Phosphorylation of eukaryotic initiation factor 2 during physiological ...

THEJOURNAL OF BIOLOGICAL CHEMISTRY

Vol. 262, No. 30, Issue of October 25, pp. 14538-14543,1987 Printed in U.S.A.

Q 1987 by The American Socirtv for Biochemistry and Molecular Biology, Inc.

Phosphorylation of Eukaryotic Initiation Factor2 during Physiological Stresses Which Affect Protein Synthesis* (Received for publication, April 3, 1987)

Kathleen A. ScorsoneS, Richard Panniers,Anne G . Rowlands, and EdgarC. Henshaw From the Cancer Center and Department of Biochemistry, University of Rochester, Rochester, New York 14642

Phosphorylation of the a subunit of eukaryotic initiation factor 2 (eIF-2) isa major mechanism regulating protein synthesisin rabbit reticulocytes. To determine whether phosphorylation of eIF-2a is a likely regulatory mechanism in the Ehrlich cell, we have measured the percentof cellular eIF-2awhich is phosphorylated incells exposed to heatshock, Z-deoxyglucose, or amino acid deprivation, conditions which rapidly decrease the concentrationof 40 S initiation complexes and inhibit protein synthesis. e 1 F - h and eIF2a(P) were separatedby isoelectric focusing and were detected by immunoblotting with a monoclonal antibody we developed for thispurpose. Under the above three inhibitory conditions, phosphorylation of eIF-2a increased rapidly, and this increase correlated in time with the rapid inhibition of protein synthesis. In heat-shocked cells which were returned to 37 ‘C, both phosphorylation and protein synthesis remained unchanged for 10 min and then returned toward control values slowly and in parallel. The close temporal correspondence between changes in protein synthesis and phosphorylation supports an important regulatory role for phosphorylation in protein synthesis. An increase of 25-35 percentage points, to 50-60% phosphorylation from control levels of 20-30% phosphorylation, correlated withan 80-100%inhibition of protein synthesis. This steep curve of inhibition is consistent witha mechanism in which eIF-Za(P) saturates and inhibits theguanine-nucleotide exchange factor.

(eIF-2)’ (9-11). In the reticulocyte lysate, the decrease in 40 S complexlevels during heme deficiency is the result of activation of heme-regulated inhibitor (HRI) (12-14), an eIF2 kinase that phosphorylates the a subunit of eIF-2, preventing the release of GDP from eIF-2. GDP.Release of GDP has been shown to require another initiation factor, GEF (15-18). Phosphorylated eIF-2 binds tightly to GEF, inhibiting it from interacting with unphosphorylated eIF-2. GDP (19, 20) and therefore inhibiting the exchange of GDP for GTP. If GTP cannot bind eIF-2, then Met-tRNA does not form a ternary complex with eIF-2.GTP, and 40 S complex levels are reduced. A similar mechanism may explain decreased 40 S initiation complexlevels in other systems. Duncan and Hershey (7) have demonstrated an increase in eIF-2a phosphorylation in heat-shocked HeLa cells compared with 37 “C cells and in serum-deprived cells (21). In a model system for amino acid starvation, we have shown that Chinese hamster ovary cells lacking tRNA synthetase activity at nonpermissive temperatures have elevated phosphorylated eIF-2 levels compared to cells at permissive temperatures (22). We have demonstrated that eIF-2 restores 40 S initiation complexes to control levels when added to lysates prepared from amino acid-starved Ehrlich cells (23). However, we have yet to demonstrate an increased phosphorylation during starvation.The results presented here show that thedegree of eIF-2 phosphorylation in Ehrlich cells increases in parallel with the increasing inhibition of protein synthesis during amino acid starvation, 2deoxyglucose substitution for glucose, and heat shock. EXPERIMENTAL PROCEDURES~

RESULTS

Translation in eukaryotic cells has been demonstrated to be an important site of regulation (1).Overall rates of polypeptide synthesis in mammalian cells are inhibited during amino acid (2), glucose (3), and serum starvation (4), heat shock (5-7), and in heme-deprived reticulocytes (8). In each of these cases, down regulation is largely the result of inhibition of polypeptide chaininitiation with aconcomitant decrease of 40 S ribosomal initiation complexes. These complexes contain the initiator Met-tRNA and GTP bound to 40 S ribosome subunits via eukaryotic initiation factor 2 * This work was supported by United States Public Health Service Grants CA-21663and CA-11198,Genetics Training Grant GM-07102, an Elon Huntington Hooker Fellowship, and the Life and Health Insurance Medical Research Fund. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. $ Present address: Louisiana State University, Choppin Hall, Baton Rouge, LA 70803-1806.

