1993;53:3034-3039.Cancer Res Helen M. Beere, Richard I. Morimoto and John A. Hickman Differentiationduring Commitment of HL-60 Cells to Granulocyte

72,000 Constitutive Heat Shock ProteinrMSynthesis of the -Methylformamide-induced Changes inNGene Expression:

Investigations of Mechanisms of Drug-induced Changes in  

  

Updated Version http://cancerres.aacrjournals.org/content/53/13/3034

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

SubscriptionsReprints and

.pubs@aacr.orgDepartment atTo order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions.permissions@aacr.orgDepartment at

To request permission to re-use all or part of this article, contact the AACR Publications

American Association for Cancer Research Copyright © 1993 on November 8, 2012cancerres.aacrjournals.orgDownloaded from

ICANCER RESEARCH 53. 3034-30.19. July 1. 1993]

Investigations of Mechanisms of Drug-induced Changes in Gene Expression:

W-Methylformamide-induced Changes in Synthesis of the Mr 72,000

Constitutive Heat Shock Protein during Commitment of HL-60Cells to Granulocyte Differentiation1

Helen M. Beere, Richard I. Morimoto, and John A. Hickman2

Cancer Research Campaign Molecular anil Cellular Pharmacology Croup, School of Biological Sciences. University of Manchester, M13 9PT. United Kingdom ¡H.M. B..J. A. H.I, and Department of Biochemistry: Molecular Biology and Cell Biology. Northwestern University, Evanston, Illinois 60201 ¡R.I. M.I

ABSTRACT

HL-60 cells were treated with the differentiating agent iV-methylforma-

mide and early changes in gene expression and protein content wereinvestigated. Analysis of protein synthesis had previously shown an early(<12 In fall in the synthesis of M, 70.000 heat shock proteins (F. M.Richards, A. Watson, and J. A. Hickman. Cancer Res., 48: 6715-6720,

1988). The changes have now been characterized in detail and their kinetics compared to those of the expression of the c-myc protein. linimi-noanalysis, using antibodies to either the stress-inducible heat shock protein hsp7(l (4G4) or a p¡m-.l/,70,000 heat shock protein antibody (3A3),

showed that there was a striking reduction in the levels of the constitutiveheat shock protein hsc70 when cells were incubated continuously with 170HIMA'-methylformamide. A reduction in the level of hsc70 RNA was ob

served within 3 h and continued thereafter. In contrast, transcription ofthe hsc70 gene was induced within 1-2 h, after which the rate returned to

basal level. There were no significant changes in the rate of transcriptionof the stress-inducible heat shock proteins hsp70 or hsp90. When A'-

methylformamide was removed from the cells, prior to commitment todifferentiation, the levels of hsc70 were reestablished, whereas after 36 hof treatment there was no recovery. Western blotting with an antibody tothe c-myc protein showed this to fall to virtually undetectable levels by 3h under the same conditions. The results suggest that the loss of list-70,

which may perform a protein chaperoning role, was mediated at bothtranscriptional and posttranscriptional levels of regulation and was anearly event closely associated with the commitment of HL-60 cells to

differentiation. The fall in lisi 711was not associated with alterations in thecell cycle, nor were the kinetics of the change suggestive of a relationshipwith the decrease in content of I--/HVCprotein.

INTRODUCTION

Structurally and mechanistically disparate pharmacological agentsare able to induce the terminal maturation of HL-60 promyelocyticleukemia cells along the granulocyte-neutrophil pathway ( 1). An in

teresting question regarding this response to such disparate agentsconcerns the nature of the mechanisms whereby interactions of drugs,at presumably different loci, are able to initiate the molecular eventsthat lead to the differentiated state (2). Of particular interest are theconserved changes in gene expression of the "immediate-early" re

sponse genes, such as c-myc (3, 4). In previous studies, observationswere made that the induction of differentiation of HL-60 cells took

place when concentrations of agents were used which were marginallybelow those which induced cytotoxicity (5). We went on to suggestthat the induction of terminal differentiation may involve a stressresponse but paradoxically we observed that one of the earliestchanges in protein observed in HL-60 cells treated with NMF' was a

Received 11/20/92; accepted 4/26/93.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore he hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely lo indicate this fact.

1Supported by grants from the Cancer Research Campaign (SPI5I8). NATO

(890957). and the NIH (6M3).- To whom requests for reprints should he addressed.1The abbreviations used are: NMF. A'-methylformamide; BSA, bovine serum albumin.

fall, rather than an increase, in the synthesis of the M, 70,000 heatshock proteins (6). We were unable to identify which specific members of the MT70,000 heat shock proteins were being modulated.

Changes in the synthesis of heat shock proteins early during differentiation have been observed in Friend erythroleukemia cellstreated with dimethyl sulfoxide (7), mouse embryonal cells treatedwith retinoic acid (8), HL-60 cells treated with phorbol esters (9, 10),THP-1 monocytic leukemia cells treated with -y-interferon or retinoic

acid (10), K562 erythroleukemia cells treated with hemin (II). and ina variety of developmental processes (12, 13). The significance ofthese changes is not clear, and the qualitative, quantitative, and temporal aspects of them differ markedly according to cell type andconditions. Some of the changes appear to be associated with alterations in the proliferative status of the cells. The expression of thestress-inducible hsp70 has been shown to be cell cycle regulated (14).

