Heat-Shock-induced Denaturation of Proteins - The Journal of ...

7 downloads 0 Views 3MB Size Report
HeLa cells was carried out as described by Capco et al. (26). Cells .... Interferon-treated cells were submitted (H, 11, F) or not (A, C, E) to a 1-h heat-shock at 44 ...
Val. 266, No. 15, Issue of May 25, pp. 9707-9711,1991 Printed in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY 1991 by The American Soclety for Biochemistry and Molecular Biology, Inc.

ic:

Heat-Shock-inducedDenaturation of Proteins CHARACTERIZATIONOF

T H E INSOLUBILIZATIONOF THEINTERFERON-INDUCED p68 KINASE* (Received for publication, November 26, 1990)

Marie FranqoiseDubois, Ara G. Hovanessianz, and Olivier Bensaude From the Groupe de Hiologie Moleculaire du Stress Ecole Normale Superieure, 46 rue d'Ulm, 7,5230 Paris, Ceder 05, and $Unite de Virulogie et Immunologie Cellulaire, Institut Pasteur, 25 rue du Dr Koux, 75015 Paris, France

Heat-shock stress causes inactivation and aggregation of various cellular proteins which become further insoluble. Previous studies have shown that the interferon-induced p68 kinase activity was greatly reduced in extracts of heat-shocked HeLa cells, and that the loss of activity wasdue to a decreased solubility of the enzyme. Here we show that the p68 kinase which is normally evenly distributed in the cytoplasm, aggregatesasa thick ring around the nucleus in heatshocked cells. The 70-kDa constitutive heat-shock proteins aremajor insolubilized proteins during stress and we find them to colocalize with the p68 kinase after stress. Treatments of cells with drugs which disrupt the cytoskeleton, such as colcemid and cytochalasin E, do nothinder the enzyme insolubilization during heatshock. On the contrary, heat-protectors such as glycerol and deuterium oxide (DzO) keep the p68 kinase under a soluble and active form during heat-shock stress. Similarly,an attenuation of the insolubilization of this enzyme is observed in cells rendered thermotolerant by a previous heat-shock, suggesting that heat-shock proteins may also contribute to the protection. During the recovery period at normal temperature after heat-shock, resolubilization occurs and most of the enzyme is again recovered under an active soluble form.

Heat-shock results in the inactivation of various cellular functionssuchas DNA replication (l),RNA splicing (2), response to hormones (3-5), and protein synthesis. Themolecular basis of these inactivations is not fully understood. Posttranslational modifications driven by the stress might lead to the inactivation of proteinfunction. For instance, modification of the phosphorylation state of the initiation factors e1F-h and eIF-4B might contribute to the protein synthesis inactivation (6, 7). Alternatively, heat denaturation of proteins has been proposed to occur during stress on the basis of calorimetric measurements (8) and of the protecting effects of substances which stabilize protein conformation (9). Such a heat denaturation may be responsible for the loss of protein solubility which has been shown to occur during heat stress.Thus cytoplasmicsolubleforeign reporter enzymes such as luciferase and /?-galactosidase have been found to be insolubilized during heat stress (10). Insolubilization of nuclear oncogenes, protooncogenes (11-13), and nuclear ribo* This work was supported by Grants ARC 6308 and 6250 from the Association pour la Recherche sur le Cancer, Villejuif, France. The costs of publication of this article were defrayed in part by the payment of page charges. Thisarticlemusttherefore be hereby marked "advertisemc,nt" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

nucleoproteins has been described (14), but this might be a consequence of the overall structural changes which lead to the increase in protein contentof nuclei after heat-shock(15). It is well known that cells can be rendered more heatresistant by a mild heat-shock treatment sufficient to induce the synthesisof the stress proteins (16). Such thermotolerant cells recover their cytoskeleton morphology and normal macromolecule synthesisfasterthancontrol cells (2, 17, 18). Foreign reporter enzymes are less susceptible to heat inactivation in thermotolerant cells (lo), but again there are little data concerning endogenous-characterized proteins. We hadpreviously noticedthat some aspects of the antiviral state induced by interferon were abolished by heat-shock (19). Thus, in an effort to understand the molecular basis of such an impairment, we analyzed the behavior during heat-shock of two interferon-inducible enzymes involved in the establishment of the antiviral state: the dsRNA'-dependent p68 kinase and the 2-5A synthetase (20). The p68 kinase which becomes autophosphorylated when activatedby dsRNA is inactivated by heat-shock, and this inactivation corresponds toa loss of solubility. Incontrast,the100-kDa form(p100) of 2-5A synthetase remains soluble in the cytoplasm during the same treatment (21). In this report, we first examined the localization of the insolubilized p68 kinase, and then we considered the effect of treatments which are known to protect against the toxic effects of heat-shock. In parallel, we have analyzed the behavior of the 70-kDa constitutive heat-shock proteins which had previously been noticed to be the major insolubilized proteins during stress (22) and which we find tocolocalize with the p68 kinase after stress. MATERIALSANDMETHODS

