rea in endometriosis_2015

Multiple Beneficial Roles of Repressor of EstrogenReceptor Activity (REA) in Suppressing theProgression of Endometriosis

Yuechao Zhao, Yiru Chen, Ye Kuang, Milan K. Bagchi, Robert N. Taylor,John A. Katzenellenbogen, and Benita S. Katzenellenbogen

Departments of Molecular and Integrative Physiology (Y.Z., Y.C., M.K.B., B.S.K.) and Chemistry (J.A.K.),University of Illinois at Urbana-Champaign, Urbana, Illinois 61801; Department of Gynecology andObstetrics (Y.K.), The Second Affiliated Hospital of Harbin Medical University, Harbin, Heilongjiang150086, China; and Department of Obstetrics and Gynecology (R.N.T.), Wake Forest School ofMedicine, Winston-Salem, North Carolina 27157

Endometriosis is an estrogen-dependent, inflammation-driven gynecologic disorder in which en-dometrial tissue creates inflammatory lesions at extrauterine sites, leading to pelvic pain andimpaired fertility. Although dysregulated estrogen receptor (ER) signaling has been implicated,understanding of this disease is incomplete and current therapies are of limited benefit. Using animmunocompetent syngeneic murine model, we used combinations of donor uterine tissue and/orrecipient host mice with partial genetic deletion of the ER coregulator, repressor of ER activity (REA)(also known as prohibitin 2), to investigate roles of REA in the contributions of donor uterine tissueand host cell influences on endometriosis establishment and progression. Ectopic lesions derivedfrom donor tissue with half the wild-type gene dosage of REA (REA�/�) grown in REA�/� hostsdisplayed enhanced proliferation, vascularization, and markedly increased neuron innervationand inflammatory responses, including elevated cytokine production, nuclear factor kappa B ac-tivation, cyclooxygenase-2 expression, and immune cell infiltration. Although lesion progressionwas greatest when REA was reduced in both donor tissue and host animals, other donor/hostcombinations indicated that distinct stimulatory inputs were derived from ectopic tissue (prolif-erative signals) and host cells (inflammatory signals). Importantly, depletion of REA in primaryhuman endometriotic stromal cells led to elevated proliferation and expression of cell cycle reg-ulators. Notably, REA was significantly lower in human endometriotic tissue versus normal humanendometrium. Thus, REA modulates cross talk among multiple cell types in the uterine tissue andhost background, serving as a brake on the estradiol-ER axis and restraining multiple aspects thatcontribute to the pathologic progression of endometriosis. (Endocrinology 157: 900–912, 2016)

Endometriosis is an estrogen-dependent and inflamma-tion-driven disorder in which endometrial tissue at-

taches at extrauterine ectopic sites, proliferates, and formsinvasive lesions. It affects 10%–14% of reproductive agewomen, with an even higher prevalence of 35%–50%among patients with pelvic pain and infertility (1–4). The

devastating effects of this disease on millions of womenand the high rate of disease recurrence after treatmentindicate the need for better mechanistic understanding ofthis disorder with the ultimate goal of developing moreeffective therapies and long-term management ofendometriosis.

ISSN Print 0013-7227 ISSN Online 1945-7170Printed in USACopyright © 2016 by the Endocrine SocietyReceived April 10, 2015. Accepted December 3, 2015.First Published Online December 14, 2015

Abbreviations: CD3, cluster of differentiation 3 T cell co-receptor; COX2, cyclooxygen-ase-2; E2, estradiol; EGFP, enhanced green fluorescent protein; ER, estrogen receptor;F4/80, macrophage marker EGF-like module-containing mucin-like hormone receptor-like1; GFP, green fluorescent protein; GL3, luciferase reporter vector GL3; IHC, immunohis-tochemistry; Ki67, proliferation marker nuclear protein Ki67; KO, knockout; PGP9.5, pro-tein gene product 9.5; PGR, progesterone receptor; p65, transcription factor p65; P-H3,phosphohistone-H3; qPCR, quantitative polymerase chain reaction; REA, repressor of ERactivity; siGL3, GL3 luciferase control siRNA; siREA, REA siRNA; siRNA, small interferingRNA; WT, wild type.

O R I G I N A L R E S E A R C H

900 press.endocrine.org/journal/endo Endocrinology, February 2016, 157(2):900–912 doi: 10.1210/en.2015-1324

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 June 2016. at 13:10 For personal use only. No other uses without permission. . All rights reserved.

Endometriosis is complex in that it is promoted by anddependent upon extensive cross talk among the numerouscell types that comprise the endometriotic lesions. Theseinclude endometrial cells from the uterus, likely from ret-rograde menstrual flow, plus immune cells that infiltrateinto the lesions, and vascular endothelial cells, blood ves-sels and nerves that grow into and support lesion estab-lishment, survival and progression. Host peritoneal cellsinto which the lesion embeds may also contribute by pro-viding a favorable environment for lesion survival. Manyof these multiple cell types express estrogen receptors(ERs) and ER coregulators (5–7) that can work together toenhance or restrain the estrogen signaling that promotesprogression of the disease.

A hallmark of endometriosis is excessive estrogen sig-naling (8, 9). This is supported by increased local produc-tion of estrogen driven by the aromatase gene, cytochromeP450, family 19, subfamily A, polypeptide 1 (2, 10), aswell as altered expression of ER� and ER� (11, 12). There-fore, current hormonal treatments, including progestins,androgens, GnRH agonists, and aromatase inhibitors, fo-cus on reducing systemic levels of estrogens. These treat-ments, however, are limited by side effects, incompleteeffectiveness, and high rates of disease recurrence aftertreatment cessation (2). To interrogate the moleculareventsunderlying the establishmentandprogressionof theuterine tissue at ectopic sites, murine models have beenwidely used recently (13). Critical roles of the ERs in en-dometriosis pathogenesis are also supported by studiesemploying ER knockout (KO) mice (14) and preclinicalanalyses with ER subtype-selective ligands (15–17). Be-cause ER coregulators are recruited to chromatin by ste-roid hormone nuclear receptors in a temporally and spa-tially specific manner for precise gene regulation, thesecomponents also emerge as contributors to and potentialtherapeutic targets for controlling the multiple hyper-estrogenic stimulatory activities that drive endometriosisestablishment and progression.

The ER coregulator, repressor of ER activity (REA) (alsoknown as prohibitin 2) (18, 19), has been shown to serve asa brake on ER activity in estrogen target tissues such as theuterus (20, 21) and mammary gland (22) and to affect cellsignaling pathways. Homozygous ablation of REA in theuterus led to infertility due to severely compromised uterinedevelopment and failure of implantation (21). However,uteri of heterozygous REA�/� mice, with half the normalwild-type (WT) level of REA, showed an accelerated andamplified decidualization process and subfertility, due to hy-perresponsiveness to estrogen signaling (20, 21, 23). Thesestudies, documenting REA to be a physiologic, protectivefactor against excessive estrogen-driven activity in uterinetissue, led us to hypothesize that REA might also play a crit-

ical role in modulating or moderating the establishment andprogression of endometriosis.

