Novel Ligands Balance Estrogen Receptor β and α Agonism for Safe and Effective Suppression of the...

Novel Ligands Balance Estrogen Receptor � and �

Agonism for Safe and Effective Suppression of theVasomotor Response in the Ovariectomized FemaleRat Model of Menopause

Chelsea McGregor, Andrea Sau, Samantha C. Ruddy, Dan Leung, Murray Webb,Tony Durst, James S. Wright, Diane Lagace, and M.A. Christine Pratt

Department of Cellular and Molecular Medicine (C.M., A.S., S.C.R., D.La., M.A.C.P.), and University ofOttawa Neuroscience Institute (D.La.), University of Ottawa, Ottawa, Ontario, Canada K1H 8M5; Centrefor Drug Research and Development (D.Le., M.W.), Vancouver, British Columbia, Canada V6T 1Z3;Department of Chemistry (T.D.), University of Ottawa, Ottawa, Ontario, Canada K1N 6N5; andDepartment of Chemistry (J.S.W.), Carleton University, Ottawa, Ontario, K1S 5B6 Canada

Vasomotor thermo-dysregulation (hot flashes) are an often debilitating symptom of menopause.Effective treatment is achieved primarily through activation of the estrogen receptor (ER)� withestrogens but is also associated with increased risk for breast and uterine cancer. In this study, wehave tested novel compounds lacking the B ring of 17-hydroxy-�-estradiol (E2) (A-CD compounds)with differing ratios of ER�:ER� binding affinities for the ability to reduce diurnal/nocturnal tail-skin temperatures (TSTs) in the ovariectomized female rat menopausal hot flash model. Normalmammary tissue expresses the predominantly antiproliferative ER�. Therefore, we hypothesizedthat a preferential ER� agonist with fractional ER� activity would safely reduce TSTs. The A-CDcompound, L17, is a preferential ER� agonist that has a ratio of ER�:ER� binding affinity relativeto E2 of 9.3 (where ER�:ER� for E2, 1.0). In the ovariectomized rat, daily administration of low doses(1 mg/kg) of the A-CD compound TD81 (ER�:ER� relative affinity, 15.2) was ineffective in temper-ature regulation, whereas L17 showed a trend toward TST reduction. Both E2 and the A-CD com-pound, TD3 (ER�:ER� relative affinity, 5.0), also reduced TSTs but had marked proliferative effectson mammary and uterine tissues. At 2 mg/kg, L17 strongly reduced TSTs even more effectively thanE2 but, importantly, had only minimal effect on uterine weight and mammary tissues. Both E2- andL17-treated rats showed similar weight reduction over the treatment period. E2 is rapidly metab-olized to highly reactive quinones, and we show that L17 has 2-fold greater metabolic stability thanE2. Finally, L17 and E2 similarly mediated induction of c-fos expression in neurons within the ratthermoregulatory hypothalamic median preoptic nucleus. Thus, the A-CD compound, L17, mayrepresent a safe and effective approach to the treatment of menopausal hot flashes.(Endocrinology 155: 2480–2491, 2014)

Studies by the Women’s Health Initiative in 2002 (1)and the United Kingdom’s Million Women Study in

2003 (2) revealed that the overall health risks associatedwith hormone replacement therapy (HRT) exceeded itsbenefits, in that women who use HRT are more likely todevelop breast cancer than those who do not. Approxi-mately 70% of women develop episodes of hot flashes

(vasomotor symptoms) resulting from decreased circulat-ing estrogen in the perimenopausal period and after meno-pause (3, 4). Although the physiology remains unclear, hotflashes are thought to arise due to disturbances in hypo-thalamic thermoregulation. Although the incidence of hotflashes decreases within a year, about 20% of women con-tinue to have these symptoms for up to 15 years after the

ISSN Print 0013-7227 ISSN Online 1945-7170Printed in U.S.A.Copyright © 2014 by the Endocrine SocietyReceived October 22, 2013. Accepted April 28, 2014.First Published Online May 13, 2014

Abbreviations: AF, activating function; E2, 17-hydroxy-�-estradiol; ER, estrogen receptor;ERE, estrogen-response element; �ERKO, ER� KO; HLM, human liver microsome; HRT,hormone replacement therapy; KO, knockout; mER, membrane receptor for E2; MnPO,median preoptic nucleus; OVX, ovariectomy, ovariectomized; TST, tail-skin temperature.

G E N E R A L E N D O C R I N O L O G Y

2480 endo.endojournals.org Endocrinology, July 2014, 155(7):2480–2491 doi: 10.1210/en.2013-1976

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2014. at 22:27 For personal use only. No other uses without permission. . All rights reserved.

onset of menopause. Hot flashes are often debilitating andassociated with sleep disturbances (5, 6). Treatment ofmenopausal symptoms is achieved with various prepara-tions of estrogens, which, unopposed by progesterone,also predisposes to uterine hyperplasia and cancer (7).Therefore, there is a clear need for an alternative therapythat addresses the common and often debilitating effectsof menopausal hot flashes without increasing the risk ofbreast cancer.

The estrogen receptors (ERs) belong to the family ofligand-dependent nuclear receptor transcription factorsthat contain DNA and ligand binding domains and 2 tran-scriptional activation functions (AF), N-terminal AF-1and C-terminal AF-2 (8). There are 2 subtypes called ER�

and ER�, which can either form homodimers or het-erodimers bound to DNA elements. Although the ER� hasa strong ligand-independent AF-1 region, the ER� AF-1function is weaker (9). There is considerable overlap inDNA-response element binding between the 2 receptors,although elements that preferentially bind to � or � havebeen described (10). The ERs can either transactivate orsilence promoters by binding coactivators or corepressorproteins, respectively. These interactions are both pro-moter and ligand dependent (11, 12). An important find-ing from studies of ER� knockout (KO) (�ERKO) micewas that 17-hydroxy-�-estradiol (E2) remains effectivefor symptoms of osteoporosis after ovariectomy (13). Thissuggests that the ER� may be able to compensate for theER� in this tissue, at least in the absence of ER�. In thenormal human and rodent mammary gland, the ER� isexpressed at higher levels than the ER� in the mammaryluminal epithelium. However, this ratio is reversed in ER�

breast cancer, where the ER� is the dominant subtype,although many breast cancers continue to express lowlevels of ER�. In the normal mammary gland, ER� nuclearstaining is found in nearly all cells in the inner layer (lumenlining) in acini and in isolated cells in the outer layer ofinterlobular ducts. ER� is expressed in the nuclei of epi-thelial cells in both inner and basal layers of acini and ductsas well as in stromal cells (14, 15). Importantly, mitoticindex is strongly increased in the �ERKO mammaryglands (14), whereas ER� activation has inhibitory effectson cyclin D1 (16). These findings, along with the fact thatthe ER� is constitutively expressed in the mammary gland,suggest that ER� plays an essential role in the normalmammary gland organization and maintenance of the dif-ferentiated phenotype (17). Consistent with this, severalgroups have demonstrated that ectopic expression of theER� in ER�� breast cancer cells results in growth inhi-bition, although the mechanism by which this occurs maybe cancer cell dependent (18, 19).

