Full text of DOI:10.1007/s004380000337

Mol Gen Genet +2000) 264: 354±362
Digital Object Identi®er +DOI) 10.1007/s004380000337
ORIGINAL PAPER
P. Silar á M. Rossignol á R. Tahar
Z. Derhy á A. Mazabraud
Informational suppressor alleles of the eEF1A gene, fertility and cell
degeneration in Podospora anserina
Received: 3 February 2000 / Accepted: 7 July 2000 / Published online: 5 September 2000
Ó Springer-Verlag 2000
Abstract Mutations that increase readthrough at a UGA review). Like its bacterial counterpart EF-Tu, eEF1A
stop codon +informational suppressor mutations) were binds aminoacyl-tRNA and GTP to form a ternary
created in the gene +AS4) that encodes translation elon- complex that delivers the charged tRNA to the A site of
gation factor eEF1A in the ®lamentous fungus Podos- the ribosome, at the expense of the hydrolysis of one
pora anserina. The results strongly suggest that the net GTP molecule. Several data suggest that the bacterial
charge of the eEF1A protein controls the accuracy of and the eukaryotic proteins di€er with respect to how
translation. Physiological analysis of the mutant strains they perform this function +Negrutskii and Elskaya
shows that some of the alleles dominantly increase life 1998), and the precise biochemical events that occur
span, while only one drastically modi®es fertility. This during the elongation step catalysed by eF1A are still
exceptional allele +AS4-56) causes a wide array of phe- unknown. Like its bacterial counterpart, and together
notypes, including a new growthcessation phenomenon
withsome ribosomal proteins and RNAs, eEF1A con-
that is di€erent from Senescence or Crippled Growth, trols the accuracy of the mRNA decoding process
previously known degenerative syndromes that are both +Sandbaken and Culberston 1988). Thus, it has been
controlled by AS4. The data emphasise the fact that shown that it in¯uences the levels of readthrough and
eEF1A exerts a complex control over cellular physiology. frameshifting +Sandbaken and Culberston 1988; Din-
man and Kinzy 1997; Farabaughand Vimaladitahn
Key words Translation elongation á Translational
accuracy á Sexual reproduction á Ageing á Podospora
anserina
1998). The precise mechanism by which eEF1A controls
accuracy +especially frameshifting) is not yet fully
understood.
eEF1A is a highly conserved protein. However, in
several organisms, multiple genes encode di€erent forms
of eEF1A. These isoforms are di€erentially regulated
under various conditions or in diverse tissues. A further
level of complexity is added by the fact that eEF1A can
be modi®ed by phosphorylation, methylation or addi-
tion of glycerolphosphoryl-ethanolamine +Tuhackova
et al. 1985; Merrick 1992; Kielbassa et al. 1995).
Phosphorylation modi®es the activity of the factor
+Tuhackova et al. 1985), but no role has been detected
for the other modi®cations. It is suspected that these
isoforms and modi®cations allow optimisation of
translation and/or of other events involving eEF1A.
Indeed, in addition to its role during translation,
eEF1A exhibits numerous unrelated activities
+Negrutskii and Elskaya 1998, for a review), including
organisation of the cytoskeleton +see Durso and
Cyr 1994; Condeelis 1995, for review), activation of
phosphatidylinositol 4-kinase +Yang et al. 1993), and
activation of the degradation of some proteins by the
ubiquitin pathway +Gonen et al. 1994). All these proper-
ties suggest that eEF1A is likely to participate in the
Introduction
eEF1A is an essential G-protein that plays a major role
in the elongation phase of cytosolic protein synthesis in
all eukaryotic cells +Negrutskii and Elskaya 1998, for a
Communicated by C. A. M. J. J. van den Hondel
P. Silar +&)
Institut de Genetique et Microbiologie de l'Universite de Paris-Sud,
CNRS UMR 8621, Bat. 400,
91405 Orsay Cedex, France
E-mail: silar@igmors.u-psud.fr
Tel.: +33-1-69154658
Fax: +33-1-69157006
M. Rossignol á R. Tahar á A. Mazabraud
Centre de Genetique Moleculaire du CNRS,
91198 Gif-sur-Yvette Cedex, France
Z. Derhy
Institut d'Enzymologie du CNRS,
91198 Gif-sur-Yvette Cedex, France

355
co-ordination of translational activity withthe demands
of the cell, in response to various growth conditions.
emphasise the complexity of the role of eEF1A in cell
physiology. They also show that the net charge of the
In light of these characteristics, it is not surprising eEF1A protein controls the accuracy of translation.
that eEF1A is essential for cell viability +Cottrelle et al.
