activin a stimulates aromatase via the alk4-smad pathway

Research ArticleActivin A Stimulates Aromatase via the ALK4-SmadPathway in Endometriosis

Juan Zheng12 Juan Qu1 Pinhong Lu1 Zhen Hou1 Yugui Cui1 Yundong Mao1

Xiaochen Qi1 Hui Ji1 and Jiayin Liu1

1State Key Laboratory of Reproductive Medicine Center of Clinical Reproductive Medicine First Affiliated HospitalNanjing Medical University Nanjing 210029 China2Center of Reproductive Medicine 174th Hospital of PLA Xiamen 361000 China

Correspondence should be addressed to Zhen Hou zhhou1981126com

Received 21 May 2016 Accepted 17 August 2016

Academic Editor Limei Qiu

Copyright copy 2016 Juan Zheng et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Endometriosis is an estrogen-dependent disease We previously found that the expression of Activin A was upregulated in theperitoneal fluid of patients with endometriosisThe results of the present study indicated that Activin A induced estradiol secretionand P450arom expression in endometrial stromal cells (ESCs) derived from endometriosis patients The mechanism of estrogenicsynthesis was regulated by the Activin-Smad pathway in endometrial lesions The data showed that the effect of Activin A on ESCswas partially abrogated by pretreatment with an inhibitor of ALK4 (the type I receptor ActRIB) and Smad4-siRNA Cumulativelythese data suggest that Activin A promotes the secretion of estradiol from ESCs by increasing the expression of P450arom via theALK4-Smad pathway These findings indicate the ALK4-Smad pathway may promote ectopic lesion survival and development

1 Introduction

Endometriosis is one of the most common gynecologicaldisorders and affects 10 of reproductive-aged women Thedisease is defined by the presence of functional endometrialtissue outside the uterine cavity The tissue is primarilylocated on the pelvic peritoneum and ovariesThemain clini-cal features include chronic pelvic pain dyspareunia andinfertility [1] Endometriosis has been traditionally definedas an estrogen-responsive disease [2ndash5] In addition endo-metrial lesions display increased estradiol biosynthesis anddecreased estradiol inactivation compared to normal endo-metria [6 7]

Estrogen in the circulation of women suffering endo-metriosis arises from three sources First estradiol secretedby the ovary reaches endometrial tissues through circulationThe follicular rupture during each ovulation causes therelease of large amounts of estradiol directly onto pelvicimplants [8] Second the aromatase P450 that resides in adi-pose tissues and skin catalyzes the conversion of circulating

androstenedione to estrone [8] This circulating androstene-dione travels to endometrial tissues and is converted locallyto estradiol Third endometrial stromal cells may synthesizeE2de novo from cholesterol by the classical steroidogenic

pathway because they express all components required for thesteroidogenic cascade which includes P450arom for estrogenbiosynthesis Peripheral or local P450arom activity may beparticularly important in the persistence of endometriosis[8] A molecular link between inflammation and estro-gen production in endometrial tissues has been previouslyreported [7 9 10] The estradiol production is mediated by apositive feedback cycle that favors the overexpression of keysteroidogenic genes including P450arom which increasesthe local production of estradiol and promotes the secretionof transforming growth factor-beta (TGF-120573) and PGE

2in

endometrial tissues [7]Our previous studies assessed the expression of different

cytokines using cytokine arrays of peritoneal fluids takenfrom subjects with minimalmild endometriosis moder-atesevere endometriosis and controls [11] ActivinAwas one

Hindawi Publishing CorporationBioMed Research InternationalVolume 2016 Article ID 5791510 10 pageshttpdxdoiorg10115520165791510

2 BioMed Research International

of the upregulated cytokines in the women with endometrio-sis [11] It is a highly conserved member of the TGF-120573 super-family This ligand initiates signal transduction pathwaysthat are critical for reproductive functions and development[12] Activin A binds to the heteromeric receptor complexconsisting of a type I (ActRIA and ActRIB ALK4) and type II(ActRIIA andActRIIB) receptors [12]The type I receptors areactivated by phosphorylation and subsequently initiate theSmad signaling pathwayThe phosphorylated receptor Smads(R-Smads) bind to the common Smad (co-Smad Smad4) andenter the nucleus as a complex In the nucleus the Smadsignaling complex regulates the transcription of target genesthrough interactionswith transcriptional regulatory elements[12] Conversely the inhibitory Smads (Smad7) have beenshown to prevent the activation of the Smad signaling path-way by occupying ligand-activated type I receptor (ActRIBALK4) and interfering with the phosphorylation of R-SmadThe inhibitor Smads interfere with the activation of TGF-120573signaling [13ndash15]

Several studies have shown thatActivin signaling throughActivin receptor-like kinase 4 (ALK4) stimulates P450aromactivity in the ovarian granulosa cells and placental tro-phoblast cells [16 17] Furthermore Activin stimulatesCYP19A gene expression in human ovarian granulosa cell-like KGN cells via the Smad2 signaling pathway Activinsignaling through the ActRIB-Smad2 pathway plays a piv-otal role in CYP19A expression and follicular develop-ment [18] However little was known about the underly-ing mechanism of Activin signaling in the pathogenesis ofendometriosis We hypothesized that Activin A stimulatesendometrial stromal cells in endometriosis by inducingthe expression of P450arom via the ALK4-Smads signalingpathway The present study was designed to investigatewhether the Activin-Smads signaling pathway is functionallypresent and upregulated in the endometrial lesions Thisstudy also examined whether the activated Activin-Smadssignaling pathway can promote the synthesis of estrogenin endometrial stromal cells Furthering our understandingof the signaling mechanisms regulating estrogen biosyn-thesis in endometriotic lesions may lead to the discoveryof new diagnostic and therapeutic targets for endometrio-sis

2 Materials and Methods

21 Patient Selection and Characterization Sixty-five womenscheduled for laparoscopy for infertility were recruited toparticipate in this study after providing written informedconsent The study was approved by the Medical ResearchReview Board of First Affiliated Hospital Nanjing MedicalUniversity (Nanjing China) All of the patients had noactive pelvic inflammatory disease hydrosalpinges or anyautoimmune disease They had no history of endometrialhyperplasia or neoplasia The subjects had not received anyanti-inflammatory or hormonal medications or received ahysterosalpingography in the past threemonths All endome-trial biopsies were in the proliferative phase which is con-firmed with a pathological examination of the endometrialbiopsy by the method described previously [19]

The diagnosis of endometriosis was confirmed in 55women (mean plusmn SD of 301 plusmn 321 years of age) by the pres-ence of glands and stromas in the endometrial lesions Endo-metriosis was diagnosed and staged according to the revisedAmerican Society for Reproductive Medicine (rASRM) clas-sification by visual inspection of the pelvis [20] Thirtywomen had the minimalmild disease (AFS stages I-II 299plusmn34 years old) with red endometriotic lesions such as redvesicles red flame-like lesions or gland-like lesions Twenty-five women had the moderatesevere disease (AFS stages III-IV 305 plusmn 296 years old) Ten women (319 plusmn 19 yearsold) that received a laparoscopic examination for infertilitywithout endometriosis served as the control group Therewere no endometrial lesions active pelvic inflammatory dis-eases or hydrosalpinges observed via laparoscopy in the con-trol group

22 Preparation of Endometrial Stromal Cells The endome-trial tissues were obtained via dilation and curettage dur-ing surgery Immediately after collection small portions ofendometrial tissues were prepared for endometrial stromalcells Endometrial stromal cells (ESCs) were separated fromthe epithelial glands by digesting the tissue fragments withcollagenase as previously described [21] The isolated ESCswere cultured in Dulbecco modified Eagle medium F-12(DMEMF12 Sigma-Aldrich St Louis Missouri) mediumcontaining 10 fetal bovine serum (FBS Gibco Australia)until reaching confluence After 24 h of culturing the cellswere characterized using immunofluorescence microscopywith antibodies to cytokeratin 19 and vimentin as previouslydescribed [22] Vimentin staining of the sample preparationdemonstrated that the purity of the ESCs is more than 95ESCs were seeded in six-well culture plates before use inexperiments

