[CANCER RESEARCH 43, 3865-3873, August 1983]

Glucocorticoid-resistant Human Acute Lymphoblastic Leukemic Cell Linewith Functional Receptor1

Robert Zawydiwski,2 Jeffrey M. Harmon,3 and E. Brad Thompson4

Laboratory ol Biochemistry, National Cancer Institute, NIH, Bethesda, Maryland 20205

ABSTRACT

A receptor-containing, steroid-resistant clone of CEM cells,CEM-C1, was isolated without selective pressure from the wild-

type population. The biological and physicochemical propertiesof glucocorticoid receptors in CEM-C1 cells were compared tothose from a clone (CEM-C7) sensitive to glucocorticoid-me-diated lysis. In a whole-cell binding assay, CEM-C1 cells exhibitedhigh affinity for [3H]dexamethasone (K,, 22 nw), nuclear trans-

location of steroid:receptor complex (nt, 43%) and were foundto contain, on the average, 12,000 receptor sites/cell (R0).Thesesteroid-binding parameters were similar to those displayed bywild-type CEM-C7 cells: «<,,19 nw; nt, 47%; and R0, -14,000sites/cell. The ion-exchange and gel permeation profiles wereindistinguishable from those of identically treated CEM-C7 cyto-sols. Thus, diethylaminoethyl cellulose chromatography of CEM-C1 cytosol showed that [3H]triamcinolone acetonide:receptor

complex was eluted at 50 mw phosphate and 220 rriM phosphateunder "activating" and "nonactivating" conditions, respectively.

Receptor complex of activated CEM-C1 cytosol bound to DNA-

cellulose and was eluted at 100 mw salt. Filtration of unactivatedCEM-C1 cytosol over Sephacryl S-300 generated a single peakof radioactivity for receptor complex with a calculated Stokes'

radius of 55 to 59 A. Dexamethasone induced glutamine synthe-tase in CEM-C1. The dose dependence (50% effective dose,~20 nw) and maximal fold increase (1.9,1 ¿¿Mdexamethasone)were comparable to those observed in CEM-C7. Since CEM-C1

cells contain apparently normal, functional cytosolic receptor, theresults suggest that resistance to glucocorticoid in these cellsinvolves a defect(s) at another locus.

INTRODUCTION

Certain immature lymphoid cells are sensitive to the lyticeffects of glucocorticoids (3). This response not only forms partof the basis for clinical use of corticosteroids in the treatment oflymphoproliferative disorders but also provides an attractivesystem in which to study the mechanism of steroid hormoneaction through the isolation and characterization of unresponsivevariants. In principle, such an analysis would be expected toyield mutants with defects in each step of the lytic pathway.Almost without exception, however, resistant variants selectedfrom sensitive clones have proven to contain receptor abnormalities (6, 48, 64). A few mouse clones with possibly normalreceptors have been described (64).

We have made use of the cultured T-cell ALL5 line, CEM, as

a model for study of human leukemia (20, 70, 71). We haveobserved that most of various unselected clones are killed byglucocorticoids (46).6 One sensitive clone, CEM-C7, has been

used to show that these cells possess glucocorticoid receptorsthe occupancy of which correlates with direct cytotoxicity ofpotent glucocorticoids (46), that cell kill does not begin until atleast 24 hr after addition of steroid and is coincident with irreversible Gìarrest (19), that resistance develops by spontaneousindependent events (22), and that resistant clones have residualbut aberrant receptors (21, 57). No resistant subclones withnormal receptors have yet been found after steroid selection,with or without prior mutagenesis. However, such cells wouldbe, in theory, of great value in studying the biochemistry andgenetics of steroid action and resistance in general and in humanleukemia in particular. Consequently, we studied CEM-C1, aunique clone of dexamethasone-resistant CEM cells found

among the original set of unselected clones derived from CEM(46).

We show here that CEM-C1 cells are karyologically, and by

membrane markers, almost certainly derived from the sameoriginal leukemic cell source as are CEM-C7. The steroid-resistant CEM-C1 cells contain glucocorticoid receptors physicallyidentical to those of wild-type cells by several tests (tests chosen

for the most part because in one or another murine system theyhave revealed receptor abnormalities in resistant cells). Furthermore, the receptors of CEM-C1 cells appear to function, in that

they mediate the steroidal induction of GS, which we have shownto correlate with the occupancy of normal receptors by activesteroids (23). Therefore, the steroid resistance of CEM-C1 cellsappears to be due to a "nonreceptor" lesion. Because they are

so similar to their sensitive clone, CEM-C7, CEM-C1 cells may

prove to be useful in the biochemical and genetic analysis ofsteroid-induced cell lysis.

MATERIALS AND METHODS

Steroids and Chemicals. [1,2-3H]DEX (20 Ci/mmol) and [1,2,4-3H]TA

(29 Ci/mmol) were obtained from Amersham/Searle Corp., ArlingtonHeights, III. Nonradioactive steroids were purchased from Sigma Chemical Co., St. Louis, Mo. Other chemicals, unless indicated otherwise,were of reagent grade and were obtained through either Sigma or FisherScientific Co., Fair Lawn, N. J.

Cells and Cell Culture Methods. CCRF-CEM (12) is a human ALLline of T-cell lineage (31, 39); CEM-C1 and CEM-C7 are DEX' and DEXS

1Presented in part at the Eleventh Annual UCLA Symposia, session on Rational

Basis for Chemotherapy, April 18 to 23, 1982, Keystone, Colo. (79).2 Recipient of a Visiting Fellowship from the National Cancer Institute.3 Present address: Department of Pharmacology, Uniformed Services University,

of the Health Sciences, 4301 Jones Bridge Road, Bethesda. Md. 20814.4To whom requests for reprints should be addressed.

Received February 11,1983; accepted April 5, 1983.

5The abbreviations used are: ALL, acute lymphoblastic leukemia; GS, glutaminesynthetase; DEX'-5, dexamethasone (9-fluoro-16«-methyl-110,17,21-trihydroxy-1,4-pregnadiene-3,20-dione)-resistant or -sensitive phenotypes; TA, triamcinoloneacetonide (11 /3,21-dihydroxy-9-fluoro-l 60,17a-isopropylidenedioxy-1,4-pregna-diene-3,20-dione); HBSS, Hanks' balanced salt solution; KPD, 5 mw potassium

phosphate:0.5 mw dithiothreitol, pH 7.2 to 7.4.6 R. Zawydiwski and E. B. Thompson, unpublished observations.

AUGUST 1983 3865

on March 3, 2019. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Zawydiwski et al.

clones independently isolated from the parent line (46). Cells were grownas stationary-suspension cultures at 37°in a humidified atmosphere of

95% air:5% CO2 in Roswell Park Memorial Institute Medium 1640 (NIHMedia Unit or Biofluids, Inc., Rockville, Md.) supplemented with 10% (v/v) heat-inactivated fetal calf serum (Biofluids, Inc., or North American

Biologicals, Miami, Fla.) or with 10% (v/v) of a 50:50 mixture of fetal andnewborn calf sera (Biocell Laboratories, Carson, Calif.). Cells were maintained in logarithmic growth by manual dilution to densities between 105and 106 cells/ml, as determined by a Model ZB Coulter Counter (Coulter

Electronics, Inc., Hialeah, Fla.). Size distributions were made by a CoulterChannelyzer calibrated with polystyrene microspheres and coupled toan X-Y recorder. Cell viability was measured by the exclusion of trypanblue (49) and by the release of 51Cr (24) essentially as previously

described.Cytogenetics. Chromosomal analyses of metaphase spreads from

Colcemid-arrested cells were performed by conventional Giemsa staining, G-banding (60), and Giemsa-11 (76) banding techniques.

Whole-Cell Steroid-Binding Assay. Specific binding of | 'H|DEX was

determined in cells harvested in midlogarithmic growth (57). Results wereplotted according to the method of Scatchard (54), and kinetic parameters («<,,ño)were calculated following linear regression of the data bythe method of least squares. Nuclear translocation of receptor complexwas measured by a minor modification (58) of the procedure of Munckand Wira (43).

