Interspecies variations of corticosteroid-binding globulin parameters

DOMESTIC ANIMAL E N D O C R I N O L O G Y Vol. 13(1):35-45, 1996

ELSEVIER

INTERSPECIES

VARIATIONS GLOBULIN

OF CORTICOSTEROID-BINDING PARAMETERS

V. Gayrard,* M. Alvinerie,** and P.L. Toutain* *Ecole Nationale Veterinaire de Toulouse Unite associre INRA de Physiopathologie et Toxicologie exp~rimentales 23 chemin des Capelles, 31 076 Toulouse, France and **INRA, Station de Pharmacologie 180 chemin de Tournefeuille, 31 300 Toulouse Received October 21, 1994

In mammalian plasma, cortisol binds to a specific oh-glycoprotein: corticosteroid-binding globulin (CBG). In this study, we measured the protein binding of cortisol by equilibrium dialysis in seven species in which plasma cortisol concentrations varied from 0.02 to 0.05 (ewe, dog, cow) to 0.1 to 0.6 (horse, human, cynomolgus monkey) to reach 1.6 IxM (squirrel monkey). No binding of cortisol to CBG was discernible in plasma from squirrel monkey. In all other species examined, we showed that the CBG maximal capacity (Bmax) was 3 (1.7 to 5.2) times more than the plasma cortisol levels, with cow, dog, ewe exhibiting the lowest and cynomolgus monkey exhibiting the highest values. We also noted the existence of a linear relationship between Bmax and the corresponding dissociation constant (Ka), Br, ax being systematically 10 (8.5 to 11.8) times more than K d. T h e low binding affinity of cortisol assigned to albumin did not differ between species. The free (6 to 14%), CBG-bound (67 to 87%), and albumin-bound (7 to 19%) cortisol fractions calculated from the estimated binding parameters and measured plasma cortisol concentrations were similar within species, except for squirrel monkey, in which half of the cortisol was albumin bound, and the other half remained protein free. Our most appealing finding was that in most species, as much as 68% of plasma CBG remained free of cortisol under physiologic conditions. These results are discussed with respect to the theories concerning the role of CBG in plasma transport and the local delivery of cortisol and free CBG as a proper hormone. INTRODUCTION Adrenal function has been described in a range of mammals, and interspecific variations in both the representativeness of one glucocorticoid (cortisol, corticosterone) and the magnitude of its plasma concentrations have been reported (1). In species in which cortisol is the major glucocorticoid, the plasma levels vary from 0.01 to 0.05 ktM in ewes (2), dogs (3,4), and cows (5) to 1.5 to 5.5 /.tM in New W o r l d primates (squirrel and marmoset monkey) (6-9), with intermediate 0.05 to 0.2 ktM levels in horses (10,11), 0.1 to 0.4/.tM levels in humans (12), and 0.5 to 0.7 ~tM levels in Old W o r l d primates (cynomolgus monkey, baboon) (6,8). The sensitivity or refractoriness of some species to cortisol could result from a control o f plasma concentrations of the biologically active form of cortisol and/or o f corticoid action at the level of target tissues. The currently accepted model of steroid action proposes that only the amount of a steroid that is unbound to plasma proteins can gain access to target tissues and bind to intracellular receptors (13). It is therefore important to evaluate the potential availability of cortisol to target cells, measured by the distribution © Elsevier Science Inc. 1996 655 Avenue of the Americas, New York, NY 10010

0739-7240/96/$15.00 SSDI 0739-7240(95)00042-D

36

GAYRARD ET AL.

of the hormone between protein-bound and protein-flee fractions. This depends on the properties of plasma cortisol-binding proteins, i.e., the high-affinity, low-capacity corticosteroid-binding globulin (CBG) and the low-affinity, high-capacity albumin. CBGbinding parameters, namely, CBG maximal capacity, Bmax (M), and its dissociation constant, Kj (M), have previously been measured in many species (14,15), including the domestic species considered in this article, by the use of a diversity of methods, which makes comparisons difficult. The purpose of this experiment was to evaluate cortisol binding to plasma proteins in the range of precited species by the use of one standardized method.

