Kinetic Pulse-Chase Labeling Study of the Glucocorticoid Receptor in Mouse Lymphoma Cells EFFECT OF GLUCOCORTICOID AND ANTIGLUCOCORTICOID HORMONES ON INTRACELLULAR RECEPTOR HALF-LIFE*

A kinetic pulse-chase labeling technique was used to measure the intracellular half-life of the glucocorticoid receptor in 549 mouse lymphoma cells. Cells were pulse-labeled with [36S]methionine for 30 min and then cultured in the presence of unlabeled methionine (chase). Labeled receptors were quantitated at periodic time points during the chase by immunoadsorption to protein A-Sepharose using the BuGR2 monoclonal an- tireceptor antibody. The decay of labeled receptors during the chase was linear on a semilog plot, consistent with first order kinetics. Receptor half-life was 9 h when cells were cultured in either phenol red-con-taining medium supplemented with fetal calf serum or in phenol red free-medium supplemented with charcoal extracted serum, indicating that endogenous steroids do not affect receptor half-life. Receptor half-life was also unchanged when cells were cultured in the presence of 0.1 p~ dexamethasone, a glucocorticoid hor- mone, or 0.1 PM RU486 (llj3-(4-dimethylamino-phenyl)-l7j3-hydroxy-l7a-(propynylestra-4,9-diene- %one), an antiglucocorticoid hormone. We conclude that the intracellular half-life of the glucocorticoid receptor in S49 mouse lymphoma

A kinetic pulse-chase labeling technique was used to measure the intracellular half-life of the glucocorticoid receptor in 549 mouse lymphoma cells. Cells were pulse-labeled with [36S]methionine for 30 min and then cultured in the presence of unlabeled methionine (chase). Labeled receptors were quantitated at periodic time points during the chase by immunoadsorption to protein A-Sepharose using the BuGR2 monoclonal antireceptor antibody. The decay of labeled receptors during the chase was linear on a semilog plot, consistent with first order kinetics. Receptor half-life was 9 h when cells were cultured in either phenol red-containing medium supplemented with fetal calf serum or in phenol red free-medium supplemented with charcoal extracted serum, indicating that endogenous steroids do not affect receptor half-life. Receptor half-life was also unchanged when cells were cultured in the presence of 0.1 p~ dexamethasone, a glucocorticoid hormone, or 0 . 1 PM RU486 (llj3-(4-dimethylaminophenyl)-l7j3-hydroxy-l7a-(propynylestra-4,9-diene-%one), an antiglucocorticoid hormone.
We conclude that the intracellular half-life of the glucocorticoid receptor in S49 mouse lymphoma cells is not regulated by either glucocorticoid or antiglucocorticoid hormones.
The glucocorticosteroid hormone receptor is an important regulatory molecule that interacts with specific enhancer sequences on DNA and thereby regulates gene transcription (1).
The role of the steroid hormone ligand appears to be that of an effector molecule that binds to the receptor and converts the receptor from an inactive, non-DNA binding form to an active, DNA binding form. The intracellular concentration of glucocorticoid receptors is an important determinant of the magnitude of the glucocorticoid response. This conclusion is based on two lines of evidence. First, there is a strong correlation between the intracellular concentration of glucocorticoid receptors and the cytolytic response of murine lymphoma cells to glucocorticoids (2-4). Second, the magnitude of tran-*This work was supported in part by Grants CA42755 and P30CA43703 from the National Institutes of Health and by a grant from the Ireland Cancer Center. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
$ scriptional regulation by glucocorticoids in cells that have been transfected with the glucocorticoid receptor gene is proportional to the level of glucocorticoid receptor expression achieved ( 5 ) . It is, therefore, possible that regulation of glucocorticoid receptor levels may be a physiological means of modulating cellular responses to glucocorticoids.