Preparation, Labeling, and Activation of Cytosols. Preparative procedures were performed at 0-4°. Cells in exponential growth (1 to 2liters, 4 to 8 x I05/rnl) were harvested, washed once in 15 ml ice-cold

HBSS, and homogenized (1.5 volumes) in KPD buffer with a TenBroeckground-glass homogenizer (2 series of 10 strokes, separated by a 30-sec interval). "Cytosols" were isolated by centrifugation of the homoge-

nate for 1 hr at 50,000 rpm (type 65 rotor) in a Beckman Model L5-65

refrigerated ultracentrifuge and collection of the supernatant fraction.Cytosols were labeled with [3H]TA at 5 x 10~8 M for 2 hr. At this

concentration, specific glucocorticoid-bound (determined as the difference in cpm between parallel incubations with and without a 200-fold

excess unlabeled TA) typically exceeded 80% of total bound radioactivity;cytosols contained 0.2 to 1 pmol receptor per mg protein. "Activation"

of receptor complex was achieved by heat-salt treatment (20°, 0.2 M

KCI). Free steroid was removed either by passage of labeled cytosolsover KPD-washed Sephadex G-25M columns (PD-10; Pharmacia FineChemicals AB, Uppsala, Sweden) or by adsorption onto dextran-coated

charcoal.Chromatography. All Chromatographie procedures were carried out

at 4°.Anion-exchange Chromatography was performed on DEAE-cellu-

lose essentially as described previously (53, 57). Briefly, cytosols (0.8ml) were applied to Whatman DE52 (W. & R. Baiston, Ltd., Maidstone,England) columns (1.2x3 cm, 3-ml bed volume) equilibrated with 5 ITIM

KPD. After extensive washing, bound radioactivity was eluted by a linear5 to 400 ITIM KPD gradient, and fractions of approximately 1 ml werecollected. Aliquots (0.4 ml) of each fraction were added to 10 ml AquasolUniversal LSC cocktail (New England Nuclear, Boston, Mass.) andcounted in a Beckman Model LS 9000 at an efficiency for 3H of about

50%. Salt concentrations were determined from electroconductivity readings (Radiometer Conductivity Meter, Copenhagen, Denmark).

DNA-cellulose Chromatography was carried out on a matrix (1-ml bed

volume) prepared (1) from calf thymus DNA (Miles Laboratories, Inc.,Elkhart, Ind.) and Whatman CF 11 cellulose powder (W. & R. Baiston,Ltd.). Cytosols were loaded, columns were eluted, and fractions wereassayed as described above.

Molecular sizing of receptor was performed by gel filtration on Se-phacryl S-300 (Pharmacia Fine Chemicals, Inc., Piscataway, N. J.).Cytosols (0.4 ml) were applied to a 1- x 49-cm column (38-ml bed volume)

equilibrated with 10 mw KPD buffer modified to contain 0.3 M KCI, 1 mwEDTA, 50 HIM Na2MoO4, and 10% (v/v) glycerol (elution buffer). Thecolumn was eluted at a rate of 6 to 8 ml/hr; 1-ml fractions were collected

and monitored for radioactivity. The column was calibrated with (A):

thyroglobulin, 85.0; ferritin, 61.0; catalase, 52.2; aldolase, 48.1; albumin,35.5; and ovalbumin, 30.5; all chemicals were obtained from Pharmacia.Void volume was determined with Blue Dextran 2000. The coefficient ofvariation for replicate column runs, determined with [mef/iy/-'4C]albumin(20 /jCi/mg; New England Nuclear), was ±2% (S.D.). Stokes' radii for

glucocorticoid: receptor complexes were estimated from a plot of (-logKav)'"versus molecular radius (33, 65) which was linear (r > 0.99) for the

range of proteins used (M, 43,000 to 669,000).Glutamine Synthetase Activity. Glutamine synthetase (EC 6.3.1.2)

was assayed spectrophotometrically (59) by measuring glutamyltrans-ferase activity essentially according to the method of Thorndike and Reif-Lehrer (72). Enzyme specific activity was expressed as nmol -y-gluta-

mylhydroxamate formed per min per mg protein and was computed fromthe "effective" molar extinction coefficient (calculated each experiment)

of L-glutamic acid-7-monohydroxamate (Sigma) measured under the

conditions of the assay. Protein was determined by the method of Lowryet al. (35) with bovine serum albumin (Sigma) as the standard.

Steroid Metabolism. Cultures, seeded at a cell density of 105 cells/ml, were treated with 1 ^M DEX containing tracer [3H]DEX (5 ¿iCi;2.5 x10~8 M final concentration) and incubated for 48 hr. Cells were collected

by centrifugation, washed once with HBSS, resuspended in distilledwater (2 ml), and sonically disrupted (Braun-sonic Model 1510; B. BraunInstruments, San Francisco, Calif.) with two 10-sec bursts at 15 watts.

The sonically disrupted cells were extracted 3 times with 5 ml ethylacetate. The organic layers were dried over anhydrous NazSCv and wereevaporated to dryness with argon (<5 ppm oxygen). The extracts werechromatographed in chloroform:ethanol (4:1) on Silica Gel GF plates (250urn thick; Analtech, Inc., Newark, Del.) previously washed (air dried) indeveloping solvent. The plates were scraped at 5-mm intervals, and the

silica was suspended in Aquasol LSC and counted. Supernatant mediaand cell washes were pooled and treated in an identical manner. Anequivalent amount of labeled and unlabeled DEX added to cell-freeincubation medium served as a control for serum-mediated steroid

transformation and/or artefacts of extraction and drying.Assay of Plasma Membrane Fragility. Cells were pulse-labeled (20

nCi/ml suspension, 2.5 hr, 37°) with Na251CrO4 (379.3 »jCi/mg; New

England Nuclear), harvested, then washed and resuspended in HBSS(¡i,0.158, calculated) at 2 x 107/rnl. Aliquots of 0.5 ml were diluted 1:20

with NaCI solutions of varying hypotonicity (fortified with MgCI2 to a finalconcentration of 1.5 mM, for maintenance of nuclear integrity), and wereallowed to stand at room temperature for 10 min. Portions were removedfor determination of cell viability by dye exclusion and 51Cr release.

Radioactivity was counted in both the supernatant and pellet fractions ina Packard Prias PGD Auto-Gamma counter (about 2% efficiency). Counts

were expressed relative to maximum label released and were correctedfor spontaneous loss of radiochromium.

Surface Membrane Properties. Binding of monoclonal antibodieswas measured by indirect immunofluorescence. Cells (100 /il, 107/ml)

were suspended in HBSS containing 0.1% albumin:0.1% NaNa andincubated with antibody («HuLyt-1,-2,-3 and «Hula; New England Nuclear) at a final dilution of 1:15 for 40 min at 0°.Cells were washed twice,

and the wet pellet was equilibrated for an additional 40 min with 10 ^lsaturated solution of fluorescein isothiocyanate-conjugated:goat anti-rabbit IgG, F(ab')2 fragment (provided by Dr. Richard Asofsky, National

Institute of Allergy and Infectious Diseases). Stained cells were washedand resuspended in supplemented HBSS (400 p\) prior to flow microfluo-rometric analysis on a FACS II (Becton-Dickinson FACS Systems, Sun

nyvale, Calif.) and a POP 11/40 computer (Digital Equipment Corp.,Marlboro, Mass.) (74).

Erythrocyte rosette formation was assayed according to the methodof Weiner ef a/. (75) with neuraminidase (Sigma)-treated sheep RBC

(Veterinary Resources Branch, NIH). Detection of soybean agglutininreceptor was performed by quantitation of the fraction of cells agglutinated by soybean lectin (P-L Biochemicals, Inc., Milwaukee, Wis.) in the

presence of homologous erythrocytes (Clinical Center Blood Bank, NIH)(51).

