S1 whereas the colonies in which the pYES2.1 plasmid was replaced by the HIS3-plasmid pCBG1202 containing the gene coding for wild-type GB2-cells or mutant but functional forms of eEF2 w
Trang 1elongation factor 2 function in yeast
Galyna Bartish1,2, Hossein Moradi1,2and Odd Nyga˚rd1
1 School of Life Sciences, So¨derto¨rns ho¨gskola, Huddinge, Sweden
2 Department of Cell Biology, Arrhenius Laboratories, Stockholm University, Sweden
Protein synthesis is one of the most complicated and
energy consuming cellular processes Approximately
150 different proteins are required to facilitate the
vari-ous processes involved in the translation process [1]
Elongation factor 2 (eEF2) is one of the key
partici-pants in the protein synthesis elongation cycle eEF2 is
a 95 kDa GTP-binding protein that binds to
pretrans-location ribosomes [2] The role of the factor, and its
eubacterial homologue, elongation factor G (EFG), is
to promote GTP-dependent translocation of the
ribo-some along the mRNA under simultaneous transfer of
peptidyl-tRNA and deacylated tRNA to the ribosomal P- and E-sites, respectively This process is presumed
to involve conformational changes in the ribosome as well as in the factor itself [2–4]
Yeast eEF2 is a protein of 842 amino acids [5] The protein is evolutionary conserved and the amino acid sequence is 66% identical and 85% homologous to the sequence of human eEF2 [5] eEF2 is an essential protein coded for by two genes, EFT1 and EFT2 [5] The cellular level of eEF2 is strictly regulated [6] and cell viability requires that at least one of the two genes is functional
Keywords
elongation factor 2; functional
complementation; osmostress;
phosphorylation; yeast
Correspondence
O Nyga˚rd, School of Life Sciences,
So¨derto¨rns ho¨gskola, S-141 89 Huddinge,
Sweden
Fax: +46 8608 4510
Tel: +46 8608 4701
E-mail: odd.nygard@sh.se
(Received 10 January 2007, revised 27 June
2007, accepted 17 August 2007)
doi:10.1111/j.1742-4658.2007.06054.x
Yeast elongation factor 2 is an essential protein that contains two highly conserved threonine residues, T56 and T58, that could potentially be phos-phorylated by the Rck2 kinase in response to environmental stress The importance of residues T56 and T58 for elongation factor 2 function in yeast was studied using site directed mutagenesis and functional comple-mentation Mutations T56D, T56G, T56K, T56N and T56V resulted in nonfunctional elongation factor 2 whereas mutated factor carrying point mutations T56M, T56C, T56S, T58S and T58V was functional Expression
of mutants T56C, T56S and T58S was associated with reduced growth rate The double mutants T56M⁄ T58W and T56M ⁄ T58V were also functional but the latter mutant caused increased cell death and considerably reduced growth rate The results suggest that the physiological role of T56 and T58
as phosphorylation targets is of little importance in yeast under standard growth conditions Yeast cells expressing mutants T56C and T56S were less able to cope with environmental stress induced by increased growth tem-peratures Similarly, cells expressing mutants T56M and T56M⁄ T58W were less capable of adapting to increased osmolarity whereas cells expressing mutant T58V behaved normally All mutants tested were retained their ability to bind to ribosomes in vivo However, mutants T56D, T56G and T56K were under-represented on the ribosome, suggesting that these non-functional forms of elongation factor 2 were less capable of competing with wild-type elongation factor 2 in ribosome binding The presence of non-functional but ribosome binding forms of elongation factor 2 did not affect the growth rate of yeast cells also expressing wild-type elongation factor 2
Abbreviations
CaMPKIII, Ca2+and calmodulin-dependent protein kinase III; eEF2, eukaryotic elongation factor 2; EFG, elongation factor G; MAP, mitogen-activated protein; SC, synthetic complete; 5-FOA, 5-fluoroortic acid.
