Identification of transformed glucocorticoid receptor from dexamethasone resistant melanoma

J. .srcrod Bidwm. Vol. 16. pp. 705 IO 71 I. I‘M Printed in Great Britain. All rights reserved

Copyr@

0022-4731 x2 060705-07503.00 0 0 19x2 Perfamon Press Ltd

IDENTIFICATION OF TRANSFORMED GLUCOCORTICOID RECEPTOR FROM DEXAMETHASONE RESISTANT MELANOMA* T. WILLIAM HUTCHENSt. EDWARD F. HAWKINS+,, + \: and FRANCIS S. MARKLAND JRt. From the tDepartments of Biochemistry. *Physiology + and Biophysics. and the ,;Comprehensive Cancer Center. University of Southern California, School of Medicine. Los Angeles, California 90033. U.S.A. SUMMARY

We have examined properties of glucocorticoid-receptor complexes which might account for dexamethasone induced alterations in the growth and morphology of melanoma target ceils. We have used cultured cells and solid tumors derived from the dexamethasone-sensitive RPM1 3460 Syrian hamster melanoma cell line together with clonal variants which are either more sensitive (c!one 6) or resistant (clone 5) to the growth inhibiting effec& of dexamethasone. Although differing markedly in their response to corticoids. each of the cell lines contains significant quantities of glucocorticoid receptor. The present studies were designed lo determine if differences in the transformability (nuclear binding ability) of the glucocorticoid receptors from these target cells could account for their sensitivity or resistance 10 glucocorticoids. Cytosolic glucocorticoid-receptor complexes were analyzed by DEAE-cellulose chromatography to identify and quantitate the relative amounts of native (unactivated) and transformed (activated-nuclear binding) receptors present. We could readily separate these two major forms of the glucocorticoid-receptor complex in cytosols from each of the melanoma cell lines and solid tumors examined. The identity of these receptor complexes as native or transformed was confirmed using both ATP-agarose and isolated nuclei: only transformed receptor complexes are bound. When cytosol is prepared in the presence of IO mM sodium molybdate. only native receptor is present. After molybdate is removed, native glucocorticoid receptor is readily transformed by increased ionic strength. We conclude that resistance to dexamethasone-induced changes in growth observed in resistant clone 5 cells and solid tumors cannot be attributed to an inability of the receptor they contain to exist as a stable. transformed complex.

INTRODUCI-ION

Details of the mechanism(s) by which steroid hormones affect the growth and morphology of hormonally-responsive cells are still not fully understood [I]. Although having received much attention, even the cytoplasmic events thought to precede changes at the genomic level remain unclear [2-4]. In an effort to ascertain the nature and contribution of receptor proteins during steroid hormone-induced alterations in cell growth, we have examined certain properties of cytosolic glucocorticoid-receptor complexes from RPM1 3460 Syrian hamster melanoma cells [S-8] and two clonal variant cell lines [9] together with solid tumors grown from these cells. Physiological concentrations of the synthetic glucocorticoid, dexamethasone. cause as least three biologically significant changes in RPM1 3460 cells: (1) growth inhibition, (2) decreased final cell density and (3) altered morphology [6-S]. Two clonal va’riants *This data was presented in preliminary form at the 72nd Annual Meering of the American Associorion for Cancer Research in Washington. DC.. April 26-29. 1981 (Abst. No. 32). To whom reprint requests should be addressed at: USC Cancer Center. Research Building One. Room 490. 1720 Zonal Avenue, Los Angeles. California 90033. U.S.A. ?I Hawkins E. F.. Hutchens T. W.. Fligiel S.. Horn D. and Markland F. S.: J. sreroirl Bindtent. 16 ( 1982) 673-68 I, \R

