Vol. 168, No. 3, 1990 May 16, 1990

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Pages 1013-1019

Heat shock induces the transcriptional activation of c-fos protooncogene F. Colotta, N. Polentarutti,

M. Staffko, G. Fincato and A. Mantovani

Istituto di.Ricerche Farmacologiche“Mario Negri”, via Eritrea 62,20157 Milano, Italy Received

March

19,

1990

SUMMARY. Treatment of murine 3T3 fibroblasts with mild elevated temperature (43’C for 45 minutes) followed by recovery at 37’C induced high levels of c-fos mRNA. The maximal c-fos induction was found after recovery at 37’C for 15 minutes. Sodium arseniteinduced both hsp 70 and c-fos transcripts. Induction of hsp70 and c-fos by heat shock did not require intact protein synthesis. c-fos mRNA induced by heat shock wasmore stable(T1/2 > 90 minutes)than that inducedby phorbol esters(t1/2-30 minutes). Northern blot analysisin the presenceof actinomycin D and nuclearmn off experiments demonstratedthat heat shock augmentedthe transcription rate of c-fos protooncogene. Human growth hormone under the control of a murine genomic c-fos fragment spanning770 bp 5 from the start of transcription is induced in transfected cells in responseto heat shock. These data indicate that heat shock induces c-fos protooncogene acting at both transcriptional and post-transcriptional(i.e. via stabilization of transcripts)levels. 01990 Academic Press, Inc.

Exposure to high temperatureand certain other stressesactivatesthe transcription of a smallsetof highly conserved specific genescalled the heat shock (hs) genes(for review seefor example ref. 1). The responseof hs genesis believed to be mediated by a factor which binds to a highly conserved sequence(heat shock element, hse) located 5’ to hs genes.It is generally accepted that heat shock proteins (hsp) may play a role at normal temperaturesin cell developmentandproliferation, andduring stressthey may provide protection from the effects inducedby stress-inducingagents(2). c-fos protooncogeneis expressedin a variety of cell types and under many different experimental conditions. c-fos transcripts have been described in actively proliferating cells (3) as well as in terminally differentiated cells and as a consequenceof functional activation even in the absenceof proliferation (4-7). It is now well establishedthat c-fos genecodesfor a protein which associateswith jun-family products forming a complex with the feature of a transcriptional activator (for review see ref. 8). Certain agentsknown to modulatec-fos expressionhave beenshownto affect the transcription rate of c-fos protooncogene:positive regulatory sequencesin 5’ untranslatedregion of c-fos have been describedwhich respondto phorbol esters,serumstimulation, cyclic AMP and conditioned medium from v-sis transfected cells (9). Post-transcriptional mechanismsalso may play a role in regulating c-fos expression. c-fos mRNA shows a very short half-life and multiple AUUUA motifs in the 3 untranslatedregion have beenimplicated in shorteningtranscript stability (10). The present study was designedto evaluate the possibility that heat shock could modulatec-fos expression.Data presentedherein demonstratethat heat shock augmentsc-fos expressionby acting at both the transcriptional andpost-transcriptional(via stabilization of transcripts)levels. 0006-291X&O

1013

$1.50

Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.