1985; Silar et al. 2000). In the mouse, a mutation in the
eEF1A2 isoform causes the ``wasted '' phenotype, char-
acterised by a set of alterations leading to degeneration
of the nervous and immune systems and death of the
organisms +Chambers et al. 1998). Interestingly, eEF1a
seems to be involved in numerous aspects of biology,
including di€erentiation and ageing. Indeed, mutations
Materials and methods
Strains, mutations, growthconditions and genetic analysis
The P. anserina strains used in this study were all derived from the
S strain +Rizet 1953), thus ensuring a homogenous genetic back-
ground. 193 and leu1-1 are UGA nonsense mutations that a€ect
a€ecting this factor may promote susceptibility to on- spore pigmentation and leucine biosynthesis, respectively. These
alleles are used to test the level of in vivo readthrough +Picard 1973;
Coppin-Raynal 1981), and hence serve to measure the translation
error rate +see below).
Standard culture conditions, as well as general genetic methods
for this fungus, have been described previously +Esser 1974).
cogenic transformation +Tatsuka et al. 1992) and a
mutated eEF1A gene is found in some cancerous cells
+Gopalkrishnan et al. 1999). During ageing, eEF1A ac-
tivity diminishes in many cell types, and this is correlated
witha decrease in the eciency of translation +Webster
1985; Cavallius et al. 1986). Following this observation,
Shepherd et al. +1989) overexpressed eEF1A in Droso-
phila melanogaster and observed an increase in longevity
in adult males. However, subsequent studies on Droso-
phila led to contradictory conclusions on this point
+Stearns and Kaiser 1993; Shikima et al. 1994). None-
theless, the decrease in eEF1A activity with ageing was
recently con®rmed in Drosophila +Shikima and Brack
1996).
DNA manipulations
DNA manipulations were performed as recommended by Ausubel
et al. +1987). The various antisuppressor AS4 alleles were cloned by
screening minibanks of genomic DNA from the corresponding
mutant strains. Mitochondrial DNA +mtDNA) was extracted by
the rapid method of Lecellier and Silar +1994). Modi®cations of the
mitochondrial genome in at least 9 senescent strains per genotype
were analyzed by restriction analysis with HaeIII, as described
+Jamet-Vierny et al. 1997).
The ®lamentous fungus Podospora anserina is a
model organisms that is extensively used in ageing
studies +Jamet-Vierny et al. 1999; Osiewacz and Kimple
1999). We have shown that mutations in AS4, the sole
gene that encodes eEF1A in P. anserina drastically in-
crease life span, diminishfertility and permit the prop-
Construction of strains carrying AS4 suppressor alleles
The AS4+ gene was mutagenised in vitro by conventional methods
+Ausubel et al. 1987) in order to change only one amino acid within
the protein. These alleles should thus be expressed from the AS4
promoter at the same level as the wild-type allele. Each mutant
agation of
a
non-conventional infectious element allele was then sequenced to ensure that only the desired mutation
had been introduced. This allowed the recovery of the mutant al-
leles AS4-55 to AS4-61 listed in Table 1. DNA fragments con-
responsible for a growthalteration called Crippled
Growth+Silar and Picard 1994; Silar et al. 1999). The
mutations located in the AS4 gene were all recovered
because they antagonise the deleterious e€ect on spor-
ulation of a UGA suppressor tRNA and were shown
globally to decrease accuracy since they also antagonise
the e€ect of an omnipotent informational suppressor
+antisuppressor mutations, Picard-Bennoun 1976).
Whereas increased translational accuracy promotes the
e€ects of the AS4 mutations on the development of
Crippled Growth+Silar et al. 1999), it is yet not clear
what causes fertility impairment and life span extension.
Here, we report the phenotype caused by new mu-
tations located in eEF1A that were obtained following
in vitro mutagenesis. These were based either on muta-
tions already known to act as suppressor mutations in
yeast +Sandbaken and Culberston 1988) or designed to
change the amino acid at positions modi®ed in the
previously obtained antisuppressor AS4 mutants. The
mutations cause a dominant increase in the incidence of
UGA readthrough +suppressor e€ect). Surprisingly,
some increase life span in a dominant fashion. However,
withthe exception of AS4-56, they do not signi®cantly
perturb sexual reproduction. Interestingly, this new al-
lele displays several unexpected properties, including the
triggering of a new type of growtharrest. These data
taining the various alleles were then cloned into pMOcosX +Orbach
1994) or pBC-Hygro +Silar 1995) and transformed into
a
Table 1 Amino acid replacements in eEF1A
Allele
Amino acid
change
Remarks
AS4-4
Lys42®Glu
His27®Arg
Arg68®His
Thr107®Ile
Gly51®Asp
Gly350®Asp
Ala433®Val
Thr431®Ala
Gly53®Asp
Glu287®Lys
Gly53®Lys
±
±
±
±
±
±
±
±
AS4-11
AS4-24
AS4-27
AS4-29
AS4-30
AS4-33
AS4-43
AS4-44
AS4-55
AS4-56
±
Equivalent to yeast TEF2-1
Same position as AS4-44 but
opposite charge
AS4-57
AS4-58
AS4-59
Thr143®Ile
Glu41®Lys
Glu163®Gln
Equivalent to yeast TEF2-7
Equivalent to yeast TEF2-3
Di€erence in the two mammalian
isoforms
AS4-60
AS4-61
Gly51®Lys
Gly350®Lys
Same position as AS4-29 but
opposite charge
Same position as AS4-30 but
opposite charge

356
P. anserina S mat+ strain. These two vectors carry a hygromycin examining total protein extracts from the various mutants and
resistance gene, which allows one to screen for transformants and from wild type fractionated on polyacrylamide gels, followed by
follow the segregation of the transgenes in crosses. For each allele, quanti®cation using the using Scion Image software +Silar et al.
