Cellular Signalling 12 (2000) 47–52

http://www.elsevier.com/locate/cellsig

Acute modulation of Ca21 influx on rat heart by 17b-estradiolClaudia Buitrago, Virginia Massheimer, Ana Russo de Boland*

Departamento de Biologia, Bioquimica & Farmacia, Universidad Nacional del Sur, San Juan 670, 8000 Bahia Blanca, Argentina

Received 21 June 1999; accepted 1 September 1999

Abstract

Estrogens initiate their action by binding to specific intracellular receptors and then acting on gene expression. In addition,there is growing evidence of a direct membrane effect via interaction with a cell surphase receptor. The aim of the present studywas to investigate the acute effects of 17b-estradiol on Ca21 fluxes through second messenger pathways in rat cardiac muscle. Ex-posure of rat ventricle to low levels of 17b-estradiol (10212–1028 M) increased 45Ca21 influx within 1 min (138%); the responsewas biphasic, peaking at 2 and 5 min (160 and 155%, respectively). The effect of the hormone on rat heart seems to be specificsince 17a-estradiol, dihydrotestosterone, and progesterone were devoid of activity. The effect of 17b-estradiol (5 min, 10210 M)was suppressed by nitrendipine (1 mM) and LaCl3 (10 mM), involving the activation of voltage-dependent Ca21 channels in theacute increase of rat heart calcium influx by the hormone. 17b-estradiol rapidly increased cAMP content and PKA activity of ratcardiac muscle in parallel to the changes in Ca21 uptake. In addition the cAMP antagonist Rp-cAMPS suppressed 17b-estradiol-dependent Ca21 influx. Altogether, the data suggest the involvement of the cAMP/PKA messenger system in the nongenomicmodulation of Ca21 influx in rat cardiac muscle by physiological levels of 17b-estradiol. 2000 Elsevier Science Inc. All rightsreserved.

Keywords: Rat heart; 17b-estradiol; Ca21 influx; cAMP; PKA; Signal transduction

1. Introduction myocytes and in ventricular myocytes of guinea-pigsand rats [8]. However, in recent years nongenomic ef-

Estrogen can influence cardiac diseases by acting di- fects of steroid hormones have been described. Theserectly on the heart or by affecting the vascular system. effects are rapid in onset and are not modified by inhib-Clinical and experimental evidence suggest that 17b- itors of transcription or protein synthesis [9]. Further-estradiol decreases the incidence of coronary heart dis- more, specific membrane binding sites for steroids haveease in humans [1,2]. Although these clinical observa- been reported in several tissues [4], including vasculartions are well known, the mechanism of estrogen action smooth muscle and endothelial cells [10–12]. Numerouson the vascular system is poorly understood. This cardio- rapid, presumably nongenomic effects for estrogens in-protective activity has been linked to estrogen-medi- clude a rapid release of cAMP [13,14], hydrolysis ofated increases in high-density lipoprotein-cholesterol polyphosphoinositides via activation of phospholipaseand decreases in low-density lipoprotein-cholesterol C [15,16], and a rapid elevation of intracellular calcium[3]. The alterations in lipid profile account for only half concentration due to an influx of extracellular Ca21

of the cardiovascular benefit observed in estrogen- [14,16,17], and/or due to a mobilisation of Ca21 from in-treated women, suggesting that other mechanisms may tracellular stores [16,18] depending on cell type.be involved [4]. The direct influences of estrogen on the heart have

Estrogen, like other steroid hormones, exerts most received less consideration. In vivo and in vitro studiesof its effects by direct binding to specific nuclear recep- have shown acute effects of estrogen on cardiac contrac-tors, which act as transcriptional activators [5–7]. The tility [19] and extravascular [20] and vascular smoothpresence of mRNA encoding the estrogen intracellular muscle contraction [21], and functional evidence for a

fast nongenomic pathway has been given by electro-receptor has been demonstrated in human atrial cardiacphysiological studies on cardiac myocytes [22–24] andmyocardial tissues [25].* Corresponding author. Tel.: 154-91-595-100, ext. 2430; fax: 154-

The purpose of this work was to study the rapid91-595-130.E-mail address: aboland@criba.edu.ar (A. R. de Boland) (nongenomic) effects of 17b-estradiol on rat myocar-

0898-6568/00/$ – see front matter 2000 Elsevier Science Inc. All rights reserved.PII: S0898-6568(99)00066-2

48 C. Buitrago et al. / Cellular Signalling 12 (2000) 47–52

dium calcium influx, activation of the cAMP signallingpathway, and its relationship to calcium influx modula-tion by the hormone.