To explore this hypothesis, we have used an immuno-competent murine model, in which estrogen hyperstimu-lation of proliferation and inflammatory signaling in ec-topic lesions recapitulate the disease in humans (24). Ourfindings reported herein using donor tissue and host ani-mals with reduced levels of REA highlight distinctive rolesthat this ER corepressor plays in the ectopic uterine tissueand host tissues: normal levels of REA in ectopic uterinetissue restrain estrogen-supported implant growth andvascularization, whereas normal REA levels in host tissuessuppress inflammatory responses associated with lesionprogression. Elevated proliferative activity of human en-dometriotic stromal cells upon loss of REA and our com-parison of human endometriotic tissue from patients vsnormal human endometrium, which revealed significantlylower REA in endometriotic lesions further support theclinical relevance of REA and the usefulness of observa-tions from our animal model. The findings highlight REAas a protective restraint on the estradiol (E2)-ER-drivenaxis in endometriosis that acts as a corepressor of multipleaspects of the pathologic progression of this disease.

Materials and Methods

Animals and immunocompetent mouse model ofendometriosis

All animals were maintained in accordance with the NationalInstitutes of Health Guide for Care and Use of Laboratory An-imals, and all procedures were approved by the University ofIllinois Institutional Animal Care and Use Committee. C57BL/6mice were purchased from Harlan Laboratories or The JacksonLaboratory (EGFP, stock number 006567). REA heterozygous(REA�/�) mice on the C57BL/6 background and their WT lit-termates were maintained and genotyped as described previously(20, 23).

Endometriotic-like lesions were surgically transplanted as de-scribed before (17, 24). Briefly, female WT, EGFP, or REA�/�

mice (8–10 wk of age) served as either donor or recipient ani-mals, or both. The uterine horns were removed from donor mice,opened longitudinally, cut into fragments using a 3-mm dermalbiopsy punch (Miltex) and transplanted onto the peritoneal wallof recipient mice by suturing. In each experimental group, uter-ine tissue was collected from at least 6 donor mice and trans-planted into 6 recipient mice. Ectopic lesion volume was calcu-lated as before (17).

To examine roles of REA in E2-supported lesion establish-ment, ovariectomized recipient mice were implanted with a pelletof E2 (Innovative Research of America) sc and underwent ecto-pic tissue transplantation on the same day. The dosage of0.125-mg E2/pellet was chosen as optimal based on our pre-vious work (17). At the times indicated, both eutopic uterinetissue and ectopic endometriotic lesions were collected forfurther analysis.

doi: 10.1210/en.2015-1324 press.endocrine.org/journal/endo 901


To interrogate functions of REA during chronic lesion pro-gression, uterine fragments from donor mice were transplantedon alternate sides of the peritoneal incision into intact femalerecipients without any hormonal administration. The ectopictissues were collected at 2, 4, and 8 weeks after transplantationsurgery for further analysis.

In order to minimize hormonal variation in cycling mice, alldonor mice used above, as well as intact recipients, underwenttissue transplantation at the diestrous stage. In addition, recip-ient mice were killed at diestrus for collection of tissues as de-scribed before (17, 24).

Primary human endometrial and endometrioticstromal cell cultures and small interfering RNA(siRNA) studies

Our studies involving human eutopic endometrial biopsies,endometriotic lesion biopsies, and primary cell cultures wereapproved by the Institutional Review Boards of the University ofIllinois, Emory University, and Wake Forest University School ofMedicine. All protocols adhere to the regulations set forth for theprotection of human subjects participating in clinical research,including the establishment of a data and safety monitoring plan.

Isolation and culture of primary human endometriotic stro-mal cells were conducted as described (23). Cells were culturedin DMEM/F-12 medium (Invitrogen) containing 5% charcoal-dextran-treated fetal bovine serum. For siRNA experiments, en-dometriotic stromal cells were transfected with REA siRNA(siREA) or GL3 luciferase control siRNA (siGL3) (Dharmacon)following the Silent-Fect kit protocol (Bio-Rad Laboratories) asbefore (23). After 24 hours of transfection, cells were exposed to20-ng/mL TNF� (R&D Systems) and 10nM E2 (Sigma-Aldrich)for the times indicated.

Histological analyses, immunohistochemistry (IHC),and immunofluorescence

IHC and immunofluorescence were performed in culturedcells or paraffin-embedded mouse or human tissue sections asdescribed (23). Primary antibodies used (Supplemental Table 2)were: REA (Millipore Co), GFP (Cell Signaling Technology),Ki67 (Bioss), platelet endothelial cell adhesion molecule (PE-CAM) (Abcam), IL-6 (Invitrogen), p65 (Cell Signaling Technol-ogy), CD3 (Abcam), F4/80 (Acris Antibodies), COX2 (Abcam),protein gene product 9.5 (PGP9.5) (Abcam), progesterone re-ceptor (PGR) (DAKO), and phosphohistone-H3 (P-H3) (Milli-pore Co). The stain signal was quantified by monitoring theaverage numbers of positively stained cells to the total number ofcells from 6 randomly chosen fields.

RNA isolation and real-time PCRTotal RNA was isolated from eutopic or ectopic tissues or

primary cells using TRIzol Reagent (Life Technologies) to pre-pare cDNA (17, 23, 24). Real-time PCR was performed to quan-tify gene expression using specific human or mouse primers (Sup-plemental Table 1) and SYBR Green kits (Bio-Rad Laboratories).After analysis by the delta cycle threshold method, data werenormalized to 36B4 gene expression (23).

Statistical analysisStatistical analyses included paired or unpaired t test, one- or

two-way ANOVA with Bonferroni’s multiple comparison testand used GraphPad Prism version 5.00 (GraphPad Software).Data are expressed as mean � SD, and P � .05 was assigned asstatistically significant.

Results

Impact of partial depletion of REA on E2-supported endometriosis-like lesion establishment

We investigated the role of REA in E2-supported lesionestablishment using different combinations of donor uter-ine tissue and recipient host background in which the levelof REA was WT (REA�/�) or reduced (REA�/�). We firstcompared in parallel donor uterine tissue with differentlevels of REA transplanted into the same recipient animal.(Because REA null uteri displayed a severely atrophic phe-notype [21], REA heterozygous [REA�/�] uteri were cho-sen for comparison with WT donor tissue with the fullcomplement of REA [REA�/�].) As shown in Figure 1, Aand B, after 2 weeks of E2 supplementation, both WT andREA�/� donor (D) uterine fragments, surgically trans-planted onto the peritoneal surface of ovariectomized WTrecipient (R) mice, were able to form endometriosis-like le-sions denoted as Dwt-Rwt and D�/�-Rwt, respectively. No-tably, the growth of D�/�-Rwt lesions was significantlygreater than Dwt-Rwt (Figure 1B), suggesting a restrainingrole for REA present in ectopic endometriosis-like lesiongrowth.

