Phosphorylation of eukaryotic protein synthesis initiation factors.

1 downloads 0 Views 1MB Size Report
Department of Biological Chemistry, School of Medicine, University of California, Davis, California 95616. Communicated by Edwin G. Krebs, October 13, 1977.
Proc. NatI. Acad. Sci. USA Vol. 75, No. 1, pp. 108-112, January 1978 Biochemistry

Phosphorylation of eukaryotic protein synthesis initiation factors (translational control/rabbit reticulocytes/cyclic AMP-independent kinase/in vitro phosphorylation/intracellular phosphorylation)

ROB BENNE, JEFFREY EDMAN, ROBERT R. TRAUT, AND JOHN W. B. HERSHEY Department of Biological Chemistry, School of Medicine, University of California, Davis, California 95616

Communicated by Edwin G. Krebs, October 13, 1977

ABSTRACT

Phosphorylation of eukaryotic initiation factors was examined both in intact cells and in vitro with purified components. Intact rabbit reticulocytes were incubated in a medium containing [IP~phosphate, and eight initiation factors were isolated and partially purified. The purified factors were analyzed on dodecyl sulfate/polyacrylamide gels and compared with highly purified nonradioactive factors. Significant amounts of radioactivity were found associated with initiation factors eIF-2, polypeptide 2 (molecular weight 53,000); eIF-3, polypeptides 2 and 4 (molecular weights 110,000 and 67,000); and eIF4B. Purified initiation factors from rabbit reticulocytes were also treated in vitro with [ly-32PJATP and a cyclic AMP-independent protein kinase isolated from rabbit erythrocytes. Only the factor polypeptides phosphorylated intracellularly were phosphorylated in vitro. The results suggest that the cyclic AMP-independent protein kinase is responsible for the phosphorylation of specific initiation factors in cells active in protein synthesis and that it may play a role in regulating translation.

cAMP-independent kinase from rabbit erythrocytes was assayed with all eight factors as substrates. The same polypeptides were phosphorylated as were found in the experiment with intact reticulocytes. The small subunit of eIF-2, which seems to be involved in the inhibition of protein synthesis caused by hemin deprivation (4-7), was not labeled in either procedure. METHODS Materials. ['y-32P]ATP (specific activity, 250 cpm/pmol) was prepared according to the method of Glynn and Chappell (18). Biochemical compounds were obtained as follows: 2-mercaptoethanol from British Drug House; acrylamide and bisacrylamide from Eastman, [32P]phosphoric acid (carrier-free) from ICN, actinomycin D-from Calbiochem, rabbit serum from Antibodies Corp., and young rabbit blood from Pel-Freeze. All chemicals were reagent grade. Sephadex G-100 (Pharmacia), DEAE-cellulose (Whatman DE-32), and phosphocellulose (Whatman P-lI) were pretreated according to the manufacturers' directions. Buffers. Buffer A: 20 mM Tris-HCI, pH 7.6/7 mM 2-mercaptoethanol/5% glycerol. Buffer B: 20 mM potassium phosphate, pH 7.2/7 mM 2-mercaptoethanol/5% glycerol. Initiation Factors. Highly purified initiation factors from rabbit reticulocytes were prepared as described (14-17). In Vivo Phosphorylation. Reticulocytes were isolated from the blood of four anemic rabbits as described by Schreier and Staehelin (19). The intact cells were incubated with [32P]phosphate essentially as described by Kabat (13). Packed cells (30 ml) were suspended in 30 ml of a medium containing 10 mM Tris-HCl (pH 7.6), 5 mM magnesium acetate, 100 mM sodium chloride, 0.03 mM each of 20 amino acids, 300 .ug of actinomycin D, 60 mg of glucose, 3 ml of rabbit serum, and 25 mCi of [32P]phosphoric acid. The mixture was incubated for 60 min at 370, by which time 95% of the radioactivity was taken up by the cells. The cells were lysed and fractionated rapidly according to the procedures of Schreier and Staehelin (19) to yield the high-salt ribosomal extract which contains the initiation factors. Each of the initiation factors was partially purified in order to remove contaminating radioactive phosphoproteins that could complicate the identification of phosphorylated initiation factors. The fractionation of the crude extract (Fig. 1) was based on procedures previously developed for the purification of all factors on a preparative scale (14-17). After ammonium sulfate fractionation of the high-salt ribosomal extract into fraction A (0-40% saturation) and fraction BC (40-70% saturation), the crude preparations were subjected to fractionation procedures based on size.