Trang 2eEF2 is subjected to post-translational
modifica-tions The C-terminal part of the protein contains a
histidine residue (H699 in yeast) that is converted to
diphthamide, a unique amino acid only found in eEF2
[7] The N-terminal part of eEF2 contains two highly
conserved threonine residues (T56 and T58 in yeast)
that can be phosphorylated The primary
phosphoryla-tion target is T56 but phosphorylaphosphoryla-tion at the second
threonine has also been observed [8,9]
Phosphoryla-tion decreases the affinity of eEF2 for pretranslocaPhosphoryla-tion
ribosomes, thereby preventing the factor from
stimu-lating translocation [10–12] The observation that
threonines T56 and T58 are highly conserved in eEF2
[5,13] has led to the suggestion that threonine
phos-phorylation may play a general role in regulating the
activity of eEF2 in eukaryotes
In mammals, an altered phosphorylation status of
eEF2 has been connected to different physiological
sit-uations and severe diseases [14] Mammalian eEF2 is
phosphorylated by a specific Ca2+ and
calmodulin-dependent protein kinase (CaMPKIII) [15,16] The
activity of the eEF2 kinase is regulated by the
mito-gen-activated protein (MAP) kinase and
mTOR-signal-ling pathways [17] These signalmTOR-signal-ling pathways activate
the eEF2 kinase in response to mitogens and other
stimuli that increase the cellular energy demand [18–
21]
Unicellular eukaryotes such as yeast appear to lack
CaMPKIII [22] However, yeast eEF2 can serve as
substrate for mammalian CaMPKIII [23] Donovan
and Bodley [23] noted that yeast eEF2 was
phosphory-lated in vivo by an endogenous kinase present in the
yeast cells Furthermore, peptide mapping suggested
that both phosphorylation by the endogenous and the
mammalian kinases occurred at the same site in yeast
eEF2 [23] The endogenous yeast kinase was identified
by Teige et al [24] as the Rck2 kinase, a Ser⁄ Thr
pro-tein kinase homologous to mammalian
calmodulin-dependent kinases Like the mammalian eEF2 kinase,
Rck2 activity is regulated via phosphorylation
Activa-tion of the Rck2 kinase is mediated by the MAP
kinase Hog1 in response to osmostress [24], an
envi-ronmental stress condition known to reduce the rate of
protein synthesis in fission yeast [25]
Site directed mutagenesis has frequently been used
to analyse the function of specific amino acids in
bac-terial EFG [26–29] To date, there are only a few
reports in which this technique has been used to
acquire information on the importance of specific
amino acids and amino acid motifs for eEF2 function
[6,13,30,31] In the present study, we have used site
directed mutagenesis to analyse the importance of
threonines T56 and T58 for cell viability in yeast
Results
Yeast eEF2 has two putative phosphorylation sites, threonines T56 and T58 We have used site directed mutagenesis to analyse the role of these two amino acids for viability of yeast cells A total of 13 eEF2 mutants were created All except three contained single amino acid substitutions The constructs were inserted
in the expression vector pCBG1202 (Table 1) under the control of the GAL1 promoter The expression plasmid contains a 3¢-located sequence coding for an inframe V5 epitope that could be used for immunode-tection of the plasmid-encoded protein All constructs were sequenced to confirm the presence of the intro-duced mutations and to assure that the correct reading frame was maintained
To ascertain that the cloned constructs were expressed, cells from the haploid yeast strain YOR133w were transformed with the expression vector pCBG1202 containing the various constructs YOR133w cells retain one of the two EFT genes normally coding for the essential protein eEF2 Viability of the cells was therefore independent of the functional properties of the plasmid-encoded eEF2 Control cells were trans-formed with the identical plasmid containing the sequence coding for V5-tagged wild-type eEF2 (GA2 cells Table 1)
As eEF2 exert its function on the ribosome, func-tional complementation studies require that the tag attached to the C-terminus of the plasmid-encoded eEF2 do not interfere with the ribosomal binding properties of the factor As shown in Fig 1A, the tagged wild-type protein was able to bind to somes Thus, the C-terminal tag did not prevent ribo-somal binding Furthermore, all mutant forms of eEF2 used in the present study were also capable of binding
to the ribosome (Fig 1A)
A closer examination of the total expression levels
of the mutant forms of eEF2 suggests that all mutants were expressed to the same level as tagged wild-type eEF2 with two exceptions (Fig 1B) The detectable levels of the double mutant T56V⁄ T58V and the single mutant T56D was 75% and 50% of the wild-type lev-els, respectively Because all constructs are identical, except for the introduced point mutations, transcrip-tion levels should be equal It is therefore possible that the lower intracellular levels of these mutant forms of eEF2 reflect increased degradation The expression lev-els of tagged wild-type eEF2 from plasmid pCBG1202
in GB2 cells (Table 1) was used as a reference for max-imum expression levels and ribosomal binding of tagged eEF2 analysed in the absence of competing eEF2 coded for by the yeast genome As shown in
Trang 3Fig 1B, the expression level was almost twice that
observed in cells also expressing genomic eEF2 The
amount of plasmid-encoded eEF2 bound to ribosomes
was also approximately double that seen in GA2 (Fig 1B) Most of the mutant forms of eEF2 were able
to bind as efficient to ribosomes as wild-type eEF2
Table 1 Strains and plasmids used in the present study Euroscarf (Frankfurt, Germany).
One Shot TOP10 cells (F- mcrA D(mrr-hsdRMS-mcrBC) /80lacZDM15 DlacX74
recA1 araD139 D(araleu) 7697 galU galK rpsL (StrR) endA1 nupG)
Invitrogen DB3.1(F–gyrA462 endA1 D(sr1-recA) mcrB mrr hsdS20(r B -, m B -) supE44 ara-14
galK2 lacY1 proA2 rpsL20(Sm R ) xyl-5 Dleu mtl1)
Invitrogen
Fig 1 Galactose induced expression levels and ribosome association of plasmid-encoded mutant and wild-type eEF2 Plasmid pCBG1202 containing mutant forms of eEF2 was inserted into Yor133w cells GA2 and GB2 cells expressing tagged wild-type eEF2 from the same plasmid was used as control (Table 1) Expression of the plasmid-encoded eEF2 was induced by incubating the transformed cells at 30 C in the presence of galactose The induced cells were harvested and an aliquot of the total cell lysate was withdrawn before isolation of ribo-somes The presence of plasmid-encoded eEF2 on isolated ribosomes was analysed by SDS gel electrophoresis and immunoblotting (A) Total expression and ribosome association of plasmid-encoded eEF2 was analysed by immunoblotting using a dot-blot technique The dot blots were quantified using computer-assisted densitometry.
