Biochimica et Biophysica Acta 1833 (2013) 1165–1172

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r .com/ locate /bbamcr

Sex-specific response of rat costochondral cartilage growth platechondrocytes to 17β-estradiol involves differential regulation of plasmamembrane associated estrogen receptors

Khairat B.Y. Elbaradie a, Yun Wang b, Barbara D. Boyan a,b,⁎, Zvi Schwartz b

a School of Biology, Georgia Institute of Technology, Atlanta, GA, USAb Wallace H. Coulter Department of Biomedical Engineering at Georgia Tech and Emory University, Georgia Institute of Technology, Atlanta, GA, USA

⁎ Corresponding author at: Department of Biomedical ETechnology, 315 Ferst Drive NW, Atlanta, GA 30332-036fax: +1 404 894 2291.

E-mail address: barbara.boyan@bme.gatech.edu (B.D

0167-4889/$ – see front matter © 2013 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.bbamcr.2012.12.022

a b s t r a c t

a r t i c l e i n f o

Article history:Received 11 September 2012Received in revised form 29 December 2012Accepted 30 December 2012Available online 7 January 2013

Keywords:Sex-specific response to 17β-estradiolEstrogen receptorDistribution of ERα and ERβ in male andfemale growth plate chondrocytes

Bothmale and female rat growth plate chondrocytes express estrogen receptors (ERs); however 17β-estradiol (E2)inducesmembrane responses leading to activation of phospholipase A2 (PLA2), phospholipase C (PLC), prostaglan-din E2 (PGE2) production, protein kinase C (PKC), and ultimately mitogen protein kinase (MAPK) only in femalecells. This study investigated if these sex-specific responses are due to differences in the actual ERs or in down-stream signaling. Western blots and flow cytometry of costochondral cartilage resting zone chondrocytes (RCs)showed 2–3 timesmore ERα in plasmamembranes (PMs) from female cells thanmale cells. Tunicamycin blockedE2-dependent ER-translocation to the PM, indicating palmitoylation was required. Co-immunoprecipitationshowed E2 induced complex formation between ER isoforms only in female RCs. To examine if the lack of responsein PKC and PGE2 inmales is due to differences in signaling,we examined involvement of ERs and the role of PLC andPLA2. Selective ERα (propylpyrazole triol, PPT) and ERβ (diarylproprionitrile, DPN) agonists activated PKC in fe-male RCs only. The PLC inhibitor, U73122 blocked E2's effect on PKC and the cytosolic PLA2 inhibitor, AACOCF3inhibited the effect on PGE2 in female RCs, confirming involvement of PLC and PLA2 in the mechanism. The PLCactivator, m-3M3FβS activated PKC and PLAA peptide increased PGE2 levels in male and female RCs, showingthat the signaling pathways are present. These data indicate that differences inmembrane ER amount, localization,translocation and interaction are responsible for the sexual dimorphic response to E2.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Longitudinal growth is regulated by the activity of chondrocytes inthe epiphyseal growth plates of long bones, including proliferation, dif-ferentiation, and apoptosis. Many hormones and growth factors play arole in the regulation of this process. Among these, sex steroids are ofcrucial importance, especially during puberty [1]. Estrogens are involvedin the initiation of the pubertal growth spurt and in fusion of the growthplate at the end of puberty in both boys and girls. The reasons for thedifferent effects of estrogens during puberty are not well understood.In vitro studies examining the effect of estradiol on growth platechondrocytes have provided conflicting information. 17β-Estradiol atphysiological concentrations (10−11 to 10−12 M) has been reportednot to affect proliferation, viability or synthesis of type X collagen in pri-mary cultures of resting, proliferative or hypertrophic chondrocytes inone set of experiments [4]; to stimulate chondrocyte proliferation inother studies [5,6]; and to inhibit chondrocyte proliferation in yet

ngineering, Georgia Institute of3, USA. Tel.: +1 404 385 4108;

. Boyan).

rights reserved.

other studies [7–9]. One possibility is that the systemic levels of thehormone, and presumably its concentration in the tissue, are factors.

Levels of systemic estrogens are important variables, but the num-ber and type of estrogen receptors (ERs) are also critical determinantsof growth plate response to the hormone [2,3]. Both ERα and ERβ arepresent in growth plates of both males and females at the mRNA andprotein level in several species, including rat, rabbit, and human[10–13], indicating that estrogens can directly regulate the tissue in ad-dition to their indirect effects via growth hormone [14–18]. Moreover,experiments using fetal rat and neonatal mouse bone organ culturesshow that growth plate development is regulated in a sex-specificman-ner [19]. The results suggested that the sexual dimorphism was at thecellular level. Although the binding affinity of E2 for ERs in growthplate chondrocytes was comparable in cells from both male and femalerats, female cells had greater numbers of receptors than male cells [20].

While receptor number could contribute to differences in responserate, it does not explain the marked differences in biological responsesobserved betweenmale and female cells.Whereas E2 caused an increasein chondrocyte differentiation in female cells, it did not do so in malecells [7]. This effect of E2 was stereospecific, indicating that it was recep-tor mediated. However, themechanism involved differed from thatme-diating the effect of E2 on chondrocyte apoptosis, whichwas the same in

1166 K.B.Y. Elbaradie et al. / Biochimica et Biophysica Acta 1833 (2013) 1165–1172

cells from both sexes [21]. Subsequent studies identified plasma mem-brane associated signaling in the sex-specific responses. E2 causedrapid increases in Ca++ ion transport [22] and activation of protein ki-nase C (PKC) [23] and phospholipase A2 [24] in female chondrocytesbut not in male cells. Importantly, these effects of E2 could also beseen in cells treated with E2-conjugated to bovine serum albumin(E2-BSA), which can't pass the plasma membrane, but did stimulatePKC. Moreover, inhibition of PKC signaling blocked the downstream ef-fect of E2 on chondrocyte differentiation [25]. Additionally, neitherdiethylstilbesterol nor ICI 182780, which are known to act via thenuclear ER, had an effect on the membrane associated response [25].

