(M) phase endometrium from women with HMB (>80 mL) and NMB (<80\nmL). Arrows indicate endothelial cells. Inset: negative control. (B)\nSemiquantitative histoscoring of TGF- β 1\nimmunohistochemistry staining. GE, glandular epithelium; SE, surface\nepithelium; St, stromal compartment. * P  <\n0.05.\nTo further investigate the posttranscriptional regulation of\nTGF- β 1, we collected primary HES cells from 3 women in the\nsecretory phase of the menstrual cycle for  in vitro  analysis.\nPerimenstrual serum progesterone and estradiol levels were not significantly\ndifferent between women with HMB or NMB ( Supplemental Table 1 ). However, we have\npreviously shown that cortisol is involved both in endometrial repair and the\nregulation of endometrial TSP-1 ( 14 ), a known\nregulator of TGF- β 1 activity ( 6 ). Cortisol or cortisol plus LSKL (a TSP-1 inhibitor) produced a\nsignificant decrease in  TGFB1  expression in HES cells\n[ P  < 0.05,  Fig.\n4(A) ], but there was no difference in the amount of latent TGFB1 secreted,\ndetected by pH activation of culture supernatants prior to detection of activated\nTGFB1 by ELISA [ Fig. 4(B) ]. However, analysis of\nunactivated cell culture supernatants revealed an increase in activation of\nTGF- β 1 protein on treatment with cortisol, which was\nprevented with cotreatment of cells with the TSP-1 inhibitor LSKL [ P \n> 0.05,  Fig. 4(C) ]. These data reveal\ncortisol does not increase the transcription or latent protein levels of stromal cell\nTGF- β 1 but has a role in the activation of latent\nTGF- β 1 in human endometrial stromal cells, via TSP-1.\nThe regulation of TGF- β 1 by cortisol in primary human\nendometrial stromal cells. (A) TGF- β 1 mRNA after\n24-hour treatment with vehicle, cortisol (1 μM), or cortisol (1\nμM) plus a TSP-1 inhibitor (5 μM LSKL). (B) Active\nTGF- β 1 protein levels in experimental culture\nsupernatants following pre-ELISA acid activation of latent\nTGF- β 1. (C) Active TGF- β 1\nprotein levels in the same culture supernatants without pre-ELISA acid\nactivation (* P  < 0.05).\nTGF- β 1 activity increases the expression and phosphorylation\nof the regulatory SMADs (SMAD2 and SMAD3). These activated pSMADs then interact with\nthe comediator SMAD4 and translocate to the nucleus to regulate transcription of\ntarget genes ( 5 ). Examination of\n SMAD2  and  SMAD3  expression revealed significant\ndecreases in women with HMB  versus  NMB during the menstrual phase of\nthe cycle [ P  < 0.05,  Fig.\n5(A)  and  5(B) ]. SMAD2 was\nsignificantly increased in women with HMB  versus  NMB during the late\nsecretory phase [ Fig. 5(A) ]. Immunohistochemical\nstaining for phosphorylated SMAD2/3 again revealed localization to the glandular\nepithelium, surface epithelial cells, stromal compartment, and endothelial cells\n[ Fig. 5(C)  and  5(D) ]. Histoscoring revealed a significant reduction in activated SMAD2/3\nprotein levels in the endometrial glandular epithelial cells in women with HMB\n versus  NMB during the late secretory phase of the menstrual cycle\n[ Fig. 5(D) ].\nSMAD2/3 in the human endometrium before and during menstruation. (A) SMAD2 mRNA\nconcentrations in endometrium from women with normal (NMB; <80 mL) and\nheavy (HMB; >80 mL) menstrual bleeding during the late secretory (LS)\nand menstrual (M) phases. (B) SMAD3 mRNA concentrations in endometrium from\nwomen with NMB and HMB in the late secretory and menstrual phases. (C)\nPhosphorylated SMAD2/3 immunohistochemical staining in late secretory\nendometrium from a woman with NMB. Inset: negative control. Arrow indicates\nendothelial cells. (D) Phosphorylated SMAD2/3 immunohistochemical staining in\nlate secretory endometrium from a woman with HMB. (E) Histoscoring of\nimmunostaining for phosphorylated SMAD2/3. GE, glandular epithelium; SE,\nsurface epithelium; St, stromal cell compartment. * P \n< 0.05.\nTo examine the functional effects of increased TGF- β 1\nactivity, primary HES were subjected to a wound scratch assay. As these cells are\nsources of TGF- β 1, they were studied in the presence of\nvehicle, SB-431542 (to block endogenously stimulated phosphorylation of SMAD\nproteins), or TGF- β 1. HES cells showed significantly\nincreased wound closure with TGF- β 1 treatment\n versus  SB-431542–treated cells ( P \n< 0.05,  Fig. 6 ).