Reagents-[y-'"PIATP (>5000 Ci/mmol) and [:"S]methionine ( X 0 0 Ci/mmol) were supplied by Amersham Corp. Human leucocyte ( a ) interferon (2 X 10' National Institutes of Health units/mg of protein) was the gift of Institut Pasteur Production. Themonoclonal antibodies specific for p68 kinase (23) and pl002-5A synthetase (24) were prepared asdescribed. Anti-HSP 70 mouse monoclonal antibody G10C9, reacted against both hsc 73 and hsx 70 proteins which are constitutively expressed in HeLa cells, and was kindly provided by Dr D. Mattei from the Institut Past.eur (25). Cells-Human HeLa cells were grown at 37 "C in Dulbecco's modified Eagle's medium (GIBCO) containing 10% newborn calf serum. Treatment of cells with interferon was at 1000 units/ml for 18 h. Treatment of Cells with Colcemid, Cytochalasin E, and HeutInterferon-treated HeLa cells were treated with colcemid (2 VM) for 5 h and cytochalasin E ( 2 pg/ml) was added for the last 2 h at 37 "C. Then cells were heated for 1 h at 44 "C, by immersion of the flasks in a water bath regulated with 0.1 "C accuracy, before lysis in buffer A (20 mM Tris-HC1, pH 7.5, 5 mM MgAc, 120 mM KCI, 1 mM dithiothreitol, 10% glycerol, and 0.5%nonionic detergentNonidet P-40).

' The abbreviations used are: dsRNA, double-strand RNA; PIPES, 1,4-piperazinediethanesulfonicacid.

9707

p68 Kinase Aggregation during

9708

Treatment of Cells with Glycerol, D20, and Heat-Subconfluent cells were exposed to D20 medium (powdered Dulbecco's modified Eagle's medium, GIBCO, dissolved in D20) or glycerol medium (Dulbecco's modified Eagle's medium adjusted to l M glycerol). After 30min preincubation at 37 "C, cells were heat-shocked during 1 h a t 44 "C in the same medium. A t the end of the heat-shock, they were lysed in buffer A or incubated in methionine-free medium containing 10% calf serum and ["S]methionine (40 pCi/ml) for 16 h. Cells were then lysedin Laemmli buffer (2% sodium dodecyl sulfate, 10% glycerol,100 mM dithiothreitol, 60 mM Tris, pH 6.8, and 0.001% bromphenol blue), and the labeled proteins were separated by onedimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis. Cell Fractionation-Cells were first washed with buffer B (35 mM Tris-HCI, pH 7.5, 140 mM NaCI, and 3 mM MgAc). Buffer A was then added for 10 min a t 4 "C. Then lysed cells were scrapped off, and thelysate was centrifuged a t 12,000 X g for 10 min. Supernatants and 12,000X g pellets (suspended in acorresponding amount of buffer A: 30 pl/106 cells) were stored a t -80 "C. Subcellular fractionation of HeLa cells was carried out as described by Capco et al. (26). Cells were washed three times with phosphate-buffered saline and cytoskeleton buffer (CSK) (100 mM NaCl, 300 mM sucrose, 10 mM PIPES, pH 6.8, 3 mM MgCln,and 0.5% Triton X-100; 1 ml for 10' cells) was added on cells for 30 min a t 4 "C. The extracted cells were washed three times with CSK buffer. Then cytoskeletal extraction buffer (10 mM NaCI, 10 mM Tris-HCI, pH 7.4,3 mM MgC12, 1% Tween-40 and 0.5%sodium deoxycholate) was added and the remaining extract scrapped off the dish and homogenized in a Dounce homogenizer. Nuclei were sedimented a t 1,500 X g and lysed in Laemmli buffer. Triton soluble and cytoskeletal fractions were precipited with 5 volumes of acetone a t 4 "C and centrifuged at 12,000 X g. The pellets were then dissolved in Laemmli buffer. Enzyme Activity Analysis-Protein kinase activity was assayed in cytoplasmic extracts incubated with 2 p~ [y-"P]ATP and poly(1). poly(C) g/ml) for 15 min a t 30 "C (19). The phosphorylation reactions were stopped by addition of 2-fold concentrated Laemmli buffer. After heating a t 95 "C for 5 min, phosphoproteins were separated by SDS-polyacrylamide (10%) gel electrophoresis and analyzed by autoradiography. Western Blots-Western blot analyses of HeLa cells extracts were carried out aspreviously described (21) using specific antibodies and "'I-labeled goat anti-mouse immunoglobulins (Du Pont-New England Nuclear). Indirect ImmunofluorescenceAnalysis-Cells were grown on tissue culture chamber/slides (Lab-TEK, Miles) and fixed with acetone/ methanol/formaldehyde (1919:2, v/v) for 5 min a t -20 "C. Then they were incubated with specific antibodies for 60 min with biotinylated anti-mouse Ig for 60 min and with fluorescein-labeled streptavidin for 15 min (Amersham). Slides were mounted with Citifluor (The City University, London) and examined using a fluorescence microscope.

