0021-972X/79/4906-0899$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright© 1979 by The Endocrine Society
Vol. 49, No. 6 Printed in U.S.A.
Relative Binding of Testosterone and Estradiol to Testosterone-Estradiol-Binding Globulin* ROBERT A. VIGERSKY, SHINZO KONO, MARK SAUER, MORTIMER B. LIPSETT, AND D. LYNN LORIAUX Reproduction Research Branch, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20014
bound with a Kd of 2.7 X 10 10 M. Examination of the dissociation kinetics of T and E2 from TeBG revealed that the mean (±SD) T1/2 of dissociation of T from plasma at 37 C (10.8 ± 2.4 min) was significantly shortened to 3.5 ± 0.4 min by saturation of plasma with dihydrotestosterone (P < 0.01), whereas that of E2 (8.9 ± 1.4 min) was not changed (9.6 ± 3.0 min). These data suggest that TeBG is not an important binder of plasma E2 at physiological temperatures and explain the observation that in diseases characterized by high TeBG levels, such as hyperthyroidism and liver disease, the MCR and free E2 levels have generally been normal. (J Clin Endocrinol Metab 49: 899, 1979)
ABSTRACT. The binding of estradiol (E2) and testosterone (T) to testosterone-estradiol-binding globulin (TeBG) was studied in vivo at 37 C by three independent methods: equilibrium dialysis, steady state polyacrylamide gel electrophoresis, and TeBG-ligand dissociation kinetics. Equilibrium dialysis was performed at 37 C with the dialysate containing human serum albumin in amounts equivalent to that of the plasma dialysand. Scatchard analysis indicated that under these conditions E2 does not measurably bind to TeBG, while T has a Kd of 3.7 x 10~10 M. Similarly, Scatchard-type analysis of E2 binding to TeBG in steady state polyacrylamide gel electrophoresis at 37 C revealed no high affinity saturable binding, while dihydrotestosterone was
T
amide gel electrophoresis, and dissociation rate analysis. These studies suggest that TeBG is not an important binder of plasma E2 at 37 C.
HE AVAILABILITY of plasma steroids to target tissues seems to be regulated to some degree by plasma protein binding. Numerous in vitro studies have demonstrated that the plasma /?-globulin for sex steroid binding, usually referred to as testosterone-estradiolbinding globulin (TeBG), binds testosterone (E2) with high affinity (1-9). These studies suggest that E2 binds to TeBG with an affinity about two thirds that of T. The in vitro demonstration of high affinity binding of a steroid for a given binding protein, however, does not necessarily correlate with the degree of in vivo binding. No studies have used classic equilibrium dialysis methods to determine the binding affinity of E2 to whole plasma at physiological temperatures (Table 1). As a first step in the development of reliable methods for the measurement of nonprotein-bound T and E2 and to assess the importance of TeBG in the regulation of E2 metabolism, the binding characteristics of plasma for T and E2 at 37 C have been studied using equilibrium dialysis, steady state polyacryl-
Materials and Methods Equilibrium dialysis Seven pools of mixed male and female human adult plasma were diluted 1:5 in 0.9% saline. In addition, individual pools of male or female plasma were similarly studied with and without the removal of endogenous steroids. Endogenous steroids were removed by stirring plasma at 20 C with 2% Darco G-60 charcoal (Atlas Chemical Industries, Wilmington, DE) for 24 h. This procedure removes more than 98% of sex steroids. One-milliliter aliquots of diluted plasma were placed in cellulose dialysis tubing (1-cm flat width; 6000-8000 mol wt cutoff) with a glass bead to speed the attainment of equilibrium. The tubing had been presoaked for 60 min in 0.9% saline. The plasma samples were dialyzed in Wheaton glass liquid scintillation counting vials (Wheaton Scientific, Millville, NJ) at 4, 22, or 37 C for 24 h against 10 ml 0.9% saline containing: 1) human serum albumin (Sigma Chemical Co., St. Louis, MO), 1 g/100 ml; 2) 7 X 104 cpm [1,2-3H]T (SA, 40 Ci/mmol) or 8 x 104 cpm [6,7-3H]E2 (SA, 48 Ci/mmol); and 3) varying amounts up to 1000-fold molar excess of either unlabeled T or E2. The counting vials were placed in a Dubnoff metabolic shaker (Fisher Scientific Co., Pittsburgh, PA) and shaken for the duration of the experiment. Equilibrium was obtained within 18 h. Recovery of radioactivity was 95-102% after dialysis. The purity of the
Received August 29,1977. Address requests for reprints to: Dr. Robert A. Vigersky, Kyle Metabolic Unit, Walter Reed Army Medical Center, Washington, D.C. 20012. * Presented in part at the 56th Annual Meeting of The Endocrine Society, June 12,1974. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or reflecting the views of the Department of the Army or the Department of Defense.
899
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
VIGERSKY ET AL.
900 TABLE 1. Studies of E2 binding to TeBG
TemperAffinity ature Ka x 108 (C)
JCE&M • 1979 VoU9 • No 6
either 0 or 37 C. Gels were sliced into 1 ± 0.2-mm sections (12) after electrophoresis, placed into 10 ml LSC Complete, and counted. The amount of steroid bound was proportional to the area under the TeBG peak, and the amount free was proportional to the area of the base of the peak, as previously described (12). Areas were determined by computer. The temperature within the resolving gel during electrophoresis was determined by placing the indicator section of a Tempa-DOT (Organon, Inc., West Orange, NJ) in a gel tube and polymerizing the gel solution around it. The Tempa-DOT indicators were also placed in the upper and lower buffer chambers. The gel temperature never rose above 98.8 F (37.1 C) and was within 0.2 F of the buffer temperatures.
Authors
Method
Pearlman et al. (6) Mercier-Bodard et al. (1) Rosner et al. (5) Rosner and Smith (14) Wu et al. (16) Murphy (4) Rosenbaum et al.