\nThe effect of TGF- β 1 on human endometrial cell wound\nrepair. (A) Average wound scratch closure distance (scratch distance at 0 hours\nminus scratch distance at 24 hours) in human primary stromal endometrial cells\nafter treatment with vehicle, the Alk receptor inhibitor SB-431542, or 1 ng\nTGF- β 1. (B) Images of wound scratch in HES cells\ntreated with 10 μg/mL SB-431542 for the following: (i) 0 hours; (ii) 24\nhours and treated with 1 ng TGF- β 1; (iii) 0 hours; and\n(iv) 24 hours. * P  < 0.05.\n\nIn this study, we detail significant differences in TGF- β 1\ndownstream of local steroid action in the endometrium of women with HMB during\nmenstruation. Endometrium from women with objectively measured HMB had decreased\nTGF- β 1 protein levels, unaltered\nTGF- β  receptor presence, and a significant reduction in both\nSMAD2 and 3 mRNA concentrations and SMAD2/3 protein phosphorylation before/during the\nmenstrual phase when compared with women with NMB. We provide mechanistic data\nsupporting TGF- β 1 protein activation by cortisol in endometrial\ncells, via TSP-1. In addition, our functional studies reveal that a suboptimal\nTGF- β  response in the local endometrial environment may\ndecrease postmenstrual repair of the stromal compartment and lead to heavy, prolonged\nmenstrual bleeding ( Fig. 7 ).\nProposed role of TGF- β 1 in the human endometrium at\nmenstruation. Red stars represent findings in women with HMB and potential impact\non endometrial function.\nPrevious studies have detailed that TGF- β 1 levels in endometrial\ntissue explants are suppressed by progesterone ( 8 ). These authors found secretory explants cultured for 24 hours in the absence\nof progesterone and estrogen, a milieu analogous to the menstrual phase, significantly\nincreased  TGFβ1  mRNA. Our results support these findings, with\nsignificantly greater  TGFβ1  mRNA prior to and during\nmenstruation when compared with the proliferative and early-mid secretory phases,\nconsistent with upregulation following progesterone withdrawal. We did not observe any\nsignificant difference in endometrial  TGFβ1  mRNA between women\nwith HMB and normal blood loss during menstruation, although we acknowledge our n\nnumbers are small. However, we did observe significantly decreased\nTGF- β  protein in the stromal compartment of women with HMB\n versus  NMB during menstruation. We acknowledge that menstrual biopsy\nn numbers are low, but these tissues are meticulously classified and have objective\nmeasurement of participant menstrual blood loss to aid precision of data. Our results\nsuggest differences in TGF- β 1 protein in women with HMB and NMB\nare not due to transcriptional regulation, but that posttranscriptional regulation may\nbe aberrant.\nInterestingly, there were no significant differences in serum progesterone or estradiol\nlevels between women with HMB and NMB. In addition, no significant differences in\nendometrial estradiol receptor or progesterone receptor expression were previously\ndetected in women with measured menstrual blood loss ( 18 ). Therefore, we hypothesized that local cortisol action may influence\nTGF- β 1 activity during menses.\nTGF- β  is synthesized as a dimeric preproprotein and is released\nin a latent form. TSP-1 is known to activate TGF- β 1 and is\nthought to do so by inducing a conformational change in the latent protein ( 6 ). Our laboratory has previously published that\nwomen with HMB have significantly reduced endometrial TSP-1 mRNA levels when compared\nwith women with normal bleeding ( 14 ). Previous\nstudies from our laboratory have also found that cortisol increases TSP-1 mRNA\nexpression in primary human endometrial stromal cells ( 14 ). Direct measurement of cortisol levels in the endometrium of women with\nHMB and NMB has not yet been carried out, but an enhanced local inactivation of cortisol\nby 11 β HSD2 may be present in the endometrium of women with heavy\nmenses ( 14 ). The 11 β HSD2\nmRNA was increased 2.5-fold in women with HMB  versus  NMB, predicting\nsubstantially lower local cortisol concentrations. Therefore, we examined whether\ncortisol was a local regulator of TGF- β 1 activity via TSP-1. On\nexamination of cell culture supernatants from HES cells treated with physiological\nlevels of cortisol ( 17 ), activated\nTGF- β 1 was significantly increased. This increase was\nabrogated by the addition of a TSP-1 inhibitor to culture. Interestingly, acid\nactivation of latent TGF- β 1 in the culture supernatant prior to\nELISA resulted in no differences in TGF- β 1 levels with any of\nthe treatments used. This is consistent with cortisol-stimulated TSP-1 production acting\non latent TGF- β 1 protein to increase its activity, rather than\nincreasing the transcription or translation of TGF- β 1. Indeed,\ncortisol and cortisol plus TSP-1 inhibitor treatment both significantly decreased\n TGFβ1 .  