3866 CANCER RESEARCH VOL. 43

on March 3, 2019. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Glucocorticoid Resistance in Human ALL Cells

RESULTS

Whole-Cell Steroid-Binding Activity. CEM-C1 cells, which

were isolated in the absence of steroid selective pressure, growin suspension culture in 1 ^M DEX without a reduction in celldensity (46). Our initial cell-free binding studies revealed specificand saturable binding of [3H]DEX and indicated that resistance

to the lytic effect of glucocorticoid was not due to the lack of ordecreased affinity for cytosolic receptor. In order to determinewhether resistance was the result of a defect at the level ofhormone uptake and/or transport, we measured steroid bindingby whole cells. The data, summarized in Table 1, show that DEXr

CEM-C1 cells accumulated steroid in both the cytosolic and

nuclear compartments to an extent slightly less than and withan apparent affinity comparable to sensitive CEM-C7 cells. In 2

parallel experiments, total cellular receptor sites were indeedfound to be slightly lower in CEM-C1 (>80% of sites in CEM-

C7). This observation is consistent with the reduced level ofcytoplasmic binding of glucocorticoid found in cell-free extractsof CEM-C1 (46). Both clones exhibited a 2-fold range in DEX-

binding sites (cf. 9,000 to 23,000 and 8,000 to 19,000 sites/cellfor DEXS and DEX' cells, respectively) when determined over a

period of about 1 year. The large experimental deviation ispossibly due to receptor regulation by variable levels of endogenous hormone (8,10, 40, 55, 61, 67) and/or other components(27) present in different batches of serum.

Physicochemical Properties of Receptor. Although CEM-C1receptor was able to bind DEX and undergo "activation" as

evidenced by nuclear accumulation, the possibility that it wasnonetheless aberrant, generating a complex incapable of association with specific nuclear acceptor sites, was subsequentlyconsidered. Presumably, such a receptor would possess alteredbiochemical and biophysical properties. We therefore analyzedreceptor by ion-exchange and gel permeation chromatography.

Receptor abnormalities in murine and human systems have beendetected by each of these techniques (37, 57, 66, 77).

DEAE-cellulose chromatography of glucocorticoid:receptor

complexes of various cell types resolves the complexes into 2forms: a low-salt eluting form (Peak I) which binds to DNA, nuclei,and chromatin and is thought to represent the "activated" form

of the hormone:receptor complex, and a high-salt eluting form

(Peak II) which is devoid of these activities and probably represents the "unactivated" form of the complex (42, 53). Chart 1depicts typical elution profiles of [3H]TA:receptor complexes

formed under activating and nonactivating conditions. Peaks I

Table 1Parameters of whole-cell steroid binding

Approximately 10' cells were incubated in eachdeterminationwith 5-10 differentconcentrations (1 to 100 nw) of [3H]DEXin the presence (50 UM)and absence ofunlabeledDEX at 37°for 1 hr. Results were expressed as specific [3H]DEXbound

and were determined as the difference in radioactivity between noncompeted andcompeted (excess unlabeled DEX) cells. Cell-binding constants were computedfollowing analysis of the data as described in "Materials and Methods"; calculation

of total binding sites per cell was based on an assumed 1:1 stoichiometry betweenDEX and receptor. Nuclear translocation (determined at 50 nM [3H]DEX±a 200-fold excess DEX)represents the percentageof specific cell counts associated withthe nucleus.

15 rCEMC1

CloneCEM-C1

CEM-C7fio

(sites/cell)12,301±4,313a (5f

14,244 ±6,382 (4)K,,(nM)22±5 (5)

19 ±2 (4)Nuclear

translocation43

±11 (3)47 ± 9(3)

SOr

I

P 25igo

5 10 15 20 25 30

FRACTION NUMBER

CEM-C7

0.4

0.3

0.2

0.1

5 10 15 20 25

FRACTION NUMBER

30

Chart 1. DEAE-cellulosechromatography of pristine (O) and activated (•)receptor complexes. Cytosols were prepared, labeledwith [3H]TA,activated (20 min),and then desalted and freed of unbound steroid by filtration over Sephadex G-25prior to chromatography as described in "Materials and Methods." cpm retained

on the column after washing with 5 mw KPD represent >70% (activated)and 90%(unactivated)of applied radioactivity for CEM-C1 cytosol (/A);Chart 1A. and >90%(both preparations) for CEM-C7 cytosol (B). Straight lines with small data pointsshow salt gradient.

and II of CEM-C1 characteristically were eluted at the same salt

concentrations (0.05 and 0.22 M potassium phosphate, respectively) as were the corresponding peaks for the cytoplasmicreceptor complexes of CEM-C7. In the experiment shown, CEM-

C1 cytosol, however, exhibited less residual Peak II activity(<10% compared to >30% for CEM-C7) and total recoverable

material (cf. 40 to 60% versus >80%). Because these experiments were not performed in parallel, it was unclear whethersuch differences were significant and indicated a subtle structuraldefect in receptor of the DEXr cells or simply reflected biological

or experimental variation. To distinguish between these possibilities, we monitored the time course of receptor activation byanalyzing sequential binding of labeled cytosols to slurries ofDNA-cellulose, DEAE-cellulose, and hydroxylapatite. DNA binding thus detects activated receptor, DEAE-cellulose, residualunactivated holoreceptor, and hydroxylapatite binds merorecep-tor (42).7 CEM-C1 and CEM-C7 showed no discernible difference

in either the rate of formation or quantity of complex bound tothe gels from 5 to 30 min postactivation (data not shown). Theincrease in receptor complex bound to DNA was accompaniedby a corresponding reduction in the species bound to DEAE-

" Mean ±S.D.0 Numbers in parentheses, number of determinations. 7N. Holbrook and A. Munck, personal communication.

AUGUST 1983 3867

on March 3, 2019. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Zawydiwski et al.

cellulose. Both clones generated a small but detectable increasein receptor associated with hydroxylapatite within the initial 5min; this fraction remained constant thereafter. In addition, mixedcytosols produced upon activation a pattern of binding whichwas unchanged. Thus, by these criteria, CEM-C1 does not

appear to have an abundance or deficiency of cytoplasmic regulatory factors governing receptor activation and/or stability.

Since CEM-C1 receptor might possess a reduced potential for

binding to nuclear components, we examined the relative affinityof steroidrreceptor complex for DNA in vitro. Recently, it hasbeen demonstrated (37) that glucocorticoid:receptor complexesof cells from some leukemic patients yield abnormal DEAE-

cellulose chromatograms, characterized either by a multiplicityof peaks or by a single-peak profile. Cytosols having the latterpattern lacked Peak II activity but showed a low-salt-eluting

fraction in the vicinity of Peak I which failed to bind DNA. Chart2 shows that the heat-salt-activated [3H]TA:receptor complex of

CEM-C1 not only bound to DNA-cellulose but also was eluted

from the matrix at the same ionic strength (0.1 M phosphate) asdid CEM-C7 receptor complex.

In the study cited above, leukemic samples exhibiting theabnormal single-peak DEAE-cellulose profiles were also found

by glycerol gradient analysis to have a lower sedimentationcoefficient than did cytosols showing normal chromatograms.Furthermore, examples of abnormalities in molecular size andshape (determined by density gradient centrifugation and gelpermeation chromatography) in receptors having altered nuclear-and DNA-binding properties have been reported for corticoste-roid-resistant variants of the S49 (64) and P1798 (66) mouselymphoma cells. Whereas CEM-C1 receptor appeared unremarkable in both its DEAE- and DNA-binding characteristics, it is

nevertheless possible that the receptor could be deformed ortruncated and yet retain these and other wild-type properties.Such a molecule might fail to properly "present" or could alto

gether lack a domain governing lytic activity. To probe for thispossibility, we sized receptor complex by gel filtration. SephacrylS-300 chromatography of unactivated, molybdate-stabilized,CEM-C1 cytosol generated a peak of specific, competable ma-cromolecular-bound radioactivity, being eluted between the cal

ibration proteins, ferritin and catalase (Chart 3). The gel filtrationprofile was virtually superimposable upon that obtained with

5 T CEM-C1 CEM-C7

10 15 20 25 30 0 5 10

FRACTION NUMBER

15 20 2Õ 30

Chart 2. DNA-cellulose chromatography of receptor complexes. Activated cytosols, filtered through Sephadex G-25, were loaded onto DNA-cellulose equilibrated with KPD buffer, and the columns were thoroughly washed and then elutedwith a shallow linear phosphate gradient. Bound radioactivity represents 42%(CEM-C1) and 40% (CEM-C7) of counts applied from crude cytosols. Straight lines

with small data points show salt gradients.