2. Materials and methods

2.1. Materials

Rp-cAMPS was obtained from Biolog (Bremen,Germany). 45CaCl2 and 125I-cyclic AMP radioimmunoas-say kit were purchased in New England Nuclear (Chi-cago, IL, USA). 17b-estradiol and all other reagentswere from Sigma Chemical Co. (St. Louis, MO, USA).

2.2. Animals

Female Wistar rats (3–6 months old) were fed withFig. 1. Time course of calcium uptake by rat cardiac tissue. Isolatedstandard rat food (1.2% Ca; 1.0% phosphorus), givenventricle slices were exposed to vehicle ethanol (,0.01%) for 1 to 10water ad libitum, and maintained on a 12-h light–12-h min. 45Ca21 uptake was measured as described in text. Results are the

dark cycle. Animals were sacrificed by cervical disloca- means 6 SD of three independent experiments (n 5 5). *P , 0.001;tion. The heart was cleaned of large vessels and adipose **P , 0.005.

tissue. The left ventricle was excised, cut in slices (15 32 mm) and maintained in cold Krebs-Henseleit solution 5.5–6.0) was used for cAMP determination by a RIAcontaining 0.2% glucose until treatment. technique [27] using a commercially available RIA kit.

2.3. In vitro treatments 2.6. Measurement of protein kinase A activity

Isolated heart slices were pre-equilibrated in Krebs– Protein kinase A activity was measured according toHenseleit-0.2% glucose for 20 min at 378C and then ex- Fujimori et al. [28]. Following each treatment, the cellsposed for short intervals (1–10 min) to different doses were homogenised in 50 mM KH2PO4, 10 mM EDTA,of 17b-estradiol or vehicle ethanol (,0.01%). When 10 mM EGTA, 2 mM IBMX, 2% Triton X-100, 20 mMcalcium channel blockers (nitrendipine, LaCl3) or the NaF, 0.125 M NaCl, 30 mM b-mercaptoethanol, 0.2 mMcAMP antagonist (Rp-cAMPS) was used, they were DTT, 0.5 mM paramethylsulfonyl fluoride, 40 mg/mladded to the medium during the pre-equilibration pe- aprotinine, 40 mg/ml leupeptine pH 7.2, and centrifugedriod. In each experiment at least four rat hearts were at 12 000 rev./min for 15 min. PKA activity was deter-employed and for a given assay condition four to six mined by the incorporation of 32P from [g-32P]-ATP intothin slices were used. kemptide, the synthetic peptide substrate for PKA [29].

Ten microliters of the supernatant were incubated for 52.4. Determination of calcium uptake min at 308C in 20 mM Tris–HCl pH 7.4, 10 mM

Mg(CH3COO)2, 10 mM NaF, 1 mM EDTA, 0.1% BSA,After the pre-equilibration period, 17b-estradiol or75 mM kemptide, and 100 mM ATP, 0.2 mCi [g-32P]-vehicle were added simultaneously with 45CaCl2 (2 mCi/ATP, in the presence or absence of 5 mM cAMP. Theml) followed by incubation for 1 to 10 min. Immediatelyreaction was stopped by transferring 25 ml of assay mix-after treatment, heart slices were washed in ice-cold un-ture into Whatman P-81 phosphocellulose paper,labelled medium, followed by solubilisation in 1 Nwashed three times with 75 mM H3PO4, dried, andNaOH. Under these conditions extracellularly boundcounted using scintillation fluid. Hormonal activation45Ca is completely removed. Aliquots were taken forof PKA was expressed as the PKA activity ratio: activ-measurement of radioactivity and protein determina-ity measured in the absence of added cAMP to that ob-

tion [26].tained in the presence of cAMP (2cAMP/1cAMP).