23 Experimental Design ESCs from patients with endo-metriosis were cultured in DMEMF12 until 80ndash90 conflu-enceThen ESCswere treatedwith various concentrations (025 25 and 50 ngmL) of Activin A for different time (0 h 2 h12 h and 24 h) Activin A was purchased from RampD Systems(Minneapolis MN USA) Real-Time PCR and Western blotwere used to determine the expression of P450arom ESCsfrom normal endometrium and patients with endometriosiswere stimulated with Activin A (25 ngmL) or testosterone(100 ngmL) for 24 h After treatment the supernatants ofthe cell culture were collected and frozen in minus80∘C for thedetection of concentrations of estradiol

ESCs from patients with endometriosis were treated with25 ngmL Activin A for 5min 15min and 1 h Western blotand immunofluorescence were used to determine the expres-sion of p-Smad3 and Smad3

After pretreatment with an inhibitor of ALK4 (the typeI receptor ActRIB) and Smad4-siRNA real-time PCR andWestern blot were used to determine the change of P450aromexpression in ESCs from patients with endometriosis stimu-lated with 25 ngmL Activin A Estradiol secretion was alsodetected

Each study was repeated with three independent samplesand each experiment was performed in triplicate

BioMed Research International 3

24 Preparation of Cell Fraction The separation and prepa-ration of cytoplasmic and nuclear extracts from ESCs wereperformed using a commercial kit following the manu-facturerrsquos instructions (KeyGEN Biotech Nanjing China)In brief cytoplasmic extraction reagents (A and B) wereadded to a cell pellet to disrupt cell membranes to releasecytoplasmic contents The integrity of the nuclei extractionwas verified by the examination with a light microscope Thesamples were lysed with a nuclear extraction reagent to yieldthe nuclear fraction by centrifugation Immunoblot analysiswas used to assess the adequacy of nuclear purificationby measuring LaminAC (nuclear protein) and GAPDH(cytoplasmic protein) content The protein concentrationswere determined using the bicinchoninic acid (BCA) proteinassay kit (KeyGEN Biotech)

25 Real-Time PCR Total cellular RNA was isolated fromcells using Trizol (Invitrogen Corporation Carlsbad CAUSA) One microgram of total RNA was subjected to reversetranscription using PrimeScript RT-PCR Kit (Takara ShuzoCo Ltd Kyoto Japan) For the PCR reaction cDNAs wereused as templates to coamplify aromatase and GAPDHmRNA The mRNA abundance of target genes and thehousekeeping geneGAPDH was quantified by real-time PCRwith an ABI 7300 (Applied Biosystems Foster City CAUSA) using a Quanti Tect SYBR Green PCR Kit (TaKaRaShuzo Co Ltd Kyoto Japan) The quantitation of the targetgene expression was evaluated by the 2-step method (DeltaDelta Ct) [23] The GAPDH gene was used as the inter-nal standard to normalize the target gene level Sequencesfor CYP19A1 primers were GACCAATGAATCGGGCTA-TGT and TCTGTGGAAATCCTGCGTCTT Sequences forGAPDH primers were GGAGTCCACTGGCGTCTTCA andATGAGTCCTTCCACGATACCAA

26 Western Blot Analysis Thirty micrograms of each pro-tein sample were separated by SDS-PAGE and transferredonto nitrocellulose membranes (Bio-Rad Laboratories IncHercules CA USA) The membranes were blocked for 1 hin Tris-buffered saline containing 001 Tween 20 with 5nonfat dried milk The membranes were then incubatedovernight at 4∘C with the relevant antibodies The followingantibodies were used in this study monoclonal antiphospho-Smad3 (Ser423425) monoclonal anti-Smad3 and poly-clonal anti-Smad4 (Cell Signaling Technology Inc Bev-erly MA USA) polyclonal anti-P450arom (BioVision IncCalifornia USA) and polyclonal LaminAC and GAPDH(SantaCruz Biotechnology Inc SantaCruz CA USA) Themembranes were washed and then incubated with the sec-ondary peroxidase-conjugated anti-IgG (SantaCruz) anti-body for 1 h The immunoreactive proteins were detectedusing Enhanced Chemiluminescence reagents (MilliporeBillerica MA USA) followed by exposure to the detectionsystem (Alpha Innotech San Leandro CA USA) QuantityOne Analysis software (Bio-Rad Laboratories Inc) was usedto determine protein density levels

27 Measurement of Estradiol ESCs (2 times 105 viable cells)were cultured in 24-well plates After treatment the levels

of E2in the culture medium were measured by the electro-

chemiluminescence immunoassay (ECLIA Beckman Coul-ter Co California USA) according to the operation manualof the Beckman company The ECLIA kit is able to detectsteroid contents (E

2ge 10 pgmL) The intra- and interassay

coefficients of variation were 15 and 20 for E2assays

28 Immunofluorescence Cells from different treatmentgroups were washed with PBS three times and then fixedin 4 paraformaldehyde for 30min at room temperatureThe samples were subsequently incubated in permeabiliza-tion buffer (05 Triton X-100 in 20mM Hepes 3mMMgCl

2 50mM NaCl 300mM sucrose and 002 NaN

3)

for 30min at 37∘C The cells were then blocked in 5 BSAfor 1 h at room temperature The cells were washed andthe primary antibody to P-smad3 or smad3 was then added(1 100 dilution in PBS) the cells were then incubated for1ndash3 hours at 37∘C The slide was washed three times withPBS and then a fluorochrome conjugated second antibody(1 100 dilution in PBS) was added after which the cells wereincubated for 1 hour at 37∘CThe cells were then washed withPBS and counterstained with 410158406-diamidino-2-phenylindole(DAPI) (Vector Laboratories Inc Burlingame CA USA) inthe mounting medium All steps using fluorochromes wereperformed under appropriate light conditions The imageswere captured using a Leica fluorescence microscope (LeicaHeidelberg Germany)

29 Targeted Knockdown of Smad 4 Two HP GenomeWidesiRNA duplexes to Smad 4 (GenBank accession numberNM 005359) were purchased from Invitrogen (Invitrogen)Smad4-1 (51015840-UAUCCAUCAACAGUAACAAUAGGGC-31015840) and Smad4-2 (51015840-GCCCUAUUGUUACUGUUGAUG-GAUA-31015840) A control siRNA (Invitrogen) (51015840-AATTCT-CCGAACGTGTCACGT-31015840) was used in all experimentsThe Smad4 small RNA interference (RNAi) was namedSmad4-siRNA and the nonsense sequence to Smad4 wasnamed negative control siRNA (NC-siRNA) The cells weretransfected with siRNA duplexes using Lipofectamine 2000Transfection Reagent (Invitrogen) All experiments wereperformed in duplicates using cells in 6-well culture platesat 70 confluence Cells were cultured in vitro for 36 h andwashed twice with PBS Lipofectamine 2000 and siRNAwereseparately diluted in serum-free medium and incubated atroom temperature for 5 minutes They were then mixed andincubated at room temperature for additional 20 minutesAliquots of the transfection mixture were then added tocell culture dishes The cells were cultured in the transfec-tion media for 6 hours and then the culture medium wasreplaced with fresh Dulbecco modified Eagle medium F-12 (DMEMF12 Sigma-Aldrich St Louis Missouri USA)medium containing 10 fetal bovine serum (FBS GibcoAustralia) The cells were then cultured for 18 hours afterwhich they were treated with Activin A (25 ngmL) for 24 hThe conditioned medium was removed and analyzed byWestern blot for P450arom