In addition to nuclear ERs, membrane receptors for E2(mERs) have been described that are able to mediate rapidhormonal responses. E2 stimulation results in rapid acti-vation of kinases and the release of calcium in some cellsthat appears to be most relevant for signaling in the brainand may be responsible for many of the effects of E2 onneurotransmission. In addition, membrane receptors havebeen revealed to potentiate nuclear transcription by ERs inthe central nervous system. The identity of E2-responsivemembrane receptors remains under study. However, itappears that different types of receptors/complexes maybe relevant for individual regions of the brain. Both theanorectic and thermoregulatory effects of E2 are thoughtto be a consequence of neuronal stimulation, at least inpart, through mERs (20). Nevertheless, the nuclear ERsare expressed in various structures of the brain, includingthe hypothalamus (arcuate nucleus and medial preopticregion) (21), where estrogen mediates regulation of energyhomeostasis and body temperature (22). Moreover, thereis substantive evidence for collaboration between mem-brane-initiated signaling and transcriptional responsethrough the ER (20, 22).

Strategies toward hormone replacement require thatthe agent has low toxicity, because use is to be continuedlong term and in otherwise healthy individuals. To thisend, we have developed and synthesized a library of com-pounds that are based on the ABCD-ring structure of E2but unlike E2 lack the B ring. In addition, we have dem-onstrated that some of these compounds, which we havenamed A-CD estrogens, showed strong hormonal potencycoupled with a spectrum of receptor selectivity (23, 24).Substituent atoms or chemical groups at positions on theA ring are predicted to resist metabolism to quinones (23),which are potential mutagens (25, 26). The reaction ofquinones with DNA can form depurinating adducts, anderror-prone base excision repair of depurinated sites maylead to the mutations that can initiate cancer (27–29).Estrogen was listed as a carcinogen by the United StatesNational Toxicology Program (30). Unlike the vast ma-jority of ER�-selective ligands, which are structurally un-related to E2 (reviewed in Ref. 31), L17 is predicted to bemore stable (24).

Because the ER� is widely expressed in normal mam-mary tissue, where it is predominantly antiproliferative,we predicted that compounds with significant ER� activ-ity would prevent proliferation in these tissues. However,selective ER� activity might be insufficient to regulatetemperature. Thus, we hypothesized that the ideal dual-purpose compound would have fractional ER� activitysufficient to alleviate temperature dysregulation, whereasstrongly activating the ER� to afford protection againstproliferation in mammary and uterine tissues. To test this

doi: 10.1210/en.2013-1976 endo.endojournals.org 2481


novel strategy, we have assessed different A-CD com-pounds having a spectrum of ER�:ER� affinities for theirability to restore thermoregulation based on diurnal/noc-turnal reductions in tail-skin temperatures (TSTs) in theovariectomized (OVX) rat model of menopausal vasomo-tor dysregulation.

Materials and Methods

Luciferase assaysHEK293 cells were plated in 6-well dishes and immediately

cotransfected with expression plasmids (pcDNA3) encoding theER� or ER�, the estrogen-response element (ERE)-thymidinekinase luciferase plasmid together with the pRL-TK vector con-taining Renilla luciferase (Promega). Transfections were doneusing polyethylenimine (Polysciences, Inc), and the next day,cultures were treated with compounds. Forty-eight hours later,whole-cell lysates were collected, and firefly and Renilla lu-ciferase activities were measured using the Dual Luciferase Re-porter Assay (Promega) according to the manufacturer’s proto-col. All transfections were performed in triplicate and repeateda minimum of twice. Wyeth compound 200070 (WAY200070)was purchased from Sigma Chemicals.

OVX rat model and TST telemetryOvariectomy in rats results in a failure to lower TST during

the nocturnal active phase. Previous reports have established atelemetry method using a surgically implanted temperatureprobe and transmitter to continuously monitor TST (32, 33). Weused a slightly modified method of those previously reported.OVX female Wistar rats (�250 g) were purchased from CharlesRiver. Rats were 2 weeks or more after OVX before surgicalimplantation. Rats were housed on a 12-hour light (6 AM to 6 PM),12-hour dark cycle at 21°C and provided water and food adlibitum. For TST measurements, a telemetric transmitter(Physio-Tel TA10TA-F40; Data Sciences International) was im-planted sc in the dorsal scapular region, and the tip of the tem-perature probe was inserted beneath the skin 2.0 cm beyond thebase of the tail. After a 7-day recovery period, continuous TSTmeasurements were initiated and recorded every 30 minutes toestablish a baseline over 3 days. Rats were injected sc with ve-hicle, E2 (Sigma Chemicals), or an A-CD estrogen synthesized aspreviously described (23, 24). All ligands were dissolved in100% ethanol as a stock solution, diluted in saline, and mixed1:1 with cottonseed oil for injection. Subcutaneous injectionswere begun 0.5 hours before the onset of the dark phase at theend of day 3 of TST recording and each day for another 4 days(5 treatments total). Animals weights were recorded daily beforetreatment. For some rats, visceral fat and/or uteri were removedand weighed immediately after killing. Handling and treatmentof animals was carried out in accordance with the standardsapproved by the Animal Care Committee at the University ofOttawa and in compliance with guidelines from the CanadianCouncil on Animal Care.

Statistical analysisComparisons between treatment groups were done using

GraphPad Prism 5.0 statistical software. In order to reduce the

potential for outlier perturbation of data, we performed linearregression analysis on TSTs for the entire experimental period toobtain trendline slopes. To calculate TST changes, the mean ofthe 3 pretreatment nocturnal periods (6:30 PM to 5:30 AM) wassubtracted from the nocturnal means of each treatment day.One-way ANOVA and a post hoc Bonferroni test were per-formed to determine significant differences between treatmentgroups. A minimum 95% confidence level was considered sig-nificant. Luciferase dose-responses curves were generated usingnonlinear regression best fit for triplicate values at each concen-tration (95% confidence).

Rat mammary whole-mount analysis andimmunohistochemistry

Inguinal mammary glands were dissected from rats andspread as thinly as possible onto glass microscope slides. Glandswere fixed overnight by laying them flat in a container containingCarnoy’s Fixative and then washed for 15 minutes in each of thefollowing solutions: 70% ethanol, 50% ethanol, 25% ethanol,and then ddH2O. Glands were defatted with 3 incubations inacetone for 30 minutes, followed by 100% and then 95% eth-anol for 20 minutes each. Glands were incubated in hematoxylinfor approximately 6 hours at room temperature to allow the stainthe fully penetrate, rinsed in tap water, and destained with 2 acidalcohol washes, first for 1 hour and then overnight. Fixed glandswere washed for 20 minutes each in the following solutions todehydrate: 70%, 95%, and 100% ethanol. Slides were stored inxylene until imaging with a dissecting microscope. Formalin-fixed paraffin-embedded mammary glands were sectioned, and10mM sodium citrate (pH 6) was used for antigen retrieval.Sections were then blocked for 1 hour in blocking solution (2%BSA, 0.05% Triton X-100). Primary antibody Ki67 (16667; Ab-cam) was diluted in antibody dilution buffer (S0809; DAKO)and incubated overnight. Secondary antibody (K4002 antirabbitIgG; DAKO) was added for 1 hour. Sections were then counter-stained with hematoxylin and dehydrated before adding per-mount mounting media.

