- Email: [email protected]
Hepatic estrogen receptors and plasma estrogen-binding activity in the Atlantic Salmon
GENERAL
AND
Hepatic
COMPARATIVE
Estrogen
Receptors and Plasma Estrogen-Binding the Atlantic Salmon
C. B. LAzIER,*‘I Departments
of “Biochemistry
57, 234-245 (1985)
ENDOCRINOLOGY
K. LONERGAN,*
and Biology,
Dalhousie
Activity
in
AND T. P. MOMMSEN
University,
Halifax,
Nova
Scotia
B3H
4H7,
Canada
Accepted April 23, 1984 Livers of male and female immature Atlantic Salmon (Salmo salur) contain specific highafhnity [3H]estradiol binding sites in cytosol (I& 2-4 n&4, concentration about 0.6 pmol/g liver). Low levels of high-affinity binding are detectable in salt extracts of nuclei of untreated fish, but injections of estradiol result in transient depletion of the cytosol binder and in accumulation of high levels of binding sites in nuclear salt extracts (Kd 5-6 nM; concentration about 6 pmol/g liver). Both the cytosol and nuclear binding sites are temperature sensitive and are optimally assayed by incubation at 2”. Both are specific for estradiol and diethylstilbestrol (DES) and no significant competition by dihydrotestosterone (DHT), progesterone, or hydrocortisone is seen. The triphenylethylene nonsteroidal antiestrogen, 4hydroxytamoxifen, exhibits an affinity comparable to that of estradiol. The nuclear binding activity sediments with a coefficient of 3.6 S in salt-containing sucrose density gradients, and is stable on storage at -20” for several months. The cytosol binder on the other hand is not stable on sucrose density gradients or on prolonged storage. Salmon plasma contains two [3H]estradiol binding components, one with a relatively high affinity for t3H]estradiol (kd 13 nM) and the other having a much lower affinity but present in high concentrations. The high-affinity plasma binder exhibits distinctive specificity with no affinity for DES or 4-hydroxytamoxifen but some affinity for DHT and progesterone. These properties serve to distinguish the plasma activity from the intrahepatic estrogen binders. The salmon liver estrogen receptor system has many features in common with typical estradiol receptors from other vertebrates. Immature salmon liver appears to be the richest source of hepatic estrogen receptor so far found for any vitellogenic SpeCieS. 0 1985 Academic Press, Inc.
The teleost liver, just as the liver of other classes of egg-laying vertebrates, responds to exogenously administered or endogenously synthesized estrogens. The prevailing and highly sensitive response to these steroid hormones is the rapid hepatic production and export of vitellogenin, the precursor of the egg yolk proteins. Recent attention has been focused mainly on the expression of the vitellogenin gene in the domestic chicken and in the amphibian, Xenopus laevis (Tata and Smith, 1979; Deeley and Goldberger, 1979; Shapiro, 1982; Wahli et al., 1982). In general, much less attention has been devoted to the fishes, although a few papers dealing with 1 To whom correspondence
should be addressed.
vitellogenesis per se have appeared (Plack et al., 1971; Korsgaard et al., 1976; Hara .and Hirai, 1978; van Bohemen et al., 1981). In teleosts, levels of circulating estrogens reveal the expected annual variations, closely reflecting the different stages in the reproductive cycle (Lambert et al., 1978). Actual concentrations of estradiol-17p, the most abundant and effective estrogen with regard to the induction of vitellogenesis, range from 22 pg/ml plasma to peak values of 4 to 60 rig/ml plasma during the vitellogenie phase of different salmonids (Crim and Idler, 1978; Fostier et aZ., 1978; van Bohemen et al., 1981). The first response of the liver of the oviparous vertebrates to estrogen appears to be the accumulation of specific nuclear re234
0016-6480185 $1.50 Copyright All rights
0 1985 by Academic Press, Inc. of reproduction in any form reserved.
ESTROGEN
BINDERS
ceptors for the hormone (Schneider and Gschwendt, 1977; Lazier and Haggarty, 1979; Shapiro, 1982). These receptors have been characterized in some detail in chickens (Schneider and Gschwendt, 1977; Lazier and Haggarty, 1979; Lazier, 1979) and in X. laevi~ (Shapiro, 1982; Wright et al,, 1983) but the only studies we are aware of in fish are one report of low levels of estradiol receptor activity in the liver nuclei of the gonadectomized goby (G&us nigeu) (LeMenn et al., 1980) and another of an increase in estradiol binding activity in the liver nuclei of vitellogenic as compared to nonvitellogenic Pacific hagfish Eptatretus stouti (Turner et al., 1981). We have been studying the regulation of liver metabolism in the spawning behavior of salmonids (Mommsen et aZ., 1980; French et al., 1983) and thus have particular interest in understanding the mechanisms involved in vitellogenesis in these species. Since vitellogenesis in Atlantic salmon is already initiated while the fish are still migrating at sea, months before they enter fresh water to spawn (Idler et aZ., 1979), we chose saltwater-adapted immature Salmo s@lar to isolate and characterize hepatic nuclear, cytosolic, and plasma estrogen binding activity in untreated and estrogen-treated animals. Our results show that salmon liver contains high concentrations of estrogen receptor, much higher than that previously reported for liver from any teleost and indeed higher than found in liver of any vitellogeaic species. MATERIALS
AND METHODS
Materials [2,4,6,7-3HJEstradio1 (104 Ciimmol) was obtained from New England Nuclear, Montreal, Quebec, Canada, and radiochemcial purity was monitored by thin-layer chromatography as previously described (Lazier and Haggarty, 1979). Steroid hormones, phenyltnethylsulfonyl fluoride, and all buffer chemicals were obtained from Sigma Chemical Company (St. Louis, MO.). Tamoxifen and 4-hydroxytamoxifen were gifts from ICI Ltd. (Macclesfield, Cheshire, U.K.).
IN SALMON
235
Animals Immature Atlantic salmon (S. salar) obtained from IMA Aquatic Farms Ltd. (Yarmouth Co., N.S. Canada), were maintained at 6-9” in a circnlating seawater aquarium. They were used for experiments at a body weight of 400 to 600 g. Fish were injected ip with estradiol-17p three times in 5 days at a dose of 5 mgi kg body wt each time. Estradiol was dissolved in ethanol (4 mgiml) and diluted with an eqnal volume of glucose-free Cortland’s fish saline (Wolf, 1963) immediately before use. Control fish received freshly prepared vehicle (ethanol/saline) alone. Both male and female immature fish were used for the experiments. The sex of each animal was noted at the time of dissection. No differences in [3H]estradiol binding in different preparations attributable to the sex of the animals were found.
Preparation
of Salmon Liver Fractions
Forty-eight hours after the last injection, a 2-ml blood sample was taken from the caudal vein and fish were sacrificed by a sharp blow to the head. Preparation of liver cytosol and nuklear fractions was essentially as described earlier for cockerel liver (Lazier and Haggarty, 1979). Briefly, minced liver was rinsed with saline and homogenized in 2.5 pro1 cytosol buffer A (0.02 ti Tris, 3 mM MgCI,, 0.33 M sucrose, IO tn.V thioglycerol, pH 7.9, containing 50 pgiml phenylmethylsulfonyl fluoride). A crude nuclear pellet was removed by centrifugation at SOOgfor 20 min and cytcsol was prepared by centrifu.gation of the supernatant fraction at 105,OOOg for 60 min. The cytosol protein concentration was 20-25 mg/ml. Ammonium sulfate fractionation of the cytosol was performed only where specifically indicated in the text. The crude nuclear pellet was washed three times as described previously (Lazier, 1978) and suspended in buffer (buffer B) containing 0.5 M KC1 and 10 & 2[tris(hydroxymethyl)methylamino] - 1 - ethanesulfonic acid (TES), 1.5 mM EDTA, 10 mM thioglycerol, pH 7.5, using 1 ml of buffer for the nuclei from 1 g tissue. The suspension was frozen fo’ 1 hr at -7o”, thawed and centrifuged at 37,000g for 30 min. The supernatant fraction constitutes the nucleai salt extract. Its protein concentration was 6-9 mgiml. The pellet was retained for assay of DNA and of salt-insoluble nuclear [3H]estradiol binding activity. Protein in the liver fractions was determined by the Bio-Rad reagent technique, (Bio-Rad Laboratories, Mississauga, Ontario, Canada),, and DNA concentrations were determined by the method of Burton (1968).
Measlcrement of [3HJEstradiol Binding Activity [3H]Estradiol
binding activity in liver cytosol and
236
LAZIER,
LONERGAN,
nuclear salt extract fractions and in plasma was measured by a charcoal adsorption assay (Lazier and Haggarty, 1979). For fractions prepared from animals which had been estrogen treated, preliminary incubation for 30 min at 2” with charcoal-dextran suspension (final concentration 1% charcoal, 0.1% dextran) was performed in order to remove endogenous free steroids. Assays were carried out by incubating the liver fractions with a range of [3H]estradiol concentrations as stated in the text in a total volume of 300 ~1. Parallel tubes containing the [3H]estradiol plus a loo-fold excess of radioinert diethylstilbestrol were also incubated for determination of nonspecific binding. All samples were incubated in duplicate. Unless otherwise noted, incubation was for 18 hr at 2”, after which time bound [3H]estradiol was separated from free by treatment at 2” for 30 min with charcoal-dextran suspension (an equal volume of 0.5% charcoal and 0.05% dextran in the appropriate salt-containing or salt-free buffer) and centrifugation for 2 min at 13,OOOg. Portions of the supernatant fractions were counted for radioactivity in Aquasol II (New England Nuclear Corp.). The salt-insoluble nuclear [3H]estradiol-binding activity was measured by the method of Snow et al. (1978). The pellets remaining after salt extraction were resuspended in TE buffer (10 mM Tris-HCl, 1.5 rnM thioglycerol, pH 7.5), incubated for 18 hr at 2” with 10 nM [3H]estradiol in the presence or absence of 1 pM DES, sedimented, and washed 3 x with TE buffer and extracted with absolute ethanol at 30”. The bound [3H]estradiol in the ethanol extract was quantitated by liquid scintillation counting.
Sucrose Density Gradient Centrifugation Liver fractions were incubated for 18 hr at 2” with 10 nM [3H]estradiol, charcoal treated by addition of one-half volume of 1% charcoal, 0.1% dextran in buffer B and applied to 5-20% sucrose gradients (prepared in buffer B plus 10% (v/v) glycerol). Centrifugation was for 3 hr at 300,OOOg in a Beckman VTi 80 rotor. [i4C]Ovalbumin and [t4C]y-globulin (New England Nuclear Corp.) were used as internal markers. Forty fractions were collected and counted for radioactivity in Aquasol II.