210 Statistical Analysis All results in the study are shownas the mean plusmn standard deviation (SD) Each experiment


was performed with triplicate samples in each treatmentcondition Data distribution was evaluated by the Kolmorov-Smirnov normality test with Prism 4 computer software(GraphPad Software La Jolla CA USA) After the normalitytest confirmed that the values did not depart significantlyfrom the normal curve they were analyzed by one-wayanalysis of variance (ANOVA) and by Fisherrsquos least significantdifference (LSD) test for multiple comparisons using SPSS190 software (SPSS Inc Chicago Illinois USA) 119901 lt 005was considered statistically significant

3 Results

31 Effects of Activin A on mRNA Level Protein Expres-sion and Activity of P450arom in ESCs Endometriosis isan estrogen-dependent disease Therefore we evaluated theeffect of Activin A on the expression of P450arom a criticalenzyme for the local production of estrogens that drive thedevelopment of the disease [24] As shown in Figure 1we conducted time-course (2 h 12 h and 24 h) and dose-response (25 ngmL 25 ngmL and 50 ngmL) experimentsto determine the effect of Activin A on P450arom expressionin ESCs from patients with endometriosis As demonstratedin Figure 1 Activin A increased the P450arom mRNA (Fig-ures 1(a) and 1(b)) and protein (Figures 1(c) and 1(d)) levels ofESCs in a time- and dose-dependent manner The maximumincrease of mRNA level was observed at the concentrationof 25 ngmL for 12 h The maximum increase of protein levelwas observed at the concentration of 25 ngmL for 24 hThese optimized conditions were used in subsequent experi-ments

The concentrations of estradiol (E2) were measured toindirectly detect the P450arom activity ESCs from normalendometria and patients with endometriosis were stimulatedwith Activin A (25 ngmL) or testosterone (100 ngmL) at20∘C for 24 h and the concentrations of estradiol (E2) weremeasured As depicted in Figure 1(e) the concentrations ofestradiol in culture media from each treatment conditionwere detectable because the level was above the lower limits ofthe assay Estradiol secretion in ESCs of subjects with normalendometria was not increased following Activin A andortestosterone treatments However estradiol secretion in ESCsof patients with endometriosis was significantly increasedfollowing Activin A andor testosterone treatments (lowast119901 lt005 versus control lowastlowast119901 lt 001 versus control) Collectivelyour data show that Activin A increased the expression ofP450arom which is required for estradiol formation inESCs This elevation of estradiol following the P450aromincrease suggests Activin mediated signal transduction maybe involved in P450arom overexpression

32 Effects of Activin A on the Smad Signaling Pathway inEutopic Endometrial Stromal Cells (ESCs) from Patients withEndometriosis ESCs were treated with 25 ngmL of ActivinA and analyzed at multiple time points The cell lysates wereprobedwith antibodies against p-Smad3 Smad3 or GAPDHFigure 2(a) shows a representative Western blot of p-smad3and smad3 in cytoplasmic and nuclear fractions from threeindependent experiments ESCs were treated with Activin

A (25 ngmL) for 5 15 and 60min The treatments resultedin a significant increase of p-Smad3 in both the cytoplasmand nucleus The amount of p-Smad3 also increased aftertreatment withActivinA for 5minutes and p-Smad3 reacheda maximum at 15 minutes The p-Smad3 levels then beganto decrease after 1 hour As illustrated in Figure 2(a) Smad3expression was significantly increased as early as 15 minutesbut slightly decreased by 1 hour only in nucleus We foundthat Smad3 was constitutively expressed in the cytoplasm ofESCs However there is no significant difference in Smad3expression at different time points

Endometrial stromal cells express TGF-120573 receptors [2526] in the Smad signaling pathwayTherefore we determinedwhether Activin Amediated its action through the activationof this pathway The phosphorylation of Smad3 and its sub-sequent nuclear translocation are critical steps in the TGF-120573 signaling [27] We examined the effects of Activin A onp-Smad3 nuclear translocation and accumulation Our dataindicated that phosphorylated Smad3 content wasmaximal at15minTherefore this time point was used in the subsequentexperiments The immunofluorescence data in Figure 2(b)show the localization of Smad3 protein in ESCs from a repre-sentative patient with endometriosis Similar results were alsofound in other patients with endometriosis and all of thedata were consistent with Western blot results (Figure 2(a))Both p-Smad3 and Smad3 were primarily localized in thecytoplasm of untreated ESCs and it was suggested that thenuclear translocation was increased following Activin Atreatment for 15min These results suggested that the effectof Activin A was partially mediated through the activation ofthe Smad pathway in the endometrium

33 Effects of Activin A on Estradiol Secretion and AromataseExpression in ESCs from Endometriosis Patients In thecanonical Activin signaling pathway the Activin ligandsbind to a type II receptor serinetyrosine kinase which thenphosphorylates a type I receptor (ALK4) [12] The activatedALK4 then propagates intracellular signals which areclassicallymediated by Smad proteins SB431542 is a selectivepotent small molecule inhibitor of ALK4 ALK5 and ALK7[28] As shown in Figure 3(a) Activin A increased estradiolsecretion In addition the effect of Activin A was abolishedby pretreating the cells for 30 minutes with SB431542 Theeffect of Activin A on the estradiol secretion suggested thatsignaling via one or more of these receptors might contributeto high P450arom expression and P450arom enzymatic activ-ity

To investigate whether Activin A affects the expression ofP450arom through Smad proteins we pretreated ESCs withSB431542 to block ALK4 As shown in Figure 3(b) the treat-ment with SB431542 significantly inhibited the expression ofP450arom compared to Activin A treatment

Smad4 is the only comediating Smad and it is translo-cated to the nucleus with phosphorylated R-Smads In thenucleus the Smad complex subsequently modulates the tran-scription of Activin A target genes in mammals To block theendogenous Activin-Smad signaling we knocked down thelevel of Smad4 in ESCs by siRNA ESCs were transfected withSmad4-siRNA in the subsequent experimentThe expression


Rela

tive P

450

arom

mRN

A le

vels

25

20

15

10

05

00Control 2 12 24

Time (hrs)

lowast lowast

Real

-tim

e PCR

(a)

Rela

tive P

450

arom

mRN

A le

vels

25

20

15

10

05

00Control 25 25 50

Activin A (ngmL)

lowast

(b)

GAPDH(36kDa)

P450arom(55kDa)

Control 2 12 24

Control 2 12 24

Time (hrs)

lowast

Wes

tern

blo

tP450

arom

GA

PDH

15

10

05

00

(c)

P450arom(55kDa)

GAPDH(36kDa)

Control 25 25 50

Control 25 25 50

Activin A (ngmL)

lowast

P450

arom

GA

PDH

15

10

05

00

(d)

ControlActivin A

Testosterone

lowastlowast

200

150

100

50

0Normal Eutopic

lowastlowast

Estr

adio

l sec

retio

n(p

mol

L)

+ testosteroneActivin A(e)

Figure 1 Effects of Activin A on mRNA level protein expression and activity of P450arom in ESCs (a) Activin A time-course data (mean plusmnSD 119899 = 3) for P450arommRNA expression ESCs were treated with Activin A (25 ngmL) for 2 12 and 24 h and total RNAwas isolated fromuntreated control and treated ESC lowast119901 lt 005 versus control (b) Activin A dose-response data (mean plusmn SD) of P450arom mRNA expressionfrom three independent experiments performed in triplicate ESCs were treated with Activin A (25 25 and 50 ngmL) for 24 h and total RNAwas isolated from untreated control and treated ESC lowast119901 lt 005 versus control (c) Activin A time-course data (mean plusmn SD) for P450aromexpression from three independent experiments performed in triplicate The top panel of (c) shows representative data The bottom panel of(c) is the mean plusmn SD data from three independent experiments performed in triplicate ESCs were treated with Activin A (25 ngmL) for 212 or 24 hrs and total RNA was isolated from untreated control and treated ESC lowast119901 lt 005 versus control (d) Activin A dose-response data(mean plusmn SD) on P450arom expression from three independent experiments performed in triplicate The top panel of (d) is representativedata The bottom panel of (d) is the mean plusmn SD from three independent experiments performed in triplicate ESCs were treated with ActivinA (25 25 or 50 ngmL) for 24 h and total RNA was isolated from untreated control and treated ESC lowast119901 lt 005 versus control (e) ESCswere incubated in the absence (control) or presence of Activin A (25 ngmL) andor testosterone (100 ngmL) The estradiol concentrationsinmedia secreted by stromal cells weremeasured in four groupsThe data displayed in the figure show themean plusmn SD from three independentexperiments performed in triplicate lowast119901 lt 005 versus control lowastlowast119901 lt 001 versus control