Histochemical analysis of uteriRat uteri were fixed in formalin and then embedded in par-

affin, and 7-�m sections were stained with hematoxylin andeosin. Images were obtained using a Zeiss Axiophot microscopeequipped with Northern Eclipse 3.0 software.

Microsomal stability assay procedureThe metabolic stability of E2 and L17 was evaluated in hu-

man liver microsomes (HLMs) (BD Biosciences). The reactionmixture was prepared by combining 0.6 mg/mL of HLM with94mM potassium phosphate, 4mM magnesium chloride, anddistilled water. Reactions were initiated by the addition of eitherE2 or L17 from ethanol stock solutions to a final concentrationof 2�M compound, and less than 0.02% ethanol, in the presenceor absence of 1mM NADPH and incubated at 37°C. At theappropriate time points, 100 �L of the reaction were withdrawnfrom the incubation and added to 100 �L of 2% formic acid,vortexed, and then stored at �80°C until extraction and anal-ysis. To identify metabolites of E2 or L17, the HLM stability wasrun as described above except that the compound was incubatedat a final concentration of 20�M and that the reactions wereterminated with 0.1% formic acid in acetonitrile. After termi-

2482 McGregor et al Balancing ER�/ER� Agonism to Treat the Hot Flash Endocrinology, July 2014, 155(7):2480–2491


nation, samples were centrifuged at 4°C and 18 000g for 15minutes. Finally, the supernatant was transferred to a 96-wellhigh pressure liquid chromatography sample plate and ana-lyzed by ultra performance liquid chromatography-massspectrometry.

Rat brain sections and c-fos immunohistochemistryOVX rats (3 per group) were treated 3 consecutive days at 5

PM with vehicle, 1 mg/kg of E2, or 2 mg/kg of L17. At 10 AM onthe fourth day, rats were killed using sodium pentothal, andbrains were perfused with 4% paraformaldehyde (wt/vol) in0.1M PBS and cryoprotected in 30% sucrose (wt/vol) in 0.1MPBS and 0.1% sodium azide (wt/vol). The brain was sectionedcoronally on a freezing microtome (Leica) at 30 �m, and 9 serialsets of sections were stored in 0.1% NaN3 in 1� PBS at 4°C untilprocessing. For staining, 1 series of sections was mounted onglass slides (Superfrost/Plus; Fisher) and dried overnight. Slideswere coded to ensure objectivity, and the code was not brokenuntil after data collection was complete. For immunohistochem-istry, sections were incubated in rabbit anti-c-fos primary anti-body (1:100, sc-52; Santa Cruz Biotechnology, Inc) in 1% nor-mal donkey serum (vol/vol) in 0.2% Tween 20 in 1� PBS for 72hours at 4°C. Subsequently, sections were incubated with bioti-nylated-donkey antirabbit IgG (1:250; Sigma) in 1% normaldonkey serum (vol/vol) in 0.2% Tween 20 in 1� PBS for 60minutes, 0.3% hydrogen peroxide (Sigma Laboratories) for 30minutes, cupric sulfate for 30 minutes, and avidin-biotin com-plex (Vector Laboratories) for 60 minutes. Staining of c-fos im-munoreactive cells was visualized using diaminobenzidine(Pierce) for 20–30 minutes, and Nuclear FastRed (Vector Lab-oratories) was used as a counterstain. c-fos-positive neuronswere enumerated on sections between bregma �0.7 to �0.9within a measured area from the region of the median preopticnucleus (MnPO) between the anterior commissure and third ven-tricle. Areas circumscribing blood vessels were subtracted fromthe total area. Cells called positive were enumerated blindly us-ing ImageJ on all sections at a consistent preestablished mini-mum staining intensity. Positive neurons were expressed as cellsper square millimeter.

Results

A-CD compoundsThe structures of TD-17 (L17), TD-81 (TD81), TD-3

(TD3), and E2 are shown in Figure 1A. The ratios of ER�:ER� binding affinities relative to E2 are shown in Figure1B, and values for individual ER affinities relative to E2are shown in Table 1. TD3 binds more strongly to the ER�

than the ER� but still has approximately 25% of the af-finity for the ER� as does E2. L17 and TD81 have nearlya 1000-fold less affinity for the ER�, although the ER�

affinity of L17 is about 70% of TD81. L17 and TD81 have0.2- and 0.4-fold affinity relative to E2 for the ER�, re-spectively. Thus, the rank order of binding preference forthe ER� over the ER� is TD3 � L17 � TD81.

We next assessed the ability of the A-CD ligands toactivate transcription via each of the ERs from a canonical

E2-response element. HEK293 cells were cotransfectedwith either the ER� or the ER� along with the ERE-lu-ciferase reporter plasmid and assayed after 48 hours ofexposure to ligand at increasing concentrations. Figure 1,C and D, and E and F, shows the ER� and ER� transac-tivation profiles, respectively, for these compounds rela-tive to E2. EC50 values using E2 as a standard are indicated(EC50E2). The ratio of ER binding affinities correlatedwith receptor preference for transcriptional activation.However, transcriptional activation EC50E2 from the ca-nonical ERE by L17 and TD3 occurred at 770- and 6.3-fold, respectively, that of E2. TD81 and the ER�-selectiveagonist, WAY200070, produced negligible transactiva-tion. The EC50E2 on the ER� for L17 and WAY200070was 48.5-fold, whereas that of TD81 and TD3 was 34-and 2.5-fold more than E2.

Effects of A-CD estrogens on OVX rat TSTFemale rats undergo diurnal temperature oscillation,

such that temperatures are highest during the 12-hourlight phase but are reduced during the 12-hour dark phase.A menopause-like state is simulated by ovariectomy, andit is proposed that the lack of estrogen lowers the set pointfor the activation of heat-dissipating mechanisms, whichwould result in an overall increase in TST, especially dur-ing the active nocturnal phase (see Ref. 34 and referencestherein). Therefore, the potential of an agent to be used toreplace E2 in this capacity can be assessed by its ability toreduce circadian TSTs. We tested the ability of A-CD es-trogens to restore thermoregulation by sc injection of 1–2mg/kg of compound once daily for 5 days into OVX ratsimplanted with a temperature sensor under the tail skincoupled to a transmitter receiver that recorded tempera-tures every 30 minutes. Baseline TSTs were recorded for 3days before the first injection. The mean nocturnal phasetemperatures were calculated from pretreatment days 1–3and for each nocturnal treatment days 1–5 (8 d total).Figure 2A shows the temperature differences between pre-treatment means and each treatment nocturnal mean forvehicle and 1 mg/kg of TD81, L17, E2, and TD3. Vehicle-treated animals showed a rise in nocturnal TST after thefirst injection, which remained steady over the treatmentperiod. A smaller increase in temperature was observed inTD81-treated rats (n � 3) but trended upward with morevariation between animals. Dosing with 1 mg/kg of L17(n � 5) resulted in a small but opposite trend in TST. Asexpected, 1 mg/kg of E2 (n � 3) strongly reduced TST byapproximately 1.5°C by the second night of treatment.TD3 (n � 3) also clearly reduced TSTs but with less effi-cacy compared with E2.