Although the heterozygous REA�/� transcript and pro-tein levels (�50%) were confirmed in REA�/� donor uter-ine tissue compared with WT uteri by qPCR (Figure 1C)and IHC (Figure 1D), it was noteworthy that D�/�-Rwt

ectopic lesions contained about 70% of the WT level ofREA mRNA (Figure 1C) and REA protein (Figure 1E),implying a contribution of REA from infiltrating hostcells. Indeed, as shown in Figure 1F, the presence of REAin host cells that infiltrate the ectopic lesion was validatedby cellular colocalization of REA and EGFP proteins inEGFP-transgenic host mice by dual immunofluorescenceanalysis when WT donor uterine tissue was transplantedinto EGFP recipients, and these recipients were given 2weeks of E2 treatment.

In order to evaluate the importance of host cell REA inectopic lesionprogression,weexamineddonortissuegrowthand phenotypic properties in E2-treated REA�/� recipientmice (Figure 2A). Although D�/�-Rwt ectopic tissues formedlarger lesions compared with Dwt-Rwt lesions (Figure 1B),WTlesionsestablished inREAheterozygousrecipients (Dwt-R�/�) displayed a growth rate similar to that of Dwt-Rwt

lesions (Figure 2B), implying that the proliferative signal in

902 Zhao et al REA in Endometriosis Endocrinology, February 2016, 157(2):900–912


http://press.endocrine.org/doi/suppl/10.1210/en.2015-1324/suppl_file/en-15-1324.pdf

http://press.endocrine.org/doi/suppl/10.1210/en.2015-1324/suppl_file/en-15-1324.pdf

lesions is affected largely by the REA content in the donortissue.

Because the host environment and cells that infiltrateinto the ectopic lesion from the host animal can contribute

to inflammatory aspects of the disease (17), we next ex-amined the expression of several cytokines known to behighly expressed in human endometriotic tissue (25–28)and regulated by E2 signaling in endometriosis (17). As

Donor UterineTissue

2 weeks ofE2 treatment

REA⁺/⁺ REA⁺/-

A

Lesi

on v

olum

e (m

m³) *

B

OVX Recipient (REA⁺/⁺)

Dwt-Rwt

Dwt-Rwt

Dwt-Rwt

Ectopic lesion

S

E

S

E

50μm

SE

D+/--Rwt Dwt-Rwt

IgGREAE

Dwt-Rwt D+/--RwtREA⁺/⁺ REA⁺/-

Ectopic lesion

a

b

a

c

mRN

A Fo

ld C

hang

e

REA

CC

C

D⁺/--R wt

D⁺/--Rwt

D

0

10

20

30

40

C

LE

S

GE LES

GELEREA

50μm

REA⁺/⁺ REA⁺/-

FREA GFP Merged

100μm

Figure 1. Ectopic lesions with a reduced level of REA show enhanced growth in E2-treated WT recipient mice. A, Endometriosis-like lesions wereestablished by transplanting donor (denoted by D) uterine fragments from REA WT (REA�/�, n � 6) and REA heterozygous (REA�/�, n � 6) miceof diestrous stage into ovariectomized (OVX) REA�/� recipient (denoted by R) mice, as described in Materials and Methods. B, Growth of ectopiclesions was monitored and lesion volume was quantified after 2 weeks of E2 treatment, P � .05 (paired t test). C, qPCR analysis of REA mRNA indonor uterine tissue and ectopic lesions after 2 weeks of E2 treatment. Levels of mRNA are expressed relative to the transcript level in REA�/�

donor uterine tissue which is set at 1.0. Different letters indicate P � .05 by one-way ANOVA with Bonferroni’s multiple comparison test. D, Donoruterine tissue from REA�/� and REA�/� mice at diestrous stage was subjected to IHC staining for REA. GE, glandular epithelium; LE, luminalepithelium; S, stromal tissue. E, IHC staining of REA in ectopic lesions after 2 weeks of E2 treatment. IgG served as negative control. C,endometriotic cyst; E, epithelial tissue; S, stromal tissue. F, Dual immunofluorescence of REA and GFP in ectopic lesions derived from WT donortissue transplanted into WT EGFP recipient mice treated with E2 for 2 weeks. D, donor; R, recipient.



shown in Figure 2D, when compared with Dwt-Rwt le-sions, levels of IL-6 (Il6), chemokine (C-C motif) ligand 2(Ccl2) and Ccl5 in Dwt-R�/� lesions were increased,whereas the expression of TNF� (Tnf�) remainedunchanged.

REA expressed in donor uterine tissue suppressesectopic cell proliferation and vascularization

To further investigate functions of donor tissue REA inlesion progression, we used a model in which both REAdonor WT (Dwt-Rwt) and REA donor heterozygous (D�/

�-Rwt) ectopic tissues were allowed to become establishedin the same WT intact recipient mice, and growth wasmonitored over 8 weeks. As seen in Figure 3A, D�/�-Rwt

ectopic lesions showed an enhanced growth rate after 2

weeks of lesion progression compared with that of Dwt-Rwt lesions. Notably, elevated ectopic lesion cell prolifer-ation was seen in 4-week D�/�-Rwt ectopic lesions vs Dwt-Rwt lesions by IHC analysis of the proliferation markerKi67 (Figure 3, B and C). By contrast, Ki67 staining anal-ysis of eutopic uterine WT REA�/� or REA�/� donor tis-sue indicated no difference in cell proliferation (Supple-mental Figure 1A), further suggesting the specific roles ofdonor REA level in lesion progression. Immunofluores-cence for the blood vessel marker, PECAM (Figure 3D),and quantification of PECAM-positive cells as an indica-tor of vascularization (Figure 3E) also documented thatD�/�-Rwt lesions were more highly vascularized com-pared with Dwt-Rwt ectopic lesions, but remained at a sim-

Ccl2 Ccl5 afnT6LI REA

*

*

**

A

mRN

A Fo

ld C

hang

e

2 weeks of E2 treatment

Dwt-Rwt Dwt-R+/-

Dwt-R+/-

Dwt-Rwt

OVX Recipient (REA⁺/-)

REA⁺/⁺

OVX Recipient (REA⁺/⁺)

C

Lesi

on v

olum

e (m

m³)

B

Dwt-Rwt Dwt-R+/-0

10

20

30

40

Figure 2. Ectopic lesions from E2-treated recipients with reduced level of REA. A, Donor uterine tissue from REA�/� mice (n � 12) wastransplanted into OVX REA�/� (n � 6) or REA�/� (n � 6) recipients, and assays were conducted after 2 weeks of E2 treatment in Dwt-Rwt and Dwt-R�/� ectopic lesions. B, Volume of ectopic lesions at 2 weeks of E2 treatment. Lesion volumes were not statistically different (unpaired t test). C,Il6, Ccl2, Ccl5, Tnf�, and REA expression levels in Dwt-Rwt and Dwt-R�/� ectopic tissues were analyzed by qPCR after 2 weeks of E2 treatment.mRNA levels are expressed relative to the transcript level in Dwt-Rwt lesions, which is set at 1.0. *, P � .05 (unpaired t test).



ilar level in eutopic tissue (data not shown). By contrast,cytokine production was similar in D�/�-Rwt lesionsand WT lesions (Figure 3F). Taken together, these find-ings suggest that growth and inflammatory signals orig-inate from distinct tissue loci, with the E2-supportedlesion growth and vascularization being largely con-trolled by REA gene dosage in the donor uterine tissue,whereas cytokine production in the ectopic lesion isprincipally determined by the host level of REA in in-filtrating cells.