(52kDa)

(52kDa)

(36GAPDH

kDa)

(70kDa)

Control 5min 15min 1h Control 5min 15min 1hCytoplasmic Nuclear

p-Smad3

Smad3

LaminAC

(a)

Control

Activin A

Control

Activin A

DAPI Merge

DAPI MergeSmad3

p-Smad3

(b)

Figure 2 Effects of Activin A on the Smad signaling pathway in eutopic endometrial stromal cells (ESCs) from patients with endometriosis(a) ESCs were treated with 25 ngmL Activin A for the indicated timesThe cell lysates were probed with antibodies against p-Smad3 Smad3or GAPDH A representativeWestern blot of p-smad3 and smad3 in cytoplasmic and nuclear fractions (b) Immunofluorescence localizationof Smad3 protein in eutopic endometrial stromal cells (ESCs) from a representative patient with endometriosis Similar results were observedin other patients with endometriosis DAPI stains the nuclei FITC staining for p-Smad or SmadsThe right column showsmerged FITCDAPIimages Control no Activin treatment Activin A the cells were treated with 25 ngmL of Activin A for 15 minutes The top 6 images displaythe p-Smad-3 expression in ESCs The bottom 6 images display Smad-3 expression in ESCs Bar = 5120583m


200

150

100

50

0A C B A + B

lowastlowast

Estr

adio

l sec

retio

n (p

mol

L)

(a)

A C B A + B

P450

arom

GA

PDH

20

15

10

05

00

lowastlowast

(55kDa)

GAPDH(36kDa)

A + BBCAP450arom

(b)

20

15

10

05

00

lowastlowast

Blank NC-siRNA

Blank NC-siRNA

(60kDa)

GAPDH(36kDa)

Smad4

Smad4-siRNA

Smad4-siRNA

Smad4

GA

PDH

(c)

lowastlowast

A + RRAC

200

150

100

50

0

Estr

adio

l sec

retio

n (p

mol

L)

R(minus)

(d)

RAC

A + R

A + R

RAC

P450arom(55kDa)

GAPDH(36kDa)

20

15

10

05

00

lowastlowast

R(minus)

R(minus)

P450

arom

GA

PDH

(e)

Figure 3 Effects of Activin A on estradiol secretion and aromatase expression in eutopic stromal cells (ESCs) from patients withendometriosis C or blank control A Activin A (25 ngmL) for 24 hr B SB431542 (10120583M) for 24 hr A + B 10120583M SB431542 for 30minfollowed by 25 ngmL Activin A for 24 hrs R smad4-siRNA R(minus) NC-siRNA A + R transfected with Smad4-siRNAThe data are expressedas mean plusmn SD from three independent experiments performed in triplicate lowastlowast119901 lt 001 versus control 119901 lt 005 versus Activin A 119901 lt 001versus Activin A (a)The estradiol level was measured in various conditions (b) P450arom expression was measured byWestern blot (the topis representative and the bottom is the mean plusmn SD from three independent experiments performed in triplicate) (c) ESCs were transfectedwith Smad4-siRNAThe silencing effect of Smad4 protein expression wasmeasured byWestern blotting (d)The estradiol level was measuredin various conditions (e) P450arom expression was measured by Western blot (top is a representative blot and the bottom is the mean plusmn SDfrom three independent experiments)


of Smad4 protein was analyzed by Western blot after trans-fection As described in Figure 3(c) the level of Smad4proteinwas decreased by approximately 71 by 48 hours aftertransfection with Smad4-siRNA The effect of Activin-Smadsignaling on steroidogenesis was then examined As shownin Figure 3(d) the transfection of Smad4-siRNA reducedthe production of E2 We knocked down the endogenousSmad4 using siRNA in ESCs to investigate whether ActivinA requires Smad signaling to regulate the expression ofP450arom Compared with Activin A treatment the knock-down of Smad4 significantly decreased the expression ofP450arom (Figure 3(e))

4 Discussion

P450arom may be involved in a positive feedback loopthat favors the expression of key steroidogenic genes inendometriosis [8] However the precise expression level ofP450arom in endometrium varies across published studies[29ndash36] Numerous reports have confirmed the expressionof P450arom in the endometrium at the mRNA transcrip-tional level [31 32 34 37] at the protein translational level[30 33 35] and in enzymatic activity [35 36] HoweverDelvoux et al [38] showed a complete absence of P450aromactivity in endometrial samples Moreover a recent studyreported that P450arom protein was not present in anytype of endometriosis and that the mRNA of P450aromwas barely detectable These results suggest that locallyproduced P450arom (within endometrial lesions) may beless important in endometriosis development [39] Consis-tent with previous results [30 40] our data demonstratedthat the P450arom expression was upregulated in both theeutopic endometrium and ectopic implants of women withendometriosis compared to those with normal endometria(Supplementary File in Supplementary Material availableonline at httpdxdoiorg10115520165791510)

Although the pathologic mechanism of endometriosisremains uncharacterized and is regulated by multiple signal-ing pathways Activin A-mediated activation of the Smad sig-naling may play a critical role in endometriosis As describedin Figure 2 Activin A strikingly activated the Smad pathwayand induced Smad3 activation and nuclear translocationin ESCs The treatment with either an inhibitor of ALK4kinase or knockdownof Smad4 strongly inhibited the Smad3-induced nuclear translocation and P450arom activity Pre-vious studies have shown that Smads are low-affinity DNA-binding proteins and that the complex of Smad234 mod-ulates gene expression by binding to Smad-binding elements(SBEs) SBEs are locatedwithin promoters containing numer-ous transcriptional coactivators and corepressors [8] Theconsensus binding sequences for Smad proteins are 5-GTCT-3 and its complement 5-AGAC-3 [41] Aberrant P450aromexpression in endometriosis is primarily mediated by theactivation of promoter II [24] There are several sequences ofSBEs in the promoter II of P450aromTherefore we proposedthat Activin A binds coactivators tomediate the transcriptionof P450arom in ESCs The molecular mechanism of ActivinA regulation of the P450arom expression in the nuclearcompartment will be studied in future investigations

Activin A has several important roles in cell proliferationinflammation immunity and fibrosis [42ndash45] Activin A canenhance the proliferation of stromal cells from endometri-omas Moreover a comprehensive review indicated thatActivin A played a complex role in controlling macrophageresponses and was predominantly proinflammatory whenacting early in an inflammatory response However therewere anti-inflammatory effects when acting on alreadyldquoinflamedrdquo cells [46] Ferreira et al have shown that TGF-120573exerts its actions in fibrosis by using Activin A as an inter-mediary [43] In addition Activin A increases matrix met-alloproteinase (MMP) activity [47 48] and promotes theinvasion of endometrial cells into peritoneum in vitro [43]A recent study has found that Activin A can stimulatehuman endometrial stromal cells to release interleukin-8 (IL-8) and vascular endothelial growth factor (VEGF) whichmay have implications for the pathogenesis of endometriosis[49] Therefore these results support several possible sites ofActivin action involved in the pathogenesis of endometrio-sis