MATERIALS AND METHODS Animals and Plasma Sampling. Plasma samples from 21 male squirrel monkeys

(Saimiri sciureus) were kindly provided by Dr. B. Bonnemains of the Pasteur Institute Primate Center (Cayenne, Guyana). Three plasma samples from anonymous human donors were supplied by the Regional Center of Blood Transfusion (Purpan, Toulouse, France). It must be noted that human CBG binding capacity values were not found to differ between males and females or between the ages of 15 and 60 years (16). For the other species, blood was obtained by venipuncture from healthy, adult, nonpregnant animals accustomed to handling and sampling sessions. The experiment was conducted in June and July. Homogeneous groups (age, gender) of animals were constituted for all but two species (horses, cynomolgus monkeys). Six French Friesan Pie Noire cows, five Manech Noire and Basco-Bearnaise ewes, four male beagle dogs, four cynomolgus monkeys, and five saddlebred horses (two females and three males) were used. A preliminary study showed that the difference in estimated free cortisol concentrations between these horses was about 12% (unpublished observations). Blood was collected in heparinized tubes. A plasma pool was prepared for each species. Two 1.5-ml fractions of the pool were immediately frozen at - 2 0 ° C and kept at this temperature until the assay of plasma cortisol and albumin levels. Endogenous corticoids were removed from the remaining plasma pool by adsorption on charcoal (17). Charcoal (500 mg) was mixed with 10 ml of plasma for 1 hr at room temperature. The charcoal was then removed by centrifugation (4 × 10 min at 3,000g). The pool plasma samples, free of steroid, were rapidly frozen at - 2 0 ° C and kept at this temperature until the study of cortisol binding. Binding Experiments. The in vitro protein binding of cortisol was measured by equilibrium dialysis with a Dianorm system (CH8135; Langenau, Zurich, Switzerland). Dialysis was performed under constant stirring (20 rpm) at 37 ° C for 1 hr in teflon half-cells separated by a semipermeable cellulose membrane that retained compounds with molecular weights of more than 10,000 (Diachema 16-10; Braun Scientetec, ZA Courtab~euL Les Ulis, France). One compartment contained cortisol (hydrocortisone; Sigma, L'Isle d'Abeau Chesnes, La Verpillibre, France) solution in 0.1 M phosphate buffer (pH = 7.4; 0.9 ml), and the other contained plasma stripped of cortisol (0.9 ml). The binding of cortisol was studied over a wide range of concentrations from 0.0055 to 5.5 laM in triplicate, except for plasma from squirrel monkey, for which a single measure per level of cortisol concentration was obtained. Twenty microliters of a 1.5 nM tritium-labeled cortisol solution (1,2,6,7[3H]cortisol, 80 Ci/mmol; Amersham International, Buckinghamshire, England) in toluene:ethanol (9:1 v/v) was added to each l-ml cortisol solution as a tracer (10,000 dpm per cell). The radiochemical purity of the [3H]cortisol solution kept at - 2 0 ° C and controlled by high-performance liquid chromatography (HPLC) before use was more than 98%. After dialysis, 200- and 4004tl aliquots from plasma and buffer solution removed from the cells, respectively, were counted in a liquid scintillation spectrometer (Kontron Beta

INTERSPECIES PLASMA CORTISOL BINDING PARAMETERS

37

V, Montigny Le Bretonneux, France) after the addition of a liquid scintillation mixture (4.5 ml; Ready Safe; Beckma Instruments, Gagny, France). The counts per minute were converted to disintegrations per minute (dpm) by the use of the external standard ratio technique. The dpm counts were converted to equilibrium concentration of unbound (free) and bound cortisol by the use of an appropriate computation (18). The plasma cortisol concentration after equilibrium (PE) was calculated from the known concentration of cortisol (PT) added to the buffer solution and the radioactivity on each side of the membrane (Equation 1).

PE = PT ×

(dpm in dialyzed plasma) (dpm in buffer) + (dpm in dialyzed plasma)

(1)

Equations 2 and 3 were used to determine the molar concentrations of F (free cortisol) and B (bound cortisol). (dpm in buffer) F = PE × (dpm in dialyzed plasma)

(2)