Vol. 166, No. 3, 1990

MATERIALS

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

AND METHODS

Cell culture and reagents: NIH-3T3 fibroblasts were cultivated in Dulbecco’s modified minimal essential medium (D-MEM, Gibco, Glasgow, United Kingdom) with 10% fetal calf serum (FCS, Gibco), 2 mM glutamine (Gibco) and 1 mM HEPES (Merck, Rahway, NJ), hereafter referred to as complete medium, at 37’C in air with 5% CO* in 25 cm* plastic flasks (Sterilin, Feltham, UK).Sodium arsenite, cycloheximide, actinomycin D and 12-phorbol-13-myristate acetate (PMA) were from Sigma, St. Louis, MO. Heat shock: Flasks containing 4 x lo6 3T3 fiboblasts grown at confluence were incubated for 45 minutes in a thermostatic bath at 43’CfO.l and subjected to RNA separation or allowed to recover at 37°C for various periods of time. Control cultures were kept at 37’C.Sodium arsenite was padded to cells in complete medium at 37’C (final concentration 80 PM) for 1 h. Northern blot: Total RNA was isolated by the method of guanidine isothiocyanate (11). 15 pg RNA were analyzed by electrophoresis through 1% agarose formaldeyde gels, followed by transfer to nylon membranes (Gene Screen Plus, New England Nuclear, Boston MA). Quality and quantity of RNA blotted on membranes was checked by short wave UV. Plasmids containing the murine genomic clone of c-fos (pc-fos 3) (12) and the human hsp70 cDNA (13) were nick-translated with a132PldCTP (3000 Ci/mmol; Amersham, Buckingamshire, UK). Membranes were pretreated and hybridized in 50% formamide (Merck) -10% dextran sulfate. Membranes were washed twice with 2X SSC-1% sodium dodecyl sulfate at 60°C for 30 min. and twice with 0.1X SSC at room temperature for 30 min. The membranes were exposed for 12-24 h. with intensifying screens at -8O’C. Nuclear run-off: Run-off experiments were carried out as described in details elsewhere (14). a132PlUTP(3000 Ci/mmol, Amersham)-labeled transcripts were isolated from 20 x lo6 3T3 fibroblasts for each experimental group, and hybridized (5 x lo6 dpm from each experimental group) to pc-fos3 and hsp70 plasmids (5 yg each) immobilized to membranes. As a negative hybridization control, we used an equal amounts of pBR322. Transfection experiments: The murine c-fos promoter region, a 772 bp SstI-AccI fragment, was excised from pc-fos(mouse)3 (12) and blunt-end ligated into the Hind111 site of the polylinker of OGH vector containing the human growth hormone (hGI-I) gene (13). We have termed this vector pc-fos GH. 3T3 fibroblasts were transfected by electroporation (11). Cells were resuspended in phosphate buffered saline (without Ca++ and Mg’+, PBS, Gibco) at 15 x lo6 cells in 0.4 ml, to which 20 ltg of pc-fos GH were added. Electroporation was performed with a single pulse from a home-made apparatus at 990 V, 450 PFd. After electroporation, cells were incubated on ice for 10 minutes and seeded in plastic flasks in complete medium. After 48-96 hours confluent cells were washed with PBS and feeded with D-MEM containing lmglml bovine serum albumine (Sigma) without FCS and glutamine. 12 hours later cells were either untreated or added with FCS (final concentration 20%) or subjected to heat shock at 43’C for 20 minutes and then at 37OC for lo-60 minutes. After each time point, supematant was withdrawn and assayed for hGH production. hGH assay: 100 ~1 medium from transfected cells were assayed in tripicate for hGH levels using the Allegro hGH kit (Nicolas Institute Dignostics, San Juan Capistrano, CA), a solid-phase RIA, acceding to manufacturer’s instructions. l*‘I-labeled antibody was measured using a gamma-counter. RESULTS To study the induction of c-fos protooncogene by heat shock, murine NIH-3T3 fibroblasts were exposed to mild elevated temperature (43’C for 45 minutes) and allowed to recover at 37’C for various lenghts of time. As a reference heat-inducible gene, we analysed the expression of heat shock protein 70 (hsp 70) gene (13), one.of the most highly conserved and expressed heat shock genes. As expected (16), untreated murine cells express a constitutive hsp 70-related mRNA of approximately 2.4 kb (Fig. lA, lane l), whereas heat-treated (43’C. 45 minutes) cells expressed a band of hybridization of 2.7 kb (Fig. lA, lane 2). The 2.4 kb transcripts expressed in cells cultivated at normal temperature belong to a heat shock cognate 70 (hsc 70) gene (16). The intensity of the heat-induced transcripts of 2.7 kb was increased when cells were allowed to recover at 37’C for 15 minutes (Fig. 1A; lane 3); thereafter, the expression of 2.7 kb decreased (Fig. 1A; lanes 4-5). After removal of the hsp 70 probe, the same filter shown in Fig. 1A was hybridized to a murine c-fos probe (Fig. 1B). As expected (3), 1014

Vol.