Collectively Activin A promotes the pathogenesis ofendometriosis through at least three distinct processes FirstActivin A facilitates the process of endometrial cell invasioninto the peritoneum to form implants Second Activin Acan increase the secretion of estradiol by regulating theexpression of P450arom via the ALK4-Smads pathway inendometrial cells The combination of estrogen and dioxin(a highly toxic environmental contaminant derived fromsources of 245-trichlorophenol) is involved in the pathogen-esis of endometriosis by promoting chemokine secretion andinvasion of endometrial stromal cells [50] Third Activin Astimulates inflammation and increases tissue fibrosis Thusthe inflammatory status influences P450arom and steroidreceptor expression in endometriosis [30]

5 Conclusions

The present data suggest that the mechanism of estro-genic synthesis is regulated by the Activin-Smad pathwayin endometrial stromal cells The Activin-Smad pathwaymay play an important role in P450arom expression and inthe pathogenesis of endometriosis The present study maysupplement the estrogen-dependent theory of endometriosisand provides the experimental basis to explore new biologicaltargets for the treatment of endometriosis

Competing Interests

The authors declared no potential conflict of interests withrespect to the research authorship andor publication of thisarticle

Acknowledgments

This study was supported by the National Natural ScienceFoundation of China (Grant no 81200424) China 973 Pro-gram (2012CB944902) and Department of Health of JiangsuProvince for Clinical Reproductive Medicine (ZX201110BL2012009 and PAPD)


References

[1] L C Giudice and L C Kao ldquoEndometriosisrdquo The Lancet vol364 no 9447 pp 1789ndash1799 2004

[2] S E Bulun S Yang Z Fang B Gurates M Tamura andS Sebastian ldquoEstrogen production and metabolism in endo-metriosisrdquoAnnals of the New York Academy of Sciences vol 955no 1 pp 75ndash85 2002

[3] T L Rizner ldquoEstrogen metabolism and action in endometrio-sisrdquo Molecular and Cellular Endocrinology vol 307 no 1-2 pp8ndash18 2009

[4] K Huhtinen M Stahle A Perheentupa and M PoutanenldquoEstrogen biosynthesis and signaling in endometriosisrdquoMolec-ular and Cellular Endocrinology vol 358 no 2 pp 146ndash1542012

[5] W Xiong L Zhang L Yu et al ldquoEstradiol promotes cells inva-sion by activating 120573-catenin signaling pathway in endometrio-sisrdquo Reproduction vol 150 no 6 pp 507ndash516 2015

[6] S E Bulun Y-H Cheng P Yin et al ldquoProgesterone resistancein endometriosis link to failure to metabolize estradiolrdquoMolec-ular and Cellular Endocrinology vol 248 no 1-2 pp 94ndash1032006

[7] Y Tremblay G E Ringler Y Morel et al ldquoRegulation of thegene for estrogenic 17-ketosteroid reductase lying on chromo-some 17cenrarrq25rdquo Journal of Biological Chemistry vol 264 no34 pp 20458ndash20462 1989

[8] S E Bulun ldquoEndometriosisrdquoTheNew England Journal of Medi-cine vol 360 no 3 pp 268ndash279 2009

[9] I Gori C Pellegrini D Staedler R Russell C Jan and G OCanny ldquoTumor necrosis factor-120572 activates estrogen signalingpathways in endometrial epithelial cells via estrogen receptor120572rdquo Molecular and Cellular Endocrinology vol 345 no 1-2 pp27ndash37 2011

[10] L Zhang W Xiong Y Xiong H Liu and Y Liu ldquo17 120573-Estra-diol promotes vascular endothelial growth factor expression viathe Wnt120573-catenin pathway during the pathogenesis of endo-metriosisrdquo Molecular Human Reproduction vol 22 no 7 pp526ndash535 2016

[11] Z Hou L Sun L Gao L Liao Y Mao and J Liu ldquoCyto-kine array analysis of peritoneal fluid betweenwomenwith end-ometriosis of different stages and those without endometriosisrdquoBiomarkers vol 14 no 8 pp 604ndash618 2009

[12] J Massague ldquoTGF-120573 signal transductionrdquo Annual Review ofBiochemistry vol 67 no 1 pp 753ndash791 1998

[13] HHayashi S Abdollah YQiu et al ldquoTheMAD-related proteinSmad7 associates with the TGF120573 receptor and functions as anantagonist of TGF120573 signalingrdquo Cell vol 89 no 7 pp 1165ndash11731997

[14] T Imamura M Takase A Nishihara et al ldquoSmad6 inhibitssignalling by the TGF-120573 superfamilyrdquoNature vol 389 no 6651pp 622ndash626 1997

[15] A Nakao M Afrakhte A Moren et al ldquoIdentification ofSmad7 a TGF120573-inducible antagonist of TGF-120573-signallingrdquoNature vol 389 no 6651 pp 631ndash635 1997

[16] Y Song J Keelan and J T France ldquoActivin-A stimulates whiletransforming growth factor 1205731 inhibits chorionic gonado-trophin production and aromatase activity in cultured humanplacental trophoblastsrdquo Placenta vol 17 no 8 pp 603ndash6101996

[17] C Mukasa M Nomura T Tanaka et al ldquoActivin signalingthrough type IB activin receptor stimulates aromatase activity

in the ovarian granulosa cell-like human granulosa (KGN)cellsrdquo Endocrinology vol 144 no 4 pp 1603ndash1611 2003

[18] M Nomura R Sakamoto H Morinaga L Wang C Mukasaand R Takayanagi ldquoActivin stimulates CYP19A gene expressionin human ovarian granulosa cell-like KGN cells via the Smad2signaling pathwayrdquo Biochemical and Biophysical Research Com-munications vol 436 no 3 pp 443ndash448 2013

[19] R W Noyes A T Hertig and J Rock ldquoDating the endometrialbiopsyrdquoAmerican Journal of Obstetrics and Gynecology vol 122no 2 pp 262ndash263 1975

[20] M Canis J G Donnez D S Guzick et al ldquoRevised Ameri-can Society for Reproductive Medicine classification of endo-metriosis 1996rdquo Fertility and Sterility vol 67 no 5 pp 817ndash8211997

[21] Z Hou J Zhou X Ma L Fan L Liao and J Liu ldquoRoleof interleukin-1 receptor type II in the pathogenesis of endo-metriosisrdquo Fertility and Sterility vol 89 no 1 pp 42ndash51 2008

[22] X Luo J Xu and N Chegini ldquoThe expression of smads inhuman endometrium and regulation and induction in endome-trial epithelial and stromal cells by transforming growth factor-120573rdquo Journal of Clinical Endocrinology andMetabolism vol 88 no10 pp 4967ndash4976 2003

[23] K J Livak and T D Schmittgen ldquoAnalysis of relative geneexpression data using real-time quantitative PCRand the 2minusΔΔ119862TmethodrdquoMethods vol 25 no 4 pp 402ndash408 2001

[24] E Attar and S E Bulun ldquoAromatase and other steroidogenicgenes in endometriosis translational aspectsrdquo Human Repro-duction Update vol 12 no 1 pp 49ndash56 2006

[25] N Chegini Y Zhao R SWilliams andK C Flanders ldquoHumanuterine tissue throughout the menstrual cycle expresses trans-forming growth factor-1205731 (TGF1205731) TGF1205732 TGF1205733 and TGF120573Type II receptor messenger ribonucleic acid and protein andcontains [125I]TGF1205731-binding sitesrdquoEndocrinology vol 135 no1 pp 439ndash449 1994

[26] T Sandberg B Casslen B Gustavsson and T J BenraadldquoHuman endothelial cell migration is stimulated by urokinaseplasminogen activator plasminogen activator inhibitor 1 com-plex released from endometrial stromal cells stimulated withtransforming growth factor 1205731 possible mechanism for para-crine stimulation of endometrial angiogenesisrdquo Biology ofReproduction vol 59 no 4 pp 759ndash767 1998

