Binding of dexamethasone to rat liver nuclei in vivo and in vitro: evidence for two distinct binding sites

Steroid binding sites in liver membranes: interplay between glucocorticoids, sex steroids, and pituitary hormones

Steroid hormones activate target cells through specific receptors that discriminate among ligands based upon recognition of distinct structural features. For most known steroids, membrane and nuclear receptors co-exist in many target cells. However, while the structure of the nuclear receptors and their function as transcriptional activators of specific target genes is generally well understood, the identity of the membrane receptors remains elusive. Using pharmacological and biochemical approaches, we are beginning to characterize receptors for glucocorticoids and anabolic-androgenic steroids in male rat liver membranes. Male rat liver endoplasmic reticulum contains two steroid binding sites which are functionally related and associated with a 90-134 kDa oligomeric protein: (1) the low-affinity glucocorticoid binding site (LAGS), composed at least in part of two peptides (37 and 53 kDa) that bind glucocorticoids and (2) the stanozolol binding protein (STBP), composed at least in part of three peptides (22, 31, and 55 kDa) that bind the synthetic androgen stanozolol. These steroid binding proteins have many properties different from those of classical nuclear receptors, with the salient differences being a failure to recognize "classical" ligands for nuclear receptors together with marked differences in biochemical properties and physiological regulation. The mechanism of interaction of glucocorticoids with the LAGS can be clearly distinguished from that with STBP. Moreover, STBP shows an extremely narrow pharmacological profile, being selective for ST and its analog, danazol, among more than 100 steroids and non-steroidal compounds that were assayed, including those that are able to displace glucocorticoids from the LAGS. The level of LAGS activity undergoes dramatic variations following changes from the physiological serum levels of thyroid hormones, glucocorticoids, GH, vitamin A, and E2. However, neither thyroid hormones nor GH have a critical role on STBP activity. The STBP is functionally related to LAGS. We have suggested a novel mechanism for STBP whereby membrane-associated glucocorticoid binding activity is targeted by stanozolol (and 16beta-hydroxylated stanozolol): stanozolol modulates glucocorticoid activity in the liver through negative allosteric modulation of the LAGS resulting in an effective increase in classical GR-signaling by increasing glucocorticoid availability to the cytosolic GR.

Vesicles in the subacrosomal space and partial diaphragms in the subacrosomal nuclear envelope of round spermatids of a rat injected intravenously with gold labeled-testosterone-bovine serum albumin conjugate: vesicular trafficking from acrosome to nucleus

Colloidal gold labeled-testosterone-bovine serum albumin conjugate (testosterone-BSA-gold) injected into the vascular system of rats is taken up by endocytosis into round spermatids. Based on observation of silver deposits indicating testosterone-BSA-gold with silver enhancement, we have suggested that testosterone-BSA-gold enters the nuclei through not only the postacrosomal nuclear envelope but also the subacrosomal nuclear envelope (SNE) via the acrosome (Nishimura and Nakano, 1997). However, it was unclear how testosterone-BSA-gold in the acrosome entered the nucleoplasm. Spermatids showing silver deposits on the subacrosomal space were observed under electron microscope without silver enhancement, to clarify the courses of translocation. In the spermatids, vesicles with the gold particles were seen in the subacrosomal space. Some of the vesicles were in contact with the SNE. A part of the outer nuclear membrane projected into the space. Furthermore, local single-bilayer nuclear membranes, which seemed to partially lack nuclear lamina, were present in the SNE. These results indicate the possibility that the vesicles mediate the transport of testosterone-BSA-gold from acrosome to nucleus, and that the vesicle membrane fuses with not only the outer nuclear membrane but also a shared bilayer in the SNE.

Nuclear localization of gold labeled-hydrocortisone-bovine serum albumin conjugate injected intravenously into the hormone-target cells of rat