Protein synthesis initiation factors* from rabbit reticulocytes are phosphorylated in vitro by a cyclic AMP (cAMP)-independent protein kinase isolated from the same source. We have shown (2) that one of the three polypeptides of eIF-2, polypeptide 2 [molecular weight (Mr) 53,000], and two of the nine polypeptides of eIF-3, polypeptides 2 and 4 (Mr 110,000 and 67,000, respectively), are phosphorylated. Traugh and coworkers (3) have demonstrated the phosphorylation of the same polypeptide of eIF-2 and also of eIF-5 with a similar protein kinase. Other experiments (4-7) indicate that the inhibition of protein synthesis observed in heme-deficient cell extracts is mediated by the phosphorylation of a different polypeptide of eIF-2, subunit 3, (Mr 35,000). The latter results suggest that phosphorylation and dephosphorylation of initiation factors may play a role in the control of protein synthesis in reticulocytes or other cell types. Because specific ribosomal proteins are also phosphorylated both in vitro and in vivo (8-12), the ribosome may also be involved in such regulatory

mechanisms. Possible phosphorylation of initiation factors has been investigated by isolating the phosphoproteins after incubation of intact cells in a medium containing [32P]phosphate. It has already been observed that a substantial proportion of phosphoproteins in reticulocytes are associated with ribosomes but may be separated from them by extraction in buffers containing a high salt concentration (11-13). Methods for the purification of eight initiation factors were developed recently (14-17) and were used to partially purify the initiation factors from reticulocytes that had been incubated in the presence of [32P]phosphate. Of the eight factors, only eIF-2, eIF-3, and eIF-4B contained significant amounts of radioactive phosphate. In an extension of earlier in vitro studies (2), a more highly purified

Abbreviations: Mr, molecular weight; cAMP, cyclic AMP. The nomenclature for eukaryotic initiation factors used here is that adopted at the International Symposium on Protein Synthesis, Bethesda, MD, in October 1976 (1).

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. ยง1734 solely to indicate this fact.

*

108

Biochemistry:

Proc. Natl. Acad. Sci. USA 75 (1978)

Benne et al.

Ribosomal wash

?

OIFAwas-eluted with the same buffer, and adsorbed protein

"

containing eIF-I and eIF-4D was eluted with buffer B and 800

Ammonium

sulfate fractionation Ammonium sulfate fractionation

eIF-3

eIF-[2+5] eIF-4A eIF-[1+4D] eIF-4C

eIF-4B

109

FIG. 1. Outline of fractionation procedure for initiation factors labeled in intact reticulocytes.

The fraction A precipitate was dissolved in buffer A containing 100 mM KCI and centrifuged through a 10-30% sucrose gradient (Beckman SW 41 rotor, 16 hr, 35,000 rpm) as described (15). Fractions (in the middle of the gradient) containing eIF-3 were pooled. Those (at the top of the gradient) containing eIF-4B were pooled and immediately passed through a column of phosphocellulose (0.5 X 5 cm). Nonadsorbed protein was eluted with buffer B containing 100 mM KCL Adsorbed protein containing eIF-4B was eluted with buffer B containing 500 mM