To assess whether reduced nocturnal minimum tem-peratures affected overall mean daily temperatures, linear



regression lines were calculated from pretreatment day 1to treatment day 5 (8 d total) across sequential diurnal andnocturnal periods. The mean regression line slope fromeach group of animals in Figure 2A is shown in Figure 2B.

Consistent with the nocturnal change in TST in treatmentgroups, the mean slopes of the TSTs from vehicle-treatedan increased across the period. In contrast, both 1-mg/kgE2 and TD3 resulted in negative trendline slopes over the

Figure 1. A, Structures of A-CD compounds and E2. B, Graph depicting the ratio of ER� to ER� binding relative to estrogen (E2) for TD-3, L17,and TD81. E2 binding is set at 100% for both ER� and ER�. C–F, Dose-response curves for A-CD compounds HEK293 cells were cotransfectedwith the consensus ERE-thymidine kinase-luciferase gene, an expression plasmid for either ER� or ER�, and a Renilla luciferase internal control.Cultures were treated with increasing concentrations of each compound for a total of 48 hours. Results were plotted showing arbitrary luciferaseunits and analyzed using GraphPad Prism 5.0. The concentration of drug required for half-maximal E2 response (EC50E2) is shown. WAY 200070was used as an ER�-selective control. Dose responses on the ER� are shown (C) for L17 and (D) for TD81 and TD3. Dose responses on the ER� areshown (E) for L17 and (F) for TD81 and TD3. Values are the means of triplicate determinations with less than or equal to 6% variation.



period of treatment. L17 (ER�-preferential/ER� partial)produced a small but negative mean slope value over,whereas TD81 (strongly ER�-preferential) resulted in asimilar magnitude but opposite effect on the mean slope.

Because L17 has a low affinity for the ER� relative toE2 but showed a promising trend toward TST reductionover the 5-day treatment period, we surmised that an in-creased dose of L17 might be more effective at TST re-duction. Therefore, we increased the dose of L17 to 2mg/kg and compared this with 2 mg/kg of E2 or the samevolume of vehicle in a larger sample size of rats (n � 6 pergroup). Again, the means of the 3 pretreatment nocturnalphases were determined, and differences between pre-treatment and treatment day nocturnal phase means werecalculated. The results in Figure 2C show that, again, ve-hicle-treated rats showed an increasing trend in TST thatplateaued after 3 days. In marked contrast, rats treatedwith either E2 and to a greater extent, L17, had reducednocturnal TSTs that peaked by 4 days of treatment. Treat-ment groups were then compared on individual days byANOVA. Both L17 and E2 produced a significant meanreduction in TST, which was maximal on day 4 at ap-proximately 1.75°C.

Examples of TST curves over the entire 3-day pretreat-ment and 5-day treatment are shown with regression linesin Figure 2, E–G, and mean slopes are graphed in Figure2H. At this dose, L17 treatment resulted in regression lineslopes with negative values that exceeded those of E2 at thesame dose. As expected, vehicle-treated rat TSTs contin-ued to increase over the experimental period. The lessereffect of E2 in our experiments compared with L17 may bea function of significant ER� down-regulation resultingfrom sustained high-dose E2 (35). Thus, E2 and TD3 at adaily dose of 1 mg/kg and L17 at 2 mg/kg are effective atsignificantly reducing TSTs in the OVX rat model over 5days of treatment.

A potential issue with long-term E2-based HRT is theinherent metabolic instability of E2, which upon succes-sive oxidations results in reactive quinone formation andthe generation of DNA adducts (27, 28). These may becarcinogenic events, particularly in proliferating breastand uterine. Therefore, increased metabolic stabilitywould be indicative of a reduced propensity for quinoneformation. Given its efficacy in TST reduction, we nextassessed the metabolic stability of L17. Using a liver mi-crosome assay, we measured the half-life of L17 and com-pared it with that of E2. The graphs in Figure 2, I and J,show decay curves in the presence or absence of NADPHand indicate that the half-life of L17 is approximatelytwice that of E2 (37 vs 18 min, respectively). Although thisdoes not directly assess catechol/quinone formation, the

enhanced stability of L17 over E2 reduces the likelihood ofoxidation product formation before elimination.

Effects of A-CD estrogens on the OVX ratmammary gland and uterus

As discussed above, estrogen-based HRTs increase riskof breast cancer and, in the absence of progesterone, en-hance the incidence of endometrial cancer, both of whichare associated with estrogen-induced hyperplasia. There-fore, successful strategies toward alleviation of meno-pausal vasomotor dysregulation would ideally have littleor no proliferative effects on either the mammary gland oruterus. Because both TD3 and L17 significantly reducedTST in this model, we next assessed the effects on uteri andmammary gland of rats treated for 5 days with vehicle, E2(1 mg/kg), TD3 (1 mg/kg), or L17 (2 mg/kg). Figure 3Ashows representative images of uteri taken from each5-day treatment group demonstrating enlarged uteri inboth TD3- and E2-treated rats. In contrast, only a smallincrease in the uterus of L17-treated rats was observed,and virtually no changes were observed in the TD81-treated rats. Uterine wet weights were also determined atthe end of the 5-day treatment. The lowest uterine weightswere observed in vehicle-treated and TD81-treated rat co-horts (Figure 3B). Both TD3 and E2 significantly increaseduterine dimensions and weights by approximately 6-foldcompared with vehicle over the 5-day treatment period.Although 2 mg/kg of L17 also increased uterine weight,the increase was approximately 2-fold greater than vehi-cle. Hematoxylin and eosin-stained uterine sections areshown in Figure 3C. Using a �5 objective, the entire uterusfrom a typical vehicle-treated rat is visible at this magni-fication, and the lumen is lined with small cuboidal epi-thelium. Only a portion of the uterine section from anE2-treated rat is visible at the same magnification. As ex-pected, E2 treatment increased luminal epithelial heightassociated with elongated columnar cell bodies, increasedthe thickness of the endometrial stroma, and promoted thedevelopment of deep endometrial glands. The myometriallayers were also increased in diameter. Interestingly, un-like E2, the increased size of the uterine horns induced byTD3 treatment did not produce the same depth and com-

Table 1. ER Binding Affinities

Ligand � �

E2 100 100TD3 27 135L17 0.186 1.73TD81 0.261 3.97

Binding affinities of AC-D compounds for the ER� and ER� relative toE2. Values are the mean of triplicate determinations as described inRef. 23.



