- Email: [email protected]
Role of Ser-53 phosphorylation in the activity of human translation initiation factor eIF-4E in mammalian and yeast cells
Gene, 163 (1995) 283-288 © 1995 Elsevier Science B.V. All rights reserved. 0378-1119/95/$09.50
283
GENE 09020
Role of Ser-53 phosphorylation in the activity of human translation initiation factor eIF-4E in mammalian and yeast cells (Cap-binding component; protein synthesis)
Yan Zhang *, Hannah L. Klein and Robert J. Schneider Department of Biochemistry, and Kaplan Cancer Center, New York University Medical Center, New York, NY10016, USA
Received by G.N. Godson: 18 Jaauary 1995; Revised/Accepted: 8 March/6 April 1995; Received at publishers: 28 April 1995
SUMMARY
Eukaryotic translation initiation factor eIF-4E is essential for protein synthesis and cell viability, eIF-4E participates in formation of an mTGTP-cap binding protein complex that mediates association of 40S ribosomal subunits with mRNAs, which occurs only when eIF-4E is phosphorylated. Regulation of eIF-4E by phosphorylation was thought to occur on Set s3, although results potentially inconsistent with phosphorylation of this site have been reported. To resolve whether Set 53 is phosphorylated, and if so whether it regulates eIF-4E activity, we directly examined whether Ser 53 is a site for phophorylation of mammalian eIF-4E in human and yeast cells. Wild-type (wt) human eIF-4E protein variants, Ser s3 ~Asp 53 or Ser 5a --*Ala53, were constructed and analyzed by overproduction in transfected human 293/T-Ag cells, or in Saccharomyces cerevisiae in which the endogenous elF-4E gene was disrupted. Wt eIF-4E and Ser 53 mutants functioned equally well in protein synthesis in both systems, and were phosphorylated to the same extent. Most importantly, the wt and Ser 53 mutants of human elF-4E produced identical tryptic phophopeptide patterns in human cells, and identical but more complicated patterns in yeast. These data demonstrate that Ser 53 is not a requisite activating site for phosphorylation of mammalian elF-4E in human or yeast cells, under conditions in which it participates in protein synthesis.
INTRODUCTION
Initiation of protein synthesis is a rate-limiting step in eukaryotes. A key target for regulation is the translation Correspondence to: Dr. R.J. Schneider, Department of Biochemistry, NYU Medical Center, New York, NY 10016, USA. Tel. (1-212) 263-6006; Fax (1-212) 263-8166; e-mail: [email protected] * Current address: Howard Haghes Medical Institute, Department of Biochemistry and Biophysics, University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA.
Abbreviations: A, absorbance (1 cm); aa, amino acid(s); Ad, adenovirus; 13Gal, 13-galactosidase;bp, base pair(s); CEN (Cen), centromere; elF, eukaryotic initiation factor; eIF-4A, elF-413; elF-4E, eIF-4~; p220, elF-47; kb, kilobase(s) or 1000 bp; LTR, long terminal repeat; nt, nucleotide(s); oligo, oligodeoxyribonucleotide; ori, origin of DNA replication; p, plasmid; PAGE, polyacrylamide-gel electrophoresis; PCR, polymerase chain reaction; r-, ribosc,mal; RSV, Rous sarcoma virus; SDS, sodium dodecyl sulfate; TCA, trichloroacetic acid; wt, wild type. SSDI 0378-1119(95)00302-9
factor eIF-4F (reviewed in Frederickson and Sonenberg, 1993). Factor eIF-4F is required for translation of most capped mRNAs. Mammalian eIF-4F consists of three subunits: a 220-kDa protein (p220) whose function is unknown, a 50-kDa protein (eIF-4A) which is an ATPdependent RNA helicase (Grifo et al., 1983) and a 24-kDa protein (eIF-4E), the cap-binding component of eIF-4F (Sonenberg et al., 1978). eIF-4F is therefore a capdependent RNA helicase, facilitating 5' unwinding of mRNA and interaction with 40S r-subunits (reviewed in Rhoads, 1988). In yeast S. cerevisiae, disruption of the elF-4E gene results in cell death (Altmann et al., 1987; Blum et al., 1989), which is prevented by expression of the mammalian elF-4E gene (Altmann et al., 1989). S. cerevisiae eIF-4E is 33% identical to mouse and human eIF-4E, and functions in a similar manner (Altmann et al., 1985). Many segments of human and yeast eIF-4E are
284 well conserved, including the Ser 53 position and its flanking sequences (Altmann et al., 1987). Reduced phosphorylation of elF-4E correlates with loss of elF-4F function and inhibition of translation during heat shock (Panniers et al., 1985; Duncan et al., 1987), mitosis (Bonneau and Sonenberg, 1987; H u a n g and Schneider, 1991 ), in cells infected by Ad (Huang and Schneider, 1991), and possibly by influenza viruses (Feigenblum and Schneider, 1993; reviewed in Zhang and Schneider, 1994). Increased phosphorylation of elF-4E enhances elF-4F function, and is observed in cells treated with mitogens or serum (Marino et al., 1989; Morley and Traugh, 1989; Kaspar et al., 1990), or transformed with oncogenes Src or Ras (Frederickson et al., 1991). Phosphorylation of elF-4E was suspected to occur on Ser 53 for a variety of experimental reasons (reviewed in Frederickson and Sonenberg, 1993). It was therefore surprising that elF-4E mutants containing Ala or Asp in place of Ser 53 were phosphorylated when overproduced in Cos cells, a monkey cell line, and did not prevent capdependent translation (Kaufman et al., 1993). In this study we therefore sought to clarify the role of Ser 53 phosphorylation in activity of mammalian elF-4E.
EXPERIMENTAL AND DISCUSSION
(a) Disruption of the endogenous yeast elF-4E gene The yeast elF-4E gene was deleted to permit comparison of the translation function and phosphorylation pattern of human eIF-4E protein in human and yeast cells. The yeast elF-4E gene was amplified by PCR, cloned into the yeast-E, coli shuttle vector, pRS316 (Fig. 1), and used todisrupt the yeast gene in a one-step procedure (Rothstein, 1983). Diploid cells were transformed with a 3.2-kb SalI-XbaI fragment from pRS316-y4Enull, which carries flanking sequences for the yeast elF-4E gene as well as the yeast HIS3 gene, which serves as a genetic marker (Fig. 2A). Disruption of the yeast elF-4E gene in His + transformants was verified by Southern blot analysis (Fig. 2B). Mutant and wt alleles were separated by tetrad analysis, and only half grew into colonies as expected if the His + phenotype is associated with the deleted allele. Loss of yeast elF-4E function was established by transformation of the heterozygous null allele strain with construct pRS316-y4E, which expresses the yeast elF-4E gene. Transformation rendered more than two out of four spores viable, indicating that the elF-4E defect was rescued by the yeast elF-4E gene. Plasmid pRS316-y4E replicates as a centromeric vector, segregating into two of four haploid spores. Some elF-4E-null spores therefore lack the plasmid and are inviable as expected, while null allele His + spores carry
Cen6
(
IX S IIBIBB BBIBX 0.8kb 0.6kb k
~ Cen6
X 1
S
S
y4E 2.2kb
ten6 X
BH / ~$3 B
X
1.8kb E
Cen6
S
B
Fig. 1. PCR cloning of yeast eIF-4E gene. Strategy used to construct vectors pRS316-y4E and pRS-y4Enull. The yeast eIF-4E coding region is indicated as y4E. The yeast centromere sequence from chromosome 6 is defined as Cen6. S, SalI; B, BamH1; X, Xbal. Methods: DNA fragments containing different parts of yeast elF-4E coding sequence were cloned by PCR amplificationof the 2.2-kb HindIII fragment with primers selected to amplify the 2.0-kb full-length coding region (nt 83-107 to 2041-2066), the 760-bp upstream regulatory region (nt 83 1-7 to 815-842), and the 571-bp fragment 3' noncoding region (nt 2041-2066 to 1496-1519). Primers included restriction sites for cloning. PCR amplified fragments were purified with Qiax (Qiagen), subjected to restriction enzymedigestions and cloned as shown in the diagram. The 2.0-kb SalI-XbaI restriction fragment was inserted into vector pRS316 (Sikorski and Hieter, 1989) to generate pRS316-y4E, the 0.76-kb SallBamHI and 0.67-kb BamHI-XbaI fragments were inserted between SalI and XbaI sites of pRS316 to generate pRS316-y4ED. The yeast HIS3 gene was inserted into pRS316-y4ED to generate pRS316-y4Enull. pRS316-y4E and are rescued. These results confirm a previous report (Altmann et al., 1987) that the yeast elF-4E gene is required for cell viability, and could therefore be used to examine the role of Ser 53 phosphorylation in eIF-4E activity in yeast and human cells.