Regulation of Phosphorylation of eIF-2a inResponse to Physiological Stimuli in the Ehrlich Cell We have usedthe IEFimmunoassay to assess the possibility that increased phosphorylation of eIF-2a is responsible for

* The abbreviations used are: eIF, eukaryotic initiation factor; HRI, heme-regulated inhibitor (the protein kinase from reticulocytes which phosphorylates the CY subunit of eIF-2);GEF, guanine-nucleotide exchange factor; IEF, isoelectric focusing; SDS, sodium dodecyl sulfate; PBS, phosphate-buffered saline; BSA, bovine serum albumin; DTT, dithiothreitol; TCA, trichloroacetic acid; MOPS, I-morpholinepropanesulfonic acid. Portions of this paper (including “Experimental Procedures,” part of “Results,” and Figs. 1-3) are presented in miniprint at the endof this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 87M-1062, cite the authors, and include a check or money order for $5.20 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

14538

eIF-2a Phosphorylation

14539

,

100

B

C

ao

-

60

- /m"--m 40

1

0

0

I

I

20

40

I

0

20

I

40

TIME (mln.)

Ty(nh3

FIG. 4. Phosphorylation of eIF-2a in heat-shocked cells. Cells were incubated with [14C]leucinea t 37 "C (0)or 43 "C (W). Fifty-pl samples were removed a t indicated times to determine the amountof leucine incorporated into protein ( A ) , separate and detect eIF-2a and eIF-Ba(P)by isoelectric focusing and immunoblot analysis ( B ) (paired lanes represent 15-p1and 30-p1cell samples, respectively), and quantitate thepercent eIF-2 phosphorylated from the autoradiographs in B by laser densitometry ( C ) .

A

I

I

I 100

100

I

B

80

B

I

Et 4 2

/

/

60

0

a 0

2

40

N

&! &?

0

(mln.)

TIME

20

40

60

Do-0

0

20

0

20

40

60

TIME (mln.)

FIG. 5. Recovery of protein synthesis and dephosphorylation of eIF-Sa in heat-shocked cells returned to 37 "C. Cells were incubated as described in Fig. 4, but one culture (0)was transferred from 43 to 37 "C after 20 min at thehigher temperature. Analyses of protein synthesis ( A ) and quantitation of levels of eIF2(P) ( B )are described in the legend to Fig. 4.

the inhibition of protein synthesis under three conditions in which the concentrationof 40 S initiation complexes is known to be depressed. On the basis of known eIF-2 characteristics, increased phosphorylation would be expected to reduce 40 S complexes. Heat Shock-Fig. 4A shows the rapid onset of severe progressive inhibition of protein synthesis in Ehrlich cells incubated at 43 "C. A significant difference in phosphorylation of eIF-2a is apparent within 8 min between control (26%) and heat-shocked cells (34%)in theautoradiograph (Fig. 4B) and laser densitometer scan of the autoradiograph (Fig. 4C). By the next time point, experimental cells have reached 60%, 36 remain steady at 26%. At 7000 while controls 2.5 min, however, there is little difference in degree of phosphorylation and no inhibition of protein synthesis is visible. Fig. 5 presents data from a similar experiment in which one culture was returned to 37 "C after 20 min of heat shock. Aswe have shown previously (6), protein synthesis recovered after a lag (Fig. 5A). Over the time period of protein synthesis recovery, the phosphorylation state of eIF-2a returnedtoward normal (Fig.

0

20

40

60

TIME(mln.)

FIG. 6. Correlation of rate of protein synthesis and increase in eIF-Ba(P) in amino acid-deprived cells. Fed cells (0)or cells deprived of glutamine (M)were incubated at 37 "C with ["Clleucine. Samples were analyzed for uptake of leucine into protein ( A ) and level of eIF-Za(P) ( B )as described.