Ihh

A

705

isolated from RPM1 3460 have been shown to be either resistant (clone 5) or more sensitive (clone 6) than the parent cell line to growth inhibition by dexamethasone [9]. Similarly, dexamethasone inhibits growth of solid tumors derived from clone 6. but not clone 5, cells$ We have previously demonstrated that the glucocorticoid resistance of clone 5 cells [9] and tumorfl cannot be explained by a reduction in the quantity of cytosolic glucocorticoid receptor. Further examination of the cytosolic receptor from the different cell lines. and the solid tumors derived from them. could help determine if the resistance of clone 5 cells and tumors to growth inhibition by dexamethasone was due to an abnormality in the structure or function of the receptor. In general, the alteration of cytosolic glucocorticoid receptor activity towards nuclei in vitro is paralleled by a &crease in receptor sizd and a striking difference in the ability of the receptor to bind to several types of cationic exchange resins [3,10. 11-J. These findings have led many investigators to postulate the existence of two principal forms of glucocorticoid-receptor complex [all. 123. These two major forms of cytosolit glucocorticoid-receptor complexes can additionally be separated by DEAE-cellulose chromatography and shown to be either native or transformed (converted to the nuclear binding form) based on their ability to bind to chromatin or DNA [IL 133. Since a possible

706

T. WILLIAM HI:TCHENS et al.

mechanism for glucocorticoid resistance in a clonal variant of the CEM-C7 human leukaemic T-cell line was recently suggested to be the inability of its glucocorticoid-receptor complex to exist as a stable transformed complex [ 143, we have examined transformation of the glucocorticoid receptor in RPM1 3460, clone 5 and clone 6 melanoma cells and tumors using selected chromatographic techniques under various experimental conditions. EXPERIMENTAL

Materials

[6,7-3H(N)]Dexamethasone (47.5 Ci/mmol) was purchased from New England Nuclear, Boston, MA. Unlabelled dexamethasone was obtained from Steraloids, Wilton, N.H. Reagent grade Ttizma base, dithiothreitol (DTT), dithioerythritol (DTE), glycerol, mono- and dibasic potassium phosphate, sodium molybdate and ATP-agarose were purchased from Sigma Chemical Co., St. Louis, MO. DEAE-cellulose was purchased as DE 52 from Whatman, Inc., Clifton, NJ. Buffers

The buffer used for tissue homogenization and during subsequent characterization of the glucocorticoid receptor by column chromatography was 10mM Tris-HCl, 1 mM dithiothreitol or dithioerythritol and 25% (v:v) glycerol, pH 7.4, at 4°C (Buffer 1). The presence of 10 mM sodium molybdate in Buffer 1 is indicated where applicable. The glycerol content of buffers used to prepare nuclei was reduced to 10% (Buffer 2) with and without 10 mM sodium molybdate as indicated. Growth and harvesting of cells

’ RPM1 3460 cells were grown in monolayer culture as described earlier [6-S]. A detailed description of the isolation and characterization of clonal variants which are either resistant (clone 5) or more sensitive (clone 6) than the parental RPM1 3460 cell line to the growth inhibiting effects of dexamethasone appears elsewhere [9]. Growth and harvesting of solid tumors

A complete description of these procedures is to appear elsewhere.7 In brief, cells grown in monolayer culture were harvested, washed and resuspended in Hank’s balanced salt solution before injection into the dorsal flank region of 4-6 week old female Syrian hamsters. RPM1 3460, clone 5 and clone 6 tumors were harvested after 11-15 days of growth, cleaned and stored under liquid nitrogen until use. Preparation of cytosol

Frozen tissue and cells were stored in liquid nitrogen and pulverized at liquid nitrogen temperature using a Thermovac tissue pulverizer (Thermovac Industries Corp., Copiague, NY). The resulting