We have suggested in a previous study using 2-nm colloidal gold labeled-testosterone-bovine serum albumin (testosterone-BSA-gold) that 2-nm gold labeled-steroid hormone-BSA conjugates would be a useful tool for analyzing the mechanism of steroid hormone action (39). In this study, we examined whether hydrocortisone-BSA conjugate (hydrocortisone-BSA) showed a similar distribution to radiolabeled hydrocortisone in vivo, by injecting 2-nm colloidal gold labeled-hydrocortisone-BSA (hydrocortisone-BSA-gold) into the rat tail vein. The hydrocortisone-BSA-gold with silver enhancement became visible as silver deposits under electron microscopy in the nuclei of hepatocytes and hepatic stellate cells but not in Kupffer cells in the liver, and in the thymocytes and thymic reticuloepithelial cells in the thymus of a rat killed 2 h postinjection. The percentage of nuclei showing deposits in the non-target cells, the epithelial cells of the seminal vesicle, was similar to the value in the seminal vesicle of a control rat injected with BSA labeled with 2-nm colloidal gold as reported previously. In the hepatocytes and thymocytes of a control rat not injected, the percentages of nuclei showing deposits were similar to those in the rat injected with testosterone-BSA-gold or BSA-gold as reported previously, but lower than those in the rat injected with hydrocortisone-BSA-gold. These results suggest that hydrocortisone-BSA-gold is useful for the morphological study of hydrocortisone target cells, and imply that BSA conjugated with hydrocortisone can enter the target cell nuclei of the rat. The present study further indicates that the fate of gold labeled-steroid hormone-BSA conjugates may be decided at the cell membrane level.

Effect of polyelectrolyte complex (PEC) on human periodontal ligament fibroblast (HPLF) functions in the presence of glucocorticoids

Cell functions in vivo are stimulated by extracellular matrices, vitamins, growth factors, and hormones. In this paper, the effects of glucocorticoids, dexamethasone (Dex), and Cortexrone (Cor) on the growth and differentiation of human periodontal ligament fibroblast (HPLF) were discussed in relation to a polyelectrolyte complex (PEC) consisting of polysaccharides (chitin, cellulose derivatives, and chitosan) as a tissue-culture material. A Dex-treatment at a concentration of 10(-)-10(-7) M inhibited one-half of HPLF growth in comparison with 10(-9) M Dex-treatment and no additive medium and produced aggregates on the chitosan-sulfated chitin PEC (SPECs) with regard to the degree of sulfate substitution. On the chitosan-sulfated cellulose PEC, 10(-7)-10(-9) M Dex-treatment promoted HPLF growth and inhibited the production of aggregates. On the other hand, a Cor-treatment, a mineral corticoid, which inhibits the interaction between Dex and its receptor, increased HPLF growth on SPEC141, but the HPLF did not construct aggregates. A Dex and Cor mixture-treatment inhibited one-third HPLF growth in comparison with 10(-5) M Dex-treatment and produced aggregates on PEC. The cooperative effect of both the culture material and hormones was found to control HPLF growth and morphology. The alkaline phosphatase (ALPase) activities of HPLF increased with an increase in the Dex and Cor concentration. The value of Dex-treated HPLF ALPase activity demonstrated a two-fold increase from that with Cor-treatment. The ALPase activity of Dex and Cor mixture-treated HPLF on PEC decreased with an increase in the Cor concentration, because Cor increased HPLF growth on PEC. In using carboxymethylated chitin derivatives as the polyanion, HPLF decreased in cell growth and produced aggregates in the absence of the additives, suggesting that PEC induces HPLF differentiation using only the stimulation of the material surface.

Expression of retinoic acid, triiodothyronine, and glucocorticoid hormone nuclear receptors is decreased in the liver of rats fed a hypercholesterolemia-inducing diet

Several studies have shown that dietary factors modulate cell signaling pathways. The aim of this study was to determine whether a hypercholesterolemia-inducing diet rich in saturated fat and cholesterol modifies rat liver expression of the nuclear receptors of retinoic acid (RAR), triiodothyronine (TR), and glucocorticoid hormone (GR), which are transcriptional factors. The experimental diet contained coconut oil 25 g/100 g as a source of lipids, cholesterol 1 g/100 g, and cholic acid 0.5 g/100 g, and the control diet contained olive oil 5 g/100 g. After 26 days of feeding the hypercholesterolemia-inducing diet, a lower binding capacity of the nuclear receptors and a smaller amount of their mRNA were observed. Moreover, the activities of malic enzyme (ME) and tyrosine aminotransferase (TAT), whose gene promotors contain a response element to TR and GR, respectively, were significantly decreased. These changes occurred in a cellular environment characterized by a high level of cholesterol and free fatty acids (FFAs). Thus, two nonexclusive hypotheses can be proposed to explain this decreased expression of nuclear receptors, one emphasizing the effect of lipidic components on the cellular amount of receptor ligands (retinoic acid [RA] and triiodothyronine [T3]), the other emphasizing a modification of the balance between nuclear receptors that could impede the upregulation of TR and RAR.