Sephadex dialysis Equilibrium dialysis
25 4
0.43 4.60
Equilibrium dialysis Equilibrium dialysis
4 37
1.1-40" 2.2*
Charcoal separation Florisil Paper electrophoresis
37 10
0.66 NDC ND
Steady state gel filtration Steady state gel filtration Equilibrium dialysis
37
6.0-6.4
Dissociation rate analysis
37
ND
4
ND
Pooled human plasma (diluted 1:5 with 0.9% saline), human serum albumin (1 g/100 ml), or 0.9% saline were incubated with 1 X 105 cpm/ml [3H]T or [3H]E2 for 30 min at 37 C. One milliliter of each was placed in Visking dialysis tubing and immersed in a beaker containing 250 ml 0.9% saline. The temperature was maintained at 37 C with a warm water bath. Constant stirring was achieved manually for the first 60 min. At 2-min intervals for the first 20 min, then at 5-min intervals for an additional 40 min, 5 ml dialysate were removed and counted in 10 ml LSC Complete. The dialysate was immediately replaced with 5 ml 0.9% saline at 37 C to keep the dialysate volume constant. Samples were obtained at 0 h and after 18 h of shaking in a Dubnoff metabolic shaker. These represented background activity and binding at equilibrium, respectively. Total activity of each sample was determined by counting 0.1 ml before dissociation. In a second set of experiments, plasma and 0.9% saline were treated as above but were first preincubated with 6 /ig/100 ml (for plasma) or 300 /ng/100 ml (for saline) of unlabeled DHT or E2 for 18 h at 37 C before incubation with either [3H]T or [3H]E2. Calculations of ti/2s of dissociation were as follows:
4
(3)
Burke and Anderson (7) Fisher and Anderson (9) Tavernetti et al. (8)
" Partially purified preparation. 6 Highly purified preparation. c ND, Not determined.
albumin was assessed by electrophoresis in polyacrylamide gel (see below) at 0 C. Gels were stained with 2 ml 0.25% Coomassie brilliant blue G 250 for 18 h after being fixed for 1 h in 10 ml 12.5% trichloroacetic acid. Two protein bands with the mobility of albumin monomer and dimer were detected. The purity of the steroids was determined by thin layer chromatography, and only steroids of greater than 98% purity were used in these studies. One milliliter of dialysate and 0.5 ml dialysand were counted in 3 ml dioxane and 10 ml liquid scintillation fluid (LSC Complete, Yorktown Research Co., South Hackensack, NJ) using a Packard Tri-Carb liquid scintillation counter (Packard, Downers Grove, IL). Calculations were made as outlined by Slaunwhite (10). It was not necessary that the albumin concentration of the dialysate exactly match that of the plasma. This was determined by performing a series of dialyses with E2 at 37 C in which the albumin concentration of the dialysate ws 50%, 75%, 100%, 150%, or 200% of the plasma albumin concentration and finding that the results were identical with those in which the concentration was 1 g/100 ml. Steady state polyacrylamide gel electrophoresis (SS-PAGE) The apparent Kd was determined by Scatchard-type analysis using a modification of the method outlined by Ritzen et al. (11) and Vigersky et al. (12). The procedure allowed the calculation of an apparent Kd for the binding of the specific protein of interest with the ligand, which is included in the gel before the sample is analyzed. Electrophoresis was performed at 0 and 37 C in resolving gels at 10% total gel concentration with 2% cross-linking using methylene-bis-acrylamide in the multiphasic buffer system D (pH 7.8), as previously described (12). Stacking gels of 3.125% total gel concentration with 20% cross-linking were used. Stacking and resolving gels were polymerized in the presence of 1-6 nM [l,2-3H]dihydrotestosterone ([3H]DHT; SA, 44 Ci/mmol) or 6-15 nM [3H]E2. Samples were preincubated with corresponding amounts of radioactive steroid for 2 h at
BT = TA - 50 (cpmt - z) where B T is the total bound steroid at any time before equilibrium, where all steroid inside the dialysis bag is, for practical purposes, considered bound; TA is the total activity of steroid in plasma; cpmt is the counts per min in dialysate at any time before equilibrium; and z is the background counts. Beq = TA - 50 (cprrieq - z) where Beq is the total bound steroid at equilibrium and cpme(, is the counts per min in dialysate at equilibrium. Ba
=
BT
~~ Beq
where Ba is the amount of bound steroid available for dissociation. a
_
(BTQVO)
-
~
°
In 2 t1/2 = — where ti /2 is the half time of dissociation and X is the negative slope of regression of log %Ba vs. time.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
TeBG BINDING OF PLASMA E2
Results
901
SS-PAGE
Equilibrium dialysis The Scatchard plots of T and E2 at 4, 22, and 37 C with a pool of normal human plasma are shown in Fig. 1. The Kds (X 1(T9) for T at 4, 22, and 37 C were 2.1, 0.31, and 0.37, respectively. Initial binding at 37 C for [3H]E2 was never sufficient to be differentiated from zero. The results of the E2 binding analyses for all seven plasma pools are presented in Table 2. E2 binding studies were also done on individual pools of male or female plasma with and without endogenous steroids removed. No detectable binding was found in any of the experiments, except in one of seven studies with stripped female plasma (Kd = 3.3 x 10"9; binding capacity, 0.1 ju,g/100 ml). It should be noted that this binding capacity was approximately 3% of that found using T as the ligand with this plasma pool.
A typical analysis of E2 and T binding to TeBG at 37 C in SS-PAGE is shown in Fig. 2. It can be seen that the Kd at 37 C for E2 was indeterminate. The Kds for DHT and E2 at 0 and 37 C, as derived from Scatchard-type analysis, are shown in Table 3. Kinetic studies The ti/2s of dissociation of E2 and T from plasma, albumin, and DHT- or E2-saturated plasma are shown in Table 4. Both T and E2 dissociate from albumin more rapidly than from plasma (P < 0.01). The ti /2 of T, but not of E2, is shortened by saturating the plasma with DHT (Fig. 3). Saturation of plasma with E2 shortens the ti/2 of both T and E2 (Table 4). Saturation of albumin with E2 or DHT does not affect the ti/2 of either T or E2 (Fig. 4). The addition of 300 /Ag/100 ml unlabeled T or E2 TABLE 2. K^s of E2 in normal human plasma Kd x 1(T8 M Pool no. I
II T-2I°
A
T-37-
o
E2-4«
3
E.-2I-
a
III IV V VI
VII Mean ± SD 0 6
4C
22C
37 C
5.0 2.6 3.3 5.9 6.7 6.7 3.7
9.1 4.0 6.3 (-) (-) (—) (-)
ND" ND (-)" (-) (-) (-) (-)
4.8 ± 1.7
6.4 ± 2.6
ND, Not done. (—), Not determinable.
• DHT
13
° E2 11 9 B/F
\ 7 5 3 o 1
0 0
°
| O° 100
2
3 4 BOUND (xlCT 8 m/l)
FIG. 1. Scatchard plots of T and E2 with a pool of human plasma at 4, 22, and 37 C, as determined by equilibrium dialysis (see text for methodological details). Bound/free, The ratio of TeBG-bound to nonTeBG-bound steroid. Binding of E 2 was indeterminate at 37 C.
° °
.