The expression of the protooncogene c-myc is also associated with

the proliferative capacity of a variety of cells (15) and a fall inexpression has been considered to herald a withdrawal from the cellcycle ( 16). A fall in the expression of c-mvr has also been observed tobe a prelude to the terminal differentiation of HL-60 cells (3, 4). Thekinetics of the change in c-myc expression was found to depend upon

the nature of the inducer used and upon the lineage to which the cellsbecome committed (17. 18). Because of a colocalization betweenc-myc protein and the stress-inducible hsp70 (19), we were interested

in determining the precise characterization of the heat shock proteinwhich changed as HL-60 cells were committed to differentiate, and

whether there might be a relationship between the kinetics of thischange and that of the expression of c-myc.

Detailed descriptions of the regulation of the expression of some ofthe genes which encode the heat shock proteins have emerged recently, most particularly those of the Mr 70,000 family (20). Their rolein protecting the cell from the effects of elevated temperatures (21)and their constitutive role as protein chaperones (22. 23) is the subjectof intensive investigation. Because of this emerging knowledge oftheir molecular and cellular biology we consider study of changes inexpression of the M, 70.000 heat shock proteins may provide clues asto how a drug might bring about changes in gene transcription as wellas perhaps clarifying what role these proteins may play in the processof differentiation. The solvent NMF was chosen as an archetypicaldifferentiating agent (24) since it had been the subject of previousstudies by us (5, 6, 25, 26).

MATERIALS AND METHODS

Materials. All materials were obtained from Sigma Chemical Co. (Poole.Dorset, United Kingdom) unless otherwise stated.

fraction VII; hsc. constitutive heat shock protein; hsp. inducible heat shock protein; PBS,phosphate-buffered normal saline. pH 7.4; SDS, sodium dodecyl sulfate; SSC. standardsaline citrate (1 X SSC. 0.15 M sodium chloride. 0.015 M sodium citrate, pH 7.0); TPA.12-0-tetradecanoylphorbol-13-acetate.

3034

American Association for Cancer Research Copyright © 1993 on November 8, 2012cancerres.aacrjournals.orgDownloaded from

Hl-AT SIKX'K I'R()TI-:INS IN DIH I-RI.NTIA 1ION

Culture and Treatment of Cells. HL-60 human promyelocytic leukemia Coomassie blue to ensure equal loading and transfer. Membranes were thencells were maintained in RPMI-1640 medium (Imperial Laboratories, Andover.

Hampshire, United Kingdom) containing 10% fetal calf serum (TechGen.London, United Kingdom) supplemented with 2 HIMi-glutamine, at 37"C. 5%

CO2, 100% humidity. Cells were routinely maintained in logarithmic phasegrowth between I x IIP and 1 X 10" cells/ml by biweekly subculture and had

an approximate doubling time of 24 h. Cells were counted by using a hema-

cytometer and short term viability was determined by the exclusion of a 0.1%solution of trypan blue. HeLa cells were grown under identical culture conditions and subcultured by trypsin-EDTA removal from tissue culture flasks

(Costar, High Wycombe, Buckinghamshire, United Kingdom). HeLa cellswere heat shocked for l h at 42°C, left for 2 h, and harvested by scraping

before protein analysis as described below. NMF (Aldrich Chemical Co.,Gillingham, United Kingdom) was added directly to the cell cultures to give aconcentration of 170 HIM.and cells were removed at predetermined time pointsfor the assessment of differentiation and protein or RNA analyses.

Assessment of Differentiation. The percentage of differentiated HL-60

cells was assessed based on the ability to produce Superoxide aniónin responseto TPA stimulation and reduce the dye nitroblue tetra/olium. as described indetail by us previously (5).

Cell Cycle Kinetics. Following drug treatment for up to 48 h. cells wereresuspended in fixative (0.1% paraformaldehyde. 0.1% Triton) at a density of1 X I06/ml for a minimum of 4 h at room temperature. Twenty j¿lpropidium

iodide solution (2.5 mg/ml) were added sequentially to each 5(X)-fil sample,

incubated for 5 min at room temperature, before cell cycle analysis using aCoulter EPICS Counter. The flow cytometer was set to excite at 4(X) mW withthe 488 nm line and red fluorescence was collected through a 630 nm longband pass filter. Cell cycle analysis was performed by using in-house software.