We hypothesized that ERs resident in the plasma membranes ofgrowth plate chondrocytes might be responsible for the differential re-sponses seen inmale and female cells. ERα [26,27] and ERβ [28–30] arethe products of separate genes. In many cells, they coexist either ashomodimers or as heterodimers [31,32]. Recent evidence indicatesthat there are specific actions of E2 that can be attributed to one receptorbut not the other [33,34]. This is supported by in vitro studieswhere es-trogen analogs have been shown to preferentially activate the two ERs[35]. These observations, coupled with the fact that ERβ can formhomodimers and heterodimers with ERα in vitro [31,32], suggest thatthe activity of estrogens may depend on whether a cell contains ERα,ERβ, or both.

While it is known that ERα and ERβ are both present in male and fe-male cells, it is not known if they are differentially expressed or activated.It is also not known if they are transported to themembrane in a compa-rable manner. Of particular importance is if the membrane associatedsignaling pathways needed to mediate the downstream responses toE2 are present inmale growth plate chondrocytes. In this study, we com-pared the expression and subcellular localization of ERα and ERβ inmaleand female rat costochondral growth plate chondrocytes; we deter-mined if their presence in the membrane involved palmitoylation ashas been shown in other systems; and we investigated differences inthe potential downstream signaling pathways.

2. Material and methods

2.1. Reagents

17β-Estradiol (E2); methyl-β-cyclodextrin (β-CD), which depleteslipid rafts and alters the caveolar microenvironment [36,37]; the phos-pholipase C (PLC) inhibitorU73122 [38]; and the PLA2 inhibitor AACOCF3[39] were purchased from Sigma Chemical Co. (St. Louis, MO). TheERα-selective agonist propyl pyrazole triol (PPT) [40–42]; the PLC ago-nist m3M3FβS [43]; and the ERβ-selective agonist diarylpropionitrile(DPN) [40–42] were obtained from Tocris Cookson Inc. (Ellisville, MO).Tunicamycin, an inhibitor of palmitoylacetyltransferase [44], was pur-chased from EMD Chemicals, Inc. (San Diego, CA). ERαmonoclonal anti-body (ab16460) and ERβ monoclonal antibody (ab16813) werepurchased from Abcam (San Francisco, CA). A polyclonal antibody tocaveolin-1 (Cav-1) was purchased from Santa Cruz Biotechnology(sc-894, Santa Cruz, CA). PKC BioTrak assay kits were obtained from GELifesciences (Piscataway, NJ) and phenol red-free Dulbecco's modifiedEagle medium (DMEM) was obtained from GIBCO-BRL (Gaithersburg,MD). The protein content of each sample was determined using thebicinchoninic acid (BCA) protein assay kit [45] obtained from PierceChemical Co. (Rockford, IL). [32P]-ATP was obtained from Perkin Elmer(Melville, NY).

2.2. Chondrocyte Cultures

The culture system used in this study has been described in detailpreviously [46]. Chondrocytes were isolated from the resting zone carti-lage of the costochondral junction of 150 g male and female Sprague–Dawley rats, plated at an initial density of 10,000 cells/cm2, andmaintained in phenol red-free DMEM containing 10% fetal bovine

serum (FBS) and 50 μg/ml ascorbic acid. The medium was replacedafter 24 h and then every 48 h until cultures reached confluence. Fourthpassage cells were used for all experiments based on previous studiesshowing that these cells preserve their chondrocyte phenotype as wellas their differential responsiveness to vitamin Dmetabolites at this pas-sage [47]. Confluent fourth passage male and female cells in 24 wellplates were treated for 90 min with 0.5 ml of ethanol at the highestconcentration used in experimental cultures or experimental media(phenol red-free DMEM+10% FBS+appropriate concentration of E2).

2.3. ER expression and subcellular localization

2.3.1. Plasma membrane and caveolae isolationPlasma membranes were isolated using a detergent-free method as

described previously [48]. Confluent, fourth passage resting zonechondrocytes isolated from female and male rats were treated with ei-ther 10−7 M E2 or vehicle for 90 min. The cell layers were washedtwo times with phosphate buffered saline (PBS) to remove any residualmedium and cells were harvested by scraping using isolation buffer(0.25 M sucrose, 1 mM EDTA, 20 mM Tricine, pH 7.8). Cells were ho-mogenized using a tissue grinder and the homogenate centrifuged at3500 rpm for 10 min to pellet large debris including the nucleus, mito-chondria, and endoplasmic reticulum. The supernatant was collectedand layered on 30% Percoll in isolation buffer (GE Healthcare,Piscataway, NJ). Plasma membranes were collected from the samplesby centrifuging at 30,500 rpm for 30 min and removing the visibleband in themiddle of the gradient column. The plasmamembrane frac-tionwas then layered over a 10–20%OptiPrep gradient (Sigma-Aldrich),and the gradient was centrifuged at 26,500 rpm for another 4 h. Plasmamembrane sub-fractions were collected from the top to the bottom ofthe tube, which resulted in isolation of 15 fractions. Caveolae were ob-served as an opaque band, which was collected in fraction 3.