Statistical Analysis Results are expressed as means + SE.
RESULTS E&radio1 Binding Activity
in Liver
Cytosol
Initial experiments with liver cytosol from untreated immature salmon showed that maximal specific binding of 10 nM
AND MOMMSEN
[3H]estradiol was achieved upon incubation at 2” for 18 hr (Table l), and that incubation at temperatures of 2.5” or above for 30 min resulted in loss of binding activity. Figure 1 shows a typical [3H]estradiol binding curve and Scatchard analysis for a salmon liver cytosol preparation incubated with varying concentrations of [3H]estradiol in the presence or absence of 100x excess of DES for 18 hr at 2”. The results suggest that the binding represents a single class of DES-competible high-affinity sites. For five different preparations of cytosol, each from two livers, the concentration of binding sites was found to be 0.64 & .03 pmol/g liver, or 60 + 5 fmol/mg protein. The mean equilibrium dissociation constant (Kd) for the [3H]estradiol binding activity was 2.9 t 1.1 n&f. The binding specificity of the cytosol [3H]estradiol binding activity is shown in Fig. 2. It can be seen that the nonsteroidal estrogen DES has a relatively high affinity for the binder, as does the nonsteroidal antiestrogen 4-hydroxytamoxifen. As for avian and mammalian estrogen receptors (Jordan et al., 1977; Lazier and Jordan, 1982), tamoxifen has considerably less affinity than the hydroxylated derivative. Nonestrogenic steroids such as progesTABLE
1
THE EFFECT OF TEMPERATURE ON SPECIFIC BINDING OF [3H]E~~~~~~~ BY SALMON LIVER Crroso~
Incubation temperature (“Cl 2 10 25 30
Specific binding (dpm x 10W3/tube) Expt 1
Expt 2
38.4 38.3 27.3 14.7
40.0 35.2 21.2 17.3
Note. Cytosol samples (150 ~1) were incubated in duplicate with 10 nM [3H]estradiol (final volume 300 ~1) at 2” for 0.5 hr followed by 0.5 hr at the indicated temperature (Expt 1), or at 2” for 18 hr followed by 0.5 hr at the higher temperature (Expt 2). After cooling to 2”, bound [3H]estradiol was separated from free by the charcoal-dextran technioue.
ESTROGEN
BINDERS
237
IN SALMON
in a vertical tube rotor) are due to excessive dissociation of the labeled ligand. The dissociation rate constant at 2” for the binder is 0.002 min- t with a half-life of 5.8 hr fresults not shown). Unlike the situation with mammalian estrogen receptor (Nui et al., 1981), inclusion of molybdate in the, salmon cytosol preparations does not affect the [3H]estradiol binding behavior on sucrose gradients. Furthermore, molybdate had no effect on the stability of cytosol estrogen binding sites to exposure to elevated temperatures or upon storage at - 20°C Estradiol Binding Activity in Salt Extracts of Liver Nuclei qf Estradiol-Treated Salmon FIG. 1. Binding of [3H]estradiol by salmon liver cytosol. (A) Cytosol (150 ~1) was incubated with [3H]estradiol (0.25-80 n@l in the absence or presence of 100x excess of DES for 18 hr at 2” (final volume 300 ~1). Bound hormone was separated from free by charcoal-dextran treatment. (0) total bound [3H]estradiol; (0) nonspecific bound; (A) specific bound. (B) Scatchard analysis of specific binding from (A). Binding site concentration, 8.5 fmol/mg protein; Kd 3.8 hf.
terone, DHT, and hydrocortisone show no appreciable affinity for the salmon liver estrogen binder. The cytosol estrogen binding activity can be enriched 4-6X by precipitation with (NH&SO, at 33% saturation, with 80100% recovery of activity in this fraction. This behavior is typical of estrogen receptors from mammalian and avian sources (Puca et al., 1975; Lazier, 1979) and differentiates the binder from the nonre.ceptor hepatic steroid-binding protein found in chicken liver, which requires 50-70% saturation with (NH&SO, for precipitation (Dower and Ryan, 1976; Lazier, 1979). We have been unable to achieve satisfactory resolution of the salmon liver cytosol estradiol binder on sucrose density gradient centrifugation under a variety of conditions. It is unlikely that the poor results on sucrose gradients (in centrifugation for 3 hr
Salt .extracts of liver nuclei from immature salmon which had not been exposed to exogenous estradiol contain variable, but relatively low, amounts of high-affinity [3H]estradiol binding activ,ity. In nine preparations, a mean concentration of 17 t 4 fmolimg protein or 148 + 35 fmolig liver was obtained. Injection of multiple high
COMPETITOR
CONCENTRATION
t “M
I
FIG. 2. Specificity of binding of [3H]estradiol to liver cytosol from untreated salmon. Increasing concentrations of the potential competitor were incubated at 2” for 18 hr with 150~1 of cytosol and 10 ~JU [3H]estmdi~l in a total volume of 300 ~1. Bound [3H]estradiol was separated from free by the charcoal-dextran method. Results are expressed as a percentage of specific binding, calculated as the net binding supressible by a 100 x excess of unlabeled estradiol-17.X (0) estradiol; (0) DES; (A) progesterone; (A) DHT; (V) hydrocortisone; (W) 4-hydroxytamoxifen; (0) tamoxifen.
238
LAZIER.
LONERGAN,
doses of estradiol-17B results in a 40 fold increase in the concentration of these sites. In order to determine optimal conditions for an exchange assay for the nuclear sites in the estrogen-treated fish, a variety of incubation conditions was tested. As a first step, nuclear salt extracts were charcoal treated at 2” in order to remove endogenous unbound steroids prior to incubation with [3H]estradiol. An extract which had been previously charcoal treated gave a specific binding activity of 400 fmol/mg protein upon incubation at 2” for 18 hr with 10 nM [3H]estradiol, whereas an extract incubated without prior charcoal treatment gave binding of 274 fmol/mg protein. Consequently, all nuclear salt extracts were routinely charcoal treated at 2” before assay. The effect of incubation temperature on the exchange assay is shown in Table 2. Incubation at 2” for 16.5 hr results in higher specific DES-competible [3H]estradiol binding than does 16 hr at 2” followed by 0.5 hr at 10”. Considerable loss of binding sites is apparent after incubation at 2” for 16 hr followed by 0.5 hr at 25” or 30”. Therefore denaturation of the binding activity compromises any benefit elevation of the incubation temperature might have in promoting exchange of bound ligand. Figure 3A illustrates the association kinetics of [3H]estradiol binding to nuclear
AND MOMMSEN
salt extracts on incubation at 2”. The second order reaction has an association rate constant (K+) of 0.0011 X lo9 mol-’ min- l. Higher incubation temperatures again were not beneficial in promoting the exchange reaction. The dissociation behavior at 2” of [3H]estradiol bound to salmon liver nuclear salt extracts is shown in Fig. 3B. Under these conditions two components are seen. One rapidly dissociates with a K- of 0.025 min-’ and a half-life of 28 min, and the other shows a K- of 0.0059 min-’ and a half-life of 117 min. Thus it is apparent that an 18 hr incubation at 2” would probably be adequate for substantial exchange of bound nuclear sites. Charcoal-treated nuclear salt extracts were thus incubated at 2” for 18 hr with varying concentrations of [3H]estradiol in the presence or absence of a 100 x excess of DES. Binding curves and a Scatchard plot are’presented in Fig. 4. It is obvious that the nuclear salt extracts contain mainly a single class of high-affinity binding sites. In five different liver preparations, the mean binding site concentration was 5.3 + 0.5 pmol/g liver or 699 + 56 fmol/mg protein with mean Kd values of 5.9 + 0.3 ti. This Kd is similar to that derived from the kinetic studies, where Kd = K-IKf =
5.4 nM.
The hormone specificity of the nuclear estradiol binding activity is similar to that THE INFLUENCE OF INCUBATION TEMPERATURE ON of the cytosol, the most notable features SPECIFIC [3H]E~~~~~~~ BINDING BY being the strong competition by DES and NUCLEAR SALT EXTRACT 4-hydroxytamoxifen (Fig. 5). Incubation temperature Specific binding The nuclear [3H]estradiol binding activity (dpm x 10-3) (“Cl differs from the cytosol binder in that it is 2 80.0 more stable on storage at -20”. The half 75.4 10 life is greater than 3 months for the nuclear 20 75.1 extracts but less than 2 weeks for the cy25 63.4 tosol. The two binders also differ in sta44.5 30 bility on sucrose gradient centrifugaNote. Samples of salt extract (100 (*l) were incution. Figure 6 shows that the nuclear bated with 10 mM [3H]estradiol in a total volume of [3H]estradiol binder sediments with a coef300 ~1 for 18 hr at 2” followed by 0.5 hr at the temficient of 3.6 S in 5-20% sucrose gradients perature shown. Specific binding was determined as containing 0.4 M KCl. This is somewhat described in Table 1. TABLE
2
ESTROGEN A P 5 10
BINDERS
.
239
IN SALMON
flect the fact that thioglycerol was present in the salt extraction buffer, which tends to release nuclear matrix-associated receptor (Barrack and Coffey, 1980). The Time Course of the Increase ilz Nuclear E&radio1 Binding Sites following a Single Injection of E&radio1
FIG. 3. Association and dissociation kinetics of [3H]estradiol binding to nuclear salt extract. (A) Specific binding of 10 n&4 [3H]estradiol incubated with 100 yl salt extract (total volume 300 ~1) at 2” for increasing periods of time. (B) Dissociation of [3H]estradiol previously bound by incubation of 100 ~1 salt extract with 10 r&f [3H]estradiol in the presence or absence of 1 @4 DES (total volume 300 ~1) at 2” for 18 hr. Dissociation was initiated by the addition of 1 IJ.M unlabeled estradiol-17S and continued incubation at 2” followed by charcoal-dextran treatment at the indicated times. Remaining specific binding is presented as the log percentage of binding at time zero (to). A control for stability showed no loss of receptor activity over the dissociation period.
less than is found for avian and mammalian estrogen receptors, but is similar to that reported for crude and partly purified preparations of the estrogen receptor in salt extracts of liver nuclei from X. Zaevis (Lazier, 1978; Gschwendt and Schneider, 1978; Jensen et al., 1982; Wright et al., 1983). We also measured [3H]estradiol binding ta the insoluble residue remaining after salt extraction of the nuclei (Snow et al., 1978). Relatively low levels of specific binding were found (0.25 pmolJg liver for estrogentreated salmon, and about 0.03 pmol/g for untreated animals). These values may re-
Fish were injected with a single dose (5 mg/kg) of estradiol and cytosol and nuclear salt extracts were prepared and assayed for [3H]estradiol binding activity at various times thereafter. Both the cytosol and nuclear fractions were charcoal treated prior to assay in order to remove endogenous estradiol. The results inFig, 7 show that cytosol receptor levels are decreased 4 hr after injection, gradually restored over the next 16 hr and are 1.5-2 times the Bevels in uninjected controls by 48-120 hr. Accu-
A
E c -I-
4
FIG. 4. Saturation analysis of [3H]estradiol binding by a salt extract of salmon liver nuclei. (A) Salt extract (100 ~1) was incubated with [3H]estradiol (OS-SO nm for 18 hr at 2” in the presence or absence of a 100X excess of DES in a final volume of 300 ~1, and binding was determined.by the charcoal-dextran method. (0) total binding of 13H]estradiol; (0) nonspecific binding; (A) specific binding. (B) Scatchard analysis of the data given in (A), specific binding. Binding site concentrations, 872 fmolVmg protein; I& 5.3 nM.
