ANALYTICAL

BIOCHEMISTRY

202,340-343

(19%)

High-Resolution One-Dimensional Polyacrylamide Isoelectric Focusing of Various Forms of Elongation Factor-2 Nicholas

T. Redpath

Department of Biochemistry, School of Medical Sciences, University University Walk, Bristol, BS8 1TD, United Kingdom

Received

Gel

November

of

Bristol,

27,199l

A system for analyzing covalent modifications of elongation factor-2 (eEF-2) by one-dimensional isoelectric focusing in slab polyacrylamide gels is described. Depending on the degree of phosphorylation, four species of eEF-2 could be resolved corresponding to the un-, mono-, bis-, and trisphosphorylated factor. Furthermore, the degree of ADP-ribosylation of the protein could also be assessed by this method. It was also shown that an acidic isoform of eEF-2 exists which appears not to be artifactual and that the relative level of this isoform appeared to vary between different cell types. By Western blotting the gels and using an antibody against eEF-2 it is possible to assess the state of phosphorylation of the faCtOr in cells. 0 is92 Academic PEWS, Inc.

Eukaryotic elongation factor-2 (eEF-2)’ catalyzes the translocation of peptidyl-tRNA from the A- to the Psite of the ribosome and the release of the uncharged tRNA from the ribosome during each round of elongation of peptide synthesis in eukaryotic cells. It is a monomeric protein of 100 kDa and requires GTP for activity and the hydrolysis of eEF-2-bound GTP is required for the release of the factor from the ribosome after each round of elongation (1). In recent years it has become evident that the activity of eEF-2 is controlled by phosphorylation. The factor is phosphorylated mainly on two threonine residues (T56 and T5*) (2) by a highly specific calcium/calmodulin-dependent kinase (3). Phosphorylation of eEF-2 inhibits ’ Abbreviations used: Chaps, 3-[(3-cholamidopropyl)dimethylammoniolpropanesulfonate; CHO, Chinese Hamster Ovary eEF-, eukaryotic elongation factor; eIF-, eukaryotic initiation IEF, isoelectric focusing; SDS, sodium dodecyl sulfate.

(cells); factor;

its activity in the poly(U)-directed synthesis of polyphenylalanine (4,5) and in the translation of endogenous mRNA in the reticulocyte lysate (6). The precise mechanism whereby phosphorylation inhibits eEF-2 is unknown although it is likely that the phosphorylated factor has a reduced affinity for the ribosome (6,7). Changes in the phosphorylation of eEF-2 in whole cells have as yet not been studied widely although growth factors in fibroblasts (S), thrombin and histamine in endothelial cells (9), and thyrotrophin-releasing hormone in GH, cells (10) have been shown to increase eEF-2 kinase activity by increasing cytoplasmic calcium. eEF-2 phosphorylation was also found to increase during the mitotic phase of amnion cells (11). The previous studies of changes in eEF-2 phosphorylation have involved labeling cells with 32Pi and analyzing the cell extract by SDS-polyacrylamide gel electrophoresis or by immunoprecipitating eEF-2 [e.g. (8,12)] or separation of the phosphorylated and unphosphorylated forms by two-dimensional electrophoresis followed by detection of the eEF-2 by staining of the gel, immunoblotting, or autoradiography of 35S- or 32P-labeled protein (11,13). These methods have the disadvantage of not being easily applicable to the analysis of a large number of samples. This is particularly the case for analysis by 2D gels especially if Western blotting and immunodetection of eEF-2 are to be carried out. Since some of these methods cannot detect unphosphorylated eEF-2, they have the drawback that the fraction of eEF-2 phosphorylated in relation to total eEF-2 is not measurable. One-dimensional polyacrylamide gel isoelectric focusing has previously been most successfully used in studying changes in the phosphorylation state of eukaryotic initiation factor-2cu (14-17). This report describes a method for studying changes in eEF-2 phosphoryla-

340 Copyright 0 1992 All rights of reproduction

0003-2697192 $3.00 hy Academic Press, Inc. in any form reserved.