BIOCHEMICAL

168, Nd. 3, 1990

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

Fig. 1: Hsp70 gene(panelA) andc-fosprotooncogene (panelB) expressionin murineNIH-3T3 fibroblastsexposedto heatshockor sodiumarsenite. RNA samples wereasfollows: 1:untreatedconfluent-arrested NIB-373 fihroblasts;2: heat-shocked cells (43’C, 4.5minutes);3: heat-shockedcells allowed to recover at 37°C for 15 minutes;4: heat-shocked cellsallowedto recoverat 37’C for 30minutes;5: heat-shocked cellsallowedto recover at 37’C for 60minutes;6: sodiumarsenite-treated cells(80p.M, 1 hourat 37‘9.

untreated confluent-arrested 3T3 fibroblasts do not express c-fos protooncogene (Fig. lB, lane 1). However, heat-treated cells express detectable levels of c-fos transcripts (Fig. lB, lane 2) which increase when cells are further cultured at 37’C for 15 minutes (Fig. lB, lane 3). After a more prolongedtime of recovery (30 and 60 minutes)the expressionof c-fos decreased(Fig lB, lanes4 and 5). Thus heat shockedmu&e 3T3 fibroblastsexpressthe c-fos protooncogenein addition to a classical heat shock gene. Heat shock genesare induced not only by heat treatment, but also by stressessuch assodium arsenite,amino acid analogues,ethanol and transition seriesmetals(1). Thus, we wished to investigate whether treatments other than heat shock are able to induce, in addition to hsp 70 gene, c-fos protooncogeneexpression.As shown in Fig. 1A and lB, lane 6, treatment of 3T3 fibroblasts with 80 PM sodium arsenite (SA) for 1 hour induced hsp 70 transcripts as well as c-fos expression. The maximal induction of hsp 70 gene by SA was found after 2 hours of culture at 37’C (Fig. 2A, lane 7), whereasc-fos expressionwas more evident after 30 minutes of culture at 37“C (Fig. 2B, lane 6) and

12345676

a

2.2

1

2

34

5

678

Kb-

Fig. Effect of metabolicinhibitorson expressionof hsp70 (panelA) andc-fos (panelB) after exposureto heatshock. RNA samples wereasfollows: 1: untreatedconfluent-arrested NIB-3T3 cells;2: heat-shocked cells (43*Cfor 45 minutes)allowedto recoverat 37*Cfor 15minutes;3: asin lane2, with theadditionof cycloheximide(10 p#nl) duringthe recovery phaseat 37’C, 4: asin lane2, with the additionof cycloheximideduringbothheatshockandrecovery;5: asin lane2. in thepresence of actinomyciuD (1 @ml); 6-8: sodiumarsenitetreatment(80@f) for, respectively.34 60and120minutesat 37’C. 1015

