May 6, 2015 - the $Nara Institute of Science and Technology, 8916-5 Takayama, Zkoma, Nara 630-01, ... Protein synthesis elongation factor 2 (EF-2) from eu-.
VOl.
THEJOURNAL OF BIOUXICAL CHEMISITY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc
269, No. 18, Issue of May 6, PP. 13497-13501, 1994 Printed in U.S.A.
Elongation Factor2 Mutants Deficient in Diphthamide Formation Show Temperature-sensitive Cell Growth" (Received for publication, November 2, 1993, and in revised form, January 31, 1994)
Yukio KimataS and Kegji KohnoSBn From the Unstitute for Molecular and Cellular Biology, Osaka University, 1-3, Yamada-oka, Suita, Osaka 565, Japan and the $Nara Institute of Science and Technology, 8916-5 Takayama, Zkoma, Nara 630-01, Japan
DT-resistant strains producing non-ADP-ribosylatable EF-2 Protein synthesis elongation factor 2 (EF-2) from euhave been isolated from cultured mammalian cells and yeast karyotes contains an unusual modified histidine resihas been shown and have been classified into two groups showingeither a redue, termed diphthamide. Diphthamide to be asite of ADP-ribosylation by bacterial toxins, but cessive or a dominant trait (6, 12-19). The former group is its functionremainsobscure. We expressedmutant defective in the enzymes responsible for the biosynthesis of genes of EF-2 with substitutionsof 19 other amino acids diphthamide (6, 16-18). These mutant strains synthesize infor His-699, whichis modified to diphthamide, in yeast complete forms of diphthamide, which are resistant to ADPcells and found that they can be classified into three ribosylation by toxins.The strains in the latter group are strucl), replacement of His- tural gene m u t a n t s of EF-2 (12, 13, 15, 19). For example in groups. In the first group (Group 699 by the basic amino acid Arg or Lys showed not only Chinesehamsterovarycells(CHO),wehavefound that a loss of EF-2 activity but also inhibitory effects on the single amino acid substitution of Arg for Gly, which is a well growth of cells co-expressing wild-type EF-2. In the secconserved amino acid located2 residues tothe carboxyl-termiond group (Group21, replacement with Gly, Pro, Ser, or Asp resulted in nonfunctionalEF-2, but it did not affect nal side of the diphthamide, frequently occurs andresults in EF-2 EF-2. In the resistance to toxins (15). This mutation also makes yeast the growth ofcells co-expressing wild-type third group (Group 3), replacement by one of the otherDT-resistant (19). However, we have failed to isolate any mu13 amino acids resulted in a functional EF-2. In the tants with replacement of the target His residue by other Group 3 mutants, EF-2 was not ribosylated by diphthe- amino acids from CHO cells(15). In this study, we haveused yeast since this is convenient for EF-2s did not form ria toxin, indicating that the mutant genetic manipulation. S. cerevisiae has two functional EF-2 diphthamide.However,theviable cells grewmore EF-2 and showed genes (EFT1 a n d EFT2) per haploid genome that encode the slowly than cells expressing wild-type same amino acid sequence(8,9). Disruption of one of t h e EF-2 temperature sensitivities. Thisresultsuggeststhat diphthamide may confer heat resistance on EF-2, al- genes has little effect on the cell growth rate, but cells do not though it still may be active without diphthamide at a survive when both genesare disrupted (8, 9). To elucidate the normal temperature. importance of the His precursor and the physiological role of diphthamide in EF-2, we constructed and characterizeda seEukaryotic elongation factor 2 (EF-2)' is an essential factor that catalyzes the GTP hydrolysis-dependent translocation of peptidyl-tRNA from the aminoacyl site to the peptidyl site on the ribosome during protein synthesis. Diphtheria toxin (DT) or Pseudomonas exotoxin A specifically inactivates EF-2 by transferring the ADP-ribose moiety of NAD+ to the histidine N1-imidazole ring of diphthamide, which is a unique amino acid and is the site of the action of the toxin that results in lethal blockage of protein synthesis (1-5). Diphthamide is synthesized in cells by a multistep post-translational modification of the His residue, encoded at codon 715 in mammalian EF-2 and at codon 699 in budding yeast Saccharomyces cerevisiae EF-2, to 2-[3-carboxyamido-3-(trimethylammonio)propyl]histidine (5-9). ArchaebacteriaharborEF-2containingdiphthamide or diphthine,a deamidated form of diphthamide, that is also recognized by DT (10, 11).
