Lymphocyte glucocorticoid receptor binding in depressed patients with hypercortisolemia

Psychoneuroendocrmology,Vol 10,No 4, pp 469-474, 1985 PrintedmGreatBritain

0306-4530/85 $3 00 + 0 00 © 1985PergamonPressLid

LYMPHOCYTE GLUCOCORTICOID RECEPTOR BINDING IN DEPRESSED PATIENTS WITH HYPERCORTISOLEMIA* JANET A . SCHLECHTE'~ a n d BARRY SHERMAN Climcal Research Center, and Department of Internal Medicine, University of Iowa, Iowa City, IA 52242, U.S.A.

(Recetved 8 August 1984; m final form II March 1985) SUMMARY Despite elevated levels of serum and urinary cortisol, patients with depressive illness manifest none of the clinical stigmata of glucocorticoid excess. This hypercorusolemia m the absence of clinical effects suggests a state of hormone resistance and could he mediated by alterations in the glucocorticoid receptor. Earlier studies have shown that small doses of glucocorticoids cause a decrease in glucocorticoid receptor binding m normal human lymphocytes. White cells from depressed patients with significant hypercortlsolemia would be expected to show a simdar change in receptor concentration if peripheral tissues are adequately exposed to and sensitive to the hormone. In this study we compared the binding of PH]dexamethasone to lymphocytes from normal subjects and depressed patients with hypercortisolemia. Lymphocytes from normal subjects had a mean receptor concentration of 10.2 _+0.66 fm/106 cells (S.E.M.) and a dissociation constant of 4.8 _+ 0.47 nM. Lymphocytes from depressed patients with abnormal 0800 h serum cortisoi after dexamethasone had a mean receptor concentration of 8.8 -+ 0.75 fm/10' cells, which was not significantly different from that in lymphocytes from normal subjects or from depressed subjects with normal post-dexamethasone cortisol levels (9.4 _+ 0.95 fm/106 cells). Lymphocytes from depressed patients with elevated urinary free cortisol excretion (UFC) also had normal receptor concentration and binding affinity for dexamethasone. The lack of a change in lymphocyte glucocorticoid receptor concentration m the presence of cortisol excess suggests the possibility that hypercortisolemia in depressive illness represents a state of peripheral glucocorticoid resistance. INTRODUCTION

FORTY tO fifty percent of patients with depressive illness have abnormalities of the pituitary adrenal axis, including elevated serum cortisol and failure to suppress serum cortisol after administration of dexamethasone (Bridges & Jones, 1966; Doig et al., 1966; Sachar et al., 1973; Schlesser et al., 1979). Some of these patients have morning cortisol values that are indistinguishable from those seen in Cushing's disease (Schlechte et al. unpublished observations), and they may have elevated urinary free cortisol excretion as well (Carroll et ai., 1976). Despite these abnormalities, patients with depression have few, if any, of the common stigmata of glucocorticoid excess. This lack of systemic effects in the face of hormone excess suggests a state of hormone resistance, as has been described for many peptide and steroid hormones (Flier et ai., 1975; Griffin et al., 1976). While

* This work was supported m part by Grant RR 59 from the General Climcal Research Centers, National Institutes of Health. t Correspondence to he addressed to: Janet A. Schlechte, M.D., Clinical Research Center, 157 MRF, Umversity of Iowa Hospitals, Iowa City, IA 52242, U.S.A. 469