TABLEI Effect of deprivationof glucose and 2-deoxyglucose on levels on protein synthesis and levels of eIF-2(P) in Ehrlich cells Cells were incubated without glucose or without glucose plus 2deoxyglucose (25mM) in minimal essential medium for 1hr. Samples were removed at this time for estimation of levels of eIF-2(P) as described under "ExperimentalProcedures." Samples of cells (100 pl) were also removed and pulse-labeled with ['4C]leucine (10 pCi/ml) for 10 min, and incorporation of label into proteinestimated by trichloroacetic acid DreciDitation as described. Protein synthesis

Control -Glucose +2-Deoxyglucose -Glucose

'IF-' rylated phospho-

cpm

%

9060

38

1500

48

5B), therecovery in eIF-2 showing a similar lag and perhaps slightly preceding the recovery of proteinsynthesis. This figure also illustrates the reproducibility of the pattern of response of eIF-2 phosphorylation.

14540


Amino Acid Deprivation-Protein synthesis falls very very low basal state may explain the inhibition of protein quickly in Ehrlich cells deprived of glutamine, and the con- synthesis in the heat-shockedcells (21). In the Ehrlichcell, a centration of 40 S initiation complexes is reduced (2, 25).Fig. crude estimate, based on the inhibition of protein synthesis 6A shows the rateof protein synthesis in such an experiment.during heat shock(Fig. 4), suggests that anincrease from 25% Fig. 6B shows that within 10 min the phosphorylationof eIF- to 60% in percent of eIF-2 which is phosphorylated is suff2a had reached a steady state of37% compared to 22% in cient to cause complete inhibition of protein synthesis. This control cells. In this experiment, phosphorylation in control implies that the GEF concentration should be roughly 50cells rose slowly over an hour from 22 to 30% even though 60%of the total eIF-2 concentration. This is much higher than has generally been reported in rabbit reticulocytes (34) the rate of protein synthesis appeared to remain constant and possiblyin HeLa cells. However, our recent measureover the sameperiod. Glucose Deprivation-In glucose-deprived cells, the inhibi- ments of GEF concentration suggest that the molar GEF tion of protein synthesis was initially lesssevere than in concentration is in fact about50% of the eIF-2 concentration why theconcentration is so amino acid-deprived cells. Although thetimecourse was in EhrlichItisnotclear somewhat variablefrom experiment to experiment, inhibitionmuch higher in Ehrlich cells than in reticulocytes or HeLa was generally only 20-35% at 1 h, presumably because of the cells, but there is no a priori reason why it should not be so, availability of other energy sources such as glycogen (Table and a wide range of concentrations may be revealed as more tissues are tested. I). Inhibition becameprogressivelymoresevere with time, Thesharpelevation of eIF-2phosphorylationinheatreaching 40-60% at 5 h (not shown). The energydeficit and the inhibition of protein synthesis could be made much more shocked cells offersan explanationfor the markeddepression cells. Using the cell-free rapid and dramatic by the addition of 2-deoxyglucose, an in protein synthesis rate in these protein-synthesizing system prepared from Ehrlich cell lyinhibitor of glycolysis, to the glucose-freemedium. In the experiment shown in Table I, protein synthesis was inhibited sates, we have shown previously that exogenously added eIF25% by glucose deprivation for 1 h and was inhibited 84% 4F, a cap-binding protein, restores control ratesof synthesis from heat-shocked cells but does not with the addition of 2-deoxyglucose. After 1 h, the level of tolysatesprepared stimulate control lysates, implying an impairment of eIF-4F phosphorylation of eIF-2 in control cells was 38% compared function as well (35).Inour cell-free system,unlikethe to 48% in the 2-deoxyglucose-treated cells, but did not inreticulocyte cell-free system, eIF-2 function is not limiting crease perceptibly in cells deprived only of glucose. In 2deoxyglucose-treated cells, inhibition of protein synthesis and over the early timecourse, explaining theability of eIF-4F to phosphorylation of eIF-2 increased in parallel and were a t a restore complete activity despite increased