Trang 4The exceptions were mutants T56D, T56G and T56K.
These mutant forms of eEF2 were under-represented
on the ribosome even after compensation for variation
in total cellular levels of plasmid-encoded factor,
sug-gesting that the mutation may have interfered with the
ribosome-binding properties of eEF2 The low
expres-sion level of the double mutant T56V⁄ T58V was not
manifested in lower levels of ribosome-bound eEF2
(Fig 1B) Instead, this mutant appears to bind well to
ribosomes This is in agreement with the lack of effect
on ribosome binding seen with the single mutants
T56V and T58V
The ability of the eEF2 mutants to functionally
complement wild-type eEF2 was analysed by
trans-forming GB1 cells (Table 1) with expression vector
pCBG1202 coding for mutant forms of eEF2 The
GB1 strain lacks both genomic genes normally coding
for eEF2 These cells are viable due to the presence of
an URA3-plasmid, pYES2.1, containing the gene
cod-ing for wild-type eEF2 (Table 1) The transformed
GB1 cells were allowed to grow on the appropriate
selective medium Colonies from each transformation
were isolated and plated onto solid media containing
5-fluoroortic acid (5-FOA) for counter selection As
shown in Fig 2, seven of the mutant eEF2 constructs
were able to support cell viability
One colony from each functional construct was
fur-ther characterized by growth on selective media The
original GB1 strain was only able to grow on plates
containing histidine (supplementary Fig S1) whereas
the colonies in which the pYES2.1 plasmid was
replaced by the HIS3-plasmid pCBG1202 containing
the gene coding for wild-type (GB2-cells) or mutant
but functional forms of eEF2 were only able to grow
in the presence of uracil (supplementary Fig S1)
Sequencing of plasmid pCBG1202 confirmed that the
surviving eEF2 constructs contained the amino acid
substitutions originally inserted in the eEF2 sequence
by PCR The results from the functional
complementa-tion assay show that the threonine at posicomplementa-tion 56 could
be replaced by cystein, methionine and serine (Fig 2)
Mutants containing asparagine, aspartic acid, glycine,
lysine or valine were nonfunctional (Fig 2) Clones
expressing eEF2 in which the adjacent threonine T58
was replaced by amino acids serine or valine were
via-ble (Fig 2)
One possibility was that threonine T56 could be
replaced by an amino acid that could not serve as
phosphate acceptor (i.e cystein or methionine) as long
as the second putative phosphorylation site T58 was
left intact To investigate this possibility, we
con-structed double mutants in which both threonines were
replaced by amino acids that could not be
phosphory-lated As shown in Fig 2, eEF2 containing the double mutants T56M⁄ T58V and T56M ⁄ T58W could replace wild-type eEF2 in yeast whereas the construct T56V⁄ T58V was nonfunctional
During the experiment, we noted that some of the clones expressing functionally active but mutant forms
of eEF2 appeared to grow slower than yeast expressing wild-type eEF2 from an otherwise identical plasmid The data presented in Fig 3A,D show that the doubling time for yeast cells expressing mutant T56M⁄ T58V was increased by approximately 75% compared to that of yeast cells expressing the tagged wild-type protein For mutants T56C, T56S and T58S, the doubling time was increased by 15–25% whereas yeast cells expressing the double mutant T56M⁄ T58W grew slightly faster than the control cells at 30C (Fig 3D) Expression of the double mutant T56M⁄ T58V resulted in a marked reduction in the number of viable cells whereas mutants T56C, T56S and T58S only caused a slight increase in cell death (Table 2)
Fig 2 Ability of mutant forms of eEF2 to functionally complement yeast cells lacking genomic copies of the eEF2 genes Yeast GB1 cells were transformed with plasmid pCBG1202 carrying wild-type (wt, positive control) or mutant forms of the eEF2 gene Cells transformed with empty plasmid pCBG1202 were used as negative control The transformed cells were grown in SC ⁄ Gal-Ura-Leu-His medium until the D 600 nm reached approximately 1 Aliquots (5 lL)
of the cell cultures (undiluted and diluted 1 : 5) were spotted onto
SC ⁄ Gal-Ura-Leu-His plates (left panel) or onto SC ⁄ Gal-Leu-His plates containing 5-FOA (right panel) The plates were incubated for
4 days at 30 C.