[27] S M Margarit H Sondermann B E Hall et al ldquoStructuralevidence for feedback activation by Ras∙GTP of the Ras-specificnucleotide exchange factor SOSrdquo Cell vol 112 no 5 pp 685ndash695 2003

[28] N J Laping E Grygielko A Mathur et al ldquoInhibition of trans-forming growth factor (TGF)-1205731-induced extracellular matrixwith a novel inhibitor of the TGF-120573 type I receptor kinaseactivity SB-431542rdquoMolecular Pharmacology vol 62 no 1 pp58ndash64 2002

[29] J Kitawaki T Noguchi T Amatsu et al ldquoExpression of aroma-tase cytochrome P450 protein and messenger ribonucleic acidin human endometriotic and adenomyotic tissues but not innormal endometriumrdquo Biology of Reproduction vol 57 no 3pp 514ndash519 1997

[30] O Bukulmez D B Hardy B R Carr R A Word and CR Mendelson ldquoInflammatory status influences aromatase andsteroid receptor expression in endometriosisrdquo Endocrinologyvol 149 no 3 pp 1190ndash1204 2008

[31] H Dassen C Punyadeera R Kamps et al ldquoEstrogen metab-olizing enzymes in endometrium and endometriosisrdquo HumanReproduction vol 22 no 12 pp 3148ndash3158 2007


[32] J-F Heilier J Donnez S Defrere V Van Kerckhove ODonnez andD Lison ldquoSerum dioxin-like compounds and aro-matase (CYP19) expression in endometriotic tissuesrdquo Toxicol-ogy Letters vol 167 no 3 pp 238ndash244 2006

[33] G Hudelist K Czerwenka J Keckstein et al ldquoExpression ofaromatase and estrogen sulfotransferase in eutopic and ectopicendometrium evidence for unbalanced estradiol production inendometriosisrdquo Reproductive Sciences vol 14 no 8 pp 798ndash805 2007

[34] H Ishihara J Kitawaki N Kado H Koshiba S Fushiki andHHonjo ldquoGonadotropin-releasing hormone agonist and danazolnormalize aromatase cytochrome P450 expression in eutopicendometrium from women with endometriosis adenomyosisor leiomyomasrdquo Fertility and Sterility vol 79 no 3 pp 735ndash7422003

[35] K Murakami K Nomura K Shinohara T Kasai M Shozuand M Inoue ldquoDanazol inhibits aromatase activity of endo-metriosis-derived stromal cells by a competitive mechanismrdquoFertility and Sterility vol 86 no 2 pp 291ndash297 2006

[36] A Purohit L Fusi J Brosens L W L Woo B V L Potter andM J Reed ldquoInhibition of steroid sulphatase activity in endo-metriotic implants by 667COUMATE a potential new therapyrdquoHuman Reproduction vol 23 no 2 pp 290ndash297 2008

[37] C A Piccinato R M Neme N Torres et al ldquoIncreased expres-sion of CYP1A1 and CYP1B1 in ovarianperitoneal endometri-otic lesionsrdquo Reproduction vol 151 no 6 pp 683ndash692 2016

[38] B Delvoux P Groothuis T DrsquoHooghe C Kyama G Dunsel-man and A Romano ldquoIncreased production of 17120573-estradiolin endometriosis lesions is the result of impaired metabolismrdquoThe Journal of Clinical Endocrinology ampMetabolism vol 94 no3 pp 876ndash883 2009

[39] S Colette J C Lousse S Defrere et al ldquoAbsence of aromataseprotein and mRNA expression in endometriosisrdquo HumanReproduction vol 24 no 9 pp 2133ndash2141 2009

[40] S E Bulun Z Lin G Imir et al ldquoRegulation of aromataseexpression in estrogen-responsive breast and uterine diseasefrom bench to treatmentrdquo Pharmacological Reviews vol 57 no3 pp 359ndash383 2005

[41] T Iempridee S Das I Xu and J E Mertz ldquoTransforminggrowth factor 120573-induced reactivation of epstein-barr virusinvolves multiple smad-binding elements cooperatively activat-ing expression of the latent-lytic switch BZLF1 generdquo Journal ofVirology vol 85 no 15 pp 7836ndash7848 2011

[42] O Yoshino G Izumi J Shi et al ldquoActivin-A is inducedby interleukin-1120573 and tumor necrosis factor-120572 and enhancesthe mRNA expression of interleukin-6 and protease-activatedreceptor-2 and proliferation of stromal cells from endometri-omardquo Fertility and Sterility vol 96 no 1 pp 118ndash121 2011

[43] M C Ferreira C A Witz L S Hammes et al ldquoActivin Aincreases invasiveness of endometrial cells in an in vitro modelof human peritoneumrdquoMolecular Human Reproduction vol 14no 5 pp 301ndash307 2008

[44] D M de Kretser R E OrsquoHehir C L Hardy and M P HedgerldquoThe roles of activin A and its binding protein follistatin ininflammation and tissue repairrdquo Molecular and Cellular Endo-crinology vol 359 no 1-2 pp 101ndash106 2012

[45] F M Reis S Luisi M S Abro et al ldquoDiagnostic value of serumactivin A and follistatin levels in women with peritoneal ovar-ian and deep infiltrating endometriosisrdquo Human Reproductionvol 27 no 5 pp 1445ndash1450 2012

[46] M P Hedger W R Winnall D J Phillips and D M deKretser ldquoThe regulation and functions of activin and follistatin

in inflammation and immunityrdquo Vitamins and Hormones vol85 pp 255ndash297 2011

[47] K Yoshinaga K Mimori H Inoue et al ldquoActivin A enhancesMMP-7 activity via the transcription factor AP-1 in an eso-phageal squamous cell carcinoma cell linerdquo International Jour-nal of Oncology vol 33 no 3 pp 453ndash459 2008

[48] R L Jones J K Findlay P G Farnworth D M Robertson EWallace and L A Salamonsen ldquoActivin A and inhibin a dif-ferentially regulate human uterine matrix metalloproteinasespotential interactions during decidualization and trophoblastinvasionrdquo Endocrinology vol 147 no 2 pp 724ndash732 2006

[49] A L L Rocha P Carrarelli R Novembri et al ldquoActivin A stim-ulates interleukin 8 and vascular endothelial growth factorrelease from cultured human endometrial stromal cells possi-ble implications for the pathogenesis of endometriosisrdquo Repro-ductive Sciences vol 19 no 8 pp 832ndash838 2012

[50] J Yu Y Wang W-H Zhou L Wang Y-Y He and D-JLi ldquoCombination of estrogen and dioxin is involved in thepathogenesis of endometriosis by promoting chemokine secre-tion and invasion of endometrial stromal cellsrdquo Human Repro-duction vol 23 no 7 pp 1614ndash1626 2008


of the upregulated cytokines in the women with endometrio-sis [11] It is a highly conserved member of the TGF-120573 super-family This ligand initiates signal transduction pathwaysthat are critical for reproductive functions and development[12] Activin A binds to the heteromeric receptor complexconsisting of a type I (ActRIA and ActRIB ALK4) and type II(ActRIIA andActRIIB) receptors [12]The type I receptors areactivated by phosphorylation and subsequently initiate theSmad signaling pathwayThe phosphorylated receptor Smads(R-Smads) bind to the common Smad (co-Smad Smad4) andenter the nucleus as a complex In the nucleus the Smadsignaling complex regulates the transcription of target genesthrough interactionswith transcriptional regulatory elements[12] Conversely the inhibitory Smads (Smad7) have beenshown to prevent the activation of the Smad signaling path-way by occupying ligand-activated type I receptor (ActRIBALK4) and interfering with the phosphorylation of R-SmadThe inhibitor Smads interfere with the activation of TGF-120573signaling [13ndash15]