Estrogen antagonism on T3 and growth hormone control of the liver microsomal low-affinity glucocorticoid binding site (LAGS)

Male rat liver microsomes contain a low-affinity glucocorticoid binding site (LAGS) capable of binding all natural glucocorticoids and progesterone with a Kd from 20 to 100 nM. The LAGS level is under endocrine control by T3, glucocorticoids and GH. These hormones act synergistically at physiological concentrations to increase the LAGS level. Since female rats show a LAGS level that is much lower than the males (0.15 vs 23 pmol/mg protein, respectively), here we investigated whether estradiol could decrease the LAGS in the male rat. Orchiectomized (OX) male rats showed a higher LAGS level than intact rats. This effect was reversed by implanting a Sylastic capsule containing testosterone. When the OX rats were implanted for 20 days with estrogen capsules that provided an estradiol level in serum of 40 pg/ml, their LAGS level decreased from 23 to 0.2 pmol/mg protein. This effect was not observed in intact male rats and can be partially reversed by testosterone implants into OX rats. Both hypophysectomized male rats and hypothyroid-orchiectomized male rats showed very low levels of LAGS. Administration of physiological doses of GH and/or T3 to these rats greatly increased their LAGS level (from 0.3 to 15 and 16 pmol/mg protein, respectively). Implantation of estrogen capsules to these rats two weeks prior to starting treatment completely inhibited the increase in the LAGS level in response to T3, and significantly decreased the response to hGH, and to a combination of hGH and T3. These results suggest that physiological estradiol levels can antagonize the LAGS induction by T3 and hGH in the male rat, and could be responsible for the low level of LAGS in the female rat. Moreover, estrogen capsules also inhibited the increase in the body and hepatic weights observed after hGH treatment, which suggests a powerful inhibitory effect of low estradiol levels on the male rat liver functions under regulation by T3 and/or GH.

Correlation between the effects of retinoic acid and dexamethasone on liver tyrosine aminotransferase

A single dose of 50 microg of trans-retinoic acid administered to rats significantly raised the level of hepatic tyrosine after a few hours. This effect was compared with that of dexamethasone and a possible correlation between these effectors was also investigated. An equal increase in enzyme activity level caused by retinoic acid was observed in adrenalectomized rats, leading to the suggestion that the effect of retinoic acid on liver tyrosine aminotransferase is independent of glucocorticoids. However, the study of the binding activity of the liver nuclear glucocorticoid receptors vs dexamethasone showed that this activity is favoured by retinoic acid, whereas no variation was evidenced for retinoic acid receptors caused by dexamethasone. In the adrenalectomized rat, the synergistic effect produced by the association of retinoic acid and dexamethasone at the lowest doses used led us to conclude that retinoic acid is an efficient effector of liver tyrosine aminotransferase. It probably affects tyrosine aminotransferase activity in a dependent and an independent way, regulated respectively by the glucorticoid status and by the provision of retinoic acid.

The influence of dietary vitamin A on triiodothyronine, retinoic acid, and glucocorticoid receptors in liver of hypothyroid rats

The properties of nuclear receptors belonging to the superfamily of receptors acting as transcription factors are modulated by nutritional and hormonal conditions. We showed recently that retinoic acid (RA) restored to normal the expression of receptors attenuated by hypothyroidism. The present study was designed to find out whether dietary vitamin A (as retinol) had the same effect. Propylthiouracil in drinking water induced both hypothyroidism and a vitamin A-deficient status in rats. The maximum binding capacity (Cmax) of triiodothyronine nuclear receptors (TR) was unchanged, while that of nuclear RA receptors (RAR) and nuclear glucocorticoid hormone receptors (GRn) was reduced in the liver of these hypothyroid rats. The reduced Cmax of RAR stemmed from a lower level of RAR mRNA, while the reduced Cmax of GRn was assumed to be due to reduced translocation of the receptor from the cytosol to the nucleus. Feeding the hypothyroid rats with a vitamin A-rich diet did not restore the Cmax of either RAR or GRn to normal. The lack of effect of dietary retinol on RAR expression may be attributed to either genomic (unoccupied TR block the expression of RAR genes) and/or extragenomic (hypothyroidism decreases the availability of retinol and/or its metabolism to RA) mechanisms. Triiodothyronine is thought to favour the translocation of glucocorticoid hormone receptors from cytosol to nucleus. These findings provide more information on the relationship between vitamin A and hormonal status, showing that a vitamin A-rich diet is without apparent effect on the expression of nuclear receptors in hypothyroid rats.