Trang 5A comparison of the growth rate at various
temper-atures using solid medium showed that all mutants
except T58V grew slightly slower than the control at
25C (Fig 3F) Cells expressing the T56M ⁄ T58V mutant failed to grow at this low temperature At
37C, cells expressing mutants T56M, T58S, T58V
Fig 3 Growth rate of yeast cells expressing plasmid born functional and nonfunctional eEF2 mutants Growth rate of yeast cells (GB) expressing wild-type or mutant functional forms of eEF2 from plasmid pCBG1202 under normal growth conditions (A,D) and under mild os-mostress (C,D) Growth rate of yeast cells (YOR133w) expressing wild-type or mutant nonfunctional forms of eEF2 from plasmid pCBG1202 (B,E) Overnight cultures were diluted to approximately D 600 nm ¼ 0.2 with SC ⁄ Gal-His medium (B,E) or with SC ⁄ Gal-His-Leu medium with-out (A,F) or with 0.4 M NaCl (C) The cells were allowed to grow at 30 C under vigorous shaking The attenuance of the yeast cultures was measured at 600 nm at the intervals indicated (A–C) and the growth rates calculated (D,E) Temperature-dependent growth of yeast cell expressing mutant but functional eEF2 (F) Cells from overnight cultures were used for serial dilution (1 : 10) in SC-His-Leu medium Aliquots (5 lL) were spotted onto solid SC-His-Leu growth medium The plates were incubated for 3 days at the temperatures indicated.
Trang 6and T56M⁄ T58W grew at nearly the same rate as cells
expressing wild-type eEF2 whereas the growth rate of
cells expressing mutants T56C, T56S and T56M⁄ T58V
was more severely affected at 37C than at 30 C
(Fig 3F) For mutants T56C and T56M⁄ T58V, the
reduced growth was partly accounted for by a marked
increase in the proportion of nonviable cells (not
shown)
The nonfunctional eEF2 constructs were able to
bind to the ribosome in the presence of wild-type eEF2
coded for by the remaining eEF2 gene in the yeast
strain YOR133w (Fig 1) It was therefore possible
that these mutants could interfere with the function of
wild-type eEF2 thereby reducing the rate of protein
synthesis and the growth rate of the transformed cells
As shown in Fig 3B, the doubling rate was only
mar-ginally affected by the presence of a nonfunctional
eEF2 Thus, there was no negative effect on the
growth rate caused by the presence of the
nonfunc-tional ribosome-binding forms of eEF2
Phosphorylation of eEF2 in yeast cells can be
trig-gered by osmostress [24] To investigate how yeast cells
expressing eEF2 lacking the putative phosphorylation
targets T56 and⁄ or T58 responded to osmostress,
con-trol cells and cells containing the mutants T56M,
T58V and T56M⁄ T58W were grown in the presence of
0.4 m NaCl As shown in Fig 4C,D, mild osmostress
had a slight negative effect on the growth rate of GB2
cells A limited effect was also seen on the growth rate
of cells expressing the mutant T58V, suggesting that
the mutation had little or no effect on the response to
increased osmolarity By contrast, yeast cells
express-ing mutants T56M and T56M⁄ T58W responded by a
reduction in the growth rate by approximately 35%
and 45%, respectively (Fig 3C,D) The fraction of
dead cells was increased to approximately the same
extent in cells expressing mutant eEF2 compared to
that in cells expressing plasmid-encoded wild-type eEF2 (not shown)
Discussion
The ability of eEF2 to promote translocation in mam-mals is regulated by phosphorylation at T56 and⁄ or T58 [8–10,12,32,33] Phosphorylation is catalysed
by an eEF2-specific Ca2+ and calmodulin-dependent kinase Threonines Thr56 and Thr58 are highly con-served in eEF2 from several different organisms (Fig 4) [5,13] Phosphorylation at the homologous threonines has therefore been assumed to play a gen-eral role in the regulation of the rate of elongation in eukaryotes Yeast cells contain a Ser⁄ Thr protein kinase called Rck2 that shows homology to the mam-malian CaM-dependent eEF2-kinase [24] The Rck2 kinase phosphorylates eEF2 in vitro The activity of the kinase is increased under environmental stress con-ditions such as increased osmolarity [24] The activa-tion is associated with elevated intracellular levels of phospho-eEF2 and reduced protein synthesis [24,25] The actual phosphorylation site(s) on eEF2 have not been identified but previous studies suggest that the target amino acids are identical to those phosphory-lated by mammalian CaMPKIII in vitro (i.e T56 and⁄ or T58) [23]
Functional complementation under standard growth conditions and under environmental stress
Our analysis of the role of the two threonines for eEF2 function in yeast cells showed that the threonine
at position 56 could be replaced with serine as well as with cystein and methionine Cells expressing mutants T56C and T56S grew markedly slower than cells expressing tagged wild-type eEF2 while mutant T56M had less effect on the growth rate The decreased growth observed with T56C and T56S was partly accounted for by a slight increase in cell mortality Amino acid substitutions were also allowed at posi-tion 58 Cells expressing mutants T58S and T58V grew slower than control cells expressing the tagged wild-type eEF2 and showed a slight decrease in the number
of viable cells Thus, T58 could be replaced by valine whereas the T56V mutation resulted in a nonfunctional eEF2 Consequently, double mutant T56V⁄ T58V was also nonfunctional The observation that the func-tional properties of double mutant T56M⁄ T58V was severely impaired was surprising because both mutants had little effect on eEF2 function, when occurring as single mutants Expression of the mutant had negative
Table 2 Determination of the fraction of viable cells expressing
functional but mutant forms of eEF2 Doubling time calculated after
compensation for variations in the percentage of viable cells The
originally observed doubling time was taken from Fig 3D.