5 10 15 20 25 30

FRACTION NUMBER

35

Chart3. Gel filtration of receptor complex. Unactivated molybdate-stabilizedlabeled cytosols with (O) and without (•)excess unlabeled TA were treated withdextran-coated charcoal and then chromatographed on Sephacryl S-300. Thecalibration proteins and their molecular dimensions are listed in "Materials andMethods." T. thyroglobulin; F. ferritin; C. catalase; A, aldolase; BSA, bovine serum

albumin; 0. ovalbumin; V0. void volume.

similarly treated CEM-C7 cytosol. Assuming these proteins tobe globular, we calculated Stokes' radii of 53 to 59 Àand 55 to59 A (n = 3) for the receptor complexes of the DEXS and DEXr

clones, respectively.GS Induction. Although the receptor of DEXr CEM-C1 seemed

normal by the biochemical criteria applied, we sought to determine whether, in fact, it was functional. GS is induced by gluco-

corticoids in a variety of cells: GH3 rat pituitary cells (68, 69),mouse L-cells (2, 69), HTC cells (32), Chinese hamster lung cells

(73), embryonic chick neural retina (41), rat glial C6 cells (50),and mouse primary astrocytes (29). We recently reported thatGS is also inducible in CEM-C7 cells and that this is a glucocor-ticoid-specific, receptor-mediated effect which occurs independently of DEX-mediated growth inhibition (23). We therefore choseto examine CEM-C1 for the expression of this steroid response.The dose-response curve for DEX (Chart 4) clearly indicates anelevation in GS activity with half-maximal and maximal activityoccurring at about 2 x 10~8 M and 10~7 M steroid, respectively.

These concentrations were similar to those required to elicitcomparable enzyme activity in CEM-C7 and essentially paralleledthe concentrations necessary for half- and full occupancy of wild-

type receptor (46). Moreover, the fold induction of GS producedby 1 (¿MDEX was found to be identical in CEM-C1 and CEM-C7

[cf. 1.9 ±0.6 (12 determinations) and 1.9 ±0.6 (9 determinations)]. As expected, the steroid specificity for GS induction, andhence for receptor, was the same in CEM-C1 and CEM-C7 (data

not shown).Steroid Metabolism. The striking similarity in receptor prop

erties of clones CEM-C1 and CEM-C7, together with the appar

ent rarity of such a variant (6, 48, 64), prompted us to consider

3868 CANCER RESEARCH VOL. 43

on March 3, 2019. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

.1001È

80<

uü600o.«

40RELATIVEG0

SCEM-C1CEM-C7.o.:

i•o•IîB5

iI1 °I}

•5Ii

i i-9 -8 -7

LOG CONCENTRATION IMI

-6

Chart 4. Dose dependence of GS induction by DEX. Cells were exposed (18hr) to various concentrations of DEX, washed in medium without glutamine, andthen resuspended in buffer (25 HIM citrate, pH 6.4) and frozen over dry ice. Cellpellets were thawed and disrupted by sonication, and the particulate-free cytosolwas assayed for enzyme activity. Data are normalized for differences in basal andmaximal (10 to 17 and 19 to 30 nmol/min/mg, respectively) GS activity, the latterset at 100. Bars, range of replicate determinations from 2 separate experiments.

alternative explanations for the resistance of CEM-C1 to DEX.

Reports in the literature of more active than usual metabolism ofcortisol by lymphocytes of murine lymphoma (11) and leukemicpatients (11, 13, 28) raised the speculation that CEM-C1 could

have an inherently normal lytic response to hormone but thatresistance is due instead to accelerated steroid biotransforma-

tion.In CEM-C7 cells, inhibition of growth (46), loss of viability, and

the concomitant arrest of cells in d (19) become apparent after24 hr of DEX addition; GS induction reaches maximum levels at12 hr (23). CEM-C7 cells can be rescued from the lethal effects

of DEX by steroid washout after 1 day (19,45) but are irreversiblyblocked in d and committed to death by 48 hr. Other wild-type

CEM clones, on the other hand, require a longer duration of DEXtreatment (e.g., up to 4 days) before full growth inhibition isachieved (71). It is possible then, that a depletion of intracellularsteroid to suboptimal concentrations by metabolic inactivationand/or elimination within the appropriate time frame might resultin the abrogation of lytic activity in sensitive cells without affectingenzyme induction. Our data, however, do not support such aninterpretation for CEM-C1. We observed no appreciable difference between CEM-C1 and CEM-C7 in the capacity to transform

DEX to more polar (extracellular fraction) or nonpolar (cellularfraction) compounds (Chart 5). The latter observation appearsto rule out conversion of DEX to a derivative possessing highreceptor affinity and antagonist activity. Furthermore, extracellular medium from DEX-treated CEM-C1 cells, when added tocultures of CEM-C7 in amounts calculated to yield DEX concen

trations between 5 and 100 nw (assuming unmetabolized steroid), produced the same degree of growth inhibition and celllysis as did identical concentrations of ethanolic DEX (data notshown). It seems unlikely then that such steroid modification asdoes occur is, in itself, a significant factor in contributing toglucocorticoid resistance in CEM-C1.

Plasma Membrane Fragility. Another potential explanationfor the resistance of CEM-C1 invokes the elaboration of lysis-

resistant cell membranes rather than a defect(s) in the biochemistry of steroid action. Based initially on observations with nucleiof corticosteroid-sensitive and -resistant murine P1798 lympho-

sarcoma cells, a hypothesis was proposed which attributed theability of the latter to better withstand the destructive effects of

Glucocorticoid Resistance in Human ALL Cells

glucocorticoids, as measured by a "nuclear fragility" assay (17),to the presence of "hardier" membranes (44). These findings

were subsequently extended to the plasma membranes of normal rat thymocytes, and the greater structural integrity of cellsmore resistant to osmotic rupture was suggested as reflecting ageneralized membrane phenomenon in cells insensitive to glu-cocorticoid-mediated lysis (78). We therefore considered thepossibility that DEXr CEM-C1 might simply be of stronger biophysical constitution than is its sister clone, DEXSCEM-C7. Using

an assay analogous to that of Young ef al. (78), we were unable,however, to demonstrate differential susceptibility of untreatedCEM-C1 and CEM-C7 cells to hypotonie shock. Both the dyeuptake and 51Cr release methods yielded curves with similar

midpoint inflections corresponding to salt concentrations (¡i,0.07to 0.08) resulting in 50% stained cells and 50% release ofradiochromium (Chart 6). We conclude that CEM-C1 does notpossess an intrinsically more lysis-resistant cell plasma mem

brane. Nevertheless, these observations do not preclude structurally dissimilar surface membranes in the sensitive and resistant cells which could be correlated with glucocorticoid cytotox-

icity (4, 5).Cell Lineage. Lymphocyte sensitivity to the lytic effects of

cEM-ci PELLET

40

12Idpm

X10*10È

12DIOACTIVI»-

CEM-C1 SUP n _. ..n-i

pTl:•CEM-C7SUP.nr-30

2.00l

]01.6M<

u- -uK

ControlsO.r1»CDCelMree Q 100n

Medium •Hi Radionuclide10 H-, j-ti „40n

n

CEM-C7 PELLET

40

1.5 3.0 4.5 6.0 7.5 0

DISTANCE (cm)

1.5 3.0 4.5 6.0 7.5

Charts. Thin-layer chromatograms of radioactivity extracted from [3H]DEX-treated cells. Cell cultures were incubated with [3H]DEX for 48 hr; the cells were

collected, washed, and disrupted by sonication. The sonically disrupted cells andsupernatant incubation media were extracted with organic solvent (at >90% and70 to 80% efficiency, respectively), and portions were chromatographed. Arrow,sample origin.