470

JANET A. SCHLECHTE AND BARRY SHERMAN

complete resistance to cortisol would be lethal, partial resistance has been described (Chrousos et al., 1982). Resistance to hormone action can be mediated by changes in receptor number and function, and changes in circulating hormone concentrations are important modulators of receptor number and binding affinity (Milgrom et al., 1973; Bar et al., 1976). Normal subjects receiving small doses of glucocorticoids have a time-dependent decrease in lymphocyte glucocorticoid receptor concentration (Schlechte et al., 1982; Shipman et al., 1983). Patients with depression and significant hypercortisolemia would be expected to show the same type of change, unless peripheral tissues are resistant to or inadequately exposed to the hormone. In this study we compared the binding of [3H] dexamethasone to lymphocytes in normal subjects and in patients with depressive illness and hypercortisolemia. PATIENTS AND METHODS After obtaining informed consent, 18 normal subjects and 20 patients with depressive dlness were evaluated m the Psychiatric Hospital and in the Chnlcal Research Center at the Umversity of Iowa. The normal subjects ranged in age from 22 to 55 years, were taking no medications, and had no history of psychiatric &sease. The depressed patients were diagnosed according to DSM III criteria (American Psychiatric Association, 1980), and 85°70 were taking psychotroplc me&cations at the time of the study. The medications included haloperldol, a variety of tricychc antidepressants, and lithium These patients ranged m age from 21 to 63 years; half were over 50 years o f age. In order to determine which of the depressed patients had significant hypercort~solemm, we utd]zed the 1 mg dexamethasone suppression test (DST) and 24 h urinary free corUsol excretion. Fourteen patients took 1 mg of dexamethasone at 2300 h and had an 0800 h serum cortisol drawn the following day. (As the patients were admitted to regular psychiatric inpatient units, the DST was not performed on all). Those individuals with morning cortisol values greater than 5 Ixg/dl were considered abnormal (Carroll et al., 1980). Fifteen of the depressed patients had measurement of 24 h urinary free cortlsol excretion Normal values for 24 h urinary free cortlsol excretion in our laboratory are less than 50 ~tg/g creatinme All blood samples for receptor determinations were drawn at 0800 h. The depressed patients were studied one day prior to the dexamethasone suppression test. Lymphocyte binding assays were performed as previously described (Schlechte et al., 1982). Mononuclear cells from heparimzed blood were separated by Ficoll - H y p a q u e centnfugation, and lymphocytes were isolated after a 45 mln incubation m Eagle's Minimal Essential Medium (10070 fetal calf serum) m plastic tissue culture flasks at 37°C. Th~s incubation allows adequate time for dissociation of endogenous cort~sol from receptor sites. 2 × 10' lymphocytes/ml were incubated with 1 nM [3H]dexamethasone and increasing concentrations o f unlabeled dexamethasone at 24°C for 8 h. Receptor concentration and dissociation constants were determined by Scatchard analysis with computer-assisted linear regression. Groups were compared using Student's t-test; data are reported as mean _+ standard error. Serum corusol and urinary free cort~sol were measured by radioimmunoassay. We obtained [1,2(n) [3H]dexamethasone (20 Ci/mmol)] from A m e r s h a m , Arlington Heights, Illinois, and unlabeled dexamethasone from Sigma, St. Louis, Missouri. RESULTS

Figure 1 shows typical Scatchard plots of the binding of [3H]dexamethasone to lymphocytes from depressed and normal subjects. These curves represent the means of duplicate samples. The linear plots are consistent with a single class of receptors of uniform binding affinity. The slopes of the two curves do not differ, and, although the receptor concentration in the depressed subjects is slightly lower, it is not significantly different from the normal subjects. Lymphocytes from normal subjects had a mean receptor concentration of 10.2 _ 0.66 fm/10' cells and a mean dissociation constant of 4.8 ___0.47 nM. Glucocorticoid receptor

GLUCOCORTICOID RECEPTOR BINDING IN DEPRESSION

471

0 0035 0 0 0025

Normal epressed

B/F 00015

00005 25

50

75

I00

B o u n d ( F M / M d h o n Cells) FIG. 1. Scatchard plots of binding o f PH]dexamethasone to lymphocytes from normal and depressed subjects.

concentration in 20 depressed subjects ranged from 5.2 to 13.2 f m / 1 0 ' cells, and the mean receptor concentration was not significantly different from the normal subjects (Table 1, Fig. 2). Lymphocytes from depressed individuals taking medications had a mean receptor concentration (9.2 _ 0.42 fm/10 ~ cells) that did not differ from the mean receptor concentration in lymphocytes from unmedicated subjects (9.3 _ 1.5 f m / 1 0 ' cells). We subdivided the patients with depression by their response to 1 mg dexamethasone to see if receptor concentration and binding affinity were different in those patients with cortisol non-suppression vs those with suppression. Nine patients had postdexamethasone serum cortisol values that exceeded 5 lxg/dl and were classified as nonsuppressors. Mean 0800 h serum cortisol in this group was 8.1 __.0.9 ~tg/dl, and individual values ranged from 5 to 11.9 ~tg/dl. Lymphocytes from these patients with elevated 0800 h cortisol levels after dexamethasone had a mean receptor concentration of 8.80 _ 0.75 fm/10 ~ cells, which was not significantly different from those in the suppressors or the normal subjects (Table I, Fig. 2). The dissociation constants were similar in all groups (Table I).

TABLE I. CONCENTRATION AND BINDING AFFINITY OF THE LYMPHOCYTE GLUCOCORTICOID RECEPTOR IN NORMAL AND DEPRESSED SUBJECTS

n Normal Depressed Nonsuppressor § Suppressor Elevated U F C II

18 20 9 5 5

Ro* 10.2 9.2 8.8 9.4 9.9

± 0.66~ ± 0.47 ± 0.75 _+ 0.95 ± 1 06

* Receptor concentration fm/106 cells. t Dissociation constant (nM). ~: _ S.E.M. § Serum cortlsol > 5 lxg/dl after 1 mg dexamethasone. II greater than 50 g g / g r creatmine.