Estimated doubling time (min)
Trang 7effects on the proportion of viable cells and on the
doubling time Double mutant T56M⁄ T58W was fully
functional and the growth rate of cells expressing this
variant of eEF2 was almost indistinguishable from that
of cells expressing the control factor In this mutant,
the sequence TDT was replaced by the homologous
motif MDW found in EFG from Escherichia coli
Interestingly none of the organisms in the alignment
shown in Fig 5 have tryptophan at the position
corre-sponding to T58 in yeast The results from the
muta-tion experiments suggest that threonines T56 and T58
are not essential for the viability of yeast cells under
normal growth conditions Both threonines could be
replaced by amino acids that cannot be
phosphory-lated by serine⁄ threonine protein kinases such as the
Rck2 kinase [24] Phosphorylation of T56 and⁄ or T58
therefore appears not to play an essential role in
regu-lating the rate of protein synthesis in yeast under
stan-dard growth conditions
Under conditions of increased osmolarity, the yeast
cells rapidly reduce protein production [25] In
Saccha-romyces cerevisiae, the HOG MAP kinase pathway is
activated under condition of increased extra cellular
osmolarity [34] Activation of Hog1 is essential for
sur-vival of yeast cells at high osmolarity Hog1 activates
Rck2, which in turn phosphorylates eEF2 [24] Mild
osmostress reduced the growth rate of yeast cells expressing plasmid-encoded wild-type eEF2 (Fig 4)
By contrast to what might have been expected, replace-ment of the two threonies with amino acids that could not serve as phosphorylation targets did not prevent the osmostress-dependent reduction in growth rate Cells expressing the T58V mutant behaved similar to cells expressing wild-type eEF2 whereas the growth rate of cells expressing mutants T56M and T56M⁄ T58W was even more reduced than that observed in the presence of wild-type eEF2 The effect
on the growth rate observed with the double mutant was probably caused by the amino acid replacement at position 56 because the effect on the growth rate was similar to that observed with the single mutation T56M The data suggest that phosphorylation at posi-tion 56 and⁄ or 58 is not critical for the cellular response to increased extra cellular osmolarity
Stress induced by increasing the growth temperature (37C) resulted in reduced growth rates for cells expressing wild-type as well as mutant forms of eEF2
Fig 4 Comparison of the amino acid context surrounding the
puta-tive phosphorylation site in eEF2 from various fungi, plants and
metazoans The position of threonines T56 and T58 (yeast
number-ing) are indicated by arrows Amino acid sequences from
(acces-sion numbers in parenthesis) Saccharomyces cerevisiae
(NP_014776), Saccharomyces castellii (AAO32487), Saccharomyces
kluyveri (AAO32562), Glugea plecoglossi (BAA11470), Ashbya
gos-sypii (AAS53513), Candida albicans (CAA70857),
Schizosaccharomy-ces pombe (CAB58373), Neurospora crassa (AAK49353), Gibberella
zeae (XP_389750), Aspergillus nidulans (XP_663934), Aspergillus
fu-migatus (XP_755686), Cryptococcus neoformans (AAW43242),
Ent-amoeba histolytica (BAA04800), Trypanosoma cruzi (BAA09433),
Dictyostelium discoideum (EAL63212), Cyanidioschyzon merolae
(BAC67668), Guillardia theta (AAK39722), Parachlorella kessleri
(P28996), Chlorella pyrcnoidosa (BAE48222), Beta vulgaris
(CAB09900), Arabidopsis thaliana (AAF02837), Oryza sativa
(NP_001052057), Blastocystis hominis (BAA11469),
Cryptosporidi-um parvCryptosporidi-um (AAC46607), PlasmodiCryptosporidi-um falciparCryptosporidi-um (BAA97565),
Tetrahymena thermophila (AAN04122), Drosophila pseudoobscura
(EAL32818), Drosophila melanogaster (P13060), Aedes aegypti
(AAK01430), Spodoptera exigua (AAL83698), Caenorhabditis
ele-gans (AAD03339), Rattus norvegicus (NP_058941), Mus musculus
(NP_031933), Cricetulus griseus (AAB60497), Pongo pygmaeus
(CAH90954), Homo sapiens (AAH06547), Gallus gallus
(NP_990699), Xenopus laevis (AAH44327), Xenopus tropicalis
(NP_001015785), Danio rerio (AAH45488), Monosiga brevicollis
(AAK27414), Pichia pastoralis (AAO39212) The arrows indicate the
position of T56 and T58.