Measurement of (iene Transcription Rates. Isolation of nuclei and transcription run-on analysis were carried out according to previously publishedmethods (11. 27). Briefly, isolated nuclei were labeled with |<:P|L'TP (Amer-

sham International, Aylesbury, Buckinghamshire. United Kingdom) and thetranscription reaction was arrested by the addition of 600 /u.1of a stop bufferand a further 2-h incubation at 45"C. After trichloroacetic acid precipitation of

nucleic acids, the labeled nascent transcripts were hyhridi/ed to plasmid DNAimmobili/ed onto nitrocellulose Hybond-C (Amersham International): pH 2.3

(human hsp70) (28); pUC 801 (human hsp90a) (29); pUC 811 (humanhsp9()|3) (29); pHG2 (human grp78) (30): pHA7.6 (human hsc70)4. pHFA-l

(human a-actin) (31 ) at 42 C for 72 h. in a buffer containing 50% formamide.6 X SSC. 10 x Denhardt's. 0.2% SDS. Filters were then washed in a series of

buffers: 6 x SSC/0.2% SDS. 1 x 15 min at room temperature; 2 x SSC/0.2%SDS, 2 x 30 min at 65°C;0.2 x SSC/0.2% SDS. 2 x 30 min at 65°Cand

exposed to X-ray film (X-OMAT AR. Kodak. Heine! Hempstead. Hertfordshire, United Kingdom) at -80°C.

Isolation of RNA and Northern Blot Analysis. Total cellular RNA wasisolated by using a modified version of the guanidinium isothiocyanate methodas described previously (32). Briefly. 5 X 107-l X 10" cells were lysed in 4 \i

guanidinium isothiocyanate solution, layered onto 5.7 MCsCl/0. l MKDTA. andcenlrifuged overnight (35.0(X) x #, 4°C,17 h). Following phenolic extraction

and overnight ethanol precipitation. 20 fig purified RNA were fractionated ona denaturing 1.2% agarose-formaldehyde gel. Hthidium bromide (40 ¿ig/ml)

was included with the RNA to allow visualization of the loading of each gel.After overnight capillary transfer to Hybond-C. RNA was immobili/ed by UV

irradiation and prehybridi/ed in formamide containing buffer (50% formamide, 6 x SSC. 5 x Denhardt's. 0.1% SDS, 50 fig/ml tRNA). at 42°Cfor

approximately 6 h. Filters were hybridi/ed overnight to the <:P random prime-

labeled (Boehringer Mannheim. Lewes. Hast Sussex, United Kingdom) 600-hase pair EcuRl fragment encoding the human 5' sequence of hsc70 (pHA 7.6).

The filters were then washed in a series of buffers, as described above foranalysis of transcription.

Immunoblot Analysis. Cells (2 x 10s) were washed in PBS. denatured by

boiling in reducing sample buffer (2% SDS. 10% glycerol. ().(X)2% bromophe-

nol blue. 40 rmi Tris, pH 6.8), and total cellular protein was separated bySDS-polyacrylamide gel electrophoresis by using 10% polyacrylamide gels.Following the electrophoretic transfer of proteins to nitrocellulose Hybond-C

Extra, the membranes were blocked overnight in 5 mg/ml BSA in PBS. andwashed in 0.1% Tween in PBS. The gels were stained for residual protein with

incubated with either anti-hsp70 specific (4G4). anti-hsp70/hsc70 (3A3).

mouse monoclonal antibodies (gifts from Dr. Shawn Murphy. NorthwesternUniversity. Evanston, IL), or rabbit antisera pan C-HIVCantibody (a gift from Dr.Gerard Evan. Imperial Cancer Research Fund. London) for 1-2 h (4G4 and3A3 antibodies were diluted 1:100 and c-mvc I : KXX)in 5 mg/ml BSA in PBS

before use) and washed with 0.1% Tween in PBS. This was followed by anadditional 1-h incubation with affinity purified goat anti-mouse IgG or goatanti-rabbit IgG conjugated to horse radish peroxidase (diluted L6IXXX)in 5

mg/ml BSA in PBS before use) and repealed washings with 0.3 and 0.1%Tween 20 in PBS. The antibody-specific proteins were visualised by using the

enhanced chemiluminescence detection system according to the recommendedprocedure (Amersham International). Filters that were to be reprobed with adifferent antibody were first incubated in stripping buffer ( IMTris. pH 6.8. 2%SDS. and 5% ß-mercaptoethanol) for 30 min at 50°C.The immunoblotting

procedure was then repeated as above.

RESULTS

Characterization of NMF-induced Changes in Levels of the Mr

70,000 Heat Shock Proteins. Immunoblot analysis showed that continuous treatment of HL-60 cells with 170 HIMNMF induced a time-

dependent decrease in the cellular amount of a heat shock protein witha molecular weight of approximately 70.000 (Fig. I). Analysis bySDS-polyacrylamide gel electrophoresis of changes in protein synthesis, after labeling with I^Slmethionine confirmed our previous find

ings (6) of a fall in the synthesis of a M, 70.000 protein after approximately 6 h (data not shown). In order to identify which member of thehsp70 family changed, we compared the Western blot results by usingtwo monoclonal antibodies, 3A3, which recognizes both the constitutive hsc7() and the inducible hsp70. and 4G4. which recognizes onlyhsp70. Protein samples from HL-60 cells exposed to NMF. and fromheat-shocked HeLa cells, were analy/.ed by using the inducible hsp70-

specific antibody, 4G4. The HeLa cells were used to provide astrongly inducible heat shock response, and therefore an increase inthe level of the stress inducible hsp70 protein (Fig. 14); Westernblotting of HL-60 cells did not show a very significant change in the

amount of this protein alter heat shock, despite our previous findingof some change in synthesis (Fig. I ). The blot was then reprobed byusing the antibody 3A3, which detects both the inducible and constitutive forms of hsp70. Both HL-60 and HeLa heat shock 70 proteins