Kdt 4.8 4.4 4.4 4.1 4 8

± 0.47 ± 0 35 ± 0.43 __. 1.1 ± 0.49

472

JANET A SCHLECHTE AND BARRY SHERMAN

16-

O

O

,,, (.3 Z O

O

12

O O

O ,3 o

._J ._1

8

8

o O O

o

O

8 8

O

O

O

.0

4 I

I

I

I

I

NORMAL

DEPR

NSUPP

SUPP

UFC

FIG 2. Glucocorticoid receptor concentration m normal and depressed subjects. NSupp = 0800 h serum cortisol >5 Ixg/dl after 1 mg dexamethasone, Supp = 0800 h serum cortisol <5 ~tg/dl after 1 mg dexamethasone; UFC = 24-h urinary free cortlsol excretion >50 Ixg/g creatmine

In order to obtain another estimate of tissue exposure to cortisol, we classified depressed individuals according to 24 h urinary free cortisol excretion. Five patients had urinary free cortisol levels greater than 50 ~tg/gr creatinine; their levels ranged from 55 to 300 ~tg/g creatinine (mean: 91 _ 36 ~tg/g creatinine). These patients with depression and elevated urinary free cortisol excretion had a mean lymphocyte glucocorticoid receptor concentration that was not significantly different from those in the depressed subjects with normal urinary free cortisol excretion or the normal controls (Table I, Fig. 2). DISCUSSION

Regulation of hormone receptors by changes in endogenous hormone concentrations is well described for both peptide and steroid hormones (Bar et al., 1976; Milgrom et al., 1973). Administration of glucocorticoids to normal subjects results in a time-dependent decrease in lymphocyte glucocorticoid receptor concentration (Schlechte et al., 1982; Shipman et al., 1983). Svec & Rudis (1981) and Cidlowski & Cidlowski (1981) also have shown that incubation of pituitary AtT-20 cells and HeLa cells in wtro with glucocorticoids leads to diminution in the number of cellular glucocorticoid receptors. In light of these findings, patients with depression who have hypercortisolemia might be expected to have decreased lymphocyte glucocorticoid receptor binding. However, our patients with abnormal dexamethasone suppression and elevated urinary free cortisol excretion had normal glucocorticoid receptor concentration and normal binding affinity for dexamethasone. This apparent lack of receptor "down-regulation" may be a manifestation of peripheral tissue resistance to glucocorticoids, just as the failure to suppress serum corisol after dexamethasone is presumed to represent pituitary resistance to the hormone. While complete resistance to cortisol would be lethal, "partial resistance" has been described in one family (Chrousos et al., 1982). In this family, two individuals had elevated serum and urinary cortisol levels without clinical evidence of glucocorticoid excess. It is of interest that these hypercortisolemic individuals demonstrated an abnormality in the affinity of

GLUCOCORTICOID RECEPTOR BINDING IN DEPRESSION

473

the lymphocyte glucocorticoid receptor for dexamethasone but no decrease in glucocorticoid receptor concentration. There are other potential explanations for our findings. The lack of change in receptor concentration in the depressed patients with hypercortisolemia may be a manifestation of insufficient exposure to cortisol. In our previous work (Schlechte et al., 1982), normal subjects who received 37.5 mg cortisone acetate, 5 mg prednisone, or 1 mg dexamethasone daily for 1 week showed a 33% decrease in glucocorticoid receptor concentration. While these are low doses which approximate (hydrocortisone, prednisone) or slightly exceed (dexamethasone) physiologic replacement, the degree of cortisol elevation in the depressed subjects may not be equivalent to the glucocorticoid doses we used in the normal subjects. We also have found that patients with depression have only mild elevations of plasma free cortisol (Schlechte et al., unpublished observations), and others have shown that the hypercortisolemic phase of depressive illness often is limited (Greden et al., 1980). It is also possible that a whole-cell binding assay is not sufficiently sensitive to detect changes in glucocorticoid receptor binding in the presence of mild, intermittent elevations of cortisol. The degree of variability in the measurement of receptor concentration in whole-cell assays can make it difficult to demonstrate significant differences between patient groups (Fig. 2). Measurement of cytoplasmic receptors in partially purified cytosol preparations and the use of ligands with higher affinity and/or specific activity would facilitate the study of the effect of glucocorticoids on the glucocorticoid receptor in depressed individuals. Our results must also be interpreted in light of medication use by the majority of the depressed patients. There is little information regarding the effect of psychotropic medications on glucocorticoid receptor binding. Van Bohemen & Rousseau (1982) showed that micromolar concentrations of trifluoperazine could compete with dexamethasone for the glucocorticoid binding site in cytosol preparations of HTC cells. Whether a similar effect takes place in vivo at lower concentrations is not known. It is often difficult to study severely depressed patients in a medication-free state. The extrapolation of findings from the peripheral lymphocyte to other glucocorticoid target tissues also is problematic. While the glucocorticoid receptor has been shown to be similar in a variety of target tissues (Feldman et al., 1978), it is impossible to determine whether the changes we describe also reflect changes in the brain and pituitary, and our results cannot be generalized. However, a cellular defect that results in glucocorticoid insensitivity at critical loci in the central nervous system is an attractive explanation for increased pituitary- adrenal activity in depressive illness. This work was supported in part by grant RR 59 from the General Clinical Research Centers National Institutes of Health. We are grateful to the nurses in the Clinical Research Center for their assistance and to Alma Vernon for typing the manuscript. REFERENCES American Psychiatric Association (1980) Diagnostic and Statistical Manual o f Mental Disorders, Third Edition (DSM liD. American Psychiatric Association, Washington, DC. Bar R S, Gorden P, Kahn C R, DeMeyts P (1976) Fluctuation in the affinity and concentration of insulin receptors on circulating monocytes of obese patients. J Chn Invest 58:1123- 1135.

