Glucocorticoid Receptor Sites Research Articles

Glucocorticoid receptors and corticosensitivity of human thymocytes at discrete stages of intrathymic differentiation.

Human thymus is composed of several discrete compartments. Stage III thymocytes, located mainly in the medulla, stain brightly with anti-T3 monoclonal antibody; stage II thymocytes, located in the cortex, are T3- but react with T6 antibodies. The earliest identifiable intrathymic cell (stage I) expresses the sheep erythrocyte glycoprotein T11 but not T6 or T3 antigens. Within the thymus a phenotypically heterogeneous pool of proliferating lymphoblasts is present. This capacity to proliferate without in vitro activation is mainly attributable to thymocytes unable to respond to mitogens and expressing the cortical T6 marker. Both T3+ and T3-T6- cells respond to mitogen. However, in order to exhibit maximal proliferative responses, T3+ but not T3-T6- thymocytes require the addition of exogenous IL 2. Thymocyte subsets at distinct stages of intrathymic differentiation were then analyzed for glucocorticoid (GC) receptor content by using a whole cell assay with 3H-triamcinolone acetonide as tracer. The least mature T3-T6- thymocyte subset contained the highest levels of GC receptors . T3+ thymocytes exhibited a receptor content higher than that found in T6+ cells and similar to that reported for peripheral blood lymphocytes. Apart from the number, the GC receptor sites in all thymocyte subsets were similar in their affinities, kinetic characteristics, specificity for steroids, and ability to undergo translocation from cytoplasm to nucleus, and they behave in all these respects like binding sites of GC receptors in lymphoid and other cells. Independently of both phenotype and GC receptor content, all in vivo activated thymocytes (i.e., spontaneously proliferating cells) were similarly sensitive to the steroid inhibitory action in vitro. Both in the presence and in the absence of exogenous IL 1 or IL 2, the PHA-induced mitogenesis of T3-T6- cells was less inhibited by GC than that of T3+ thymocytes. Exogenous IL 1 and IL 2 were equally effective in removing, although not completely, the GC inhibition on T3-T6- proliferative responses to PHA. Relative to T3+ cell mitogenesis, only exogenous IL 2 was able to antagonize the steroid inhibitory action. The capacity observed in vitro of GC to differentially affect the proliferative potential or the cell viability of thymocytes belonging to functionally distinct subsets suggests that these hormones could regulate the intrathymic maturative pathways. Finally, although at present the physiologic relevance of the highest expression of GC receptors in intrathymic precursor cells remains unclear, the receptor density may be considered a marker of differentiation for the T lymphoid lineage.

Feedback action and tonic influence of corticosteroids on brain function: A concept arising from the heterogeneity of brain receptor systems

Two types of corticosteroid receptors can be distinguished in rat brain. The type 1 receptor resembles the kidney mineralocorticoid receptor and has two functional expressions in brain, i.e. type 1 corticosterone (CORT) preferring sites (CR) and mineralocorticoid receptors (MR). The type 2 receptor is similar to the liver glucocorticoid receptor (GR). CORT binds to both CR and GR. The localization, binding specificity, and capacity of the receptor systems have served as criteria to evaluate steroid dependent events in brain biochemistry and behaviour. The GR is widely distributed in neurons and glial cells, with the highest density in frontal brain regions. The GR becomes occupied concomitant with rising plasma CORT levels after stress and as part of the circadian rhythm. The GR mediates the feedback action of CORT on stress-activated brain processes. The CR has its predominant localization in neurons of the septo-hippocampal complex and has a ten-fold higher affinity for CORT than that of the GR. The CR is, at all times of intact adrenocortical secretion, 90% or more occupied by endogenous hormone. The CR mediates a tonic influence exerted with stringent specificity by CORT on hippocampus-associated functions, e.g. cognition, mood, and affect. CORT, via the CR, thus contributes to hippocampus function in interpretation of sensory information, leading to appropriate neuroendocrine and behavioural responses, which are themselves subsequently subject to feedback action via the GR. The MR mediates the mineralocorticoid effect on salt and water balance and its behavioural corollary of salt appetite. The anatomical localization of the MR system is as yet ill-defined, although functional studies suggest circumventricular organs as mineralocorticoid target sites. The CR and the MR have in common the high affinity for mineralocorticoids, but the CR is defined by its exclusive responsiveness to CORT as its agonist. The CR and MR probably represent the same chemical receptor modality (type 1), which is expressed differentially depending on the presence of extravascular corticosteroid binding globulin (CBG) in the vicinity of the receptor. GR capacity is subject to autoregulation. Chronic stress, senescence, and chronic CORT administration reduce GR number, with, as a consequence, a less efficient feedback signal. The CR number seems not to be under the control of corticosteroids, probably since the receptor sites are extensively occupied by endogenous hormones. The CR number displays a circadian rhythm and is reduced during senescence. The reduced number of CR sites, but not of GR sites, is restored to the same number as in the hippocampus of young control animals after chronic treatment with the behaviourally active ACTH(4–9) analogue, ORG 2766. CR capacity thus seems to be regulated by neural factors.