were immunologically reactive with 3A3 (Fig. 20). Because of theslightly inducible nature of the constitutive heat shock protein. hsc70,and the severultbld induction of the stress inducible hsp7() proteinfollowing a heat shock, these two proteins from HeLa cells are notfully resolved. The detection of an additional upper band followingincubation with 3A3 is, however, evident and is labeled accordingly(Fig. 2, Lane f>). This result suggested that there was a high basalexpression of the constitutive!)' expressed hsc70 protein which wasreduced following NMF treatment. Although 1-dimensional analysisfailed to detect the heat/stress-inducihle protein hsp70 in HL-60 cells,small amounts of this protein were detected by using 2-dimensional

170mM

u)12h 15h I8h 2lh 24h 36h

Hsc70

4C. Hunt, unpublished dala.

Fig. I. Levels »Ihsc70 protein in HL-60 cells following Ireuunem with a difl'erenlialini:

dose of NMF. Cell lysates were separated by SDS-polyacrylumide gel electrophoresis andheal shock proteins iminunologieally reactive to the anli-hsp7() antibody. 3A3. delected b\Western hlol analysis.

3035

American Association for Cancer Research Copyright © 1993 on November 8, 2012cancerres.aacrjournals.orgDownloaded from

HEAT SHOCK PROTEINS IN DIFFERENTIATION

Jta

1V)

4h 8h 12h 24h

-Hsp70

-Hsc70'Hsp70

Fig. 2. Detection of hsc7() protein expression in HL-6() cells using a combination oftwo mouse monoclonal antibodies specific for more that one member of the heat shockprotein 70 family. Western blot analysis of total cellular protein from 2 X IO5 cells/lanewas first carried out by using (A ) the monoclonal antibody, 4G4, specific for the stress-inducihle form of the MT 70.000 heat shock protein. hsp70. A protein sample fromheat-shocked (42CC. 1 h) HeLa cells was included as a positive control. The filter was

stripped and rcprobcd (0 ) with the use of the monoclonal antibody 3A3 to detect both theconstitutive (hsc) and inducible (hsp) forms of hsp70.

lh 2h 4h 8h

VECTOR

Hsp70Hsp90a

Hsp90ß

Grp78

Hsc70

Actin

Fig. .V In vitro transcription in nuclei of HL-60 cells isolated during NMF treatment,and alter a heat shock of 42°C for 1 h. Following isolation of nuclei and in vitrotranscription reactions. [ 1;P|UTP-labeled transcripts were hybridized to filler houndDNA: pH 2.3 (human hsp70); pL'C 801 (human hspWcr); pL'C 811 (human

pHG2 (human grp78) pHA7.fi (human hsc70): pHFA-l (human a-actin).

Western analysis, and the levels remained essentially unchangedthroughout NMF treatment (data not shown). The onset of the selective decrease in the constitutive protein hscVO was evident after 12 hof NMF exposure. The level of hsc70 fell to almost undetectablelevels by 36 h, concomitant with the irreversible commitment of thecells to terminal differentiation (see below).

Changes in Heat Shock Gene Expression. Exposure to 170 nuiNMF consistently induced the transient and selective increase in thetranscription rate of the Iisc70 gene within 1-2 h (Fig. 3). Significantly, the transcription of its stress-inducible counterpart. hsp70. and

another heat shock gene h.spVO,were unaffected. Following the transient induction of Iisc70 transcription, there was a return to controllevels by 4-8 h. Fig. 3 shows that heat and NMF treatment of HL-60

cells induced transcriptional changes of the a-actin gene, so that the

basal and steady state levels of transcription of the other heat shockgenes were taken to represent steady state levels of transcription. Theexpression of the hsc70 gene was also examined by Northern blotanalysis. It was observed that hsc70 message levels were significantlyreduced by 3 h (Fig. 4), despite the maintenance of basal levels oftranscription (Fig. 3). Ethidium bromide staining of the gel prior totransfer (shown in Fig. 4, bottom) ensured equal loading of RNA.Significant levels of protein were detectable (Fig. I ) up until after 12h. suggesting either that the protein had a significantly longer half-lifethan its message or that its half-life was extended in the presence of

NMF.Relationship between Changes in hsc70, c-myc Levels. Growth

Arrest, and Commitment to Differentiation. The conditions foroptimal HL-60 differentiation have been described previously (5): a96-h continuous incubation with 170 ITIMNMF induced maximumdifferentiation to granulocyte-like cells. These parameters were rein-

vestigated for the purpose of current experiments, and furthered todetermine the time after which the cells became irreversibly committed to a terminally differentiated phenotype, the so-called commitment

time. It was found that a minimum exposure time of between 36 and48 h to NMF was required to commit cells, as assessed by nitrobluetetrazolium reduction (Fig. 5-4). Flow cytometric analysis, during thefirst 48 h of NMF exposure, showed that there was no significantchange in the distribution of cells in the cell cycle, at which time thecells were committed. Analysis of cell numbers confirmed this (TableI ). Interestingly, cell cycle progression of HL-60 cells occurred independently of much reduced levels of c-myc protein. We observed arapid decline in the levels of c-myc protein well before proliferationceased (Fig. 5ß).Although appreciable levels of c-myc were unde

tectable by approximately 3 h. cells were still capable of initiatinganother full round of cell division (Table 1).