powder was mixed with 3-5 vol. of ice cold Buffer 1. Homogenization was carried out at 4’C using either a Polytron PT-1OST tissue homogenizer (Brinkman Instruments, Inc., Westbury, NY) or a glass Potter homogenizer with a motor driven Teflon pestle. In both cases, homogenization was interrupted at 10-20 s intervals for 1-2 min cooling periods in an ice-water bath. The homogenate was centrifuged for 1 h at 4°C in a Beckman L5-65 ultracentrifuge using a Ty65 rotor at 4O,OOOrev/min. The supernatant, cytosol, was decanted and filtered through two layers of Nitex bolting cloth (35 pm, Tetko, Inc., Elmsford, NY) before labelling with 20 nM [3H]-dexamethasone for 6-20 h at 04°C. Ion exchange chromatography DEAE-cellulose was precycled and equilibrated with Buffer 1 using procedures described by the manufacturer. All operations were carried out at 4°C. Cytosol previously labelled with C3H]-dexamethasone was applied to a small DEAE-cellulose column (2.6 x 2cm) and the column was washed with 3-5 column volumes of Buffer 1. Elution was achieved using a linear gradient of O-O.4 M KPO., in Buffer 1 at a flow rate of 50-60 ml/h. Fractions of 2 ml each were collected. Radioactivity was determined in alternate fractions using a Beckman model 3150T liquid scintillation counter (Beckman Instruments, Inc., Palo Alto, CA). Recovery of C3H]dexametbasone was 87-98x and counting efficiency varied from 34-37x. Conductivity in every fifth fraction was measured using a Radiometer conductivity meter (Radiometer, Copenhagen, Denmark), model CDM 2s with a type CDC 114 electrode. Receptor containing fractions eluted from DEAE-cellulose were desalted by Sephadex G-25 chromatography using the appropriate buffer prior to incubation with isolated nuclei or chromatography on ATP-agarose columns. ATP-agarose (ATP attached through the ribose hydroxyls to beaded agarose via a six carbon spacer) contained approximately 2 pmol of ATP per ml of settled gel. After washing with Buffer 1 containing 0.5 M KCl, the gel was poured into a 1.3 cm diameter column and packed to a volume of 2 ml. The packed gel was washed with 20-30 ml of Buffer 1 before use. Samples, previously labelled with [3H]dexametbasone were applied to the column and unbound radioactivity was eluted by washing with Buffer 1. Bound radioactivity was eluted with 0.4 M KCI in Buffer 1. Fractions of 1.0 ml each were collected and the radioactivity was determined as described above for DEAE-cellulose chromatography. Nuclear uptake and binding

Nuclei were prepared from tumor tissue stored in liquid nitrogen. Tissue was pulverized at liquid nitrogen temperature with the Thermovac tissue pulverizer, mixed immediately with 4 vol. of ice cold B&r 2 containing molybdate and homogenized using a Potter homogenizer with a mechanically driven Teflon

Receptors in corticoid resistant melanoma pestle. The homogenate was monitored by phase contrast microscopy to determine the amount of homogenization that was necessary to obtain maximum cell lysis with minimal damage to nuclei and other organelles. Centrifugation of the homogenate in a Beckman model TJ-6 centrifuge using a Beckman JS7.5 rotor at 2ooO rev./min (75Og at R,,,.) for 15 min resulted in the formation of a nuclear pellet. Using a Dounce homogenizer with a loose fitting m pestle. the nuclear pellet was resuspended 10-15 ml of Buffer 2 containing molybdate. then repelleted during each of 3 successive washes. Nuclei were incubated with [‘HI-dexamethasone alone or with C3H]-dexamethasone prelabelled receptor fractions in molybdate containing buffer for varying times at l5’C. The nuclei were then washed three times with ice cold Buffer 2 containing molybdate as described above. Nuclear bound radioactivity was extracted with 2 x 1 ml absolute ethanol and quantitated by scintillation counting. The DNA content of the nuclear pellets was determined according to the method of Burton[ 15-J. The above procedures were also carried out using Buffer 2 in the absence of sodium molybdate. Protein determination

Protein concentrations were determined using the Coomassie Brilliant Blue binding technique described by Bradford[16] employing a kit obtained from BioRad, Richmond, CA. RESULTS

Cytosolic glucocorticoid-receptor complexes from both dexamethasone-sensitive and -resistant melanoma appear similar to previously activated glucocor-