2.5. Measurement of cyclic AMP levels 2.7. Statistical evaluation

Immediately after treatment tissue slices were trans- The significance of the results was evaluated by Stu-ferred to ice-cold tricholoracetic acid (TCA, 6% final con- dent’s t-test [30].centration), homogenised, and centrifuged at 1200 3 g(15 min, 48C). The pellet was dissolved with NaOH and

3. Resultsaliquots were taken for protein measurements [26], andthe supernatant was extracted four times with 5 vol of We first evaluated the kinetics of calcium uptake in

isolated ventricle slices (Fig. 1). In the absence of hor-water-saturated diethylether. The extract (final pH:

C. Buitrago et al. / Cellular Signalling 12 (2000) 47–52 49

Table 1Specificity of the effect of 17b-estradiol on calcium uptake by rat hearta

Treatment Calcium uptake(nmol/mg protein)

Vehicle 6.4 6 0.417b-estradiol, 0.1 nM 9.6 6 0.8*17b-estradiol, 10 nM 12.4 6 0.9*17a-estradiol, 0.1 nM 6.1 6 0.317a-estradiol, 10 nM 6.9 6 0.9Progesterone, 0.1 nM 6.3 6 0.2Progesterone, 10 nM 6.6 6 0.4DHT, 0.1 nM 5.6 6 1.0DHT, 10 nM 5.8 6 0.8

a Results are the means 6 SD of three independent experiments(n 5 5).

* P , 0.001.Fig. 2. Time course of the effects of 17b-estradiol on calcium uptakeby rat heart. Isolated ventricle slices were exposed to 0.1 nM 17b-estradiol or vehicle ethanol (,0.01%) for 1 to 10 min. 45Ca21 uptake protein) and the maximal response was reached at 1028

was measured as described in text. Results are the means 6 SD of M (1100%).three independent experiments (n 5 5). *P , 0.005; **P , 0.01.The effects of 17b-estradiol on calcium uptake was

specific (Table 1); after 5 min, the rapid stimulation ofcalcium uptake was not observed when ventricle slicesmonal stimulus, 45Ca21 uptake increased linearly withwere incubated in the presence of the stereoisomer oftime, reaching a plateau at 5 min (5.8 6 0.4 nmol Ca21/17b-estradiol, 17a-estradiol, a biologically inactive iso-mg protein) and remained elevated until 10 min. Asmer that is identical to the native estrogen except forshown in Fig. 2, physiological concentrations (10210 M)the configuration of a single hydrogen atom [31], or theof 17b-estradiol rapidly stimulates 45Ca21 influx in ratsex steroid hormones progesterone and dihydrotestos-myocardium. The stimulatory action of the hormone onterone.calcium uptake was already seen at 1 min. (138%), fol-

To evaluate whether the activation of calcium chan-lowed by two pronounced peaks at 2 min (160%) andnels are involved in the rapid stimulation by 17b-estra-at 5 min (155%). The response then declined to valuesdiol of rat heart calcium influx, the effects of the hor-near basal after 10 min of hormonal exposure. The stim-mone on 45Ca21 uptake were measured in the presenceulation of calcium influx induced by 17b-estradiol afterof 1 m M nitrendipine or 10 m M La Cl3. As shown in5 min was between 10212 and 1028 M (Fig. 3). HormoneFig. 4, both calcium channel blockers abolished the in-concentrations as low as 10212 M increased calcium up-crement in calcium uptake elicited by the hormone (5take 45% above control values (6.4 6 0.5 nmol Ca21/mgmin, 10210 M).

Fig. 3. Dose-response profile of the rapid effects of 17b-estradiol on Fig. 4. Effects of calcium channel blockers on the rapid increase inheart calcium uptake induced by 17b-estradiol. Rat ventricles wererat heart calcium uptake. Ventricle slices isolated from rat heart were

incubated for 5 min with different concentrations of 17b-estradiol or treated for 5 min with 0.1 nM 17b-estradiol in the absence or presenceof the calcium channel blockers nitrendipine (1 mM) or LaCl3 (10vehicle ethanol (,0.01%). 45Ca21 uptake was measured as described

in text. Results are the means 6 SD of three independent experi- mM). 45Ca21 uptake was measured as described in text. Results are themeans 6 SD of three independent experiments (n 5 5). *P , 0.005.ments (n 5 5). *P , 0.001.