phosphorylation plateau level by 20 min (not shown). More prolonged depri- of eIF-20. In the absence of better information concerning in intactcontrolandheatvation of glucose alone led t o more severe inhibition of protein which stepsarerate-limiting shocked cells, it is not possible to determine how much the synthesis and to some increase in phosphorylation of eIF-2. of The increase in phosphorylation was not, however, as great impairment of eIF-4F activity contributes to the inhibition protein synthesis in vivo, compared to phosphorylation of eIFas in heat-shocked cells a t comparable level of inhibition of 2. However, the data in this paper are compatible witha protein synthesis. Emetine-We have also treated the cells with emetine, an strong dependence of protein synthesis upon eIF-2 function inhibitor of polypeptide chain elongation (33) which would in intactcells, and a clear effectof impaired eIF-4F is to cause not be anticipated t o produce a decrease in 40 S initiation preferential translationof mRNAs (36) forheat shock protein (35), as theyhave less dependence uponeIF-4F. We favor the complexes. While protein synthesis was inhibited96%,in hypothesis that phosphorylationof eIF-2 is primarily responcontrast to the results under the previous conditions, phossible for the overall inhibition of protein synthesis in heatphorylation of eIF-Sa did not rise, and, in fact, fell from 23 shocked cells and impairmentof eIF-4F is responsible for the to 16% within10 min (data not shown). preferential synthesis of heat shock proteins. We have shown previously that initiation in the cell-free DISCUSSION protein-synthesizing system from Ehrlich cells is very sensiInhibition of eIF-2 function in Ehrlich cells exposed to heat tive to the GDP/GTP ratio, over a narrow range of values shock, amino acid deprivation, and 2-deoxyglucose has been (37). This is expected on the basis of our knowledge of the implied previously by the depression of 40 S initiation com- competition between GDP and GTP in the GEF-catalyzed plexes. The increase in phosphorylation of eIF-2a under these GDP/GTP exchange reaction. The in vivo GDP/GTP ratio three conditions provides an explanation for the impaired falls over the same narrow range in glucose-deprived Ehrlich function of eIF-2 because the guanine-nucleotide exchange cells and, to a lesser extent, in amino acid-deprived cells (3). factor, GEF, is able t o release GDP only from eIF-2, not eIF- We are unable to assess with precision the relative contribu2a(P). Thus, phosphorylated eIF-2 is unable to function cy- tions of fall in energy charge and increase in eIF-2 phosphoclically in protein synthesis. However, the fractional loss in rylation to the inhibitionof protein synthesis in amino acidrate of protein synthesis is much greater than the fractional deprived cells. A simple assumption is that both contribute. increase in phosphorylated eIF-2,so that an increasein phos- However, in the glucose-deprived cells early during the time phorylation from 26 t o 63% during heat shock, for instance, course (1 h), the changein phosphorylation was often impercaused a 96% fall in the rateof protein synthesis. Thus, small ceptible (Tabie I) and could not explain the inhibition of changes in phosphorylation are correlated with large inhibi- protein synthesis (Table I). In this circumstance, the fall in tions of protein synthesis. This phenomenon has been clearly GTP/GDP ratio appears to be the major regulatory mechademonstrated in the reticulocyte lysate protein-synthesizing nism. With the more severe energy deficit caused by 2-deoxsystem (30-32) and is due to a large excessof eIF-2 over GEF yglucose, phosphorylation of eIF-2 is increased to a greater in the lysate, and to the fact that eIF-2dP). GDP binds GEF extent (Table I), but the increase is still less than that in much more effectively than eIF-2. GDP, thus competing for heat-shocked cells(Fig. 5)in which inhibition of protein GEF and effectively inhibiting its GDP exchange function. Similarly, even though phosphorylation of eIF-2 reachesonly A. G. Rowlands, K. S. Montine, E. C. Henshaw, and R. Panniers, a low level in heat-shocked HeLa cells, the increase from a manuscript in preparation.