Trang 8However, cells expressing mutants T56C, T56S and
T56M⁄ T58V were clearly more affected than cells
expressing the other functional mutants The increased
growth temperature also influenced the number of
via-ble cells For most mutants, the effect was similar to
that seen in cells expressing plasmid-encoded wild-type
eEF2 The exception was cells expressing the double
mutant T56M⁄ T58V These cells showed markedly
increased cell death The observation that mutants
T56M, T58V and T56M⁄ T58W did not alter the
abil-ity of the yeast cells to respond to temperature stress
suggests that the ability to phosphorylate eEF2 at T56
and⁄ or T58 is not crucial for regulating translation in
response to temperature stress
One possibility is that environmental stress induces
phosphorylation at an alternative site in eEF2 A
recent large-scale characterization of nuclear
phospho-proteins in HeLa cells showed the presence of eEF2
phosphorylated at Ser485 (yeast numbering) located in
the so-called hinge region of the factor [35] The
pres-ence of serine-phosphorylated eEF2 has also been
demonstrated in vitro upon activation of a yeast kinase
homologous to the type II Ca2+ and
calmodulin-dependent kinases [36]
It has been speculated that the general role of eEF2
phosphorylation may not be a massive shut-down of
protein synthesis but rather a mechanism to promote
translation of specific mRNAs that have difficulties in
competing with more translation efficient mRNAs by
slowing down the elongation rate [19,37] The situation
would be analogous with that observed on translation
after administration of limited concentrations of
cyclo-heximide [38–42] Our results cannot rule out the
pos-sibility that a limited reduction of the elongation rate
through phosphorylation at T56 is necessary to
pro-mote translation of mRNAs needed under specific
stress situations
An alignment of the amino acid sequences from a
variety of eukaryotic organisms showed that Thr56
often is replaced by methionine in fungal eEF2
whereas Thr58 is much more conserved (Fig 4) It
should be noted that none of the listed eEF2 sequences
have amino acids S or V in position 58 and none of
the sequences have C or S in position 56 The latter
could be explained by the slower growth rate of yeast
cells expressing eEF2 carrying these mutations The
slower growth of the T58S mutant could also be an
evolutionary disadvantage and hence explain the lack
of serine at position 58, even if the resulting protein is
functional However, the absence of valine at
posi-tion 58 is notable because replacement of T58 with
valine had limited effect on eEF2 function as
deter-mined by the effect of the mutation on the growth rate
under both normal growth conditions and conditions
of increased environmental stress
The T56M mutation had little if any effect on the growth rate of yeast cells under standard laboratory growth conditions However, yeast cells expressing the T56M mutant (or the double mutant T56M⁄ T58W) have considerable difficulties in coping with environ-mental stress situations as demonstrated by the effect
of increased osmolarity Thus, the better ability to adapt to environmental stress may have constituted a strong evolutionary pressure in favour of threonine at position 56 in eEF2
Properties of the eEF2 mutants eEF2 is a GTP-binding protein that interacts with pre-translocation ribosomes and promotes ribosomal trans-location along the mRNA under GTP-hydrolysis [2] All mutant forms of eEF2 described here (i.e even the mutants that were unable to functionally complement wild-type eEF2) were able to bind to ribosomes in cells also expressing wild-type eEF2 from one of the remaining eEF2 coding genes Because binding of eEF2 to the ribosome is dependent on the preforma-tion of a guanosine nucleotide-factor complex [43], the observation suggests that the mutant forms of the fac-tor were also able to interact with guanosine nucleo-tides Three of the nonfunctional mutants, T56D, T56G and T56K, were under-represented on the ribo-some even after adjusting for variations in the intra-cellular concentrations of plasmid-encoded eEF2, indicating that these mutations had a negative effect
on the ability of the factor to bind to ribosomes The other nonfunctional mutants had approximately the same ability to bind to ribosomes as the plasmid-encoded wild-type eEF2, suggesting that the mutations interfere with the ability of the factor to sustain elon-gation rather than with the ability to associate with ribosomes
The function of eEF2 in translocation requires recip-rocation between two conformational states associated with the phosphorylation status of the bound guanosine nucleotide [2] The two putative phosphorylation sites are located in the so-called switch I region (also known
as the effector-domain) [44,45], a flexible region known
to be involved in the dynamic properties of elongation factors [46] Due to its flexible nature, the peptide sequence containing threonines T56 and T58 is missing
in the crystal structure of yeast eEF2 [3] It is therefore difficult to estimate the structural effects caused by the introduced point mutations The observed phenotypic effects of the analysed eEF2 mutants, an inability to functionally complement wild-type eEF2 and the