BIOCHEMICAL

Vol. 168, No. 3, 1990

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

decreasedat 1 and 2 hours (Fig. 2B, lanes7-8). Thesedata demonstratethat, in addition to heat shock, sodiumarsenite,a classicalinducer of heat shock geneexpression,is able to induce both hsp 70 and c-fos transcripts.Similar resultswere obtained(not shown) by treating 3T3 fibroblsts with 5 x 10” M Cu- and Zn-sulphate for lh. Induction of heat shock genesdoes not require an active protein synthesis, suggestingthat induction involves modification of an existing protein (1). Therefore we examinedwhether induction of c-fos by heat shockwas influenced by the simultaneouspresenceof cycloheximide. As expected, heat shock induceshsp 70 geneexpressionalso when cycloheximide (10 ~g/rnl) was presentduring both the “induction” (at 43’C for 45 minutes) and the “recovery” phase(at 37°C for 15 minutes, Fig. 2A, lanes3 and 4). Similarly, c-fos expressionwas not inhibited by treatment of 3T3 fibroblasts with the protein synthesis inhibitor (Fig. 2B). The expression of c-fos remained unchanged when cycloheximide was added during the “recovery” phase at 37’C after heat shock (Fig. 2B, lane 3), whereas the hybridization signal was more intense when cycloheximide was present both during “induction” and “recovery” phase(Fig. 2B, lane 4): this effect could be related to a superinductionof c-fos expressionby 1 hour treatmentwith cycloheximide (17) in heat shockedcells. Treatment of 3T3 cells with cycloheximide at 37°C for 1 hour did not induce detectable levels of c-fos (not shown). Thus, as for classicalheat shock genes,the induction by heat shock of c-fos doesnot require an intact protein syntesisof target cells. To understandthe mechanism(s)of c-fos induction by heat shock, cells were exposedto 43OC and then incubated at 37’C in the presenceof actinomycin D for various periodsof time, in order to evaluate the influence of heat on transcript stability. Whereas PMA-induced c-fos transcriptsin 3T3 cells have the expected short half-life (-30 minutes) (3), the half-life of c-fos transcripts induced by heat shockin 3T3 fibroblastswas found to be longer than 90 minutes(Fig 3). To assesswhether heat shock augment c-fos expression via enhanced transcription, we followed three experimental approaces.First, cells were heat shockedin the presenceof actinomycin D. As predicted, the induction of hsp 70 gene expression(the heat inducible band of 2.7 kb) by heat treatment was completely blocked by actinomycin D( Fig. 2A, lane 5). The sameresult wasobtained when we examined c-fos expression (Fig. 2B, lane 5). Second, we carried out nuclear run off experiments on nuclei isolated from control and heat-shocked 3T3 fibroblsts. As shown in Fig. 4,

A 1234

5670

6

C

9 I 2

0

3

4

Fip(. Determination of c-fostranscripthalf life in 3T3cellstmatedwith PMA or heatshock. RNA samples wereasfollows:1: untreated3T3 fibmblasts;2: cellstmatedwith PMA (10rig/ml)for 1 hour.;3-5: asiu lane2, followedby treatmentwith ActD (l@nl) for 15minutes,30and60minutes respectively.;6: cellstratedat 43’C for 45minutesandthenallowedto recoverat 37% for 15minutes.; 7-9: asin lane6, followedby treatmentwith ActD (1 p&l) for 1530 aud60 minutesrespectively. Fi .4: Nuclearrun-off of c-fostranscriptsin heattreated3T3 fibroblasta. +*ranscnptswerefrom fibroblatsin serumfreemediumwithouttreatment(A), tmatedat 43°Cfor 45 mm.andthenallowedto recoverat 37’C for 15min. (B), or addedwith 20%FCSfor 15mm.at 37°C (C). DNA samples hybridizedto membranes were:1: hsp70;2: c-fos;3: pBR322. 1016

Vol.

BIOCHEMICAL

168, No. 3, 1990

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

Table 1: Production of humangrowthhormon(hGH) in 3T3cellstransfected with c-foshGH

vector andtreatedwith fetalcalf serumor heatshock

Treatment

Heatshockb FCSC

10

SecretedhGHa:Timeafterstimulus(minutes) 20 30 45

lOZt5d

79f7

85flO

104fll

115*12 63zt5

2oof19

273f21

126rtll

18of15

29lf32 209f21

60 25Wl 546k37 526k45

a: 48 hoursaftertransfection(seeMaterialsandMethods),cellswerewashedandfeededwith serum-freemedium.After 12h. cellswereeitheruntreatedor addedwith FCSor heatedasdetailed below.Then, after appropriateperiodsof time, 0.1 ml of mediumwasassayedfor hGH levelsin triplicate. The resultantcountsper minuteof I lz5-labeledanti-hGH antibody houndto the solid supportareshown.Reportedvaluesarecorrectedcountsper minute@pm)=total cpm- backgmud cpm. b: 3T3 cells weretreatedat 43’C for 20 minutes;thenthe mediumwaschangedandcells incubatedat 37’C for the variousperiodsof time, asreportedin the Tab1e.c:3T3 cellswereadded with 20%(final concentration) FCS.d: Standarddeviationof threesamples.