Host REA restrains inflammation and innervationof ectopic endometriotic lesions

Next, both WT and REA�/� mice were used as recip-ients to interrogate the impact of REA in the host tissue onendometriotic lesions that developed from REA WT uter-ine tissue. Consistent with our observations for cytokineexpression in WT lesions from E2-treated heterozygousrecipients (Dwt-R�/�) (Figure 2C), higher Il6, Ccl2, Ccl5,and Tnf� mRNA levels were seen by qPCR analysis in WTlesions established in ovary intact REA�/� recipients(Dwt-R�/�) and monitored over 8 weeks of progression(Figure 4A). Immunostaining also demonstrated a greatlyincreased level of IL6 protein and p65 protein in Dwt-R�/�

lesions vs Dwt-Rwt lesions (Figure 4, B and C). The greater

than 2-fold increase in the number of nuclear p65-positivecells in Dwt-R�/� ectopic lesions supports stronger nuclearfactor kappa B activity in these lesions (Figure 4C). Im-munofluorescence for the T-cell marker CD3 and macro-phage marker F4/80 also revealed more immune cells inthe Dwt-R�/� lesions (Figure 4D). COX2 protein, anotherhallmark of endometriotic lesion-associated inflamma-tion (2), was also present at a higher level in Dwt-R�/� vsDwt-Rwt lesions (Figure 4E). Also notable was the in-creased innervation of Dwt-R�/� lesions, observed bystaining of the pan neuron marker, PGP9.5 (Figure 4E).Taken together, these data support suppressive roles ofREA in host cells and tissues in multiple inflammatoryresponses and in lesion innervation that accompanies le-sion progression. Notably, however, Dwt-R�/� ectopic tis-sues showed similar lesion growth rate (Figure 4F) andcell proliferation activity indicated by Ki67 staining(Supplemental Figure 1B) to that of Dwt-Rwt lesions,which is distinctly different from what was observed inD�/�-Rwt lesions. These findings highlight the impor-tance of REA status in donor uterine tissue on prolif-erative drive, in addition to distinctive contributionsfrom the host environment in the inflammatory aspectsof endometriosis.

*

*

Dwt-Rwt D+/--Rwt

Dwt-Rwt D+/--Rwt

ALe

sion

vol

ume

(mm

³)

DWeeks

*

*

B C

S

S

BV

BV

BV

BV

50μm

PECAM

Dwt-RwtD+/--Rwt

Ectopic lesion (4 weeks, intact recipient)

S

E

S

E

50μm

Ki67

D+/--RwtDwt-Rwt

Intact recipient


F

per

fiel

d

E

D+/--RwtDwt-Rwt

F

Ccl2 Ccl5IL6 Tnfa

mRN

A F

old

Chan

ge

Dwt-Rwt

D+/--Rwt

Figure 3. Impact of REA level in donor uterine tissue on lesion progression. Donor uterine tissue from REA WT (REA�/�, n � 6) and REAheterozygous (REA�/�, n � 6) mice were transplanted into intact WT recipient mice (n � 6) and were followed over 8 weeks for formation ofDwt-Rwt and D�/�-Rwt lesions, respectively. Both donor and recipient mice underwent transplantation surgery at the diestrous stage. A, Growth ofDwt-Rwt and D�/�-Rwt lesion volume over time was quantified as shown. *, P � .05 (two-way ANOVA with Bonferroni’s multiple comparison test).B, IHC staining for Ki67 in Dwt-Rwt and D�/�-Rwt lesions after 4 weeks of progression in intact WT recipients. C, Quantification of Ki67 stainingsignals in lesions. *, P � .05 (paired t test). D, PECAM staining of vasculature in ectopic lesions at 4 weeks of progression in intact WT recipients. E,Quantification of PECAM-positive cells at 4 weeks. *, P � .05 (paired t test). BV, blood vessel; C, endometriotic cyst; E, epithelial tissue;S, stromal tissue. F, Il6, Ccl2, Ccl5, and Tnf� mRNA levels in Dwt-Rwt and D�/�-Rwt ectopic tissues were analyzed by qPCR at 4 weeks of lesionprogression. Transcript levels are expressed relative to the transcript level in Dwt-Rwt lesions which is set at 1.0. No significant difference wasdetected.



Lesion growth and inflammation are mostincreased when REA level is reduced in bothdonor and recipient host tissues

To further understand the actions of REA in the crosstalk between donor and host tissues, REA�/� donor uter-

ine tissue was transplanted into heterozygous REA hostmice (D�/�-R�/�) and lesions were collected after 8weeks. Quantification of lesion volume (Figure 5A)showed that the reduced gene dosage of REA in both do-nor and host tissues (D�/�-R�/�) resulted in the greatest

*

Am

RNA

Fol

d Ch

ange

S

E

SC

EC

IL6

B

C

CD3

F4/80

S

S S

S

S

S S

SCOX2

F4/80

PGP9.5

D

E




D wt-Rwt

Dwt-R+/-

50 μm

100 μm

C CE E

S

S

p65

*

afnT5lcC2lcC6LI

**

** *

* *

Weeks0 2 4 6 80 2 4 6 80 2 4 6 80 2 4 6 8

0000

10

20

30

40

10

100

150

100

200

300

400

500

20

40

60

80

100

Weeks

F

Lesi

on v

olum

e (m

m³)