100

80

60

40

20

0.16 0.14 0.12 0.10 008 006

IONIC STRENGTH (u>

0.04 0.02

Chart 6. Plasma membrane fragility as a function of ionic strength of medium.Cells were suspended in salt solutions of varying hypotonicity, allowed to stand atroom temperature, and then assayed for cell viability by trypan blue exclusion and51Cr release.

AUGUST 1983 3869

on March 3, 2019. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Zawydiwski et al.

Table 2Cellular properties oíCEM-C1 and CEM-C7

Growth andsizeCEM-C1

(DEX1)CEM-C7 (DEX-)Doubling

time(hr)25

±3a (6)6

24 ±3 (6)Modal

sphericaldiameter

(lim)Karyotype9.3

±0.2 (7) 47,XX,inv(9),del(9Xp1 2),+209.2 ±0.3 (7) 47,XX,inv(9),deK9Xp1 2),+20Surface

membrane markers(% ofcells)Monoclonal

antibodyspecificityHuLyt-142

20HuLyt-2>99>99HyLyt-35 33Hula4 9Soybean

ag-Sheep RBC glutinin recep-

receptorstors0

<2>99 >993 Mean ±S.D.' Numbers in parentheses, number of determinations.

CEM C7

r *

(a) (b)

Fig. 1. Giemsa-11 stained chromosome 9 of CEM-C1 (a) and CEM-C7 (b). Theabnormal homologue is the right-hand member of each chromosome pair.

corticosteroids varies with the state of cell maturation and/ordifferentiation and morphological or immunological classification(9, 25, 56). Since CEM-C1 insensitivity could be ascribed to

these factors and not to an abnormality of steroid response, weundertook to establish the origin of the resistant phenotype.

CCRF-CEM was initially obtained as an uncloned ALL cell line.

It exhibited a bimodal chromosome distribution (38) and expressed T-cell-specific membrane characteristics (31, 39). CEM-C1 and CEM-C7 were independently isolated from the parentline (46); the DEX' variants described to date, however, have all

been derived from CEM-C7 (22) and were determined to have

receptor abnormalities (22, 57). It seemed a distinct possibility,then, that CEM-C1 might not be a clonal relative of CEM-C7

which had developed resistance through random spontaneousmutation (22, 62) but had instead evolved from a second leu-kemic population that was innately insensitive to glucocorticoid-

mediated cytolysis. However, several lines of evidence suggestthat CEM-C1 and CEM-C7 share a common ancestry, (a) The

morphological and growth properties of both clones were foundto be indistinguishable. Upon microscopic examination, eachclone showed intracellular features (not shown) typical of thelymphoblasts from which they were derived (12). Furthermore,the cells were found to be comparable in size and were observedto grow in stationary-suspension culture at identical rates (Table2). (b) Cytogenetic analysis showed no difference in cell kary-otype (Table 2), CEM-C1 having the same modal chromosomenumber and chromosome abnormality as does CEM-C7. The

unique structural rearrangement of chromosome 9 (Fig. 1), involving both a pericentric inversion and partial deletion of theshort arm,8 provides strong evidence that these cells arose from

the same transformed clone. This abnormality has been identifiedin all clones of CEM that we have studied thus far including DEXr

*r3h,'2

Li1

oJf—U(01

"ioti13•2

•1

-1"TnHuLyt-1—

20*»200

400 600 8001i

oHuLyt3—

-33°=V^»j»-«.r-

... |4321000

4321L".

OjToHuLyt-2àf\l\•

99%\200

400 600 8001•Hub_

j^L

...000200 400 600 800 1000 200 400 600 800 1000

CHANNEL NUMBER

CEM Cl

4

3

2 •

uHuLyM

200 400 600 800 1000

oHuLyt3

oHuLyt-2

200 400 600 800 1000

»Hub

8 D. E. Moore, R. Zawydiwski, and E. B. Thompson, manuscript in preparation.

200 400 600 800 1000 200 400 600 800 1000

CHANNEL NUMBER

Chart?. Indirect flow microfluorometric analysis of binding of monoclonal antibodies. Fluorescence frequency profiles for cells stained with fluorescein isothio-cyanate-conjugated goat anti-rabbit IgG. F(ab')u fragment, in the presence (solid

line) and absence (broken line) of mouse anti-human IgG.

variants isolated from a DEXSclone other than CEM-C7. (c) both

clones expressed similar biochemical markers on their surfacemembranes. When screened against a panel of 4 monoclonalantibodies, CEM-C1 and CEM-C7 yielded a qualitatively similar

spectrum of reactivity (Chart 7). Virtually all the cells of eachclone bound «HuLyt-2; a subpopulation of both clones, some-

3870 CANCER RESEARCH VOL. 43

on March 3, 2019. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Glucocorticoid Resistance in Human ALL Cells

what variable in number (Table 2), exhibited lower binding of«HuLyt-1and «HuLyt-3. These immunoglobulins detect differentantigens found largely on normal human T-cells and T-cell leu-kemias (18, 30, 36). A minor but comparable fraction of CEM-C1 and CEM-C7 cells reacted with «Hula,which recognizes adeterminant found predominantly on the non-T-cell population,i.e., B-cells, monocytes, and null cells (18). CEM-C1 failed toform erythrocyte rosettes, and only a small number of CEM-C7cells did so (Table 2). CEM-C1 and CEM-C7 alike were strongly

aggregated by soybean agglutinin, a response characteristic ofthe helper subset of T-lymphocytes (51). Collectively, these

observations reveal no dramatic differences in the membrane orin cellular or biological properties of either clone, thus lendingsupport to our interpretation of a common lineage for these cells.The membrane phenotype, in particular, the inability to formerythrocyte rosettes, suggests that CEM-C1 is not a more

differentiated lymphocyte and argues against DEX resistance byreason of cell maturation.

DISCUSSION

The phenotype of DEXr clone CEM-C1 is atypical of steroid-

selected, mutagenized, or drug-induced lymphoid variants whichare resistant to glucocorticoid-induced cytolysis. Resistant cellsof established murine lymphoma lines S49 and WEHI-7 (W7)either lack glucocorticoid-binding activity altogether (r) (48, 63)or exhibit aberrant nuclear translocation (nr, nt1)(7, 16, 26, 48,

77). Cells in the latter category possess receptors with alteredaffinity for DNA or altered molecular size or shape (16, 64, 77);they fail to demonstrate positive complementation of defects insomatic cell hybrids (15, 47). Resistant cells of the P1798 trans-

plantable mouse lymphoma contain an abnormally small receptorwhich displays greater than parental nuclear- and DMA-binding

properties (66).In the human ALL cell line CEM, spontaneously resistant

clones derived from DEXS clone CEM-C7 are receptor positive

but exhibit a wide range in content (22). Preliminary characterization of some clones reveals diminished or nonexistent nucleartransfer (21 ) which, in at least one example (57), can be attributedto the inability to form stable activated steroid-receptor complex(act1). Hence, while the DEXr derivatives of murine and human

lymphoid cells show considerable phenotypic variability, physi-

cochemical and genetic evidence alike indicates that they allcontain defects in receptor. Resistant clone CEM-C1, on theother hand, has near wild-type levels of receptor, comparable

affinity for DEX, and efficiency of nuclear transfer. Biochemically,the receptor of CEM-C1 is indistinguishable from that of sensitiveclone CEM-C7 as assessed by ion-exchange and gel filtrationchromatography. Furthermore, CEM-C1 receptor can mediateanother glucocorticoid-specific effect, GS induction. None of thereceptor-containing DEXr mutants of CEM examined to date has

demonstrated this response. Thus, CEM-C1 receptor not only

appears normal but is also functional, at least by this test.The presence of a functional receptor in CEM-C1 raised the

possibility that resistance may not, in fact, be due to a lesion(s)in the mechanism of glucocorticoid hormone action. We consequently considered and subsequently ruled out as unlikely several alternative explanations, among them steroid metabolism,"hardier" plasma membranes, cell ploidy, growth rate, and cell

lineage. Collectively, these observations suggest that resistance

in clone CEM-C1 is not associated with a receptor defect, butresults from a postreceptor lesion in a steroid-sensitive cell line.CEM-C1 possesses properties of the murine "deathless" variant

(63) and may prove to be the first bona fide example of thisvariant in human lymphoid cells, the murine equivalents of whichhave seen only limited study (64).