TGFβ1  mRNA was not\nsignificantly different in the endometrium of women with NMB  versus  HMB\nduring the perimenstrual phase, but there was a trend toward increased\nTGF- β 1 mRNA concentrations in women with HMB at this time,\nconsistent with lower endometrial cortisol levels ( 14 ).\nNext, we examined the functional significance of TGF- β 1 protein\nlevels on endometrial cells. After shedding, endometrial cells migrate to cover the\nexposed surface of the endometrium and the stromal compartment regenerates ( 19 ). The wound scratch assay mimics this process\n in vitro , providing a means of quantifying stromal cell migration\nacross a wounded surface. We found that TGF- β 1 increased wound\nhealing of primary stromal cell cultures. As we detected reduced phosphorylation of\nSMAD2/3 in the endometrium of women with HMB  versus  NMB, we blocked\nTGF- β –mediated activation of SMAD proteins with SB\n431542 and showed a decrease in stromal cell wound migration, which was significantly\nless than that seen with the addition of TGF- β 1. We propose that\nwomen with HMB may have defective or delayed repair of the stromal cell compartment\nfollowing shedding of their functional endometrium at menses.\nIn addition to its functional role in proliferation, it is clear that the\nTGF- β  superfamily plays an important role in endothelial cell\nfunction and blood loss. Greater than 50% of TGF- β 1 knockout\nmice die during embryogenesis due to yolk sac defects affecting vasculogenesis and\nresulting in vessel fragility ( 20 ). In humans,\nmutation of the TGF- β  receptor I activin receptor-like kinase I\nor of the endothelial accessory receptor endoglin causes hereditary hemorrhagic\ntelangiectasia, an autosomal dominant vascular disease ( 21 ). The resulting aberrant TGF- β  superfamily\nsignaling results in epistaxis, telangiectasia, and arteriovenous malformations.\nInterestingly, previous histochemical and microscopic examination of endometrial blood\nvessels from women with normal and HMB revealed increased endothelial gaps in women with\nheavy loss ( 22 ). The role of the\nTGF- β  superfamily in this pathology remains to be determined,\nbut the observational data contained in this work suggest that low late\nsecretory/menstrual TGF- β 1 protein levels and decreased pSMAD2/3\nmay be involved. Previous results from our center support a role for\nTGF- β 1 in the generation of vasoactive factors in women with\nendometriosis ( 23, 24 ) and it may have a\nsimilar, if more regulated, role in the endometrium to ensure physiological\nmenstruation.\nWe have previously shown that cortisol is angiostatic, preventing endothelial tubelike\nstructure formation  in vitro  ( 14 ). Furthermore, small interfering RNA silencing of TSP-1 in uterine\nendothelial cells reversed the antiangiogenic effect. In combination with data contained\nin this work, we propose that cortisol may activate endometrial\nTGF- β 1 via TSP-1 during menses to prevent an excessive\nangiogenic response and increase vascular integrity. Further experiments are required to\ndefinitively test this hypothesis.\nIn conclusion, we show that women with objectively measured HMB have decreased\nendometrial TGF- β 1 protein and downstream SMADs during the late\nsecretory/menstrual phase when compared with women with NMB. This may partially explain\nthe increased menstrual blood loss experienced by many women. In addition, we show that\ncortisol has a mechanistic role in the activation of endometrial\nTGF- β 1 at this time ( Fig.\n7 ). Our  in vitro  results are consistent with\nTGF- β 1 having a functional role in repair of the denuded\nendometrial surface at menstruation, and we propose that women with HMB may benefit from\ntherapies that increase TGF- β  during menses.","source_license":"CC0","license_restricted":false}