D wt-Rwt Dwt-R+/-

D wt-Rwt Dwt-R+/-

D wt-Rwt Dwt-R+/-

D wt-Rwt Dwt-R+/-

D wt-Rwt Dwt-R+/-

Dwt-R+/-

D wt-Rwt

D wt-Rwt Dwt-R+/-


IL6

p65

CD3 F4/80

Dwt-Rwt Dwt-R+/-Dwt-Rwt Dwt-R+/-

0

10

20

30

40

0

10

20

30* *

0

5

10

15

20

25

0

5

10

15

20COX2 PGP9.5

Dwt-Rwt Dwt-R+/- Dwt-Rwt Dwt-R+/-

**

* *

Figure 4. Impact of reduced host REA level on lesion progression. Donor uterine tissue from REA�/� mice at diestrous stage (n � 12) wasengrafted into intact REA�/� (n � 6) or REA�/� (n � 6) recipients at diestrus and followed over 8 weeks for lesion progression, forming Dwt-Rwt

and Dwt-R�/� implants, respectively. A, Il6, Ccl2, Ccl5, and Tnf� transcript levels in Dwt-Rwt and Dwt-R�/� ectopic tissues were profiled by qPCRover 8 weeks of lesion progression in intact recipients. mRNA levels are expressed relative to the transcript level in REA�/� eutopic donor tissue,which is set at 1.0. *, P � .05 (two-way ANOVA with Bonferroni’s multiple comparison test). IHC staining of (B) IL6 or (C) p65 in Dwt-Rwt and Dwt-R�/� ectopic lesions at 8 weeks. The IHC signals were quantified. *, P � .05 (unpaired t test). C, endometriotic cyst; E, epithelial tissue; S, stromaltissue. D, Immunofluorescence of immune cell markers CD3 and F4/80 and of (E) Cox2 as well as the pan-neuron marker PGP9.5 were performedand quantified (*, P � .05) (paired t test) in Dwt-Rwt and Dwt-R�/� ectopic lesions at 8 weeks. Quantitation of CD3, F4/80, COX2, and PGP9.5 areshown (at the right). F, Growth of WT lesions in intact WT (Dwt-Rwt, n � 6) and heterozygous (Dwt-R�/�, n � 6) recipients was monitored over 8weeks. No significant difference was detected (two-way ANOVA with Bonferroni’s multiple comparison test).



increase in lesion size, compared with donor REA�/� im-plants in recipient WT hosts (D�/�-Rwt), which werelarger than WT donor uterine implants grown in WT hosts(Dwt-Rwt) or heterozygous hosts (Dwt-R�/�). qPCR anal-ysis (Figure 5B) demonstrated significantly higher tran-script levels for most of the inflammation-associated cy-tokines examined in D�/�-R�/� lesions compared withDwt-Rwt and D�/�-Rwt lesions. In addition, althoughmRNA expression levels of ER� or ER� were similar inectopic tissue (Supplemental Figure 2A), IHC analysis ofPGR, which serves as an indicator of estrogen signalingactivity, showed enhanced PGR with reduced REA level in

the donor tissue (D�/�-Rwt) whichwas further elevated in D�/�-R�/�

lesions (Supplemental Figure 2B),demonstrating the restraining role ofREA in estrogen signaling during le-sion progression. Therefore, the en-hanced progression observed in D�/

�-R�/� lesions reveals that REA isnot only able to suppress multiplecharacteristic aspects of endometri-osis, but that it also critically con-tributes in estrogen signaling, andthe cross talk among the multiple celltypes in the donor uterine tissue andthe host background, whichare important in endometriosisprogression.

REA regulates proliferation ofhuman endometriotic stromalcells

To examine the functional signif-icance of REA in human endometri-otic cells from patient samples, weemployed siRNA knockdown ofREA in primary human endometri-otic stromal cells cultured in vitro.Treatment with siREA resulted ingreatly reduced levels of REAmRNA (Figure 6A) and protein(Supplemental Figure 3). When cellswere exposed to TNF� and E2 tomimic the in vivo hyperestrogenicand inflammatory microenviron-ment characteristic of endometriosis(29, 30), the mRNA levels of severalkey cell cycle regulators, such as cy-clin-dependent kinase 2 (CDK2), cy-clin B2 (CCNB2), cyclin D2(CCND2), and minichromosomemaintenance complex 2, were more

markedly elevated in siREA treated than in control siGL3-treated human endometriotic stromal cells, especially incells treated with E2 � TNF� (Figure 6A). By contrast, thelevel of ER� was not changed (Figure 6A). Increased cellproliferation was evident from immunofluorescence as-says of Ki67, P-H3, and by quantitation of cell numbers incells with knockdown of REA (Figure 6B). Thus, REAnormally acts as a suppressor of human endometriotic cellproliferation, reflected by enhanced expression of key cellcycle regulators and proliferation markers when REA wasreduced in the cells.

Il6

Ccl5

Ccl2

Tnfa

B

mR

NA

Fo

ld C

han

ge

a a

c

a abb

c

bb

a aa a

a

b

mR

NA

Fo

ld C

han

ge

Dwt-Rwt D+/--Rwt Dwt-R+/- D+/--R+/-

ALe

sion

vol

ume

(mm

³)

Dwt-Rwt Dwt-R+/-D+/--Rwt D+/--R+/-

a a

b

c

Dwt-Rwt D+/--Rwt Dwt-R+/- D+/--R+/-

0

20

40

60

80

100

0.0

0.5

1.0

1.5

2.0

0

5

10

15

0

1

2

3

4

0.0

0.5

1.0

1.5

2.0

2.5

Figure 5. Impact of reduced REA levels in both donor uterine tissue and host animals on lesionprogression. Donor tissues from REA�/� mice at diestrous stage (n � 6) were transplanted intointact REA�/� recipients (n � 6) to form D�/�-R�/� lesions. At 8 weeks after transplantation,quantification of (A) lesion volume and (B) cytokine mRNA levels by qPCR were monitored in Dwt-Rwt, D�/�-Rwt, Dwt-R�/�, and D�/�-R�/� lesions. Donors and recipients were intact animalschosen from the diestrous stage. Different letters indicate P � .05 by one-way ANOVA withBonferroni’s multiple comparison test.



REA is lower in human endometriotic tissue vsnormal human endometrium

As shown in Figure 7, we compared by IHC the presenceof REA protein in normal eutopic endometrium (n � 4women, ages 25–44) and in ectopic endometriosis samples

from women with the disease (n � 12 patients, ages 22–50).Quantitation of REA in tissue sections (3 tissue blocks perpatient and 6 fields quantitated per section from each block)revealed that REA was significantly lower in endometriosissamples compared with normal eutopic endometrium.

Cell

num

ber/

field

E2 + TNFα

siGL3 siREA siGL3 siREA

Veh

a a

b

c

A

B

mRN

A F

old

Chan

gem

RNA

Fol

d Ch

ange

siGL3 siREA siGL3 siREA siGL3 siREA siGL3 siREA siGL3 siREA siGL3 siREAVeh E2+TNF α Veh E2+TNF α Veh E2+TNFα

CDK2

MCM2REA

CCNB2 CCND2

siGL3 siREA siGL3 siREA siGL3 siREA siGL3 siREA siGL3 siREA siGL3 siREAVeh E2+TNF α Veh E2+TNF α Veh E2+TNFα

siGL3

siREA

αFNT + 2EheVKi67 + DAPI Ki67 + DAPI

E2 + TNF αsiGL3 siREA siGL3 siREA

Veh

Ki67

siGL3

siREA

P-H3 + DAPI P-H3 + DAPIαFNT + 2EheV

siGL3 siREA siGL3 siREA

Phospho-H3

E2 + TNFαVeh

20μm

20μm

a

b b

a

a

b b

c

c

a

ba,b

ab b

c

ab b

c

a a

b

c

a ab

c

a aaa

ERα

Figure 6. Impact of REA knockdown in human endometriotic stromal cells on cell proliferation and expression of cell cycle regulators. 24 hoursafter control siGL3 or siREA transfection, human endometriotic stromal cells were treated with E2 (10nM) and TNF� (20 ng/mL), or vehicle. A,qPCR analysis for the indicated RNAs after 24 hours of ligand or vehicle treatment. B, Immunofluorescence for Ki67 and P-H3 at 48 hours aftertreatment. DAPI staining indicates nuclei. Quantification of staining signals and measurement of cell numbers are shown (right panels). Differentletters indicate P � .05 by one-way ANOVA with Bonferroni’s multiple comparison test.