240
LAZIER,
10 COMPETITOR
102
12
CONCENTRATION
LONERCAN,
ld I nM I
5. Binding specificity of [3H]estradiol binding by nucelar salt extracts from estrogen-treated salmon. Nuclear preparations (100 ~1) were incubated with 10 nM [3H]estradiol in the absence or presence of potential competitors. Incubation conditions, assay of specific binding, and the symbols used were exactly as described in Fig. 2. FIG.
AND
MOMMSEN
An important feature of both the highand low-affinity [3H]estradiol binding sites in plasma is that neither is competed for by DES. Figure 9 shows that DES does not compete for the high-affinity sites (measured using 20 nM [3H]estradiol). We also found no competition at all for [3H]estradiol binding by a 100 x excess of DES at concentrations of the labeled ligand of up to 500 nit4 (results not shown). Figure 9 further shows that the triphenylethylene antiestrogens do not compete for [3H]estradiol binding to the high-affinity plasma component. In addition, DHT has considerable affinity for the plasma binder, and both progesterone and estrone compete to a lesser but measurable extent. This distinctive hormone specificity clearly differentiates the plasma binder from the DES and antiestrogen-competible binding activ-
mulation of nuclear binding sites is clear by 12 hr and continues over the 4-day period of the experiment. This pattern of estradiol binding site distribution following estrogen treatment is similar to that seen in chickens (Lazier and Haggarty, 1979). The very high concentration of nuclear receptor presumably arises through estradiol-induced synthesis or stabilization of its own receptor.
I
I
[3HjEstradiol Binding by Salmon Plasma
Although liver preparations were well rinsed with saline prior to homogenization, it was possible that some of the intrahepatic estrogen binding arose from contamination with plasma components. Figure 8 shows that salmon plasma does indeed contain [3H]estradiol binding activity, a portion of which appears to have a relatively high affinity (Kd - 13 n&4) and to be saturable. A large number of sites not saturated under the conditions used are also present. In separate experiments these latter sites were found to be saturable at [3H]estradiol concentrations of 200-400 nM.
0
G
1
1
4 \ t
FRACTION
NUMBER
FIG. 6. Sucrose gradient centrifugation of salmon liver nuclear salt extract. A sample of salt extract (100 ~1) from an estradiol-treated fish was incubated in a total volume of 300 ~1 with 10 nM 13H]estradiol in the presence or absence of 1 p.M DES for 18 hr at 2”, charcoal-treated, and applied to a 5-20% sucrose gradient. Centrifugation was for 3 hr at 300,OOOg in a vertical tube rotor. (0)-[14C]Ovalbumin marker, 3.6 S; (G)-[14C]y-globulin marker, 6.6 S.
ESTROGEN
BINDERS
IN SALMON
ities present in liver cytosol and nuclei. The plasma binder instead resembIes the “sex steroid binding protein” described by others in a variety of fishes as well as in higher vertebrates (Martin, 1980). The lack of affinity of DES and of the triphenylethylenes is characteristic of the human plasma sex steroid binding globulin (King and Mainwaring, 1974). Close examination of Fig. 2 shows that a small fraction of cytosol binding sites occupied by 10 nM [3H]estradiol are not competed for by high conentrations of DES or 4-hydroxytamoxifen. This could represent a minor degree of contamination of the cytosol preparation with plasma. DISCUSSION The results presented here show that liver of the Atlantic salmon contains f3H]estradiol binding activity with many
FIG. 8. [3H]estradiol binding by plasma. Varying concentrations of 13H]estradiol in the absence or presence of a 100 x excess of unlabeled estradioL17B were incubated with 285 ~1 of a IO-fold dilution of salmon plasma in cytosol buffer A in a total volume of 300 ~1. Incubatian was for 18 hr at 2” and bound 13H]estradiol was separated from free by charcoal-dextran treatment. (A) Binding curves (@), total; (O), nonspecific; and (A) specific binding of [3Hlestradiol. (R) Scatchard analysis of specific binding from (A). Concentration of the high-affinity component: 16.8 nM in plasma, Kd 13 nM by the Rosenthal method (Rosen-
thal,
24 TIME
48 AFTER
72 INJECTION
96
120 I h I
FIG. 7. The effect of injection of a single dose of estradiol on intracellular distribution of high-affinity estradiol binding sites. ‘Salmon were injected with 5 mgikg estradiol ip and cytosol and nuclear salt extracts were prepared after various time intervals up to 108 hr. Livers from two fish were pooled for each point. The concentration of specific estradiol binding sites in the charcoal-treated extracts was measured by incubation with 10 nM [3]estradiol in the presence and absence of DES at 2” for 18 hr and by correction for the degree of saturation considering the mean K, values previously observed for the cytosol and nuclear binding sites. (e) Nuclear salt extracts; (0) cytosol.
1967).
similarities to the estrogen receptor system target tissues of mammals, birds, amphibia, and reptiles (Gschwendt and Schneider, 1978; Lazier, 1979; Martin, 198Q; Shapiro, 1982). The cytosol receptor in untreated immature fish has the appropriate binding affinity and specificity. It is clearly distinguishable from estradiol-binding activity in plasma on the basis of its relativeiy higher affinity for [3H]estradiol, its high affinity for DES and 4-hydroxytamoxifen, and its lack of affinity for DHT. The concentration of the cytosol estrogen receptor is n&ably high compared to that reported for estrogen receptor in liver from cockerels, X. iuevis, or turtles. In cockerels for example, estrogen receptor is barely detectable in LD-
242
LAZIER,
c_
LONERGAN,
04fa 10 COMPETITOR
IO2 CONCENTRATION
12
IO4 ( “M
I
FIG. 9. Specificity of [3H]estradiol binding by salmon plasma. Plasma, diluted as in Fig. 8, was incubated with 20 nM [3H]estradiol for 18 hr at 2” along with varying concentrations of a potential competitor. Bound [3H]estradiol was separated from free by the charcoal-dextran method. Data are expressed as a percentage of the total [3H]estradiol bound in the absence of competitor. The symbols used are the same as in Fig. 2.
fractionated preparations of liver cytosol, ana only on concentration by ammonium sulfate precipitation with concomitant removal of the nonreceptor hepatic steroid binding protein, can levels of about 20-40 fmol/mg protein or 200-400 fmol/g liver be deteited (Gschwendt, 1977; Lazier and Haggarty, 1979; De Boer et al., 1982). Higher concentrations are transiently observed during chicken embryogenesis at the 19th day of development (Gschwendt, 1977; Lazier, 1980). In X. laevis, workers have generally reported receptor concentrations in terms of sites/cell based on a figure of 1.3 x lo8 cells/g liver (Westley and Knowland, 1978; Hayward et al., 1980). Thus, liver cytosol of untreated male frogs contains about 100 sites of estrogen receptor or 22 fmol/g. In livers of animals that have been estrogentreated and withdrawn for 70 days a cytosol receptor concentration of 1200 sites/cell (260 fmol/g) has been found (Hayward et al., 1980). High-affinity binding of estradiol can be detected in liver cytosol of the adult female turtle, but the concentration is only
AND MOMMSEN
3 fmol/mg protein (Heisermann et al., 1980). Injection of large doses of estradiol, sufficient to stimulate substantial levels of vitellogenin synthesis (Plack et al., 1971; Korsgaard et al., 1976; LeMenn et al., 1980) results in accumulation of a high concentration of estrogen receptor in salmon liver nuclei. The concentrations found here are the highest yet reported for the liver of any oviparous vertebrate on estrogenic stimulation. In untreated cockerels, salt extracts of liver nuclei contain about 200 fmol of estradiol binding sites/g liver (Lazier and Haggarty, 1979; Lazier, 1979). This is similar to that found in the immature salmon and about 1.5 times that in male X. laevis liver nuclei (Westley and Knowland, 1978). Injection of a single high dose of estradiol results in the accumulation of about 2.5 pmol of binding sites/g liver in cockerels (Lazier and Haggerty, 1979; Lazier, 1979) and 0.4 pmollg in X. laevis (Hayward et al., 1980). In the experiments reported here, we find that multiple injections of estradiol result in accumulation of more than 5 pmol of high-affinity sites per gram of tissue 48 hr after the last injection. A single dose of estradiol(5 mg/kg) results in some elevation at 12 hr, followed by a linear rise over the succeeding 4 days with a level of about lo12 pmol/g liver being reached at that time. Cytosol receptors fall to almost undetectable levels 4 hr after estrogen administration, but are fully replenished by 48 hr and almost twice control levels after 4 days. We have not yet studied any period after 4 days. It will be of interest to determine if estrogen treatment results in any long-term changes in receptor distribution and concentration, as appears to be the case in X. laevis (Hayward et al., 1980). We would like to express our data in terms of the number of binding sites/hepatocyte, but are uncertain of the DNA content per liver cell in S. salav or of the number of cells per gram liver. DNA values ranging from 2.5 to 6.5 pg per haploid cell
ESTROGEN
BINDERS
and of 5.6 pg per erythrocyte have been reported for various Salmo species (Fasman, 1976). We determined that the mean DNA content of the liver from untreated S. salar was 2.06 _+ .16 mgig and for the estrogen-treated fish was 1.91 + 0.24 mgig. This gives a mean nuclear receptor concentration of 3.16 t 0.65 pmoli mg DNA for the animals given multiple injections of estradiol. Conservatively assuming a mean hepatocyte DNA content of 6 pg, a figure of about 11,000 sites/cell can be calculated. Even higher values were achieved in the time-course experiment where assays were conducted 4 days after a single injection of estradiol. The high concentration of estrogen receptor ,sites in liver nuclei of the estrogentreated salmon indicate that this might be an excellent source for purification of liver receptor from an oviparous vertebrate. The relative stability of the salt-extracted nuclear receptor on storage at -20” further substantiates this point. Under the influence of estrogen, hepatic metabolism of egg-laying animals undergoes a far-reaching and quite rapid reorganization, predominantly in protein and lipid metabolism (Tata and Smith, 1979; Shapiro, 1982). In teleost fishes the metabolic demand placed on the liver following the binding of estrogen is further compounded through cessation of feeding during spawning migration. We are currently analyzing the metabolic and hormonal control of liver metabolic pathways duri.ng such spawning migration in salmonids (Mommseq et al., 1980). The emphasis has been placed on hepatic gluconeogenesis, since the spawning process per so is known to be fueled exclusively by carbohydrate (French et a!., 1983), spawning occurring at a time when lipid reserves ‘have been depleted. At any point during the migration, vitellogenesis, is a major, direct competitive pathway for gluconeo~enic precursors as well as for ‘I avarlable fatty acids. Therefore’ these studies on estrogen receptors were initiated
IN SALMON
243
as a prerequisite toward understanding the regulation of estrogen responsiveness and induction of vitellogenesis in the framework of hepatic carbohydrate metabolism. ACKNOWLEDGMENTS These studies were supported by a grant to C.B.L. from the Medical Research Couhcil of Canada and by a grant to T.P.M. from the Natural Sciences and Engineering Research Council of Canada.