Glucocorticoid receptors and corticosensitivity in the human monocytic cell line CM-S: Changes during phorbol ester-induced differentiation

CM-S is a line of human precursor mononuclear phagocytes inducible to macrophage differentiation in response to the tumor promoter phorbol ester 12- O-tetradecanoylphorbol-13-acetate (TPA). Untreated CM-S cells expressed a single class of high-affinity ( K D = 4.5 ± 2.3 n M) glucocorticoid receptor sites (GCr) (27,530 ± 3752 sites/cell) as measured by a whole-cell assay at 37 °C using [ 3H]triamcinolone acetonide as tracer, while CM-S cells induced to macrophage differentiation by 10 −7 M TPA showed reduced levels of GCr (10,729 ± 2135 sites/cell). Kinetic studies indicated that this reduction was progressive, reaching about 34% of the original value 96 hr after TPA addition. The GCr in untreated and TPA-induced cells were similar in their specificity for corticosteroids. In the precursor cell population dexamethasone enhanced both the rate of protein synthesis and the production of autostimulatory growth factor(s), while in TPA-induced cells it inhibited the rate of protein synthesis in a dose-dependent manner. Our data with the CM-S cell line thus suggest that in the monocytic cell line lineage both the number of GCr and the cell response to glucocorticoids depend on the degree of cell maturation.

The mature, mitogen-responsive, human thymocyte population comprises two functionally distinct T-cell subsets differing in exogenous interleukin requirements for mitogenesis and in sensitivity to glucocorticoids.

Human thymocytes were separated into 10 fractions by continuous Percoll density gradient centrifugation. Almost all the cells able to proliferate in response to phytohaemagglutinin (PHA) and about 80% of the thymocytes carrying the mature T3 marker were contained within the first two lightest fractions (Fr1 and Fr2). However, cells in Fr1 and in Fr2 were clearly different in terms of their interleukin requirement for mitogenesis, i.e. in the presence of interleukin-1 (IL-1), Fr1 thymocytes showed proliferative responses to PHA that were three to five times greater than those of the Fr2 cells, whereas in the presence of exogenous IL-2 both Fr1 and Fr2 cells had mitogenic responses of a similar magnitude. These differences could not be explained by accessory cell contamination or by different kinetics of the proliferative responses. Furthermore, in the presence of exogenous IL-1, Fr1 thymocytes showed a greater capacity for producing IL-2. In contrast, there was no difference in the expression of functional IL-2 receptors between the two fractions. It can be concluded that the mature thymocyte subpopulation contains at least two functionally distinct subsets, which differ in density, and that during intrathymic maturation the capacities to produce and to respond to IL-2 do not develop simultaneously. Although both the number and the binding characteristics of glucocorticoid (GC) receptor sites did not differ significantly between Fr1 and Fr2 thymocytes, dexamethasone was more effective in inhibiting the PHA-induced mitogenesis of Fr2 than of Fr1 cells. Thus, the GC sensitivity of T-cell mitogenesis is not directly correlated with receptor-related variables. In contrast with observations made on peripheral T cells, the inhibitory activity of GC on thymocyte mitogenesis was only partially reversed by the addition of exogenous IL-2. Thus, the inhibitory action of GC on thymocyte mitogenesis cannot be explained only by the steroid-induced inhibition of IL-1--IL-2 production; it must also be due to a GC-inhibitory effect on the IL-2-responsive cell proliferation.

Glucocorticoid receptors and depression.

The number of glucocorticoid receptor sites in lymphocytes was estimated and plasma cortisol concentrations measured in 17 depressed patients, 12 patients with chronic schizophrenia, and 31 healthy control subjects. The number of receptor sites was significantly lower in the depressed patients than in either the controls or the patients with chronic schizophrenia, but there were no differences between the groups in the dissociation constants of the glucocorticoid receptors or the plasma cortisol concentrations. When two control subjects were studied intensively over 28 hours a slight diurnal variation in the number of glucocorticoid receptors was detected. The lower numbers of glucocorticoid receptors in the lymphocytes of depressed patients may explain why such patients, who often have hypercortisolaemia, do not show the clinical features of excess production of cortisol.