50 C. Buitrago et al. / Cellular Signalling 12 (2000) 47–52

Fig. 6. Effect of the cAMP antagonist Rp-cAMPS on the rapid in-crease in heart calcium uptake induced by 17b-estradiol. Rat ventricleslices were preincubated with 200 mM Rp-cAMPS and then exposedFig. 5. Time course of the effects of 17b-estradiol on intracellular0.1 nM 17b-estradiol or vehicle ethanol (,0.01%) for 1 and 5 min.cAMP levels of heart tissue. Rat ventricle slices were treated with 0.145Ca21 uptake was measured as described in text. Results are thenM 17b-estradiol or vehicle ethanol (,0.01%) at the times indicated.means 6 SD of three independent experiments (n 5 5). *P , 0.005.cAMP levels were measured as described in text. Results are the

means 6 SD of three independent experiments (n 5 4). *P , 0.005.

sable cAMP analogue that selectively blocks PKA acti-vation by binding to the regulatory subunit [33], the in-It is well known that voltage-dependent calciumcrement in calcium uptake induced by 17b-estradiolchannels (L-type) are sensitive to dihydropiridines,was completely suppressed.such as nitrendipine, and are modulated by intracellular

second messenger signalling systems [32]. The potentialinvolvement of the cAMP-dependent protein kinase A 4. Discussionpathway in the 17b-estradiol rapid stimulation of heart

The current results provide evidence of a rapid non-calcium influx was investigated. To that end, hormone-genomic action of estrogen in rat cardiac muscle ininduced changes in heart cAMP content and the effectsvitro. We demonstrated that exposure of isolated ratof inhibitors of this messenger system on rat ventricleventricle to very low concentrations of 17b-estradiol in-calcium uptake were studied. Incubation of ventricleduced a rapid stimulation of calcium influx within 1 toslices with 10210 M 17b-estradiol rapidly increased cAMP10 min. An interesting feature of the time course oflevels (Fig. 5). Significant elevations of intracellularmodifications in 45Ca21 uptake by heart ventricle elic-cAMP content were observed in ventricles exposed toited by 17b-estradiol was the decreased response at 3the hormone for 1 and 5 min (100 and 65% above con-min. This profile may reflect increased Ca21 efflux fromtrol values, respectively). The basal cAMP level of ratcardiac muscle following an initial elevation of Ca21 in-ventricle was 0.25 6 0.02 pmol cAMP/mg protein. Fur-flux and/or release from intracellular stores by the hor-thermore, in vitro exposure to 10210 M 17b-estradiol in-mone. However, additional experimental work is re-creases by two-fold PKA activity within 1 to 5 min (Ta-quired to test this hypothesis further.ble 2). As shown in Fig. 6, when calcium uptake was

The changes induced by the hormone in heart 45Ca21measured in the presence of Rp-cAMPS, a nonhydroly-uptake were suppressed by nitrendipine and LaCl3, im-plying the involvement of voltage-dependent Ca21

Table 2 channels in the fast modulation of calcium influx byAcute effect of 17b-estradiol on protein kinase A (PKA) activity 17b-estradiol. In agreement with our results, studiesin rat hearta

with other cell types have shown a rapid effect of physi-Time PKA activity (pmol pi/mg protein/min) ological concentrations of 17b-estradiol on both cal-(min) cium-channel activation and intracellular calcium mobi-Control 17b-estradiol

lisation [14,16–18]. Contrary to these observations, in1 1.84 6 0.05 3.49 6 0.83*vascular smooth muscle cells pharmacological concen-3 1.34 6 0.71 2.03 6 0.76**

5 1.06 6 0.07 2.42 6 0.44* trations of estrogens primarily reduces Ca21 influxthrough inhibition of L-type Ca21-channels [34]. Elec-a Results are the means 6 SD of three independent experimentstrophysiological studies demonstrated a calcium antag-(n 5 4).

* P , 0.001, ** P , 0.005. onistic action of 17b-estradiol in ventricular myocytes

C. Buitrago et al. / Cellular Signalling 12 (2000) 47–52 51

[22,24,35]. These effects, however, occur in vitro with Finally, it remains to be determined whether otherintracellular effectors (e.g., protein kinase C) are alsohormone concentrations within the supraphysiological

range. involved in transducing the rapid effect of estrogen inrat heart. Nevertheless, the present investigation sug-The rapid stimulation of Ca21 influx in mammalian

heart induced by 17b-estradiol appeared to be specific gests a role of 17b-estradiol in the rapid stimulation ofCa21 influx in cardiac tissue. This role may be importantsince other sex steroid hormones, such as progesterone

and dihydrotestosterone and the inactive analogue 17a- in understanding certain aspects of cardiovascular dys-function associated with estrogen deficiency.estradiol, were devoid of activity. However, we cannot

exclude the participation of other estrogen metabolites.Although most of the oxidative metabolism of estro-