B = PE -- F (3) Assays. Plasma levels of cortisol were determined after methylene chloride extraction (19) by HPLC. The HPLC (except for the column) was from Kontron (Montigny Le Bretonneux, France) and consisted of a 465 autoinjector (100-~tl sample loop), a 420 pump, a 432 UV detector (~, = 254 nm) with computer control by a 450 data system (version 2.00). The column (Nucleosil C18; 3 ~tm; 150 × 4.6 mm with a guard column) was from Interchim (Montluqon, France). The eluent was a mixture of water and methanol (45/55, v/v) and the flow rate was 0.6 ml/mn. The level of quantification of the assay was 5 ng/ml, and the intra-assay coefficient of variation was less than 5%. Proteins fractionated by electrophoresis on cellulose acetate in Veronal buffer (pH = 8.6, kt = 0.05 M(ionic force); Sebia, Issy-les Molineaux, France) were analyzed by dry chemistry procedures with a Kodak Ektachem XR700 (Kodak, Paris, France). Data Analysis. Protein-bound cortisol concentrations were plotted against unbound ones. For all of the species examined, except for squirrel monkey, the profile indicated the presence of both saturable and nonsaturable protein binding. These data were fitted by use of the Rosenthal relationship (20) (Equation 4). In the case of the squirrel monkey, only nonsaturable protein binding was observed. Hence, data were fitted according to Equation 5.

B=

Bmax × F x Ka + NS × F 1 +KaxF

B = NS x F

(4) (5)

where B and F are the molar concentrations of bound and free cortisol, respectively. Bmax (M) and Ka (M -j) are the CBG maximal capacity and the association constant of binding, respectively. Bmax is equivalent to the product of the number of binding sites of the specific binding protein (CBG) and its molar concentration, i.e., the concentration of the CBG-binding sites. NS is a dimension-less nonspecific binding constant of cortisol related to albumin. Initial binding parameters (B . . . . Ka, and NS), estimated by a Scatchard analysis, were

38

GAYRARD ET AL.

optimized by a computerized nonlinear least squares regression program (Micropharm, Version 1.6; INSERM, Paris, France). K d was then calculated as the inverse of Ka. The concentration of free cortisol was calculated from the quadratic equation derived by Tait and Burstein (21) (Equation 6). Ka x (1 + NS)

×

F 2 + (1 + NS + Ka

x Bma x -

Ka x TOT) x F - TOT = 0

(6)

where TOT is the concentration of total plasma cortisol. The albumin-bound cortisol concentration (BalD was calculated from Equation 7. Bal b =

NS x F

(7)

The concentration of CBG-bound cortisol (BcBG) was determined by the difference between total cortisol and the sum of free and albumin-bound cortisol concentrations. Statistical Analysis. The relationship between species plasma cortisol levels vs. CBG binding parameters and species CBG capacity vs. Ka of binding was analyzed by a linear least squares regression with a statistical program (Statgraphics, STSC, IMC, Rockville, MD). The square of the coefficient of correlation (R2), which indicates the percentage of variation explained by the model, was calculated. RESULTS

Figure 1 illustrates interspecific variations in measured cortisol plasma levels. Figure 2 shows the percentage of cortisol in plasma that was bound to plasma proteins as a function of total plasrha cortisol concentrations in the seven investigated species. In all species except squirrel monkeys, protein-binding profiles were biphasic. The first slope of the curves indicates the presence of the high-affinity binding sites of CBG; the second slope is the result of the nonsaturable cortisol binding to albumin. No specific binding of cortisol was discernible in squirrel monkey plasma, which showed only nonsaturable binding. Plasma protein-binding parameters for cortisol and plasma levels of cortisol and albumin are presented in Table 1. In all species examined except squirrel monkeys, physiologic plasma cortisol levels were linearly related to CBG-binding parameters, namely, CBG maximal capacity (B . . . . R 2 = 0.98) and its dissociation constant for cortisol ( K d, R 2 = 0.97) (Figure 3). Species

Squirrel Monkey CynomolgusMonkey Human Horse Ewe Dog Cow 0

0'.2

0'.4

0'.6

" //'

116

CONTROLPLASMACORTISOL(BM) Figure 1. Interspecific variations in cortisol plasma concentrations. Plasma cortisol levels were measured in the plasma pool of each species.

INTERSPECIES PLASMA CORTISOL BINDING PARAMETERS

39

95 z D 85 o

75 --...,.

,.,/ © 65' u

.~

z r..) 5 5 e~

_

--,

45" 35

0

1

2

3

4

5

CORTISOL CONCENTRATION (/,tM) Figure 2. Percentage of cortisol bound to plasma proteins (mean + standard deviation) as a function of hormone concentrations in seven species: horses (O), cows (*), ewes (O), dogs (O), humans ( I ) , cynomolgus monkeys ([]), and squirrel monkeys (A). The binding of cortisol was studied in triplicate for all plasma species except for plasma from squirrel monkey, in which a single measure per level of cortisol concentration was obtained. The bound fraction of cortisol decreases from 92 to 90% to 63 to 43% as the hormone concentration increases from 0.0055 to 1.38 I.tM (horses, dogs, ewes, and cows) or 2.76 ~M (cynomolgus monkeys and humans). Cortisol binding then remains almost constant as the hormone concentrations reach higher values. Percent bound cortisol does not vary with hormone concentrations in squirrel monkey plasma.