plexity of endometrial gland formation as E2, althoughthe myometrium was similarly increased in size, perhapsreflecting differential ER� and ER� activation relative toE2. In contrast, nearly the entire lumen of L17-treated rats

was visible at this magnification. In some regions, the lu-minal epithelial cells had elongated morphology, and al-though the thickness of the stromal epithelium and myo-metrium was increased relative to vehicle, these regions

Figure 2. Effects of A-CD compounds in the OVX rat TST model and L17 stability. OVX adult female rats were implanted sc between the scapulawith the DSI transmitter, and a temperature probe was inserted under the tail skin as described in Materials and Methods. One week afterimplantation, baseline temperatures were monitored for 3 days (pretreatment), and daily sc injections were initiated at 5 PM on day 4 andcontinued at the same time for 5 consecutive days. TST readings were obtained every 30 minutes. The mean nocturnal TST was subtracted fromthe mean nocturnal TST for each treatment day, and the differences were plotted. A, Graph showing mean differences after treatment withcompounds TD81 (n � 3), L17 (n � 5), TD3 (n � 3), and E2 (n � 3) at 1 mg/kg of each compound or vehicle (veh). B, Graph of the means oflinear regression line slopes derived from TST measurements across the entire experimental treatment period from the animals in A are show onthe graph. C, Graph showing mean differences after treatment with compounds L17 (2 mg/kg) and E2 (2 mg/kg) or vehicle (n � 6). D, Results of ANOVAstatistical analysis with post hoc Bonferroni comparing each daily treatment group in C. Examples of telemetry and linear regression trendlines over thepretreatment and treatment periods are shown for (E) vehicle, (F) E2, and (G) L17. Bars along the abscissa indicate nocturnal phases. H, The mean slopesof regression lines from 6 rats per treatment group in C are plotted. Significance of differences was determined by ANOVA with post hoc Bonferroni test.*, P � .05; ***, P � .0001. Scale bars are SE. Stability of E2 (I) and L17 (J) during incubation in HLMs in the presence or absence of NADPH. Error barsrepresent the SD of 2 replicate values. Regression lines are plotted for a single phase first order decay process.



Figure 3. Effects of A-CD compounds on mammary gland and uterus. A, Rat uteri were isolated after 5 days of treatment. Shown arerepresentative images of gross morphology vehicle (veh) after 5 days of treatment with vehicle, E2 (1 mg/kg), L17 (2 mg/kg), TD3 (1 mg/kg), orTD81 (1 mg/kg). B, Graph of mean wet weights of uteri (Veh, n � 8; TD81, n � 4; TD3, n � 4; E2 1 mg/kg, n � 3; L17 2 mg/kg, n � 8; E2 2 mg/kg, n � 7 rats per group). Vertical bars are SE. Statistical comparisons: ANOVA with Bonferroni post hoc test. *, P � .05; **, P � .01. C,Representative images of uterine hematoxylin and eosin stained paraffin sections from rats treated with the indicated compounds as above.Regions of the uterus are shown in the section from the E2-treated rat. MM, myometrium; ES, endometrial stroma; LE, luminal epithelium; V,vehicle. D, Mammary gland whole-mounts were performed to analyze ductal epithelial proliferation. The fourth inguinal glands were removedfrom rats after day the 5 of treatment. Typical mammary glands from each group are shown (�5 objective). E, Ki67 immunohistochemistry onparaffin sections from rats treated with 2-mg/kg E2 or 2-mg/kg L17 or vehicle for 5 days. Representative images are shown (�40 objective). F,Graph depicting percentage of Ki67-positive cells after E2, L17, or vehicle treatment as in E. A minimum of 300 cells were enumerated from 3separate rat mammary gland sections. Scale bars are SE. *, P � .05; ***, P � .0001; n.s., not significant.



were clearly thinner relative to uteri of E2-treated rats.Moreover, endometrial invaginations were shallow andpoorly developed compared with those in the E2 group.

We also performed whole-mount analysis of mammaryglands (Figure 3D). As expected, mammary glands fromvehicle-treated rats showed fine ductal branching consis-tent with mature virgin females, whereas E2 treatmentresulted in ductal expansion and lobular development.Consistent with preferential binding and transcriptionalactivation of the ER�, TD3 had a similar effect on mam-mary tissue. In contrast, despite its ability to reduce TST,2 mg/kg of L17 had no discernible effect on mammarymorphology. The mammary glands of rats treated with thestrongly selective ER� agonist, TD81, also closely resem-bled those from vehicle-treated animals, although someglands appeared to have a less dense branching pattern. Todirectly assess the effects of compounds on mammary cellproliferation we immunostained paraffin sections withanti-Ki67. The immunohistochemistry in Figure 3E showsthat although vehicle-treated mammary glands had fewproliferating cells, a small but significant increase was ob-served in tissue from L17-treated animals. These resultscontrast with those obtained after E2 treatment, whichresulted in a profound and significant increase in prolif-eration involving more than 50% of cells within the mam-mary ducts (Figure 3D). Consistent with these results, inother work, we have shown that L17 reduces the expres-sion of E2-responsive genes involved in proliferation (cy-clin D1) and antiapoptosis (Bcl-2) in breast cancer cellsexpressing a reduced level of ER� relative to ER� (36).

Overall, none of the A-CD estrogens tested fully reca-pitulated the effects of E2 on the uterus or mammarygland. Importantly, although a 2-mg/kg dose of L17 ef-fectively reduced TSTs in OVX rats, it had a significantlyreduced effect compared with E2 on proliferation of bothuterine and mammary tissue.

Effects of L17 on body mass and adiposityER�KO mice, but not ER�KO mice, demonstrate obe-

sity associated with increased visceral adiposity and de-creased energy expenditure (37, 38), whereas activation ofthe ER� reduces weight gain (39) and food intake (38). Wetherefore tested and compared the effects 2-mg/kg L17,vehicle, and 2-mg/kg E2 on weight gain and adiposity.Rats were weighed before initiation and each day of treat-ment. As shown in Figure 4A, by day 4 of treatment witheither E2 or L17, a significant reduction in weight wasobserved compared with vehicle controls, which wasmaintained into day 5. In addition, visceral fat was re-moved from rats killed after 5 consecutive days of treat-ment, weighed, and expressed as a percentage of bodyweight. No significant changes in adiposity were observed

in any group after the 5-day treatments. These results sug-gest that L17 behaves similar to E2 with respect to atten-uation of weight gain after ovariectomy, whereas effectson adiposity may require longer treatment periods.

L17 activates neurons in the hypothalamic MnPORecently, it was shown that effects of E2 on thermo-

regulation in the rat were associated with activation ofc-fos in neurons within the MnPO (40). Importantly, thisregion is enriched for expression of the ER� but relativelyweak expression of the ER� (41). We therefore predictedthat A-CD compound-mediated reduction in TST wouldbe associated with activation of c-fos expression in MnPOneurons. Brains from rats treated with vehicle, and 2 mg/kgof E2 or L17 for 3 days were collected, sectioned, and sub-jected to immunohistochemistry for c-fos. Figure 5A is alow-power image depicting the region of the brain sur-rounding the MnPO. Representative images of c-fos im-munostaining after 3 days of treatment with 2-mg/kg L17,2-mg/kg E2, or vehicle are shown in Figure 5B and areenumerated in Figure 5C. The results indicate approxi-mately twice the number of c-fos-positive nuclei in theMnPO of E2- and L17-treated rats relative to vehicle con-sistent with E2- and L17-induced neuronal signaling inthis region.