14541

10. Gupta, N. K., Woodley, C. L., Chen, Y. C., andBose, K. K. (1973) J. Bwl. Chem. 248,4500-4511 11. Levin. D. H... Kvner. " , D.., and Acs. G. (1973) J. Biol. Chem. 248, 6416-6425 12. Levin, D. H., Ranu, R. S., Ernst, V., Fifer, M. A., and London, I. M. (1975) Proc. Natl. Acad. Sci. U.S. A. 72,4849-4853 13. Kramer, G., Cimadevilla, J. M., and Hardesty. B. (1976) Proc. Natl. Acad. Sci. U.S. A . 73,3078-3082 14. Farrell, P. J., Balkow, K., Hunt, T., Jackson, R. J., and Trachsel, H. (1977) Cell 11, 187-200 15. Konieczny, A., and Safer, B. (1983) J. Biol. Chem. 2 5 8 , 34023408 16. Pain, V. M., and Clemens,M. J. (1983) Biochemistry 2 2 , 726733 17. Panniers, R., and Henshaw, E. C. (1983) J. Biol. Chem. 2 5 8 , 7928-7934 18. Siekierka. J.. Mitsui. K. I.. and Ochoa., S. (1981) Proc. Natl. Acad. Sci. U.S . A . 78,220-223 19. Siekierka, J., Manne, V., Mauser, L., and Ochoa, S. (1983) Proc. Natl. Acad. Sci. U.S. A. 80, 1232-1235 20. Clemens, M. J., Pain, V. M.,-Wong, S. T., and Henshaw, E. C. (1982) Nature 2 9 6 , 93-95 21. Duncan, R., and Hershey, J. W.B. (1985) J. Bwl. Chem. 2 6 0 , as 5493-5497 22. Clemens, M. J., Galpine,A., Austin, S. A., Panniers, R., Henshaw, E. C., Duncan, R., Hershey, J. W. B., and Pollard, J. W. (1987) J. Biol. Chem. 2 6 2 , 767-771 and Clemens, 23. Pain, V. M., Lewis,J. A., Huvos, P., Henshaw, E. C., M. J. (1980) J. Bwl. Chem. 255,1486-1491 24. Kohler, G., and Milstein, C. (1975) Nature 256,495-497 25. Pain, V., and Henshaw, E. C. (1975) Eur. J. Biochem. 5 7 , 335342 Acknowledgments-We thank Dr. Rosemary Jagus for a generous gift of sheep antiserum against reticulocyte eIF-2 and Dr. Edith Lord 26. Wong, S. T., Mastropaolo, W., and Henshaw,E.C. (1982) J . Bwl. Chem. 257,5231-5238 and Lee Harwell for assistance in preparing monoclonal anti-eIF-2. We would also liketo thank Kathleen Montine for useful discussions 27. Laemmli, U. K. (1970) Nature 227,680-685 and Eileen Stewart forexcellenttechnicalassistance. The word 28. Lloyd, M. A., Osbourne, J. C., Jr., Safer, B., Powell, G. M., and Merrick, W. C. (1980) J. Bioi. Chem. 256, 1189-1193 processing expertise of Mary LeRoy-Jacobs and Lynne Palmiere is 29. Towbin, H., Staehelin, T., and Gordon,J. (1979) h o c . Natl. Acad. greatly appreciated. Sci. U.S. A . 76,4350-4354 30. Farrell, P., Hunt, T., and Jackson, R. J. (1979) Eur. J. Biochem. REFERENCES 89,517-521 1. Revel, M., and Groner, Y.(1987) Annu. Reu. Biochem. 47, 107931. Ernst, V.,Levin, D. H., and London, I. M. (1979) Proc. Natl. 1126 Acad. Sci. U. S. A . 76,2118-2122 2. van Venrooij, W. J. W., Henshaw, E. C., and Hirsch, C. A. (1970) 32. Leroux, A., and London, I. M. (1982) Proc. Natl. Acad. Sci. U, S. A. 7 9 , 2147-2151 J. Bwl. Chem. 245,5947-5953 3. vanvenrooij, W. J., Henshaw, E. C., and Hirsch, C. A. (1972) 33. Grollman, A. P. (1968) J. Biol. Chem. 243,4089-4094 34. Thomas, N. S. B., Matts, R. L., Petryshyn, R., and London,I. M. Biochim. Biophys. Acta 259,127-137 (1984) Proc. Natl. Acad. Sci. U. S. A . 8 1 , 6998-7002 4. Kaminskas, E. (1973) J. Cell Physiol. 82,475-488 5. McKenzie, S. L., Henikoff, S., and Meselson, M. (1975) Proc. 35. Panniers, R., Stewart, E. B., Merrick,W. C., and Henshaw, E. C. (1985) J . Biol. Chem. 260,9648-9653 Natl. Acad. Sci. U. S. A . 7 2 , 1117-1121 Brendler, T. G., Ayda, A., Daniels-McQueen, S., 6. Panniers, R., and Henshaw, E. C. (1984) Eur. J. Biochem. 140, 36. Ray,B.K., Kelvin Miller, J., Hershey,J. W. B., Griffo, J. A., Merrick, W. 209-214 C., and Thach, R. E. (1983) Proc. Natl. Acad. Sci. U.S. A. 80, 7. Duncan, R., and Hershey, J. W. B. (1984) J. Bwl. Chem. 259, 663-667 11882-11889 8. Bruns, G. P., and London, I. M. (1965) Biochem. Biophys. Res. 37. Hucul, J. A., Henshaw, E. C . , and Young, D. A. (1985) J. Biol. Chem. 2 6 0 , 11585-11591 Commun. 18, 236-242 9. Safer, B., Anderson, W. F., and Merrick, W.C. (1975) J. Biol. Chem. 250,9067-9075