Western blots were used to determine if ERα and/or ERβ werepresent in whole cell lysates and the distribution of the ER isoformsin plasmamembranes and caveolae in the male and female cells. Plas-ma membranes, the caveolae fraction and the nuclear fraction wereseparated on 4-20% gradient acrylamide gels. Blots of the gels wereprobed with either the mouse monoclonal antibody against ERα orthe mouse monoclonal antibody against ERβ. In addition, blots wereprobed using a polyclonal antibody to caveolin-1 (Cav-1). Immunore-active bands were visualized using goat anti-mouse horseradishperoxidase-conjugated secondary antibodies (Bio-Rad, Hercules,CA). The Super Signal West Pico Chemiluminescent System (ThermoFisher Scientific) was used to treat the membrane and then imagedwith the VersaDoc imaging system (Bio-Rad).

2.3.2. PalmitoylationTo assess the role palmitoylation in localizing ERα and ERβ to the

plasma membrane as well as the role of E2 in regulating ERα and ERβpalmitoylation, male and female cultures were pre-incubated with30 μM tunicamycin for 2 h. The cells were then treated with 10−7 ME2 for 90 min. Plasmamembranes were isolated and analyzed as above.

2.3.3. Receptor complex formationTo determine if the ER isoforms formed complexes in response to E2,

fourth passage chondrocytes from female rats were treated with andwithout 10−7 M E2 for 90 min. At harvest the cell layers were washedwith PBS and the cells were lysed and sonicated in RIPA buffer (50 mMTris–HCl, 150 mM NaCl, 5 mM disodium EDTA, 1% Nonidet P-40)containing 100 mM NaF, protease inhibitor cocktail (Sigma-Aldrich)and 1 mM phenylmethylsulfonyl fluoride (PMSF). Protein sampleswere mixed with anti-ERα antibody and incubated at 4 °C overnightwith continuous agitation. Immunoprecipitation was measured in pro-tein samples using a commercially available Dynabeads® Protein kit fol-lowing the manufacturer's directions (Invitrogen, Grand Island, NY). Inorder to lower the background in the images, a ONE-HOUR IP-Western

Fig. 1. Difference in the amount and subcellular localization of ERα and ERβ betweenmale and female chondrocytes. Subcellular localization of ERα in female resting zone chondrocytes (A), ormale resting zone chondrocytes (B). Subcellular localization of ERβ in female resting zone chondrocytes (C), or male resting zone chondrocytes (D). Whole cell lysates (WCL), plasma mem-branes (PM), caveolae fraction (F3), and nuclear fraction (NF) were isolated and subjected to Western blot. Female cultures were pre-incubated for 30 min with 5 mMmethyl-β-cyclodextrin (β-CD) and then treated with 10−7 M E2 for 90 min. PKC specific activity was measured in cell layer lysates. Values are the mean±SEM for six independent culturesfor each variable. *Pb0.05, treatment vs. control, ^Pb0.05, E2+ β-CD vs. E2 alone.

1167K.B.Y. Elbaradie et al. / Biochimica et Biophysica Acta 1833 (2013) 1165–1172

Kit (Genscript) was used. The immunoprecipitated samples were thensubjected to Western blot analysis.

2.3.4. Receptor numberThe number of the ERs on the plasmamembrane and intracellularly

was evaluated by flow cytometry using the Quantum Simply CellularMicrobeads Kit (Bangs Laboratories, Inc. Fishers, IN). This kit providesa method for the evaluation of the number of molecules of antibodyper cell. Briefly, this kit is comprised of populations of uniformcell-sized microspheres with different calibrated binding capacities ofgoat anti-mouse IgG (FC-specific) on their surface. The kit includes 4coated populations, each with different antibody binding capacity(ABC) for mouse monoclonal antibodies, and 1 blank population withno specific binding capacity for mouse IgG. When the bead populationsare labeled in the samemanner as the cells to be analyzed, they provide

Fig. 2. Difference in ERα content betweenmale and female chondrocytes. Flow cytometry show(plasma membrane and nuclear receptors) in permeabilized (B). Values are the mean±SEM f

a means of constructing a QuickCal® calibration curve (ABC values ver-sus fluorescence intensity), from which samples may be analyzed.

The number of ERα cell surface receptors was determined by incu-bating male and female cell suspensions (5∗105 cells) for 30 min in150 μl 1% bovine serum albumin (BSA) with a 1:50 dilution of theERα monoclonal antibody. Cells were then washed with 1 ml 1% BSA,centrifuged for 5 min at 1000 rpm, and resuspended in 150 μl 1% BSAcontaining a 1:500 dilution of goat-anti mouse antibody conjugated toAlexa Fluor® (Invitrogen, Grand Island, NY) for 30 min in the dark.Cells were centrifuged at 1000 rpm for 10 min and resuspended in300 μl PBS. After collecting the cells by filtration, samples were readyfor analysis.

To determine the total number of receptors (surface and cytoplas-mic), cells were fixed by incubating with 4% paraformaldehyde in PBSfor 20 min at 4 °C. Then cells were collected by centrifugation at1000 rpm for 10 min and resuspended in permeabilization solution

ing plasmamembrane ERα content in non permeabilized cells (A), and total ERα numberor six independent cultures for each variable. *Pb0.05, female vs. male.

Fig. 3. Palmitoylationof themembrane receptor is required for E2-dependentPKCactivity only in female chondrocytes. Female cellswere either treated for 90 minwith10−7 ME2 alone orwith30 μMtunicamycin for 2 h and then treatedwith10−7 ME2 for 90 min. Presence of ERα and ERβ inwhole cell lysates andplasmamembranes of female chondrocytes (A) ormale chondrocytes(B) was examined byWestern blot.

1168 K.B.Y. Elbaradie et al. / Biochimica et Biophysica Acta 1833 (2013) 1165–1172

(0.5% Triton X-100 in PBS) at room temperature for 15 min. Perme-abilized cells were incubated with the primary antibody and furtherprocessed as described above.Microspheres (50 μL per tube)were incu-bated with the secondary antibody, washed and further processed onthe flow cytometer under the same conditions as the samples.