Fraction

707

rat liver and kidney [ 12, 13, 171 in being heterogeneous with respect to chromatographic properties on DEAE-cellulose. Figure I illustrates the typical chromatographic behavior observed upon chromatography of [3H]-dexamethasone labelled cytosols from resistant clone 5 cells (Fig. 1A) and the solid tumors derived from them (Fig. IB). Elution of adsorbed glucocorticoid-receptor complexes with a linear salt gradient resolves two major regions of glucocorticoid receptor activity in both cytosols; an early eluting region (designated peak I) and a late eluting region (designated peak II). Peak I and peak II receptor complexes were consistently found in nearly equal amounts in cytosols from both resistant clone 5 cells and solid tumors. While this also appears to be true for receptor in cytosols from sensitive clone 6 cells and tumors, interestingly. peak I is typically the predominant species (up to 70”,,) when RPMI 3460 cytosols are prepared under identical conditions (data not shown). The relative amounts of peak I and peak II glucocorticoid-receptor complexes can be altered by varying the experimental conditions. The presence of 10mM sodium molybdate in Buffer 1 during preparation of cytosol. equilibration of DEAE-cellulose and subsequent chromatography, dramatically alters the elution profile of the receptor. The profile shown in Fig. 2A using [3H]-dexamethasone labelled cytosol from resistant clone 5 tumors is typical of that obtained when both sensitive and resistant melanoma cells and tumors are similarly analyzed in the presence of molybdate. In each case peak I glucocorticoid-receptor complexes are completely absent and only the later eluting peak II complexes are observed. This absence is not caused by a physical interference of molybdate which prevents binding of peak I glucocorticoid-receptor complexes to DEAE-

ticoid receptors from

Number

Fig. 1. DEAE-cellulose chromatography of [3H]-dexamethasone labelled cytosols prepared from clone 5 cells (A) and sold tumors (B) derived from them. Cytosols were prepared in Buffer 1. Incubation of cytosols with [‘HI-dexamethasone at O’C and chromatography on DEAE-cellulose are as described in Experimental. Regions labelled I and II represent transformed and native receptor complexes. respectively.

708

T.

I\

WILLIAM

HUTCHENS

PO A.(Ml

C

I

PO4 04) .--. 0.4

500

exactly the same manner (data not shown). It has been our experience that the continued presence of 1OmM sodium molybdate is necessary to maintain glucocorticoid receptors as peak II complexes during purification by multiple consecutive chromatographic procedures. Indeed. whenever peak II glucocorticoidreceptors are passed through a Sephadex G-25 column equilibrated with Buffer I so that both salt and molybdate are removed. and are then rechromatographed on a second DEAE-cellulose column in the continued absence of molybdate. peak I receptor complexes become the predominant (sometimes only) species (Fig. 2C). When cytosol from RPM1 3460, clone 5 or clone 6 cells (or tumors) is incubated with C3H]-dexamethasone in the presence of a lOO-fold molar excess of unlabelled dexamethasone and chromatographed on DEAE-cellulose, the [3H]-dexamethasone binding activity of peak I and peak II is completely suppressed (data not shown). These experiments also showed that nonreceptor bound [‘HI-dexamethasone and/or free steroid elutes from the column before application of the salt gradient. Separate experiments have shown that free [3H]-dexamethasone is recovered quantitatively from DEAE-cellulose in fractions eluted prior to initiation of the salt gradient. Subsequent experiments were designed to determine the nuclear binding properties of the glucocorticoid-receptor complexes eluting at peak I or peak II

L_L!L_A

moo

i

t’t ul.

/

c- ‘/I;

0

(0

z10 30 40 00 00 70 00

FRACTION

0.3 0.2

0.1

NUMBER

Fig. 2. DEAE-cellulose chromatography of [‘HI-dexamethasone labelled clone 5 tumor cytosol. Both prep aration of cytosol and chromatography on DEAE-cellulose was with Buffer I containing 10 mM sodium molybdate (A). Peak II containing fractions were then pooled. desalted on Sephadex G-25 and rechromatographed on a second, then third DEAE-cellulose column (B) using Buffer I with IOmM sodium molybdate throughout. Peak II activity from (A) was also passed through a Sephadex G-25 column equilibrated with Buffer I (so that both salt and molybdate were removed) and then rechromatographed on DEAEcellulose (C). Procedures are as outlined in Experimental. Peaks labelled I and II represent transformed and native receptor complexes, respectively.