CBG capacity values were linearly related to corresponding K d values (R 2 = 0.99) (Figure 3). The low-affinity binding of cortisol assigned to albumin did not differ between species. Free cortisol, CBG-bound cortisol, and albumin-bound cortisol fractions were calculated from the CBG-binding parameters and measured plasma cortisol levels (Figure 4). CBG distribution between the cortisol-free and cortisol-bound CBG fractions was estimated in six species (Figure 5). In all species examined except squirrel monkeys, the major fraction of cortisol was bound to CBG (from 67.3 to 86.5%), whereas the free and albumin-bound cortisol concentrations represented 5.9 to 14.0% and 6.9 to 18.7% of the hormone plasma levels, respectively (Figure 4). Most of the CBG in these species re-

T A B L E 1. P L A S M A PROTEIN-BINDING PARAMETERS AND PLASMA LEVELS OF CORTISOL IN SEVEN SPECIES. a

CBG

Albumin

B ....

Ka

SPECIES

(~tM)

(~M)

Horse Ewe Dog Cow Cynomolgus monkey Human Squirrel monkey

0.22 0.078 0.082 0.11 1.18 0.38 ND b

0.019 0.0092 0.0079 0.013 0.10 0.039

NS

Plasma Albumin (~tM)

(~M)

Plasma Cortisol (~M)

1.23 1.34 1.06 0.63 0.88 1.21 1.06

543.9 636.4 540.9 598.5 763.6 580.3 601.1

442.2 474.9 510.3 880.1 867.7 479.6 375.7

0.13 0.050 0.023 0.021 0.56 0.14 1.6

Kd

a Bm,x and K a are the CBG maximal capacity and its dissociation constant. Bmax is equivalent to the product of the number of binding sites by the molar concentration of the protein, i.e., the concentration of the CBG-binding site concentration. NS is a dimension-less proportionality constant of the nonspecific binding of cortisol. Cortisol and albumin plasma concentrations were measured in each species plasma pool. The Ka of albumin is the dissociation constant of albumin calculated by the ratio between measured albumin plasma levels and estimated NS, assuming one site of cortisol binding per molecule of albumin. b ND, undetectable. No binding of cortisol to CBG was discernible in plasma from squirrel monkeys.

40

GAYRARD

ET AL.

A Cynomolgus Monkey ~' 1 0 0 -

<

5040-

u

20.

Human Horse Cow

10

A•Dog

<

• Ewe

i

n

u

i

i

2

5

10

20

5O

C O N T R O L P L A S M A C O R T I S O L ( 1 0 "2 IzM)

_ -~ ~

B Cynomolgus Monkey

IO.

Z

Human © Z O

F_<

Horse •

2- Cow •

Ewe • Dog

•

121 i

i

r

i

i

2

5

10

20

50

C O N T R O L P L A S M A C O R T 1 S O L ( 1 0 -2 paVl)

-c Cynomolgus Monkey I00 ~

<

5o

Horse

4020-

× <

10-

Human

Cow

Dog •

• • Ewe i

t

1

2 DISSOCIATION CONSTANT

u

4

n

i

5

10

(10 -2 I.tM)

Figure 3. Relationship between plasma cortisol concentrations and CBG B.... (A), between plasma cortisol concentrations and K a (B), and between the K d and CBG B..... (C). Plasma cortisol levels were measured in the plasma pool of each species. mained free o f cortisol (68.2%). In contrast, in squirrel monkey, half of the plasma cortisol was bound to albumin; the other half remained free. Figure 6 illustrates the relative variation in free and total cortisol concentrations in a horse. Free cortisol concentrations were calculated for various plasma cortisol concentrations ranging from 30 to 250% of C B G Bmax, including nycthemeral variations of plasma cortisol levels (36.4 to 77.3% of Bmax) (10) and values reached during a ride (77.3 to 250% of Bmax) (V. Lassourd et al., unpublished observations). Total and free cortisol concentrations were expressed as a percentage of the physiologic value, i.e., 0.015 and 0.13 ~tM for free and total hormone, respectively. The free cortisol concentration variations followed that of total ones under physiologic conditions but increased more than proportionally to total cortisol when cortisol plasma levels reached higher than rest values.