Immunoblot analysis of proteins isolated from cells which had beenincubated for increasing times with 170 ITIMNMF. then washed beforea 24-h drug-free recovery period, showed that the effects of NMF to

suppress hsc levels were reversible up to, but not beyond the point of

lh 3h 4h 6h

28S

hsc70

18S

EthidiumBromide

*•»••¿�••¿�*Fig. 4. Northern blot analysis of hsc70 gene expression during NMF treatment (0-6 h).

Total cellular RNA was isolated at the indicated times using the guanidiniumisothiocyanate/CsCI method. Twenty ¿igRNA were separated on a denaturing formaldehyde gel. transferred to nitrocellulose, and hybridi/ed to a l2P-labeled 600-base pair DNA

fragment of hsc70 (pHA7.6).

3036

American Association for Cancer Research Copyright © 1993 on November 8, 2012cancerres.aacrjournals.orgDownloaded from

HEAT SHOCK PROTEINS IN DIFFERENTIATION

60 -I

50-

40 H

O % 4 8 24 32 36 48

NMF EXPOSURE (h)

B

170mM

1h 2h 3h 16h 20h 24h 32h 48h

c-Myc

Fig. 5. A. communient til HL-60 cells to granulocytic differentiation following treatment with 170 nisi NMF. assessed by nitroblue tetrazolium reduction. Limited exposureto NMF (0—18h) followed by drug-free incubation up to % h allowed determination ofthe commitment time to terminal differentiation. Controls included cells incubated in thepresence or absence of NMF for 96 h. Viable cell numbers were also assessed at each timepoint. B. Western blot analysis of c-mvt1 protein levels in HL-60 cells following NMFtreatment. Total cellular protein from 2 X 10s cells/lane was analyzed throughout the

period of commitment to differentiation by using the pan-inyc rabbit anti-sera, at theindicated times. Antibody-specific proteins were visualized by using the enhanced chemi-luminescence detection system (see "Materials and Methods").

commitment to granulocytic differentiation (Fig. 6). The reduction inhscVO protein content after a 36-h exposure to NMF was not restoredto control level after a 24-h recovery period.

DISCUSSION

HL-60 cells treated with NMF underwent an early fall in cellular

levels of one of the members of the M, 70.000 heat shock family. Thiscorresponded to our previous observations regarding the fall in synthesis of heat shock proteins (6). Immunoblot analysis using twoantibodies, which discriminated in the pattern of recognition of themembers of the Mr 70,000 hsps, suggested, by elimination, that theheat shock protein which was reduced in amount was the constitutivehsc70. Analysis of the proteins by two-dimensional electrophoresis

followed by immunoblotting. showed that there was no change in thestress/heat-inducible hsp70 (data not shown). Although a number of

previous studies have identified changes in the synthesis of M, 70.000family of heat shock proteins during differentiation, neither the preciseidentity of the protein whose expression was being modulated norchanges in cellular levels were defined. This is important since eachmember of this family of proteins may serve a different role in the cell(13). The heat shock protein Mr 70,000 family is the most highlyconserved of the heat shock proteins, containing between 8 and 10

members, some of which are expressed constitutively, while others areinduced in response to a variety of stresses. Although the precisefunction of hsc70 remains unclear, reports have identified a range ofhsc70-specific characteristics, including clathrin-uncoating ATPase

activity (33, 34). Suggested functional roles for hsc70 also includeinvolvement in the lysosomal degradation of intracellular proteins(35), the chaperoning of proteins from the cytoplasm to the nucleus(36), and as a binding protein for peroxisome proliferators (37). Incomparison to the stress-inducible heat shock protein 70. regulation of

the expression of hsc70 has not been investigated.The change in the cellular level of hsc70 was accompanied by

selective changes in the transcriptional rate of the hsc70 gene, incomparison to the stress-inducible hsp70 (Fig. 4). Nuclear run-on

analysis, interestingly, showed a transient but significant change intranscriptional rate after 1-2 h of NMF incubation. Thereafter, the

transcriptional rate appeared to fall back to control levels. The significance of this transient increase in transcription is unclear. Two recentreports observed transient increases in the expression of hsp27 (9) inHL-60 cells treated with the phorbol ester TPA, and in c-myc duringinterleukin 6-induced differentiation of MI myeloid leukemia cells(38. 39). The transient increase in c-myc RNA preceded a subsequent

reduction to undetectable levels and, although this was noted by theauthors, no mechanistic explanation of the changes was offered. Atransient increase in hsc70 protein synthesis or amount (by immunoblotting). corresponding to the rise in transcription rate, was not observed (data not shown). The disparity observed between transcriptionrate and the levels of RNA and protein suggest that the transcriptsdetected by nuclear run-on analysis may be incomplete and therefore

not able to be fully processed to produce mature mRNA to be used fortranslation. This does not eliminate the possibility of the involvementof changes in the rate of transcription in some regulatory mechanism.The transcription rate of hsc70 was low (Fig. 3), but the constitutivelevels of mRNA appeared to be considerable (Fig. 4), which is suggestive of the production of a stable message. The reduction in theamount of hsc70 mRNA to a level below that of control, underconditions of continued basal transcription, therefore suggests thatNMF may affect message stability. Previous reports have shown thatheat-inducible hsp70 expression is regulated at both transcriptionaland posttranscriptional levels (40). Although down-regulation of

hsc70 protein levels were also observed, these occurred at a time afterthe changes in RNA levels (compare Figs. 1 and 4). This suggests thateither the stability of the protein is greater, or that RNA stability waschanged after NMF treatment. This is under investigation.