There is considerable evidence that glucocorticoid receptor levels in cells are regulated by glucocorticoid hormones. Glucocorticoid hormones have been found to down-regulate the level of their own receptor in a number of different cell lines (3, 6-11), in target tissues of intact rats (12, 13), and in lymphocytes of steroid-treated human volunteers (14). More recently, glucocorticoid receptor levels have been shown to be upregulated in a human leukemia cell line, suggesting that the pattern of glucocorticoid receptor autoregulation may be tissue-specific (15). Anti-glucocorticoid hormones, such as RU486,' do not appear to regulate the level of glucocorticoid receptors, despite their ability to bind with high affinity to the glucocorticoid receptor (16).
Glucocorticoids could down-regulate glucocorticoid receptors by decreasing the rate of receptor synthesis or by increasing the rate of receptor degradation. To investigate the latter possibility, we have employed a novel pulse-chase labeling technique to measure the intracellular half-life of the glucocorticoid receptor, both in the presence and absence of hormone. Our results, which are described here, indicate that neither glucocorticoid nor anti-glucocorticoid hormones have a significant effect on the intracellular half-life of the glucocorticoid receptor in mouse lymphoma cells.

Measurement of Receptor
Half-life-A kinetic pulse-chase labeling technique was used to measure the intracellular halflife of the glucocorticoid receptor in wild type, glucocorticoidresponsive S49.1 mouse lymphoma cells. Cells were pulselabeled with [35S]methionine for 30 min and then cultured in The abbreviations and trivial names used are: RU486, llo-(4dimethylaminophenyl)-l7~-hydroxy-l7a-(propynylestra-4,9diene-3-one; PBS, phosphate-buffered saline; dexamethasone, 901fluoro-16a-methyl-ll~, 17a, 21-trihydroxypregna-1,4-diene-3,20dione; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; HSP90, heat shock protein 90; DMEM, Dulhecco's modified Eagle's medium. Portions of this paper (including "Experimental Procedures" and Figs. 2-7) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that is available from Waverly Press. medium supplemented with 1 mg/ml unlabeled methionine for a 24-h period referred to as the chase. Cell viability, determined by measuring trypan blue dye exclusion, was greater than 90% throughout the chase. To insure that [35S] methionine incorporation did not continue during the chase, cytosol prepared from cells at the beginning of the chase (0 h) and at different time points during the chase was subjected to SDS-PAGE (Fig. 1A). The amount of labeled protein did not increase during the chase, indicating that incorporation of [35S]methionine into newly synthesized protein was effectively blocked by culturing cells in excess unlabeled methionine.
At different time points during the chase, cytosol was prepared from cells and labeled receptors were immunoadsorbed to protein A-Sepharose using the BuGR-2 monoclonal antireceptor antibody. Immunoadsorbed proteins were subjected to SDS-PAGE and analyzed by autoradiography (Fig.  1B). The band opposite the 97.4-kDa standard corresponds to intact glucocorticoid receptor. Lower molecular weight receptor fragments were not detected in this or subsequent experiments. To quantitate the amount of labeled receptor recovered at each time point in the chase, the density of the receptor band on autoradiograms was measured by scanning densitometry (Fig. IC). The density of the receptor band decreased with time, indicating that the amount of labeled receptor that was recovered from cytosol decreased during the chase.
The validity of scanning densitometry for measuring relative amounts of labeled protein on autoradiograms is documented in Fig. 2. In this experiment, serial dilutions of an immunoadsorbed [35S]methionine-labeled protein (the 90-kDa heat shock protein, HSP9O) were subjected to gel electrophoresis and duplicate gels were analyzed by two different methods. In one method, the gel was analyzed by autoradiography and the absorbance of bands on the autoradiogram was measured by scanning densitometry. In the other method, the wet gel was sliced and the amount of radioactivity in gel slices was measured using a liquid scintillation counter. As shown in Fig. 2, there is a linear relationship between absorb- ance and radioactivity over the range of laser densitometry slit widths employed in this study.
Multiple separate pulse-chase labeling experiments were performed to test the reproducibility of the observation shown in Fig. 1. In one experiment, the amount of labeled receptor was quantitated at frequent intervals throughout the chase (Fig. 3A). The decay of labeled receptor appears to be linear (correlation coefficient = 0.93), consistent with the kinetics of a first order reaction, although more complex kinetics cannot be completely excluded. Recovery of labeled receptors at 4, 6, 8, and 22 h into the chase was quantitated in 10 separate experiments (Fig. 3B). Based on these data, the intracellular half-life of the glucocorticoid receptor is 9 h.