cellulose. When

peak 1 glucocorticoid-receptor

A

II

/

com-

by DEAE-cellulose chromatography in the absence of molybdate and are then exposed to 10mM molybdate and rechromatographed on a second molybdate equilibrated DEAEcellulose column, they are adsorbed and eluted as peak I glucocorticoid-receptor complexes (data not shown). Furthermore, since 10mM molybdate does not significantly alter the [3H]-dexamethasone binding capacity of these cytosols. it is unlikely that the absence of peak I receptor complexes is due to their inability to become labeled in the presence of molybdate. The peak II glucocorticoid-receptor complexes plexes are prepared

CPM -

tooo

and isolated

prepared with molybdate and isolated as described in Fig. ZA can be desalted and rechromatographed on a

second, then third DEAE-cellulose column using Buffer I with molybdate throughout. As seen in Fig. 2B for clone 5. under these conditions. native. peak II glucocorticoid-receptor complexes are preserved. Peak II glucocorticoid-receptor complexes isolated from RPM1 3460 and clone 6 cytosols behave in

0

soor

0

4

3

12

16

20

24

23 32

3

12

16

20

24

28 32

B_

4

FRACTION NUMBER

Fig. 3. (A), ATP-agarose chromatography of [‘HI-dexamethasone labelled peak I receptor isolated from clone 5 tumor cytosol by DEAE-cellulose chromatography. Buffer I was used throughout. After adding desalted sample, the column was washed with IO vol. of ButTer I before elution with 0.4 M KCI in Buffer I (as indicated by the arrow). Methods are as described in Experimental. (B), ATP-agarox chromatography of [‘HI-dexamethasone labelled peak II receptor isolated from clone 5 cytosol by DEAE-cellulose chromatography. In contrast to Fig. 3A. Buger 1 containing IOmM sodium molybdate was used throughout. The procedure for ATP-agarose chromatography was as described for Fig. 3A and in Experimental.

709

Receptors in corticoid resistant melanoma positions from DEAE-cellulose. We have demonstrated that peak 1 glucocorticoid-receptor complexes represent transformed. nuclear binding (activated) receptor species. As shown in Fig. 3A for the case of solid tumor. peak I gluc~orti~oid-r~ptor complexes from both clone 5 cell and tumor cytosols bind to ATP-agarose. Peak 1 receptor complexes generated from molybdate stabilized peak II receptor complexes (as described for Fig. 2C) also bind to ATP-agarose. Peak I receptor complexes isolated from RPM1 3440 and clone 6 tumor cytosols also bind to ATP-agarose (data not shown). Molybdate does not interfere with this binding. In contrast. peak II glucocorticoidreceptor complexes isolated by DEAE-cellulose chromatography, and maintained as such by including lOmM molybdate in the buffer. do not bind to ATP-agarose (Fig. 3B). Uptake and binding of glucocorticoid-receptor complexes to isolated nuclei was also measured to distinguish transformed from native glucocorticoidreceptor complexes. Results shown in Table 1 reveal Table I.

Uptake of isolated ~lucocorticoid receptor fractions by nuclei itt rifro

Experiment A

Source Clone

5

Fraction added

Percent bound

to nuclei

to nuclei*

Peak I Peak II

(+

Cytosol

MoD,f

Free steroid Clone 6

3

Clone 5

I -MOO,)

1 1

Peak I Peak II Cytosol Free steroid

60+6 1

Peak I Peak II

50 + 4

Cytosol Clone 6

61 & 3 1

1 I

t

+ +

Free steroid

I

Peak I Peak II Cytosol Free steroid

52 + 3 t :

I

Recombination of various glucocorticoid receptor containing fractions (240-260 fmol) with excess isolated nuclei (500 p(e DNA per incubation). In part A with the exception of the actual transformation of glucocorticoid-receptor complexes (generation of peak I and peak if). all phases of each experiment including isolation of nuclei, contained 1OmM sodium moiyhdate. In part B all phases of each experiment were performed in the absence of sodium molybdate. Methods for isolation of receptor peaks I and II by DEAE-cellulose chromatography as well as preparation of isolated nuclei are as described in Experimental. Data obtained using receptor fractions isolated from cell and tumor cytosols were similar and combined. * Values represent the mean f SEM of 2-4 determinations in duplicate. + Peak II glucocorticoid-receptor complexes cannot be isolated and maintained as such without molybdate. 2 In contrast to cytosol prepared as for part A. cytosol prepared in the absence of molybdate is a heterogeneous mixture of Peak I und Peak II and was therefore not included here.