The temporal change in hsc70 level induced by NMF was stronglysuggestive of it playing a role in the maintenance of the block indifferentiation which characterizes leukemic HL-60 cells. A compar

ison of the kinetics of the reduction in hsc70 protein levels with thoseof commitment (Fig. 5/4) showed that as levels fell below thosedetectable by immunoblotting. cells were becoming committed to

Table I F/mr cylitmetric itnal\sis of the tiisirìhiiiiotìiif HL-f>()ceils in the cell cvcletoUowiiìfìinttuction i)f iliffen'titititian with NMh

Cells were harvested at the indicated times after addition of NMF. fixed, and analyzedfor DNA content after staining with propidium iodide.

Control8h24

h48h»of

G,56.06

55.354.153.354.7

55.089.7

89.6<7r

ofS26.5

27.721.6

25.128.0

28.24.44.1%

ofG2/M18.6

18.325.3

22.617.7

17.75.9

6.3Cell

densityI.25x

lO-Vml1.48X

I05/ml2.8

xloVml3.55

x 10-Vml

3037

American Association for Cancer Research Copyright © 1993 on November 8, 2012cancerres.aacrjournals.orgDownloaded from

HI:ATSHOCKPKOTLINSIN DIHI.RI.NIIAIION

Fig. 6. Reversibility of the fall in hsc70 levelsafter NMF treatment for various times. Followingexposure of HL-60 eells to 170 IHMNMF for thetimes indicated, cells were allowed a 24-h drug-freeperiod before proteins were separated by SDS-polyacrylamide gel electrophoresis and immunob-lotted with the use of the antibody 3A3 (see "Materials and Methods").

irreversible commitment to terminal differentiation (Fig. 6).Early changes in transcription and in mRNA were followed by a fall

in hscVO protein levels by about 12 h. We were concerned that thesefalls in hsc7() levels might simply reflect accumulating numbers ofcells moving out the cell cycle and to growth arrest, since the stress-

inducible hsp7() had been shown to be cell cycle regulated (41).Analysis of the cell cycle and measurement of cell numbers did notsupport this hypothesis (Table I ). Additional experiments, not reported here, have shown that hscVO protein levels remained unchanged when HL-60 cells were treated with 300 RIMNMF. which

induced a rapid cytostasis. Recently, it was reported that TPA inducedan increase in the stress-inducible hsp70 mRNA in HL-60 cells com

mitted to monocyte differentiation (9). These cells were rapidlygrowth arrested and it is possible that these changes may reflectchanges in cell cycle rather than differentiation.

The loss of proliferative capacity of NMF-treated HL-60 cells co

incided with their commitment to differentiate, which has been relatedto changes in the expression of the protooncogene c-myc (3. 4). We

were interested in the temporal relationship, if any. between changesin the nuclear c-mvc protein and hsc70, because hsc70 has been

implicated in the translocation of proteins from the cytoplasm to thenucleus (36). Additionally, there has been an association reportedbetween a heat shock protein and c-myc (19, 42). This raised the

possibility that the fall in hsc70 might be associated with changes inc-myc expression in differentiating HL-60 cells. Immunoblottingshowed that the fall in cellular levels of c-myc after NMF treatment

occurred within 2 h (Fig. 5B), whereas changes in hsc70 levels did notoccur until after 12 h (Fig. I ). This disparity of kinetic change does notsupport the idea that hsc7() may be involved with changes in c-mvc

levels.Interestingly, the rapid tall in c-myc protein levels occurred well

before any changes were observed in growth or differentiation of theHL-60 cells. Furthermore, the cells initiated another full round ofdivision in the absence of detectable levels of c-myc protein. Similar

observations have been reported by other workers who have suggestedthat the kinetics of the decline in c-myc varies both with the nature of

the inducing agent and the lineage to which the cells become committed (16. 17. 43).

The loss of a constitutively expressed protein, which has a role asa molecular chápenme, from cells committed to differentiate by NMFraises a number of important questions. How NMF brings about thischange and how general it is for agents which induce granulocyte-

neutrophil differentiation remains to be answered. Our immediate aimis to discover which prolein(s) is normally associated with hsc7()under basal conditions and whether the loss of their protein chaperonemay play a critical role in the block in differentiation, indicative of themalignant phenotypc. Additionally, we are interested in how a simplemolecule like NMF can initiate subtle and selective changes in geneexpression and posttranscriptional control.

i38S 8h?

?