(b) Human elF-4E does not require Ser53 phosphorylation to permit translation in yeast and human cells Since mouse eIF-4E substitutes for the yeast homolog in vivo by permitting yeast cell growth (Altmann et al., 1989), and mouse and human eIF-4E proteins differ by only one amino acid, they should be interchangeable. An eIF-4E Ser53~Ala53 human variant (Ala 53) and an eIF-4E Ser 53 --*Asp53 variant (Asp 53) were constructed to study the importance of phosphorylation at residue Ser 53. Wt and variant human elF-4E genes were expressed from the low-copy yeast C E N vector pGAD2, under GALIO promoter control and T R P 1 gene selection (Fig. 3). Transformants were sporulated, dissected onto rich plates containing Gal to induce expression of human eIF-4E and viable His + spores synthesizing human wt (SerS3),
285 A
B
1.8kb H H
"7 o
II 0.76 kb
~'
Transform
c Lu
03
N
>-
0.57 kb
diploid his3"
H H
I I X
=1= t-~|
elF-4E
X
= 2.2 kb
!
~H
JiL
2.2 kb
Double crossover
•'9[-- 1.6 kb i
!
....
~'--
1.4
kb
r
0.2 kb
C YZ-y4Enull
RS3'I6y4E
Fig. 2. Method and analysis of yeast elF-4E gene disruption. (A) Scheme for single-step disruption of yeast eIF-4E gene. H, HindlII. (B) Southern blot analysis of S. cerevisiae elF-4E gene. Arrows indicate positions of the 2.2-kb (undisrupted allele), 1.6 and 1.4-kb (disrupted allele) restriction fragments. (C) Tetrad analysis. Diploid strains were sporulated and dissected on YEPD rich media. Strain YZ-y4Enull carries a disrupted endogenous elF-4E gene. Strain RS316y4E is a YZ-y4EnuU strain transformed with plasmid pRS316y4E. Methods: Yeast elF-4E gene disruption. Yeast strains containing disrupted elF-4E alleles were constructed by single-step gene transplacement (Rothstein, 1983). pRS316y4Enull was digested with SalI+XbaI, and transformed into the diploid yeast strain w303-D (Ito et al., 1983), His + transformants selected and sporulated for 3 d at room temperature. Tetrads were treated with Zymolase (Sigma), streaked on YEPD plates, and dissected (Sherman et al., 1986). Two days later, tetrads were replicated toplates containing SC medium without His. The verified diploid yeast strain carrying one disrupted allele of the endogenous elF-4E gene was named YZ-y4Enull. Yeast genornic DNA was prepared, digested with restriction enzymes, fractionated on 0.8% agarose gels, transferred to nitrocellulose and hybridized with a2p-labeled SalI-XbaI DNA fragments from pRS316-y4E (Sambrook et al., 1989).
Ala 53 or Asp 53 forms of eIF-4E protein obtained. All strains were equally viable, grew into colonies of the same size, anddisplayed equivalent doubling times (3 h at 30°C; data not shown). Yeast cell growth rates were 75% that of complementation by yeast elF-4E (RS316-Y4E). Mammalian elF-4E proteins therefore permit yeast cell growth despite the absence of phosphorylated Ser 53. Complementation of protein synthetic rates by the human elF-4E proteins was determined by [35S]Met labeling of yeast null mutants expressing Ser 53, Asp 53 or Ala 53 constructs (Table I). All yeast displayed similar protein synthetic rates. Human eIF-4E proteins all efficiently complement yeast protein synthesis regardless of whether a Ser is located at postion 53. Expression of elF-4E proteins was examined by labeling human 293/T-Ag cel]s or yeast in vivo with 35S-Met and mVGDP-affinity chromatography (Fig. 4A-C). All three forms of human e[F-4E protein were synthesized
in yeast cells to similar levels, 3-5-fold more than endogenous yeast eIF-4E when produced from low copy plasmids (Fig. 4B), or more than 50-fold from high copy plasmids (Fig. 4C). Replacement of Ser 53 with Ala53 or Asp 53 did not alter binding to mVGDP-structures. Human eIF-4E proteins also enter into a complex with the yeast 150-kDa eIF-4FT-related polypeptide at a level consistent with their abundance (Fig. 4B). The ability of human Ser s3 and Ala 53 mutants to complement translation was studied in human 293/T-Ag produced cells. Overproduced Ser 53 or Ala5a eIF-4E proteins in transfected 293/T-Ag cells accumulated to over 20-fold higher levels than endogenous eIF-4E (Fig. 4A). Overproduced Ala s3 and Ser 53 eIF-4E proteins also incorporated [32p] PO4 to a 20-fold higher level, consistent with their abundance (Fig. 4A). Translation of cotransfected reporter mRNA ([3Gal) controlled by the RSV LTR and 5' noncoding region, which is elF-4F-
286 dependent, was enhanced approx. 80% by overproduction of Ser 53 or Ala 53 eIF-4E variants (Table II), without a detectable increase in mRNA abundance (data not shown). It is not known why a large increase in eIF-4E abundance produces only a slight increase in reporter translation, but it may reflect the fact that these mRNAs are already well translated. These results demonstrate that mammalian elF-4E protein functions well in protein synthesis without a phosphorylated Ser 53.
Cen6
pGAlfn~
elF-4E
t
BN
B N r-=
,Ao. .~.
Sm
t
B elF-4E
Cen6
Y B
B
paA
B tADH
B
S elF-4 I:
eiF-4E
pGAL10
\
"~ /
/
Cen6 PB
pGAL
lO~tAO H
pGALIO
IS
IF-4E
B eIF-4E B
Fig. 3. Construction of plasmids expressing the wt or variant human elF-4E gene in yeast. Strategy used to construct expression vectors. Details of the cloning strategy are described in Methods. Different forms of the human elF-4E coding region are indicated as elF-4E. The yeast GALIO promoter is indicated as pGAL10. The 2~t plasmid ori is shown as 2g, and the yeast centromere sequence from chromosome 6 as Cen6. Restriction enzyme sites: B, BamHI; N, NotI; S, SalI; Sm, Sinai. Methods: Site-directed mutagenesis and cloning of human elF-4E gene: An EcoRI-BamHI fragment containing the elF-4E coding region was subcloned from pGEM-4Ewt (gift from Dr. R. Rhoads, LSU) into M13mpl9 to create construct M13mpl9-4Ewt. Oligos were used to convert Ser to Asp or Ala at position 53, generating M13mutants mpl9-4Easp and mpl9-4Eala, respectively. Mutagenized sequences were verified by DNA sequencing. A fragment containing the yeast ADH1 termination signal from pGBT9 was inserted in pGAD2 to generate construct pGAD2T, as shown. The SmaI site upstream from the elF-4E coding region in M13mpl9-4Ewt, mpl9-4Easp and mpl9-4Eala was converted to BamHI, then the BamHl fragment (human elF-4E coding region) subcloned to generate pGAD4Ewt, pGAD4Easp and pGAD4Eala. The fragment containing the human elF-4E coding region from pGem-4Ewt was subcloned into the yeast 211 vector pSJ101 (Johnson, 1991 ), to create pSJ4Ewt. The Sinai sites of M 13mpl 9-4Easp and mpl9-4Eala were converted to SalI sites, and the SalI-BamHI fragment inserted into pJS101 to generate pSJ4Easp and pSJ4Eala.