KCI (16). Ammonium sulfate fraction BC was fractionated by molecular sieve chromatography on a column of Sephadex G-100 (2.0 X 40 cm) equilibrated in buffer B and 100 mM KCG as described (16). The eluted protein was divided into two fractions based on prior analysis of similar but larger columns (16): (i) the excluded protein peak, shown previously to contain eIF-2 and eIF-5;- and (i) the globin peak region, shown to contain eIF-1, eIF-4A, eIF-4C, and eIF-4D (17). The excluded protein fractions were pooled and passed through a phosphocellulose (0.5 X 5 cm) column. The column was washed with buffer B and 250 mM KGI; adsorbed protein containing eIF-2 and eIF-5 was eluted with buffer B and 1000 mM KCG. The fractions that contained elF-1, eIF-4A, eIF-4C, and eIF-4D (16) were applied to a phosphocellulose column (0.5 X 5 cm) equilibrated in buffer B and 250 mM KCI. Nonadsorbed protein containing

mM KCL. Elution of this column with biffer B and 1200 mM KCI yielded a fraction containing eIF-4C. Finally, the nonadsorbed fraction from the phosphocellulose column was applied to a column of DEAE-cellulose (0.5 X 5 cm) equilibrated in buffer A and 100 mM KCG. The column was washed with the same buffer, and adsorbed protein containing eIF-4A was eluted with buffer A and 500 mM KCl. The amounts of radioactivity and protein obtained at each step of the purifications are given in Table 1. All factors were dialyzed against 7.5% propionic acid and lyophilized prior to analysis by gel electrophoresis. cAMP-Independent Protein Kinase. This was isolated from rabbit erythrocytes and partially purified as described (21) with the following modification. The 0-60% saturated ammonium sulfate fraction was applied to a DEAE-cellulose column that was eluted first with a linear 0-400 mM KCI gradient and then with 1 M KCI (22). The cAMP-independent protein kinase activity eluting at 1 M KCI was further purified from contaminating cAMP-dependent protein kinase activity by phosphocellulose column chromatography (21). This enzyme fraction was highly active with acidic proteins (e.g.., casein and phosvitin) as substrates and was relatively inactive with basic proteins (e.g., histones) (J. Edman and R. R. Traut, unpublished data); it is functionally indistinguishable from the casein kinase activity eluting from DEAE-cellulose at 280 mM KCI which was shown previously to phosphorylate eIF-2 and eIF-3 (2). It was assayed with casein throughout the purification. In Vitro Phosphorylation. Initiation factors (1.0 Ag of eIF-1; 2.5 ,ug of eIF-4A, eIF-4B, eIF-4C, eIF-4D, or eIF-5; 12.5 Ag of eIF-2; 25 ,g of eIF-3) were incubated in 100-,gl reaction mixtures as described for eIF-2 and eIF-3 (2) for 60 min at 370 in the presence of the following components: 50 mM 4-morpholinoethanesulfonic acid, pH 6.9; 7 mM 2-mercaptoethanol; 75 mM KCI; 50 mM NaCI; 10 mM magnesium acetate; 0.5 Mig of protein kinase; and 0.1 mM [y-32P]ATP (specific activity, 250 cpm/pmol). The amounts of protein kinase and [-y-32P]ATP were saturating; plateau levels of phosphorylation were obtained within 60 min. The reaction mixture was then brought to 0.1% in sodium dodecyl sulfate and 5% in glycerol, heated for 10 min at 65', and directly applied to a dodecyl sulfate/ polyacrylamide slab gel for electrophoresis.

Table 1. Partial purification of labeled initiation factors from intact reticulocytes incubated with [32P]phosphate

Fraction

Vol., ml

Protein, mg

Radioactivity, cpm X 10-3

Specific radioactivity, cpm X 10-3/mg of protein

Fraction A 0.75 4.25 4457 1048 eIF-3 (sucrose gradient) 4.0 0.56 472 843 eIF-4B (sucrose gradient) 4.0 3.12 2432 780 eIF-4B (phosphocellulose) 3.0 0.78 666 853 Fraction BC 0.75 6.63 3864 462 eIF[2 + 5] (Sephadex G-100) 10.5 3.78 2709 716 eIF[2 + 5] (phosphocellulose) 3.0 1.59 2052 1290 eIF[1 + 4A + 4C + 4D] (Sephadex G-100) 27.0 4.86 699 144 eIF-4C (phosphocellulose) 3.0