that under conditions of nuclear excess, isolated peak II glucocorticoid-receptor complexes are not bound. Similarly, when cytosol is prepared in Buffer 1 plus 10mM molybdate and prelabelled with [3HJ-dexamethasone, the resulting native (peak II) gfucocotticoid-receptor complexes do not bind to isolated nuclei. In contrast, approximately N-60”/, of transformed (peak I) glucocorticoid-receptor complexes reproducibly bind to isolated nuclei regardless of the presence or absence of 1OmM sodium molybdate. The ability of peak 1 receptor complexes to bind to isolated nuclei appears to be independent of whether these complexes were isolated directly from cytosol by DEAE-cellulose chromatography or generated from molybdate stabilized peak 11 receptor complexes as described for Fig. 2C. When equal amounts of receptor isolated as peak I gluc~orticoid-re~ptor complexes from either clone 5 or clone 6 cytosols are added to equivalent aliquots of their own isolated nuclei. the same percentage of peak I glucocorticoidreceptor complexes binds in each case. Separate experiments revealed that [3Hf-dexamethasone does not bind to nuclei under these conditions.

DISCUSSION

Cytosolic glucocorticoid-receptor complexes have been repeatedly shown to be heterogeneous with respect to size. charge, and nuclear binding capacity. This appears to be true for glucocorticoid receptors prepared from both normal and malignant tissues [ 10,12,13, IS-201. The ability to resolve heterogeneous. cytosolic gluc~orti~id-r~eptor complexes into native (unacdvated) and transformed (activated) components by DEAE-cellulose chromatography has facilitated studies aimed at elucidating the relative contribution of these two receptor species to the mechanism(s) by which glucocorticoids affect their target cells. This technique has recently been used for analysis of glucocorticoid receptor from the human lymphoid CEM-C7 cell line and its glucocorticoid resistant 4R4 clonal variant [14]. In sharp contrast to the glucocorticoid sensitive C7 clone, glucocorticoid receptor f.:pared from the resistant 4R4 subclone did not exist as a stable, transformed species. Indeed. giucocorticoid resistance of the 4R4 subclone was attributed to its inability to produce (or sustain) a transformed (activated) glucocorticoid-receptor complex. Using DEAE-cellulose chromatography. we have demonstrated the presence of stable, transformed glucocorticoid-receptor complexes (peak I) in cytosols prepared from gluc~orticoid resistant (clone 5) melanoma celis and tumors. Furthermore. as evidenced by Figs 1A and IB. the quantity of receptor eluting from DEAE-cellulose in the position of transformed receptor is substantial, typically representing 50”” of the total glucocorticoid receptor present. Similarly. cytosots prepared from parental RPM1 3460 cells. gluco-