Evidence for the presence of functional glucocorticoid receptorin some lymphoid cells inherently insensitive to lysis has recentlybeen presented. The human lymphoblastoid line, IM-9, while notgrowth inhibited by DEX, responds with a receptor-mediatedstimulation of S'-nucleotidase (52). SAK8, a murine T-cell line

derived from AKR leukemic cells, shown previously to containapparently normal steroid receptor (34), responds with cell aggregation and an increase in mRNA and leukemia virus proteins(14). Although likewise not susceptible to the lethal activity ofglucocorticoids, SAK8 has receptor capable of complementingthe lytic defect in somatic cell hybrids formed with a receptoriessvariant of the W7 cell line. In addition, treatment with 5-azacyti-

dine generates DEX sensitivity. It has therefore been suggestedthat glucocorticoid resistance in this line involves inactivation ofnon-receptor genes through methylation of DNA at a "lysis" locus

(14). Methylation and/or demethylation may, of course, be causing much more general effects. Nevertheless, the probability thatglucocorticoid resistance in human clone CEM-C1 can be reversed by 5-azacytidine has also been explored. Such treatment,however, failed to restore DEX sensitivity.9 Thus, there is a clear

difference in phenotype between the 2 receptor-positive but lysis-resistant lines, the CEM-C1 (human) and the SAK8 (mouse). This

may reflect different mechanisms of resistance between the 2lines, and may also reflect a tendency of the 2 species to acquireresistance by different mechanisms, as other data suggest (19-

23,37,70,71).CEM-C1 is but one of over 100 spontaneously resistant and

mutagenized clones isolated from CEM in this laboratory. Themajority of these DEXr variants contain drastically reduced

(<30% wild-type) receptor levels; a few contain 50% or moreand, of these, at least 2 possess greater than parental (CEM-

C7) levels of receptor (22). Because only a limited number havebeen studied, it is not possible to estimate the frequency of thisnew phenotype. Presumably, many of these cells would exhibitthe recently characterized "activation-labile" receptor defect.

However, it is also conceivable that some of the above-mentioned clones possessing near wild-type steroid binding might

have functional receptor. It remains to be determined, therefore,whether the CEM-C1 phenotype constitutes a larger proportion

of the resistant population than is presently indicated. Furthermore, it would be of considerable interest to learn whether thisclass of resistance has a restricted distribution (is confined tothe CEM line) or is widely manifest in human leukemias.

The CEM-C1 phenotype of glucocorticoid resistance has important implications for the hormonal management of lymphopro-

liferative disorders. First and foremost, it indicates that resistanceneed not necessarily be associated with diminished content ofand/or structural abnormality in receptor. Thus, biochemical assays for altered receptor properties might not be reliable indicators of cytolytic sensitivity and therapeutic response. Also, itquestions the utility of GS induction as a prognosticator ofcytolytic sensitivity. We have suggested previously that GS

9J. C. Gasson and S. Bourgeois, personal communication.

AUGUST 1983 3871

on March 3, 2019. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Zawydiwski et al.

induction may provide a useful marker for receptors capable ofmediating cell death in leukemic cells (59). Clearly, the data showthat cell death and GS induction in CEM-C1 cells are noncoor-

dinately regulated. Nonetheless, the infrequency with which wehave isolated cells expressing the CEM-C1 phenotype suggests

that this phenotype may be the exception rather than the rule.Only direct analysis of samples from leukemic patients will provide the answer as to the usefulness of GS induction as a markerfor therapeutic response.

Although our data suggest that CEM-C1 contains a "postreceptor" lesion, they do not exclude entirely the possibility that

resistance is due to a subtle receptor defect. Confirmation awaitsthe demonstration of either structural identity with wild-type

receptor or functional capacity to mediate cytolysis. Studiestoward resolving this question are currently in progress. ShouldCEM-C1 prove to possess a nonreceptor defect, it would beinvaluable in the elucidation of the molecular basis of resistanceand in the identification of genetically distinct components ofglucocorticoid hormone action in somatic cell hybrids betweenvariants of similar phenotypes.

While these cells do not meet the genetic ideal of being deriveddirectly from a sensitive clone by single-step selection, they

nevertheless appear to be of considerable potential value in thestudy of steroid action. They are closely related to their sensitiveclone, have functional receptors, and yet are steroid resistant.CEM-C1 cells could ultimately prove to be a well-defined model

for the chemotherapy of certain classes of leukemias refractoryto glucocorticoids.

ACKNOWLEDGMENTS

The authors wish to thank Susan O. Sharrow (Immunology Branch, NationalCancer Institute) for performing the FACS analysis, Billie P. Wagner for technicalassistance with the enzyme assays, Dorothy E. Moore for karyotyping cells, andJean W. Regan for editorial assistance in the preparation of this manuscript.

REFERENCES

1. Alberts, B., and Herrick, G. DNA-cellulose chromatography. Methods Enzymol.,21: 198-217, 1971.

2. Barnes, P. R., Youngberg, D., and Kitos, P. A. Factors affecting the productionof glutamine in cultured mouse cells. J. Cell. Phystol, 77, 135-144,1971.

3. Baxter. J. D., Harris, A. W., Tomkins, G. M., and Cohn, M. Glucocorticoidreceptors in lymphoma cells in culture: relationship to glucocorticoid killingactivity. Science (Wash. D. C.), 171: 189-191.1971.

4. Behrens, U. J., and Hollander, V. P. Cell membrane sialoglycopeptides ofcorticoid-sensitive and -resistant lymphosarcoma P1798. Cancer Res., 36:172-180,1976.

5. Behrens, U. J., Mashbum, L. T., Stevens, J., Hollander, V. P., and Lampen, N.Differences in cell surface characteristics between glucocorticoid-sensitive and-resistant mouse lymphomas. Cancer Res., 34: 2926-2932,1974.

6. Bourgeois, S. Glucocorticoid-mduced lymphocytolysis: state of the geneticanalysis. J. Supramol. Struct., 73.-401-410,1980.

7. Bourgeois, S., Newby, R. F., and Huet, M. D. Glucocorticoid resistance inmurine lymphoma and thymoma lines. Cancer Res., 38: 4279-4284,1978.

8. Cidlowski. J. A., and Cidlowski. N. B. Regulation of glucocorticoid receptorsby glucocorticoids in cultured HeLa S3 cells. Endocrinology, 709: 1975-1982,1981.

9. Claman, H. N. Corticosteroids and lymphoid cells. N. Engl. J. Med., 287: 388-397, 1972.

10. Coulson, P. B., Thomthwaite, J. T., Skafar, D. F., and Seaver, S. S. Modulationof glucocorticoid hormone receptor levels in chicken lymphoid tissue followingtreatment with androgens In vivo. J. Steroid Biochem., 77:1-9,1982.

11. Dougherty, T. F., Berliner, M. L., and Berliner, D. L Hormonal control oflymphocyte production and destruction. Prog. Hematol., 3: 155-169, 1962.

12. Foley, G. E., Lazarus, H., Farber, S., Uzman, B. G., Boone, B. A., andMcCarthy, R. E. Continuous culture of human lymphoblasts from peripheralblood of a child with acute leukemia. Cancer (Phila.), 78: 522-529, 1965.