INTERSPECIES PLASMA CORTISOL BINDING PARAMETERS

41

100 80

7,

7~ F/

60

r. i.~

.~

i/

40"

m 20"

Horse (I4.9)

Ewe (7.0)

Dog (2.3)

Cow

Cynomolgus monkey Human

(2.3)

(60.8)

(15.3)

Squirrel monkey (776.7)

Figure 4. Percentage of free ([Z), albumin-bound ( I ) and CBG-bound (W) cortisol in the plasma (37 ° C) of the seven species. Percentages were calculated from the estimated plasma protein-binding parameters and measured plasma cortisol levels. Plasma cortisol levels were measured in the plasma pool of each species. Values in parentheses represent estimated free cortisol (nanomolar) concentrations.

DISCUSSION

This experiment showed, in a range of species, that physiologic plasma cortisol concentrations were related to CBG-binding parameters, namely, CBG Bma x and K d. Species cortisol plasma levels were within the range of physiologic concentrations: ewes (2), dogs (3,4), cows (5) squirrel monkeys (9), horses (10,11), humans (12), and cynomolgus monkeys (6,8), indicating the representativeness of measurements, from the plasma pool prepared for each species. In all of the investigated species except squirrel monkey, CBG Bma x w a s consistently 3 (1.6 to 5.2) times higher than measured plasma cortisol levels. With respect to diurnal variations of plasma cortisol levels described in most of the examined species: ewes (22), cows (23), horses (10,11), humans (12), and rhesus monkeys (24), CBG Bma x w a s about near to or higher than the upper range of cortisol plasma levels reached in the early hours of the morning. In addition, the CBG Kd for cortisol, i.e., the free cortisol concentration for which half of the CBG was saturated, was 10 (8.5 to 11.8)

1 -A

¢o

& o

Horse

Ewe

Dog

Cow

Cynomolgus Human monkey

Figure 5. Distribution of CBG-binding sites between cortisol-free ( i ) and cortisol-bound ([]) fractions in the plasma (37 ° C) of six species. Percentages were calculated from the estimated plasma CBG-binding parameters and measured plasma cortisol levels.

42

GAYRARD ET AL.

lOOO

80o 600 ~ 8

400

N

2oo o 30

50

100

200

Relative cortisol concentrations (percent) Figure 6. Relative total (13) and free (*) cortisol plasma concentrations in a horse for total cortisol concentrations ranging from 30 to 250% of CBG Bmax. Free cortisol plasma concentrations were calculated from the estimated individual plasma protein-binding parameters and measured plasma cortisol levels. Total and free cortisol concentrations are expressed as percentages of change of the respective measured and calculated concentrations. The shaded area represents nycthemeral variations in plasma cortisol levels (0.08 to 0.17 ~tM) (10). The hatched area represents the values of plasma cortisol levels reached during a ride (Lassourd et al., unpublished observations).

times lower than the CBG Bmax. In contrast, CBG binding was not detected in squirrel monkeys despite their very high plasma cortisol levels. The low-affinity binding of cortisol assigned to albumin was not related to species physiologic plasma cortisol levels. Our results can be compared with those previously obtained for the same species using different techniques. In this experiment, CBG Bin, x values for horse, cow, and dog were similar to those previously estimated with prednisolone, a synthetic analogue of cortisol, unique in that it binds to CBG (17). Similarly, our estimations of CBG B . . . . nonspecific binding constant values, or free fraction of cortisol are in good agreement with those determined by others: horse (25), ewe (26), dog (4,27), human (28,29), and cynomolgus monkey (8). The absence of discernible CBG-binding activity in plasma from squirrel monkey is not surprising with regard to earlier reports. Indeed, CBG-binding capacities were lower in squirrel monkeys than in Old World primate species, being less than 0.2 gM (8). Furthermore, unlike the other species, a broad spectrum of CBG-binding capacities was observed, ranging from 20 to 4,700 nM in 35 separate pools of squirrel monkey plasma (13). Absolute values of protein-binding parameters determined from charcoaltreated plasma must be noted with caution because charcoal could have removed fatty acid-induced changes in protein conformation (30). To our knowledge, an inverse relationship between CBG Bin, x and CBG affinity has never been reported. Unfortunately, we could give no satisfactory biologic explanation for such a relationship. An artefactual origin of this relationship as the result of the structure of the equation, which is a Michaelis Menten type equation, cannot be excluded (see Ref. 31). Whatever the explanation, our estimated parameters adequately fitted the data and were appropriate to simulate the distribution of cortisol between free and protein-bound fractions. Using CBG parameters and measured physiologic plasma cortisol levels, we have shown that the major fraction of cortisol is bound to CBG (76.5%), tree and albuminbound fractions being minor in all of the investigated species, except squirrel monkey. Indeed in this specie, half of the plasma cortisol was bound to albumin, whereas the other half remained free. There thus appeared to be 50 times more free cortisol in the plasma