1
200
300
\
400
BOUND (pg)
FIG. 2. Scatchard plot of DHT and E2 binding to TeBG, as determined by SS-PAGE. Aliquots (lO/il) of plasma were applied to gels containing 1-6 nM [ 3 H]DHT or 6-15 nM [3H]E2, and electrophoresis was performed at 4 or 37 C. Plasma was preincubated at 4 or 37 C with the corresponding amount of steroid.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
VIGERSKY ET AL.
902
JCE&M • 1979 Vol49 • No6 3
TABLE 3. Kds of DHT and E2 in SS-PAGE at 0 and 37 C
0C (M) 1.5 X 1(T9 3.3 X 10"10
DHT E2
37 C (M) 2.7 X 10"
40
' Not determinable.
20
TABLE 4. Half-times of dissociation (in minutes) of E2 and T from plasma, albumin, and E2-saturated or DHT-saturated plasma
E2 T
Plasma (n = 4)
Albumin (n = 7)
DHT-plasma
8.9 ± 1.4 10.8 ± 2.4
3.6 ± 0.8" 2.6 ± 1.2°
9.6 ± 3.0 3.5 ± 0.4"
3
H-Testosterone
100 80 iV 60 - V
E2-plasma (n = 4)
%Ba
10
H-Estradiol
- * - Albumin •-••• E2-Albumin - • - DHT-Albumin
1 *
—+- Albumin ••«--• E2 -Albumin -•-DHT-Albumin
" r i
5 \
\
4.9 ± 1.2" 2.4 ± 0.4° 1 i . i . i
Values given are the mean ± SD. a P < 0.01 compared to plasma.
2
6 10 14 18
I iI•I ,
25 30
2
6 10 14 18
25 30
MINUTES 3
40
%B a
vx
\ -
\ \ \
10
\
5
H-Estradiol -+— Plasma •••••• E2-Plasma - • - DHT-Plasma
Plasma E2-Plasma DHT-Plasma
100 80 60
20
3
H-Testosterone
> ' \ \ \ \ *\ • \
V
\
\
\ 1 I.I.I 2
, \ . I ,\
6 10 14 18
I 25
30
2
I i I • 6 10 14 18
I 25
30
MINUTES
FIG. 3. Dissociation of [3H]T and [3H]E2 at 37 C from untreated plasma or plasma preincubated with saturating amounts of DHT or E2 at 37 C for 18 h before analysis. Ba, Amount of steroid available for dissociation.
to [3H]T or [3H]E2 in 0.9% saline did not increase the rate of exit of labeled steroids from the dialysis bag, indicating that there was no binding to the dialysis tubing. Discussion These data suggest that TeBG is not an important binder of plasma E2 at physiological temperatures. Three independent in vitro methods yield the same conclusion. The method of dialysis used in these studies differs from most of those previously reported in that albumin has been included in the dialysate. However, Vermeulen and Verdonck (2) in examining TeBG binding properties and Slaunwhite (10) in examining the binding characteristics of transcortin used albumin in the dialysate. We have calculated the Kd in a manner analogous to theirs.
FIG. 4. Dissociation of [3H]T and [3H]E2 at 37 C from untreated human albumin or albumin preincubated with saturating amounts of DHT or E2 at 37 C for 18 h before analysis. Ba, Amount of steroid available for dissociation.
It was not necessary to precisely equalize the albumin concentrations of dialysate and dialysand, as shown by the failure to influence estradiol binding at 37 C using a wide range of albumin concentrations. This is not surprising given the essentially unlimited steroid-binding capacity of the albumin. The inclusion of albumin in the dialysate not only reduces the amount of steroid bound at low concentrations of ligand (as would be expected), but in the case of nonalbumin T binding, unmasks nonlinearity of the Scatchard plot. The latter may represent either heterogeneity of binding sites or negative cooperativity. With E2 no binding could be detected at 37 C that could not be attributed to albumin. This corroborates the findings of Bishoff and Stauffer (13), who found that at 37 C under conditions of nonequilibrium dialysis E2 binding to human serum could be attributed to albumin alone. Three previous studies have investigated E2 binding to TeBG at 37 C (7,14,15), but only Rosner and Smith (14) used equilibrium dialysis. They found that the association constant for E2 for purified TeBG was 6.0 ± 1.1 X 108 M" 1 at 37 C. This is not necessarily in conflict when the markedly different dialysis conditions (purified TeBG and no albumin in the dialysate) are considered. The dialysis system we have used was designed to simulate in vivo conditions. In this system, one might predict that most of the binding for E2 would be accounted for by albumin and most of the binding for T would be accounted for by TeBG since: 1) the binding affinity of E2 to TeBG is 1 order of magnitude less than that of T (Fig. 1); 2) the binding affinity of E2 for albumin is 1 order of magnitude greater than that for T (10); and 3) albumin is present in much greater concentrations than TeBG. Our data, performed under these simulated physiological
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
TeBG BINDING OF PLASMA E2
conditions, would tend to confirm this prediction. SS-PAGE permits the observation of the binding characteristics of the specific protein of interest when separated from other plasma proteins. The KdS for DHT in SS-PAGE at 0 and 37 C are within 1 order of magnitude of those determined by equilibrium dialysis of purified TeBG (14), suggesting that the Kd obtained by the method of equilibrium dialysis used here does not represent a composite of TeBG and other high affinity, nonTeBG proteins. These results also suggest that no significant damage occurs to TeBG during the 37 C electrophoresis. The kinetic data (Table 4) indicate that the DHTbinding site on TeBG is not (quantitatively) an important binder of plasma E2 at 37 C. This is seen in the experiments where the t1/2 of dissociation of E2 from plasma is not changed by DHT saturation of binding sites. The finding that E2 significantly shortens the ti/2 of E2 dissociation from plasma suggests that saturable, nonandrogen-binding sites do exist in plasma. In general, our data agree with those of Heyns and DeMoor (15), who, using a different experimental design, showed that the ti /2 s of dissociations of E2 from TeBG at 37 C were too rapid to be measured. Our findings do not agree with those of Wu et al. (16), who, using a 30-min incubation at 37 C and separating bound from free hormones with dextrancoated charcoal, found that E2 bound to TeBG with a Ka of 6.58 x 107 M""1 and that TeBG had a binding capacity for E2 of 16.6 ± 0.43 ng/ml. They also found that 38.4 ± 0.7% of E2 was bound to TeBG. It should be noted, however, that the 30-min incubation at 37 C followed an incubation performed at 4 C for 10 min. They found, nevertheless, that the percentage of free E2 determined by equilibrium dialysis was the same at 4, 22, and 7 C, suggesting that TeBG