13. Forker. A. D.. Solinger, R. E., Morris. J. H., and Lawson, W. E. Metabolism of

cortisol-"C by human peripheral leukocyte cultures from leukemic patients.Metabolism, 72: 751-759, 1963.

14. Gasson, J. C., and Bourgeois, S. A new determinant of glucocorticoid sensitivity in lymphoid cell lines. J. Cell Biol . 96: 409-415,1983.

15. Gehring, U. Specific receptors control steroid sensitivity in lymphoma cellhybrids. Mol. Cell. Endocrinol., 20: 261-274, 1980.

16. Gehring, U., and Tomkins, G. M. A new mechanism for steroid unresponsive-

ness: loss of nuclear binding activity of a steroid hormone receptor. Cell, 3:301-306,1974.

17. Giddings, S. J., and Young, D. A. An in vitro effect of physiological levels ofcortisol and related steroids on the structural integrity of the nucleus in ratthymic lymphocytes as measured by resistance to lysis. J. Steriod Biochem.,5:587-595,1974.

18. Hansen. J. A., Martin, P. J., and Nowinski, R. C. Monoclonal antibodiesidentifying a novel T cell antigen and la antigens of human lymphocytes.Immunogenetics, 70: 247-260,1980.

19. Harmon, J. M., Norman, M. R., Fowlkes, B. J.. and Thompson, E. B. Dexa-

methasone induces irreversible d arrest and death of a human lymphoid cellline. J. Cell. Pnysiol., 98: 267-278,1979.

20. Harmon, J. M., Norman, M. R., and Thompson, E. B. Human leukemic cells inculture—a model system for the study of glucocorticoid-induced lymphocytolysis. In: E. B. Thompson and M. E. Uppman (eds.). Steroid Receptors andthe Management of Cancer, Vol. 2, pp. 113-129. Boca Raton, Fla.: CRC

Press, Inc., 1979.21. Harmon, J. M., Schmidt, T. J.. and Thomspon, E. B. Defective steroid receptors

in a glucocorticoid resistant clone of a human leukemic cell line. In: W. W.Leavitt (ed.), Hormones and Cancer, pp. 301-313. New York: Plenum Publish

ing Corp., 1982.22. Harmon, J. M., and Thompson, E. B. Isolation and characterization of dexa-

methasone-resistant mutants from human lymphoid cell line CEM-C7. Mol.Cell. Biol., 7:512-521,1981.

23. Harmon, J. M., and Thompson, E. B. Glutamine synthetase induction byglucocorticoids in the glucocorticoid-sensitive human leukemic cell line CEM-C7. J. Cell. Physiol., 770: 155-160, 1982.

24. Holden, H. T., Lichter, W., and Sigei, M. M. Quantitative methods for measuringcell growth and death. In: P. F. Kruse, Jr., and M. K. Patterson, Jr. (eds.),Tissue Culture: Methods and Applications, pp. 408-412. New York: Academic

Press, Inc., 1973.25. Homo, F., Duval, D., Hatzfeld, J., and Evrard, C. Glucocorticoid sensitive and

resistant cell populations in the mouse thymus. J. Steroid Biochem., 73: 135-

143,1980.26. Huet-Minkowski, M., Gasson, J. C., and Bourgeois, S. Induction of glucocor-

ticoid-resistant variants in a murine thymoma line by antitumor drugs. CancerRes.. 47: 4540-4546,1981.

27. lacobelli, S., Ranelletti, F. O., Natoli, C., and Longo, P. Glucocorticoid receptorsin resting and growing 3T3 cells, in. L. Jimenez de Asua, R. Levi-Montalcini,R. Schieids, and S. lacobelli (eds.). Control Mechanisms in Animal Cells, pp.121-133. New York: Raven Press, 1980.

28. Jenkins, J. S., and Kemp, N. H. Metabolism of cortisol by human leukemiccells. J. Clin. Endocrinol. Metab., 29: 1217-1221,1969.

29. Juurtink, B. H., Schousboe A., Jürgensen,0. S., and Hertz, L. Induction byhydrocortisone of glutamine synthetase in mouse primary astrocyte cultures.J. Neurochem., 36: 136-142, 1981.

30. Kamoun, M., Martin, P. J., Hansen, J. A., Brown, M. A., and Nowinski, R. C.Identification of monoclonal antibody of a p55 surface polypeptide on humanT lymphocytes associated with E-rosette receptor. J. Exp. Med., 753: 207-

212, 1981.31. Kaplan, J., Ravindranath, Y., and Peterson, W. D., Jr. T and B lymphocyte

antigen-positive null cell leukemias. Blood, 49: 371-378,1977.32. Kulka, R. G., Tomkins, G. M., and Crook, R. B. Clonal differences in glutamine

synthetase activity of hepatoma cells. J. Cell Biol., 54: 175-179, 1972.33. Laurent, T. C.. and Killander, J. A theory of gel filtration and its experimental

verification. J. Chromatogr., 74: 317-330,1964.34. Lippman, M. E., Perry, S., and Thompson, E. B. Cytoplasmic glucocorticoid-

binding proteins in glucocorticoid-unresponsive human and mouse leukemiccell lines. Cancer Res., 34: 1572-1576,1974.

35. Lowry, 0. H., Rosebrough, N. J., Fan, A. L., and Randall, R. J. Proteinmeasurement with the Folin phenol reagent. J. Biol. Chem., 793: 265-275,1951.

36. Martin, P. J., Hansen, J. A., Nowinski, R. C., and Brown, M. A. A new humanT cell differentiation antigen: unexpected expression on chronic lymphocyticleukemia cells. Immunogenetics, 77: 429-439, 1980.

37. McCaffrey. R., Lillquist, A., and Bell. R. Abnormal glucocorticoid receptors inacute leukemia cells. Blood, 59: 393-400, 1982.

38. McCarthy, R. E., Junius, V., Farber, S., Lazarus, H., and Foley, G. E. Cyto-genetic analysis of human lymphoblasts in continuous culture. Exp. Cell Res.,40: 197-200,1965.

39. Minowada, J. Markers of human leukaemia-lymphoma cell lines reflect haematopoietic cell differentiation. In: B. Serrou and C. Rosenfeld (eds.), HumanLymphocyte Differentiation: Its Application to Cancer, pp. 337-344. Amsterdam: Elsevier/North-Holland BiomédicalPress, 1978.

40. Morishige, W. K. Thyroid hormone influences glucocorticoid receptor levels in

3872 CANCER RESEARCH VOL. 43

on March 3, 2019. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Glucocorticoid Resistance in Human ALL Cells

the neonatal rat lung. Endocrinology, 711:1017-1019, 1982.41. Moscona, M., Frenkel, N., and Moscona, A. A. Regulatory mechanisms in the

induction of glutamine synthetase in the embryonic retina: immunochemicalstudies. Dev. Biol., 28: 229-241,1972.

42. Munck, A., and Foley, R. Activated and non-activated glucocorticoid-receptorcomplexes in rat thymus cells: kinetics of formation and relation to steroidstructure. J. Steroid Biochem., 12: 225-230,1980.

43. Munck, A., and Wira, C. Methods for assessing hormone-receptor kineticswith cells in suspension: receptor-bound and nonspecifically bound hormone;cytoplasmic-nuclear translocation. Methods Enzymol., 36: 255-264,1975.

44. Nicholson, M. L, and Young, D. A. Effect of glucocorticoid hormones in vitroon the structural integrity of nuclei in corticosteroid-sensitive and -resistantlines of lymphosarcoma P1798. Cancer Res., 38: 3673-3680,1978.

45. Norman, M. R., Harmon, J. M., and Thompson, E. B. Use of a human lymphoidcell line to evaluate interactions between predmsolone and other chemother-apeutic agents. Cancer Res., 38: 4273-4278,1978.

46. Norman, M. R., and Thompson, E. B. Characterization of a glucocorticoid-sensitive human lymphoid cell line. Cancer Res., 37: 3785-3791,1977.

47. Pfahl, M., and Bourgeois, S. Analysis of steroid resistance in lymphoid cellhybrids. Somatic Cell Genet., 6: 63-74, 1980.