2.4. Signaling pathways

2.4.1. Protein kinase C activityConfluent fourth passage male and female cells in 24 well plates

were treated for 90 min with 0.5 ml of ethanol at the highest concen-tration used in experimental cultures or experimental media (phenolred-free DMEM+10% FBS+10−7 to 10−10 M E2). To determinewhich ERs were involved in this pathway, male and female cultureswere treated for 90 min with 10−7 to 10−9 M ERα-selective agonistPPT or ERβ-selective agonist DPN, respectively. To determine if theincrease in PKC was caveolae dependent, female cultures werepre-incubated for 30 min with 5 mM methyl-β-cyclodextrin (β-CD)[36], and then treated for 90 min with 10−7 M E2.

We previously showed that phosphatidylinositol-specific PLC medi-ated the rapid increase in E2-dependent PKC [25]. To establish if PLC alsoplays a role in the increase in PKC at 90 min, confluent cultures of femalecells were treated with E2±the PLC inhibitor U73122. To determine ifthe downstream signaling pathway was functional in male cells, male

Fig. 4. ERαwas immunoprecipitated and subjected to Western blot. The membranes were incubgroup (10−7 M E2 for 90 min).

and female cells were treatedwith PLC activatorm-3M3FβS and PKC ac-tivity was measured. After the incubation period, cell layers werewashedwith PBS to remove any excess medium and then lysed in solu-bilization buffer (50 mM Tris–HCl, pH 7.5, 150 mM NaCl, 5 mM EDTA,1 mMPMSF and 1%NP-40). 25 μl of the cell lysates was assayed for pro-tein content [45]. PKC activity was measured in cell lysates using a kitfollowing the manufacturer's directions (Amersham Bioscience). Thiskit is optimized for measurement of phospholipid and Ca2+-dependentPKC [49,50].

2.5. Regulation of PGE2 release

PGE2 production is a downstream consequence of PLA2 activationand can activate the PKC signaling pathway leading to ERK1/2 activation[51]. To determine if PGE2 release into the medium is produced inresponse to E2, female cells were pre-incubated for 30 min with10−6–10−5 M of the PLA2 inhibitor AACOCF3, and then cells weretreated with 10−7 M E2 for 90 min. Male and female cells were alsotreated for 30 min with 10−6–10−5 M PLAA peptide, which activatesPLA2 [52]. At the end of the incubation, the media were acidified andPGE2 was measured using a commercially available kit (ProstaglandinE2 [125I]-RIA kit, NEK020001K, Perkin Elmer, Waltham, MA). PGE2data were normalized to total DNA (Quant-iT™ PicoGreen® dsDNAAssay kit, P11496, Invitrogen).

ated with anti-ERα and anti-ERβ antibodies. C: control group (with vehicle only), T: treated

Fig. 5. Effect of PPT and DPN on PKC specific activity in resting zone chondrocytes. Confluent fourth passage female andmale chondrocytes were treated for 90 min with 10−7 to 10−9 M PPT(A) or 10−7 to 10−9 MDPN (B). PKC specific activitywasmeasured in cell layer lysates. Values are themean±SEM for six independent cultures for each variable. *Pb0.05, treatment vs. controlat each time point.

1169K.B.Y. Elbaradie et al. / Biochimica et Biophysica Acta 1833 (2013) 1165–1172

2.6. Statistical management of the data

For each experiment, each value represents the mean±the standarderror of themean (SEM)of the cell layers of six independent cultures. Sta-tistical significance was determined by analysis of variance (ANOVA).Pb0.05 was considered significant. Each experiment was repeated twoor more times to ensure the validity of the data. The data representedare from a single representative experiment.

3. Results

3.1. Membrane ER localization

Western blots of the cell lysates demonstrated that ERαwas presentin female and male resting zone chondrocytes (data not shown). Thenuclear fraction from both cell types exhibited all three ERα isoforms:ERα68, ERα46 and ERα36 (Fig. 1A,B). Plasma membranes from femalechondrocytes had all three isoforms, but male cells lacked ERα36. Thiswas also the case for caveolae. ERβ59 was present in plasma mem-branes and nuclear fraction from female andmale cells andwas presentin caveolae from female cells but not male cells (Fig. 1C,D).

Sex-specific differences in the amount and distribution of ERαwereconfirmed by flow cytometry. There wasmore than twice asmuch plas-mamembraneERα in non-permeabilized female cells than inmale cells.Total ERα in permeabilized cells was not statistically different, however

Fig. 6. Activation of PKC by E2 is dependent onPLC signaling. Confluent, fourth passage resting z10−6–10−5 M PLC inhibitor U73122 (A) or 10−6–10−5 M PLC activatorm-3M3FβS (B). *Pb0.017β-estradiol.

(Fig. 2A,B). Interestingly, the percentage of positive cells was not signif-icantly different in both permeabilized and non-permeabilized femaleand male cells.

3.2. Effect of E2 on ER localization

Tunicamycin reduced the amount of ERα68 in whole cell lysates offemale chondrocytes but had no effect on the male cells (Fig. 3A,B). E2caused an increase in plasma membrane ERα46 and ERα36, as well asERβ59 in female cells, all of which were reduced by pretreatmentwith tunicamycin. In contrast, E2 reduced the levels of ERα68, ERα46and ERβ59 in the plasma membranes from male cells and this was un-affected by tunicamycin treatment. These observations were confirmedby co-immunoprecipitation of ERα. Treatment of female chondrocytesfor 90 min with E2 increased the amount of ERα68, ERα46 and ERβ59that was present in the immunoprecipitates, but inmale cells, the levelsof all three isoforms were reduced (Fig. 4).