TABLE I Protein synthetic rates of yeast containing human or yeast elF-4E genesa Yeast strain
elF-4E genes (mutants)
cpm x 10-3/i.tg Complementation protein per min (%)
RS316-Y4E GAD4E-wt GAD4E-Ala GAD4E-Asp
yeast plasmid human Ser53 human Ala 5a human Asp53
24.5 14.3 15.1 14.7
100 58 62 60
a Yeast were transformed with plasmids and tetrads were selected. Yeast were labeled with [35S]methionine and extracts prepared as described in the legend to Fig. 4. Protein synthetic rates were calculated by determining the specific activity of TCA-precipitable protein in extracts (Sambrook et al., 1989).
(c) Human elF-4E is not phosphorylated at Ser s3 in human or yeast cells To unambiguously determine whether Ser 53 is phosphorylated, two dimensional tryptic phosphopeptide mapping was performed on the Ser 53 and Ala 53 proteins produced in human and yeast cells by comparing patterns obtained using equal amounts of labeled, endogenous elF-4E and overproduced Ala 53 proteins. This was done to avoid the argument that a Ser 53 protein kinase might be limiting and unable to phosphorylate more than a small fraction of the overproduced wt elF-4E protein at Ser 53. Human elF-4E proteins were labeled with [32p]PO 4 in human 293/T-Ag cells for wt endogenous elF-4E, and in cells transfected with the Ala 53 variant, resolved by SDS-PAGE, eluted from gels and equal amounts of labeled protein digested with trypsin and fractionated (Boyle et al., 1991). Endogenous elF-4E isolated from human 293/T-Ag cells produced 2 or 3 isoelectric forms (Fig. 5). Importantly, the tryptic phosphopeptide pattern of the overproduced Ala 53 variant was identical to that of the endogenous wt protein. Since the Ala 53 variant was overproduced and labeled with [32p]PO 4 to a level more than 20-times that of the endogenous wt protein, the tryptic phosphopeptide pattern of the Ala 53 variant cannot be more than 5% contaminated by wt protein. These data demonstrate that Ser 53 is not a site for phosphorylation of human elF-4E in human cells. Ser 53 and Ala 53 human elF-4E purified from yeast cells were treated in an identical manner. In yeast, the human proteins produced identical but more complex patterns of phosphopeptides than elF-4E in human cells. In addition, none of the spots comigrated with the elF-4E phosphopeptide isolated from human cells (Fig. 5), even when plates were overexposed for long periods of time (data not shown). Because Ser 53 and Ala 53 human elF-4E proteins displayed phosphopeptide patterns identical in composition and intensity, it is apparent that in human and yeast cells Ser 53 is not detectably phosphorylated. (d) Conclusions (1) This study established that Ser 53 is not a site for phosphorylation of the mammalian elF-4E protein in human or yeast cells, despite the high degree of sequence
287 A. human 293/T-Ag cells [:3~]-methionine
3~PO 4
_
== '~ tit
_
' o
B. Yeast
==
' ~
C. Yeast
~" LU
--~ UJ
o
o
[3sS]-methionine
a2PO4
:', o,) ~
=~
~
elF-4F~
4, elF-4E
aP
elF-4E elF-4E
Fig. 4. Synthesis of human wt and Ser53 variant elF-4E proteins in human and yeast cells. (A) Abundance and phosphorylation of eIF-4E in human 293/T-Ag cells untransfected or transfected with the Ala53 variant (pelF-4E-Ala 53) overexpressor. (B) Level of elF-4E synthesized in yeast strains synthesizing endogenous yeast e]IF-4E (W303-1a), human wt elF-4E (GAD4Ewt), Asp 53 (GAD4Easp), or Ala sa (GAD4Eala) variants of human eIF-4E. Yeast were grown in YEI:'/Gal medium to mid-log phase and then labeled with [3sS] methionine. The position of elF-4E and elF-4Fy (p150) migration are indicated. (C) Abundance and phosphorylation of elF-4E in yeast strains synthesizing endogenous elF-4E W(303-1a), human wt elF-4E (SJ4Ewt), or human Ala ~3 variant (SJ4Eala). Methods: Yeast exponentially growing haploid cells at 30°C were centrifuged, washed and resuspended in SC(-Met) labeling medium wilh 80 pCi/ml of [35S]methionine for 3 h, or in YEP/Gal ( - P O 4 ) with 250 pCi/ml of [32p]PO 4 for 4 h. Cells were collected, washed, lysed, and protein extracts prepared as described (Blum et al., 1989). For human 293/T-Ag cells, 10-cm plates were transfected at low confluency (20%) with 20 gg plasmids containing the SV40 ori and expressing wt or Ser53--*Ala5a mutant of elF-4E, and transfection efficiencies monitored by cotransfection of a plasmid expressing 13Gal controlled by the RSV promoter, and immunofluorescent staining of cells in situ on cover slips. 48 h later, cells were labeled with 100 gCi/ml with [35S]methionine, or with 250 pCi/ml 32PO4, elF-4E was purified by mTGDP-attinity chromatography from equal amounts of proteins as described (Marino et al., 1989), and resolved by 0.1% SDS-15% PAGE, followed by autoradiography. TABLE II Protein synthetic activity of Ser5:~ and Ala 53 elF-4E in human cellsa Transfection
[3Gal activity A42o/200 gg extract per h
none 13Gal only 13Gal+ Ser53 elF-4E 13Gal+ Ala 53 elF-4E
0.03 0.45 0.80 0.77
A. Human 293/T-Ag Cells
endogenous
a 293/T-Ag were untransfected (none), transfected with I gg pRSV-13Gal only, or pRSV-13Gal+pelF-4E-wt or pelF-4E-Ala 53, and activity of 13Gal measured on soluble S10 p[otein extracts (Sambrook et al., 1989).
conservation in this region ineukaryotes. A previous study was not able to firmly establish whether wt endogenous elF-4E was phosphorylated at Ser 53 (Kaufman et al., 1993). By comparing the tryptic phosphopeptide pattern of endogenous human elF-4E to an overproduced Ala 53 variant, our results demonstrate that Ser 53 is not a target for phosphorylation in the wt protein as normally expressed in cells (Fig. 5). (2) This study also demonstrated that human elF-4E protein, when mutated from Ser53--*Ala 53, or from Ser53--*Asp 53, promotes translation initiation equally well in both human and yeast cells. (3) Previous studies found that mutation of S e r 53 impaired certain elF-4E activities, such as participation in cell transformation (B,eBenedetti and Rhoads, 1990; Lazaris-Karatzas et al., 1990), whereas others such as cap-binding and translatJ~on initiation are not changed (Kaufman et al., 1993; Joshi-Barve et al., 1990; this report). We suggest that it is mutation of position 53
AlaS~ e l F . 4 E
wt eIF4E
B.
SJRF:wt
Transfected Yeast
SJdFa[a
electroph.