INTERSPECIES PLASMA CORTISOL BINDING PARAMETERS

43

of squirrel monkey than in human plasma, whereas the ratio between total cortisol concentrations was 11.4. Consequently, the magnitude of interspecific variations in free cortisol concentrations closely follows that of measured total hormone plasma levels in most species. In contrast, the difference in adrenal function between squirrel monkey and the other examined species is amplified when free cortisol concentrations are considered. Interspecific variations in free cortisol concentrations greatly exceeded the previously observed interindividual variations in horses (12%; unpublished observations), except in dogs and cows, in which both the similar low cortisol plasma levels and the CBG Bmax resulted in the same free cortisol concentrations. The percentage of free cortisol in plasma was assumed to be underestimated as the result of the possible interference of endogenous CBG ligands, such as corticosterone and progesterone, with cortisol binding. CBG was not detected in plasma from the species exhibiting the highest plasma cortisol levels, and in other species, most of the CBG remained in the cortisol-free state (68%). These observations raise several questions with regard to the physiologic significance of this high-affinity binding protein. First, it is clear that the presence in plasma of a highaffinity binding protein does not lead to any standardization of the active (free) cortisol concentrations within species. This suggests that the sensitivity or refractoriness of species to cortisol action is more likely the result of a regulatory mechanism exerted at the level of target tissues (e.g., via interspecific variations in glucocorticoid receptor affinity or the cellular metabolism of cortisol [32]) than of the protein transport system. Indeed, some, although not all, in vitro studies showed a marked decrease of the affinity of dexamethasone for the glucocorticoid receptor in squirrel monkey compared with Old World primate species (6). According to those authors, a resistance at the level of the glucocorticoid receptor led to a compensatory increase in cortisol secretion in response to the hypothalamic-pituitary-adrenal feedback axis. Second, because the species' CBG B~a × was higher than the physiologic plasma cortisol levels in most species, the CBG was not saturated by physiologic plasma cortisol fluctuations. Under physiologic conditions, free cortisol concentrations increase almost proportionally to total cortisol concentrations. In contrast, in emergency situations, because the total plasma cortisol levels are near to or higher than the CBG Bma×, increases in total cortisol may cause a more than proportional rise in free cortisol. In this view, the saturability of CBG could represent a mechanism to amplify the adrenal response in emergency situations. Finally, because cortisol is quantitatively the most important CBG ligand (the other endogenous steroid amounts bound to CBG being relatively low when compared with glucocorticoids, as demonstrated in humans [33]), major CBG-binding sites may remain unoccupied under physiologic conditions. This observation does not argue in favor of a single steroid carrier protein role for CBG. Indeed, previous reports showed specific interactions of human CBG with membranes prepared from human prostate tissue (34), human liver, decidual endometrium, and placental syncytiotrophoblast (35). There has been a parallelism in recent developments between CBG and SHBG (sex hormonebinding globulin): the moiety that binds to the specific receptor could be unliganded CBG (36), as demonstrated for SHBG (37). Powerfully bound steroids prevent CBG from binding to its receptor, but subsequently, the binding of steroid to preformed CBGreceptor complexes is required to induce the accumulation of cyclic AMP (36). The involvement of this interaction in the facilitation of cortisol uptake, suggested by some authors, has not been demonstrated. The activation of a second messenger system rather suggests that free CBG could act as a hormone ligand for cell surface-binding sites in cortisol target tissues and that the meaning of interspecific differences in CBG-binding parameters must also be investigated in terms of plasma free CBG concentrations. Indeed,

44

GAYRARD ET AL.