REFERENCES Barrack, E. R., and Coffey, D. S. (1980). The spec& binding of estrogens and androgens to the nuclear matrix of sex hormone responsive tissues. .?. Bial. Chern. 255, 1265-7275. Burton, K. (1968). Determination of DNA concentration with diphenylamine.,In “Methods in Enzymology” (L. Grossman and F. Moldave, eds.), Vol. 12, pp. 163-166. Academic Press, New York. Crim, L. W., and Idler, D. R. (1978). Plasma gonadotropin, estradiol and vitellogenin and gonad phosvitin levels in relation to seasonal reproductive cycles of female brown trout. Ann. Bid. Anim. Biochim. Biopkys. 18, 1001-1005. De Boer, W., Snippe, L., AB, G., and Gruber, M. (1982). Estrogen receptor in avian embryo and adult liver: Estrogen receptor activation and dissociation kinetics of estradiol, ethynyl estradiol and moxestrol. Endocrinology 110, 1217-1224. Deeley, 92. G., and Goldberger, R. E (1979). Regulation of expression of the vitellogenin gene in avian liver. In “Ontogeny of Receptors and Reproductive Hormone Action”’ (T. H. Hamilton, J. H. Clark, and W. A. Sadler, eds.), pp. 291-308. Raven Press, New York. Dower, W. J., and Ryan, K. .I. (1976). A cytoplasmic estrone-specific binding protein in hen liver. Fed. Proc. 35, 1366 (abstract). Fasman, G. D. (1976). “CRC Handbook of Biochemistry and Molecular Biology.” pp. 293. Chemical Rubber Co., Cleveland/ Ohio. Fostier, A., Weil, C., Terqui, M., Breton. B., and Jalabert, B. (1978). Plasma estradiol-17B and gonadotropin during ovulatian in rainbow trout (Salmo gairdneri R.). Ann. Bibl. Biockim. Biopkys. 18, 929-935. French, C. J,, Hochachka, P. W., and Mommsen, T. I? (1983). Metabolic organization of liver during spawning migration of sockeye salmon. Amer. J. Pkysiol. 245, R827-R830. Gschwendt, M. (1977). A cytoplasmic high-affinity estrogen-binding protein in the embryonic chick liver. Ecru. J. Biockem. 80, 461-468. Gschwendt, M., and Schneider, W. (1978). Estrogen-
LAZIER,
244
LONERGAN,
binding sites of chicken liver. Preliminary characterization of nuclear components. Ear. J. Biochem. 91, 139-149. Hara, A., and Hirai, H. (1978). Comparative studies on immunochemical properties of female-specific serum protein and egg proteins in rainbow trout (Salmo
gairdneri).
Comp.
Biochem.
Physiol.
b
59, 339-344. Hayward, M. A., Mitchell, T. A., and Shapiro, D. J. (1980). Induction of estrogen receptor and reversal of the nuclearicytoplasmic receptor ratio during vitellogenin synthesis and withdrawal in Xenopus laevis. J. Biol. Chem. 255, 1130811312. Heiserman, G. J., Ho, S. M., and Callard, I. P. (1980). Estrogen binding activity in the liver of the eastern painted turtle Chrysemys picta. Amer. Zool. 20, 360 (abstract). Idler, D. R., Hwang, S. J., and Crim, L. W. (1979). Quantification of vitellogenin in Atlantic salmon (Salmo salar) plasma by radioimmunoassay. J. Fish Res. Board
Canad.
36, 574-578.
Jensen, E. V., Green, G. L., Closs, L. E., DeSombre, E. R., and Nadji, M. (1982). Receptors reconsidered: A 20-year perspective. Rec. Prog. Horm. Res. 38, l-33. Jordan, V. C., Collins, M., Rowsby, L., and Prestwich, G. (1977). A monohydroxyl metabolite of tamoxifen with potent antioestrogenic activity. J. Endocrinol. 75, 305-316. King, R. J. B., and Mainwaring, W. I. P. (1974). “Steroid-cell Interactions.” p. 193. Univ. Park Press, Baltimore. Korsgaard, B., Emmersen, J., and Petersen, I. M. (1976). Natural occurrence, and experimental induction by estradiol-17B, of a lipophosphoprotein (vitellogenin) in flounder (Platichtys Jesus). Comp.
Biochem.
Physiol.
B 64, 443-446.
Lambert, J. G. D., Bosman, G. I. C. G. M., van der Hurk, R., and van Oordt, P. G. W. J. (1978). Annual cycle of plasma estradiol-17B in the female trout Salmo gairdneri. Ann. Biol. A&m. Biochim. Biophys. 18, 923-927. Lazier, C. B. (1978). Ontogeny. of the vitellogenic response to oestradiol and of the soluble nuclear oestrogen receptor in embryonic chick liver. Biochem. J. 174, 143-152. Lazier, C. B. (1979). Estrogen-binding proteins in avian liver: Characteristics, and ontogenesis. Zn “Ontogeny of Receptors and Reproductive Hormone Action” (T. H. Hamilton, J. H. Clark, and W. A. Sadler, eds.), pp. 353-370. Raven Press, New York. Lazier, C. B. (1980). The development of estrogen receptors and of the vitellogenic response to estradiol in embryonic chick liver. Adv. Biosci. 25, 125-139.
AND MOMMSEN Lazier, C. B., and Haggarty, A. J. (1979). A high-affinity oestrogen-binding protein in cockerel liver cytosol. Biochem. J. 180, 347-353. Lazier, C. B., and Jordan, V. C. (1982). High affinity binding of antioestrogen to the chick liver nuclear oestrogen receptor. Bbchem. J. 206, 387-394. Le Menn, F., Rochefort, H., and Garcia, M. (1980). Effect of androgen mediated by estrogen receptor of fish liver: Vitellogenin accumulation. Steroids 35, 315-328. Martin, B. (1980). Steroid-protein interactions in nonmammalian vertebrates. Zn “Steroids and Their Mechanism of Action in Nonmammalian Vertebrates” (G. Delvio and J. Brachet, eds.), pp. 6373. Raven Press, New York. Mommsen, T. P., French, C. J., and Hochachka, P. W. (1980). Protein and amino acid utilisation during the spawning migration of salmon. Canad. J. Zool. 58, 1785-1799. Niu, E.-M., Neal, R. M., Pierce, V. K., and Sherman, M. R. (1981). Structural similarities of molybdatestabilized steroid receptors in human breast tumors, uteri and leukocytes. J. Steroid Biochem. 15, l-10. Plack, P. A., Pritchard, D. J., and Fraser, N. W. (1971). Egg proteins in cod serum. Biochem. J. 121, 847-856.
Puca, G. A., Nola, E., Sica, V.., and Brescani, F. (1975). Purification of estrogen receptors. In “Methods in Enzymology” (B. W. O’Malley and J. G. Hardman, eds.), Vol. 36, pp. 331-349. Academic Press, New York. Rosenthal, H. E. (1967). A graphic method for the determination and presentation of binding parameters in a complex system. Anal. Biochem. 20, 525-532. Salhanick, A. R., Vito, C. C., Fox, T. O., and Callard, I. P (1979). Estrogen-binding proteins in the oviduct of the turtle Chrysemys picta: Evidence for a receptor species. Erzdocrinology 105, 13881395. Schneider, W., and Gschwendt, M. (1977). Kinetics of the appearance of nuclear estrogen binding sites in the chick liver. Hoppe Seyler’s Z. Physiol. Chem. 358, 1583-1589. Shapiro, D. (1982). Steroid hormone regulation of vitellogenin gene expression. CRC Crit. Rev. Biochem. 8, 187-203. Snow, L. D., Erikkson, H., Hardin, J. W., Chan, L., Jackson, R. L., Clark, J. H., and Means, A. R. (1978). Nuclear estrogen receptor in the avian liver: Correlation with biologic response. J. Steroid Biochem. 9, 1017-1023. Tata, J. R., and Smith, D. F. (1979). Vitellogenesis: A versatile model for hormonal regulation of gene expression. Recent Prog. Horm. Res. 35, 47-95. Turner, R. T., Dickhoff, W. W., and Gorbman, A.
ESTROGEN
BINDERS
(1981). Estrogen binding to hepatic nuclei of pacific hagfish Eptatrerus stouti. Gen. Comp. Endocrinol.
45, 26-29.
van Bohemen, C. G., Lambert, J. G. D., and Petite, 9. (1981). Annual changes in plasma and liver in relation to vitellogenesis in the female rainbow trout, Salmo gairdneri. Gen. Camp. Endocrinol. 44, 94- 107. Wahli, W., David, I. B., Ryfell, G. U., and Weber, R. (1982). Vitellogenesis and the vitellogenin gene family. Science (Wbshington, D.C.) 212, 298304.
IN SALMON
24.5
Westley, B., and Knowland, J. (1978). An estrogen receptor from Xenopus luevis liver possibly connected with vitellogenin synthesis. Ceil lS, 367374. Wolf, K. (1963). Physiological salines for fresh water teleosts. Prog. Fish Cult. 25, 135-138. Wright, C. V. E., Wright, S. C., and Knowland, .I. (1983). Partial purification of estradiol receptor from Xenopus taevis liver and levels of receptor in relation to estradiol concentration. EWO S. 2, 973-977.