Concentration of glucocorticoid receptor sites in normal human lymphocytes.

We used a whole-cell glucocorticoid receptor assay to examine characteristics of the glucocorticoid receptor in the lymphocytes of normal human donors. We measured binding of [3H]dexamethasone to the lymphocytes of four different donors on several different occasions; the variation about the mean for the assays was +/- 15%. Whole-cell assays in 15 normal subjects showed a mean value of 6.18 fmol/10(6) cells or 3722 sites per cell, with a somewhat higher level in men (7.67 fmol/10(6) cells, or 4620 sites per cell) than women (4.48 fmol/10(6) cells, or 2698 sites per cell). We saw no correlation between donor age and receptor values, in either group. Assays in which we used [3H]prednisolone demonstrated similar binding properties as with [3H]dexamethasone. The mean glucocorticoid receptor value for normal human T-cells from three donors was 2.52 fmol/10(6) cells, or 1518 sites per cell.

Glucocorticoid receptors and glucocorticoid resistance in human leukemia in vivo and in vitro.

Clinical measurements quantitating glucocorticoid receptor sites in leukemic blasts may give useful prognostic information. In childhood ALL, where the most data is available, "high" receptor content in peripheral or marrow blasts correlated with likelihood of remission on therapy, longer durations of remission and better prognosis generally. In lymphomas and CLL as well, high receptor content correlated with likelihood of response to steroid therapy, though the number of studies is less. In AML the correlation with receptor site content is moot, and in other leukemias the reports are less complete. A model system for childhood ALL is provided by CEM cells, a glucocorticoid sensitive human cell line from a patient with the disease. These cells have glucocorticoid receptors which must be filled by hormone for greater than 24 hr for cell lysis to begin. Four types of glucocorticoid resistance have been identified thus far in clones of these cells. Their distinctive properties are described and their relevance to clinical situations briefly discussed.

Perineal muscles: possible androgen regulation of glucocorticoid receptor sites in the rat.

The glucocorticoid receptor population of the androgen-dependent levator ani-bulbo cavernosus muscle complex in response to both androgen withdrawal and androgen replacement was examined in rats. After castration the concentration of receptor sites more than doubled. The increase in the sites began within the first 24 h after castration and reached a plateau level within 7 days. Parallel dry weight studies demonstrated that no loss in muscle mass is observed until 48 h after castration. In a second series of experiments, both the regrowth of the muscle and the number of receptor sites in response to androgen injections were examined. By day 11 of injection, the muscle had increased in dry weight to a point where it was equal to the size of this muscle in normal intact male controls. During this process, the concentration of receptor sites began to decline 5 days after the beginning of injections and reached control or below control values by 11 days.

Paradoxical involvement of glucocorticoid receptors in the aldosterone-induced impairment of acth secretion from perifused pituitary glands

The present investigation was aimed at examining whether interaction of aldosterone with specific mineralocorticoid receptors at the level of the pituitary gland may account for the inhibitory effect of that steroid on ACTH secretion. By using pituitaries from neonatal rats, which we show to completely lack specific mineralocorticoid receptors but to contain a functional glucocorticoid receptor system, we demonstrated the persistence of aldosterone-induced inhibition of ACTH release from perifused glands. Conversely, when the glucocorticoid receptors sites were blocked in pituitaries from mature rats by means of a potent antiglucocorticoid (RU 38486), thus leaving unaltered mineralocorticoid binder, aldosterone no longer dampened hormonal output. We conclude that the latter steroid affected corticotropic activity by interacting not with its proper and specific receptor, but rather with the glucocorticoid binding sites.

Inhibition of programmed cell death in mouse embryonic palate in vitro by cortisol and phenytoin: receptor involvement and requirement of protein synthesis.

In an in vitro model cortisol and phenytoin inhibit the precisely timed process of palatal development, the lysosomally mediated cell death of the medial edge palatal epithelium. This inhibition of programmed cell death of the palatal midline epithelium by each drug is virtually completely blocked by the antiglucocorticoid, cortexolone, whose blocking action results from competitive binding of the glucocorticoid receptor site. The inhibition produced by each of these drugs is prevented by the protein synthesis blocker, cycloheximide. Thus, blockade of programmed cell death by each of these drugs involves the glucocorticoid receptor site and requires protein synthesis.