Heat-Shock

A

"T

512

C HS

-0

B-"

-C HS

p12

C HS - 4 ~ 6 8

1

2

3

C HS

C HS

C HS

FIG. 1. Characterization of p68 kinase insolubilization during heat-shock. Interferon-treated cells were submitted ( H S )or not ( C ) to a 44 "C heat-shock for 1 h. A t the end of the heat-shock, cells were lysed in buffer A; Laemmli buffer was added to totallysate (T), to pellets obtained after centrifugation a t 12,000 X g (PIP) and to the corresponding supernatants (S12). Samples were then analyzed by Western blot using monoclonal antibody specific for p68 kinase (A). In B, subcellular fractionation of interferon-treated cells heatshocked (HS)or not ( C )was performed as described under "Materials and Methods." Soluble cytosolic fraction (lanes I ), cytoskeleton fraction (lanes 2) and nuclei (lanes 3 ) were analyzed by Western blot using monoclonal antibodies specific for both p68 kinase and plOO 2-5A synthetase.

RESULTS

Localization of thep68 Kinase in Heat-Shocked HeLa CellsWe have previously shown that, during a 44 "C heat-shock, most of the interferon-induced p68 kinase present in HeLa cells becomes concomitantly insoluble in nonionic detergents and inactive. To further characterize the behavior of the p68 kinase during a heat-shock stress, a fractionation of interferon-treated HeLa cells, submitted or not to a heat-shock, was carried out using several procedures of extraction. As shown on Fig. lA,by Western blot analysis, using monoclonal anti-p68 antibodies, the amount of the p68 kinase was not significantly modified in the totallysate of cells submitted to a 44 "C heat-shock. On the other hand, the amount of this protein was greatly decreased in the 12,000 X g supernatant of cell lysate prepared with nonionic detergent Nonidet P-40, and most of the p68 kinase was recovered with the nuclear pellet after centrifugation of the lysate at 2,000 x g (data not shown). Fractionation of the p68 kinase was further refined by the procedure described by Capco et al. (26) which utilizes a series of washes of the material remaining attached ontothe culture dish. We show that the amount of p68 kinase was greatly

FIG. 2. Distribution of p68 kinase, plOO 2-5A synthetase, and hsp 70 kDa in interferon-treated cells submitted to a heat-shock as determined by immunofluorescence analysis. Interferon-treated cells were submitted (H, 11, F ) or not (A, C, E ) to a1-h heat-shock a t 44 "C, then fixed and processed for indirect immunofluorescence analysis as described under"Materials and Methods." Cells were stained with a monoclonal antibody specific for the p68 kinase (A, B ) for plOO 2-5A synthetase (C, D )or for the hsp 70 proteins ( E , F ) .