was not an important factor in determining the bound to free ratio. The observation that T but not E2 binds to TeBG at 37 C may shed light on the findings of several clinical studies. Galvao-Teles et al. (17) found that in patients with liver disease and elevated plasma TeBG levels, plasma free T was low while free E2 was normal. Olivo et al. (18) found no decrease in the MCR of E2 in a similar group of patients. Another disorder in which high plasma TeBG is found is hyperthyroidism. Chopra and Tulchinsky (19) observed in such patients that, while elevations in total T were associated with a normal free T level, elevations in total E2 were associated with elevated free E2 levels. Additionally, Ruder et al. (20) showed that in three normal male volunteers made hyperthyroid with T3, the associated rise in TeBG was accompanied by a fall in the MCR of T without a change in the MCR of E2. In hirsutism, a disorder associated with low TeBG levels, Fisher and Anderson (9) showed that the free T was higher than that of normal women, but free E2 was no
903
different. Thus, these data are in general accord with the proposition that TeBG binding of E2 is of little physiological significance in man. On the other hand, Ridgway et al. (21) have recently shown that the MCR of E2 is diminished in hyperthyroidism. The MCR of E2 returned to normal with resolution of the hyperthyroidism. The decreased MCR of E2 may be explained, in part, by the decrease in adipose tissue and muscle mass in hyperthyroidism. Muscle and fat are actively involved in E2 metabolism (21), and a decrease in the mass of these tissues may explain the decreased MCR of E2 in hyperthyroidism, and why it is about half that found in liver disease (18, 21). The lack of significant E2 binding to TeBG at 37 C may explain the coexistence of gynecomastia and normal or slightly elevated plasma E2 levels in several disorders associated with elevated TeBG, such as hyperthyroidism, cirrhosis, and Klinefelter's syndrome as well as in puberty and senescence. Free T is decreased while free E2 is unchanged. Conversely, in disorders associated with decreased TeBG levels, such as idiopathic hirsutism or hirsutism secondary to acromegaly or Cushing's syndrome, the free T is increased but free E2 is unchanged. Thus, a common denominator in these disorders may be an abnormal free E2 to free T ratio that results from the binding characteristics of these steroids to TeBG at physiological temperatures. References 1. Mercier-Bodard, C, A. Alfsen, and E. E. Baulieu, Sex-steroid binding plasma protein, Acta Endocrinol [Suppl] (Kbh) 147: 204, 1970. 2. Vermeulen, A., and L. Verdonck, Studies on the binding of testosterone to human plasma, Steroids 11: 609,1968. 3. Rosenbaum, W., N. P. Christy, and W. G. Kelly, Electrophoretic evidence for the presence of an estrogen binding ^-globulin in human plasma, J Clin Endocrinol Metab 26: 1399, 1966. 4. Murphy, B. E. P., Binding of testosterone and estradiol in plasma, CanJBiochem 46: 299, 1968. 5. Rosner, W., N. P. Christy, and W. G. Kelley, Partial purification and preliminary characterization of estrogen-binding globulin from human plasma, Biochemistry 8: 3100, 1969. 6. Pearlman, W. H., I. F. F. Fong, and J. H. Tou, A further study of a testosterone-binding component of human pregnancy plasma, J Biol Chem 244: 1373, 1969. 7. Burke, C. W., and D. C. Anderson, Sex-hormone-binding-globulin is an oestrogen amplifier, Nature 240: 38, 1972. 8. Tavernetti, R. R., W. Rosenbaum, W. G. Kelley, N. P. Christy, and M. S. Roginsky, Evidence for the presence in human plasma of an estrogen-binding factor other than albumin: abnormal binding of estradiol in men with hepatic cirrhosis, J Clin Endocrinol Metab 27: 920,1967. 9. Fisher, R. A., and D. C. Anderson, Simultaneous measurement of unbound testosterone and estradiol fractions in undiluted plasma at 37°C by steady-state gel filtration, Steroids 24: 809, 1974. 10. Slaunwhite, Jr., W. R., The binding of estrogens, androgens and progesterone by plasma proteins in vitro, In Antonaides, H. S. (ed.), Hormones in Human Plasma, Little, Brown, and Co., Boston, 1960, p. 478. 11. Ritzen, E. M., F. S. French, S. C. Weddington, S. N. Nayfeh, and V. Hansson, Steroid binding in polyacrylamide gels. Quantitation at
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
904
VIGERSKY ET AL.
steady state conditions, J Biol Chem 249: 6597, 1974. 12. Vigersky, R. A., D. L. Loriaux, S. S. Howards, G. B. Hodgen, M. B. Lipsett, and A. C. Chrambach, Androgen binding proteins of the testis, epididymis, and plasma in man and monkey, J Clin Invest 58: 1061, 1976. 13. Bishoff, F., and R. D. Stauffer, Orientation of circulating human estrogens by albumin, Am J Physiol 191: 313, 1957. 14. Rosner, W., and R. N. Smith, Isolation and characterization of the testosterone-estradiol binding globulin from human plasma. Use of a novel affinity column, Biochemistry 14: 4813, 1975. 15. Heyns, W., and P. DeMoor, Kinetics of dissociation of 17 /?-hydroxysteroids from the steroid binding /J-globulin of human plasma, J Clin Endocrinol Metab 32: 147, 1971. 16. Wu, C. H., T. Motohashi, H. A. Abdel-Rahman, G. Flickinger, and G. Mikhail, Free and protein-bound plasma estradiol-17/? during the menstrual cycle, J Clin Endocrinol Metab 43: 436, 1976.
JCE&M Vol49
i 1979
No 6
17. Galvao-Teles, A., D. C. Anderson, C. W. Burke, J. C. Marshall, C. S. Corker, R. L. Brown, and M. L. Clark, Biologically active androgens and oestradiol with chronic liver disease, Lancet 1: 173, 1973. 18. Olivo, J., G. G. Gordon, F. Rafil, and A. L. Southren, Metabolic clearance rates (MCR) of estradiol (E2) in cirrhosis and hyperthyroidism, Endocrinology (Suppl) 94,1974 (Abstract 282). 19. Chopra, J. J., and D. Tulchinsky, Status of estrogen-androgen balance in hyperthyroid men with Grave's disease, J Clin Endocrinol Metab 38: 269, 1974. 20. Ruder, H., P. Corvol, J. A. Mahoudeau, G. T. Ross, and M. B. Lipsett, Effects of iriduced hyperthyroidism on steroid metabolism in man, J Clin Endocrinol Metab 33: 382, 1971. 21. Ridgway, E., C. C. Longcope, and F. Maloof, Metabolic clearance and blood production of estradiol in hyperthyroidism, J Clin Endocrinol Metab 41: 491, 1975.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
Vol. 49, No. 6 Printed in U.S.A.