Discussion

A wide variety of therapies targeting the menopausal hotflash have been investigated, ranging from natural prod-ucts, such as soy isoflavones and black cohosh, to a num-ber of synthetic drugs with varying efficacies and side ef-fects. The synthetic prohormone, tibolone, has weakestrogenic and progestinic activity and has been shown toalleviate hot flashes but is associated with increased breastcancer risk (42). Nonhormonal therapies include the �-ad-

Figure 4. L17 attenuates OVX-associated weight increase. A, Ratweights were recorded daily before injections. The weight before thefirst injection (initial weight) of 2-mg/kg E2 or L17 was subtracted fromweights on each treatment day and expressed as a percentage changefrom initial weight. B, Visceral white adipose was removed from rats atkilled and weighed. Adiposity is expressed percent visceral fat:totalbody weight; n � 6 for all experiments. Veh, vehicle. Statisticalcomparisons were performed by ANOVA and post hoc Bonferroni test.***, P � .0001; n.s., not significant.



renergic agonist clonidine, gabapentin, and serotonin re-uptake inhibitors (5, 43). However, their effectivenessdoes not approach that of E2, and their use is associatedwith significant side effects. One report showed that theER�-selective agonist ER�-19 could alleviate the rise inTST in OVX rats (44), suggesting a possible role for ER�

in control of vasomotor symptoms. However, a separatestudy found an ER�-specific ligand inactive in this model(45). Consistent with the latter study, the A-CD-prefer-ential ER� agonist, TD81, did not reverse the increasingTST trend in this model. However, tissue-specific effects ofpreferential or specific ER�/� ligands may not be accu-rately predicted based on receptor binding assays andtranscriptional activation on a consensus reporter geneand likely involve cell-specific interaction with coactiva-tors/repressors on specific genes (46–48). Recently, a syn-thetic ER modulator called bazedoxifene has been foundeffective in the prevention and treatment of osteoporosisbut is an antagonist in uterus and brain (49, 50). A com-bination of conjugated estrogens with bazedoxifene hasbeen evaluated in a 2-year clinical trial and found to beeffective in treatment of hot flushes without untowardeffects to date (51). Bazedoxifene does not stimulate the

proliferation of MCF-7 cells, although it has a very lowbioavailability relative to conjugated estrogens (52), andlong-term effects on breast tissue are not known.

Our study provides evidence for the utility of a novelstrategy for the treatment of menopausal vasomotor dys-regulation that involves using an ER� agonist with partialER� activity that we suggest is sufficient to induce theactivation of neuronal activity in the thermoregulatorycenter of the hypothalamus resulting in reduction in OVXrat TST. Estrogen depletion by OVX in either ER�KO orER�KO mice results in an increase in TST that is sup-pressed by E2 treatment (13). This suggests that only 1 ofthe ERs is required for thermoregulation or, alternatively,that the predominant effect of E2 is mediated through anonnuclear/membrane receptor. Regarding the formerpossibility, because L17 is a preferential ER� agonist, wecannot fully discount the possibility that, at least, some ofthe effect on TST could have been attributed to ER� ac-tivation. However, the inability of TD81 to have any effecton TSTs when compared with L17 at the same concen-tration suggests that this is unlikely. With respect to thelatter possibility, a Gq-coupled mER has also been iden-tified, and recent studies have shown that ligand selectivefor Gq-coupled mER can reduce core body temperature inan OVX guinea pig model (53). This ligand was previouslyshown to inhibit gamma-aminobutyric acid release fromhypothalamic neurons in male ER�/�KO mice by a mech-anism that was prevented by the antiestrogen ImperialChemical Industries-182,780 (54). Whether the recep-tor(s) responsible for thermoregulation in the OVX rat aresimilar is not known. However, in the present study, thedecrease in TSTs after 1 mg/kg of L17 contrast with theincreased TSTs after TD81 dosing indicates that, despitehighly similar chemical structures, TD81 may be less likelyto stimulate mERs than L17. Other studies found that aphytoestrogen also reversed OVX-induced rise in TST inan Imperial Chemical Industries-182,780-sensitive man-ner (55). Based on the limited blood-brain barrier access ofthis ER antagonist, this group attributed at least some ofthis agent’s effects to a peripherally mediated mechanism.

E2 is also well documented to have an anorectic effect,preventing weight gain after OVX through a mechanismthat is inhibited by antiestrogen, which is thought to be theresult of neurotransmission within both the MnPO andthe arcurate nucleus (56). Use of noncell permeable E2-BSA conjugates produces similar effects, supporting a rolefor mERs in this activity (57). Our results suggest that L17can also mediate a significant reduction in weight gainalbeit at a consistently lower level than E2. Aside fromappetite control, E2 signaling through the ER� plays a rolein adiposity illustrated by the finding that female and male�ERKO mice show significant accumulation of white ad-

Figure 5. L17 activates c-fos expression in MnPO neurons. FemaleOVX rats were treated with daily injections of vehicle, 2-mg/kg E2, or2-mg/kg L17 for 3 consecutive days (3 rats per group). Animals werekilled and brains sectioned and immunostained with c-fos as describedin Materials and Methods. A, Low-power image of MnPO region. ac,anterior commisare; 3V, third ventricle. Bar, 200 �m. B, Representativec-fos-immunostained sections from indicated treatment groups. t, tear;bv, blood vessel. Bar, 20 �m. C, A minimum intensity was establishedusing ImageJ, and c-fos-positive MnPO neurons were enumerated overa circumscribed area between the anterior commissure and the thirdventricle. Results are expressed as cells per square millimeter. Scalebars are SE. Comparisons between groups by ANOVA and post hocBonferroni test. *, P � .05.



ipose tissue (37). In contrast, �ERKO mice have normalbody weight and adipose tissue, compared with wild-typemice (38). This suggests that adipose tissue is regulated bynuclear ERs. We did not note any changes in the visceralfat to total weight ratios with either compound, althoughlonger treatment periods may be required to observe al-terations in adipose accumulation.

A critical feature of L17 is the preferential activation ofthe ER�. Because nuclear ER� function principally medi-ates both the proliferative uterine and mammary glandresponses to E2, the ER� signals in large part in oppositionto the ER� in these tissues. Although E2 strongly increaseduterine weight and mammary cell proliferation, these ef-fects were significantly less after L17 treatment. Based onmammary gland whole-mount analysis, L17 appeared tohave no effect on mammary ductal proliferation, whereasthe ER�-preferential agonist, TD3, and E2 resulted inmarked expansion of ductal structures. Although moremammary cells were mitotic after L17 treatment, the per-centage of Ki67-positive cells was significantly less thanafter E2 treatment. Moreover, because L17 is metaboli-cally more stable than E2, it is less likely to form reactivecompounds that, over long-term treatment, might pro-mote DNA damage. Numerous studies have shown ben-eficial effects of ER�-selective ligands in experimentalmodels of endometriosis, inflammatory diseases, includ-ing rheumatoid arthritis, and cardiovascular and centralnervous system conditions (see Ref. 45 and referencestherein). Based on this, long-term treatment with L17 hasthe potential for similar advantages.