reduced after heat-shock in the soluble cytosolic fraction (Fig. 1B). Most of the enzyme sedimented with both the cytoskeletonand the nuclear fraction. Meanwhile, the plOO2-5A synthetase remained mostly soluble in heat-shocked HeLa cells. As another approach, we investigated the localization of the p68 kinase using indirect immunofluorescence analysis. As shown on Fig. 2, in unstressed HeLa cells, the fluorescence was evenly distributed in the cytoplasm, whereas after a 1-h heat-shock at 44 "C, it aggregated as a thick fibrous-like ring around the nucleus; no p68 kinase was detected inside the nucleus. In contrast, the distribution of the plOO2-5A syn-

p68 Kinase Aggregation during Heat-Shock

9709

thetase remained diffuse in the cytoplasm before and after A c - HS 1 2 3 1 2 3 heat-shock. Noteworthy, both hsc 73 and hsx 70 proteins also aggregated around the nucleus in a structure similar to that of the p68 kinase. Similar observations were obtained with cells submitted to a 46 "C heat-shock for 15 min. Thus during a nonlethal heat-shock, the cytoplasmic p68 kinase of HeLa cells is rendered insolubleand accumulates around thenucleus like the 70-kDa constitutive heat-shock proteins. Distribution of p68 Kinase in Cells Containing a Disrupted Cytoskeleton-It has been shown that the intermediate fila30 ment network collapsed around the nucleus upon heat-shock (27). To determine if the p68 kinase change in solubility during heat-shock was dependent upon this collapse, we pretreated HeLa cells withtwo drugs which have well documentedeffects on the three majorcytoskeletalnetworks. Incubation of cells with colcemid for 5 h causes the disappearance of the microtubules and collapse of the intermediate 2 filaments, while cytochalasin E disrupts the actin-containing microfilaments. As shown in Fig. 3 by Western blot analysis, 31 the p68 kinase is insolubilized at the same extent by heatstress both in control cells and in cells lacking an organized cytoskeleton. Interestingly we observe that the 70-kDa con1" -4 stitutive heat-shock proteins are also partially recovered in 2 -4 the insoluble pellet from heat-shockedcells and thatcolcemid hsp70 and cytochalasin E treatments do not alter their insolubili3. 4 zation (data notshown). D20and Glycerol Protect p68 Kinase from Insolubilization FIG.4. Protection of p68 kinase and hsp 70 by glycerol or during Heat-Shock-We have studied thebehavior of the p68 kinase in HeLa cells treated with glycerol or D2O. Cells were DnO treatment. Interferon-treated cells were pretreated with glyclysed in nonionic detergent immediately after heat-shock and erol ( 2 ) , Dz0 (31, or not ( I ) before being submitted to a 1-h 44 "C heat-shock ( H S ) or not ( C ) . The dsRNA-dependentactivity was the soluble extracts were analyzed for protein phosphoryla- assayed in cytoplasmic extracts ( A ) . In B, supernatants ( S I 2 ) and tion. As shown on Fig. 4 A , the p68 kinase autophosphorylation 12,000 X g pellets ( P I 2 )were analyzed by Western blot using specific is not modified in extracts of unstressed cells treated with monoclonal antibodies against p68 kinase or hsp 70 proteins. glycerol or DnO.As previously observed, in extractsof control cells submitted toa 1-h heat-shock a t 44 "C, this phosphoryl- teins were also rendered insoluble in nonionic detergents by clearly that following heat-shock as 35% of the proteinwas recovered in the pellet. ation is totally impaired. But it appears the same heat-shock, the glycerol treatment maintainsa high While glycerol treatment modified only slightly their solubillevel of p68 kinase phosphorylation, while the D20 treatment ity, DeO treatment protected very efficiently these heat-shock maintains the full kinase activity in the soluble cytoplasmic proteins by keeping them in thesoluble cytosolic fraction. Since glycerol and DrO had been shown to attenuate the extracts. The samecytoplasmic extracts and thecorresponding pel- heat-shock response (9), we analyzed this attenuation in our lets were analyzed by Western blot, using monoclonal anti- experimental conditions. The patterns of newly synthesized p68 antibodies.As shown on Fig. 4B, the large amount of p68 proteins from control and stressed HeLa cells incubated in kinase in the supernatant does not vary when cells treated normal medium, D,O medium, and glycerol medium were with glycerol or D,O are kept a t 37 "C. Only a small amount compared on one-dimensional polyacrylamide gels, and the sedimented withthe pellet, and theDzOtreatment maintained resulting fluorograms of the [:%]methionine-labeled proteins the enzyme soluble. When cells were submitted to a 44 "C are shown in Fig. 5. In normal medium, the synthesis of 70heat-shock, only asmall amountof the enzyme was recovered and 90-kDa heat-shock proteins is clearly visible after a heatin the supernatant (30%) and most of it switched to the shock a t 43 "C, while the normal protein pattern is hardly insoluble pellet (70%). After glycerol treatment, the protein affected. After a 44 "C heat-shock, hsp 70 kDa and hsp 90 was evenly distributed between the soluble extract and the kDa are the major synthesized proteins as the shut-off of insoluble phase, while after D,O treatment, thewhole protein normal proteins synthesis is very important. In cells pretreated with glycerol or D,O before heat-shock, no hsp synremained soluble. The constitutive 70-kDa heat-shock prothesis was observed after a 43 "C heat-shock, and only aweak 512 Pl2 synthesis occurred after a 44 "C heat-shock. Heat-shock imC HS C HS pairment of protein synthesiswas also attenuated as no shutoff was observed. Attenuation of the Insolubilization of p68 Kinase in Thermotolerant Cells-In order to test the behavior of the p68 a p68 kinase in thermotolerant cells, interferon-treated HeLa cells were submitted to a 1-h heat-shock a t 43 "C, 6 h before the FIG. 3. p68 kinase insolubilization in cellswith a disorgan- 1-h challenging heat-shock a t 44 "C. As shown on Fig. 6, the ized cytoskeleton. Interferon-treated cells were pretreated (+) or dsRNA-dependent p68 kinase autophosphorylation was comn o t (-) with colcemid and cytochalasin E before a 1-h heat-shock at pletely impaired in cytoplasmicextracts of cells subr~itted to 44 "C. Extracts were prepared in buffer A at the end of heat-shock. Supernatants ( S I 2 ) and pellets ( P I 2 ) obtained after a 12,000 X g a unique 44 "C heat stress as previously described (21). On centrifugation were analyzed by Western blot using specific mono- the contrary,a 43"C heat-shock did not abolish the p68 kinase clonal anti-pG8 antibodies. activity. It is clear that in cytoplasmic extracts of cells ren'