ISOELECTRIC

tion by one-dimensional vantages of this method cussed. MATERIALS

AND

FOCUSING

OF ELONGATION

isoelectric focusing. The adover previous methods are dis-

METHODS

Chemicals and biochemicals were obtained as previously reported (2,6). Urea (Ultrapure) was from BRL. Acrylamide, IV, N-methylenebisacrylamide, and ampholines were obtained from Pharmacia and Chaps was from Fluka. Enhanced chemiluminescence kit was from Amersham. Cell Preparations

Rabbit reticulocyte lysates were prepared and incubations performed as in (6). eEF-2 and eEF-2 kinase were prepared from reticulocyte lysate as described (2). Hepatocytes were prepared as described (18). Swiss mouse 3T3 cells were maintained as previously reported (18). CHO cells were maintained in the same way except that Hams F-12 medium was used. For analysis of cellular eEF-2 by IEF, cell monolayers were washed with 5 ml of buffer containing 20 mM Hepes (pH 7.5), 100 mM NaCl, 50 mM sodium+glycerophosphate, 2 mM EDTA, 10% glycerol, 0.5 mM benzamidine, and 1 pg ml-’ each of pepstatin, leupeptin, and antipain. The buffer was aspirated off and 2 ~1 of 20 I.LM microcystin was added to the Petri dish then the cells were scraped off using a rubber policeman. Cell suspension (20 ~1) was then added to sample buffer and solid urea was added to give a concentration of 9 M. Two microliters were applied to the IEF gel. Isoelectric Focusing

The gel solution was freshly prepared containing 9 M (w/v) urea, 4.8% (w/v) acrylamide, 0.2% (w/v) bisacrylamide, and 4% (w/v), pH 3-10, ampholines in a final volume of 6 ml, which is sufficient for one 0.75-cm-thick gel using the Bio-Rad minigel system. Following degassing, solid Chaps was dissolved in the solution to give a concentration of 2%. Fifteen microliters of 10% ammonium persulfate and 6 ~1 of tetramethylethylenediamine were then added and the gel was poured and allowed to set for at least 1 h. The wells were then filled with 6 M urea and underlaid with sample buffer which consisted of 9 M urea, 4%, pH 3-10, ampholines, 2% Chaps, and a small amount of bromophenol blue. The cathode electrolyte was 20 mM NaOH and the anode electrolyte was 10 mM phosphoric acid. The polarity was reversed so that the upper chamber was the anode and the lower one the cathode. The gel was prefocused at 200,300, and 400 V for 15 min each. The samples were prepared by adding 5 ~1 of 7X concentrated sample buffer without urea and 25 mg of solid urea to 20 ~1 of sample. If the gel was run at constant voltage the initial voltage setting was 550 V,

FACTOR-2

FORMS

341

which was increased gradually to 700 V over 3 h followed by 1000 V for 30 min. Alternatively, the gel was run at a constant power of 2 W for 3 h followed by 1000 V for 30 min. The phosphoric acid was replaced at least once with cold acid during the run to prevent overheating of the gel. After electrophoresis the gel was either Western blotted or silver stained. Blots were incubated with anti eEF-2 antibody (19) except that whole serum was used at a 1:lOOO dilution followed by anti-rabbit IgG peroxidase conjugate. eEF-2 was visualized by enhanced chemiluminescence. If the gel was to be stained, it was first washed overnight in 10% trichloroacetic acid to remove the Chaps and ampholines. If required, the relative levels of the different forms of eEF-2 could be assessed by laser densitometry using a Joyce-Loebl Chromoscan 3 scanner. RESULTS

AND

DISCUSSION

Isoelectric focusing of purified rabbit reticulocyte eEF-2 using the method described above showed that eEF-2 exists as two isoforms (Fig. 1, lane 1). The minor form (referred to here as eEF-2’) constitutes lo-15% of the total eEF-2 and has a more acidic isoelectric point than eEF-2, approximately midway between those of unphosphorylated and monophosphorylated eEF-2. It is unlikely that eEF-2’ is formed artifactually during the purification as discussed later. In addition, the inclusion of 1 mM dithiothreitol in the IEF gel or addition of thioglycolic acid to the cathode electrolyte to scavenge free radicals did not reduce the amount of eEF-2’ present, suggesting that eEF-2’ is not the product of -XH group oxidation (not shown). Marzouki et al. (13) also reported the existence of a minor isoform of rat liver eEF-2 detected using 2D gel electrophoresis. Incubation of eEF-2 with eEF-2 kinase resulted in the appearance of eEF-2 focusing at lower pH values corresponding to mono-, bis-, and trisphosphorylated factor. These bands have previously been shown to be eEF-2 varying solely in the degree of phosphorylation by 32Plabeling studies and by mapping of the phosphorylation sites (2). The maximum level of phosphorylation was about 2 mol phosphate/m01 eEF-2; the level of trisphosphorylation in vitro is rarely more than is shown in Fig. 1. Evidence (not shown) suggests that the additional site of phosphorylation in the trisphosphorylated factor is threonine 53. It is also evident that eEF-2’ can also be phosphorylated as the level of this form of eEF-2 was reduced as the incubation proceeded. The monophosphorylated form of eEF-2’ was not evident, however. It is unlikely that eEF-2’ did not undergo phosphorylation and it is possible that monophosphorylated eEF-2’ runs close to and is obscured by bisphosphorylated eEF-2. Additional bands above trisphosphorylated eEF-2 are also visible. These appeared late on in the incubation