[3H]dexamethasone binding activity in liver microsomes is modulated differently by 17 alpha-alkylated androgens and testosterone in vivo

Rat liver microsomes contain a single class of steroid binding sites, capable of binding various glucocorticoids and progesterone. In a previous article, we have described the in vitro interaction of several androgens with this binding site. Unlike natural androgens, the 17 alpha-alkyl derivatives stanozolol and danazol were capable of interacting with this binding site through a negative allosteric pattern. Now, the effects these steroids exert on the microsomal [3H]dexamethasone binding site have been studied in vivo. The administration of a single dose of stanozolol to rats provoked a significant reduction in the microsomal [3H]dexamethasone binding capacity. This effect was maximal two hr after stanozolol administration and persisted for six hr. The restoration of the [3H]dexamethasone binding level after stanozolol administration was dependent on protein synthesis, since it was blocked by the concomitant administration of cycloheximide. None of the other androgens tested (danazol, methyltestosterone, fluoxymesterone, and testosterone propionate) was capable of provoking a similar effect when administered 2 or 24 hr prior to sacrifice. In rats treated for seven days with a daily dose of diverse androgens and sacrificed 24 hr after the last treatment, none of the 17 alpha-alkyl androgens assayed provoked significant changes in the microsomal [3H]dexamethasone binding level, although stanozolol, danazol, and methyltestosterone provoked a significant increase in glucocorticoid receptor concentration. In contrast, the administration of testosterone propionate provoked a 50% reduction in the [3H]dexamethasone binding level without causing changes in the glucocorticoid receptor concentration. These results provide new evidence on the existence of different effects on the liver of 17 alpha-alkyl androgens, compared to the effects produced by natural androgens.

Retinoids modulate the binding capacity of the glucocorticoid receptor and its translocation from cytosol to nucleus in liver cells

The binding capacity (Cmax) of the glucocorticoid hormone receptor (GR) was affected by vitamin A status in rat liver. In rats fed on a vitamin A-overloaded diet as well as in rats administered with retinoic acid (RA) there was an increased ratio Cmax of nuclear GR (expressed as fmol/mg liver): Cmax of cytosolic GR (expressed as fmol/mg liver) while in rats fed on a vitamin A-deficient diet this ratio was decreased. These results suggested that an increased amount of RA, resulting from either metabolization of an increased amount of dietary retinol or RA administration, enhanced the translocation of GR from the cytosolic compartment to the nuclear compartment. Moreover such an increased amount of RA could also induce the observed decreased Cmax of the total GR that we observed. These observations were similar to the well known effects of dexamethasone administration on the properties of GR. It is probable that RA, similarly to dexamethasone treatment, induces a dissociation of the tetrameric form of the cytosolic GR and thus enhances translocation of the monomeric form from cytosol to nucleus and also resulting in an increased proteolytic degradation of the receptor.

The estradiol induction of the microsomal low-affinity glucocorticoid binding sites (LAGS) in the male rat liver is independent of the endocrine status

The low-affinity glucocorticoid binding sites (LAGS) are entities present in the microsomal fraction of the rat liver, capable of binding several glucocorticoids and progesterone with low affinity. The present work focuses on the demonstration that estradiol exerts a powerful stimulatory effect on the LAGS concentration. For this purpose, we studied the effect of this hormone in immature, hypothyroid, and hypophysectomized rats, three experimental models which present a very low level of LAGS. In all of them, estradiol showed ability to significantly increase the level of LAGS. The positive results obtained in hypophysectomized rats point to a direct action of estradiol on the liver. In immature rats, the estradiol induction of the LAGS was shown to be especially slow, 3-4 days after estradiol administration being necessary to obtain a significant rise in the level of LAGS. Moreover, the dose of estradiol necessary to obtain the LAGS induction in these rats (0.5 mg/100 g body weight) was clearly supraphysiological. From these data we concluded that: (A) estradiol is a powerful stimulator of the LAGS concentration, its effect probably being exerted directly on the liver; and (B) to elicit its effect, estradiol does not need the participation of other hormones known to be implicated in the endocrine regulation of the LAGS.