"

+

p68 Kinase Aggregation during Heat-Shock

9710 A 1 2 3

1 ..

0 2

C 3

1

2

3

97 69

I

p68

4343 30

B

E

HS

0 0 3 6 1 6

-

" "

FIG.5. Heat-shock response in cells pretreated with glycerol or DrO. Interferon-treated cells were kept in normal medium ( A ), incubated in DzOmedium ( B ) or in glycerol medium (C) for 30 min before being submitted to a 1-h heat-shock a t 43 "C (lanes2) or 44 "C (lanes 3 ) . In lane 1 , cells were kept a t 37 "C.Then cells were incubated for16 h with [:"S]methionine, lysed in Laemmli buffer, and proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and visualized by autoradiography.

- ++++ 1

2

3

" "

69,

4

dsRNA

' 4pm

4330-

FIG.6. p68 kinase activity in extracts of cells rendered thermotolerant. dsRNA-dependent p68 kinase activitywas assayed in cytoplasmic extracts of cells kept a t 37 "C (lanesI ) , submitted to a 1-h heat-shock a t 43 "C (lane 2 ) or a t 44 "C (lane 3), or rendered t hermotolerant by a 43 "C heat-shock 6 h before the 44 "C heat-shock (lane 4 ) . dered thermotolerant by a 43 "C heat-shock 6 h before the 44 "C heat-shock, the p68 kinase activity, detected by its autophosphorylation, was maintained to a highlevel. The maintenance of the enzyme under a soluble cytoplasmic form in primed cells submitted to a subsequent heat-shock is correlated with the preservation of the kinase activity (data not shown). Partial Recovery of the p68 Kinase Activity after HeatShock-To determine the fate of the p68 kinase during the recovery period at 37 "C, interferon-treated cells were heatshocked for 1 h at 44 "C, then incubated at 37 "C in medium depleted of interferon but containingcycloheximide to inhibit de novo synthesis of the p68 kinase. Cytoplasmic extracts were prepared at the end of the heat-shock or 3,6, and 16 h later. As expected, the p68 kinase activity was completely absent in cytoplasmic extracts prepared at the end of the heat-shock (Fig. 7A). After 6-h recovery, 40% of the initial activity reappeared, and the same enzymatic activity was still detected 16 h afterheat-shock. This recovery of the p68 kinase activity after heat-shock is not due to de novo synthesis of the enzyme, as cycloheximide at 50 pg/ml completely inhibited protein synthesis. Thus, the reappearance of p68 kinase activity after heat-shock could be due either to the reactiva-