710 corticoid-sensitive

T. WILLIAM HUTCHENSet al clone 6 cells and solid tumors

de-

rived from each of them, all contain significant quantities of peak I glucocorticoid-receptor complex. It is interesting that such significant quantities of peak I appears without prior (intentional) activation of the cytosol. We are aware of the many in vitro manipulatons which can lead to glucocorticoid receptor activation [ 111. However, these conditions have been intentionally avoided prior to DEAE-cellulose chromatography. Nevertheless, we believe that the significant percentage (up to 70%) of receptor eluting from DEAE-cellulose in the peak I position may reflect an unusual succeptibility of these receptors to undergo transformation in vitro. In our laboratory, glucocorticoid-receptor complexes from lactating goat mammary tissue have been extensively characterized*. Cytosol from this tissue is regularly prepared and analyzed by DEAE-cellulose chromatography in a manner identical to that described in Fig. 1 for melanoma cells and tumors. Interestingly, and in sharp contrast to results shown in Fig. I, only native peak II receptor complexes are routinely observed with the mammary glucocorticoid receptor*. It is possible that after saturation with [‘Hl-dexamethasane, receptor from the melanoma cells. being more susceptible to transformation in vitro than the glucocorticoid receptor from goat mammary, undergoes transformation as a result of exposure to relatively high concentrations of basic amines while binding to DEAE-cellulose. We are continuing to investigate the mechanism(s) by which such a consistently large percentage of glucocorticoid-receptor complexes from these melanoma target cells becomes activated. Isolated native peak II receptor complexes can be readily transformed with increased ionic strength to peak I complexes. The transformed status of these peak I glucocorticoid-receptor complexes was confirmed by demonstrating their ability to bind both to ATP-agarose and to isolated nuclei. Activation has been shown to be an absolute requirement for the with interaction of steroid-receptor complexes in contrast to results ATP [21]. Furthermore, obtained using DNA-cellulose and phosphocellulose [l 1J, the interaction of activated steroid-receptor complexes with ATP-Sepharose closely parallels their interaction with isolated nuclei [21]. The identity of DEAE-cellulose peaks I and II as transformed and native glucocorticoid-receptor complexes, respectively, is particularly evident when these complexes are analyzed in the presence then absence of sodium molybdate. When cytosols, from all three cell lines and tumors derived from them, are prepared with 10mM sodium molybdate, subsequent analysis by DEAE-cellulose chromatography reveals only peak II glucocorticoid-receptor complexes. Furthermore, these complexes do not bind to ATP-agarose or isolated nuclei. The receptor can be maintained in the * Hutchens T. W. and Markland publication.

F. S.: Submitted for

peak II form by the continued presence of molybdate, even during subsequent rechromatography on DEAEcellulose. It would appear that the inclusion of 10 mM sodium molybdate prevents the transformation of native (peak II) receptor to the activated (peak I) form (Fig. 2) during cytosol preparation or DEAE-cellulose chromatography. When molybdate is removed, native peak II receptor complexes are readily converted to transformed peak I receptor complexes either by rechromatography on DEAE-cellulose or by increased ionic strength. The elution properties of partially-purified, transformed, peak I glucocorticoid-receptor complexes appear resistant to any further ionic strength dependent changes in the presence or absence of 1OmM sodium molybdate. Separate experiments have also shown that 10mM sodium molybdate does not interfere with the interaction between peak I receptor complexes and ATP-agarose or isolated nuclei. In fact, our data (and those of others) indicate that binding of activated receptor complexes to isolated nuclei may be enhanced by the ‘presence of sodium molybdate [22]. We conclude that the resistance of clone 5 cells and tumors to growth inhibition by dexamethasone is not correlated with the inability to form stable transformed receptors or the inability of those receptors to bind to or be retaiined by nuclei in vitro. Furthermore, the use of molybdate together with DEAE-cellulose and ATP-agarose chromatography and nuclear binding studies appears to be an excellent way to prevent the transformation of glucocorticoid-receptor complexes which otherwise can occur as a result of these procedures.

Acknowledgements-This study was supported in part by PHS Grant CA 22910 from the National Cancer Institute, DHHS to F. S. Markland and by a USC Biomedical Research Support Award to E. F. Hawkins and by a generous donation from Mr and Mrs W. F. Fisher of Arcadia. California. We are especially grateful to Ms. Dieumy Thai for technical assistance and to Dr D. Horn and Mr R. L. Buzard of the USC Department of Biology for gowing cells for portions of these experiments.

REFERENCES

Yamamoto K. R. and Alberts B. M.: Steroid receptors: elements for modulation of eukaryotic transcription. Ann. Rev. Biochem. 45 (1976) 721-746. Gschwendt M.: The general validity of the subunit model of the progesterone receptor from chick oviduct appears questionable. Comparison of progesterone and estrogen receptor. Molec. cell. Endocr. 19 (1980)57-67. Bloom E.. Matulich D. T., Lan N. C., Higgins S. J., Simons S. S. and Baxter J. D.: Nuclear binding of glucocorticoid receptors: relations between cytosol binding, activation and the biological response. J. strroid Eiochmm.I2 ( 1980) I75- 184. 4. Barnett C. A.. Schmidt T. J. and Litwack G.: Elects of calf intestinal alkaline phosphatase. phosphatase inhibitors, and phosphorylated compounds on the rate of activation of glucocorticoid-receptor complexes.