* This work was supported in part by the Toray Science Foundation
and by Grants 04833017 and 04262206 from the Ministry of Science, Education, and Cultureof Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement"in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. This article is dedicated to the late Professor Tsuyoshi Uchida who died of cancer on May 3, 1989. ll To whom correspondence should be addressed. Nara Institute of Science and Technology, 8916 Takayama, Ikoma, Mara 630-01, Japan. Tel.: 81-7437-2-5640; Fax: 81-7437-2-5649. The abbreviations used are: EF-2, elongation factor 2; DT, diphtheria toxin; CHO, Chinese hamster ovary; 5-FOA, 5-fluoroorotic acid.
ries of structural gene mutantsof EF-2 in which His-699 was replaced by 19 other amino acids (EF-2s(His-699 X ) ) . "-f
MATERIALSANDMETHODS Plasmids-The methods used for plasmid construction were as described by Sambrook et al. (20). Plasmid pEFT2, a pUC118-based plasmid, cames a 3.8-kilobase long EcoRI-EcoRI fragment containing the entire yeast EF-2 gene (EFT2), including its own promoter (19). Plasmid pRS314 is a yeast centromeric vector containing the TRPl selectable marker as described by Sikorski and Hieter (21).As shown in Fig. I A , PED plasmids, which are centromeric plasmids containing various EF-2s(His-699 X)genes, were constructed from pEFT2 and pRS314. Plasmid pEFT2A was generated by oligonucleotide-mediatedmutagenesis of pEFT2 to create an SphI site and to eliminate a BstXI site
-
(CATTTTGGATGTTACTTTACATGCCGATGCTATC+CATTTTGGATGTTACTTTGCATGCCGATGCTATC (the SphI site is underlined),
AAGAAATTCGGTGTCGACAAGGEAAGATGATaACAGATT-AAGAAATTCGGTGTCGATAAGGCTAAGATGATGGACAGATT (the BstXI site is underlined)) (Fig. IA, a ) . To eliminate the multicloning site in pRS314, exceptthe EcoRI site, it was digested with KpnI and Sac1 and ligated with a double-stranded synthetic oligonucleotide ( 5 ' - m ~TTTGGGGTTTCTCGAGCT-3',3"CATGCTTAAGAAACCCCAAAGAGC-5' (the EcoRI site is underlined)) (Fig. l.4, b ) .ABstXI site (on the TRPl marker) of the resulting plasmid was eliminated by oligonucleotide-mediated mutagenesis to generate a plasmid pRS314B(GGC-
ATTGATATCGTWCTGCATSAGATG-GGCATTGATATCGTTCAACTGCATGGAGATG (the BstXI site is underlined)) (Fig. IA, 6 ) . A 3.8-kilobase fragment (EFT2),which was prepared by EcoRI digestion of pEFTPA, was inserted into the EcoRI site of pRS314B to construct PED-H (Fig. L4, c). To change the 699th codon of the EF-2 gene, PED-H was digested with BstXI and SphI, ligated with a mixture of synthetic double-stranded oligonucleotides (5'-CCGATGCTATC(G/A/l'/C)(G/A/T/
13497
13498
Temperature-sensitive EF-2 Mutants Lacking Diphthamide
CXAfI7GAG-3' 3'-GTACGGCTACGATAG(G/APT/C)(G/APT/C)(A/T)-5'), and introduced into competent Escherichia coli DH5a (Fig. lA, d). By sequencing plasmids in 30 independently isolated transformants, PED-C, -N, -T, -F, -Y, -E, -A, -G, -P, -S, -D, and -R and PED-ter were obtained (Table I). To generate other PED plasmids, PED-H was digested with BstXI and SphI and ligated with synthetic double-stranded oligonucleotidescontaining unique mutations (Fig. lA, d, Table I). Plasmids pRS316-EFT2 and pRS316-EFT2(G/R701)are centromeric plasmids containing the URA3 selectable marker and the wild-type and DT-resistant EF-2 gene, respectively, as described previously (19). Culture Media and Yeast Manipulation-Yeast strains were grown under nonselectiveconditions in YPD medium containing 1%yeast extract (Difco), 2% Bacto-peptone (Difco), and 2% dextrose and under selective conditionsin SD medium containing 0.67% Bacto-yeast nitrogen base without amino acids (Difco) and 2% dextrose. To make agar plates, 2% BiTek Agar (Difco) was added into liquid medium. For the selection of ura3 cells, 5-fluoroorotic acid (5-FOA) (PCRagar Inc.) plates (SD plates containing 0.1% 5-FOA and 20 pg/ml uracil) were used. Genetic manipulations of yeast cells