Acknowledgmentsgens takes place in the liver [36], some estrogen meta-bolising isoenzymes that are usually expressed at low or This research was supported by grants from the Con-undectectable levels in liver are selectively expressed in sejo Nacional de Investigaciones Cientificas y Tecnicascertain extrahepatic tissues [37]. However, the func- (CONICET), Universidad Nacional del Sur, and thetional role or importance of estrogen metabolism in the Consejo Nacional de Investigaciones Cientificas de laheart remains unexplored. It is possible that certain es- Provincia de Buenos Aires (CIC).trogen metabolites may function as chemical mediatorsor as secondary hormones.

Various lines of evidence have indicated that the ac- Referencestivity of cardiac calcium channels is modulated by phos-

[1] Colditz GA, Willett WC, Stamfer MJ, Rosner B, Speizer FE,phorylation [38,39]. We showed evidence of the partici-Hennekens CH. N Engl J Med 1987;316:1105–10.pation of the cAMP pathway in 17b-estradiol-mediated

[2] Henderson BE, Paganini-Hill A, Ross RK. Am J Obstet Gyne-calcium influx. A temporal correlation between changes col 1988;159:312–7.in Ca21 influx, cAMP levels, and PKA activity in re- [3] Gruchow HW, Anderson AJ, Barboriak JJ, Sobocinski KA. Amsponse to estrogen was observed. In agreement with Heart J 1988;115:954–63.

[4] Bush TL. Ann NY Acad Sci 1990;592:263–71.these observations, fast increases in cAMP levels either[5] Beato M. Cell 1989;56:335–44.after in vitro or in vivo treatment with 17b-estradiol[6] Weiz A, Cicatiello L, Persicot E, Scalona M, Bresciani F. Molwere observed in rat intestinal, bone, and uterine cells, Endocrinol 1990;4:1041–50.

and intact uterus and human breast cancer cells [14,40– [7] O’Malley BW, Tsai MJ. Biol Repro 1992;46:163–67.42]. Moreover, we found that the effects of estrogen [8] Meyer R, Linz KW, Surges R, Meinardus S, Vees J, Hoffmann

A, Windholz O, Grohe C. Exptl Physiol 1998;83:305–21.were suppressed by the cAMP antagonist Rp-cAMPs,[9] McEwen BS. Trends Pharmacol Sci 1991;12:141–7.suggesting that PKA-dependent phosphorylation is

[10] Wehling M. Annu Rev Physiol 1997;59:365–93.part of the mechanism involved in 17b-estradiol-depen- [11] Giscard V, Miller VM, Vanhoutte PM. J Pharmacol Exp Therdent calcium influx in rat heart. 1987;224:19–22.

It has been proposed that the cAMP generated by [12] Farhat MI, Vargas R, Dingaan B, Ramwell PW. Br J Pharmacol-ogy 1992;107:679–83.the nongenomic pathway synergises with steroid hor-

[13] Aronica SM, Kraus WL, Katzenellenbogen S. Proc Natl Acadmone receptor-mediated activation of gene transcrip-Sci USA 1994;91:8517–21.tion by phosphorylation of the receptor or associated [14] Picotto G, Massheimer V, Boland R. Molec Cell Endocrinol

transcription factors [43,44]. 1996;119:129–34.Furthermore, the presence of estrogen receptor [15] Graber R, Sumida C, Vallette G, Nunez EA. Cell Signal 1993;

5:181–6.mRNA has been recently demonstrated in rat ventricle[16] Liberherr M, Grosse B, Kackacke M, Balsan S. J Bone Mineralmyocytes [8]. There is evidence supporting the concept

Res 1993;8:1365–76.that steroid hormones control cellular function by[17] Audy MC, Vacher P, Dufy B. Eur J Endocrinol 1996;135:367–73.

means other than the nuclear receptor steroid binding [18] Morley P, Whitfield JF, Vanderhyden BC, Tsaang BK, Schwartzmechanism. The rapidity and specificity of 17b-estra- JL. Endocrinology 1992;131:1305–12.