Several studies have shown thatActivin signaling throughActivin receptor-like kinase 4 (ALK4) stimulates P450aromactivity in the ovarian granulosa cells and placental tro-phoblast cells [16 17] Furthermore Activin stimulatesCYP19A gene expression in human ovarian granulosa cell-like KGN cells via the Smad2 signaling pathway Activinsignaling through the ActRIB-Smad2 pathway plays a piv-otal role in CYP19A expression and follicular develop-ment [18] However little was known about the underly-ing mechanism of Activin signaling in the pathogenesis ofendometriosis We hypothesized that Activin A stimulatesendometrial stromal cells in endometriosis by inducingthe expression of P450arom via the ALK4-Smads signalingpathway The present study was designed to investigatewhether the Activin-Smads signaling pathway is functionallypresent and upregulated in the endometrial lesions Thisstudy also examined whether the activated Activin-Smadssignaling pathway can promote the synthesis of estrogenin endometrial stromal cells Furthering our understandingof the signaling mechanisms regulating estrogen biosyn-thesis in endometriotic lesions may lead to the discoveryof new diagnostic and therapeutic targets for endometrio-sis

2 Materials and Methods

21 Patient Selection and Characterization Sixty-five womenscheduled for laparoscopy for infertility were recruited toparticipate in this study after providing written informedconsent The study was approved by the Medical ResearchReview Board of First Affiliated Hospital Nanjing MedicalUniversity (Nanjing China) All of the patients had noactive pelvic inflammatory disease hydrosalpinges or anyautoimmune disease They had no history of endometrialhyperplasia or neoplasia The subjects had not received anyanti-inflammatory or hormonal medications or received ahysterosalpingography in the past threemonths All endome-trial biopsies were in the proliferative phase which is con-firmed with a pathological examination of the endometrialbiopsy by the method described previously [19]

The diagnosis of endometriosis was confirmed in 55women (mean plusmn SD of 301 plusmn 321 years of age) by the pres-ence of glands and stromas in the endometrial lesions Endo-metriosis was diagnosed and staged according to the revisedAmerican Society for Reproductive Medicine (rASRM) clas-sification by visual inspection of the pelvis [20] Thirtywomen had the minimalmild disease (AFS stages I-II 299plusmn34 years old) with red endometriotic lesions such as redvesicles red flame-like lesions or gland-like lesions Twenty-five women had the moderatesevere disease (AFS stages III-IV 305 plusmn 296 years old) Ten women (319 plusmn 19 yearsold) that received a laparoscopic examination for infertilitywithout endometriosis served as the control group Therewere no endometrial lesions active pelvic inflammatory dis-eases or hydrosalpinges observed via laparoscopy in the con-trol group

22 Preparation of Endometrial Stromal Cells The endome-trial tissues were obtained via dilation and curettage dur-ing surgery Immediately after collection small portions ofendometrial tissues were prepared for endometrial stromalcells Endometrial stromal cells (ESCs) were separated fromthe epithelial glands by digesting the tissue fragments withcollagenase as previously described [21] The isolated ESCswere cultured in Dulbecco modified Eagle medium F-12(DMEMF12 Sigma-Aldrich St Louis Missouri) mediumcontaining 10 fetal bovine serum (FBS Gibco Australia)until reaching confluence After 24 h of culturing the cellswere characterized using immunofluorescence microscopywith antibodies to cytokeratin 19 and vimentin as previouslydescribed [22] Vimentin staining of the sample preparationdemonstrated that the purity of the ESCs is more than 95ESCs were seeded in six-well culture plates before use inexperiments

23 Experimental Design ESCs from patients with endo-metriosis were cultured in DMEMF12 until 80ndash90 conflu-enceThen ESCswere treatedwith various concentrations (025 25 and 50 ngmL) of Activin A for different time (0 h 2 h12 h and 24 h) Activin A was purchased from RampD Systems(Minneapolis MN USA) Real-Time PCR and Western blotwere used to determine the expression of P450arom ESCsfrom normal endometrium and patients with endometriosiswere stimulated with Activin A (25 ngmL) or testosterone(100 ngmL) for 24 h After treatment the supernatants ofthe cell culture were collected and frozen in minus80∘C for thedetection of concentrations of estradiol

ESCs from patients with endometriosis were treated with25 ngmL Activin A for 5min 15min and 1 h Western blotand immunofluorescence were used to determine the expres-sion of p-Smad3 and Smad3

After pretreatment with an inhibitor of ALK4 (the typeI receptor ActRIB) and Smad4-siRNA real-time PCR andWestern blot were used to determine the change of P450aromexpression in ESCs from patients with endometriosis stimu-lated with 25 ngmL Activin A Estradiol secretion was alsodetected

Each study was repeated with three independent samplesand each experiment was performed in triplicate












3)






3 Results










Rela

tive P

450

arom

mRN

A le

vels

25

20

15

10

05

00Control 2 12 24

Time (hrs)

lowast lowast

Real

-tim

e PCR

(a)

Rela

tive P

450

arom

mRN

A le

vels

25

20

15

10

05

00Control 25 25 50

Activin A (ngmL)

lowast

(b)

GAPDH(36kDa)

P450arom(55kDa)

Control 2 12 24

Control 2 12 24

Time (hrs)

lowast

Wes

tern

blo

tP450

arom

GA

PDH

15

10

05

00

(c)

P450arom(55kDa)

GAPDH(36kDa)

Control 25 25 50

Control 25 25 50

Activin A (ngmL)

lowast

P450

arom

GA

PDH

15

10

05

00

(d)

ControlActivin A

Testosterone

lowastlowast

200

150

100

50

0Normal Eutopic

lowastlowast

Estr

adio

l sec

retio

n(p

mol

L)




(52kDa)

(52kDa)

(36GAPDH

kDa)

(70kDa)


p-Smad3

Smad3

LaminAC

(a)

Control

Activin A

Control

Activin A

DAPI Merge

DAPI MergeSmad3

p-Smad3

(b)



200

150

100

50

0A C B A + B

lowastlowast

Estr

adio

l sec

retio

n (p

mol

L)

(a)

A C B A + B

P450

arom

GA

PDH

20

15

10

05

00

lowastlowast

(55kDa)

GAPDH(36kDa)

A + BBCAP450arom

(b)

20

15

10

05

00

lowastlowast

Blank NC-siRNA

Blank NC-siRNA

(60kDa)

GAPDH(36kDa)

Smad4

Smad4-siRNA

Smad4-siRNA

Smad4

GA

PDH

(c)

lowastlowast

A + RRAC

200

150

100

50

0

Estr

adio

l sec

retio

n (p

mol

L)

R(minus)

(d)

RAC

A + R

A + R

RAC

P450arom(55kDa)

GAPDH(36kDa)

20

15

10

05

00

lowastlowast

R(minus)

R(minus)

P450

arom

GA

PDH

(e)




4 Discussion





5 Conclusions


Competing Interests


Acknowledgments



References

































































3)






3 Results










Rela

tive P

450

arom

mRN

A le

vels

25

20

15

10

05

00Control 2 12 24

Time (hrs)

lowast lowast

Real

-tim

e PCR

(a)

Rela

tive P

450

arom

mRN

A le

vels

25

20

15

10

05

00Control 25 25 50

Activin A (ngmL)

lowast

(b)

GAPDH(36kDa)

P450arom(55kDa)

Control 2 12 24

Control 2 12 24

Time (hrs)

lowast

Wes

tern

blo

tP450

arom

GA

PDH

15

10

05

00

(c)

P450arom(55kDa)

GAPDH(36kDa)

Control 25 25 50

Control 25 25 50

Activin A (ngmL)

lowast

P450

arom

GA

PDH

15

10

05

00

(d)

ControlActivin A

Testosterone

lowastlowast

200

150

100

50

0Normal Eutopic

lowastlowast

Estr

adio

l sec

retio

n(p

mol

L)




(52kDa)

(52kDa)

(36GAPDH

kDa)

(70kDa)


p-Smad3

Smad3

LaminAC

(a)