Fig. 5. Two-dimensional tryptic phosphopeptide analysis of human Ser5~ and Ala 53 variant e[F-4E proteins produced in human and yeast cells. (A) [n vivo [a2p] PO4-labeled endogenous wt human eIF-4E and overproduced Ala 53 variant proteins were labeled and purified as described in the legend to Fig. 4, resolved by 0.1% SDS-15% PAGE, eluted, digested with trypsin, and the resulting phosphopeptides resolved by thin-layer plate electrophoresis. (B) Tryptic phospbopeptide analysis of in vivo labeled e[F-4E proteins from yeast cells. Tryptic peptide mapping wasperformed by excising 32p-labeled eIF-4E, eluting from unfixed gels, and equal numbers of cpm corresponding to the e]uted proteins subjected to performic acid oxidization and trypsin digestion. Thin-layer plate electrophoresis was carried out for 45 rain at 1.5 kV in pH 1.9 buffer at 1°C, and chromatography was for 15 h in isobutyric acid buffer (Boyle et al., 1991). The dotted area marks the pattern of tryptic phosphopeptides obtained with endogenous e[F-4E from 293 cells.
288
rather than loss of phosphorylation at this site which impairs some but not all eIF-4E activities. This study also suggests although Ser 53 may be a site for phosphorylation of eIF-4E in vitro, in rabbit reticulocyte lysates, it is not a likely site used in vivo.
ACKNOWLEDGEMENTS
We thank Latika Khatri for expert technical assistance in parts of this work, and Robert E. Rhoads for the human wt eIF-4E gene. This work was supported by Public Health Service grants CA-42357 (to R.J.S.) and GM-30439 (to H.L.K.).
REFERENCES Altmann, M., Edery, I., Sonenberg, N. and Trachsel, H.: Purification and characterization of protein synthesis initiation factor eIF-4E from the yeast Saccharomyces cerevisiae. Biochemistry 24 (1985) 6085-6089. Altmann, M., Handschin, C. and Trachsel, H.: mRNA Cap-binding protein: Cloning of the gene encoding protein synthesis initiation factor eIF-4E from Saccaromyces cerevisiae. Mol. Cell. Biol. 73 (1987) 998-1003. Altmann, M., Sonenberg, N. and Trachsel, H.: Translation in S. cerevisiae: initiation factor 4E-dependent cell free system. Mol. Cell. Biol. 9 (1989) 4467-4471. Blum, S., Mueller, M., Schmid, S.R,, Linder, P. and Trachsel, H.: Translation in S. cerevisiae: initiation factor 4A-dependent cell free system. Proc. Natl. Acad. Sci. USA 86 (1989) 6043-6046. Bonneau, A.M. and Sonenberg, N.: Involvement of the 24kd capbinding protein in regulation of protein synthesis in mitosis. J. Biol. Chem. 262 (1987) 11134-11139. Boyle, W., van dernGeer, P. and Hunter, T.: Phosphopeptide mapping and phosphoamino acid analysis by two-dimensional separation on thin layer cellulose. Methods Enzymol. 201 (1991) 110-149. DeBenedetti, A. and Rhoads, R.E.: Overexpression of eukaryotic protein synthesis initiation factor 4E in Hela cells results in aberrant growth and morphology. Proc. Natl. Acad. Sci. USA 87 (1990) 8212-8216. Duncan, R., Milburn, S.C. and Hershey, J.W.B.: Regulated phosphorylation and low abundance of Hela cell initiation factor eIF-4F suggest a role in translational control. J. Biol. Chem. 262 (1987) 380 388. Feigenblum, D. and Schneider, R.J.: Modification of eukaryotic initiation factor 4F during infection by influenza virus. J. Virol. 67 (1993) 3027-3035. Frederickson, R.M., Montine, K.S. and Sonenberg, N.: Phosphorylation of eucaryotic translation intiation factor 4E is increased in Srctransformed cell lines. Mol. Cell. Biol. 11 (1991) 2896 2900. Frederickson, R.M. and Sonenberg, N.: eIF-4E phosphorylation and the regulation of protein synthesis. In: Ilan, J. (ed.), Translational
regulation of gene expression, Vol. 2. Plenum Press, New York, NY, 1993, pp. 143 162. Grifo, J.A., Tahara, S.M., Morgan, M.A., Shatkin, A.J. and Merrick, W.C.: New initiator activity required for globin mRNA translation. J. Biol. Chem. 258 (1983) 5804-5810. Huang, J. and Schneider, R.J.: Adenovirus inhibition of cellular protein synthesis involves inactivation of cap binding protein. Cell 65 (1991) 271-280. Ito, H., Fukuda,Y., Murata, K. and Kimura, A.: Transformation of intact yeast cells treated with alkali cations. J. Bacteriol. 153 (1983) 163 168. Johnson, S.L.: Structure and Function Analysis of the Yeast CDC4 Gene Product. Ph.D. Thesis, University of Washington, Seattle, WA, 1991. Joshi-Barve, S., Rychlik, W. and Rhoads, R. E.: Alteration of the major phosphorylation site of eukaryotic protein synthesis initiation factor 4E prevents its association with the 48S initiation complex. J. Biol. Chem. 265 (1990) 2979-2983. Kaspar, R., Rychlik, W., White, M.W., Rhoads, R.E. and Morris, D.R.: Simultaneous cytoplasmic redistribution of ribosomal protein L32 mRNA and phosphorylation of eukaryotic initiation factor 4E after mitogenic stimulation of Swiss 3T3 cells. J. Biol. Chem. 265 (1990) 3619-3622. Kaufman, R.J., Murtha-Riel, P., Pittman, D.D. and Davies, M.V.: Characterization of wild-type and Ser ~3mutant eukaryotic initiation factor 4E overexpression in mammalian cells. J. Biol. Chem. 268 (1993) 11902-11909. Lazaris-Karatzas, A., Montine, K. S. and Sonenberg, N.: Malignant transformation by a eukaryotic initiation factor subunit that binds to mRNA 5' cap. Nature 345 (1990) 544-547. Marino, M.W., Pfeffer, L.M., Guidon, P.T. and Donner, D.B.: Tumor necrosis factor inducesphosphorylation of a 28kd mRNA capbinding protein in human cervical carcinoma cells. Proc. Natl. Acad. Sci. USA 86 (1989) 8417-8421. Morley, S.J. and Traugh, J.A.: Phorbol esters stimulate phosphorylation of eukaryotic initiation factors 3, 4B and 4F. J. Biol. Chem. 264 (1989) 2401 2404. Panniers, R., Stewart, E.B., Merrick, W.C. and Henshaw, E.C.: Mechanism of inhibition of polypeptide chain initiation in heat shocked Ehrlich cells involves reduction of eukaryotic initiation factor 4F activity. J. Biol. Chem. 260 (1985) 9648-9653. Rhoads, R.E.: Cap recognition and the entry of mRNA into the protein synthesis initiation cycle. Trends Biochem. Sci. 13 (1988) 52-56. Rothstein, RJ.: One-step gene disruption in yeast. Methods Enzymol. 101 (1983) 202-211. Sambrook, J., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989. Sherman, F., Fink, G.R. and Hicks, J.B.: Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1986. Sikorski, R.S. and Hieter, P.: A system of shuttle vectors and yeast host strains designed for effeicient manipulation of DNA in Saceharomyces cerevisiae. Genetics 122 (1989) 19-27. Sonenberg, N., Morgan, M.A., Merrick, W.C. and Shatkin, A.J.: A polypeptide in eukaryotic initiation factors that crosslinks specially to the Y-terminal cap in mRNA. Proc. Natl. Acad. Sci. USA 75 (1978) 4843 4847. Zhang, Y. and Schneider, R.J.: Adenovirus inhibition of cell translation facilitates release of virus particles and enhances degradation of the cytokeratin network. J. Virol. 68 (1994) 2544-2555.