-OOO

-

4p68(P12)

FIG.7. Recovery of p68 kinase activity after heat-shock. Interferon-treated cells were submitted (HS) or not ( C )to a 1-h heatshock a t 44 "C. At the end of the heat-shock, cells were washed and incubated a t 37 "C in medium containing 50 pg/ml cycloheximide. Cytoplasmic extracts were prepared at the end of the heat-shock or 3, 6, and 16 h later. dsRNA-dependent p68 kinase activity was then assayed ( A ) as described under "Materials and Methods." Supernatants ( S I 2 )and pellets (PI2) obtained after a 12,000 X g centrifugation were analyzed by Western blots using monoclonal anti-p68 and anti-p100 antibodies (B). tion of the enzyme which remained soluble (about 20%) but inactive, after heat-shock (Fig. 1A) or to a partial resolubilization of the p68 kinase which aggregated during the stress. By Western blot analysis, we show that theamount of kinase molecules remains low in the soluble extract during the first 6 h following the heat-shock; meanwhile it is still very high in the pellet (Fig. 7 B ) . Thus, it is likely that most of the activity recovered in the soluble extracts at 6 h after heatshock is due to the reactivation of the enzyme remaining soluble after heat-shock, as resolubilization occurs later. Indeed, after 16 h recovery, most of the enzyme is present in the supernatant (74%) and only a small amount in the pellet. DISCUSSION

Experiments presented here further characterize and define the behavior of an endogenous cellular enzyme: the interferoninduced p68 kinase, during heat-shock stress of human HeLa cells. We show that the p68 kinase becomes resistant to nonionic detergent extraction during heat-shock while 2-5A synthetase remains mostly soluble. It is interesting to note that thep68 kinase insolubilization is correlated with its inactivation. Whereas in unstressed cells the p68 kinase is evenly distributed in the cytoplasm, after heat treatment it aggregates around the nucleus. Meanwhile, the otherinterferon-induced enzyme 2-5A synthetase remains evenly distributed in the cytoplasm. The different behavior of both enzymes might be due to their interaction with distinct subcellular structures. Indeed, we have shown by gradient centrifugation that most of the pl00 2-5A synthetase cosediments with ribosomes, whereas a large amount of p68 kinase is not ribosome-associated (28). Though heat-shock is known to provoke a collapse in the intermediate filament network (27), here we show that drug treatments which disorganize the cytoskeleton, do not hinder the insolubilization of the p68 kinase. It appears as a rule that hyperthermia-induced cell damages are minimized in thermotolerant cells (2, 18). Indeed in cells rendered thermotolerant by a previous mild heat-shock at 43 "C, whichinduces the synthesis of heat-shock proteins but