Glucocorticoids modulate the response of ornithine decarboxylase to unilateral removal of the dorsal hippocampus

he effect of unilateral removal of the dorsal hippocampus and of glucocorticoid administration was measured on the activity of ornithine decarboxylase (ODC) in the remaining contralateral hippocampus lobe. Unilateral hippocampectomy (Hx) resulted in a rapid rise of ODC activity in the contralateral lobe. The effect on ODC was maximal at 6 h after surgery and lasted two days. In the absence of the adrenals the effect of Hx on the enzyme was more potent and more prolonged. Elevated ODC activity was still detectable at 5 days after surgery, but not at 10 days. Chronic replacement with dexamethasone (DEX) offered in drinking water decreased the Hx-induced ODC response of ADX rats at 3 days after surgery to the level of enzyme activity observed in the S-ADX Hx subject. The effect of the steroid seemed related to the extent of occupation of the pool of glucocorticoid receptor sites in cytosol of rat hippocampus. In contrast, a single injection of a high dose of DEX to Hx-ADX animals at 3 days after surgery increased ODC activity in addition to the lesion-induced ODC in the contralateral lobe. It is concluded that after unilateral removal of the dorsal hippocampus ODC is a biochemical marker for cellular responses takin place in the contralateral lobe. Glucocorticoids modulate the lesion-induced ODC response.

Ketoconazole binds to glucocorticoid receptors and exhibits glucocorticoid antagonist activity in cultured cells.

We have recently found that ketoconazole inhibits adrenal steroidogenesis; in this paper we investigated whether imidazole antimycotic drugs additionally interact with glucocorticoid receptor sites in target tissues. Our approach was to assess the ability of three drugs: ketoconazole, clotrimazole, and RS 49910, to inhibit [3H]dexamethasone binding to hepatoma tissue culture (HTC) cell cytosol. The results indicated dose-dependent, competitive displacement of [3H]dexamethasone binding that was in the potency sequence: clotrimazole greater than ketoconazole greater than RS 49910. We then examined the functional response of this binding by measuring tyrosine aminotransferase (TAT) activity in HTC cells. The antimycotics did not exhibit TAT agonist activity and inhibition of basal enzyme levels was not detected. However, the drugs were potent antagonists of dexamethasone-induced TAT activity and the effect was temporally reversible. This antagonist activity was in the same sequence and closely correlated with the binding potency of the three drugs. We conclude that ketoconazole and other imidazole antimycotic drugs possess glucocorticoid antagonist activity by virtue of occupancy of glucocorticoid receptor sites in target tissues.

Mineralcorticoid receptor in rat kidney and hippocampus: characterization and quantitation by isoelectric focusing.

The mineralcorticoid receptor in cytosol from the kidneys and hippocampi of adrenalectomized rats was characterized and quantified by use of isoelectric focusing in slabs of polyacrylamide gel. The mineralcorticoid receptor was labeled with [3H]aldosterone after selective blockade of the glucocorticoid receptor sites with a synthetic pure glucocorticoid RU 26988 [11 beta, 17 beta-dihydroxy-17 alpha-(1-propionyl)androsta-1,4,6-trien-3-one]. The apparent dissociation constant for aldosterone was 5.1 X 10(-9) M, and the ligand specificity studies revealed a mineralcorticoid pattern. The glucocorticoid and mineralocorticoid receptors were compared by limited trypsin digestion, followed by isoelectric focusing. Completely different fragmentation patterns were obtained for the two receptors in the same cytosol from either kidney or hippocampus. However, comparison of the mineralcorticoid receptor from the kidney with that from the hippocampus gave rise to identical fragmentation patterns. An antiglucocorticoid receptor antibody showed no crossreactivity with the mineralcorticoid receptor. Cytosol from kidney and hippocampus contained 28 +/- 24 and 155 +/- 115 fmol mineralcorticoid receptors/mg cytosol protein, respectively. Significant amounts of RU 26988-resistant [3H]aldosterone-binding sites were also present in cerebrum, cerebellum, spleen, and colon.