283
GENE 09020
Role of Ser-53 phosphorylation in the activity of human translation initiation factor eIF-4E in mammalian and yeast cells (Cap-binding component; protein synthesis)
Yan Zhang *, Hannah L. Klein and Robert J. Schneider Department of Biochemistry, and Kaplan Cancer Center, New York University Medical Center, New York, NY10016, USA
Received by G.N. Godson: 18 Jaauary 1995; Revised/Accepted: 8 March/6 April 1995; Received at publishers: 28 April 1995
SUMMARY
Eukaryotic translation initiation factor eIF-4E is essential for protein synthesis and cell viability, eIF-4E participates in formation of an mTGTP-cap binding protein complex that mediates association of 40S ribosomal subunits with mRNAs, which occurs only when eIF-4E is phosphorylated. Regulation of eIF-4E by phosphorylation was thought to occur on Set s3, although results potentially inconsistent with phosphorylation of this site have been reported. To resolve whether Set 53 is phosphorylated, and if so whether it regulates eIF-4E activity, we directly examined whether Ser 53 is a site for phophorylation of mammalian eIF-4E in human and yeast cells. Wild-type (wt) human eIF-4E protein variants, Ser s3 ~Asp 53 or Ser 5a --*Ala53, were constructed and analyzed by overproduction in transfected human 293/T-Ag cells, or in Saccharomyces cerevisiae in which the endogenous elF-4E gene was disrupted. Wt eIF-4E and Ser 53 mutants functioned equally well in protein synthesis in both systems, and were phosphorylated to the same extent. Most importantly, the wt and Ser 53 mutants of human elF-4E produced identical tryptic phophopeptide patterns in human cells, and identical but more complicated patterns in yeast. These data demonstrate that Ser 53 is not a requisite activating site for phosphorylation of mammalian elF-4E in human or yeast cells, under conditions in which it participates in protein synthesis.
INTRODUCTION
Initiation of protein synthesis is a rate-limiting step in eukaryotes. A key target for regulation is the translation Correspondence to: Dr. R.J. Schneider, Department of Biochemistry, NYU Medical Center, New York, NY 10016, USA. Tel. (1-212) 263-6006; Fax (1-212) 263-8166; e-mail: [email protected] * Current address: Howard Haghes Medical Institute, Department of Biochemistry and Biophysics, University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA.
Abbreviations: A, absorbance (1 cm); aa, amino acid(s); Ad, adenovirus; 13Gal, 13-galactosidase;bp, base pair(s); CEN (Cen), centromere; elF, eukaryotic initiation factor; eIF-4A, elF-413; elF-4E, eIF-4~; p220, elF-47; kb, kilobase(s) or 1000 bp; LTR, long terminal repeat; nt, nucleotide(s); oligo, oligodeoxyribonucleotide; ori, origin of DNA replication; p, plasmid; PAGE, polyacrylamide-gel electrophoresis; PCR, polymerase chain reaction; r-, ribosc,mal; RSV, Rous sarcoma virus; SDS, sodium dodecyl sulfate; TCA, trichloroacetic acid; wt, wild type. SSDI 0378-1119(95)00302-9
factor eIF-4F (reviewed in Frederickson and Sonenberg, 1993). Factor eIF-4F is required for translation of most capped mRNAs. Mammalian eIF-4F consists of three subunits: a 220-kDa protein (p220) whose function is unknown, a 50-kDa protein (eIF-4A) which is an ATPdependent RNA helicase (Grifo et al., 1983) and a 24-kDa protein (eIF-4E), the cap-binding component of eIF-4F (Sonenberg et al., 1978). eIF-4F is therefore a capdependent RNA helicase, facilitating 5' unwinding of mRNA and interaction with 40S r-subunits (reviewed in Rhoads, 1988). In yeast S. cerevisiae, disruption of the elF-4E gene results in cell death (Altmann et al., 1987; Blum et al., 1989), which is prevented by expression of the mammalian elF-4E gene (Altmann et al., 1989). S. cerevisiae eIF-4E is 33% identical to mouse and human eIF-4E, and functions in a similar manner (Altmann et al., 1985). Many segments of human and yeast eIF-4E are
284 well conserved, including the Ser 53 position and its flanking sequences (Altmann et al., 1987). Reduced phosphorylation of elF-4E correlates with loss of elF-4F function and inhibition of translation during heat shock (Panniers et al., 1985; Duncan et al., 1987), mitosis (Bonneau and Sonenberg, 1987; H u a n g and Schneider, 1991 ), in cells infected by Ad (Huang and Schneider, 1991), and possibly by influenza viruses (Feigenblum and Schneider, 1993; reviewed in Zhang and Schneider, 1994). Increased phosphorylation of elF-4E enhances elF-4F function, and is observed in cells treated with mitogens or serum (Marino et al., 1989; Morley and Traugh, 1989; Kaspar et al., 1990), or transformed with oncogenes Src or Ras (Frederickson et al., 1991). Phosphorylation of elF-4E was suspected to occur on Ser 53 for a variety of experimental reasons (reviewed in Frederickson and Sonenberg, 1993). It was therefore surprising that elF-4E mutants containing Ala or Asp in place of Ser 53 were phosphorylated when overproduced in Cos cells, a monkey cell line, and did not prevent capdependent translation (Kaufman et al., 1993). In this study we therefore sought to clarify the role of Ser 53 phosphorylation in activity of mammalian elF-4E.
EXPERIMENTAL AND DISCUSSION
(a) Disruption of the endogenous yeast elF-4E gene The yeast elF-4E gene was deleted to permit comparison of the translation function and phosphorylation pattern of human eIF-4E protein in human and yeast cells. The yeast elF-4E gene was amplified by PCR, cloned into the yeast-E, coli shuttle vector, pRS316 (Fig. 1), and used todisrupt the yeast gene in a one-step procedure (Rothstein, 1983). Diploid cells were transformed with a 3.2-kb SalI-XbaI fragment from pRS316-y4Enull, which carries flanking sequences for the yeast elF-4E gene as well as the yeast HIS3 gene, which serves as a genetic marker (Fig. 2A). Disruption of the yeast elF-4E gene in His + transformants was verified by Southern blot analysis (Fig. 2B). Mutant and wt alleles were separated by tetrad analysis, and only half grew into colonies as expected if the His + phenotype is associated with the deleted allele. Loss of yeast elF-4E function was established by transformation of the heterozygous null allele strain with construct pRS316-y4E, which expresses the yeast elF-4E gene. Transformation rendered more than two out of four spores viable, indicating that the elF-4E defect was rescued by the yeast elF-4E gene. Plasmid pRS316-y4E replicates as a centromeric vector, segregating into two of four haploid spores. Some elF-4E-null spores therefore lack the plasmid and are inviable as expected, while null allele His + spores carry
Cen6
(
IX S IIBIBB BBIBX 0.8kb 0.6kb k
~ Cen6
X 1
S
S
y4E 2.2kb
ten6 X
BH / ~$3 B
X
1.8kb E
Cen6
S
B
Fig. 1. PCR cloning of yeast eIF-4E gene. Strategy used to construct vectors pRS316-y4E and pRS-y4Enull. The yeast eIF-4E coding region is indicated as y4E. The yeast centromere sequence from chromosome 6 is defined as Cen6. S, SalI; B, BamH1; X, Xbal. Methods: DNA fragments containing different parts of yeast elF-4E coding sequence were cloned by PCR amplificationof the 2.2-kb HindIII fragment with primers selected to amplify the 2.0-kb full-length coding region (nt 83-107 to 2041-2066), the 760-bp upstream regulatory region (nt 83 1-7 to 815-842), and the 571-bp fragment 3' noncoding region (nt 2041-2066 to 1496-1519). Primers included restriction sites for cloning. PCR amplified fragments were purified with Qiax (Qiagen), subjected to restriction enzymedigestions and cloned as shown in the diagram. The 2.0-kb SalI-XbaI restriction fragment was inserted into vector pRS316 (Sikorski and Hieter, 1989) to generate pRS316-y4E, the 0.76-kb SallBamHI and 0.67-kb BamHI-XbaI fragments were inserted between SalI and XbaI sites of pRS316 to generate pRS316-y4ED. The yeast HIS3 gene was inserted into pRS316-y4ED to generate pRS316-y4Enull. pRS316-y4E and are rescued. These results confirm a previous report (Altmann et al., 1987) that the yeast elF-4E gene is required for cell viability, and could therefore be used to examine the role of Ser 53 phosphorylation in eIF-4E activity in yeast and human cells.