Trang 9reduced growth rates obtained after expression of
func-tional eEF2 mutants may be related to a loss of the
dynamic properties of the factor Such a loss could lead
to a reduced ability of the factor to participate in the
elongation cycle, resulting in a reduced growth rate
It has previously been shown that the level of eEF2
in yeast cells is tightly regulated [6] The regulation
involves a post-transcriptional mechanism that keeps
the cellular level of eEF2 constant Thus,
over-expres-sion of mutated eEF2 in cells also expressing plasmid
encoded wild-type eEF2 result in decreased levels of
the wild-type protein as the proportion of mutated
fac-tor increases As a consequence, nonfunctional
mutants of eEF2 (e.g point mutations at V488 and
H699) cause a dominant negative phenotype when
expressed in cells also expressing wild-type eEF2 [6]
The nonfunctional eEF2 mutants described in the
pres-ent study were capable of binding to the ribosome and
could therefore have been expected to interfere with
the function of wild-type eEF2 even if expression of
these mutants would have no effect on the intracellular
level of wild-type eEF2 However, the nonfunctional
mutants did not interfere with the growth rate of yeast
cells also expressing wild-type eEF2 Only the
expres-sion of mutant T56V had a slight effect on the growth
rate Thus, no dominant negative effect of the
non-functional mutants could be observed In the present
study, the nonfunctional mutants were expressed in
yeast cells retaining one of the two genes normally
coding for eEF2 in wild-type cells It is possible that
these cells have a sub-optimal content of wild-type
eEF2 If this is the case, the presence of nonfunctional
eEF2 in the ribosomal fraction without noticeable
effects on the growth rate may reflect an increased
population of ‘hungry’ pretranslocation ribosomes
waiting to interact with a functional eEF2
Experimental procedures
Chemicals
BP clonase enzyme mix, LR clonase enzymes mix, Reading
frame cassette C, DNA polymerase (Klenow fragment) and
anti-V5-HRP serum were obtained from Invitrogen
(Carls-bad, CA, USA) Restriction nucleases AatII, ClaI, BamHI
and XhoI were obtained from Roche (Mannheim, Germany)
Alkaline phosphatase, Ready-To-Go ligation kit, ECL
wes-tern blotting detection kit was obtained from Amersham
Pharmacia Biotech Inc (Uppsala, Sweden) 5-FOA was
pro-vided by Larodan Fine Chemicals (Malmo, Sweden)
Ampi-cillin, kanamycin, chloramphenicol and synthetic dropout
medium supplement lacking histidine, leucine, tryptophan
and uracil were obtained from Sigma (St Louis, MO, USA)
Taq DNA polymerase, Pfu DNA polymerase, RNasine were obtained from SDS Promega (Madison, WI, USA) Yeast nitrogen base without amino acids and agar were from BD (Franklin Lakes, NJ, USA) Ammonium sulphate and amino acids were from Merck (Darmstadt, Germany)
Strains, plasmids and primers The strains and plasmids used are listed in Table 1 Plasmid pFA6a-HIS3MX6 was kindly provided by C Sjo¨gren (Department of Cell and Molecular Biology, Karolinska Institutet, Stockholm, Sweden) All primers were synthe-sized by CyberGene AB (Huddinge, Sweden) The primers used are listed in supplementary Tables S1–S2
Growth media Escherichia colicells were grown in LB containing the proper antibiotics Yeast strains were grown on synthetic complete medium, SC, containing 0.67% (weight by volume) bacto-yeast nitrogen base without amino acids, 0.14% (w⁄ v) yeast synthetic drop-out medium without histidine, leucine, trypto-phan and uracil, 0.5% (w⁄ v) ammonium sulphate) The medium was supplemented with uracil (20 lgÆmL)1) and the appropriate amino acids: histidine (20 lgÆmL)1) and leucine (60 lgÆmL)1) as indicated Galactose (2% weight by volume) was added as carbon source unless noted
For counter selection, we used SC-Leu-His media supple-mented with 5-FOA (1 gÆL)1) and uracil (50 lgÆmL)1) Solid growth media contained 2% (w⁄ v) agar
Construction of a conditional null strain For cloning of the yeast eEF2 gene, total yeast DNA was prepared form strain YDR385w as described by Hoffman and Winston [47] The gene for eEF2 was amplified by PCR using primers eEF2F and eEF2R (supplementary Table S1) The 2.5 kb PCR-product was introduced into the TOPO vector pYES2.1 and the resulting plasmid was transformed into strain YOR133w carrying only one of the two genomic alleles for eEF2 The transformed cells were plated onto SC-Ura and a positive colony (YOR133w; pYES2.1⁄ URA3 ⁄ eEF2) was selected This strain is referred
to as GA1 (Table 1)
For deletion of the remaining genomic copy of the eEF2 gene, the LEU2 gene was amplified from plasmid pAT3 using primers Leu2F and Leu2R (supplementary Table S1) These two primers contained 20 nucleotides that matched the 5¢- and 3¢-sequence of LEU2, and 40 nucleotides with a sequence identical to the 5¢- and 3¢-sequences flanking the genomic eEF-2 in strain YOR133w The purified PCR frag-ment was introduced into the GA1 cells and the genomic eEF2 coding sequence replaced by the LEU2 gene via homologous recombination The transformed cells were
Trang 10plated on SC-Ura-Leu for selection of positive colonies.