48. Pfahl, M., Kelleher, R. J., Jr., and Bourgeois, S. General features of steroidresistance in lymphoid cell lines. Mol. Cell. Endocrinol., 10:193-207,1978.

49. Phillips, H. J. Dye exclusion tests for cell viability. In: P. F. Kruse, Jr., and M.K. Patterson, Jr. (eds.). Tissue Culture: Methods and Applications, pp. 406-

408. New York: Academic Press, Inc., 1973.50. Pishak, M. R., and Phillips, A. T. Glucocorticoid stimulation of glutamine

synthetase production in cultured rat glioma cells. J. Neurochem., 34: 866-

872, 1980.51. Reisner, Y., Pahwa, S., Chiao, J. W., Sharon, N., Evans, R. L, and Good, R.

A. Separation of antibody helper and antibody suppressor human T cells byusing soybean agglutinin. Proc. Nati. Acad. Sei. U. S. A., 77: 6778-6782,

1980.52. Rousseau, G. G., Cambrón, P., and Amar-Costesec, A. Glucocorticoid recep

tor-mediated stimulation of 5'-nucleotidase in human lymphoblastoid IM-9

cells. FEBS Lett., 727: 249-252,1980.53. Sakaue, Y., and Thompson, E. B. Characterization of two forms of glucocor

ticoid hormone-receptor complex separated by DEAE-cellulose column chro-matography. Biochem. Biophys. Res. Commun., 77: 533-541,1977.

54. Scatchard, G. The attractions of proteins for small molecules and ions. Ann.N. Y. Acad. Sci., 57: 660-672,1949.

55. Schlechte, J. A., Ginsberg, B. H., and Sherman, B. M. Regulation of theglucocorticoid receptor in human lymphocytes. J. Steroid Biochem., 76: 69-

74,1982.56. Schlesinger, M. Antigens of the thymus. Prog. Allergy, J6: 214-299,1972.57. Schmidt, T. J., Harmon, J. M., and Thompson, E. B. "Activation-labile" gluco

corticoid-receptor complexes of a steroid-resistant variant of CEM-C7 humanlymphoid cells. Nature (Lond.), 286: 507-510, 1980.

58. Schmidt, T. J., Kim, K. J., and Thompson, E. B. Glucocorticoid sensitivity andreceptors in BALB/c T cell lymphoma lines expressing restricted patterns ofLy differentiation antigens. J. Steroid Biochem., 73:13-22,1980.

59. Schmidt, T. J., and Thompson, E. B. Glucocorticoid receptors and glutaminesynthetase in leukemic Sézarycells. Cancer Res., 39: 376-382,1979.

60. Seabright, M. A rapid banding technique for human chromosomes. Lancet, 2:971-972,1971.

61. Shipman, G. F., Bloomfield, C. D., Smith, K. A., Peterson, B. A., and Munck,A. The effects of glucocorticoid therapy on glucocorticoid receptors in leukemia

and lymphoma. Blood, 58: 1198-1202,1981.62. Sibley, C. H., and Tomkins, G. M. Isolation of lymphoma cell variants resistant

to killing by glucocorticoids. Cell, 2: 213-220,1974.63. Sibley, C. H., and Tomkins, G. M. Mechanisms of steroid resistance. Cell, 2:

221-227,1974.64. Sibley, C. H., and Yamamoto, K. R. Mouse lymphoma cells: mechanisms of

resistance to glucocorticoids. Monogr. Endocrinol., 72: 357-376,1979.

65. Siegel, L. M., and Monty, K. J. Determination of molecular weights and frictionalratios of proteins in impure systems by the use of gel filtration and densitygradient centrifugation: application to crude preparations of sulfite and hydrox-ylamine reductases. Biochim. Biophys. Acta, 772: 346-362,1966.

66. Stevens, J., and Stevens, Y.-W. Physicochemical differences between gluco-corticoid-binding components from the corticoid-sensitive and -resistant strainsof mouse lymphoma P1798. Cancer Res., 39: 4011-4021, 1979.

67. Svec, F., and Rudis, M. Glucocorticoids regulate the glucocorticoid receptorin the AtT-20 cell. J. Biol. Chenr, 256: 5984-5987,1981.

68. Tashjian, A. H., Jr., and Hoyt, R. F., Jr. Transient controls of organ-specific

functions in pituitary cells in culture. In: M. Sussman (ed.), Molecular Geneticsand Developmental Biology, pp. 353-387. Englewood Cliffs, N. J.: Prentice-

Hall, Inc., 1972.69. Thompson, E. B., Dannies, P. S., Buckler, C. E., and Tashjian, A. H., Jr.

Hormonal control of tyrosine aminotransferase, prolactin and growth hormoneinduction in somatic cell hybrids. J. Steroid Biochem., 72: 193-210, 1980.

70. Thompson, E. B., Harmon, J. M., Norman, M. R., and Schmidt, T. J. Glucocorticoid actions in a human acute lymphoblastic leukemia, T-cell line: a modelsystem for understanding steroid therapy. In: S. lacobelli, H. R. Lindner, R. J.B., King, and M. E. Lippman (eds.), Hormones and Cancer, pp. 89-98. New

York: Raven Press, 1980.71. Thompson, E. B., Harmon, J. M., and Zawydiwski, R. Corticosteroid effects

on an acute lymphoblastic leukemic cell line: a model for understanding steroidtherapy. In: S. Murphy and J. Gilbert (eds.). Proceedings of the InternationalSymposium on Leukemia Cell Biology and Therapy, St. Jude Children's

Research Hospital, Memphis, Tenn., New York: Elsevier, 1982.72. Thomdike, J., and Reif-Lehrer, L. A sensitive assay for glutamyltransferase.

Enzyme (Basel), 72: 235-241, 1971.73. Tiemeier, D. C., and Milman, G. Regulation glutamine synthetase in cultured

hamster cells. J. Biol. Chem., 247: 5722-5727,1972.74. Titus. J. A., Sharrow, S. O., Connolly, J. M., and Segal, D. M. FcflgG) receptor

distributions in homogeneous and heterogeneous cell populations by flowmicrofluorometry. Proc. Nati. Acad. Sei. U. S. A., 78: 519-523,1981.

75. Weiner, M. S., Bianco, C., and Nussenzwig, V. Enhanced binding of neuramin-idase-treated sheep erythrocytes to human T lymphocytes. Blood, 42: 939-

946,1973.76. Wyandt, H. E., Wysham, D. G., Minden, S. K., Anderson, R., and Hecht, F.

Mechanisms of Giemsa banding of chromosomes. 1. Giemsa banding withazure and eosin. Exp. Cell Res., 702: 85-94,1976.

77. Yamamoto, K. R., Stampfer, M. R., and Tomkins, G. M. Receptors fromglucocorticoid-sensitive lymphoma cells and two classes of insensitive clones:physical and DNA-binding properties. Proc. Nati. Acad. Sei. U. S. A., 77:3901-

3905, 1974.78. Young, D. A., Nicholson, M. L., Voris, B. P., and Lyons, R. T. Mechanisms

involved in the generation of the metabolic and lethal actions of glucocorticoidhormones in lymphoid cells. Prog. Cancer Res. Ther., 74:135-155,1980.

79. Zawydiwski, R., Harmon, J. M., and Thompson, E. B. Functional receptor in ahuman lymphoid line resistant to glucocorticoid-mediated cytolysis. J. Cell.

Biochem. Suppl. 6, p. 366. 1982.

AUGUST 1983 3873

on March 3, 2019. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1983;43:3865-3873. Cancer Res   Robert Zawydiwski, Jeffrey M. Harmon and E. Brad Thompson  Cell Line with Functional ReceptorGlucocorticoid-resistant Human Acute Lymphoblastic Leukemic

  Updated version

  http://cancerres.aacrjournals.org/content/43/8/3865

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/43/8/3865To request permission to re-use all or part of this article, use this link

on March 3, 2019. © 1983 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from