synthesis is comparable. It perhaps is not surprising that the GTP/GDP ratio would fall and would be the major regulator of protein synthesis in energy-deprived cells. It is not clear why this ratio should fall in cells deprived of one essential amino acid. We have shown,however, that glucose utilization actually falls in amino acid-deprived Ehrlichcells (3). The increased phosphorylation of eIF-2 helps explain the decreased 40 S initiation complexes in heat-shocked and 2deoxyglucose-treated or amino acid-deprived cells, and the excellent correlation between the time course of increased phosphorylation and decreasedproteinsynthesissuggests thattheincreasedphosphorylationisresponsibleforthe inhibition of protein synthesis. However, other explanations for the correlation are possible, as suggested by the result with emetine. In this case, an inhibition of chain elongation A possible explacaused a decrease in eIF-2 phosphorylation. nation is that a backup of 40 S subunits, with a shift in proportion of various intermediates, causes accumulation of eIF-2 in a form which has an increase in susceptibility to phosphatase relative to kinase, favoring dephosphorylation an event secondary to inhibitionof protein synthesis. At this time, it cannot be excluded that the rises in phosphorylation during heat shock and nutrient deprivation are also secondary, for instance to loss of eIF-4F activity and fall in energy charge, respectively.

I

I

.

~

~~


14542 Supplemental Material to

Phosphorylation Of eIF-2 during Physiological Stresses which AfEect Protein Synthesis Kathleen A. Scorsone. RlChard Panniers, Anne G. Rowlands and Edgar C. Henshaw EXPERIMEWAL

corresponding to the position of the eIF-2 D subunit visualized by Coonassie Blue dye staining or by phosphorylation with (32P)ATP and H R I (the reticulocyte eIF-2a kinase) (data not shown). Bands corresponding were stained by antibody on to eIF-20 and eIF-2r~lP) protein bands imunoblots Of IEF gels, and the eIF-2a band disappeared wlth a comensurate increase in the eIF-2m(P) band upon phosphorylation with HRI (Fig. 1).

PROCEDURES

Cell Culture EhrliCh ascites tumor cells were grown in spinner culture at 37OC in Eagles minimal essential medium supplemented with mM 20 MOPS (pH 7.35) and 108 calf serum IMicrobiological Assoc.). Cultures were diluted daily to maintainthe cell number between 2 x 105 and 8 x 105 ce11s/m1.

4

Measurement of Cellular Protein Synthesis m i n o Acid Starvation: exponentially growing Ehrlich ascites tumor cells were centrifuaed at 5 0 0 x a for 7 minutes. washed once with PBS and resuspended with brewarmed serim-free minimal essential nedlvm minus glutamine at a density of 3 x 106 cells/ml. Samples (1-2 mls) were placed in snap-cap tubes which contahned either glutamine (final concentration, 20 mM). or the appropriate volume Of water and incubated at 37Oc during which 95% air/5% CO2 was Continuously present. 1 1 4 C l l e u ~ i nwas ~ added to each tube to a final concentration of 2.5 uCi/ml and at various times duplicate5 0 u1 samples were placed on Whatnan no. 3 filter discs. IAnOther two 50 el samples were added to tubes containing 17.5 u 1 of ice cold isoelectrlc sample buffer and 70 ng urea to stop metabolic processe5 immediately. eIF-2- and eIF-2c~lPl were separated by isoelectrbc focusing gel electrophoresis and detected as described below). Filters were washed in a 5 % trichloroacetic acid and 1% casamino acid solution andboiled in an identical Solution for 10 nin. After sequential washes With 10% trichloroacetic acid, 95% ethanol. 958 ethanol/acetone (50:50), and acetone, the filters were dried. and radioactivity estimated by liquid scintillation Spectrophotometry. Heat Shock: the procedure was the same as the starvation experiments except the cells wgre resuspended in minimal essential medium prewarned to 37% or 43 C and hncubation was carried Out at each respective temperature. Slab Gel