(b) Human elF-4E does not require Ser53 phosphorylation to permit translation in yeast and human cells Since mouse eIF-4E substitutes for the yeast homolog in vivo by permitting yeast cell growth (Altmann et al., 1989), and mouse and human eIF-4E proteins differ by only one amino acid, they should be interchangeable. An eIF-4E Ser53~Ala53 human variant (Ala 53) and an eIF-4E Ser 53 --*Asp53 variant (Asp 53) were constructed to study the importance of phosphorylation at residue Ser 53. Wt and variant human elF-4E genes were expressed from the low-copy yeast C E N vector pGAD2, under GALIO promoter control and T R P 1 gene selection (Fig. 3). Transformants were sporulated, dissected onto rich plates containing Gal to induce expression of human eIF-4E and viable His + spores synthesizing human wt (SerS3),
285 A
B
1.8kb H H
"7 o
II 0.76 kb
~'
Transform
c Lu
03
N
>-
0.57 kb
diploid his3"
H H
I I X
=1= t-~|
elF-4E
X
= 2.2 kb
!
~H
JiL
2.2 kb
Double crossover
•'9[-- 1.6 kb i
!
....
~'--
1.4
kb
r
0.2 kb
C YZ-y4Enull
RS3'I6y4E
Fig. 2. Method and analysis of yeast elF-4E gene disruption. (A) Scheme for single-step disruption of yeast eIF-4E gene. H, HindlII. (B) Southern blot analysis of S. cerevisiae elF-4E gene. Arrows indicate positions of the 2.2-kb (undisrupted allele), 1.6 and 1.4-kb (disrupted allele) restriction fragments. (C) Tetrad analysis. Diploid strains were sporulated and dissected on YEPD rich media. Strain YZ-y4Enull carries a disrupted endogenous elF-4E gene. Strain RS316y4E is a YZ-y4EnuU strain transformed with plasmid pRS316y4E. Methods: Yeast elF-4E gene disruption. Yeast strains containing disrupted elF-4E alleles were constructed by single-step gene transplacement (Rothstein, 1983). pRS316y4Enull was digested with SalI+XbaI, and transformed into the diploid yeast strain w303-D (Ito et al., 1983), His + transformants selected and sporulated for 3 d at room temperature. Tetrads were treated with Zymolase (Sigma), streaked on YEPD plates, and dissected (Sherman et al., 1986). Two days later, tetrads were replicated toplates containing SC medium without His. The verified diploid yeast strain carrying one disrupted allele of the endogenous elF-4E gene was named YZ-y4Enull. Yeast genornic DNA was prepared, digested with restriction enzymes, fractionated on 0.8% agarose gels, transferred to nitrocellulose and hybridized with a2p-labeled SalI-XbaI DNA fragments from pRS316-y4E (Sambrook et al., 1989).
Ala 53 or Asp 53 forms of eIF-4E protein obtained. All strains were equally viable, grew into colonies of the same size, anddisplayed equivalent doubling times (3 h at 30°C; data not shown). Yeast cell growth rates were 75% that of complementation by yeast elF-4E (RS316-Y4E). Mammalian elF-4E proteins therefore permit yeast cell growth despite the absence of phosphorylated Ser 53. Complementation of protein synthetic rates by the human elF-4E proteins was determined by [35S]Met labeling of yeast null mutants expressing Ser 53, Asp 53 or Ala 53 constructs (Table I). All yeast displayed similar protein synthetic rates. Human eIF-4E proteins all efficiently complement yeast protein synthesis regardless of whether a Ser is located at postion 53. Expression of elF-4E proteins was examined by labeling human 293/T-Ag cel]s or yeast in vivo with 35S-Met and mVGDP-affinity chromatography (Fig. 4A-C). All three forms of human e[F-4E protein were synthesized
in yeast cells to similar levels, 3-5-fold more than endogenous yeast eIF-4E when produced from low copy plasmids (Fig. 4B), or more than 50-fold from high copy plasmids (Fig. 4C). Replacement of Ser 53 with Ala53 or Asp 53 did not alter binding to mVGDP-structures. Human eIF-4E proteins also enter into a complex with the yeast 150-kDa eIF-4FT-related polypeptide at a level consistent with their abundance (Fig. 4B). The ability of human Ser s3 and Ala 53 mutants to complement translation was studied in human 293/T-Ag produced cells. Overproduced Ser 53 or Ala5a eIF-4E proteins in transfected 293/T-Ag cells accumulated to over 20-fold higher levels than endogenous eIF-4E (Fig. 4A). Overproduced Ala s3 and Ser 53 eIF-4E proteins also incorporated [32p] PO4 to a 20-fold higher level, consistent with their abundance (Fig. 4A). Translation of cotransfected reporter mRNA ([3Gal) controlled by the RSV LTR and 5' noncoding region, which is elF-4F-
286 dependent, was enhanced approx. 80% by overproduction of Ser 53 or Ala 53 eIF-4E variants (Table II), without a detectable increase in mRNA abundance (data not shown). It is not known why a large increase in eIF-4E abundance produces only a slight increase in reporter translation, but it may reflect the fact that these mRNAs are already well translated. These results demonstrate that mammalian elF-4E protein functions well in protein synthesis without a phosphorylated Ser 53.
Cen6
pGAlfn~
elF-4E
t
BN
B N r-=
,Ao. .~.
Sm
t
B elF-4E
Cen6
Y B
B
paA
B tADH
B
S elF-4 I:
eiF-4E
pGAL10
\
"~ /
/
Cen6 PB
pGAL
lO~tAO H
pGALIO
IS
IF-4E
B eIF-4E B
Fig. 3. Construction of plasmids expressing the wt or variant human elF-4E gene in yeast. Strategy used to construct expression vectors. Details of the cloning strategy are described in Methods. Different forms of the human elF-4E coding region are indicated as elF-4E. The yeast GALIO promoter is indicated as pGAL10. The 2~t plasmid ori is shown as 2g, and the yeast centromere sequence from chromosome 6 as Cen6. Restriction enzyme sites: B, BamHI; N, NotI; S, SalI; Sm, Sinai. Methods: Site-directed mutagenesis and cloning of human elF-4E gene: An EcoRI-BamHI fragment containing the elF-4E coding region was subcloned from pGEM-4Ewt (gift from Dr. R. Rhoads, LSU) into M13mpl9 to create construct M13mpl9-4Ewt. Oligos were used to convert Ser to Asp or Ala at position 53, generating M13mutants mpl9-4Easp and mpl9-4Eala, respectively. Mutagenized sequences were verified by DNA sequencing. A fragment containing the yeast ADH1 termination signal from pGBT9 was inserted in pGAD2 to generate construct pGAD2T, as shown. The SmaI site upstream from the elF-4E coding region in M13mpl9-4Ewt, mpl9-4Easp and mpl9-4Eala was converted to BamHI, then the BamHl fragment (human elF-4E coding region) subcloned to generate pGAD4Ewt, pGAD4Easp and pGAD4Eala. The fragment containing the human elF-4E coding region from pGem-4Ewt was subcloned into the yeast 211 vector pSJ101 (Johnson, 1991 ), to create pSJ4Ewt. The Sinai sites of M 13mpl 9-4Easp and mpl9-4Eala were converted to SalI sites, and the SalI-BamHI fragment inserted into pJS101 to generate pSJ4Easp and pSJ4Eala.