whereasthe transcriptionsof both hsp70and c-fos geneswere undetectablein untreatedcells, a strong hybridization signal was found in heat shockedcells (Fig. 4, B). Finally, we have subcloneda murine c-fos 5’ flanking region (spanning770 bp from the start of transcription) near the human growth hormone(hGH) gene.This vector (pcfos-hGH) hasbeenuansfectedin 3T3 fibroblasts. Serum-starved cells were then exposed to fetal calf serum(a classical inducer of c-fos promoter, ref. 3) or to heat shock, and levels of secretedhGH determined. A representativeexperiment is shown in Table 1, and similar results have been obtained in six other different experiments. As expected (3), the addition of seruminducedincreasedlevels of hGH comparedto thoseproducedby untreatedcells; along the same line, high levels of hGH were produced by transfectedcells exposedto heat shock. After 30 minutes we observed a 2 fold (for serum) and 3.2 fold (for heat shock) increase of secreted hGH levels comparedto untreatedcells. All thesedata strongly supportthe conclusionthat heat shockinducesthe transcriptionalactivation of c-fos protooncogene.

DISCUSSION The data presentedhere demonstratethat heat shock and stresseswhich activate the heat shock response(e.g. sodium arsenite)induce the expressionof the c-fos protooncogene.Induction of c-fos expression dependsupon, on the one hand, an increasedstability of c-fos transcripts, and, on the other, an augmentedtranscription of c-fos gene.The induction of c-fos by heat shock sharesa number of properties (kinetics, lack of requirement of protein synthesis,transcriptional control of induction) with activation of expressionof classicalhsp70 gene.Moreover, by transient transfection experiments we demonstratedthat the c-fos promoter is inducible by heat shock. c-fos protooncogene is expressedin several different cell types and in responseto a large variety of external stimuli which promote cell proliferation (3), cell differentiation and functional activation of terminally differentiated cells (4-7). c-fos codes for a protein which associateswith jun-family coded factors, forming a heterodimerwith the characteristicsof a transcriptional activator (8). Thus, induction of c-fos may play a key role in mediating the molecular events in responseto 1017

Vol.

168, No. 3, 1990

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

external stimuli. Stimuli of c-fos expressionhave been shown to act either via posttranscriptional mechanisms(e.g. augmentedhalf life of transcripts) or activation of c-fos transcription. c-fos mRNA has a very short half life (approximately 30 minutes) and multiple AUUUA motifs in the 3 untranslatedregion which have implicated in shortening transcript stability (10). Certain treatments, including the inhibition of protein synthesis,increasec-fos mRNA stability, presumably affecting a labile protein which hasbeen involved in c-fos transcript degradation. Two previous studieshave shown that heat shock augmentsc-fos transcript half-life (18,19), probably as a consequenceof the degradation of proteins and of the inhibition of protein synthesiswhich follow heat treatment. The presentstudy, though confirming that heat shock augmentsc-fos transcript half-life, indicatesthat this is not the only nor the major mechanismof induction of c-fos sinceheat shockinducesa greatincrease of c-fos

transcription

rate

as assessed

by Northern

blot

analysis

in

the presence

of a transcriptional

block (e.g. with actinomycin D) and by nuclear run off experiments. The transcription rate of c-fos induced by heat shock is several folds higher than untreated cells (see Fig. 4) being in the latter undetectableeven after long autoradiographicexposure,thus indicating that this mechanismis relevant to the overall induction of expressionof c-fos. Induction of transcription of c-fos has been shown to be regulated by positive regulatory elementslocatedin the 5’ untranslatedregion of the gene(9). Transcriptionalactivation of classicalheat shock genesis mediatedby a short highly conserved sequencetermed heat shock element (HSE) (1). The finding reportedhere that a constructcontaining murine c-fos promoterupstreamto a reporter gene is expressedin transfectedcells exposedto heat shockis suggestivethat c-fos promoter could contain a sequenceresponsiveto stress.Computer analysis of murine and human c-fos sequencesrevealed a region 410 and 406 bp respectively upstreamto TATA box with homology to HSE. Whether this region is indeedresponsiveto stressremainsto be elucidated. In addition to being induced by a seriesof external stimuli related to cell proliferation (3), differentiation and functional activation (4-7), c-fos protooncogeneis induced by a seriesof diverse conditionsranging from wounding of cell monolayer (20), to induction of seizuresin the central nervous system(21), to treatment with barium (22), to birth (23). The resultspresentedhere, by showing that c-fos expressionis part of the stereotypedresponseto certain stresses,known asheat shockresponse, provides a conceptual framework in which to interpret the induction of transcription of this protooncogene by such diverse stimuli. Moreover, these findings, by establishing a relationship between one protooncogeneand heat shock genes,raisethe interestingpossibility, which needsto be explored, that sucha relationshipmay extend beyond the particular caseexaminedin this study.