Glucocorticoids and Lymphocytes. III. Effects of Glucocorticoid Administration on Lymphocyte Glucocorticoid Receptors

To determine the effects of glucocorticoid administration on the number of measured lymphocyte glucocorticoid receptor sites and the duration of such effects, seven normal volunteers were studied. Glucocorticoid receptor levels of the lymphocytes circulating in the blood of each volunteer were determined. Glucocorticoid was then administered in a regimen of a total of four doses of dexamethasone 4 mg p.o. every 6 hr. Determinations of the number of receptors were performed at 6 hr and at various subsequent times after the end of dexamethasone administration. When compared to baseline receptor numbers, six volunteers showed a decrease in receptor number after glucocorticoid administration (median maximum decrease 2,046 sites/cell). The fall in receptor number occurred rapidly, reaching a nadir within 30 hr from the end of glucocorticoid administration. The return of receptor number to baseline was more gradual, requiring from 3 to as long as 17 days in one subject. Our results suggest that in order to accurately interpret glucocorticoid receptor numbers in human lymphoid cells, glucocorticoid should not have been administered for 3 wk prior to determinations of receptor levels.

Glucocorticoids and lymphocytes. III. Effects of glucocorticoid administration on lymphocyte glucocorticoid receptors

To determine the effects of glucocorticoid administration on the number of measured lymphocyte glucocorticoid receptor sites and the duration of such effects, seven normal volunteers were studied. Glucocorticoid receptor levels of the lymphocytes circulating in the blood of each volunteer were determined. Glucocorticoid was then administered in a regimen of a total of four doses of dexamethasone 4 mg p.o. every 6 hr. Determinations of the number of receptors were performed at 6 hr and at various subsequent times after the end of dexamethasone administration. When compared to baseline receptor numbers, six volunteers showed a decrease in receptor number after glucocorticoid administration (median maximum decrease 2,046 sites/cell). The fall in receptor number occurred rapidly, reaching a nadir within 30 hr from the end of glucocorticoid administration. The return of receptor number to baseline was more gradual, requiring from 3 to as long as 17 days in one subject. Our results suggest that in order to accurately interpret glucocorticoid receptor numbers in human lymphoid cells, glucocorticoid should not have been administered for 3 wk prior to determinations of receptor levels.

Determination of occupied cytoplasmic glucocorticoid receptor sites by an exchange assay in rat muscles

We have established an assay to determine total cytoplasmic glucocorticoid receptor sites in heart, vastus and levator-ani-bulbocavernosus muscles from rats. We have tested the exchange of endogenously bound hormone with labeled triamcinolone acetonide by performing long incubation at 0–4°C. This procedure appeared to produce an adequate exchange of exogenously added unlabeled corticosterone in the range of 16 nM (normal unstressed concentration of corticosterone in plasma) with vastus and levator ani-bulbocavernosus muscles. Beyond that level, the displacement of unlabeled corticosterone by labeled triamcinolone acetonide was less efficient. By contrast, in heart and especially in liver cytosols, the exchange was adequate even in the presence of high concentrations of corticosterone (64 and 460 nM respectively). Using the exchange assay conditions (20 h of incubation at 0–4°C with 16 nM of labeled triamcinolone acetronide), we found no significant difference between glucocorticoid receptor levels in intact male and female rats. In both sexes, these glucocorticoid receptor values were not affected by gonadectomy, but increased more than 2-fold after adrenalectomy. No seasonal variations were recorded in intact or adrenalectomized rats.

Heterogeneity and ontogenesis of progestin receptors in rabbit lung

In order to evaluate the possible role of progesterone in fetal lung development, the presence of specific pulmonary progestin receptors and their ontogenesis were investigated in the rabbit fetus. Scatchard analysis of binding in lung cytosol from 29-day fetuses over a wide range of [ 3H]-R5020 concentrations indicates the presence of at least two binding sites. One of these sites, type I, is of very high affinity ( K D = 0.12 nM) and low capacity (26fmol per mg protein). The second binding site, type II, is of lower affinity ( K D = 36 nM) and higher capacity (240 fmol per mg protein). These two binding sites can be distinguished by sucrose density gradient centrifugation, the type I component sedimenting at 7.1 S and the type II component sedimenting at 4.5 S. Similar type I and type II sites are present in adult lung cytosol except that the type II binding component in adult lung sediments at 2.8 S rather than 4.5 S. Progesterone and R5020 compete well with [ 3H]-R5020 for binding to both sites while dexamethasone and cortisol do not compete. Thus the type I and type II binding sites appear to represent specific progestin receptors distinct from transcortin or the glucocorticoid receptor. The concentration of the type I sites increases significantly between the 20th and 29th day of gestation, with a further increase being observed in adult animals. The type II site is not measurable until 26 days of gestation and attains adult levels by day 29. Among a large number of fetal tissues examined, the lung contained the highest concentration of type I progestin receptor sites. Although cortisol and dexamethasone, even at very high concentrations, do not compete with [ 3H]-R5020 for binding to lung cytosol, the binding of [ 3H]-dexamethasone is inhibited significantly by nonlabeled progesterone or R5020 and this inhibition appears to be due to dissociation of [ 3H]-dexamethasone-receptor complexes. These results indicate that, in addition to type I and type II progestin receptor sites, fetal lung cytosol contains a third binding site, type III, which appears to be different from the glucocorticoid receptor site. Occupation of the type III site by progestins interferes with the binding of glucocorticoids to glucocorticoid receptors perhaps by increasing the rate of dissociation of glucoeortieoid-receptor complexes.