AND
Hepatic
COMPARATIVE
Estrogen
Receptors and Plasma Estrogen-Binding the Atlantic Salmon
C. B. LAzIER,*‘I Departments
of “Biochemistry
57, 234-245 (1985)
ENDOCRINOLOGY
K. LONERGAN,*
and Biology,
Dalhousie
Activity
in
AND T. P. MOMMSEN
University,
Halifax,
Nova
Scotia
B3H
4H7,
Canada
Accepted April 23, 1984 Livers of male and female immature Atlantic Salmon (Salmo salur) contain specific highafhnity [3H]estradiol binding sites in cytosol (I& 2-4 n&4, concentration about 0.6 pmol/g liver). Low levels of high-affinity binding are detectable in salt extracts of nuclei of untreated fish, but injections of estradiol result in transient depletion of the cytosol binder and in accumulation of high levels of binding sites in nuclear salt extracts (Kd 5-6 nM; concentration about 6 pmol/g liver). Both the cytosol and nuclear binding sites are temperature sensitive and are optimally assayed by incubation at 2”. Both are specific for estradiol and diethylstilbestrol (DES) and no significant competition by dihydrotestosterone (DHT), progesterone, or hydrocortisone is seen. The triphenylethylene nonsteroidal antiestrogen, 4hydroxytamoxifen, exhibits an affinity comparable to that of estradiol. The nuclear binding activity sediments with a coefficient of 3.6 S in salt-containing sucrose density gradients, and is stable on storage at -20” for several months. The cytosol binder on the other hand is not stable on sucrose density gradients or on prolonged storage. Salmon plasma contains two [3H]estradiol binding components, one with a relatively high affinity for t3H]estradiol (kd 13 nM) and the other having a much lower affinity but present in high concentrations. The high-affinity plasma binder exhibits distinctive specificity with no affinity for DES or 4-hydroxytamoxifen but some affinity for DHT and progesterone. These properties serve to distinguish the plasma activity from the intrahepatic estrogen binders. The salmon liver estrogen receptor system has many features in common with typical estradiol receptors from other vertebrates. Immature salmon liver appears to be the richest source of hepatic estrogen receptor so far found for any vitellogenic SpeCieS. 0 1985 Academic Press, Inc.
The teleost liver, just as the liver of other classes of egg-laying vertebrates, responds to exogenously administered or endogenously synthesized estrogens. The prevailing and highly sensitive response to these steroid hormones is the rapid hepatic production and export of vitellogenin, the precursor of the egg yolk proteins. Recent attention has been focused mainly on the expression of the vitellogenin gene in the domestic chicken and in the amphibian, Xenopus laevis (Tata and Smith, 1979; Deeley and Goldberger, 1979; Shapiro, 1982; Wahli et al., 1982). In general, much less attention has been devoted to the fishes, although a few papers dealing with 1 To whom correspondence
should be addressed.
vitellogenesis per se have appeared (Plack et al., 1971; Korsgaard et al., 1976; Hara .and Hirai, 1978; van Bohemen et al., 1981). In teleosts, levels of circulating estrogens reveal the expected annual variations, closely reflecting the different stages in the reproductive cycle (Lambert et al., 1978). Actual concentrations of estradiol-17p, the most abundant and effective estrogen with regard to the induction of vitellogenesis, range from 22 pg/ml plasma to peak values of 4 to 60 rig/ml plasma during the vitellogenie phase of different salmonids (Crim and Idler, 1978; Fostier et aZ., 1978; van Bohemen et al., 1981). The first response of the liver of the oviparous vertebrates to estrogen appears to be the accumulation of specific nuclear re234
0016-6480185 $1.50 Copyright All rights
0 1985 by Academic Press, Inc. of reproduction in any form reserved.
ESTROGEN
BINDERS
ceptors for the hormone (Schneider and Gschwendt, 1977; Lazier and Haggarty, 1979; Shapiro, 1982). These receptors have been characterized in some detail in chickens (Schneider and Gschwendt, 1977; Lazier and Haggarty, 1979; Lazier, 1979) and in X. laevi~ (Shapiro, 1982; Wright et al,, 1983) but the only studies we are aware of in fish are one report of low levels of estradiol receptor activity in the liver nuclei of the gonadectomized goby (G&us nigeu) (LeMenn et al., 1980) and another of an increase in estradiol binding activity in the liver nuclei of vitellogenic as compared to nonvitellogenic Pacific hagfish Eptatretus stouti (Turner et al., 1981). We have been studying the regulation of liver metabolism in the spawning behavior of salmonids (Mommsen et aZ., 1980; French et al., 1983) and thus have particular interest in understanding the mechanisms involved in vitellogenesis in these species. Since vitellogenesis in Atlantic salmon is already initiated while the fish are still migrating at sea, months before they enter fresh water to spawn (Idler et aZ., 1979), we chose saltwater-adapted immature Salmo s@lar to isolate and characterize hepatic nuclear, cytosolic, and plasma estrogen binding activity in untreated and estrogen-treated animals. Our results show that salmon liver contains high concentrations of estrogen receptor, much higher than that previously reported for liver from any teleost and indeed higher than found in liver of any vitellogeaic species. MATERIALS
AND METHODS
Materials [2,4,6,7-3HJEstradio1 (104 Ciimmol) was obtained from New England Nuclear, Montreal, Quebec, Canada, and radiochemcial purity was monitored by thin-layer chromatography as previously described (Lazier and Haggarty, 1979). Steroid hormones, phenyltnethylsulfonyl fluoride, and all buffer chemicals were obtained from Sigma Chemical Company (St. Louis, MO.). Tamoxifen and 4-hydroxytamoxifen were gifts from ICI Ltd. (Macclesfield, Cheshire, U.K.).
IN SALMON
235
Animals Immature Atlantic salmon (S. salar) obtained from IMA Aquatic Farms Ltd. (Yarmouth Co., N.S. Canada), were maintained at 6-9” in a circnlating seawater aquarium. They were used for experiments at a body weight of 400 to 600 g. Fish were injected ip with estradiol-17p three times in 5 days at a dose of 5 mgi kg body wt each time. Estradiol was dissolved in ethanol (4 mgiml) and diluted with an eqnal volume of glucose-free Cortland’s fish saline (Wolf, 1963) immediately before use. Control fish received freshly prepared vehicle (ethanol/saline) alone. Both male and female immature fish were used for the experiments. The sex of each animal was noted at the time of dissection. No differences in [3H]estradiol binding in different preparations attributable to the sex of the animals were found.
Preparation
of Salmon Liver Fractions
Forty-eight hours after the last injection, a 2-ml blood sample was taken from the caudal vein and fish were sacrificed by a sharp blow to the head. Preparation of liver cytosol and nuklear fractions was essentially as described earlier for cockerel liver (Lazier and Haggarty, 1979). Briefly, minced liver was rinsed with saline and homogenized in 2.5 pro1 cytosol buffer A (0.02 ti Tris, 3 mM MgCI,, 0.33 M sucrose, IO tn.V thioglycerol, pH 7.9, containing 50 pgiml phenylmethylsulfonyl fluoride). A crude nuclear pellet was removed by centrifugation at SOOgfor 20 min and cytcsol was prepared by centrifu.gation of the supernatant fraction at 105,OOOg for 60 min. The cytosol protein concentration was 20-25 mg/ml. Ammonium sulfate fractionation of the cytosol was performed only where specifically indicated in the text. The crude nuclear pellet was washed three times as described previously (Lazier, 1978) and suspended in buffer (buffer B) containing 0.5 M KC1 and 10 & 2[tris(hydroxymethyl)methylamino] - 1 - ethanesulfonic acid (TES), 1.5 mM EDTA, 10 mM thioglycerol, pH 7.5, using 1 ml of buffer for the nuclei from 1 g tissue. The suspension was frozen fo’ 1 hr at -7o”, thawed and centrifuged at 37,000g for 30 min. The supernatant fraction constitutes the nucleai salt extract. Its protein concentration was 6-9 mgiml. The pellet was retained for assay of DNA and of salt-insoluble nuclear [3H]estradiol binding activity. Protein in the liver fractions was determined by the Bio-Rad reagent technique, (Bio-Rad Laboratories, Mississauga, Ontario, Canada),, and DNA concentrations were determined by the method of Burton (1968).
Measlcrement of [3HJEstradiol Binding Activity [3H]Estradiol
binding activity in liver cytosol and
236
LAZIER,
LONERGAN,
nuclear salt extract fractions and in plasma was measured by a charcoal adsorption assay (Lazier and Haggarty, 1979). For fractions prepared from animals which had been estrogen treated, preliminary incubation for 30 min at 2” with charcoal-dextran suspension (final concentration 1% charcoal, 0.1% dextran) was performed in order to remove endogenous free steroids. Assays were carried out by incubating the liver fractions with a range of [3H]estradiol concentrations as stated in the text in a total volume of 300 ~1. Parallel tubes containing the [3H]estradiol plus a loo-fold excess of radioinert diethylstilbestrol were also incubated for determination of nonspecific binding. All samples were incubated in duplicate. Unless otherwise noted, incubation was for 18 hr at 2”, after which time bound [3H]estradiol was separated from free by treatment at 2” for 30 min with charcoal-dextran suspension (an equal volume of 0.5% charcoal and 0.05% dextran in the appropriate salt-containing or salt-free buffer) and centrifugation for 2 min at 13,OOOg. Portions of the supernatant fractions were counted for radioactivity in Aquasol II (New England Nuclear Corp.). The salt-insoluble nuclear [3H]estradiol-binding activity was measured by the method of Snow et al. (1978). The pellets remaining after salt extraction were resuspended in TE buffer (10 mM Tris-HCl, 1.5 rnM thioglycerol, pH 7.5), incubated for 18 hr at 2” with 10 nM [3H]estradiol in the presence or absence of 1 pM DES, sedimented, and washed 3 x with TE buffer and extracted with absolute ethanol at 30”. The bound [3H]estradiol in the ethanol extract was quantitated by liquid scintillation counting.
Sucrose Density Gradient Centrifugation Liver fractions were incubated for 18 hr at 2” with 10 nM [3H]estradiol, charcoal treated by addition of one-half volume of 1% charcoal, 0.1% dextran in buffer B and applied to 5-20% sucrose gradients (prepared in buffer B plus 10% (v/v) glycerol). Centrifugation was for 3 hr at 300,OOOg in a Beckman VTi 80 rotor. [i4C]Ovalbumin and [t4C]y-globulin (New England Nuclear Corp.) were used as internal markers. Forty fractions were collected and counted for radioactivity in Aquasol II.