Relative Binding of Testosterone and Estradiol to Testosterone-Estradiol-Binding Globulin* ROBERT A. VIGERSKY, SHINZO KONO, MARK SAUER, MORTIMER B. LIPSETT, AND D. LYNN LORIAUX Reproduction Research Branch, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20014
bound with a Kd of 2.7 X 10 10 M. Examination of the dissociation kinetics of T and E2 from TeBG revealed that the mean (±SD) T1/2 of dissociation of T from plasma at 37 C (10.8 ± 2.4 min) was significantly shortened to 3.5 ± 0.4 min by saturation of plasma with dihydrotestosterone (P < 0.01), whereas that of E2 (8.9 ± 1.4 min) was not changed (9.6 ± 3.0 min). These data suggest that TeBG is not an important binder of plasma E2 at physiological temperatures and explain the observation that in diseases characterized by high TeBG levels, such as hyperthyroidism and liver disease, the MCR and free E2 levels have generally been normal. (J Clin Endocrinol Metab 49: 899, 1979)
ABSTRACT. The binding of estradiol (E2) and testosterone (T) to testosterone-estradiol-binding globulin (TeBG) was studied in vivo at 37 C by three independent methods: equilibrium dialysis, steady state polyacrylamide gel electrophoresis, and TeBG-ligand dissociation kinetics. Equilibrium dialysis was performed at 37 C with the dialysate containing human serum albumin in amounts equivalent to that of the plasma dialysand. Scatchard analysis indicated that under these conditions E2 does not measurably bind to TeBG, while T has a Kd of 3.7 x 10~10 M. Similarly, Scatchard-type analysis of E2 binding to TeBG in steady state polyacrylamide gel electrophoresis at 37 C revealed no high affinity saturable binding, while dihydrotestosterone was
T
amide gel electrophoresis, and dissociation rate analysis. These studies suggest that TeBG is not an important binder of plasma E2 at 37 C.
HE AVAILABILITY of plasma steroids to target tissues seems to be regulated to some degree by plasma protein binding. Numerous in vitro studies have demonstrated that the plasma /?-globulin for sex steroid binding, usually referred to as testosterone-estradiolbinding globulin (TeBG), binds testosterone (E2) with high affinity (1-9). These studies suggest that E2 binds to TeBG with an affinity about two thirds that of T. The in vitro demonstration of high affinity binding of a steroid for a given binding protein, however, does not necessarily correlate with the degree of in vivo binding. No studies have used classic equilibrium dialysis methods to determine the binding affinity of E2 to whole plasma at physiological temperatures (Table 1). As a first step in the development of reliable methods for the measurement of nonprotein-bound T and E2 and to assess the importance of TeBG in the regulation of E2 metabolism, the binding characteristics of plasma for T and E2 at 37 C have been studied using equilibrium dialysis, steady state polyacryl-
Materials and Methods Equilibrium dialysis Seven pools of mixed male and female human adult plasma were diluted 1:5 in 0.9% saline. In addition, individual pools of male or female plasma were similarly studied with and without the removal of endogenous steroids. Endogenous steroids were removed by stirring plasma at 20 C with 2% Darco G-60 charcoal (Atlas Chemical Industries, Wilmington, DE) for 24 h. This procedure removes more than 98% of sex steroids. One-milliliter aliquots of diluted plasma were placed in cellulose dialysis tubing (1-cm flat width; 6000-8000 mol wt cutoff) with a glass bead to speed the attainment of equilibrium. The tubing had been presoaked for 60 min in 0.9% saline. The plasma samples were dialyzed in Wheaton glass liquid scintillation counting vials (Wheaton Scientific, Millville, NJ) at 4, 22, or 37 C for 24 h against 10 ml 0.9% saline containing: 1) human serum albumin (Sigma Chemical Co., St. Louis, MO), 1 g/100 ml; 2) 7 X 104 cpm [1,2-3H]T (SA, 40 Ci/mmol) or 8 x 104 cpm [6,7-3H]E2 (SA, 48 Ci/mmol); and 3) varying amounts up to 1000-fold molar excess of either unlabeled T or E2. The counting vials were placed in a Dubnoff metabolic shaker (Fisher Scientific Co., Pittsburgh, PA) and shaken for the duration of the experiment. Equilibrium was obtained within 18 h. Recovery of radioactivity was 95-102% after dialysis. The purity of the
Received August 29,1977. Address requests for reprints to: Dr. Robert A. Vigersky, Kyle Metabolic Unit, Walter Reed Army Medical Center, Washington, D.C. 20012. * Presented in part at the 56th Annual Meeting of The Endocrine Society, June 12,1974. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or reflecting the views of the Department of the Army or the Department of Defense.
899
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
VIGERSKY ET AL.
900 TABLE 1. Studies of E2 binding to TeBG
TemperAffinity ature Ka x 108 (C)
JCE&M • 1979 VoU9 • No 6
either 0 or 37 C. Gels were sliced into 1 ± 0.2-mm sections (12) after electrophoresis, placed into 10 ml LSC Complete, and counted. The amount of steroid bound was proportional to the area under the TeBG peak, and the amount free was proportional to the area of the base of the peak, as previously described (12). Areas were determined by computer. The temperature within the resolving gel during electrophoresis was determined by placing the indicator section of a Tempa-DOT (Organon, Inc., West Orange, NJ) in a gel tube and polymerizing the gel solution around it. The Tempa-DOT indicators were also placed in the upper and lower buffer chambers. The gel temperature never rose above 98.8 F (37.1 C) and was within 0.2 F of the buffer temperatures.
Authors
Method
Pearlman et al. (6) Mercier-Bodard et al. (1) Rosner et al. (5) Rosner and Smith (14) Wu et al. (16) Murphy (4) Rosenbaum et al.