Heat-Shock p68 Aggregation during Kinase

9711

Acknowledgments-We would like to thankM. Morange for helpful whichdoes not insolubilize thep68kinase, we observe a advice and critical reading of the manuscript, M. Pinto for fruitful protection of the enzyme remaining mostly soluble in the discussions, and A. Grkgoire for excellent technical assistance. cytoplasm. Strong correlative data are in favor of a role of heat-shock proteins in the acquisition of thermotolerance (29REFERENCES 32). In thermotolerant cells, the large quantity Of hsp present 1. Mivechi, N. F., and Dewey, W. C. (1985) Radiat. Res. 103, 337-350 2. Yost,H. J . , andLindwist, s. (1986) Cell 45,185-193 at the time of the stress would prevent the aggregation of 3. Calderwood, S. K., and Hahn, G. M. (1983) Biochim. Biophys. Acta 756, unfolded proteins. However, the transient association of the 1-8 unfolded p68 kinase with any of the heat-shock proteins 4. Magun, B. E.,andFennie, C. W. (1981) Radiat. Res. 8 6 , 133-146 5. Wolffe, A. P., Perlman, A. d., and Tata, J. R. (1984) EMBO J. 3 , 2763remains to be demonstrated. Interferenceof heat-shock pro2770 teins with aggregation of other proteins has already been 6. Duncan, R., andHershey, J. W.B. (1984) J . Biol. Chem. 259,11882-11889 7. Duncan, R. F., and Hershey, J. W. B. (1989) J. Cell. Hid. 109,1467-1481 suggested for Escherichia Coli in which heat-dlock response 8. Lepock, J. R., Frey,H.E.,Rodahl, A. M.,andKruuv, J . (1988) J . Cell. promotestheproper assembly of overproduced RNA polymPhysiol. 137, 14-24 9. Edington, B.V., Whelan, S. A,, and Hightower,I,. E. (1989) J . Cell. Physiol. erase (33). 1 3 9 , 219-228 During the recovery period, after a 44 “C heat-shock, we 10. Nguyen, V. T.,Morange,M.,andBensaude, 0.(1989) J. Biol. Chem. 2 6 4 , 10487-10492 Observe a reappearance of the P68 kinase in the 11. Evan, G. I., andHancock, D. C. (1985) Cell 43, 253-261 cytoplasmic extract. This seems to be due to a reactivation of 12. Littlewood, T. D., Hancock, D. C., and Evan, G. I. (1987) J . Cell Sci. 8 8 , 65-72 the enzyme remaining but inactivated during heat- 13. Luscher, B., and Eisenman, R. N. (1988) Mol. Cell. Biol. 8 , 2504-2512 shock as the resolubilization of the enzyme aggregated by the 14. Lutz, Y., Jacob, M., and Fuchs, J. P. (1988) Exp. Cell Res. 1 7 5 , 109-124 Roti, J. L., Uygur, N., and Higashikubo, R. (1986) Radiat. Res. 107, stress later. Whateverthealternative may be, a role of 15. Roti 250-261 the heat-shock proteins in this process has to be considered. 16. Li, G. C., and Laszlo, A. (1985) in Changes in Eukaryotic Gene Expression i n Response to Enoironmental Stress (Atkinson, B. G., and Walden, D. Indeed, it has recently been shown that heat-shock proteins B., eds) pp. 227-254, Academic Press, New York promote the reactivationby refolding of heat-denatured bac- 17. Sciandra, J. J., and Subjeck, J. R. (1984) Cancer Res. 4 4 , 5188-5194 teriophage x repressor (34). Further biochemical analyses are 18. Welch, w. J., and Mizzen, L. A. (1988) J . Cell Bid. 106,1117-1130 19. Dubois, M. F., Ferrieux, C., Robert, N., Lebon, P., and Hovanessian, A. G. under way to determine if heat-shock proteins are involved (1987) Ann. Inst. Pasteur Virol. 1 3 8 , 345-353 in the process of reactivation of the interferon-induced p68 ~ b ~ ~ f , ~ : ~Hovanessian, . ~ A, G, ~ kinase. (1989) J. Bid. Chem. 264,12165-12171 Heat-shock protein hsp 70 is also partly rendered insoluble 22. Bensaude, o., Pinto, M., Dubois, M. F., Nguwn, V. T., and Morange, M. (1991) in Heat Shock Proteins (Schlesinger, M. J., and Santoro, G., eds) by heat-shock. “Sticky” internal domainsof cellular proteins Springer Verlag, Berlin, in press might be unraveled by unfolding during heat-shock and hsp 23. Laurent, A. G., Krust, B., Galabru, J., Svab, J., and Hovanessian, A. G. (1985) Proc. Natl. Sei. U. S. A . 