3.3. Role of ERs and membrane associated signaling

Intact caveolae were required for E2 activation of PKC in femalecells. Treatment with methyl β-cyclodextrin blocked the stimulatoryeffect of the hormone (Fig. 1E).

Both ERα and ERβ were involved in the E2-dependent activation ofPKC in female cells. Treatment of the cells with the ERα agonist PPT

one cells from female rats (A,B) ormale rats (B)were treated for 90 minwith 10−7 ME2±5, treatment vs. control, ^Pb0.05, E2+U73122 vs. E2 alone at a particular concentration of

Fig. 7. Production of PGE2 by E2 is dependent on PLA2 signaling. Confluent, fourth passage resting zone cells from female rats (A,B) or male rats (B) were treated for 90 minwith 10−7 M E2±10−6–10−5 M cytosolic PLA2 inhibitor ACCOCF3 (A) or 10−6–10−5 M PLA2 activator PLAA peptide (B). *Pb0.05, treatment vs. control, ^Pb0.05, E2+ACCOCF3 vs. E2 alone at a particularconcentration of E2.

1170 K.B.Y. Elbaradie et al. / Biochimica et Biophysica Acta 1833 (2013) 1165–1172

(Fig. 5A) or the ERβ agonist DPN (Fig. 5B) had the same stimulatoryeffect on PKC as E2. In contrast, activation of ERα or ERβ using therespective agonists had no effect on PKC in male cells.

Inhibition of PLC blocked the stimulatory effect of E2 on PKC in thefemale cells, even at 90 min (Fig. 6A). Conversely, the PLC activatorm-3M3FβS induced an increase in PKC activity in the female cells;male cells also responded to the PLC activator with an increase in PKC,but the effect was seen only at the highest concentration (Fig. 6B).

PGE2 production in response to E2 was blocked by the PLA2 inhibitorAACOCF3 (Fig. 7A). Activation of PLA2with PLAA also caused an increasein PGE2 release in the female cells; male cells exhibited an increase inPGE2 release when treated with PLAA, but the stimulatory effect wasless robust than was observed in the female cells (Fig. 7B).

4. Discussion

This study provides evidence that the sex-specific difference in E2-dependent membrane-associated signaling in male and female growthplate chondrocytes is due to differential regulation of ER isoforms inthe plasmamembrane and specifically, in the caveolae.Whereas plasmamembranes and caveolae isolated from female cells possess ERα68,ERα46 and ERα36 as well as ERβ59, plasma membranes from malecells lack ERα36. Moreover, caveolae from male cells lack both ERα36and ERβ59. Caveolae are critical to the ability of E2 to activate rapidPKC signaling in the female cells, as evidenced by the loss of an E2 effectwhen caveolae were disrupted by treatment with β-cyclodextrin.

We did not specifically knockdown ERα36 or ERβ59, but our resultsstrongly implicate both receptors in the failure of male cells to respondto E2. Male caveolae lack both isoforms but possess ERα68 and ERα46.Although agonists for both ERα and ERβ activated PKC in femalechondrocytes, neither agonist was able to do so in male chondrocytes,despite the presence of two ERα isoforms as well as evidence compo-nents of the signalingpathways involved in PKC activationwere presentand could be stimulated to the same extent as in female cells.

ER trafficking may be an important variable in the differential re-sponse of male and female chondrocytes to E2 as well. Others haveshown that palmitoylation is required for the plasmamembrane associ-ation of ERs [44,53]. This was the case for female chondrocytes in thepresent study. E2 increased the association of ERswith the plasmamem-brane, particularly ERα46, ERα36 and ERβ59 whereas tunicamycin re-duced the association of all three ERα isoforms and ERβ59 with the

plasma membrane. In male cells, however, E2 caused a marked reduc-tion in the association of ERα68, ERα46 and ERβ59 with the plasmamembrane. Why the receptor trafficking in the male cells was contrato that of the female cells is not clear. In female cells, E2 elicits formationof a multi-receptor complex that involves both ERα and ERβ, but this isnot the case for male cells, suggesting that a scaffolding unit may beabsent. In other studies, caveolin-1 provides scaffolding for ER transloca-tion to the plasmamembrane [37,54,55] and palmitoylation is requiredfor caveolin-1 to interact with the plasmamembrane [56]. This suggeststhat a difference between male and female cells may be related to thepalmitoylation mechanism. It is also possible that E2 treatment in malecells may lead to ER translocation to the nucleus.

Our results support our previous study indicating that female cellsmay have more ERs than male cells [20] and show that it is specificallythe pool of ERs that are on the surface of the cell that contribute to thisincrease. The percentage of positive female cellswas not significantly dif-ferent than the percentage of positive male cells in non-permeabilizedcells, although the total number of ERα on the surface was significantlyhigher in female cells. This is in agreement with our previous study,which showed that there was no difference in E2-binding affinity be-tween females and males, but females had more receptors [20]. No dif-ferences in total ERα number were found when assessing ERα contentof permeabilized cells whereas ERα content of non-permeabilizedcells, i.e., the cell surface, was more than two times greater in femalecells. This is an important observation given the fact that ERα36 waspresent only on the plasma membranes of female cells, whereas allother isoforms of ERα as well as ERβ59 were present on the plasmamembranes of male cells. While this strongly supports the hypothesisthat ERα36 plays a critical role, it is not yet definitive proof. However,recent studies in our lab examining the role of ERα36 in E2 dependentactivation of PKC in breast cancer cells showed that specific antibodiesto this isoform blocked the stimulatory effect of the hormone [57].