were performed as described by Rose et al. (22). Yeast Strains-The diploid strain SDDl (MATalMATa uru3-52/ ura3-52 leu2-3,112/leu2-3,112 his3-A200Ihis3-A200 trpl-A90lltrplA901 EFTl leftl::LEU2 eft2::HZS3 IEFT2) described previously (19) was transformed with pRS316-EFT2. Oneof the meiotic segregants of this transformant was designated as KKUl (MATa ura3-52 leu2-3,112 his3-A200 trpl-AgOl eftl::LEU2 eft2::HIS3 [pRS316-EFT21). As shown SE2-GR, (MATa in Fig. U?, MKK strains were derived from -1. ura3-52 leu2-3,112 his3-A200trpl-A901eftl::LEU2 eft2::HZS3 [pRS314EFT2(G/R701)1)(Fig. U?),is aDT-resistant strain containing EF-2(Gly701 + Arg), as described previously (19). Zn Vitro ADP-ribosylation Assay-Incorporation of ADP-ribose into crude cell extracts by DT fragment A was measured as described previously (19). GrowthRates-Doubling times were determined from changes in optical density at 600 nm of cells growingin YPD or SD medium at the indicated temperature. Detection of EF-2 by Zmmunoblotting-Cell extracts were prepared as described by Jazwinski (23). Protein concentrations in the extracts were determined using a Bio-Rad assay kit (Bio-Rad). Then the cell extracts (15 pg of protein) were subjected to electrophoresis in a 7.0% SDS-polyacrylamide gel. Proteins were transferred to nitrocellulose, and then the blot was incubated with anti-EF-2 rabbit serum, kindly provided by Dr. A. Nairn (24, 25). Rabbit antibody binding t o the blot was detected with an ECL Western blotting detection kit (Amersham, United Kingdom).
Ed1
/
J.
Transformation with PED plasmids
RESULTS
Mutant EF-2s with Substitution of Lys or Arg for His-699 Znhibit Growth of Yeast Cells-A series of centromeric plasmids, designatedas PED plasmids, were constructed as shown in Fig. 1 A . These plasmids contain TRPl selectable markers and mutant EF-2s genes in which residue 699 has been substituted by all 19 other amino acid residues (Table I). Strain KKU1, whose chromosomal EF-2 genes (EFT1 and EFT2) have been disrupted, is rescued by expression of a wild-type EF-2 gene on plasmid pRS316-EFT2. PED plasmids were introduced into KKUl cells using a lithium acetate transformation technique, and transformants containing both plasmids were selected for TRPl and URA3 functions (Fig. lB). The doubling times of these transformants were measured SD in medium at 30.0 "C. SD medium was used in this experiment because it does not containuracil or tryptophan anddoes not allow growth of yeast cells that have lost pRS316-EFT2 (URA3 marker) or the introduced PED plasmid (TRPl marker). Most transformants grew at a rate identical to that of the KKUl strain carrying thecontrol vector plasmid pRS314 (Fig. lA). However, transformants harboring PED-R (expressing EF-2(His-699 + Arg)) or PED-K (expressing EF-2(His-699 Lys)) grew slower than the other transformants (Table I). Of the mutant EF-2s examined, only EF-2(His-699 + Arg) and EF-2(His-699 + Lys) prevented co-expressed wild-type EF-2 function in vivo. Mutant EF-2s with Substitutionsof 13 Amino Acids for His699 Are Active-To investigate whether mutant EF-2s were
-
n-selective growth for Ura+ Replica platlng for CFOA selection
MKK strain
/
PRS314-EFT2(G/R701)
FIG.1. Plasmids and yeast strains used in this study. A, construction of PED plasmids. Plasmids used in this study wereconstructed as described under "Materials and Methods." B , yeast strains. Strain KKU-1, containing disrupted EF-2 genes on chromosomesand a functional wild-type EF-2 geneon a plasmid with a URA3 marker, was transformed with a mutant EF-2 gene on a PED plasmid. The transformants were inoculated onto SD agar plates containing uracil for nonselection of the URA3 plasmid at 30 "C for 3 days. Then the plates were replicated onto 5-FOA plates on which cells that had lost the URA3 plasmid were selected (MKK cells). DT-resistant strain SE2-GR is also shown. kb, kilobase.