Control

Activin A

Control

Activin A

DAPI Merge

DAPI MergeSmad3

p-Smad3

(b)



200

150

100

50

0A C B A + B

lowastlowast

Estr

adio

l sec

retio

n (p

mol

L)

(a)

A C B A + B

P450

arom

GA

PDH

20

15

10

05

00

lowastlowast

(55kDa)

GAPDH(36kDa)

A + BBCAP450arom

(b)

20

15

10

05

00

lowastlowast

Blank NC-siRNA

Blank NC-siRNA

(60kDa)

GAPDH(36kDa)

Smad4

Smad4-siRNA

Smad4-siRNA

Smad4

GA

PDH

(c)

lowastlowast

A + RRAC

200

150

100

50

0

Estr

adio

l sec

retio

n (p

mol

L)

R(minus)

(d)

RAC

A + R

A + R

RAC

P450arom(55kDa)

GAPDH(36kDa)

20

15

10

05

00

lowastlowast

R(minus)

R(minus)

P450

arom

GA

PDH

(e)




4 Discussion





5 Conclusions


Competing Interests


Acknowledgments



References
























































3 Results










Rela

tive P

450

arom

mRN

A le

vels

25

20

15

10

05

00Control 2 12 24

Time (hrs)

lowast lowast

Real

-tim

e PCR

(a)

Rela

tive P

450

arom

mRN

A le

vels

25

20

15

10

05

00Control 25 25 50

Activin A (ngmL)

lowast

(b)

GAPDH(36kDa)

P450arom(55kDa)

Control 2 12 24

Control 2 12 24

Time (hrs)

lowast

Wes

tern

blo

tP450

arom

GA

PDH

15

10

05

00

(c)

P450arom(55kDa)

GAPDH(36kDa)

Control 25 25 50

Control 25 25 50

Activin A (ngmL)

lowast

P450

arom

GA

PDH

15

10

05

00

(d)

ControlActivin A

Testosterone

lowastlowast

200

150

100

50

0Normal Eutopic

lowastlowast

Estr

adio

l sec

retio

n(p

mol

L)




(52kDa)

(52kDa)

(36GAPDH

kDa)

(70kDa)


p-Smad3

Smad3

LaminAC

(a)

Control

Activin A

Control

Activin A

DAPI Merge

DAPI MergeSmad3

p-Smad3

(b)



200

150

100

50

0A C B A + B

lowastlowast

Estr

adio

l sec

retio

n (p

mol

L)

(a)

A C B A + B

P450

arom

GA

PDH

20

15

10

05

00

lowastlowast

(55kDa)

GAPDH(36kDa)

A + BBCAP450arom

(b)

20

15

10

05

00

lowastlowast

Blank NC-siRNA

Blank NC-siRNA

(60kDa)

GAPDH(36kDa)

Smad4

Smad4-siRNA

Smad4-siRNA

Smad4

GA

PDH

(c)

lowastlowast

A + RRAC

200

150

100

50

0

Estr

adio

l sec

retio

n (p

mol

L)

R(minus)

(d)

RAC

A + R

A + R

RAC

P450arom(55kDa)

GAPDH(36kDa)

20

15

10

05

00

lowastlowast

R(minus)

R(minus)

P450

arom

GA

PDH

(e)




4 Discussion





5 Conclusions


Competing Interests


Acknowledgments



References























































Rela

tive P

450

arom

mRN

A le

vels

25

20

15

10

05

00Control 2 12 24

Time (hrs)

lowast lowast

Real

-tim

e PCR

(a)

Rela

tive P

450

arom

mRN

A le

vels

25

20

15

10

05

00Control 25 25 50

Activin A (ngmL)

lowast

(b)

GAPDH(36kDa)

P450arom(55kDa)

Control 2 12 24

Control 2 12 24

Time (hrs)

lowast

Wes

tern

blo

tP450

arom

GA

PDH

15

10

05

00

(c)

P450arom(55kDa)

GAPDH(36kDa)

Control 25 25 50

Control 25 25 50

Activin A (ngmL)

lowast

P450

arom

GA

PDH

15

10

05

00

(d)

ControlActivin A

Testosterone

lowastlowast

200

150

100

50

0Normal Eutopic

lowastlowast

Estr

adio

l sec

retio

n(p

mol

L)




(52kDa)

(52kDa)

(36GAPDH

kDa)

(70kDa)


p-Smad3

Smad3

LaminAC

(a)

Control

Activin A

Control

Activin A

DAPI Merge

DAPI MergeSmad3

p-Smad3

(b)



200

150

100

50

0A C B A + B

lowastlowast

Estr

adio

l sec

retio

n (p

mol

L)

(a)

A C B A + B

P450

arom

GA

PDH

20

15

10

05

00

lowastlowast

(55kDa)

GAPDH(36kDa)

A + BBCAP450arom

(b)

20

15

10

05

00

lowastlowast

Blank NC-siRNA

Blank NC-siRNA

(60kDa)

GAPDH(36kDa)

Smad4

Smad4-siRNA

Smad4-siRNA

Smad4

GA

PDH

(c)

lowastlowast

A + RRAC

200

150

100

50

0

Estr

adio

l sec

retio

n (p

mol

L)

R(minus)

(d)

RAC

A + R

A + R

RAC

P450arom(55kDa)

GAPDH(36kDa)

20

15

10

05

00

lowastlowast

R(minus)

R(minus)

P450

arom

GA

PDH

(e)




4 Discussion





5 Conclusions


Competing Interests


Acknowledgments



References























































(52kDa)

(52kDa)

(36GAPDH

kDa)

(70kDa)


p-Smad3

Smad3

LaminAC

(a)

Control

Activin A

Control

Activin A

DAPI Merge

DAPI MergeSmad3

p-Smad3

(b)



200

150

100

50

0A C B A + B

lowastlowast

Estr

adio

l sec

retio

n (p

mol

L)

(a)

A C B A + B

P450

arom

GA

PDH

20

15

10

05

00

lowastlowast

(55kDa)

GAPDH(36kDa)

A + BBCAP450arom

(b)

20

15

10

05

00

lowastlowast

Blank NC-siRNA

Blank NC-siRNA

(60kDa)

GAPDH(36kDa)

Smad4

Smad4-siRNA

Smad4-siRNA

Smad4

GA

PDH

(c)

lowastlowast

A + RRAC

200

150

100

50

0

Estr

adio

l sec

retio

n (p

mol

L)

R(minus)

(d)

RAC

A + R

A + R

RAC

P450arom(55kDa)

GAPDH(36kDa)

20

15

10

05

00

lowastlowast

R(minus)

R(minus)

P450

arom

GA

PDH

(e)




4 Discussion





5 Conclusions


Competing Interests


Acknowledgments



References























































200

150

100

50

0A C B A + B

lowastlowast

Estr

adio

l sec

retio

n (p

mol

L)

(a)

A C B A + B

P450

arom

GA

PDH

20

15

10

05

00

lowastlowast

(55kDa)

GAPDH(36kDa)

A + BBCAP450arom

(b)

20

15

10

05

00

lowastlowast

Blank NC-siRNA

Blank NC-siRNA

(60kDa)

GAPDH(36kDa)

Smad4

Smad4-siRNA

Smad4-siRNA

Smad4

GA

PDH

(c)

lowastlowast

A + RRAC

200

150

100

50

0

Estr

adio

l sec

retio

n (p

mol

L)

R(minus)

(d)

RAC

A + R

A + R

RAC

P450arom(55kDa)

GAPDH(36kDa)

20

15

10

05

00

lowastlowast

R(minus)

R(minus)

P450

arom

GA

PDH

(e)




4 Discussion





5 Conclusions


Competing Interests


Acknowledgments



References
























































4 Discussion





5 Conclusions


Competing Interests


Acknowledgments



References























































References






















































activin a stimulates aromatase via the alk4-smad pathway

Documents