E2 activates PKC in growth plate chondrocytes via two signalingpathways, PLC and PLA2. Our results indicate that the failure of malecells to respond to E2 is not due to an intrinsic defect in either pathway.Activation of PLC bym-3M3FβS could be achieved to the same extent inmale and female cells. Similarly, both male and female cells were capa-ble of generating PGE2 when PLAA was used to activate PLA2. The effectof PLAAwas less robust inmale cells, raising the possibility that the PLA2

pathway was reduced in some manner, but it is difficult to draw toomany conclusions given the fact that PGE2 production is a downstream

1171K.B.Y. Elbaradie et al. / Biochimica et Biophysica Acta 1833 (2013) 1165–1172

surrogate for the enzyme itself. These data do show that the differencein response occurs upstream from the enzymes triggering the signalingpathways and supports the conclusion that the difference is at the re-ceptor level.

Acknowledgements

This work was supported by Children's Healthcare of Atlanta, thePrice Gilbert, Jr. Foundation and Sector Missions of Cultural Relationsat the Ministry of Higher Education Arab Republic of Egypt.

References

[1] S.J. Holmes, S.M. Shalet, Role of growth hormone and sex steroids in achieving andmaintaining normal bone mass, Horm. Res. 45 (1996) 86–93.

[2] E.P. Smith, J. Boyd, G.R. Frank, H. Takahashi, R.M. Cohen, B. Specker, T.C. Williams, D.B.Lubahn, K.S. Korach, Estrogen resistance caused by amutation in the estrogen-receptorgene in a man, N. Engl. J. Med. 331 (1994) 1056–1061.

[3] J.E. Compston, Sex steroids and bone, Physiol. Rev. 81 (2001) 419–447.[4] C. Rodd, N. Jourdain, M. Alini, Action of estradiol on epiphyseal growth plate

chondrocytes, Calcif. Tissue Int. 75 (2004) 214–224.[5] D. Somjen, Y.Weisman, A. Harell, E. Berger, A.M. Kaye, Direct and sex-specific stimula-

tion by sex steroids of creatine kinase activity andDNA synthesis in rat bone, Proc. Natl.Acad. Sci. U.S.A. 86 (1989) 3361–3365.

[6] D. Somjen, Y.Weisman, Z.Mor, A. Harell, A.M. Kaye, Regulation of proliferation of ratcartilage and bone by sex steroid-hormones, J. Steroid Biochem.Mol. Biol. 40 (1991)717–723.

[7] E. Nasatzky, Z. Schwartz, B.D. Boyan,W.A. Soskolne, A. Ornoy, Sex-dependent effects of17-beta-estradiol on chondrocyte differentiation in culture, J. Cell. Physiol. 154 (1993)359–367.

[8] Z. Schwartz, Y. Finer, E. Nasatzky,W.A. Soskolne, D.D. Dean, B.D. Boyan, A. Ornoy, Theeffects of 17β-estradiol on chondrocyte differentiation aremodulated by vitamin D3metabolites, Endocrine 7 (1997) 209–218.

[9] Z. Schwartz, V.L. Sylvia, T. Hughes, Z. Chang, D.D. Dean, B.D. Boyan, 17 beta-estradiolmediates its sex-specific effects on chondrocyte differentiation by activation of pro-tein kinase C, J. Bone Miner. Res. 12 (1997) T473.

[10] K. Pettersson, J.A. Gustafsson, Role of estrogen receptor beta in estrogen action, Annu.Rev. Physiol. 63 (2001) 165–192.

[11] V. Kusec, A.S. Virdi, R. Prince, J.T. Triffitt, Localization of estrogen receptor-alpha inhuman and rabbit skeletal tissues, J. Clin. Endocrinol. Metab. 83 (1998) 2421–2428.

[12] J. Kennedy, C. Baris, J.A. Hoyland, P.L. Selby, A.J. Freemont, I.P. Braidman, Immunoflu-orescent localization of estrogen receptor-alpha in growth plates of rabbits, but notin rats, at sexual maturity, Bone 24 (1999) 9–16.

[13] O. Nilsson, V. Abad, D. Chrysis, E.M. Ritzen, L. Savendahl, J. Baron, Estrogenreceptor-alpha and -beta are expressed throughout postnatal development inthe rat and rabbit growth plate, J. Endocrinol. 173 (2002).

[14] B. Carlsson, O. Forsberg,M. Bengtsson, T.H. Totterman,M. Essand, Characterization ofhuman prostate and breast cancer cell lines for experimental T cell-based immuno-therapy, Prostate 67 (2007) 389–395.

[15] Z. Schwartz, W.A. Soskolne, T. Neubauer, M. Goldstein, S. Adi, A. Ornoy, Direct andsex-specific enhancement of bone-formation and calcification by sex steroids in fetalmice long-bone in vitro (biochemical and morphometric study), Endocrinology 129(1991) 1167–1174.

[16] C.L. Boguszewski, B. Carlsson, L.M. Carlsson, Mechanisms of growth failure innon-growth-hormone deficient children of short stature, Horm. Res. 48 (Suppl. 4)(1997) 19–22.

[17] A. Ornoy, S. Giron, R. Aner, M. Goldstein, B.D. Boyan, Z. Schwartz, Gender dependenteffects of testosterone and 17 beta-estradiol on bone growth andmodelling in youngmice, J. Bone Miner. Res. 24 (1994) 43–58.

[18] Z. Schwartz, W.A. Soskolne, T. Neubauer, M. Goldstein, S. Adi, A. Ornoy, Direct andsex-specific enhancement of bone formation and calcification by sex steroids infetal mice long bone in vitro (biochemical andmorphometric study), Endocrinology129 (1991) 1167–1174.