§ 12h §116h I 24h36h

+24hl36h—

«*- -_evels

was reversible onlynich corresponded to theREFERENCES1.Birnie. G. D. The HL-60 eelline: a model system tor slutying human ilyeloid cell

10.

14.

15.

16.

17.

18.

19.

20.

23.

24.

25.

26.

3038

differentiation. Br. J. Cancer (SuppU V: 41-45. 1988.Hickman. J. A., and Friedman. R. M. Pharmacology and mechanisms of action ofdifferentiating agents. In: S. Waxman. G. B. Rossi, and F. Takaku (eds.K The Statusof Differentiation Therapy, pp. 231-260. New York: Raven Press. 1988.Filtnus, J.. and Buick. R. N. Relationship of c-m\r expression to differentiation andproliferation of HL-W) cells. Cancer Res., -¿5:822-825, 1985.Collins. S. J. The HL-60 promyelocytic leukemia cell line: proliferation, differentiation and cellular oncogene expression. Blood. 70: 1233-1244. 19X7.Langdon. S. P., and Hickman, J. A. Correlation between the molecular weight andpotency of polar compounds which induce the differentiation of HL-60 human promyelocytic leukemia cells. Cancer Res., 47: 140-144, 1987.Richards. F. M.. Watson. A., and Hickman. J. A. Investigation of the effects of heatshock and agents which induce a heat shock response on the induction of differentiation of HL-60 cells. Cancer Res.. 48: 6715-6720. 1988.Hensold. J. O.. and Housman. D. E. Decreased expression of the stress protein hsp70is an early event in murine erythroleukemic cell differentiation. Mol. Cell Biol. K:2219-2223, 1988.

Bensaude. O.. Babinet. O. Morange. M.. and Jacob. F. Heal shock proteins, firstmajor products of zygotic gene activity in mouse embryo. Nature (Lond.). JOS:331-333, 1983.

Shakoori. A. R.. Oberdorf. A. M.. Owen. T. A.. Weber. L. A.. Hickey. E.. Stein. J. L..Lian. J. B.. and Stein. G. S. Expression of heat shock genes during differentiation ofmammalian osteoblasts and promyelocytic leukemia cells. J. Cell Biochem.. 48:277-287, 1992.

Perm. M. T.. Soderstrom. K.. Jindal. S.. Gronberg. A.. Ivanyi, J.. Young. R., andKiessling. R. Induction of human Iisp60 expression in monocytic cell lines. Int.Immunol.. 4: 305-311. 1992.Theodorakis. N. G., Zand. D. J.. Kotzbauer. P. T.. Williams. G. T.. and Morimoto. R.I. Hemin-induced transcriplional activation of the /i.v/>7()gene during erythroid maturation in K562 cells is due to a heat shock factor mediated stress response. Mol. CellBiol., 9: .3166-3173. 1989.Lindquist. S. The heat shock response. Annu. Rev. Biochem.. 55: 1151-1191, 1986.Morimolo, R. I., and Milarski. K. L. Expression and function of vertebrate hsp70genes, in: R. I. Morimoto. A. Tissieres. and C. Georgopoulos (eds.). Stress Proteinsin Biology and Medicine, pp. 323-359. Cold Spring Harbor. NY: Cold Spring HarborLaboratory. 1990.Milarski. K. L.. Welch. W. J., and Morimoto. R. I. Cell cycle-dependent associationof hsp70 with specific cellular proteins. J. Cell Biol., /OH: 413-423. 1989.Eilers, M.. Picard. D.. Yamamoto. K. R., and Bishop, M. J. Chimaeras of Myconcoprotein and steroid receptors cause hormone-dependent transformation of cells.Nature (Lond.). J4Ü:66-68. 1989.Waters. C.. Littlewood. T.. Hancock. D.. Moore. J.. and Evan, G. c-myc proteinexpression in untransformed fibroblasts. Oncogene. 6: 101-109. |oi)|Vass, J. K.. Neill. R.. Jamieson, T.. and Birnie, G. D. Regulation of the relativeabundances of c-mvf niRNAs in human promyelocytic HL-60 cells. Differentiation,45: 49-54. 1990.

Mitchell. L. S.. Neill. R. A., and Birnie. G. D. Temporal relationships between inducedchanges in C-/MVCmRNA abundance, proliteration, and differentiation in HL-60 cells.Differentiation! 49: 119-125. 1991.Koskinen. P. J.. Sistonen. L.. Evan. G.. Morimoto. R.. and Alitalo. K. Nuclearcolocalizalion of cellular and viral myc proteins with HSP70 in myc-overexpressingcells. Mol. Cell Biol.. 65: 842-851. Ì99I.Morimoto. R. I. Heat shock: the role of transient inducible responses in cell damage,transformation, and differentiation. Cancer Cells (Cold Spring Harbor), 3: 295-301,1991.Riabowol. K. T.. Mizzen. L. A., and Welch. W. J. Heat shock is lethal to fibroblastsmicroinjecled with antibodies against hsp7(). Science (Washington DC). 242: 433-436. 1988.Gething. M. J.. and Sambrook. J. Protein folding in the cell. Nature (Lond.). 355:33-45. 1992.Ellis, R. J.. and van der Vies. S. M. Molecular chaperones. Annu. Rev. Biochem.. M:321-347. 1991.