342

NICHOLAS

T.

REDPATH

123456

7

FIG. 1. Purified rabbit reticulocyte lysate eEF-2 was electrophoresed on an IEF gel (1.6 pg per lane) or after phosphorylation with eEF-2 kinase for 2,5,10,30, or 60 min at 30°C (lanes 2-6, respectively). P,, and P, refer to un-, mono-, bis-, and trisphosphorylated eEF-2 and Pb and P; refer to the equivalent incubated with 10 pg ml-’ ADP-ribosyl transferase fragment of diphtheria toxin and 1 mM NAD’ for (lane 7). The gel was then silver stained. I& and R, refer to the unmodified and ADP-ribosylated factor, equivalent forms of eEF-2’. ”

and are likely to be trisphosphorylated eEF-2’ and possibly eEF-2 phosphorylated on four sites. Electrophoresis of the kinase alone on the IEF gel illustrated that none of the stained bands were proteins present in the kinase preparation (not shown). The previous method used in this laboratory for isoelectric focusing of eEF-2 in one-dimensional gels used a narrow pH range (7-9) (19). The focused bands, however, were poorly resolved, distorted, and ill defined. By using a wide pH range (3-10) in the method described here, the gradient formed is steeper and therefore the focused bands are sharper and highly defined. Diphtheria toxin-catalyzed ADP-ribosylation (which occurs only on one site) causes a shift in the isoelectric point of eEF-2 to one similar to that of monophosphorylated eEF-2 (Fig. 1, lane 7). It is also evident that eEF2’ can also be ADP-ribosylated. It has been previously reported that prior phosphorylation or ADP-ribosylation of eEF-2 has no effect on subsequent ADP-ribosylation or phosphorylation, respectively (13,20), although recently Marzouki et al. (21) have suggested that ADP-ribosylated eEF-2 can be phosphorylated three times faster than the unmodified factor. The production of specific polyclonal antibodies against eEF-2 has enabled quantification of the degree of phosphorylation of eEF-2 in crude cell extracts and whole cells. Figure 2 shows the effect of incubating reticulocyte lysates with the protein phosphatase inhibitor okadaic acid on eEF-2 phosphorylation. Okadaic acid has previously been shown to inhibit reticulocyte lysate protein synthesis by increasing the phosphorylation of eEF-2 (6). Again, this improved method of isoelectric focusing provides greater resolution of the various forms of eEF-2 over methods previously reported as the presence of bisphosphorylated eEF-2 under control

without prior phosphorylation (lane 1) The gel was then silver stained. P,, P,, forms of eEF-2’. eEF-2 (1.6 pg) was also 30 min at 30°C then run on an IEF gel respectively, and & and R’r refer to the

conditions and trisphosphorylated eEF-2 in the presence of okadaic acid was evident. The isoelectric focusing of eEF-2 from various cell types was also performed (Fig. 3). The factor from rat hepatocytes, Swiss 3T3 cells, and CHO cells all had the same isoelectric point. The basal level of phosphorylation of eEF-2 in these three cell types was, however, very low. Both the Swiss 3T3 cells and the CHO cells contained low levels of eEF-2’ as did reticulocytes. In hepatocytes, however, about 50% of eEF-2 was eEF-2’, which is more than that detected by Marzouki et al. (13) when using two-dimensional gel electrophoresis. As mentioned previously, it is unlikely that eEF-2’ is produced artifactually since in this case the cells were scraped from the plates and immediately lysed in urea-containing buffer without any processing of eEF-2 prior to elec-

12345

FIG. 2.