After 4 days at 30C, positive colonies were selected and
the replacement of the genomic copy of eEF2 by LEU2 was
confirmed by PCR and sequencing The resulting yeast
strain is referred to as GB1 (Table 1)
Construction of plasmid for counter selection
Vector pYES3⁄ CT was digested with restriction enzymes
AatII and ClaI to remove the TRP1 gene The digestion
products were separated by agarose gel-electrophoresis and
the vector without TRP1 was isolated The HIS3 gene was
amplified from plasmid pFA6a-HIS3MX6 using primers
HisF, HisR, HisFL and HisRL (supplementary Table S1)
The latter primer set was used to create overhangs, which
were complementary to the overhangs generated after AatII
and ClaI digestion of the vector The PCR products
obtained using the two sets of primers were pooled, heated
to 95C for 10 min and allowed to gradually anneal by
stepwise lowering of the temperature The annealing
mix-ture was ligated into the digested pYES3⁄ CT plasmid The
ligated plasmid was transformed into TOP10 cells The
presence of the HIS3 gene in the plasmid was confirmed by
PCR analysis
The new vector, pYES3⁄ CT ⁄ HIS3, was used for
construc-tion of a destinaconstruc-tion vector suitable for use in the Gateway
technology cloning system [48–51] For this purpose, the
vec-tor was digested with restriction enzymes BamHI and XhoI
and treated with the Klenow polymerase fragment followed
by treatment with alkaline phosphatase Reading frame
cas-sette C.1 was ligated with the digested vector and the
result-ing plasmid was transformed into DB3.1a E coli cells, which
were plated onto LB plates containing chloramphenicol
Positive colonies were selected and the presence of the
read-ing frame cassette in the new destination plasmid referred to
as pCBG1202 was confirmed by restriction analysis
Site directed mutagenesis
All mutants were generated using the mega-primer method
described by Brons-Poulsen et al [52] Primer GateEF2F
was used in combination with one of the reverse primers
carrying the point mutation to produce a short PCR
frag-ment (supplefrag-mentary Table S2) This fragfrag-ment was used as
a mega-primer together with primer GateEF2R
(supple-mentary Table S2) for amplifying the full-length gene The
PCR products were inserted into the donor vector
pDONR221 using BP clonase The presence of the
muta-tion was confirmed by sequence analysis Mutant eEF2
genes were transferred to the vector pCBG1202 by
recombi-nation using LR clonase The destirecombi-nation vector was
trans-formed into strain YOR133w for confirming gene
expression, and into strain GB1 for functional analysis by
plasmid shuffling A copy of the wild-type eEF2 gene
obtained by PCR amplification using primers GateEF2F
and GateEF2R was cloned into the pCBG1202 vector as described above This plasmid served as control
Cell transformation Bacterial transformations were performed according to standard methods [53] Yeast cells were transformed using the lithium acetate method, as described by Soni et al [54]
Detection of eEF-2 expression by immunoblotting Yeast strain YOR133w containing plasmid pCBG1202 with
a wild-type or mutated eEF2 gene was grown overnight at
30C in 5 mL of SC-His medium containing 2% (w ⁄ v) glu-cose The cells were collected by centrifugation, washed and resuspended in 30 mL of SC-His medium with galactose After induction during approximately 20 h at 30C, the cells were harvested, washed in 20 mm Hepes-KOH (pH 7.4), 2 mm Mg(CH3COO)2, 100 mm KCl and 1 mm dithiothreitol, and suspended in the same buffer containing
1 mm PMSF and 4000 U RNasine The cell suspension was mixed with glass beads and the yeast cells lysed as described [55] The crushed cells were centrifuged for 5 min
at 5000 g with a Haereus Biofuge (Berlin, Germany) An aliquot of the supernatants were withdrawn for analysis of the total level plasmid-encoded eEF2 The remaining super-natants were transferred to new tubes and centrifuged for another 15 min at 15000 g The supernatants were used for preparation of ribosomes Deoxycholate and Triton X-100 were added at a final concentration of 1% (w⁄ v) each The supernatants (1 mL), were layered onto 2 mL sucrose cush-ions containing 0.75 m sucrose in 75 mm KCl, 20 mm Tris⁄ HCl, pH 7.6, 2 mm Mg(CH3COO)2and 15 mm 2-mer-captoethanol The material was centrifuged in a TLA100.3 rotor (Beckman Instruments, Palo Alto, CA, USA) for 150 min, at 198 000 g and 4C The ribosomal pellets were dissolved in 0.25 m sucrose, 25 mm KCl,
30 mm Hepes-KOH (pH 7.6), 2 mm Mg(CH3COO)2 and
1 mm dithiothreitol Dissolved ribosomes and the post-ribo-somal supernatants were stored in aliquots at )80 C until used
For detection of total cellular levels of plasmid-encoded eEF2 crude cell lysates, 40 lg protein in 2 lL, were spot-ted on nitrocellulose membranes For estimation of the ribosomal binding capacity of plasmid-encoded eEF2 iso-lated ribosomes, 40 lg ribosomes in 2 lL, were spotted on nitrocellulose membranes The dried membranes were for immunoblotting The ribosome bound eEF2, 50 lg of ribosomes, was also analysed by SDS gel electrophoresis
on 10% (w⁄ v) polyacrylamid gels [56] The separated pro-teins were transferred to a nitrocellulose membrane, and the membrane was incubated with anti-V5-HRP serum Bound antibodies were detected using the ECL western blotting detection kit