In summary, we have developed a ligand with a balanceof ER� and ER� activity that is metabolically more stablethan E2 and which suppresses the vasomotor response toE2 depletion. Critically, as an ER�-preferential agonist,L17 has minimal effects on the primary peripheral estro-gen-responsive target tissues, which are associated withenhanced risk for breast and uterine cancers. We concludethat L17 represents a viable approach for the treatment ofhot flashes and could be of particular benefit in womenwho are at elevated risk for breast cancer.

Acknowledgments

We thank David Carr, Keren Leviel Kumar, and Angela Nguyenfor technical expertise and Lingge Li for providing assistancewith the microsomal stability assays. We also thank Dr Mary-Ellen Harper for assisting with adiposity determinations andMiguel Cabrita for critical reading of the manuscript.

Address all correspondence and requests for reprints to: DrM.A. Christine Pratt, Department of Cellular and Molecular

Medicine, University of Ottawa, 451 Smyth Road, Ottawa, On-tario, Canada K1H 8M5. E-mail: [email protected].

This work was supported by the Canadian Institutes ofHealth Research PPP Grant 122888 (to M.A.C.P. and T.D.) andby the Canadian Breast Cancer Foundation (Ontario Region)(M.A.C.P.).

Disclosure Summary: The authors have nothing to disclose.

References

1. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits ofestrogen plus progestin in healthy postmenopausal women: princi-pal results From the Women’s Health Initiative randomized con-trolled trial. JAMA. 2002;288:321–333.

2. Beral V, Million Women Study Collaborators. Breast cancer andhormone-replacement therapy in the Million Women Study. Lancet2003;362:419–427.

3. Greendale GA, Reboussin BA, Hogan P, et al. Symptom relief andside effects of postmenopausal hormones: results from the Post-menopausal Estrogen/Progestin Interventions Trial. Obstet Gyne-col. 1998;92:982–988.

4. Constantine GD, Pickar JH. Estrogens in postmenopausal women:recent insights. Curr Opin Pharmacol. 2003;3:626–634.

5. Sassarini J, Lumsden MA. Hot flushes: are there effective alterna-tives to estrogen? Menopause Int. 2010;16:81–88.

6. Freeman EW, Sammel MD, Lin H, Gracia CR, Kapoor S. Symptomsin the menopausal transition: hormone and behavioural correlates.Obsts Gynecol. 2008;111:127–136.

7. North American Menopause Society. Management of osteoporosisin postmenopausal women: 2006 position statement of the NorthAmerican Menopause Society. Menopause. 2006;13:340–367; quiz368–369.

8. Edwards DP. The role of coactivators and corepressors in the biol-ogy and mechanism of action of steroid hormone receptors. J MammGl Biol Neopl. 2000;5:307–324.

9. Delaunay F, Pettersson K, Tujague M, Gustafsson JA. Functionaldifferences between the amino-terminal domains of estrogen recep-tor � and �. Mol Pharmacol. 2000;58:584–590.

10. Liu Y, Gao H, Marstrand TT, et al. The genome landscape of ER�

and ER�-binding DNA regions. Proc Natl Acad Sci USA. 2008;105:2604–2609.

11. Klinge CM. Estrogen receptor interaction with co-activators andco-repressors. Steroids. 2000;65:227–251.

12. O’Malley BW. Coregulators: from whence came these “mastergenes”. Mol Endocrinol. 2007;21:1009–1013.

13. Opas EE, Gentile MA, Kimmel DB, Rodan GA, Schmidt A. Estro-genic control of thermoregulation in ER�KO and R�KO mice. Ma-turitas. 2006;53:210–216.

14. Park BW, Kim KS, Heo MK, et al. Expression of estrogen receptor-�in normal mammary and tumor tissues: is it protective in breastcarcinogenesis? Breast Cancer Res Treat. 2003;80:79–85.

15. Li S, Han B, Liu G, et al. Immunocytochemical localization of sexsteroid hormone receptors in normal human mammary gland. J His-tochem Cytochem. 2010;58:509–515.

16. Liu MM, Albanese C, Anderson CM, et al. Opposing action ofestrogen receptors � and � on cyclin D1 gene expression. J BiolChem. 2002;277:24353–24360.

17. Förster C, Mäkela S, Wärri A, et al. Involvement of estrogen receptor� in terminal differentiation of mammary gland epithelium. ProcNatl Acad Sci USA. 2002;99:15578–15583.

18. Ström A, Hartman J, Foster JS, Kietz S, Wimalasena J, GustafssonJA. Estrogen receptor � inhibits 17-� estradiol-stimulated prolifer-ation of the breast cancer cell line T47D. Proc Natl Acad Sci USA.2004;101:1566–1571.



mailto:[email protected]

19. Paruthiyil S, Parmar H, Kerekatte V, Cunha GR, Firestone GL, Leit-man DC. Estrogen receptor � inhibits human breast cancer cell pro-liferation and tumor formation by causing a G2 cell cycle arrest.Cancer Res. 2004;64:423–428.

20. Vasudevan N, Pfaff DW. Membrane-initiated actions of estrogens inneuroendocrinology: emerging principles. Endocr Rev. 2007;28:1–19.

21. Weiser MJ, Foradori CD, Handa RJ. Estrogen receptor � in thebrain: from form to function. Brain Res Rev. 2008;57:309–320.

22. Roepke TA. Oestrogen modulates hypothalamic control of energyhomeostasis through multiple mechanisms. J Neuroendocrinol.2009;21:141–150.

23. Wright JS, Shadnia H, Anderson JM, et al. A-CD estrogens. I. Sub-stituent effects, hormone potency, and receptor subtype selectivity ina new family of flexible estrogenic compounds. J Med Chem. 2011;54:433–448.

24. Asim M, El-Salfiti M, Qian Z, et al. Deconstructing the estradiolABCD ring structure: a new family of A-CD compounds which arepotent and subtypeselective estrogen receptor agonists. BioOrganicMed Chem Lett. 2009;19:1250–1253.

25. Cavalieri E, Chakravarti D, Guttenplan J, et al. Catechol estrogenquinones as initiators of breast and other human cancers. BiochimBiophys Acta. 2006;1766:63–78.

26. Liehr JG. Genotoxicity of the steroidal oestrogens oestrone and oes-tradiol: possible mechanism of uterine and mammary cancer devel-opment. Hum Reprod Update. 2001;10:273–281.

27. Zahid M, Kohli E, Saeed M, Rogan E, Cavalieri E. The greaterreactivity of estradiol-3,4-quinone vs. estradiol-2,3-quinone withDNA in the formation of depurinating adducts: implications fortumor-initiating activity. Chem Res Toxicol. 2006;19:164–172.