Although the decay of labeled receptors described in the preceding experiments is likely to represent intracellular receptor degradation, an alternative explanation might be that labeled receptors gradually shift to the nucleus during the chase and are therefore not recovered in cytosol extracts. To exclude this possibility, the same pulse-chase labeling method was applied to a glucocorticoid-unresponsive variant of the S49 mouse lymphoma cell line, S49.22r. The receptors in this variant are defective in terms of nuclear translocation and DNA binding due to a single amino acid substitution in the DNA-binding region (28,29). We found the rate of decay of labeled receptors is the same in S49.22r cells and in wild type S49 cells (data not shown), indicating that nuclear translocation is not responsible for the decrease in recovery of labeled receptors from cells during the chase. The pulse-chase labeling technique was used to determine the effect of pharmacologic concentrations of the glucocorticoid hormone dexamethasone on receptor half-life. Addition of dexamethasone to pulse-labeled cells at the beginning of the chase induced a concentration dependent decrease in the recovery of labeled receptors from cytosol at 4 h into the chase (Fig. 5). The effect of dexamethasone appeared to be specific for the glucocorticoid receptor since dexamethasone did not decrease the recovery of a control protein (the 90-kDa heat shock protein HSP9O) from the same cells. Two points are worthy of note in regard to experimental design. First, dexamethasone was added to cells after completion of the pulse-labeling period. Thus, dexamethasone did not affect the incorportion of [3SS]methionine into protein and the amount of labeled protein present in cytosol extracts of dexamethasone-treated and control cells was the same (data not shown). Second, both liganded and unliganded receptors are immunoadsorbed by the BuGR-2 antibody with equal efficiency (data not shown). For the following reasons, we conclude that the dexamethasone-induced decrease in recovery of labeled receptors is due to nuclear translocation and tight nuclear binding such that labeled receptors are not recovered in the cytosol fraction. First, the dexamethasone-induced decrease in recovery of labeled receptors from wild type S49 cells was detected as early as 30 min into the chase (data not shown), consistent with earlier evidence that dexamethasone-induced nuclear localization occurs within minutes of adding dexamethasone to cells (30). Second, dexamethasone did not decrease the recovery of labeled receptors from nuclear transfer defective S49.22r cells (data not shown).
Following the dexamethasone-induced shift of labeled receptors into the nucleus, the recovery of labeled receptors from dexamethasone-treated and control cells decreased at the same rate, indicating that the intracellular half-life of the glucocorticoid receptor is not altered by dexamethasone (Fig.  6). In these experiments, control and dexamethasone-treated cells were cultured under cortisol-free conditions (phenol redfree medium and charcoal-extracted serum). Dexamethasone also failed to change receptor half-life in cells that were cultured in the presence of unextracted fetal calf serum (data not shown).
Effect of RU486 on Receptor Dynamics-The same pulsechase labeling technique was used to determine the effect of the antiglucocorticoid hormone RU486 on glucocorticoid receptor half-life in cells that were cultured under cortisolfree conditions (phenol red-free medium supplemented with charcoal-extracted fetal calf serum). Cells were pulse-labeled with [35S]methionine and RU486 was added to cultures at the beginning of the chase to give a final concentration of 0.1 pM. Based on earlier studies, this concentration of RU486 is sufficient to achieve receptor saturation. The recovery of labeled receptors from cells during the chase decreased at the same rate in the presence and absence of RU486, indicating that RU486 does not affect the intracellular half-life of the glucocorticoid receptor (Fig. 7). In contrast to dexamethasone, RU486 did not induce a rapid decrease in the recovery of labeled receptors from cells. Thus, unlike dexamethasonereceptor complexes, RU486-receptor complexes do not appear to associate tightly with the nucleus and are recovered in the cytosol fraction.