Temperature-sensitive EF-2 Mutants Lacking Diphthamide active, we tested the viabilities of cells harboring the mutant EF-2 genes only. Transformants containing both plasmids (pRS316-EFT2and one of the PED plasmids) were cultured on SD agar plates supplemented with 20 pg/ml uracil (Fig. 1B). Since these plates are not selective for URA3, cells mitotically segregated for pRS316-EFT2 were also able to grow if they were viable. Then cells on these plates were transferred to 5-FOA agar plates and incubated further at 30.0 "C. On the latter plates, Ura3- clones should appear as 5-FOA-resistant papillae (Fig. 1B). No papillae were formed on the plates by TABLE I Characteristics of cells containing both pRS316-EFT2 (wild-type EF-2) and a PED plasmid (mutant EF-2) Strain KKU-1, containing disrupted EF-2genes on chromosomes and a functional wild-type EF-2 gene on a plasmid with a URA3 marker, was transformed with the indicated PED plasmid. PED plasmid
@no acid resldue 699"
Doubling timeb
PED-H PED-C PED-L PED" PED-Q PED-N PED-I PED-T PED-V PED-W PED-F PED-Y PED-E PED-A PED-G PED-P PED-S PED-D PED-R PED-K PED-ter pRS314
His (CAOd Cys (TGT) Leu (CW) Met (ATG) Gln (CAA) Asn (AAT) Ile ( A n ) Thr (ACT) Val (GTA) "rp (TGG) Phe ( " T ) Tyr (TAT) Glu (GAA) Ala (GCT) Gly (GGA) Pro (CCT) Ser (AGT) Asp (GAT) Arg (AGA) Lys (AAA) Termination (TAA)
183 2 4 180 2 4 181-c 5 184 f 3 182 f 3 185 f 2 180 f 4 182 +_ 5 180 2 3 184 2 4 183 f 2 184 f 3 183 f 3 186 f 8 187 f 1 185 f 2 186 f 3 183 -c 4 229 f 10 351 f 49 185 6 184 3
rnin
-
Growth on 5-FOA plates'
+
+ + +
+ + + + + + + + + +
Amino acid residue 699 of EF-2 encoded by the PED plasmid and its nucleotide codon. * Doubling time of transformants in SD medium at 30.0 "C. The values are means S.E. of determinations of six (for PED-R and PED-K) or three (for other plasmids) independently isolated transformants. e Transformants were transferred t o 5-FOA plates and incubated at 30.0 "C for 3 days (cf.Fig. 1B). Wild-type EF-2. e pRS314 does not contain the EF-2 gene.