Discussion

Estrogen and inflammatory signaling, which are con-trolled by nuclear receptors and their coregulators, areessential for the survival of endometriotic tissue and fordisease progression (31, 32). Endometriotic tissue, likenormal uterine tissue, is reliant on estrogen, but endome-triosis is unique in that the endocrine milieu and hormonereceptor status of the endometriotic lesions are very dif-ferent from those in normal reproductive tissues. In par-ticular, estrogen production and ER regulation are alteredin endometriotic lesions. The ectopic tissue overexpressesaromatase and COX2 (2, 33), thereby causing continuouslocal production of estrogens and prostaglandins. Alsoendometriotic lesions have increased levels and increasedactivity of ERs which elicit a state of hyperstimulation (11,12, 34) that drives progression of the disease. Coregula-tors partner with ERs to control receptor activity and, inthis study, we have found that the corepressor, REA, func-tions as a restraint on ER to suppress the estrogen-stim-ulated proliferative drive of endometriotic lesions. Thus,when REA was reduced, it exacerbated and promotedpathologic progression of the disease.

Our observations highlight and support the existence ofextensive cross talk among various cell types that collab-orate to support the growth and phenotypic properties ofthe endometriotic lesions; these include the ectopic uterineendometrial cells and supporting cells from the immune,nervous and vascular systems (17, 35) that are found in-filtrating the ectopic lesions as they develop (1, 36). Ourstudies using host EGFP transgenic mice demonstratedinfiltration of host cells into the ectopic lesion and an en-hanced macrophage-monocyte complement in heterozy-

gous REA�/� host animals. Our findings indicate thatREA modulates this cross talk between donor uterine-de-rived cells and infiltrating host cells, and that reduction ofthe REA level in both donor and host tissues most greatlyaccelerates the growth and inflammatory signaling in en-dometriotic lesions.

Notably, as shown in the model in Figure 8, our find-ings suggest that the growth and inflammatory signals thatcontribute to endometriotic lesion progression originateprincipally from distinct tissue loci, with stimulatory in-puts from the ectopic uterine tissue primarily responsiblefor control of lesion proliferation and vascularization, andhost cells and tissues primarily responsible for control ofinflammation and neurogenesis in lesions. Our use of dif-ferent combinations of donor tissue and host back-grounds, that allow modulation of REA gene dosage ineach, enabled us to specify distinct stimulatory inputsfrom the ectopic uterine donor and host cells. However,the donor/recipient experimental observations also haverevealed that the donor and recipient tissues impact eachother in ways that influence the progression of endome-triosis, because lesion growth and inflammatory signalingwere greatest when both the uterine donor tissue and therecipient host mice were heterozygous for REA.

Of note, our studies in primary human endometrialstromal cells in which REA levels were experimentally re-duced revealed that REA normally restrains proliferationso that its depletion resulted in elevated proliferative ac-tivity and enhanced expression of cell cycle regulators.Furthermore, in clinical specimens, REA was found to besignificantly lower in human endometriotic lesions frompatients compared with normal endometrium. The find-

REA-Endometriosis samples

REA-Normal endometrium

50μm

REA

pos

ive

cel

ls/c

m2

Normalendometrium

Endometriosissamples

P = 0.0077

0

4000

3000

2000

1000

BA

Control IgG

CG CG CG

CG

SSS

SS

S

EG EG EG

Figure 7. REA protein by IHC in clinical human endometriosis samples and in human eutopic normal endometrium. Formalin-fixed and paraffin-embedded sections of endometriosis tissue or normal eutopic endometrium were examined for REA protein by IHC using REA antibody or IgG. A,Representative images are shown for REA staining in endometriosis samples from 3 patients and normal eutopic endometrium from 2 women notaffected by endometriosis. CG, cystic gland; EG, endometrial gland; S, stroma. B, Quantitation of REA in tissue samples. P � .0077 by unpaired ttest.



ings with clinical samples support observations made inour preclinical mouse endometriosis model and suggest acritical role for REA in the pathologic progression ofendometriosis.

Previous studies have documented that REA repressesER signaling (21) and exerts modulatory roles on path-ways controlling cell survival and metabolism consistentwith its name also as prohibitin 2 (18, 19, 37). REA hasbeen established as a key ER corepressor in the mammarygland and female reproductive tract, as well as in breastcancer cells (20–23). In the current study, we have high-lighted the pleiotropic ability of REA to suppress lesionprogression by modulating multiple aspects of estrogen-mediated signaling in endometriosis. The findings provideevidence that this coregulator acts as a restraint on ERactivities, repressing ER signaling that contributes to thepathologic molecular milieu in endometriotic lesions.Some proteins that function as coactivators of ER havealso been shown to impact endometriosis. For example,the coactivator steroid receptor coactivator-1 is cleaved byTNF�-activated matrix metallopeptidase 9 into a cyto-

plasmic 70-kDa shortened isoform, which notably pre-vents TNF�-mediated apoptosis in ectopic endometrioticcells (38). Also of interest, endometrial deficiency of thetranscription factor Krüppel-like factor 9, which acts as aregulator of ER� signaling, promoted endometriotic le-sion establishment and affected notch-, hedgehog-, andsteroid receptor-regulated pathways (39). Thus, both co-activators and corepressors appear to regulate key aspectsin the pathogenesis of endometriosis.

Endometriosis is associated with chronic inflammation(1–4) and in recent results from murine models, Burns etal reported that compared with WT lesions transplantedinto WT hosts, WT lesions were proliferative in ER�KOrecipient mice but showed decreased inflammatory re-sponses upon E2 treatment (14). Consistent with this, thenovel ER ligand, oxabicycloheptene sulfonate, with pref-erential affinity for ER�, lost its suppressive effects in WTectopic lesion-associated inflammatory responses inER�KO recipients (17) or in WT recipient mice depletedof macrophages with clodronate liposomes (17), suggest-ing critical roles of ER� and host myeloid responses.