Acknowledgment: Biotecnologie.

This work was supported by Consiglio Nazionale delle Ricerche, p.f.

REFERENCES 1. 2. 3. 4.

Lindquist, S. (1986) Annu. Rev. B&hem. 55, 11.51-l191. Pelham, H. (1988) Cell 46,959-961. Muller. R. (1986) Biochim. Biophy. Acta 823,207-225. Introna, M., Hamilton, T.A., Kaufman, R.E.. Adams, D.O., and Bast Jr., R.C. (1986) J. Immunol. 137,2711-2715. 5. Colotta, F., Wang, J.M., Polentarutti, N., and Mantovani, A. (1987) 3. Exp. Med. 165, 1224-1229. 1018

Vol.

168,

No.

3, 1990

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

6. Radzioch, D., Bottazzi, B., and Varesio, L. (1987) Mol. Cell. Biol. 7,595-599. 7. Colotta, F., Lampugnani, M.G., Polentarutti, N., Dejana, E., and Mantovani, A. (1988) B&him. Biophys. Res. Comm. 152, 1104-1110. 8. Cm-ran, T., and Franza Jr., B.R. (1988) Cell 55, 395-397. 9. Verma, I.M., and Sassone-Corsi, P. (1987) Cell 51,513-514. 10. Shaw, G., and Kamen, R. (1986) Cell 46,659-667. 11. Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular cloning. A laboratory manual. Second edition. Cols Spring Harbor Laboratry Press. New York. 12. Miller, A.D., Curran, T., and Verma, I.M. (1984) Cell 36,51-60. 13. Wu, B.J., and Morimoto, R.I. (1985) Proc. Natl. Acad. Sci. USA 82,6070-6074. 14. Bertani, A., Polentarutti, N., Sica, A., Rambaldi, A., Mantovani, A., and Colotta, F. (1989) Blood 74, 1811-1816. 15. Selden, R.F., Howie, K.B., Rowe, M.E., Goodman, H.M., and Moore, D.D. (1986) Mol. Cell. Biol. 6, 3173-3179. 16. Lowe, D.G., and Moran, L.A. (1984) Proc. Natl. Acad. Sci. USA 81,2317-2321. 17. Mitchell, R.L., Zokas, L., Schreiber, R.D., and Verma, I.M. (1985) Cell 40,209-215. 18. Andrews, G.K., Harding M.A., Calvet, J.P., and Adamson, E.D. (1987) Mol. Cell. Biol. 7, 3452-3458. 19. Hollander, M.C., and Fornace Jr., A.J. (1989) Cancer Res. 49, 1687-1692. 20. Verrier, B., Muller, D., Bravo, R., and Muller, R. (1986) EMBO J. 5,913-917. 21. Dragunow, M., and Robertson, H.A. (1987) Nature 329,441~442. 22. Curran, T., and Morgan, J.I. (1986) Proc. Natl. Acad. Sci. USA 83, 8521-8524. 23. Kasik, J.W., Wan, Y.-J., and Ozato, K. (1987) Mol. Cell. Biol. 7, 3349-3352.

1019

Heat shock induces the transcriptional activation of c-fos protooncogene.

Treatment of murine 3T3 fibroblasts with mild elevated temperature (43 degrees C for 45 minutes) followed by recovery at 37 degrees C induced high lev...
1MB Sizes 0 Downloads 0 Views