in squirrel m o n k e y , the h i g h levels o f c o r t i s o l m a y result in the saturation o f C B G - b i n d i n g sites and the s u b s e q u e n t i n h i b i t i o n o f C B G - s p e c i f i c r e c e p t o r interaction. In c o n c l u s i o n , w e h a v e d e m o n s t r a t e d that p l a s m a C B G d o e s not lead to a s t a n d a r d i z a tion o f free cortisol c o n c e n t r a t i o n s across s p e c i e s and that in a r a n g e o f species, m o s t p l a s m a C B G r e m a i n e d in the c o r t i s o l - f r e e state, w h e r e a s the m a j o r cortisol f r a c t i o n b o u n d to C B G . T h e s e results s u p p o r t the attractive h y p o t h e s i s o f others that C B G c o u l d act as a p r o p e r h o r m o n e a n d that the m e a n i n g o f i n t e r s p e c i f i c d i f f e r e n c e s in C B G - b i n d i n g c h a r a c t e r i s t i c s m u s t also b e s o u g h t in t e r m s o f free C B G c o n c e n t r a t i o n s rather t h a n in the c o n c e n t r a t i o n s o f free c o r t i s o l alone. ACKNOWLEDGMENTS/FOOTNOTES The authors thank J.F, Sutra and V. Laroute for expert assistance. We also thank Dr. B. Bonnemains for kindly supplying samples of squirrel monkey plasma and J. Barr6 for critical reading of the manuscript. Author for correspondence: P.L. Toutain, Ecole Nationale V6t6rinaire, 23 chemin des Capelles, 31076 Toulouse, France.

REFERENCES 1. Bush IF. Species differences in adrenocortical secretion. J Endocrinol 9:95-100, 1953. 2. Fulkerson WJ, Jamieson PA. Pattern of cortisol release in sheep following administration of synthetic ACTH or imposition of various stressor agents. Aust J Biol Sci 35:215--222, 1982. 3. Thun R, Eggenberger E, Zerobin Z. 24-hour profiles of plasma cortisol and testosterone in the male dog: Absence of circadian rhythmicity, seasonal influence and hormonal interrelationships. Reprod Dom Anim 25:68-77, 1990. 4. Bamberg-Thal6n B, Nyberg L, Fackler L, Edqvist LE. Cortisol binding capacity of corticosteroid binding globulin in hyperadrenocorticoid and healthy dogs. Res Vet Sci 52:363-366, 1992. 5. Verkerk GA, Macmillan KL, Mcleay LM. Adrenal cortex response to adrenocorticotropic hormone in dairy cattle. Domest Anita Endocrinol 11 : 115-123, 1994. 6. Chrousos GP, Renquist D, Brandon D, Eil C, Pugeat M, Vigersky R, Cutler JR, Loriaux DL, Lipsett MB. Glucocorticoid hormone resistance during primate evolution: Receptor-mediated mechanisms. Proc Natl Acad Sci USA 79:2036-2040,1982. 7. Chrousos GP, Loriaux DL, Brandon D, Tomita M, Vingerhoeds ACM, Merriam GR, Johnson EO, Lipsett MB. Primary cortisol resistance: A familial syndrome and an animal model. J Steroid Biochem 19:567-575, 1983. 8. Klosterman LL, Murai JT, Siiteri PK. Cortisol levels, binding, and properties of corticosteroid-binding globulin in the serum of primates. Endocrinology 118:424-434, 1986. 9. Champoux M, Zanker D, Levine S. Food search demand effort effects on behavior and cortisol in adult female squirrel monkeys. Physiol Behav 54:1091-1097, 1993. 10. Toutain PL, Oukessou M, Autefage A, Alvinerie M. Diurnal and episodic variations of plasma hydrocortisone concentrations in horses. Domest Anim Endocrinol 5:55 59, 1988. 11. Irvine CHG, Alexander SL. Factors affecting the circadian rhythm in plasma cortisol concentrations in the horse. Domest Anita Endocrinol 11:227-238, 1994. 12. Follenius M, Brandenberger G, Hietter B. Diurnal cortisol peaks and their relationship to meals. J Clin Endocrinol Metab 55:757-761, 1982. 13. Siiteri PK, Murai JT, Hammond GL, Nisker JA, Raymoure WJ, Kuhn RW. The serum transport of steroid hormones. Recent Prog Horm Res 38:457-510, 1982. 14. Seal US, Doe RP. Corticosteroid-binding globulin: species distribution and small-scale purification. Endocrinology 73:371-376, 1963. 15. Seal US, Doe RP. Corticosteroid-binding globulin: Biochemistry, physiology and phylogeny. In: Steroid Dynamics, Pincus G, Nakao T, and Tait JF (eds). Academic Press, New York, p. 63-90, 1965. 16. Westphal. Corticosteroid-binding globulin (CBG) of man and other species. In: Steroid-Protein Interactions, Gross F, Labhart A, Mann T, Samuels LT, and Zander J (eds). Springer-Verlag, New York, p. 164-225, 1971. 17. Alvinerie M, Houin G, Toutain PL. Prednisolone binding to plasma proteins in domestic species. J Pharm Sci 77:937-938, 1988. 18. Rose JQ, Yurchak AM, Jusko WJ. Dose dependent pharnlacokinetics of prednisone and prednisolone in man. J Pharmacokinet Biopharm 9:389-417, 1981.