[19] P. Raz, E. Nasatzky, B.D. Boyan, A. Ornoy, Z. Schwartz, Sexual dimorphism of growthplate prehypertrophic and hypertrophic chondrocytes in response to testosteronerequires metabolism to dihydrotestosterone (DHT) by steroid 5-alpha reductasetype 1, J. Cell. Biochem. 95 (2005) 108–119.

[20] E. Nasatzky, Z. Schwartz, W.A. Soskolne, B.P. Brooks, D.D. Dean, B.D. Boyan, A. Ornoy,Evidence for receptors specific for 17 beta-estradiol and testosterone in chondrocytecultures, Connect. Tissue Res. 30 (1994) 277–294.

[21] M. Zhong, D.H. Carney, B.D. Boyan, Z. Schwartz, 17beta-estradiol regulates ratgrowth plate chondrocyte apoptosis through amitochondrial pathway not involvingnitric oxide or MAPKs, Endocrinology 152 (2011) 82–92.

[22] J. Ekstein, E. Nasatzky, B.D. Boyan, A. Ornoy, Z. Schwartz, Growth-platechondrocytes respond to 17beta-estradiol with sex-specific increases in IP3 andintracellular calcium ion signalling via a capacitative entry mechanism, Steroids70 (2005) 775–786.

[23] V.L. Sylvia, T. Hughes, D.D. Dean, B.D. Boyan, Z. Schwartz, 17beta-estradiol regulation ofprotein kinase C activity in chondrocytes is sex-dependent and involves nongenomicmechanisms, J. Cell. Physiol. 176 (1998) 435–444.

[24] Z. Schwartz, P.A. Gates, E. Nasatzky, V.L. Sylvia, J. Mendez, D.D. Dean, B.D. Boyan, Effectof 17 beta-estradiol on chondrocyte membrane fluidity and phospholipid metabolism

is membrane-specific, sex-specific, and cell maturation-dependent, Biochim. Biophys.Acta 1282 (1996) 1–10.

[25] V.L. Sylvia, B.D. Boyan, D.D. Dean, Z. Schwartz, Themembrane effects of 17β-estradiolon chondrocyte phenotypic expression aremediated by activation of protein kinase Cthrough phospholipase C and G-proteins, J. Steroid Biochem. Mol. Biol. 73 (2000)211–224.

[26] G.L. Greene, L.E. Closs, H. Fleming, E.R. DeSombre, E.V. Jensen, Antibodies to estrogenreceptor: immunochemical similarity of estrophilin from various mammalian spe-cies, Proc. Natl. Acad. Sci. U. S. A. 74 (1977) 3681–3685.

[27] G.L. Greene, N.B. Sobel, W.J. King, E.V. Jensen, Immunochemical studies of estrogenreceptors, J. Steroid Biochem. 20 (1984) 51–56.

[28] J.H. Krege, J.B. Hodgin, J.F. Couse, E. Enmark,M.Warner, J.F.Mahler,M. Sar, K.S. Korach,J.A. Gustafsson, O. Smithies, Generation and reproductive phenotypes of mice lackingestrogen receptor beta, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 15677–15682.

[29] S. Mosselman, J. Polman, R. Dijkema, Er beta: Identification and characterizationof a novel human estrogen receptor, FEBS Lett. 392 (1996) 49–53.

[30] G.B. Tremblay, A. Tremblay, N.G. Copeland, D.J. Gilbert, N.A. Jenkins, F. Labrie, V.Giguere, Cloning, chromosomal localization, and functional analysis of the murineestrogen receptor beta, Mol. Endocrinol. 11 (1997) 353–365.

[31] S.M. Cowley, S. Hoare, S. Mosselman, M.G. Parker, Estrogen receptors α and β formheterodimers on DNA, J. Biol. Chem. 272 (1997) 19858–19862.

[32] K. Pettersson, K. Grandien, G.G. Kuiper, J.A. Gustafsson,Mouse estrogen receptor betaforms estrogen response element-binding heterodimers with estrogen receptor α,Mol. Endocrinol. 11 (1997) 1486–1496.

[33] D.W. Schomberg, J.F. Couse, A.Mukherjee, D.B. Lubahn,M. Sar, K.E.Mayo, K.S. Korach,Targeted disruption of the estrogen receptor-alpha gene in female mice: characteri-zation of ovarian responses and phenotype in the adult, Endocrinology 140 (1999)2733–2744.

[34] O. Vidal, M.K. Lindberg, K. Hollberg, D.J. Baylink, G. Andersson, D.B. Lubahn, S.Mohan, J.A. Gustafsson, C. Ohlsson, Estrogen receptor specificity in the regulationof skeletal growth and maturation in male mice, Proc. Natl. Acad. Sci. U.S.A. 97(2000) 5474–5479.

[35] K. Paech, P. Webb, G.G. Kuiper, S. Nilsson, J. Gustafsson, P.J. Kushner, T.S. Scanlan,Differential ligand activation of estrogen receptors ERalpha and ERbeta at AP1sites, Science 277 (1997) 1508–1510.

[36] B.D. Boyan, K.L. Wong, L. Wang, H. Yao, R.E. Guldberg, M. Drab, H. Jo, Z. Schwartz,Regulation of growth plate chondrocytes by 1,25-dihydroxyvitamin D3 requirescaveolae and caveolin-1, J. Bone Miner. Res. 21 (2006) 1637–1647.

[37] M. Razandi, G. Alton, A. Pedram, S. Ghonshani, P. Webb, E.R. Levin, Identification ofa structural determinant necessary for the localization and function of estrogenreceptor alpha at the plasma membrane, Mol. Cell. Biol. 23 (2003) 1633–1646.

[38] J.E. Bleasdale, G.L. Bundy, S. Bunting, F.A. Fitzpatrick, R.M. Huff, F.F. Sun, J.E. Pike, In-hibition of phospholipase C dependent processes by U73122, Adv. ProstaglandinThromboxane Leukot. Res. 19 (1989) 590–593.