474

JANET A. SCHLECHTE AND BARRY SHERMAN

Bridges P K, Jones M T (1966) The dmrnal rhythm of plasma cortlsol concentration in depression Br J Psychtatry 112: 1 2 5 7 - 1261. Carroll B, Curtis G, Davies G, Mendels J, Sugerman A (1976) Urinary free corusol excretion m depression Psychol Med 6 : 4 3 - 50. Carroll B J, Feinberg M, Greden J F, James N M, Sterner M, T a n k a J (1980) Diagnosis of endogenous depression. J Affect Dlsord 2 : 1 7 7 - 194. Chrousos G, Vingerhoeds A, Brandon D, Ed C, Pugeat M, L o n a u x D, Llpsett M (1982) Primary cortlsol resistance in man" a glucocortlcold receptor-me&ated disease. J Chn Invest 69:1261 - 1269 CldlowskJ J A, Cidlowski N B (1981) Regulation of glucocortlcold receptors by glucocortlcolds m cultured HeLa $3 cells. Endocrinology 109:1975 - 1982 Dolg J F, M u m m e r y R V, Wills M R, Elkes A (1966) Plasma cortlsol levels m depression. Br J Psychtatry 112: 1263- 1267 Feldman D, Funder J, Loose D (1978) Is the glucocortlcold receptor identical m ~anous target organs 9 J SteroM Btochem 9:141 - 145. Fher J S, Kahn C R, Roth J, Bar R S (1975) Antibodies that impair m s u h n receptor binding m an unusual dmbet~c syndrome with severe m s u h n resistance. Scwnce 190: 6 3 - 65. Greden J F, Albala A A, Haskett R F, James N M, G o o d m a n L, Sterner M, Carroll B J (1980) Normahzatlon of dexamethasone suppression test: a laboratory index of recovery from endogenous depression. Btol Psychtatry 15:449 - 458. Griffin J E, Punyashthlti K, Wilson J D (1976) Dlhydrotestosterone binding by cultured h u m a n fibroblasts Comparison of cells from control subjects and from patients with hereditary male pseudohermaphrodlt~sm due to androgen resistance. J Chn Invest 57: 1 3 4 2 - 1351 Mdgrom E, Thl L, Atger M, Bauller E E (1973) M e c h a m s m s regulating the concentration and c o n f o r m a t m n of progesterone receptors in the uterus. J Btol Chem 248: 6 3 6 6 - 6374 Sachar E J, Hellman L, Roffwarg H P, Halpern F S, Fukushima D K, Gallagher T F (1973) Disrupted 24-hour patterns of cortisol secretion in psychotic depression. Arch Gen Psychtatry 28: 1 9 - 2 4 Schlechte J A, Ginsberg B H, Sherman B M (1982) Regulation of the glucocortlcold receptor m h u m a n lymphocytes J Sterotd Btochem 16: 6 9 - 74 Schlesser M, Winokur G, Sherman B (1979) Genetic subtypes of umpolar primary depresswe dlness dlstingmshed by h y p o t h a l a m l c - p i t u i t a r y - a d r e n a l axis activity. Lancet i: 739 741. Shipman G F, Bloomfield C D, GajI-Peczalska K J, Munch A U, Smith K A (1983) Glucocortlcolds and lymphocytes. Ill. Effects of glucocorticoid administration on lymphocyte glucocortlcold receptors. Blood 61: 1 0 8 6 - 1090. Svec F, Rudls M ( 1981) Glucocortlcolds regulate the glucocortlcold receptor in the AtT-20 cell. J Btol Chem 256: 5 9 8 4 - 5987. Van Bohemen C G, Rousseau G G (1982) C a l m o d u h n antagomsts competitively mhlblt dexamethasone b m d m g to the glucocorticoid receptor. FEBS Lett 143:21 - 25.