Sephadex dialysis Equilibrium dialysis
25 4
0.43 4.60
Equilibrium dialysis Equilibrium dialysis
4 37
1.1-40" 2.2*
Charcoal separation Florisil Paper electrophoresis
37 10
0.66 NDC ND
Steady state gel filtration Steady state gel filtration Equilibrium dialysis
37
6.0-6.4
Dissociation rate analysis
37
ND
4
ND
Pooled human plasma (diluted 1:5 with 0.9% saline), human serum albumin (1 g/100 ml), or 0.9% saline were incubated with 1 X 105 cpm/ml [3H]T or [3H]E2 for 30 min at 37 C. One milliliter of each was placed in Visking dialysis tubing and immersed in a beaker containing 250 ml 0.9% saline. The temperature was maintained at 37 C with a warm water bath. Constant stirring was achieved manually for the first 60 min. At 2-min intervals for the first 20 min, then at 5-min intervals for an additional 40 min, 5 ml dialysate were removed and counted in 10 ml LSC Complete. The dialysate was immediately replaced with 5 ml 0.9% saline at 37 C to keep the dialysate volume constant. Samples were obtained at 0 h and after 18 h of shaking in a Dubnoff metabolic shaker. These represented background activity and binding at equilibrium, respectively. Total activity of each sample was determined by counting 0.1 ml before dissociation. In a second set of experiments, plasma and 0.9% saline were treated as above but were first preincubated with 6 /ig/100 ml (for plasma) or 300 /ng/100 ml (for saline) of unlabeled DHT or E2 for 18 h at 37 C before incubation with either [3H]T or [3H]E2. Calculations of ti/2s of dissociation were as follows:
4
(3)
Burke and Anderson (7) Fisher and Anderson (9) Tavernetti et al. (8)
" Partially purified preparation. 6 Highly purified preparation. c ND, Not determined.
albumin was assessed by electrophoresis in polyacrylamide gel (see below) at 0 C. Gels were stained with 2 ml 0.25% Coomassie brilliant blue G 250 for 18 h after being fixed for 1 h in 10 ml 12.5% trichloroacetic acid. Two protein bands with the mobility of albumin monomer and dimer were detected. The purity of the steroids was determined by thin layer chromatography, and only steroids of greater than 98% purity were used in these studies. One milliliter of dialysate and 0.5 ml dialysand were counted in 3 ml dioxane and 10 ml liquid scintillation fluid (LSC Complete, Yorktown Research Co., South Hackensack, NJ) using a Packard Tri-Carb liquid scintillation counter (Packard, Downers Grove, IL). Calculations were made as outlined by Slaunwhite (10). It was not necessary that the albumin concentration of the dialysate exactly match that of the plasma. This was determined by performing a series of dialyses with E2 at 37 C in which the albumin concentration of the dialysate ws 50%, 75%, 100%, 150%, or 200% of the plasma albumin concentration and finding that the results were identical with those in which the concentration was 1 g/100 ml. Steady state polyacrylamide gel electrophoresis (SS-PAGE) The apparent Kd was determined by Scatchard-type analysis using a modification of the method outlined by Ritzen et al. (11) and Vigersky et al. (12). The procedure allowed the calculation of an apparent Kd for the binding of the specific protein of interest with the ligand, which is included in the gel before the sample is analyzed. Electrophoresis was performed at 0 and 37 C in resolving gels at 10% total gel concentration with 2% cross-linking using methylene-bis-acrylamide in the multiphasic buffer system D (pH 7.8), as previously described (12). Stacking gels of 3.125% total gel concentration with 20% cross-linking were used. Stacking and resolving gels were polymerized in the presence of 1-6 nM [l,2-3H]dihydrotestosterone ([3H]DHT; SA, 44 Ci/mmol) or 6-15 nM [3H]E2. Samples were preincubated with corresponding amounts of radioactive steroid for 2 h at
BT = TA - 50 (cpmt - z) where B T is the total bound steroid at any time before equilibrium, where all steroid inside the dialysis bag is, for practical purposes, considered bound; TA is the total activity of steroid in plasma; cpmt is the counts per min in dialysate at any time before equilibrium; and z is the background counts. Beq = TA - 50 (cprrieq - z) where Beq is the total bound steroid at equilibrium and cpme(, is the counts per min in dialysate at equilibrium. Ba
=
BT
~~ Beq
where Ba is the amount of bound steroid available for dissociation. a
_
(BTQVO)
-
~
°
In 2 t1/2 = — where ti /2 is the half time of dissociation and X is the negative slope of regression of log %Ba vs. time.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
TeBG BINDING OF PLASMA E2
Results
901
SS-PAGE
Equilibrium dialysis The Scatchard plots of T and E2 at 4, 22, and 37 C with a pool of normal human plasma are shown in Fig. 1. The Kds (X 1(T9) for T at 4, 22, and 37 C were 2.1, 0.31, and 0.37, respectively. Initial binding at 37 C for [3H]E2 was never sufficient to be differentiated from zero. The results of the E2 binding analyses for all seven plasma pools are presented in Table 2. E2 binding studies were also done on individual pools of male or female plasma with and without endogenous steroids removed. No detectable binding was found in any of the experiments, except in one of seven studies with stripped female plasma (Kd = 3.3 x 10"9; binding capacity, 0.1 ju,g/100 ml). It should be noted that this binding capacity was approximately 3% of that found using T as the ligand with this plasma pool.
A typical analysis of E2 and T binding to TeBG at 37 C in SS-PAGE is shown in Fig. 2. It can be seen that the Kd at 37 C for E2 was indeterminate. The Kds for DHT and E2 at 0 and 37 C, as derived from Scatchard-type analysis, are shown in Table 3. Kinetic studies The ti/2s of dissociation of E2 and T from plasma, albumin, and DHT- or E2-saturated plasma are shown in Table 4. Both T and E2 dissociate from albumin more rapidly than from plasma (P < 0.01). The ti /2 of T, but not of E2, is shortened by saturating the plasma with DHT (Fig. 3). Saturation of plasma with E2 shortens the ti/2 of both T and E2 (Table 4). Saturation of albumin with E2 or DHT does not affect the ti/2 of either T or E2 (Fig. 4). The addition of 300 /Ag/100 ml unlabeled T or E2 TABLE 2. K^s of E2 in normal human plasma Kd x 1(T8 M Pool no. I
II T-2I°
A
T-37-
o
E2-4«
3
E.-2I-
a
III IV V VI
VII Mean ± SD 0 6
4C
22C
37 C
5.0 2.6 3.3 5.9 6.7 6.7 3.7
9.1 4.0 6.3 (-) (-) (—) (-)
ND" ND (-)" (-) (-) (-) (-)
4.8 ± 1.7
6.4 ± 2.6
ND, Not done. (—), Not determinable.
• DHT
13
° E2 11 9 B/F
\ 7 5 3 o 1
0 0
°
| O° 100
2
3 4 BOUND (xlCT 8 m/l)
FIG. 1. Scatchard plots of T and E2 with a pool of human plasma at 4, 22, and 37 C, as determined by equilibrium dialysis (see text for methodological details). Bound/free, The ratio of TeBG-bound to nonTeBG-bound steroid. Binding of E 2 was indeterminate at 37 C.
° °
.