Glucocorticoid receptors in hairy-cell leukemia.

Glucocorticoid receptors were measured from leukemic cells of five patients with hairy-cell leukemia as well as a cell line (JOK-1) derived from one of the patients studied and grown in continuous culture in vitro. Receptor analyses were accomplished using a whole-cell binding assay with [3H]dexamethasone as the ligand. Glucocorticoid receptor sites in the leukemic hairy cells ranged from 5,710 to 11,400 sites/cell, and the mean dissociation constant (KD) between the receptor sites and [3H]dexamethasone was 0.67 +/- 0.03 X 10(-8) M. The glucocorticoid receptor nature of the binding activity was verified by demonstration of an appropriate order of competing ability for a selected group of ligands. The cell line JOK-1 contained 9830 receptor sites/cell with a KD of 0.78 X 10(-8) M. The in vitro growth pattern of this cell line was characterized by an almost complete insensitivity to glucocorticoids. We conclude that the postulated general presence of glucocorticoid receptors in various types of normal and malignant leukocytes extends to hairy-cell leukemia, a disease often characterized by resistance to antileukemic drugs, including corticosteroids. The sole presence of the glucocorticoid receptor thus does not seem to signify steroid responsiveness.

The effects of glucocorticoid therapy on glucocorticoid receptors in leukemia and lymphoma

Measurement of glucocorticoid receptors appears to be useful for selecting which patients with leukemia and lymphoma should receive glucocorticoid therapy. To determine the effect of recent or concurrent glucocorticoid therapy on the number of measured tumor glucocorticoid receptor sites, 18 patients with leukemia and lymphoma were studied. Baseline determinations of numbers of glucocorticoid receptors were performed on the malignant cells circulating in the patients' peripheral blood. Glucocorticoid therapy was then instituted consisting of dexamethasone 4 mg p.o. every 6 hr. Repeat determinations of the number of glucocorticoid receptor sites were performed within 24 hr and at various subsequent times from the start of therapy. When compared to baseline receptor numbers, 16 of 18 patients demonstrated a decrease in receptor number (median decrease 1651 sites/cell) after the start of glucocorticoid therapy. The magnitude of the change in receptor number was independent on the initial number of receptors. Our results suggest that in order accurately interpret glucocorticoid receptor numbers in patients with leukemia and lymphoma, glucocorticoid should not be administered for 3 wk prior to determinations of receptor levels.

The Effects of Glucocorticoid Therapy on Glucocorticoid Receptors in Leukemia and Lymphoma

Measurement of glucocorticoid receptors appears to be useful for selecting which patients with leukemia and lymphoma should receive glucocorticoid therapy. To determine the effect of recent or concurrent glucocorticoid therapy on the number of measured tumor glucocorticoid receptor sites, 18 patients with leukemia and lymphoma were studied. Baseline determinations of numbers of glucocorticoid receptors were performed on the malignant cells circulating in the patients' peripheral blood. Glucocorticoid therapy was then instituted consisting of dexamethasone 4 mg p.o. every 6 hr. Repeat determinations of the number of glucocorticoid receptor sites were performed within 24 hr and at various subsequent times from the start of therapy. When compared to baseline receptor numbers, 16 of 18 patients demonstrated a decrease in receptor number (median decrease 1651 sites/cell) after the start of glucocorticoid therapy. The magnitude of the change in receptor number was independent on the initial number of receptors. Our results suggest that in order accurately interpret glucocorticoid receptor numbers in patients with leukemia and lymphoma, glucocorticoid should not be administered for 3 wk prior to determinations of receptor levels.