28. Bolton JL. Quinoids, quinoid radicals, and phenoxyl radicals fromestrogens and antiestrogens: role in carcinogenesis? Toxicology.2002;177:55–65.

29. Cavalieri E, Rogan E, Chakravarti D.2004 The role of endogenouscatechol quinones in the initiation of cancer and neurodegenerativediseases. Methods Enzymol. 2004;382:293–319.

30. National Toxicology Program. 10th Report on Carcinogens. UnitedStates Department of Health and Human Services; 2002. http://ehp03.niehs.nih.gov.

31. Nilsson S, Gustafsson JÅ. Estrogen receptors: therapies targeted toreceptor subtypes. Clin Pharmacol Ther. 2011;89:44–55.

32. Berendsen HH, Weekers AH, Kloosterboer HJ. Effect of tiboloneand raloxifene on the tail temperature of oestrogen-deficient rats.Eur J Pharmacol. 2001;419:47–54.

33. Leventhal L, Cosmi S, Deecher D. Effect of calcium channel mod-ulators on temperature regulation in ovariectomized rats. Pharma-col Biochem Behav. 2005;80:511–520.

34. Girbig D, Keller K, Prelle K, Patchev V, Vonk R, Igl BW. A dynamicmodel of circadian rhythms in rodent tail skin temperature for com-parison of drug effects. J Circadian Rhythms. 2012;10:1.

35. Reid G, Hübner MR, Métivier R, et al. Cyclic, proteasome-mediatedturnover of unliganded and liganded ER� on responsive promotersis an integral feature of estrogen signaling. Mol Cell. 2003;11:695–707.

36. Ruddy SC, Lau R, Cabrita MA, et al. Preferential estrogen receptor� ligands reduce Bcl-2 expression in hormone-resistant breast cancercells to increase autophagy [published online April 30, 2014]. MolCancer Ther. doi:1158/1535-7163.MCT-13.1066.

37. Heine PA, Taylor JA, Iwamoto GA, Lubahn DB, Cooke PS. In-creased adipose tissue in male and female estrogen receptor-�knockout mice. Proc Natl Acad Sci USA. 2000;97:12729–12734.

38. Geary N, Asarian L, Korach KS, Pfaff DW, Ogawa S. Deficits inE2-dependent control of feeding, weight gain, and cholecystokininsatiation in ER-� null mice. Endocrinology. 2001;142:4751–4757.

39. Thammacharoen S, Lutz TA, Geary N, Asarian L. Hindbrain ad-ministration of estradiol inhibits feeding and activates estrogen re-ceptor-�-expressing cells in the nucleus tractus solitarius of ovari-ectomized rats. Endocrinology. 2008;149:1609–1617.

40. Dacks PA, Rance NE. Effects of estradiol on the thermoneutral zoneand core temperature in ovariectomized rats. Endocrinology. 2010;151:1187–1193.

41. Mitra SW, Hoskin E, Yudkovitz J, et al. Immunolocalization ofestrogen receptor � in the mouse brain: comparison with estrogenreceptor �. Endocrinology. 2003;144:2055–2067.

42. Kenemans P, Bundred NJ, Foidart JM, et al. LIBERATE StudyGroup. Safety and efficacy of tibolone in breast-cancer patients withvasomotor symptoms: a double-blind, randomised, non-inferioritytrial. Lancet Oncol. 2009;10:135–146.

43. Guttuso JT Jr. Gabapentin’s effects on hot flashes and hypothermia.Neurology. 2000;54:2161–2163.

44. Opas EE, Scafonas A, Nantermet PV, et al. Control of rat tail skintemperature regulation by estrogen receptor-� selective ligand. Ma-turitas. 2009;64:46–51.

45. Minutolo F, Macchia M, Katzenellenbogen BS, KatzenellenbogenJA. Estrogen receptor� ligands: recent advances and biomedical ap-plications. Med Res Rev. 2011;31:364–442.

46. Likhite VS, Stossi F, Kim K, Katzenellenbogen BS, KatzenellenbogenJA. Kinase-specific phosphorylation of the estrogen receptorchanges receptor interactions with ligand, deoxyribonucleic acid,and coregulators associated with alterations in estrogen and tamox-ifen activity. Mol Endocrinol. 2006;20:3120–3132.

47. Tee MK, Rogatsky I, Tzagarakis-Foster C, et al. Estradiol and se-lective estrogen receptor modulators differentially regulate targetgenes with estrogen receptors � and �. Mol Biol Cell. 2004;15:1262–1272.

48. Meegan MJ, Lloyd DG. Advances in the science of estrogen receptormodulation. Curr Med Chem. 2003;10:181–210.

49. Komm BS, Kharode YP, Bodine PV, Harris HA, Miller CP, LyttleCR. Bazedoxifene acetate: a selective estrogen receptor modulatorwith improved selectivity. Endocrinology. 2005;146:3999–4008.

50. Kharode Y, Bodine PV, Miller CP, Lyttle CR, Komm BS. The pairingof a selective estrogen receptor modulator, bazedoxifene, with con-jugated estrogens as a new paradigm for the treatment of meno-pausal symptoms and osteoporosis prevention. Endocrinology.2008;149:6084–6091.

51. Stovall DW, Tanner-Kurtz K, Pinkerton JV. Tissue-selective estro-gen complex bazedoxifene and conjugated estrogens for the treat-ment of menopausal vasomotor symptoms. Drugs. 2011;71:1649–1657.

52. Shen L, Ahmad S, Park S, et al. In vitro metabolism, permeability,and efflux of bazedoxifene in humans. Drug Metab Dispos. 2010;38:1471–1479.

53. Roepke TA, Bosch MA, Rick EA, et al. Contribution of a membraneestrogen receptor to the estrogenic regulation of body temperatureand energy homeostasis. Endocrinology. 2010;151:4926–4937.

54. Qiu J, Bosch MA, Tobias SC, et al. A G-protein-coupled estrogenreceptor is involved in hypothalamic control of energy homeostasis.J Neurosci. 2006;26:5649–5655.

55. Bowe J, Li XF, Kinsey-Jones J, et al. The hop phytoestrogen, 8-pre-nylnaringenin, reverses the ovariectomy-induced rise in skin tem-perature in an animal model of menopausal hot flushes. J Endocri-nol. 2006;191:399–405.

56. Santollo J, Torregrossa AM, Eckel LA. Estradiol acts in the medialpreoptic area, arcuate nucleus, and dorsal raphe nucleus to reducefood intake in ovariectomized rats. Horm Behav. 2011;60:86–93.

57. Santollo J, Marshall A, Daniels D. Activation of membrane-associ-ated estrogen receptors decreases food and water intake in ovariec-tomized rats. Endocrinology. 2013;154:320–329.



Novel Ligands Balance Estrogen Receptor β and α Agonism for Safe and Effective Suppression of the...

Documents

Transcript of Novel Ligands Balance Estrogen Receptor β and α Agonism for Safe and Effective Suppression of the...