1
200
300
\
400
BOUND (pg)
FIG. 2. Scatchard plot of DHT and E2 binding to TeBG, as determined by SS-PAGE. Aliquots (lO/il) of plasma were applied to gels containing 1-6 nM [ 3 H]DHT or 6-15 nM [3H]E2, and electrophoresis was performed at 4 or 37 C. Plasma was preincubated at 4 or 37 C with the corresponding amount of steroid.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
VIGERSKY ET AL.
902
JCE&M • 1979 Vol49 • No6 3
TABLE 3. Kds of DHT and E2 in SS-PAGE at 0 and 37 C
0C (M) 1.5 X 1(T9 3.3 X 10"10
DHT E2
37 C (M) 2.7 X 10"
40
' Not determinable.
20
TABLE 4. Half-times of dissociation (in minutes) of E2 and T from plasma, albumin, and E2-saturated or DHT-saturated plasma
E2 T
Plasma (n = 4)
Albumin (n = 7)
DHT-plasma
8.9 ± 1.4 10.8 ± 2.4
3.6 ± 0.8" 2.6 ± 1.2°
9.6 ± 3.0 3.5 ± 0.4"
3
H-Testosterone
100 80 iV 60 - V
E2-plasma (n = 4)
%Ba
10
H-Estradiol
- * - Albumin •-••• E2-Albumin - • - DHT-Albumin
1 *
—+- Albumin ••«--• E2 -Albumin -•-DHT-Albumin
" r i
5 \
\
4.9 ± 1.2" 2.4 ± 0.4° 1 i . i . i
Values given are the mean ± SD. a P < 0.01 compared to plasma.
2
6 10 14 18
I iI•I ,
25 30
2
6 10 14 18
25 30
MINUTES 3
40
%B a
vx
\ -
\ \ \
10
\
5
H-Estradiol -+— Plasma •••••• E2-Plasma - • - DHT-Plasma
Plasma E2-Plasma DHT-Plasma
100 80 60
20
3
H-Testosterone
> ' \ \ \ \ *\ • \
V
\
\
\ 1 I.I.I 2
, \ . I ,\
6 10 14 18
I 25
30
2
I i I • 6 10 14 18
I 25
30
MINUTES
FIG. 3. Dissociation of [3H]T and [3H]E2 at 37 C from untreated plasma or plasma preincubated with saturating amounts of DHT or E2 at 37 C for 18 h before analysis. Ba, Amount of steroid available for dissociation.
to [3H]T or [3H]E2 in 0.9% saline did not increase the rate of exit of labeled steroids from the dialysis bag, indicating that there was no binding to the dialysis tubing. Discussion These data suggest that TeBG is not an important binder of plasma E2 at physiological temperatures. Three independent in vitro methods yield the same conclusion. The method of dialysis used in these studies differs from most of those previously reported in that albumin has been included in the dialysate. However, Vermeulen and Verdonck (2) in examining TeBG binding properties and Slaunwhite (10) in examining the binding characteristics of transcortin used albumin in the dialysate. We have calculated the Kd in a manner analogous to theirs.
FIG. 4. Dissociation of [3H]T and [3H]E2 at 37 C from untreated human albumin or albumin preincubated with saturating amounts of DHT or E2 at 37 C for 18 h before analysis. Ba, Amount of steroid available for dissociation.
It was not necessary to precisely equalize the albumin concentrations of dialysate and dialysand, as shown by the failure to influence estradiol binding at 37 C using a wide range of albumin concentrations. This is not surprising given the essentially unlimited steroid-binding capacity of the albumin. The inclusion of albumin in the dialysate not only reduces the amount of steroid bound at low concentrations of ligand (as would be expected), but in the case of nonalbumin T binding, unmasks nonlinearity of the Scatchard plot. The latter may represent either heterogeneity of binding sites or negative cooperativity. With E2 no binding could be detected at 37 C that could not be attributed to albumin. This corroborates the findings of Bishoff and Stauffer (13), who found that at 37 C under conditions of nonequilibrium dialysis E2 binding to human serum could be attributed to albumin alone. Three previous studies have investigated E2 binding to TeBG at 37 C (7,14,15), but only Rosner and Smith (14) used equilibrium dialysis. They found that the association constant for E2 for purified TeBG was 6.0 ± 1.1 X 108 M" 1 at 37 C. This is not necessarily in conflict when the markedly different dialysis conditions (purified TeBG and no albumin in the dialysate) are considered. The dialysis system we have used was designed to simulate in vivo conditions. In this system, one might predict that most of the binding for E2 would be accounted for by albumin and most of the binding for T would be accounted for by TeBG since: 1) the binding affinity of E2 to TeBG is 1 order of magnitude less than that of T (Fig. 1); 2) the binding affinity of E2 for albumin is 1 order of magnitude greater than that for T (10); and 3) albumin is present in much greater concentrations than TeBG. Our data, performed under these simulated physiological
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
TeBG BINDING OF PLASMA E2
conditions, would tend to confirm this prediction. SS-PAGE permits the observation of the binding characteristics of the specific protein of interest when separated from other plasma proteins. The KdS for DHT in SS-PAGE at 0 and 37 C are within 1 order of magnitude of those determined by equilibrium dialysis of purified TeBG (14), suggesting that the Kd obtained by the method of equilibrium dialysis used here does not represent a composite of TeBG and other high affinity, nonTeBG proteins. These results also suggest that no significant damage occurs to TeBG during the 37 C electrophoresis. The kinetic data (Table 4) indicate that the DHTbinding site on TeBG is not (quantitatively) an important binder of plasma E2 at 37 C. This is seen in the experiments where the t1/2 of dissociation of E2 from plasma is not changed by DHT saturation of binding sites. The finding that E2 significantly shortens the ti/2 of E2 dissociation from plasma suggests that saturable, nonandrogen-binding sites do exist in plasma. In general, our data agree with those of Heyns and DeMoor (15), who, using a different experimental design, showed that the ti /2 s of dissociations of E2 from TeBG at 37 C were too rapid to be measured. Our findings do not agree with those of Wu et al. (16), who, using a 30-min incubation at 37 C and separating bound from free hormones with dextrancoated charcoal, found that E2 bound to TeBG with a Ka of 6.58 x 107 M""1 and that TeBG had a binding capacity for E2 of 16.6 ± 0.43 ng/ml. They also found that 38.4 ± 0.7% of E2 was bound to TeBG. It should be noted, however, that the 30-min incubation at 37 C followed an incubation performed at 4 C for 10 min. They found, nevertheless, that the percentage of free E2 determined by equilibrium dialysis was the same at 4, 22, and 7 C, suggesting that TeBG was not an important factor