Statistical Analysis Results are expressed as means + SE.
RESULTS E&radio1 Binding Activity
in Liver
Cytosol
Initial experiments with liver cytosol from untreated immature salmon showed that maximal specific binding of 10 nM
AND MOMMSEN
[3H]estradiol was achieved upon incubation at 2” for 18 hr (Table l), and that incubation at temperatures of 2.5” or above for 30 min resulted in loss of binding activity. Figure 1 shows a typical [3H]estradiol binding curve and Scatchard analysis for a salmon liver cytosol preparation incubated with varying concentrations of [3H]estradiol in the presence or absence of 100x excess of DES for 18 hr at 2”. The results suggest that the binding represents a single class of DES-competible high-affinity sites. For five different preparations of cytosol, each from two livers, the concentration of binding sites was found to be 0.64 & .03 pmol/g liver, or 60 + 5 fmol/mg protein. The mean equilibrium dissociation constant (Kd) for the [3H]estradiol binding activity was 2.9 t 1.1 n&f. The binding specificity of the cytosol [3H]estradiol binding activity is shown in Fig. 2. It can be seen that the nonsteroidal estrogen DES has a relatively high affinity for the binder, as does the nonsteroidal antiestrogen 4-hydroxytamoxifen. As for avian and mammalian estrogen receptors (Jordan et al., 1977; Lazier and Jordan, 1982), tamoxifen has considerably less affinity than the hydroxylated derivative. Nonestrogenic steroids such as progesTABLE
1
THE EFFECT OF TEMPERATURE ON SPECIFIC BINDING OF [3H]E~~~~~~~ BY SALMON LIVER Crroso~
Incubation temperature (“Cl 2 10 25 30
Specific binding (dpm x 10W3/tube) Expt 1
Expt 2
38.4 38.3 27.3 14.7
40.0 35.2 21.2 17.3
Note. Cytosol samples (150 ~1) were incubated in duplicate with 10 nM [3H]estradiol (final volume 300 ~1) at 2” for 0.5 hr followed by 0.5 hr at the indicated temperature (Expt 1), or at 2” for 18 hr followed by 0.5 hr at the higher temperature (Expt 2). After cooling to 2”, bound [3H]estradiol was separated from free by the charcoal-dextran technioue.
ESTROGEN
BINDERS
237
IN SALMON
in a vertical tube rotor) are due to excessive dissociation of the labeled ligand. The dissociation rate constant at 2” for the binder is 0.002 min- t with a half-life of 5.8 hr fresults not shown). Unlike the situation with mammalian estrogen receptor (Nui et al., 1981), inclusion of molybdate in the, salmon cytosol preparations does not affect the [3H]estradiol binding behavior on sucrose gradients. Furthermore, molybdate had no effect on the stability of cytosol estrogen binding sites to exposure to elevated temperatures or upon storage at - 20°C Estradiol Binding Activity in Salt Extracts of Liver Nuclei qf Estradiol-Treated Salmon FIG. 1. Binding of [3H]estradiol by salmon liver cytosol. (A) Cytosol (150 ~1) was incubated with [3H]estradiol (0.25-80 n@l in the absence or presence of 100x excess of DES for 18 hr at 2” (final volume 300 ~1). Bound hormone was separated from free by charcoal-dextran treatment. (0) total bound [3H]estradiol; (0) nonspecific bound; (A) specific bound. (B) Scatchard analysis of specific binding from (A). Binding site concentration, 8.5 fmol/mg protein; Kd 3.8 hf.
terone, DHT, and hydrocortisone show no appreciable affinity for the salmon liver estrogen binder. The cytosol estrogen binding activity can be enriched 4-6X by precipitation with (NH&SO, at 33% saturation, with 80100% recovery of activity in this fraction. This behavior is typical of estrogen receptors from mammalian and avian sources (Puca et al., 1975; Lazier, 1979) and differentiates the binder from the nonre.ceptor hepatic steroid-binding protein found in chicken liver, which requires 50-70% saturation with (NH&SO, for precipitation (Dower and Ryan, 1976; Lazier, 1979). We have been unable to achieve satisfactory resolution of the salmon liver cytosol estradiol binder on sucrose density gradient centrifugation under a variety of conditions. It is unlikely that the poor results on sucrose gradients (in centrifugation for 3 hr
Salt .extracts of liver nuclei from immature salmon which had not been exposed to exogenous estradiol contain variable, but relatively low, amounts of high-affinity [3H]estradiol binding activ,ity. In nine preparations, a mean concentration of 17 t 4 fmolimg protein or 148 + 35 fmolig liver was obtained. Injection of multiple high
COMPETITOR
CONCENTRATION
t “M
I
FIG. 2. Specificity of binding of [3H]estradiol to liver cytosol from untreated salmon. Increasing concentrations of the potential competitor were incubated at 2” for 18 hr with 150~1 of cytosol and 10 ~JU [3H]estmdi~l in a total volume of 300 ~1. Bound [3H]estradiol was separated from free by the charcoal-dextran method. Results are expressed as a percentage of specific binding, calculated as the net binding supressible by a 100 x excess of unlabeled estradiol-17.X (0) estradiol; (0) DES; (A) progesterone; (A) DHT; (V) hydrocortisone; (W) 4-hydroxytamoxifen; (0) tamoxifen.
238
LAZIER.
LONERGAN,
doses of estradiol-17B results in a 40 fold increase in the concentration of these sites. In order to determine optimal conditions for an exchange assay for the nuclear sites in the estrogen-treated fish, a variety of incubation conditions was tested. As a first step, nuclear salt extracts were charcoal treated at 2” in order to remove endogenous unbound steroids prior to incubation with [3H]estradiol. An extract which had been previously charcoal treated gave a specific binding activity of 400 fmol/mg protein upon incubation at 2” for 18 hr with 10 nM [3H]estradiol, whereas an extract incubated without prior charcoal treatment gave binding of 274 fmol/mg protein. Consequently, all nuclear salt extracts were routinely charcoal treated at 2” before assay. The effect of incubation temperature on the exchange assay is shown in Table 2. Incubation at 2” for 16.5 hr results in higher specific DES-competible [3H]estradiol binding than does 16 hr at 2” followed by 0.5 hr at 10”. Considerable loss of binding sites is apparent after incubation at 2” for 16 hr followed by 0.5 hr at 25” or 30”. Therefore denaturation of the binding activity compromises any benefit elevation of the incubation temperature might have in promoting exchange of bound ligand. Figure 3A illustrates the association kinetics of [3H]estradiol binding to nuclear
AND MOMMSEN
salt extracts on incubation at 2”. The second order reaction has an association rate constant (K+) of 0.0011 X lo9 mol-’ min- l. Higher incubation temperatures again were not beneficial in promoting the exchange reaction. The dissociation behavior at 2” of [3H]estradiol bound to salmon liver nuclear salt extracts is shown in Fig. 3B. Under these conditions two components are seen. One rapidly dissociates with a K- of 0.025 min-’ and a half-life of 28 min, and the other shows a K- of 0.0059 min-’ and a half-life of 117 min. Thus it is apparent that an 18 hr incubation at 2” would probably be adequate for substantial exchange of bound nuclear sites. Charcoal-treated nuclear salt extracts were thus incubated at 2” for 18 hr with varying concentrations of [3H]estradiol in the presence or absence of a 100 x excess of DES. Binding curves and a Scatchard plot are’presented in Fig. 4. It is obvious that the nuclear salt extracts contain mainly a single class of high-affinity binding sites. In five different liver preparations, the mean binding site concentration was 5.3 + 0.5 pmol/g liver or 699 + 56 fmol/mg protein with mean Kd values of 5.9 + 0.3 ti. This Kd is similar to that derived from the kinetic studies, where Kd = K-IKf =
5.4 nM.
The hormone specificity of the nuclear estradiol binding activity is similar to that THE INFLUENCE OF INCUBATION TEMPERATURE ON of the cytosol, the most notable features SPECIFIC [3H]E~~~~~~~ BINDING BY being the strong competition by DES and NUCLEAR SALT EXTRACT 4-hydroxytamoxifen (Fig. 5). Incubation temperature Specific binding The nuclear [3H]estradiol binding activity (dpm x 10-3) (“Cl differs from the cytosol binder in that it is 2 80.0 more stable on storage at -20”. The half 75.4 10 life is greater than 3 months for the nuclear 20 75.1 extracts but less than 2 weeks for the cy25 63.4 tosol. The two binders also differ in sta44.5 30 bility on sucrose gradient centrifugaNote. Samples of salt extract (100 (*l) were incution. Figure 6 shows that the nuclear bated with 10 mM [3H]estradiol in a total volume of [3H]estradiol binder sediments with a coef300 ~1 for 18 hr at 2” followed by 0.5 hr at the temficient of 3.6 S in 5-20% sucrose gradients perature shown. Specific binding was determined as containing 0.4 M KCl. This is somewhat described in Table 1. TABLE
2
ESTROGEN A P 5 10
BINDERS
.
239
IN SALMON
flect the fact that thioglycerol was present in the salt extraction buffer, which tends to release nuclear matrix-associated receptor (Barrack and Coffey, 1980). The Time Course of the Increase ilz Nuclear E&radio1 Binding Sites following a Single Injection of E&radio1
FIG. 3. Association and dissociation kinetics of [3H]estradiol binding to nuclear salt extract. (A) Specific binding of 10 n&4 [3H]estradiol incubated with 100 yl salt extract (total volume 300 ~1) at 2” for increasing periods of time. (B) Dissociation of [3H]estradiol previously bound by incubation of 100 ~1 salt extract with 10 r&f [3H]estradiol in the presence or absence of 1 @4 DES (total volume 300 ~1) at 2” for 18 hr. Dissociation was initiated by the addition of 1 IJ.M unlabeled estradiol-17S and continued incubation at 2” followed by charcoal-dextran treatment at the indicated times. Remaining specific binding is presented as the log percentage of binding at time zero (to). A control for stability showed no loss of receptor activity over the dissociation period.
less than is found for avian and mammalian estrogen receptors, but is similar to that reported for crude and partly purified preparations of the estrogen receptor in salt extracts of liver nuclei from X. Zaevis (Lazier, 1978; Gschwendt and Schneider, 1978; Jensen et al., 1982; Wright et al., 1983). We also measured [3H]estradiol binding ta the insoluble residue remaining after salt extraction of the nuclei (Snow et al., 1978). Relatively low levels of specific binding were found (0.25 pmolJg liver for estrogentreated salmon, and about 0.03 pmol/g for untreated animals). These values may re-
Fish were injected with a single dose (5 mg/kg) of estradiol and cytosol and nuclear salt extracts were prepared and assayed for [3H]estradiol binding activity at various times thereafter. Both the cytosol and nuclear fractions were charcoal treated prior to assay in order to remove endogenous estradiol. The results inFig, 7 show that cytosol receptor levels are decreased 4 hr after injection, gradually restored over the next 16 hr and are 1.5-2 times the Bevels in uninjected controls by 48-120 hr. Accu-
A
E c -I-
4
FIG. 4. Saturation analysis of [3H]estradiol binding by a salt extract of salmon liver nuclei. (A) Salt extract (100 ~1) was incubated with [3H]estradiol (OS-SO nm for 18 hr at 2” in the presence or absence of a 100X excess of DES in a final volume of 300 ~1, and binding was determined.by the charcoal-dextran method. (0) total binding of 13H]estradiol; (0) nonspecific binding; (A) specific binding. (B) Scatchard analysis of the data given in (A), specific binding. Binding site concentrations, 872 fmolVmg protein; I& 5.3 nM.