[19] Raddino R, Manca C, Poli E, Bolognesi R, Visioli O. Arch Intdiol effects on heart 45Ca21 uptake could possibly be ex-Pharmacodyn Ther 1986;281:57–65.plained by the presence of an estrogen receptor on the

[20] Yasay GD, Kau ST, Li JH. Pharmacology 1995;51:273–80.cell membrane. High affinity binding sites have been[21] Thomas G, Ito K, Zikic E, Bhatti T, Han C, Ramwell PW. J Pha-

identified in membranes from other estrogen target rmac Exp Ther 1995;273:1544–50.cells [45–47]. Moreover, specific estrogen binding pro- [22] Jiang C, Poole-Wilson PA, Sarrel PM, Mochizuki S, Collins P,

McLeod KT. Br J Pharmacol 1992;106:739–45.teins different from classical nuclear estrogen receptor[23] Eckstein N, Nadler E, Barnea O, Shavit G, Ayalon D. Am J Ob-were found in the cellular membrane of rat brain [48].

stet Gynecol 1994;171:844–8.Therefore, it is then possible that 17b-estradiol partici-[24] Berger F, Borchard U, Hafner D, Putz I, Weis TM. Naunyn-

pates in hormonal regulation of calcium metabolism Schmiedeberg’s Arch Pharmacol 1997;356:788–96.acting in heart muscle through genomic and nongeno- [25] Sitzler G, Lenz O, Kilter H, La Rosee K, Bohm M. Br J Pharma-

col 1996;119:43–8.mic pathways.

52 C. Buitrago et al. / Cellular Signalling 12 (2000) 47–52

[26] Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. J Biol Chem [38] Bkaily G, Sperelakis N. Am J Physiol 1984;246:H630–4.[39] Hosey MM, Lazdunski M. J Membr Biol 1988;104:81–105.1951;195:265–77.[39] Aronica SM, Kraus WL, Katzenellenbogen BS. Proc Natl Acad[27] Holmegaard SN. Acta Endocrinol 1982;101:1–46.

Sci USA 1994;91:8517–21.[28] Fujimori A, Cheng SL, Avioli L, Civitelli R. Endocrinology[40] Szego CM, Davis JS. Proc Natl Acad Sci USA 1967;58:1711–8.1992;130:29–35.[41] Fiorelli G, Gori F, Frediani U, Franceschelli F, Tanini A, Tosti-[29] Kemp BE, Clark MG. J Biol Chem 1978;253:5147–54.

Guerra C, Benvenuti S, Gennari L, Becherini L, Baldi ML. J Ste-[30] Snedecor GW, Cocharn WG. Statistical Methods. Ames, IA:roid Biochem Mol Biol 1996;59:233–40.The Iowa State University Press, 1967. pp. 120–34.

[42] Fuyimoto N, Katzenellenbogen BS. Mol Endocrinol 1994;8:[31] Morley P, Whitfield JF, Vanderhyden BC, Tsang BK, Schartz296–304.JL. Endocrinology 1992;121:1305–12.

[43] LeGoff P, Montano MM, Schodin DJ, Katzenellenbogen BS. J[32] Catteral WA. Science 1988;242:50–61.

Biol Chem 1994;269:4458–66.[33] Erneux C, Van Sande J, Jastorff B, Dumont JE. Biochem J 1986; [44] Parikh I, Anderson WL, Neame P. J Biol Chem 1980;255:

234:193–7. 10266–70.[34] Kitazawa T, Hamada E, Kitazawa K, Gaznabi AKM. J Physiol [45] Berthois Y, Pourreau-Schneider N, Gandilhon P, Miltre H, Tub-

1997;499:497–511. iana N, Martin P. J Steroid Biochem 1986;6:963–72.[35] Meyer R, Linz KW, Surges R, Meinardus S, Vees J, Hoffmann [46] Ramirez VD, Zheng J, Siddique KM. Cell Mol Neurobiol 1996;

A, Windholz O, Grohe C. Experimental Physiol 1998;83:305–21. 16:175–98.[36] Martucci CP, Fishman J. Pharmacol Ther 1993;57:237–57. [47] Zheng J, Ramirez VD. J Steroid Biochem Mol Biol 1997;

62:327–36.[37] Fishman J, Norton B. Endocrinology 1975;96:1054–9.