[39] S. Morelli, R. Boland, A.R. de Boland, 1,25(OH)2-vitamin D3 stimulation of phospholi-pases C and D inmuscle cells involves extracellular calcium and a pertussis-sensitiveG protein, Mol. Cell. Endocrinol. 122 (1996) 207–211.

[40] C.M. Klinge, K.A. Blankenship, K.E. Risinger, S. Bhatnagar, E.L. Noisin, W.K.Sumanasekera, L. Zhao, D.M. Brey, R.S. Keynton, Resveratrol and estradiol rapidlyactivate mapk signaling through estrogen receptors alpha and beta in endothelialcells, J. Biol. Chem. 280 (2005) 7460–7468.

[41] W.R. Harrington, S. Sheng, D.H. Barnett, L.N. Petz, J.A. Katzenellenbogen, B.S.Katzenellenbogen, Activities of estrogen receptor alpha- and beta-selective ligandsat diverse estrogen responsive gene sitesmediating transactivation or transrepression,Mol. Cell. Endocrinol. 206 (2003) 13–22.

[42] S. Liu, C. Le May, W.P. Wong, R.D. Ward, D.J. Clegg, M. Marcelli, K.S. Korach, F.Mauvais-Jarvis, Importance of extranuclear estrogen receptor-α and membrane Gprotein-coupled estrogen receptor in pancreatic islet survival, Diabetes 58 (2009)2292–2302.

[43] K.E. Akerman, R. Shariatmadari, J. Krjukova, K.P. Larsson,M.J. Courtney, J.P. Kukkonen,Ca2+-dependent potentiation of muscarinic receptor-mediated Ca2+ elevation, CellCalcium 36 (2004) 397–408.

[44] L. Li, M.P. Haynes, J.R. Bender, Plasma membrane localization and function of theestrogen receptor alpha variant (ERα46) in human endothelial cells, Proc. Natl.Acad. Sci. U.S.A. 100 (2003) 4807–4812.

[45] P.K. Smith, R.I. Krohn, G.T. Hermanson, A.K. Mallia, F.H. Gartner, M.D. Provenzano,E.K. Fujimoto, N.M. Goeke, B.J. Olson, D.C. Klenk, Measurement of protein usingbicinchoninic acid, Anal. Biochem. 150 (1985) 76–85.

[46] B.D. Boyan, Z. Schwartz, L.D. Swain, D.L. Carnes Jr., T. Zislis, Differential expressionof phenotype by resting zone and growth region costochondral chondrocytes invitro, Bone 9 (1988) 185–194.

[47] Z. Schwartz, D.L. Schlader, V. Ramirez, M.B. Kennedy, B.D. Boyan, Effects of vitamin-dmetabolites on collagen production and cell-proliferation of growth zone and restingzone cartilage cells-invitro, J. Bone Miner. Res. 4 (1989) 199–207.

[48] E. Nasatzky, D. Grinfeld, B.D. Boyan, D.D. Dean, A. Ornoy, Z. Schwartz, Transforminggrowth factor-beta1modulates chondrocyte responsiveness to 17β-estradiol, Endo-crine 11 (1999) 241–249.

[49] Y.A. Hannun, R.M. Bell, Phorbol ester binding and activation of protein kinase c on tri-ton x-100 mixed micelles containing phosphatidylserine, J. Biol. Chem. 261 (1986)9341–9347.

[50] R.M. Bell, Y.A. Hannun, C.R. Loomis, Mechanism of regulation of protein kinase Cby lipid second messengers, Symp. Fundam. Cancer Res. 39 (1986) 145–156.

[51] J. McMillan, S. Fatehi-Sedeh, V.L. Sylvia, V. Bingham, M. Zhong, B.D. Boyan, Z.Schwartz, Sex-specific regulation of growth plate chondrocytes by estrogen isvia multiple MAP kinase signaling pathways, Biochim. Biophys. Acta 1763(2006) 381–392.

1172 K.B.Y. Elbaradie et al. / Biochimica et Biophysica Acta 1833 (2013) 1165–1172

[52] Z. Schwartz, E.J. Graham, L. Wang, S. Lossdorfer, I. Gay, T.L. Johnson-Pais, D.L.Carnes, V.L. Sylvia, B.D. Boyan, Phospholipase A2 activating protein (PLAA) isrequired for 1α,25(OH)2D3 signaling in growth plate chondrocytes, J. Cell. Physiol.203 (2005) 54–70.

[53] L. Bjornstrom, M. Sjoberg, Mechanisms of estrogen receptor signaling: convergence ofgenomic andnongenomic actions on target genes,Mol. Endocrinol. 19 (2005) 833–842.

[54] A. Pedram, M. Razandi, R.C. Sainson, J.K. Kim, C.C. Hughes, E.R. Levin, A conservedmechanism for steroid receptor translocation to the plasma membrane, J. Biol.Chem. 282 (2007) 22278–22288.

[55] E.R. Levin, Cellular functions of plasma membrane estrogen receptors, Steroids 67(2002) 471–475.

[56] A. Uittenbogaard, E.J. Smart, Palmitoylation of caveolin-1 is required for cholesterolbinding, chaperone complex formation, and rapid transport of cholesterol to caveolae,J. Biol. Chem. 275 (2000) 25595–25599.

[57] R.A. Chaudhri, R. Olivares-Navarrete, N. Cuenca, A. Hadadi, B.D. Boyan, Z.Schwartz, Membrane estrogen signaling enhances tumorigenesis and metastaticpotential of breast cancer cells via estrogen receptor-alpha36 (ERα36), J. Biol.Chem. 287 (2012) 7169–7181.