DISCUSSION
The mechanism of glucocorticoid receptor autoregulation has been the subject of investigation in a number of different laboratories. Several laboratories have reported that the level of glucocorticoid receptor mRNA in a variety of types of cells is decreased in response to glucocorticoids (31-33). It appears from these studies that glucocorticoids directly regulate transcription of the glucocorticoid receptor gene. On the basis of this information, one would predict that glucocorticoids decrease the rate of glucocorticoid receptor synthesis in cells. However, other investigators have recently reported that glucocorticoids upregulate the level of glucocorticoid receptor mRNA in human leukemia cells, suggesting that patterns of glucocorticoid receptor autoregulation may be different in different types of cells (15).
Other investigators have sought to understand glucocorticoid receptor autoregulation by investigating the effect of glucocorticoids on the rate of glucocorticoid receptor turnover in cells. Svec and Rudis (6) found that glucocorticoid receptors in AtT-20 cells are depleted more rapidly when cells are incubated with both cycloheximide and dexamethasone than when cells are incubated with cycloheximide alone, suggesting that the half-life of the glucocorticoid receptor is shorter in the presence of hormonal ligand than in the absence of hormonal ligand. However, cycloheximide may disrupt normal metabolic processes in the cell and may itself alter the rate of intracellular receptor degradation (34). The effect of glucocorticoid hormones on intracellular receptor half-life have been extensively investigated by Samuels and coworkers using the technique of dense amino acid labeling (8, 11). Their results indicate that glucocorticoid receptor half-life in GH1 pituitary tumor cells is significantly shorter when the cells are cultured in the presence of glucocorticoid hormone than when the cells are cultured in the absence of glucocorticoid hormone. A similar conclusion regarding the effect of estrogens on estrogen receptor half-life has been reached by Katzenellenbogen and coworkers (35) using similar techniques. However, one potential difficulty associated with measuring receptor half-life by the dense amino acid labeling technique is that radiolabeled hormonal ligands are required to detect and quantitate receptors. Thus, it may be difficult to discriminate between changes in the amount of receptor present in cells and alterations in the ligand binding properties of receptors. This is an important consideration since glucocorticoid receptors are able to reversibly convert from an "active" form that binds hormone to an "inactive" form that does not bind hormone (36)(37)(38)(39).
Because of potential limitations associated with present techniques for measuring glucocorticoid receptor half-life in cells, the present study was undertaken to measure intracellular glucocorticoid receptor half-life by means of a kinetic pulse-chase labeling method. This method has two advantages over previous methods. First, it does not involve use of protein synthesis inhibitors. Second, glucocorticoid receptors are synthetically labeled with [36S]methionine and quantitated by reaction with monoclonal antireceptor antibody. Thus, receptor quantitation does not require binding of radiolabeled hormonal ligand to the receptor, enabling detection of both active and inactive forms of the receptor as well as comparison of the effects of different hormonal ligands on receptor halflife.
The results of the present study indicate that the intracellular half-life of the glucocorticoid receptor in S49 mouse lymphoma cells is 9 h. Receptor half-life appears to be the same in both wild type, glucocorticoid responsive cells (S49.1), and a glucocorticoid resistant variant in which glucocorticoid receptors are unable to undergo nuclear translocation (S49.22r). Glucocorticoid receptor half-life was unaffected by culture conditions. Thus, receptor half-life was the same when cells were cultured in the presence of fetal calf serum that contains cortisol and when cells were cultured in the presence of fetal calf serum that had been extracted to remove cortisol. In addition, the presence or absence of phenol red in the culture medium did not affect receptor half-life.
Our studies also show that pharmacologic concentrations of a glucocorticoid hormone, dexamethasone, and an antiglucocorticoid hormone, RU486, do not affect receptor half-life in S49 mouse lymphoma cells. This conclusion is different from that of dense amino acid labeling studies discussed above in which glucocorticoid hormones were found to shorten the half-life of the glucocorticoid receptor. The discrepancy between the findings of the present study and earlier studies may relate to differences in the methods used to measure receptor half-life or may represent a difference in receptor regulation in different types of cells.
In summary, we find that the pulse-chase labeling technique is an effective method for measuring glucocorticoid receptor half-life in cells. Using this technique, we conclude that receptor half-life is not regulated by glucocorticoid or antiglucocorticoid hormones in S49 mouse lymphoma cells.