Glucocorticoid type II receptors of the spinal cord show lower affinity than hippocampal type II receptors: binding parameters obtained with different experimental protocols

We have used three experimental protocols to determine binding parameters for type I and type II glucocorticoid receptors in the spinal cord and hippocampus (HIPPO) from adrenalectomized rats. In protocol A, 0.5-20 nM [3H]dexamethasone (DEX) was incubated plus or minus a 1000-fold excess of unlabeled DEX, assuming binding to a two-site model. In protocol B, [3H]DEX competed with a single concentration of RU 28362 (500 nM), whereas in protocol C, we used a concentration of RU 28362 which varied in parallel to that of [3H]DEX, such as 500 x. Results of protocols A and C were qualitatively similar, in that: (1) Bmax for type I receptors favored the HIPPO, while the content of type II sites was comparable in the two tissues; (2) Kd was consistently lower for type I than for type II sites in both tissues; and (3) type II receptors from the spinal cord showed lower affinity than their homologous sites from HIPPO. This last result was also obtained when using protocol B. In contrast, protocol B yielded binding data indicating that type II sites were of similar or higher affinity than type I sites. Computer simulation of the binding protocols demonstrated that protocols A and C were the most theoretically reliable for estimating the Kd and Bmax of type I sites, and the predicted error was smaller for protocol C, in comparison with protocol B. We suggest that the noted differences in the Kd of type II receptors between the spinal cord and HIPPO could account for a difference in sensitivity of the two systems in the physiological adrenal hormone range.

Characterization of high affinity and low affinity dexamethasone binding sites on male rat liver nuclear envelopes

Steroids must traverse the nuclear envelope before exerting their action at the chromatin. However, few studies have been done to elucidate the mechanism by which steroids traverse this membrane barrier. As first steps towards investigating the mechanism, we have characterized the binding sites for dexamethasone on male rat liver nuclear envelopes. The nuclear envelopes, prepared in the presence of dithiothreitol, were isolated from purified nuclei after treatment with DNase 1 at high pH. Binding of dexamethasone to the nuclear envelopes was measured after 16 h of incubation at 0-4 degrees C. At pH 7.4, only a single high capacity, low affinity binding site for dexamethasone was identified. However, at pH 8.6, two sites were identified; a low capacity, high affinity site and a high capacity, low affinity site. Adrenalectomy of the animal before preparation of the membranes caused loss of the high affinity site and reduction in the number of the lower affinity sites. Acute dexamethasone treatment of adrenalectomized rats resulted in the reappearance of the high affinity site but long term treatment with dexamethasone was required for complete restoration of the high affinity sites and reappearance of any of the low affinity sites. The steroid specificity of these nuclear envelope binding sites was different from that of the cytosolic glucocorticoid receptor, generally showing broader specificity. However, triamcinolone acetonide, which is a potent competitor for binding to the glucocorticoid receptor, did not complete effectively. The binding sites were sensitive to protease treatment and salt extraction studies revealed that the dexamethasone binding sites do not represent proteins non-specifically bound to the nuclear envelope. The affinity and the hormone responsiveness of the high affinity site are similar to those of the nuclear glucocorticoid receptor. Therefore, the nuclear envelope may be a site of action of glucocorticoids.

Adrenocorticoid action in the spinal cord: some unique molecular properties of glucocorticoid receptors

1. Glucocorticoid hormones affect several functions of the spinal cord, such as synaptic transmission, biogenic amine content, lipid metabolism, and the activity of some enzymes (ornithine decarboxylase, glycerolphosphate dehydrogenase), indicating that this tissue is a target of adrenal hormones. 2. Corticosterone, the main glucocorticoid of the rat, is detected at all regional levels of the spinal cord, and cold stress increases this steroid, predominantly in the cervical regions. 3. Intracellular glucocorticoid receptors have been found in the spinal cord, with higher concentrations in the cervical and lumbar enlargements. Prima facie, these receptors presented biochemical, stereospecifical, and physicochemical properties similar to those of receptors found in other regions of the nervous system. The prevalent form in the spinal cord is the type II receptor, although type I is also present in small amounts. 4. The type II glucocorticoid receptor of the spinal cord shows an affinity lower (Kd 3.5 nM) than that of the hippocampal type II site (Kd 0.7 nM) when incubated with [3H]dexamethasone. This condition may impair the nuclear translocation of the spinal cord receptor. 5. Another peculiar property of spinal cord type II site is a greater affinity for DNA-cellulose binding than the hippocampal receptor during heat-induced transformation. Also, the spinal cord receptor shows resistance to the action of RNAse A, an enzyme which increases DNA-cellulose binding of the hippocampal receptor, indicating that both receptors may be structurally different. 6. Therefore, it is possible that a different subclass of type II, or "classical glucocorticoid receptor," is present in the spinal cord. This possibility makes the cord a useful system for studying diversity of glucocorticoid receptors of the nervous system, especially the relationship between receptor structure and function.