Isoelectric Focusing

Rapidly Stopped samples Of cells generated above were analyzed by gel iSOeleCtr~Cfocusing in the presence of 9.2 M urea as described by Wong et al. (26) except Chat Nonidet PI0 was not used. The samples consisted of 30 ul of cells, 42 ng urea and 10.5 e l of isoelectric sample buffer which contained 40% B-mercaptoethanol and 32% of Isoelectric focusing Ampholine mixture ILKBI (pH 5-7:pH 3.5-10, 4:l). was performed at 3 watts Iconstant wattage) for 15 hours followed by BOO V (constant voltage) €Or 1 hour. The pH gradient Which was formed ranged from pH 4 . 7 - 6 . 5 . SDS-Polyacrylamide Gel ElectroDhoresis Sodium dodecyl sulfate polyacrylamidegel electrophoresis was a modification OE the method Of Laemli (27) in a 1.5 m slab gel containing 15% acrylamide. 0.098 bisacrylamide and 0.1% SDS. Samples for the electrophoresis were prepared by mixing with an equal volume of SDS sample buffer (0.125 H Tris, pH 6.8, 12% glycerol, 2 . 4 % SDS, 1.1 M 8-mercaptoethanol, 0.01% bromophenol blue). The alpha. beta. and gama subunits of EhrliCh c e l l eIF-2 were identified by their nlgration in relation to molecular weight standards and their comigration with the subunits of purified rabbit reticulocyte eIF-2 128). Phosphorylation Of eIF-2 12 ug of purified eIF-2 and 5 ug of partially purified HRA (14) were incubated in a final volume Of 4 0 u l for 20 minutes at 30 C. Final concentrations Of Other CODStitUentS were 20 mM MOPS (pH 7 . 6 ) . 5 mM MgAc2. 100 nU4 KC1. 1 mJ8 DTT. 0.1 mM EDTA. and 10 yM (y32-P) ATP I 6 Ci/mol) (261.

2

1 Fig. 1

Imunodetection of eIP-2m and eIF-2 m(P) in isoelectric focusing gels Autoradiograph Showing the relative positions of eIF-201P) and eIF-2ol in lsoelectrlc focusing gels. Purified =IF-2 IO.5ug). Lane 2 , and eIF-2 after incubation with HRI and ATP, Lane 1, were electrophoresed, blotted and Imunodetected as described in materials and methods.

In an experiment measuring the recovery of 32P-labeled eIF-2- on immunoblots Of IEF gels. 87% of the subunit was transferred from the gel to the nitrocellulose filter, and the transferred material remained stably bound on the glutaradehyde-fixed filter even after the extensive washes required for the imuno-detection Of elF-2o bv an enzyme linked

Quantitation was by reference to a standard curve (Fig. 2). Over the range 0 . 0 4 uq to 0.2 ug of eIF-2a. the absorbance of the lase. densitometer-scanned autoradiograms vas approximately linear Wlth the amount of eIF-2 applied. For each experlmental point several concentrations were run so that quantitation could always be performed in this range. The specific absorbances ofthe phosphorylated and unphoSphorylated forms were equal (Fig. 2 ) .

Immunization Protocol and Antibody Purification Calf brain eIF-2, approximately 70% pure was emulsified in an equal volume Of complete Freund's adjuvant. Mice were immunized by a n initial injection of 21 ug of eIF-2 and two subsequent injectmns of 42 ua each ldivided amona multiDle Sites1 at m m t h l v intervals. Four &ks after the last injectibn, the mbuse with tie hlghest Serum titre of antibody to eIF-2 war given an intravenous injection of 1 4 yg of eIF-2 dissolved in normal saline on three successive days. On the fourth day, the spleen vas removed. a single cell su5pension vas prepared, and the cells were fused using 3 5 e polyethylene glycol in a 2:l ratio with BALB/c myeloma P3-X63 A48.653.3.1, according to standard procedures ( 2 4 ) . The hybridoma supernitants from 96 well plates were assayed for Specific antibody production by a microtitre assay using purified Ehrlich c e l l eIF-2 as the substrate and a n avidin-biotinperoxidase labeled second antibody (horse anti-mouseIgG which was Obtained from Vector Staln Laboratories Inc.). A stable clone producing a high titre Of antibody to the alpha Subunit was re-cloned twice. For antzbody production, the hybridoma was grown in the ascites form in pristane-treated mice. The antlbody vas purified from the ascites fluid by amonlum Sulfate precipitation and chromatography on a "DEAE Affi-Gel Blue" (Bio-Rad) column and was greater than 901 pure. Inmunoblottinq Cellular proteins were separated on isoelectcic focusing gels and electroblotted Onto nitrocellulose sheets(0.22 urn, Millipore) following the method of Towbin et al. (29). The proteins were fixed to the nitrocellulose by incubation for 20 minutes wlth 0.5% glutaraldehyde ~n phosphate buffered saline IPSS)10.15 M NaC1, 0.01 M sodium phosphate, pH 7 . 4 ) . followed by three 5 minute washer with TriDbuffered saline 10.15 M NaCl. 20 mM Trir-HC1. OH 7 . 4 1 to Inactivate remaining glutarildehyde. Nbn-speciEic binding site; on these sheets yere blocked dgrkng incubation With PBS containing 2% v/v BSA for 30 mlnutes at 37 C. The nitrocellulose sheet was incubated vlth purified monoclonal anti eIF-20 11.6 n d m l l €or 1 hour at roam temperature after which unbound antibody was removed by Several washings with PBS containing 0 . 1 8 v/v Tween 20. The nitrocellulose was then incubated in a Solution of PBS containing 3% BSA, 2.5% calf Serum, and 0 . 5 uCi/ml 1251 -labeled rabbit anti-mouse IgG (NEN). After 5 washes vlth PBS Contaming 0.1% Tween 20, the nitrocellulose sheet was dried under a heat lamp and placed against Kodak XAR5 x-ray fllm €or 16 hours.