240
LAZIER,
10 COMPETITOR
102
12
CONCENTRATION
LONERCAN,
ld I nM I
5. Binding specificity of [3H]estradiol binding by nucelar salt extracts from estrogen-treated salmon. Nuclear preparations (100 ~1) were incubated with 10 nM [3H]estradiol in the absence or presence of potential competitors. Incubation conditions, assay of specific binding, and the symbols used were exactly as described in Fig. 2. FIG.
AND
MOMMSEN
An important feature of both the highand low-affinity [3H]estradiol binding sites in plasma is that neither is competed for by DES. Figure 9 shows that DES does not compete for the high-affinity sites (measured using 20 nM [3H]estradiol). We also found no competition at all for [3H]estradiol binding by a 100 x excess of DES at concentrations of the labeled ligand of up to 500 nit4 (results not shown). Figure 9 further shows that the triphenylethylene antiestrogens do not compete for [3H]estradiol binding to the high-affinity plasma component. In addition, DHT has considerable affinity for the plasma binder, and both progesterone and estrone compete to a lesser but measurable extent. This distinctive hormone specificity clearly differentiates the plasma binder from the DES and antiestrogen-competible binding activ-
mulation of nuclear binding sites is clear by 12 hr and continues over the 4-day period of the experiment. This pattern of estradiol binding site distribution following estrogen treatment is similar to that seen in chickens (Lazier and Haggarty, 1979). The very high concentration of nuclear receptor presumably arises through estradiol-induced synthesis or stabilization of its own receptor.
I
I
[3HjEstradiol Binding by Salmon Plasma
Although liver preparations were well rinsed with saline prior to homogenization, it was possible that some of the intrahepatic estrogen binding arose from contamination with plasma components. Figure 8 shows that salmon plasma does indeed contain [3H]estradiol binding activity, a portion of which appears to have a relatively high affinity (Kd - 13 n&4) and to be saturable. A large number of sites not saturated under the conditions used are also present. In separate experiments these latter sites were found to be saturable at [3H]estradiol concentrations of 200-400 nM.
0
G
1
1
4 \ t
FRACTION
NUMBER
FIG. 6. Sucrose gradient centrifugation of salmon liver nuclear salt extract. A sample of salt extract (100 ~1) from an estradiol-treated fish was incubated in a total volume of 300 ~1 with 10 nM 13H]estradiol in the presence or absence of 1 p.M DES for 18 hr at 2”, charcoal-treated, and applied to a 5-20% sucrose gradient. Centrifugation was for 3 hr at 300,OOOg in a vertical tube rotor. (0)-[14C]Ovalbumin marker, 3.6 S; (G)-[14C]y-globulin marker, 6.6 S.
ESTROGEN
BINDERS
IN SALMON
ities present in liver cytosol and nuclei. The plasma binder instead resembIes the “sex steroid binding protein” described by others in a variety of fishes as well as in higher vertebrates (Martin, 1980). The lack of affinity of DES and of the triphenylethylenes is characteristic of the human plasma sex steroid binding globulin (King and Mainwaring, 1974). Close examination of Fig. 2 shows that a small fraction of cytosol binding sites occupied by 10 nM [3H]estradiol are not competed for by high conentrations of DES or 4-hydroxytamoxifen. This could represent a minor degree of contamination of the cytosol preparation with plasma. DISCUSSION The results presented here show that liver of the Atlantic salmon contains f3H]estradiol binding activity with many
FIG. 8. [3H]estradiol binding by plasma. Varying concentrations of 13H]estradiol in the absence or presence of a 100 x excess of unlabeled estradioL17B were incubated with 285 ~1 of a IO-fold dilution of salmon plasma in cytosol buffer A in a total volume of 300 ~1. Incubatian was for 18 hr at 2” and bound 13H]estradiol was separated from free by charcoal-dextran treatment. (A) Binding curves (@), total; (O), nonspecific; and (A) specific binding of [3Hlestradiol. (R) Scatchard analysis of specific binding from (A). Concentration of the high-affinity component: 16.8 nM in plasma, Kd 13 nM by the Rosenthal method (Rosen-
thal,
24 TIME
48 AFTER
72 INJECTION
96
120 I h I
FIG. 7. The effect of injection of a single dose of estradiol on intracellular distribution of high-affinity estradiol binding sites. ‘Salmon were injected with 5 mgikg estradiol ip and cytosol and nuclear salt extracts were prepared after various time intervals up to 108 hr. Livers from two fish were pooled for each point. The concentration of specific estradiol binding sites in the charcoal-treated extracts was measured by incubation with 10 nM [3]estradiol in the presence and absence of DES at 2” for 18 hr and by correction for the degree of saturation considering the mean K, values previously observed for the cytosol and nuclear binding sites. (e) Nuclear salt extracts; (0) cytosol.
1967).
similarities to the estrogen receptor system target tissues of mammals, birds, amphibia, and reptiles (Gschwendt and Schneider, 1978; Lazier, 1979; Martin, 198Q; Shapiro, 1982). The cytosol receptor in untreated immature fish has the appropriate binding affinity and specificity. It is clearly distinguishable from estradiol-binding activity in plasma on the basis of its relativeiy higher affinity for [3H]estradiol, its high affinity for DES and 4-hydroxytamoxifen, and its lack of affinity for DHT. The concentration of the cytosol estrogen receptor is n&ably high compared to that reported for estrogen receptor in liver from cockerels, X. iuevis, or turtles. In cockerels for example, estrogen receptor is barely detectable in LD-
242
LAZIER,
c_
LONERGAN,
04fa 10 COMPETITOR
IO2 CONCENTRATION
12
IO4 ( “M
I
FIG. 9. Specificity of [3H]estradiol binding by salmon plasma. Plasma, diluted as in Fig. 8, was incubated with 20 nM [3H]estradiol for 18 hr at 2” along with varying concentrations of a potential competitor. Bound [3H]estradiol was separated from free by the charcoal-dextran method. Data are expressed as a percentage of the total [3H]estradiol bound in the absence of competitor. The symbols used are the same as in Fig. 2.
fractionated preparations of liver cytosol, ana only on concentration by ammonium sulfate precipitation with concomitant removal of the nonreceptor hepatic steroid binding protein, can levels of about 20-40 fmol/mg protein or 200-400 fmol/g liver be deteited (Gschwendt, 1977; Lazier and Haggarty, 1979; De Boer et al., 1982). Higher concentrations are transiently observed during chicken embryogenesis at the 19th day of development (Gschwendt, 1977; Lazier, 1980). In X. laevis, workers have generally reported receptor concentrations in terms of sites/cell based on a figure of 1.3 x lo8 cells/g liver (Westley and Knowland, 1978; Hayward et al., 1980). Thus, liver cytosol of untreated male frogs contains about 100 sites of estrogen receptor or 22 fmol/g. In livers of animals that have been estrogentreated and withdrawn for 70 days a cytosol receptor concentration of 1200 sites/cell (260 fmol/g) has been found (Hayward et al., 1980). High-affinity binding of estradiol can be detected in liver cytosol of the adult female turtle, but the concentration is only
AND MOMMSEN
3 fmol/mg protein (Heisermann et al., 1980). Injection of large doses of estradiol, sufficient to stimulate substantial levels of vitellogenin synthesis (Plack et al., 1971; Korsgaard et al., 1976; LeMenn et al., 1980) results in accumulation of a high concentration of estrogen receptor in salmon liver nuclei. The concentrations found here are the highest yet reported for the liver of any oviparous vertebrate on estrogenic stimulation. In untreated cockerels, salt extracts of liver nuclei contain about 200 fmol of estradiol binding sites/g liver (Lazier and Haggarty, 1979; Lazier, 1979). This is similar to that found in the immature salmon and about 1.5 times that in male X. laevis liver nuclei (Westley and Knowland, 1978). Injection of a single high dose of estradiol results in the accumulation of about 2.5 pmol of binding sites/g liver in cockerels (Lazier and Haggerty, 1979; Lazier, 1979) and 0.4 pmollg in X. laevis (Hayward et al., 1980). In the experiments reported here, we find that multiple injections of estradiol result in accumulation of more than 5 pmol of high-affinity sites per gram of tissue 48 hr after the last injection. A single dose of estradiol(5 mg/kg) results in some elevation at 12 hr, followed by a linear rise over the succeeding 4 days with a level of about lo12 pmol/g liver being reached at that time. Cytosol receptors fall to almost undetectable levels 4 hr after estrogen administration, but are fully replenished by 48 hr and almost twice control levels after 4 days. We have not yet studied any period after 4 days. It will be of interest to determine if estrogen treatment results in any long-term changes in receptor distribution and concentration, as appears to be the case in X. laevis (Hayward et al., 1980). We would like to express our data in terms of the number of binding sites/hepatocyte, but are uncertain of the DNA content per liver cell in S. salav or of the number of cells per gram liver. DNA values ranging from 2.5 to 6.5 pg per haploid cell
ESTROGEN
BINDERS
and of 5.6 pg per erythrocyte have been reported for various Salmo species (Fasman, 1976). We determined that the mean DNA content of the liver from untreated S. salar was 2.06 _+ .16 mgig and for the estrogen-treated fish was 1.91 + 0.24 mgig. This gives a mean nuclear receptor concentration of 3.16 t 0.65 pmoli mg DNA for the animals given multiple injections of estradiol. Conservatively assuming a mean hepatocyte DNA content of 6 pg, a figure of about 11,000 sites/cell can be calculated. Even higher values were achieved in the time-course experiment where assays were conducted 4 days after a single injection of estradiol. The high concentration of estrogen receptor ,sites in liver nuclei of the estrogentreated salmon indicate that this might be an excellent source for purification of liver receptor from an oviparous vertebrate. The relative stability of the salt-extracted nuclear receptor on storage at -20” further substantiates this point. Under the influence of estrogen, hepatic metabolism of egg-laying animals undergoes a far-reaching and quite rapid reorganization, predominantly in protein and lipid metabolism (Tata and Smith, 1979; Shapiro, 1982). In teleost fishes the metabolic demand placed on the liver following the binding of estrogen is further compounded through cessation of feeding during spawning migration. We are currently analyzing the metabolic and hormonal control of liver metabolic pathways duri.ng such spawning migration in salmonids (Mommseq et al., 1980). The emphasis has been placed on hepatic gluconeogenesis, since the spawning process per so is known to be fueled exclusively by carbohydrate (French et a!., 1983), spawning occurring at a time when lipid reserves ‘have been depleted. At any point during the migration, vitellogenesis, is a major, direct competitive pathway for gluconeo~enic precursors as well as for ‘I avarlable fatty acids. Therefore’ these studies on estrogen receptors were initiated
IN SALMON
243
as a prerequisite toward understanding the regulation of estrogen responsiveness and induction of vitellogenesis in the framework of hepatic carbohydrate metabolism. ACKNOWLEDGMENTS These studies were supported by a grant to C.B.L. from the Medical Research Couhcil of Canada and by a grant to T.P.M. from the Natural Sciences and Engineering Research Council of Canada.