Rabbit reticulocyte lysates were incubated without further additions (lane 1) or in the presence of 250,150, or 100 nM okadaic acid (lanes 2-4, respectively) or 200 pM of the calmodulin antagonist, trifluoperaxine (lane 5). Samples were removed and run on an IEF gel followed by Western blotting and immunodetection of eEF-2. The figure is an autoradiograph.

ISOELECTRIC

1

2

FOCUSING

OF

3

ELONGATION

FACTOR-2

Russell, 258. 3. Nairn, Acad.

R. J. M.,

FORMS and Proud,

C. G. (1991)

A. C., Bhagat, B., and Sci. USA 82, 7939-7943.

4. Ryazanov, A. G., Shestakova, Nature (London) 334,170-173. 5. Nairn, A. C., and Palfrey, 14,265-14,272. 6. Redpath, N. T., and Proud,

FIG. 3. Samples of CHO cells (lane l), Swiss 3T3 cells (lane 2), and hepatocytes (lane 3) were run on an IEF gel followed by transfer and immunodetection of eEF-2. The figure is an autoradiograph.

trophoresis. It is not known whether eEF-2’ is functionally different in any way from eEF-2 or why hepatocytes should contain more eEF-2’ than other cell types. In conclusion, this report describes a method of onedimensional isoelectric focusing of eEF-2 from different cell types which is a major improvement over previous one- and two-dimensional electrophoresis methods. Because a wide pH range is used it should be possible to analyze changes in the phosphorylation of other proteins of interest such as the translational factors eEFICC, eIF-2q or eIF-2B under the same conditions, and possibly on the same gel. ACKNOWLEDGMENT This work was supported the Science and Engineering

by a project Research

grant to Dr. C. G. Proud Council.

from

1. Moldave, 2. Price,

K. (1985) N. T., Redpath,

Annu.

Reu. &o&em.

N. T., Severinov,

54,

1109-1149.

K. V., Campbell,

D. G.,

H. C. (1987) C. G. (1989)

Biochem.

8. Palfrey, H. C., Nairn, (1987) J. Biol. Chem.

A. C., Muldoon, 262,9785-9792.

9. Mackie, K. P., Nairn, (1989) J. Biol. Chem.

A. C., Hampel, G., Lam, 264, 1748-1753.

11. Celis, Acad.

Martin,

J. E., Madsen, P., and Ryazanov, Sci. USA 87,4231-4235.

12. End, D., Tolson, N., Hashimoto, Chem. 258,6549-6555.

Chem.

J. B&&en.

G., and Jaffe, J. Biol.

Chem.

A. G. (1990)

S., and Guroff,

262,

J. 262,69-75. Eur.

L. L., and Villereal,

T. F. J. (1982)

Nutl.

P. G. (1988)

J. Biol.

0. (1990)

282,253Proc.

E. A., and Natapov,

A., and Nygird,

10. Drust, D. S., and 7566-7573.

Lett.

H. C. (1985)

U., Nilsson, 7. Carlberg, 191,639-645.

PFOC.

G. (1983)

M. L. E. A. 257, N&Z. J. Biol.

13. Marzouki, A., Lavergne, J. P., Reboud, J. P., and Reboud A. M. (1989) FEBS Lett. 255,72-76. 14. Cox, S., Redpath, N. T., and Proud, C. G. (1988) FEBS Lett. 239, 333-338. 15. Scorsone, K. A., Panniers, E. C. (1987) J. Biol. Chem.

R., Rowlands, A. C., and Henshaw, 262, 14,538-14,543.

16. Rowlands, R. (1988)

K. S., Henshaw, 175,93-99.

A. G., Montine, Eur. J. Biochem.

E. C., and Panniers,

17. Kimball, S. R., Antonetti, D. A., Brawley, R. M., L. S. (1991) J. Biol. Chem. 266, 1969-1976. 18. Redpath, N. T., and Proud, C. G. (1990) Biochem. 180.

and Jefferson, J. 272,

175-

19. Redpath, 1093,36-41.

Biophys.

Acta

N. T., and Proud,

20. Nilsson, A., Carlberg, 195,377-383.

REFERENCES

Palfrey,

FEBS

C. G. (1991)

U., and

Nyglrd

Biochim. (1991)

Eur.

J. Biochem.

21. Marzouki, A., Sontag, B., Lavergne, J. P., Vidonne, C., Reboud, J. P., and Reboud, A. M. (1991) Biochimie 73,1151-1156.