TABLE I Protein synthetic rates of yeast containing human or yeast elF-4E genesa Yeast strain
elF-4E genes (mutants)
cpm x 10-3/i.tg Complementation protein per min (%)
RS316-Y4E GAD4E-wt GAD4E-Ala GAD4E-Asp
yeast plasmid human Ser53 human Ala 5a human Asp53
24.5 14.3 15.1 14.7
100 58 62 60
a Yeast were transformed with plasmids and tetrads were selected. Yeast were labeled with [35S]methionine and extracts prepared as described in the legend to Fig. 4. Protein synthetic rates were calculated by determining the specific activity of TCA-precipitable protein in extracts (Sambrook et al., 1989).
(c) Human elF-4E is not phosphorylated at Ser s3 in human or yeast cells To unambiguously determine whether Ser 53 is phosphorylated, two dimensional tryptic phosphopeptide mapping was performed on the Ser 53 and Ala 53 proteins produced in human and yeast cells by comparing patterns obtained using equal amounts of labeled, endogenous elF-4E and overproduced Ala 53 proteins. This was done to avoid the argument that a Ser 53 protein kinase might be limiting and unable to phosphorylate more than a small fraction of the overproduced wt elF-4E protein at Ser 53. Human elF-4E proteins were labeled with [32p]PO 4 in human 293/T-Ag cells for wt endogenous elF-4E, and in cells transfected with the Ala 53 variant, resolved by SDS-PAGE, eluted from gels and equal amounts of labeled protein digested with trypsin and fractionated (Boyle et al., 1991). Endogenous elF-4E isolated from human 293/T-Ag cells produced 2 or 3 isoelectric forms (Fig. 5). Importantly, the tryptic phosphopeptide pattern of the overproduced Ala 53 variant was identical to that of the endogenous wt protein. Since the Ala 53 variant was overproduced and labeled with [32p]PO 4 to a level more than 20-times that of the endogenous wt protein, the tryptic phosphopeptide pattern of the Ala 53 variant cannot be more than 5% contaminated by wt protein. These data demonstrate that Ser 53 is not a site for phosphorylation of human elF-4E in human cells. Ser 53 and Ala 53 human elF-4E purified from yeast cells were treated in an identical manner. In yeast, the human proteins produced identical but more complex patterns of phosphopeptides than elF-4E in human cells. In addition, none of the spots comigrated with the elF-4E phosphopeptide isolated from human cells (Fig. 5), even when plates were overexposed for long periods of time (data not shown). Because Ser 53 and Ala 53 human elF-4E proteins displayed phosphopeptide patterns identical in composition and intensity, it is apparent that in human and yeast cells Ser 53 is not detectably phosphorylated. (d) Conclusions (1) This study established that Ser 53 is not a site for phosphorylation of the mammalian elF-4E protein in human or yeast cells, despite the high degree of sequence
287 A. human 293/T-Ag cells [:3~]-methionine
3~PO 4
_
== '~ tit
_
' o
B. Yeast
==
' ~
C. Yeast
~" LU
--~ UJ
o
o
[3sS]-methionine
a2PO4
:', o,) ~
=~
~
elF-4F~
4, elF-4E
aP
elF-4E elF-4E
Fig. 4. Synthesis of human wt and Ser53 variant elF-4E proteins in human and yeast cells. (A) Abundance and phosphorylation of eIF-4E in human 293/T-Ag cells untransfected or transfected with the Ala53 variant (pelF-4E-Ala 53) overexpressor. (B) Level of elF-4E synthesized in yeast strains synthesizing endogenous yeast e]IF-4E (W303-1a), human wt elF-4E (GAD4Ewt), Asp 53 (GAD4Easp), or Ala sa (GAD4Eala) variants of human eIF-4E. Yeast were grown in YEI:'/Gal medium to mid-log phase and then labeled with [3sS] methionine. The position of elF-4E and elF-4Fy (p150) migration are indicated. (C) Abundance and phosphorylation of elF-4E in yeast strains synthesizing endogenous elF-4E W(303-1a), human wt elF-4E (SJ4Ewt), or human Ala ~3 variant (SJ4Eala). Methods: Yeast exponentially growing haploid cells at 30°C were centrifuged, washed and resuspended in SC(-Met) labeling medium wilh 80 pCi/ml of [35S]methionine for 3 h, or in YEP/Gal ( - P O 4 ) with 250 pCi/ml of [32p]PO 4 for 4 h. Cells were collected, washed, lysed, and protein extracts prepared as described (Blum et al., 1989). For human 293/T-Ag cells, 10-cm plates were transfected at low confluency (20%) with 20 gg plasmids containing the SV40 ori and expressing wt or Ser53--*Ala5a mutant of elF-4E, and transfection efficiencies monitored by cotransfection of a plasmid expressing 13Gal controlled by the RSV promoter, and immunofluorescent staining of cells in situ on cover slips. 48 h later, cells were labeled with 100 gCi/ml with [35S]methionine, or with 250 pCi/ml 32PO4, elF-4E was purified by mTGDP-attinity chromatography from equal amounts of proteins as described (Marino et al., 1989), and resolved by 0.1% SDS-15% PAGE, followed by autoradiography. TABLE II Protein synthetic activity of Ser5:~ and Ala 53 elF-4E in human cellsa Transfection
[3Gal activity A42o/200 gg extract per h
none 13Gal only 13Gal+ Ser53 elF-4E 13Gal+ Ala 53 elF-4E
0.03 0.45 0.80 0.77
A. Human 293/T-Ag Cells
endogenous
a 293/T-Ag were untransfected (none), transfected with I gg pRSV-13Gal only, or pRSV-13Gal+pelF-4E-wt or pelF-4E-Ala 53, and activity of 13Gal measured on soluble S10 p[otein extracts (Sambrook et al., 1989).
conservation in this region ineukaryotes. A previous study was not able to firmly establish whether wt endogenous elF-4E was phosphorylated at Ser 53 (Kaufman et al., 1993). By comparing the tryptic phosphopeptide pattern of endogenous human elF-4E to an overproduced Ala 53 variant, our results demonstrate that Ser 53 is not a target for phosphorylation in the wt protein as normally expressed in cells (Fig. 5). (2) This study also demonstrated that human elF-4E protein, when mutated from Ser53--*Ala 53, or from Ser53--*Asp 53, promotes translation initiation equally well in both human and yeast cells. (3) Previous studies found that mutation of S e r 53 impaired certain elF-4E activities, such as participation in cell transformation (B,eBenedetti and Rhoads, 1990; Lazaris-Karatzas et al., 1990), whereas others such as cap-binding and translatJ~on initiation are not changed (Kaufman et al., 1993; Joshi-Barve et al., 1990; this report). We suggest that it is mutation of position 53
AlaS~ e l F . 4 E
wt eIF4E
B.
SJRF:wt
Transfected Yeast
SJdFa[a
electroph.