Steroid induction of low-affinity glucocorticoid binding sites in rat liver microsomes

Rat liver contains two glucocorticoid binding sites: the high-affinity or glucocorticoid receptor (GR) and the low-affinity glucocorticoid binding sites, or LAGS. The Kd of LAGS predicts that they can be half-saturated by plasma corticosteroids in some physiological circumstances and, therefore, that they can play relevant roles in the rat liver. [3H]dexamethasone was used as a ligand in exchange assays, to study the relative abundance of GR and LAGS in cell fractions of rat liver. GR were found in the cytosol, but not in the purified nuclei, the mitochondria, or the microsomes. LAGS were found in all the particulate fractions, being more abundant in the smooth-surfaced microsomes, but they were not found in the cytosol. The LAGS of microsomes and purified nuclei showed the same Kd and also the same broad range of steroid competition with [3H]dexamethasone (cortisol = progesterone greater than dexamethasone greater than or equal to corticosterone greater than R5020 greater than DHEA greater than testosterone = estradiol). LAGS were found in liver, placenta and kidney, but not in other GR-containing organs. This suggests that the LAGS could be involved in physiological functions related to the metabolism of steroid hormones. The liver microsome LAGS were undetectable at rat birth, and became present in the 25-day-old rat. The level of LAGS then increased progressively, reaching its maximum level in the 2-3-month-old rats (10 pmol/mg protein), and declining afterwards to reach the adulthood level (5 pmol/mg protein) in 6-month-old rats. LAGS are mainly controlled by the corticoadrenal steroids, which is shown by their dramatic decrease after adrenalectomy, and especially after hypophysectomy. Many steroid hormones, like estradiol, testosterone, and corticosterone (but not progesterone) induce LAGS, estradiol being the most effective. A combination of T4 and corticosterone was more effective in inducing LAGS than when the two hormones were injected separately. It is possible to conclude that rat liver LAGS are mainly microsomal proteins, whose concentration is regulated by a multihormone system under pituitary control.

Two distinct mechanisms for regulation of gamma-glutamyl transpeptidase in cultured rat hepatocytes by glucocorticoid-like steroids

Adult rat hepatocytes maintained in primary monolayer culture with defined medium were used to characterise two effects of glucocorticoid-like steroids in regulating gamma-glutamyltranspeptidase (GGT). Low concentrations of glucocorticoids alone had little effect on GGT but synergistically enhanced induction of the enzyme by liver tumor-promoting xenobiotics such as 1,1,1-trichloro-2,2-bis-(4-chlorophenyl)-ethane (DDT) and hexachlorocyclohexane. The enhancing effect appears to be mediated by the classical glucocorticoid hormone receptor since structural requirements and concentration-dependence for enhancement were similar to those for induction of tyrosine aminotransferase in parallel cultures. Higher concentrations (1-100 microM) of various glucocorticoids alone increased GGT activity. Most glucocorticoids induced GGT but their order of potency did not parallel that for induction of tyrosine aminotransferase under similar culture conditions. Among the most potent glucocorticoids, triamcinolone was a weak GGT inducer and cortivazol appeared to act as an antagonist of GGT induction by steroids. Some non-glucocorticoids including pregnenolone 16 alpha-carbonitrile, and some progestins, also induced but required addition of 30 nM dexamethasone for maximal effect. Some specific steroid structural features were identified which increased (presence of a 16 alpha methyl group) or impaired GGT-inducing activity. Although interpretation is complicated by differential metabolism of individual steroids in culture, the results suggest that GGT induction by pharmacological levels of steroids may be mediated, directly or indirectly, by one or more relatively specific receptors distinct from the classical glucocorticoid receptor.