REA(+/+)

REA(+/+)

ERE2

Endometriotic Cells

Infiltrating Host Cells

Proliferation

Vascularization

Neurogenesis

Inflammation

DiseaseProgression

REA(+/-)

REA(+/-)

ERE2

Endometriotic Cells

Infiltrating Host Cells

Proliferation

Vascularization

Neurogenesis

Inflammation

EnhancedDisease

Progression

Figure 8. Model depicting the cross talk and interrelationships between cells of the endometriotic lesion and host tissues, and the impact of REA.Our findings suggest that the growth and inflammatory signals that contribute to endometriotic lesion progression originate principally fromdistinct tissue loci, with stimulatory inputs from the ectopic uterine tissue primarily responsible for control of lesion proliferation andvascularization, and host cells and tissues primarily responsible for control of inflammation and neurogenesis in lesions. These influences werequeried using different combinations of donor tissue and host background that were either WT (full complement of REA) or heterozygous(expressing reduced REA levels). Further, donor and recipient tissues impact each other such that lesion growth and inflammatory signaling weregreatest when both the uterine donor tissue and the recipient host mice were heterozygous for REA.



In the current work, we have used an immune-intactsyngeneic murine model, in which the impacts of donorand host REA could be clearly distinguished and com-pared. We found that E2-supported chronic inflammatoryresponses in intact animals that mimic clinical findings.These include cytokine production, nuclear factor kappaB activation, Cox2 expression and immune cell infiltra-tion, all of which were elevated upon partial loss of hostbut not donor REA. By contrast, donor uterine tissue REAlevel was most important in the control of ectopic lesionproliferation and vascularization. These findings under-line the cellular and functional complexity of endometri-osis lesions and support separate contributions of the do-nor tissue and the host environment in the proliferativeand inflammatory aspects of endometriosis driven by theestrogen-ER axis.

Lesion innervation is thought to be involved in endo-metriosis-associated pain (40–42), and we showed pre-viously that treatment with dual antiestrogenic and anti-inflammatory compounds suppressed the innervation ofmurine endometriotic-like lesions (17). Interestingly, neu-roangiogenesis, a critical process driving the disease, hasbeen shown to be regulated by E2 signaling by Greaves etal (43). Moreover, E2 is not only able to stimulate mac-rophage infiltration into ectopic lesions but also to acti-vate interactions of macrophages and nerves, and thusmay exacerbate endometriosis-associated pain (44). Thecurrent findings of reduced REA increasing lesion inflam-matory signaling and nerve innervation suggest that byimpacting E2-ER signaling, host-derived REA might serveas a potential regulator of pain in endometriosis.

Current medical management of endometriosis pa-tients, which is primarily focused on suppressing E2 pro-duction (2), has not proven to be fully satisfactory. ERs,which are known to be essential and dysregulated in thepathogenesis of endometriosis, and their coregulatorsemerge as promising therapeutic targets. For example, ournovel ER ligands, oxabicycloheptene sulfonate andchloroindazole, displayed dual suppression of estrogenicand inflammatory activities and were effective in prevent-ing the establishment and progression of endometrioticlesions in mice (17). The selective ER modulators, baze-doxifene (15) and ERB-041 (16), have also been shown tosuppress endometriotic lesion growth. Because REA/pro-hibitin 2 suppresses the proliferation of human endometri-otic stromal cells and endometriosis-like lesion progres-sion in the preclinical mouse model, and is reduced inhuman endometriotic tissue compared with its level innormal human endometrium, it appears that maintenanceof adequate levels of REA may be important in preventingthe development of this disease. The clinical relevance ofour study is also highlighted by our observation that neu-

ron innervation, which may be involved in the chronicpelvic pain of endometriosis (40–42), was also suppressedby host REA. Therefore, our findings provide new insightsinto critical roles of coregulators in endometriosis, andimply that novel therapeutic approaches based on modu-lation of such coregulators might hold future potential forimproving medical care of women with this challengingdisease.

Acknowledgments

Address all correspondence and requests for reprints to: DrBenita S. Katzenellenbogen, Department of Molecular and In-tegrative Physiology, University of Illinois at Urbana-Cham-paign, Urbana, IL 61801. E-mail: [email protected].

This work was supported by the National Institutes of Health(NIH) Grant U54 HD055787 as part of the Eunice KennedyShriver National Institute of Child Health and Human Devel-opment/NIH Centers Program in Reproduction and InfertilityResearch (to B.S.K., M.K.B., and R.N.T.) and by the NIH GrantPHS 5R01DK015556 (to J.A.K.).

Disclosure Summary: The authors have nothing to disclose.

References

1. Asante A, Taylor RN. Endometriosis: the role of neuroangiogenesis.Annu Rev Physiol. 2011;73:163–182.

2. Bulun SE, Utsunomiya H, Lin Z, et al. Steroidogenic factor-1 andendometriosis. Mol Cell Endocrinol. 2009;300:104–108.

3. Giudice LC, Kao LC. Endometriosis. Lancet. 2004;364:1789–1799.

4. Taylor HS, Osteen KG, Bruner-Tran KL, et al. Novel therapies tar-geting endometriosis. Reprod Sci. 2011;18:814–823.

5. Calippe B, Douin-Echinard V, Delpy L, et al. 17�-estradiol pro-motes TLR4-triggered proinflammatory mediator productionthrough direct estrogen receptor � signaling in macrophages in vivo.J Immunol. 2010;185:1169–1176.

6. Lélu K, Laffont S, Delpy L, et al. Estrogen receptor � signaling in Tlymphocytes is required for estradiol-mediated inhibition of Th1and Th17 cell differentiation and protection against experimentalautoimmune encephalomyelitis. J Immunol. 2011;187:2386–2393.

7. Umetani M, Ghosh P, Ishikawa T, et al. The cholesterol metabolite27-hydroxycholesterol promotes atherosclerosis via proinflamma-tory processes mediated by estrogen receptor �. Cell Metab. 2014;20:172–182.

8. Missmer SA, Cramer DW. The epidemiology of endometriosis. Ob-stet Gynecol Clin North Am. 2003;30:1–19, vii.

9. Meldrum DR, Chang RJ, Lu J, Vale W, Rivier J, Judd HL. “Medicaloophorectomy” using a long-acting GNRH agonist–a possible newapproach to the treatment of endometriosis. J Clin EndocrinolMetab. 1982;54:1081–1083.

10. Noble LS, Simpson ER, Johns A, Bulun SE. Aromatase expressionin endometriosis. J Clin Endocrinol Metab. 1996;81:174–179.

11. Brandenberger AW, Lebovic DI, Tee MK, et al. Oestrogen receptor(ER)-� and ER-� isoforms in normal endometrial and endometri-osis-derived stromal cells. Mol Hum Reprod. 1999;5:651–655.

12. Xue Q, Lin Z, Cheng YH, et al. Promoter methylation regulatesestrogen receptor 2 in human endometrium and endometriosis. BiolReprod. 2007;77:681–687.



mailto:[email protected]

13. Grümmer R. Animal models in endometriosis research. Hum Re-prod Update. 2006;12:641–649.

14. Burns KA, Rodriguez KF, Hewitt SC, Janardhan KS, Young SL,Korach KS. Role of estrogen receptor signaling required for endo-metriosis-like lesion establishment in a mouse model. Endocrinol-ogy. 2012;153:3960–3971.