Fig. 5. Two-dimensional tryptic phosphopeptide analysis of human Ser5~ and Ala 53 variant e[F-4E proteins produced in human and yeast cells. (A) [n vivo [a2p] PO4-labeled endogenous wt human eIF-4E and overproduced Ala 53 variant proteins were labeled and purified as described in the legend to Fig. 4, resolved by 0.1% SDS-15% PAGE, eluted, digested with trypsin, and the resulting phosphopeptides resolved by thin-layer plate electrophoresis. (B) Tryptic phospbopeptide analysis of in vivo labeled e[F-4E proteins from yeast cells. Tryptic peptide mapping wasperformed by excising 32p-labeled eIF-4E, eluting from unfixed gels, and equal numbers of cpm corresponding to the e]uted proteins subjected to performic acid oxidization and trypsin digestion. Thin-layer plate electrophoresis was carried out for 45 rain at 1.5 kV in pH 1.9 buffer at 1°C, and chromatography was for 15 h in isobutyric acid buffer (Boyle et al., 1991). The dotted area marks the pattern of tryptic phosphopeptides obtained with endogenous e[F-4E from 293 cells.
288
rather than loss of phosphorylation at this site which impairs some but not all eIF-4E activities. This study also suggests although Ser 53 may be a site for phosphorylation of eIF-4E in vitro, in rabbit reticulocyte lysates, it is not a likely site used in vivo.
ACKNOWLEDGEMENTS
We thank Latika Khatri for expert technical assistance in parts of this work, and Robert E. Rhoads for the human wt eIF-4E gene. This work was supported by Public Health Service grants CA-42357 (to R.J.S.) and GM-30439 (to H.L.K.).
REFERENCES Altmann, M., Edery, I., Sonenberg, N. and Trachsel, H.: Purification and characterization of protein synthesis initiation factor eIF-4E from the yeast Saccharomyces cerevisiae. Biochemistry 24 (1985) 6085-6089. Altmann, M., Handschin, C. and Trachsel, H.: mRNA Cap-binding protein: Cloning of the gene encoding protein synthesis initiation factor eIF-4E from Saccaromyces cerevisiae. Mol. Cell. Biol. 73 (1987) 998-1003. Altmann, M., Sonenberg, N. and Trachsel, H.: Translation in S. cerevisiae: initiation factor 4E-dependent cell free system. Mol. Cell. Biol. 9 (1989) 4467-4471. Blum, S., Mueller, M., Schmid, S.R,, Linder, P. and Trachsel, H.: Translation in S. cerevisiae: initiation factor 4A-dependent cell free system. Proc. Natl. Acad. Sci. USA 86 (1989) 6043-6046. Bonneau, A.M. and Sonenberg, N.: Involvement of the 24kd capbinding protein in regulation of protein synthesis in mitosis. J. Biol. Chem. 262 (1987) 11134-11139. Boyle, W., van dernGeer, P. and Hunter, T.: Phosphopeptide mapping and phosphoamino acid analysis by two-dimensional separation on thin layer cellulose. Methods Enzymol. 201 (1991) 110-149. DeBenedetti, A. and Rhoads, R.E.: Overexpression of eukaryotic protein synthesis initiation factor 4E in Hela cells results in aberrant growth and morphology. Proc. Natl. Acad. Sci. USA 87 (1990) 8212-8216. Duncan, R., Milburn, S.C. and Hershey, J.W.B.: Regulated phosphorylation and low abundance of Hela cell initiation factor eIF-4F suggest a role in translational control. J. Biol. Chem. 262 (1987) 380 388. Feigenblum, D. and Schneider, R.J.: Modification of eukaryotic initiation factor 4F during infection by influenza virus. J. Virol. 67 (1993) 3027-3035. Frederickson, R.M., Montine, K.S. and Sonenberg, N.: Phosphorylation of eucaryotic translation intiation factor 4E is increased in Srctransformed cell lines. Mol. Cell. Biol. 11 (1991) 2896 2900. Frederickson, R.M. and Sonenberg, N.: eIF-4E phosphorylation and the regulation of protein synthesis. In: Ilan, J. (ed.), Translational
regulation of gene expression, Vol. 2. Plenum Press, New York, NY, 1993, pp. 143 162. Grifo, J.A., Tahara, S.M., Morgan, M.A., Shatkin, A.J. and Merrick, W.C.: New initiator activity required for globin mRNA translation. J. Biol. Chem. 258 (1983) 5804-5810. Huang, J. and Schneider, R.J.: Adenovirus inhibition of cellular protein synthesis involves inactivation of cap binding protein. Cell 65 (1991) 271-280. Ito, H., Fukuda,Y., Murata, K. and Kimura, A.: Transformation of intact yeast cells treated with alkali cations. J. Bacteriol. 153 (1983) 163 168. Johnson, S.L.: Structure and Function Analysis of the Yeast CDC4 Gene Product. Ph.D. Thesis, University of Washington, Seattle, WA, 1991. Joshi-Barve, S., Rychlik, W. and Rhoads, R. E.: Alteration of the major phosphorylation site of eukaryotic protein synthesis initiation factor 4E prevents its association with the 48S initiation complex. J. Biol. Chem. 265 (1990) 2979-2983. Kaspar, R., Rychlik, W., White, M.W., Rhoads, R.E. and Morris, D.R.: Simultaneous cytoplasmic redistribution of ribosomal protein L32 mRNA and phosphorylation of eukaryotic initiation factor 4E after mitogenic stimulation of Swiss 3T3 cells. J. Biol. Chem. 265 (1990) 3619-3622. Kaufman, R.J., Murtha-Riel, P., Pittman, D.D. and Davies, M.V.: Characterization of wild-type and Ser ~3mutant eukaryotic initiation factor 4E overexpression in mammalian cells. J. Biol. Chem. 268 (1993) 11902-11909. Lazaris-Karatzas, A., Montine, K. S. and Sonenberg, N.: Malignant transformation by a eukaryotic initiation factor subunit that binds to mRNA 5' cap. Nature 345 (1990) 544-547. Marino, M.W., Pfeffer, L.M., Guidon, P.T. and Donner, D.B.: Tumor necrosis factor inducesphosphorylation of a 28kd mRNA capbinding protein in human cervical carcinoma cells. Proc. Natl. Acad. Sci. USA 86 (1989) 8417-8421. Morley, S.J. and Traugh, J.A.: Phorbol esters stimulate phosphorylation of eukaryotic initiation factors 3, 4B and 4F. J. Biol. Chem. 264 (1989) 2401 2404. Panniers, R., Stewart, E.B., Merrick, W.C. and Henshaw, E.C.: Mechanism of inhibition of polypeptide chain initiation in heat shocked Ehrlich cells involves reduction of eukaryotic initiation factor 4F activity. J. Biol. Chem. 260 (1985) 9648-9653. Rhoads, R.E.: Cap recognition and the entry of mRNA into the protein synthesis initiation cycle. Trends Biochem. Sci. 13 (1988) 52-56. Rothstein, RJ.: One-step gene disruption in yeast. Methods Enzymol. 101 (1983) 202-211. Sambrook, J., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989. Sherman, F., Fink, G.R. and Hicks, J.B.: Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1986. Sikorski, R.S. and Hieter, P.: A system of shuttle vectors and yeast host strains designed for effeicient manipulation of DNA in Saceharomyces cerevisiae. Genetics 122 (1989) 19-27. Sonenberg, N., Morgan, M.A., Merrick, W.C. and Shatkin, A.J.: A polypeptide in eukaryotic initiation factors that crosslinks specially to the Y-terminal cap in mRNA. Proc. Natl. Acad. Sci. USA 75 (1978) 4843 4847. Zhang, Y. and Schneider, R.J.: Adenovirus inhibition of cell translation facilitates release of virus particles and enhances degradation of the cytokeratin network. J. Virol. 68 (1994) 2544-2555.