eIF-2 Fig. 2 Standard

(ug)

curve for eIP-2 and eIP-2alP) Standard EUTVIS yere generated for eIF-2 and elF-2a(P) by isoelectric focusing purified =IF-2 or eIP-2lP) 1100% phosphorylated) and Scanning imunodetected bands in autoradiographs for optical density using a laser densitometer. The total optical density per band versus amount of eIF-2- ( 0 )or elF-ZalP) I. 1 electrophoresed is shown.

The Optical densityof bands representing eIF-2o. and eIF-2alP) on the autoradiograms vas estimated by a LKB Instruments Inc. Ultroscan XL Laser densitometer. The level Of the phosphorylated form of eIF-2 was calculated as a Dercentaae OE the total ODtiCal densitv of eIF-lo and eIF-2mlP) on the'autoradiogram. The amounts of protein represented by each band were determined Erom standard curves siniliar to those depicted by Figure 2 . The percent of eIF-2 phosphorylated is given as100 x eIF-2m(P) / (eIF-2o + eIF-2a(P)). RESULTS Validation of the measurement Of elF-2s DhoSDhorYlation. Small changes in phOsphorylatiDn can cause large changes in the rate of protein synthesis (see Discussion). so that the validity of the measurements Of the phosphorylation state ofeIF-2- is crucial. We have utilized isoelectric focusina in 9.2 H urea on wlvacrvlamide eels to separate phosphorylated and unbhosphorylatedm subun;ts Hnd havedeveloped a monoclonal antibody to the 0. Subunit in order to quantitate the protein in crude cell extracts.

On immunoblots ("Western" blots) Of SDS polyacrylamidegels Of EhrliCh cell extracts, monoclonal antibody stained Only one band,

TO assess our method under known conditions we measured the state OE phosphorylation of eIF-2 in a reticulocyte lysate proteinSynthesxring system Incubated with and without hemin or HRI. Figure 3A Shows that synthesis rate was linear for at least 20 minutes in the presence of hemin, fell by about 6 minutes in the absence of hemin, and fell by 3 minutes in the presence Of H R I . The isoelectric gel pattern is shown in Fig. 38 and quantitated in 3C. Increases in phosphorylation occurred at times corresponding to the fall in rate of protein synthesis, in accordance with the measurementsOf other Investigatocs 00-32).


0

30

20

10

14543

TIME ( r n i n . )

6

+hemin

-hemin

" o 2

4

6

io

o 2

20 30

4

6 10

20 30

min

elF-2a elF-2aP +HRI

o 2

4

6

io

20

min.

- 0 0

10

20

30

TIME ( r n i n . ) Fig. 3

Phosphorylation Of eIP-2.a in reticulocyte lysate in the absence of hemin or presence of BRI

is

A.

Protein synthesis was followed the reticulocyte C leucine into TCA lysate by the incorporation Of precipitable material in 10 "1 samples as described 114). Reactions contained 20 uM hemin ( 0 ), no hemin I ) . hemin plus HRI .1 ).

8.

Samples I10 ul) removed at indicated times were used for the separation and innunodetection Of eIF2 0 andl$JF-20(P) as described. Autoradiograph Shows I secondary antibody labelled eIF-2.

C.

Bands representing eIF-2a and eIF-Zm(P) in 8. were quantitated by laser densitometry and are expressed as percent eIF-2lP) relative to total.