15. Kulak J Jr, Fischer C, Komm B, Taylor HS. Treatment with baze-doxifene, a selective estrogen receptor modulator, causes regressionof endometriosis in a mouse model. Endocrinology. 2011;152:3226–3232.

16. Harris HA, Bruner-Tran KL, Zhang X, Osteen KG, Lyttle CR. Aselective estrogen receptor-� agonist causes lesion regression in anexperimentally induced model of endometriosis. Hum Reprod.2005;20:936–941.

17. Zhao Y, Gong P, Chen Y, et al. Dual suppression of estrogenic andinflammatory activities for targeting of endometriosis. Sci TranslMed. 2015;7:271ra279.

18. Mishra S, Murphy LC, Murphy LJ. The Prohibitins: emerging rolesin diverse functions. J Cell Mol Med. 2006;10:353–363.

19. Thuaud F, Ribeiro N, Nebigil CG, Désaubry L. Prohibitin ligands incell death and survival: mode of action and therapeutic potential.Chem Biol. 2013;20:316–331.

20. Park SE, Xu J, Frolova A, Liao L, O’Malley BW, KatzenellenbogenBS. Genetic deletion of the repressor of estrogen receptor activity(REA) enhances the response to estrogen in target tissues in vivo.Mol Cell Biol. 2005;25:1989–1999.

21. Park S, Yoon S, Zhao Y, et al. Uterine development and fertility aredependent on gene dosage of the nuclear receptor coregulator REA.Endocrinology. 2012;153:3982–3994.

22. Park S, Zhao Y, Yoon S, et al. Repressor of estrogen receptor activity(REA) is essential for mammary gland morphogenesis and func-tional activities: studies in conditional knockout mice. Endocrinol-ogy. 2011;152:4336–4349.

23. Zhao Y, Park S, Bagchi MK, Taylor RN, Katzenellenbogen BS. Thecoregulator, repressor of estrogen receptor activity (REA), is a cru-cial regulator of the timing and magnitude of uterine decidualiza-tion. Endocrinology. 2013;154:1349–1360.

24. Zhao Y, Li Q, Katzenellenbogen BS, et al. Estrogen-induced CCN1is critical for establishment of endometriosis-like lesions in mice.Mol Endocrinol. 2014;28:1934–1947.

25. Akoum A, Lemay A, McColl S, Turcot-Lemay L, Maheux R. Ele-vated concentration and biologic activity of monocyte chemotacticprotein-1 in the peritoneal fluid of patients with endometriosis. Fer-til Steril. 1996;66:17–23.

26. Rong R, Ramachandran S, Santanam N, Murphy AA, Parthasara-thy S. Induction of monocyte chemotactic protein-1 in peritonealmesothelial and endometrial cells by oxidized low-density lipopro-tein and peritoneal fluid from women with endometriosis. FertilSteril. 2002;78:843–848.

27. Hornung D, Ryan IP, Chao VA, Vigne JL, Schriock ED, Taylor RN.Immunolocalization and regulation of the chemokine RANTES inhuman endometrial and endometriosis tissues and cells. J Clin En-docrinol Metab. 1997;82:1621–1628.

28. Bergqvist A, Bruse C, Carlberg M, Carlström K. Interleukin 1�,

interleukin-6, and tumor necrosis factor-� in endometriotic tissueand in endometrium. Fertil Steril. 2001;75:489–495.

29. Wieser F, Yu J, Park J, et al. A botanical extract from channel flowinhibits cell proliferation, induces apoptosis, and suppresses CCL5in human endometriotic stromal cells. Biol Reprod. 2009;81:371–377.

30. Sakamoto Y, Harada T, Horie S, et al. Tumor necrosis factor-�-induced interleukin-8 (IL-8) expression in endometriotic stromalcells, probably through nuclear factor-� B activation: gonadotropin-releasing hormone agonist treatment reduced IL-8 expression. J ClinEndocrinol Metab. 2003;88:730–735.

31. Han SJ, O’Malley BW. The dynamics of nuclear receptors and nu-clear receptor coregulators in the pathogenesis of endometriosis.Hum Reprod Update. 2014;20:467–484.

32. Lonard DM, Lanz RB, O’Malley BW. Nuclear receptor coregula-tors and human disease. Endocr Rev. 2007;28:575–587.

33. Attar E, Bulun SE. Aromatase and other steroidogenic genes in en-dometriosis: translational aspects. Hum Reprod Update. 2006;12:49–56.

34. Hayashi A, Tanabe A, Kawabe S, et al. Dienogest increases theprogesterone receptor isoform B/A ratio in patients with ovarianendometriosis. J Ovarian Res. 2012;5:31.

35. Hull ML, Escareno CR, Godsland JM, et al. Endometrial-peritonealinteractions during endometriotic lesion establishment. Am JPathol. 2008;173:700–715.

36. Beste MT, Pfaffle-Doyle N, Prentice EA, et al. Molecular networkanalysis of endometriosis reveals a role for c-Jun-regulated macro-phage activation. Sci Transl Med. 2014;6:222ra216.

37. Bavelloni A, Piazzi M, Raffini M, Faenza I, Blalock WL. Prohibitin2: at a communications crossroads. IUBMB Life. 2015;67:239–254.

38. Han SJ, Hawkins SM, Begum K, et al. A new isoform of steroidreceptor coactivator-1 is crucial for pathogenic progression of en-dometriosis. Nat Med. 2012;18:1102–1111.

39. Heard ME, Simmons CD, Simmen FA, Simmen RC. Kruppel-likefactor 9 deficiency in uterine endometrial cells promotes ectopiclesion establishment associated with activated notch and hedgehogsignaling in a mouse model of endometriosis. Endocrinology. 2014;155:1532–1546.

40. Anaf V, Simon P, El Nakadi I, et al. Hyperalgesia, nerve infiltrationand nerve growth factor expression in deep adenomyotic nodules,peritoneal and ovarian endometriosis. Hum Reprod. 2002;17:1895–1900.

41. Berkley KJ, Rapkin AJ, Papka RE. The pains of endometriosis. Sci-ence. 2005;308:1587–1589.

42. McAllister SL, McGinty KA, Resuehr D, Berkley KJ. Endometriosis-induced vaginal hyperalgesia in the rat: role of the ectopic growthsand their innervation. Pain. 2009;147:255–264.

43. Greaves E, Collins F, Esnal-Zufiaurre A, Giakoumelou S, HorneAW, Saunders PT. Estrogen receptor (ER) agonists differentiallyregulate neuroangiogenesis in peritoneal endometriosis via the re-pellent factor SLIT3. Endocrinology. 2014;155:4015–4026.

44. Greaves E, Temp J, Esnal-Zufiurre A, Mechsner S, Horne AW,Saunders PT. Estradiol is a critical mediator of macrophage-nervecross talk in peritoneal endometriosis. Am J Pathol. 2015;185:2286–2297.



rea in endometriosis_2015

Documents