82, 4341-4345 molecules would bind readily these ‘‘sticky” domains. Indeed, 24. Hovanessian, A. G., Laurent, A. G., Chebath, J., Galabru, J., Robert, N., hsp havebeenshown to“chaperon”proteinsduringbiosynandSvab, J. (1987) E M B O J. 6,1273-1280 25. Mattei, D., Scherf, A,, Bensaude, O., and Pereira da Silva, L. (1989) Eur. thesis andbefore assembly into functional structures(35-38). J. Immunol. 19,1823-1828 Under stress conditions, an increasing numberof hsp would 26. Capco, D. G., Wan, K. M., and Penman, S. (1982) Cell 29,847-858 27. Welch, W. J., and Feramisco, J. R. (1985) Mol. Cell. Bid. 5, 1571-1581 be required to prevent aggregation Of the unfolded PolYK’ep28. Dubois, M. F., and Hovanessian, A. G. (1990) Virology 1 7 9 , 591-598 tides. When the pool of free hsp would be depleted, partly 29. Li, G. C., and Werb, 2. (1982) Proc. Natl. Acad. Sci. U. S. A . 7 9 , 32183222 unprotected proteins start to aggregate and trap the 30. Subjeck, J. R., and Sciandra, J. J. (1982) in Heat Shock From Bacteria to Man (Schlesinger, M. J., Ashburner, M., andTissiGres, A,, eds) pp. 405previously bound heat-shock proteins (22). The presence of 411, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY proteins is thought to trigger the heat-shock re- 31. Landry, J., Bernier, D., ChrBtien, P., Nicole, L. M., and Tanguay, R. M. (1982) Cancer Res. 61, 428-437 sponse (9,39-41). Depletion of free hsp may be an important 32. Carper, s.W.3 Duffy, J. J., and Gerner, E.w. (1987) Cancer Res. 4 7 , 5 2 4 9 step in triggering the heat-shock response and accountfor the 5255 self-regulation of hsp synthesis. In prokaryotes, the function- 33. Kashlev, M. V., Gragerov, A. I., and Nikirov, V. G. (1989) Mol. Gen. Genet. 216,469-474 Of the heat-shock transcription factor is antagonized by 34. Gaitanaris, G. A., Papavassiliou, A. G., Rubock, P., Silverstein, S. J., and excess of heat-shock proteins (42). In Saccharomyces cereuisGottesman, M. E. (1990) Cell 61, 1013-1020 E.s.,Lissin, N. M., and Girshovich, A. S. (1988) Nature 3 3 6 , iae the 7 0 - k ~heat-shock ~ protein is (43), 35. Bochkareva, 254-257 while in mammalian cells, injection of neutralizing anti-hsp 36. Ellis, R. J., and Hemmingsen, S. M. (1989) Trends Biochem. Sei. 1 4 , 339342 70 antibodies is reported to induce a heat-shock response (44)‘ 37. Finley, D., Ciechanover, A., and Varshavsky, A. (1984) Cell 3 7 , 4 3 - 5 5 Protection of cells from lethal effects of elevated temperature 38. Pelham, H. R. B. (19%) Cell 46,959-961 39. Gaff. s. A., and Goldberg, A. L. (1985) Cell 4 1 , 587-595 by D,O (45-47) and polyhydroxyl alcohols (48.50) was sug- 40. Hightower, L. E. (1980) J. Cell. Physiol. 1 0 2 , 407-427 gested to result from their ability to stabilize Proteins and to 41. Ananthan, J., Goldberg, A. L., and Voellmy, R. (1986) Science 2 3 2 , 522524 minimize heat-induceddenaturation.Indeed,amongother 42. Straus, D., Walter, W., and Gross, C. A. (1990) Genes 8. Deo. 4, 2202-2209 effects, we find that D 2 0 or glycerol in the culture medium 43. Stone, D. E., and Craig, E. A. (1990) Mol. Cell. Biol. 10, 1622-1632 prevent inactivation and insolubilization of the p68 kinase 44. Riahowol, K. T., Mizzen, L. A., and Welch, W. J. (1988) Science 2 4 2 , 4 3 3 436 within the living Thus, D2° Or glycerol in the 45. Li, G. C., Fisher, G. A., and Hahn, G. M. (1982) Radiat. Res. 92, 541-551 medium would minimize the heat-shock response, decreasing 46. Azzam, E. I., George, I., and Raaphorst, G. P. (1982) Radiat. Res. 9 0 , 644648 the synthesis Of heat-shockproteins and maintaining the 47. Raaphorst, G. P., and Azzam, E. I. (1982) J . Therm. Biol. 7,147-154 normal level of protein synthesis, because these two com- 48. Henle, K. J. (1981) Radiat. Res. 8 8 , 3 9 2 - 4 0 2 Lin, p.-s.,Kwock, L., and Hefter, K. (1981) J . Cell. Physiol. 108,439-443 pounds preventaggregation of proteins and thereforeinsolu- 49. 50. Massicotte-Nolan, P., Glofcheski, D. d., Kruuv, J., and Lepock,J . R. (1981) bilization of the constitutive hsp70 proteins upon heat-shock. Radiat. Res. 8 7 , 284-299

;;:

~

~