INTERSPECIES PLASMA CORTISOL BINDING PARAMETERS

45

19. Alvinerie M, Toutain PL. Simultaneous determination of corticosterone, hydrocortisone, dexamethasone in dog plasma using high performance liquid chromatography. J Pharm Sci 71:816-818, 1982. 20. Rosenthal HE. A graphic method for the determination and presentation of binding parameters in a complex system. Anal Biochem 73:371-376, 1967. 21. Tait JF, Burstein S. In vivo studies of steroid dynamics in man. In: The Hormones, Pincus G, Thimann KV, Astwood EB (eds). Academic Press, New York, Vol 5, p. 441-557, 1964. 22. McNatty KP, Cashmore M, Young A. Diurnal variations in plasma cortisol levels in sheep. J Endocrinol 54:361-362, 1972. 23. Wagner WC, Oxenreider SL. Adrenal function in the cow. Diurnal changes and the effects of lactation and neurohypophyseal hormones. J Anim Sci 34:63ff4535, 1972. 24. Quabbe HJ, Gregor M, Bumke-Vogt C, Hiirdel C. Pattern of plasma cortisol during the 24-hour sleep/wake cycle in the rhesus monkey. Endocrinology 110:1641-1646, 1982. 25. Irvine CHG, Alexander SL. Measurement of free cortisol and the capacity and association constant of cortisol-binding proteins in plasma of foals and adult horses. J Reprod Fertil [Suppl] 35:19-24, 1987. 26. Paterson JYF, Hills F. The binding of cortisol by ovine plasma proteins. J Endocrinol 37:261-268, 1967. 27. Meyer HP, Rothuizen J. Determination of the percentage of free cortisol in plasma in the dog by ultrafiltration/dialysis. Domest Anim Endocrinol 10:45-53, 1993. 28. Robinson PA, Hawkey C, Hammond GL. A phylogenetic study of the structural and functional characteristics of corticosteroid binding globulin in primates. J Endocrinol 104:251-257, 1985. 29. Schlechte JA, Hamilton D. The effect of glucocorticoids on corticosteroid binding globulin. Clin Endocrinol 27:197-203, 1987. 30. Haourigui M, Martin ME, Thobie N, Benassayag C, Nunez EA. Stimulation of the binding properties of adult rat corticosteroid-binding globulin by a lipolysis-induced rise in plasma free fatty acids. Endocrinology 133:183-191, 1993. 31. Seakins JWT, Hjelm M, Wang KJ. Statistical properties of the Michaelis-Menten equation and their implication for inborn errors of metabolism. J Inherited Metab Dis 15:1-4, 1992. 32. Abel SM, Back DJ, Maggs JL, Park BK. Cortisol Metabolism In vitro-II. Species difference. J Steroid Biochem Mol Biol 45:445-453, 1993. 33. Dunn JF, Nisula BC, Rodbard D. Transport of steroid hormones: Binding of 21 endogenous steroids to both testosterone-binding globulin and corticosteroid-binding globulin in human plasma. J Clin Endocrinol Metab 53:58-68, 1981. 34. Rosner W. Plasma steroid-binding proteins. Endocrinol Metab Clin North Am 20:697-720, 1991. 35. Strel'chyonok OA, Avvakumov GV. Interaction of human CBG with cell membranes. J Steroid Biochem Mol Biol 40:795-803, 1991. 36. Maitra US, Khan MS, Rosner W. Corticosteroid-binding globulin receptor of the rat hepatic membrane-solubilization, partial characterization, and the effect of steroids on binding. Endocrinology 133:1817-1822, 1993. 37. Rosner W. The functions of corticosteroid-binding globulin and sex hormone-binding globulin: Recent advances. Endocr Rev 11:80-91, 1990.