in determining the bound to free ratio. The observation that T but not E2 binds to TeBG at 37 C may shed light on the findings of several clinical studies. Galvao-Teles et al. (17) found that in patients with liver disease and elevated plasma TeBG levels, plasma free T was low while free E2 was normal. Olivo et al. (18) found no decrease in the MCR of E2 in a similar group of patients. Another disorder in which high plasma TeBG is found is hyperthyroidism. Chopra and Tulchinsky (19) observed in such patients that, while elevations in total T were associated with a normal free T level, elevations in total E2 were associated with elevated free E2 levels. Additionally, Ruder et al. (20) showed that in three normal male volunteers made hyperthyroid with T3, the associated rise in TeBG was accompanied by a fall in the MCR of T without a change in the MCR of E2. In hirsutism, a disorder associated with low TeBG levels, Fisher and Anderson (9) showed that the free T was higher than that of normal women, but free E2 was no
903
different. Thus, these data are in general accord with the proposition that TeBG binding of E2 is of little physiological significance in man. On the other hand, Ridgway et al. (21) have recently shown that the MCR of E2 is diminished in hyperthyroidism. The MCR of E2 returned to normal with resolution of the hyperthyroidism. The decreased MCR of E2 may be explained, in part, by the decrease in adipose tissue and muscle mass in hyperthyroidism. Muscle and fat are actively involved in E2 metabolism (21), and a decrease in the mass of these tissues may explain the decreased MCR of E2 in hyperthyroidism, and why it is about half that found in liver disease (18, 21). The lack of significant E2 binding to TeBG at 37 C may explain the coexistence of gynecomastia and normal or slightly elevated plasma E2 levels in several disorders associated with elevated TeBG, such as hyperthyroidism, cirrhosis, and Klinefelter's syndrome as well as in puberty and senescence. Free T is decreased while free E2 is unchanged. Conversely, in disorders associated with decreased TeBG levels, such as idiopathic hirsutism or hirsutism secondary to acromegaly or Cushing's syndrome, the free T is increased but free E2 is unchanged. Thus, a common denominator in these disorders may be an abnormal free E2 to free T ratio that results from the binding characteristics of these steroids to TeBG at physiological temperatures. References 1. Mercier-Bodard, C, A. Alfsen, and E. E. Baulieu, Sex-steroid binding plasma protein, Acta Endocrinol [Suppl] (Kbh) 147: 204, 1970. 2. Vermeulen, A., and L. Verdonck, Studies on the binding of testosterone to human plasma, Steroids 11: 609,1968. 3. Rosenbaum, W., N. P. Christy, and W. G. Kelly, Electrophoretic evidence for the presence of an estrogen binding ^-globulin in human plasma, J Clin Endocrinol Metab 26: 1399, 1966. 4. Murphy, B. E. P., Binding of testosterone and estradiol in plasma, CanJBiochem 46: 299, 1968. 5. Rosner, W., N. P. Christy, and W. G. Kelley, Partial purification and preliminary characterization of estrogen-binding globulin from human plasma, Biochemistry 8: 3100, 1969. 6. Pearlman, W. H., I. F. F. Fong, and J. H. Tou, A further study of a testosterone-binding component of human pregnancy plasma, J Biol Chem 244: 1373, 1969. 7. Burke, C. W., and D. C. Anderson, Sex-hormone-binding-globulin is an oestrogen amplifier, Nature 240: 38, 1972. 8. Tavernetti, R. R., W. Rosenbaum, W. G. Kelley, N. P. Christy, and M. S. Roginsky, Evidence for the presence in human plasma of an estrogen-binding factor other than albumin: abnormal binding of estradiol in men with hepatic cirrhosis, J Clin Endocrinol Metab 27: 920,1967. 9. Fisher, R. A., and D. C. Anderson, Simultaneous measurement of unbound testosterone and estradiol fractions in undiluted plasma at 37°C by steady-state gel filtration, Steroids 24: 809, 1974. 10. Slaunwhite, Jr., W. R., The binding of estrogens, androgens and progesterone by plasma proteins in vitro, In Antonaides, H. S. (ed.), Hormones in Human Plasma, Little, Brown, and Co., Boston, 1960, p. 478. 11. Ritzen, E. M., F. S. French, S. C. Weddington, S. N. Nayfeh, and V. Hansson, Steroid binding in polyacrylamide gels. Quantitation at
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
904
VIGERSKY ET AL.
steady state conditions, J Biol Chem 249: 6597, 1974. 12. Vigersky, R. A., D. L. Loriaux, S. S. Howards, G. B. Hodgen, M. B. Lipsett, and A. C. Chrambach, Androgen binding proteins of the testis, epididymis, and plasma in man and monkey, J Clin Invest 58: 1061, 1976. 13. Bishoff, F., and R. D. Stauffer, Orientation of circulating human estrogens by albumin, Am J Physiol 191: 313, 1957. 14. Rosner, W., and R. N. Smith, Isolation and characterization of the testosterone-estradiol binding globulin from human plasma. Use of a novel affinity column, Biochemistry 14: 4813, 1975. 15. Heyns, W., and P. DeMoor, Kinetics of dissociation of 17 /?-hydroxysteroids from the steroid binding /J-globulin of human plasma, J Clin Endocrinol Metab 32: 147, 1971. 16. Wu, C. H., T. Motohashi, H. A. Abdel-Rahman, G. Flickinger, and G. Mikhail, Free and protein-bound plasma estradiol-17/? during the menstrual cycle, J Clin Endocrinol Metab 43: 436, 1976.
JCE&M Vol49
i 1979
No 6
17. Galvao-Teles, A., D. C. Anderson, C. W. Burke, J. C. Marshall, C. S. Corker, R. L. Brown, and M. L. Clark, Biologically active androgens and oestradiol with chronic liver disease, Lancet 1: 173, 1973. 18. Olivo, J., G. G. Gordon, F. Rafil, and A. L. Southren, Metabolic clearance rates (MCR) of estradiol (E2) in cirrhosis and hyperthyroidism, Endocrinology (Suppl) 94,1974 (Abstract 282). 19. Chopra, J. J., and D. Tulchinsky, Status of estrogen-androgen balance in hyperthyroid men with Grave's disease, J Clin Endocrinol Metab 38: 269, 1974. 20. Ruder, H., P. Corvol, J. A. Mahoudeau, G. T. Ross, and M. B. Lipsett, Effects of iriduced hyperthyroidism on steroid metabolism in man, J Clin Endocrinol Metab 33: 382, 1971. 21. Ridgway, E., C. C. Longcope, and F. Maloof, Metabolic clearance and blood production of estradiol in hyperthyroidism, J Clin Endocrinol Metab 41: 491, 1975.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:54 For personal use only. No other uses without permission. . All rights reserved.