REFERENCES Barrack, E. R., and Coffey, D. S. (1980). The spec& binding of estrogens and androgens to the nuclear matrix of sex hormone responsive tissues. .?. Bial. Chern. 255, 1265-7275. Burton, K. (1968). Determination of DNA concentration with diphenylamine.,In “Methods in Enzymology” (L. Grossman and F. Moldave, eds.), Vol. 12, pp. 163-166. Academic Press, New York. Crim, L. W., and Idler, D. R. (1978). Plasma gonadotropin, estradiol and vitellogenin and gonad phosvitin levels in relation to seasonal reproductive cycles of female brown trout. Ann. Bid. Anim. Biochim. Biopkys. 18, 1001-1005. De Boer, W., Snippe, L., AB, G., and Gruber, M. (1982). Estrogen receptor in avian embryo and adult liver: Estrogen receptor activation and dissociation kinetics of estradiol, ethynyl estradiol and moxestrol. Endocrinology 110, 1217-1224. Deeley, 92. G., and Goldberger, R. E (1979). Regulation of expression of the vitellogenin gene in avian liver. In “Ontogeny of Receptors and Reproductive Hormone Action”’ (T. H. Hamilton, J. H. Clark, and W. A. Sadler, eds.), pp. 291-308. Raven Press, New York. Dower, W. J., and Ryan, K. .I. (1976). A cytoplasmic estrone-specific binding protein in hen liver. Fed. Proc. 35, 1366 (abstract). Fasman, G. D. (1976). “CRC Handbook of Biochemistry and Molecular Biology.” pp. 293. Chemical Rubber Co., Cleveland/ Ohio. Fostier, A., Weil, C., Terqui, M., Breton. B., and Jalabert, B. (1978). Plasma estradiol-17B and gonadotropin during ovulatian in rainbow trout (Salmo gairdneri R.). Ann. Bibl. Biockim. Biopkys. 18, 929-935. French, C. J,, Hochachka, P. W., and Mommsen, T. I? (1983). Metabolic organization of liver during spawning migration of sockeye salmon. Amer. J. Pkysiol. 245, R827-R830. Gschwendt, M. (1977). A cytoplasmic high-affinity estrogen-binding protein in the embryonic chick liver. Ecru. J. Biockem. 80, 461-468. Gschwendt, M., and Schneider, W. (1978). Estrogen-
LAZIER,
244
LONERGAN,
binding sites of chicken liver. Preliminary characterization of nuclear components. Ear. J. Biochem. 91, 139-149. Hara, A., and Hirai, H. (1978). Comparative studies on immunochemical properties of female-specific serum protein and egg proteins in rainbow trout (Salmo
gairdneri).
Comp.
Biochem.
Physiol.
b
59, 339-344. Hayward, M. A., Mitchell, T. A., and Shapiro, D. J. (1980). Induction of estrogen receptor and reversal of the nuclearicytoplasmic receptor ratio during vitellogenin synthesis and withdrawal in Xenopus laevis. J. Biol. Chem. 255, 1130811312. Heiserman, G. J., Ho, S. M., and Callard, I. P. (1980). Estrogen binding activity in the liver of the eastern painted turtle Chrysemys picta. Amer. Zool. 20, 360 (abstract). Idler, D. R., Hwang, S. J., and Crim, L. W. (1979). Quantification of vitellogenin in Atlantic salmon (Salmo salar) plasma by radioimmunoassay. J. Fish Res. Board
Canad.
36, 574-578.
Jensen, E. V., Green, G. L., Closs, L. E., DeSombre, E. R., and Nadji, M. (1982). Receptors reconsidered: A 20-year perspective. Rec. Prog. Horm. Res. 38, l-33. Jordan, V. C., Collins, M., Rowsby, L., and Prestwich, G. (1977). A monohydroxyl metabolite of tamoxifen with potent antioestrogenic activity. J. Endocrinol. 75, 305-316. King, R. J. B., and Mainwaring, W. I. P. (1974). “Steroid-cell Interactions.” p. 193. Univ. Park Press, Baltimore. Korsgaard, B., Emmersen, J., and Petersen, I. M. (1976). Natural occurrence, and experimental induction by estradiol-17B, of a lipophosphoprotein (vitellogenin) in flounder (Platichtys Jesus). Comp.
Biochem.
Physiol.
B 64, 443-446.
Lambert, J. G. D., Bosman, G. I. C. G. M., van der Hurk, R., and van Oordt, P. G. W. J. (1978). Annual cycle of plasma estradiol-17B in the female trout Salmo gairdneri. Ann. Biol. A&m. Biochim. Biophys. 18, 923-927. Lazier, C. B. (1978). Ontogeny. of the vitellogenic response to oestradiol and of the soluble nuclear oestrogen receptor in embryonic chick liver. Biochem. J. 174, 143-152. Lazier, C. B. (1979). Estrogen-binding proteins in avian liver: Characteristics, and ontogenesis. Zn “Ontogeny of Receptors and Reproductive Hormone Action” (T. H. Hamilton, J. H. Clark, and W. A. Sadler, eds.), pp. 353-370. Raven Press, New York. Lazier, C. B. (1980). The development of estrogen receptors and of the vitellogenic response to estradiol in embryonic chick liver. Adv. Biosci. 25, 125-139.
AND MOMMSEN Lazier, C. B., and Haggarty, A. J. (1979). A high-affinity oestrogen-binding protein in cockerel liver cytosol. Biochem. J. 180, 347-353. Lazier, C. B., and Jordan, V. C. (1982). High affinity binding of antioestrogen to the chick liver nuclear oestrogen receptor. Bbchem. J. 206, 387-394. Le Menn, F., Rochefort, H., and Garcia, M. (1980). Effect of androgen mediated by estrogen receptor of fish liver: Vitellogenin accumulation. Steroids 35, 315-328. Martin, B. (1980). Steroid-protein interactions in nonmammalian vertebrates. Zn “Steroids and Their Mechanism of Action in Nonmammalian Vertebrates” (G. Delvio and J. Brachet, eds.), pp. 6373. Raven Press, New York. Mommsen, T. P., French, C. J., and Hochachka, P. W. (1980). Protein and amino acid utilisation during the spawning migration of salmon. Canad. J. Zool. 58, 1785-1799. Niu, E.-M., Neal, R. M., Pierce, V. K., and Sherman, M. R. (1981). Structural similarities of molybdatestabilized steroid receptors in human breast tumors, uteri and leukocytes. J. Steroid Biochem. 15, l-10. Plack, P. A., Pritchard, D. J., and Fraser, N. W. (1971). Egg proteins in cod serum. Biochem. J. 121, 847-856.
Puca, G. A., Nola, E., Sica, V.., and Brescani, F. (1975). Purification of estrogen receptors. In “Methods in Enzymology” (B. W. O’Malley and J. G. Hardman, eds.), Vol. 36, pp. 331-349. Academic Press, New York. Rosenthal, H. E. (1967). A graphic method for the determination and presentation of binding parameters in a complex system. Anal. Biochem. 20, 525-532. Salhanick, A. R., Vito, C. C., Fox, T. O., and Callard, I. P (1979). Estrogen-binding proteins in the oviduct of the turtle Chrysemys picta: Evidence for a receptor species. Erzdocrinology 105, 13881395. Schneider, W., and Gschwendt, M. (1977). Kinetics of the appearance of nuclear estrogen binding sites in the chick liver. Hoppe Seyler’s Z. Physiol. Chem. 358, 1583-1589. Shapiro, D. (1982). Steroid hormone regulation of vitellogenin gene expression. CRC Crit. Rev. Biochem. 8, 187-203. Snow, L. D., Erikkson, H., Hardin, J. W., Chan, L., Jackson, R. L., Clark, J. H., and Means, A. R. (1978). Nuclear estrogen receptor in the avian liver: Correlation with biologic response. J. Steroid Biochem. 9, 1017-1023. Tata, J. R., and Smith, D. F. (1979). Vitellogenesis: A versatile model for hormonal regulation of gene expression. Recent Prog. Horm. Res. 35, 47-95. Turner, R. T., Dickhoff, W. W., and Gorbman, A.
ESTROGEN
BINDERS
(1981). Estrogen binding to hepatic nuclei of pacific hagfish Eptatrerus stouti. Gen. Comp. Endocrinol.
45, 26-29.
van Bohemen, C. G., Lambert, J. G. D., and Petite, 9. (1981). Annual changes in plasma and liver in relation to vitellogenesis in the female rainbow trout, Salmo gairdneri. Gen. Camp. Endocrinol. 44, 94- 107. Wahli, W., David, I. B., Ryfell, G. U., and Weber, R. (1982). Vitellogenesis and the vitellogenin gene family. Science (Wbshington, D.C.) 212, 298304.
IN SALMON
24.5
Westley, B., and Knowland, J. (1978). An estrogen receptor from Xenopus luevis liver possibly connected with vitellogenin synthesis. Ceil lS, 367374. Wolf, K. (1963). Physiological salines for fresh water teleosts. Prog. Fish Cult. 25, 135-138. Wright, C. V. E., Wright, S. C., and Knowland, .I. (1983). Partial purification of estradiol receptor from Xenopus taevis liver and levels of receptor in relation to estradiol concentration. EWO S. 2, 973-977.