13499
cultures of KKUl transformed with PED-G, -P, -S, -D, -R, and -K containing the mutant EF-2 genes in which His-699 was replaced by Gly,Pro, Ser, Asp, Arg,and Lys, respectively (Table I). These results indicate that all these mutant EF-2s are inactive in uiuo. In contrast, many papillae were formed by cultures of KKUl transformants with PED plasmids containing the wild-type gene (on PED-H) or the mutant EF-2 genes (on PED-C,-L, -M, -Q, -N, -I, -T, -V, -W, -F, -Y, -E, and-A) (Table I). These results indicate that cells were viable when they expressed one of the mutant EF-2s only, in which His-699 was replaced by Cys, Leu, Met, Gln, Asn, Ile, Thr, Val, Trp, Phe, Tyr, Glu, or Ala. We isolated the viable cells from the papillae on the 5-FOA plates. These clones were named MKK strains (Fig. 1B) and were further characterized. These MKK strains mustcarry one of the mutant EF-2s only (Table 11) encoded on the PED plasmid. To confirm the absence of wild-type EF-2 in MKK cells, we assayed ADP-ribose incorporation byDT as described under "Materials and Methods." ADP-ribosewas incorporated into an extract of MKK-H cells containing the wild-type EF-2 gene but not into extracts of other MKK cells (Table 11). This result confirmed that loss of pRS316-EFT2 was as expected and that the mutant EF-2s did not contain a site for ADP-ribosylation. In addition, ADP-ribose was not incorporated into an extract of SE2-GR cells containing another type of DT-resistant EF-2 as described previously (Table11)(19). Cells withEF-2 Mutated at Diphthamide Show Temperaturesensitive Growth-MKK-H cells harboring wild-type EF-2 grew with a doubling time of about 100 min at 30.0 "C in YPD medium, which was less than thatof any MKK cells harboring the mutant EF-2 (Table 11). The superiority of the effect of wildtype EF-2 on cell growth increased at higher temperatures. On SD agar plates, MKK-H grewslightly more slowlyat 37.0,38.2, and 39.4 "C than at 30.0 "C, whereas other MKK cells grew faintly or didnotgrow at higher temperatures (Fig. 2 A ) . SE2-GR cellscontaining another type of DT-resistant EF-2 also showed growth retardation at temperaturesgreaterthan 37.0 "C, though they grew as fast as MKK-H cells at 30.0 or 37.0 "C (Fig. 2, Table 11).To elucidate the differences in growth rates, we measured the doubling times of MKK cells and SE2-GR cellsat 30.0 and 38.2 "C in YPD liquid medium (Table 11). Growth retardation of cells harboring mutant EF-2s was also observed in this liquid medium, though it was not so obvious as thaton agar plates. The ratio of the doubling times at
I1 TABLE Doubling times of strains with mutant EF-2 and their sensitivities to DT Strain
EF-2
Inco oration ofdLbose" pmol lmg protein
MKK-H MKK-C MKK-L MKK-M MKK-Q MKK-N MKK-I MKK-T MKK-V MKK-W MKK-F MKK-Y His-699 MKK-E MKK-A Gly-701SE2-GR
(A) 30.0 "C
(B) 38.2 "C ~~
rRNA) .",
rnin
160 f 8 1.60 100 f 1 116 2 3 232 f 25 2.00 116 2 2 2.07 240 f 10 113 f 1 226 f 11 2.17 115 f 2 245 5 8 2.13 120 f 4 317 2 29 2.64 311 f 18 2.33 133 2 8 129 2 8 298 2 14 2.31 2.12 274 f 24 129 f 2 124 f 3 251 2 20 2.02 297 f 9 2.15 138 2 1 143 f 1 308 f 16 2.15 152 2 6 303 2 14 2.00 468 f 51 2.87 163 2 4 +k g 101 f 3 197 f 2 1.95 'Cells were grown in YPD medium at 30.0 "C and harvested in theexponential growth phase. Incorporation of ADP-ribose into crude extracts of the harvested cells was assayed as described under "Materials and Methods." Doubling times in YPD medium at 30.0 and 38.2 "C were assayed as described under "Materials and Methods." Values are means 2 S.E. of determinations of three independently isolated clones.
'
type His-699 + Cys His-699 + Leu His-699 + Met His-699 + Gln His-699 + Asn His-699 + Ile His-699 + Thr His-699 + Val His-699 + Trp His-699 + Phe His-699 + Tyr Glu His-699 + Ala
Doubling timeb
-
103 2 5