Spremulli. E. N.. and Dexter. D. N. Polar solvents: a novel class of antineoplasticagents. J. Clin. Oncol.. 2: 227-241. 1984.Langdon. S. P.. and Hickman. J. A. Alkylformamides as inducers of tumour celldifferentiation—a minireview. Toxicology. 43: 239-249. 1987.Langdon. S. P.. Richards. F. M.. and Hickman. J. A. Investigations of the relationship

American Association for Cancer Research Copyright © 1993 on November 8, 2012cancerres.aacrjournals.orgDownloaded from

HEAT SHOCK PROTEINS IN DIFFERENTIATION

between cell proliferation and differentiation of HL-60 cells induced to differentiateby /V-mclhylformatnide. Leuk. Res., 12: 211-216. 1988.

21. Banerji. S. S., Theodorakis. N. G.. and Morimoto. R. I. Heal shocked-induced Irans-

lational control of hsp70 and globin synthesis in chicken reticulocytes. Mol. CellBiol., 4: 2437-2448. 1984.

28. Wu. B.. Hunt, C.. and Morimolo. R. I. Structure and expression of the gene encodingthe major heat shock protein HSP70. Mol. Cell Biol.. 5: 330-341. 1985.

29. Mickey, E., Brandon. S. E.. Samle, G.. Lloyd. D., and Weber. K. A. Sequence andregulation of a gene encoding a human 89-kilodalton heat shock protein. Mol. CellBiol.. V: 2615-2626. 1989.

30. Watowich. S. S. Ph.D. Dissertation. Northwestern University. Evanston, IL. 1990.31. Gunning. R.. Ponte. P.. Okayama. H.. Engel, J.. Blau, H., and Kedes, L. Isolation and

characterisation of full-length c-DNA clones for human a-, ß-and 7-actin mRNAs:skeletal but not cytoplasmic actins have an amino-terminal cysteine that is subsequently removed. Mol. Cell Biol.. 3: 787-795. 1983.

32. Chirgwin, F., Pryzbyla. A., McDonald, R., and Rotter. W. Isolation of biologicallyactive ribonucleic acid from sources enriched in ribonuclease. Biochemistry, IS:5295-5295. 1979.

33. Ungewickcll. E. The 70-kD mammalian heat shock proteins are structurally and

functionally related to the uncoating protein that releases clathrin triskelia from coatedvesicles. EMBO J.. 4: 3385-3391, 1985.

34. Chappell. T. G.. Welch. W. J.. Schlossman. D. M., and Rolhman, J. E. UncoatingATPase is a member of the 70 kilodallon family of stress proteins. Cell, 45: 3-13,1986.

35. Chiang. H. L.. Terlecky. S. R.. Plant. C. P.. and Dice, J. F. A role for a 70-kilodalton

heat shock protein in lysosomal degradation of intracellular proteins. Science (Washington DCl. 24f>: 382-385. 1989.

36. Shi. Y., and Thomas, J. O. The transport of proteins into the nucleus requires the70-kilodalton heat shock protein or its cytosolic cognate. Mol. Cell Biol., 12: 2186-

2192, 1992.37. Alvares, K.. Carrillo. A.. Yuan. P. M.. Kawano. H.. Morimoto. R. I., and Reddy. J. K.

Identification of cytosolic peroxisome proliferator binding protein as a member of theheat shock protein family. Proc. Nati. Acad. Sci. USA, 87: 5293-5297, 1990.

38. Hoffman-Liebermann. B.. and Lieberman. D. A. Inlerleukin-6- and leukemia inhibitory factor-induced terminal differentiation of myeloid leukemia cell is blocked at anintermediate stage by constitutive c-mvr. Mol. Cell Biol., //: 2375-2381. 1991.

39. Selvakumaran, M., Lieberman, D. A., and Hoffman-Liebcrman. B. Deregulated C-MV/Jdisrupts interleukin-6 or leukemia inhibitory factor-induced myeloid differentiationprior to c-mvr: role in leukemogenesis. Mol. Cell Biol.. 12: 2493-251X). 1992.

40. DiDomenico. B.. Bugaisky. G. E.. and Lindquist. S. The heat shock response isself-regulated at both the transcriptional and posttranscriptional levels. Cell, 31:593-603, 1982.

41. Milarski, K. L.. and Morimolo. R. I. Expression of human hsp7() during the syntheticphase of the cell cycle. Proc. Nati. Acad. Sci. USA. S3: 9517-9521, 1986.

42. Kingston. R. E.. Baldwin. A. S.. Jr.. and Sharp. P. A. Regulation of heal shock prolein70 gene expression by c-myr. Nature (Lond.), 312: 280-283. 1984.

43. Pedrazzoli. P.. Bains, M. A., Watson. R., Fisher, J., Hoy. T. G.. and Jacobs, A. c-mycand c-m\7> oncoproleins during induced maturation of myeloid and erythroid humanleukemic cell lines. Cancer Res.. 49: 6911-6916. 1989.

3039

American Association for Cancer Research Copyright © 1993 on November 8, 2012cancerres.aacrjournals.orgDownloaded from