Biochemistry 19 (1980) 5446-5455.

Receptors in corticoid resistant melanoma 5. Moore G.: In uitro cultures of a pigmented hamster melanoma cell line. Exp. Cell Res. 36 (1964) 422-423, 6. Hawkins E. F.. Horn D. and Markland F. S.: Recep tors for glucocorticoids in RPMI 3460 melanoma cells. Cancer Res. 40 ( 1980) 2 174-2 178. 7. Markland F. S. and Horn D.: Steroid hormone receptor studies in melanoma model systems. J. supramolec. Struct. 13 (1980) 35-46. 8. Horn D. and Buzard R. L.: Growth inhibition by glucocorticoids in RPM1 3460 melanoma cells. Cancer Res. 41 (1981) 3155-3160. 9. Buzard R. L., Hutchens T. W.. Hawkins E. F., Markland F. S. and Horn D.: Clonal variants of a melanoma cell line sensitive to growth inhibition by dexamethasone. Exp. Cell Res. in press (1982). separIO. Atger M. and Milgrom E.: Chromatographic at& on phospho&lulose of activated- and nonactivated forms of steroid-receptor complex. Purification of the activated complex. Biochemistry 15 (1976) 4298-4304. Il. Bailly A.. Atger M.. Lefevre B.. Savouret J. F. and Milgrom E.: Activation of steroid-receptor complexes. In Pharmacological Modulation qf Steroid Action (Edited by E. Genazzani et al.). Raven Press, New York (1980) pp. 181-186. 12. Parchman L. G. and Litwack G.: Resolution of activated and unactivated forms of the glucocorticoid receptor from rat liver. Arch. hiochem. Biophys. 183 (1977) 374-382. I3 Sakaue Y. and Thompson E. B.: Characterization of two forms of glucocorticoid hormone-receptor separated by DEAE-cellulose chromacomplex tography. Biochem. hiophy.t. Rex Comnwn. 77 (1977) 533-541.

711

14. Schmidt T. J., Harmon J. M. and Thompson E. B.: ‘Activation-labile’ glucocorticoid-receptor complexes of a steroid-resistant variant of CEM-C7 human lymphoid cells. Narure 286 (1980) 507-510. 15. Burton K.: Determination of DNA concentration with diphenylamine. Merhods Enrymol. 128 (1968) 163166. 16. Bradford M. M.: A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principal of protein-dye bindinn Analvt. BioC/&I. 72.( 1978) 248-i54. . 17. Markovic R. D. and Litwack G.: Activation of liver and kidney glucocorticoid-receptor complexes occurs in vioo. Arch. biochem. Biophys. 202 (1980) 374-379. 18. Stevens J. and Stevens Y. W.: Physicochemical differences between glucocorticoid-binding components from the corticoid-sensitive and -resistant stiains of mouse lymphoma P1798. Cancer Res. 39 (1979) 401 I-4021. 19. Stevens J., Eisen H. J.. Stevens Y.-W.. Haubenstock H., Rosenthal R. L. and Artishevsky A.: lmmunochemical differences between glucocorticoid receptors from corticoid-sensitive and -resistant malignant lymphocytes. Cancer Res. 41 (1981) 134-137. 20. Sherman M. R., Pickering L. A., Rollwagen F. M. and Miller L. K.: Meroreceptors: proteolytic fragments of receptors containing the steroid-binding site. Fedn. Proc. 37 (1978) 167-173. 21. Miller J. B. and Toft D. 0.: Requirement for activation in the binding of progesterone receptor to ATPsepharose. Biochem. 17 (1978) 173-177. 22. Auricchio F. and Migliaccio A.: In vitro inactivation of oestrogen receptor by nuclei. Prevention by phosphatax inhibitors. FEBS Left. 117 (1980) 224-226.