six transformants were crossed witha leu1-1 DAS4 mat-/leu1-1 2000). The data +not shown) indicated that all mutant alleles are
AS4+ mat+ heterokaryotic strain +Silar et al. 2000). Among the expressed at the same level as the wild-type form.
progeny of these crosses, we could recover homokaryotic strains
withthe following genotypes: leu 1-1 {hygR, AS4)} DAS4 mat-,
{hygR, AS4)} DAS4 mat-, leu1-1 {hygR, AS4)} AS4+ mat+ and
Physiological analysis
{hygR, AS4)} AS4+ mat+, where {hygR, AS4)} designates the
Longevity was measured on M2 minimal medium at 27 °C in th e
dark as described +Belcour and Begel 1980). The longevity given for
mutant transgenic copy. Except in the case of AS4-59 and AS4-55,
two independent integrated copies derived from two di€erent pri-
mary transformants that were able to ensure cell viability were then
selected for further studies. The strains were then crossed with S
mat- in order to obtain {hygR, AS4)} AS4+ mat- strains. These
a
particular strain is expressed as the average length +in
cm‹standard deviation) reached by independent cultures before
the onset of senescence. At least nine cultures derived from at least
two di€erent ascospores of the appropriate genotype were analysed
for eachstrain +i.e., the values listed in Table 2 are derived from at
least 4 ´ 9=36 independent cultures). The fertility of a particular
AS4 mutant strains was measured after fertilization withthe con-
idia of a strain of the opposite mating type that carried the same
AS4 allele+s), as described +Coppin et al. 1993).
were crossed witha ehterokaryotic
leu1-1 DAS4 mat+/leu1-1
AS4+ mat- strain +Silar et al. 2000). Among the progeny of these
crosses, we recovered leu1-1 {hygR AS4)} DAS4 mat+, {hygR,
AS4)} DAS4 mat+, leu1-1 {hygR, AS4)} AS4+ mat-, and {hygR,
AS4)} AS4+ mat- strains. The crosses thus allowed us to obtain
strains that carry either the DAS4 allele at the endogenous locus
and a transgenic copy of the mutant allele or the AS4+ allele at the
endogenous locus and a transgenic copy of the mutant allele
+partial diploid heterozygous strains). All the desired strains were Suppression analysis
then subjected to phenotypic analysis.
For AS4-59, only one integrated copy was selected and analysed The suppression eciency of the new AS4 alleles was measured
as described, since this allele was found to display the same prop- according to the following strategies. +1) The 193 mutation is a
erties as wild type. In the case of AS4-55, none of the six integrated suppressible UGA nonsense mutation that a€ects ascospore colour
copies ®rst tested were able to promote viability when combined +Picard 1973). Wild-type ascospores are black and 193 ascospores
withthe deletion +i.e., ascospores withthe genotype
leu1-1 {hygR, are white. The readthrough level at the 193 mutation site can be
AS4)} DAS4 mat- or {hygR, AS4)} DAS4 mat- were not recovered), estimated in the progeny of a {hygR, AS4)} DAS4 ´ 193 {hygR,
suggesting that this allele was lethal. We thus crossed 41 additional AS4)} AS4+ cross based on the intensity of the green pigmentation
transformants withthe 193 tester strain. Among these crosses, two in 193 {hygR AS4)} DAS4 and 193 {hygR AS4)} AS4+ ascospores.
clearly yielded green spores. This was taken as an indication that The leu1-1 mutation is a UGA nonsense mutation that prevents
the two transformants carried an expressed copy of AS4-55. At- growthon M2 minimal medium and is suppressed by informational
tempts to recombine these copies with the deletion failed, indicating suppressors +Coppin-Raynal 1981). The leu1-1 {hygR, AS4)} DAS4
that AS4-55 is unable to restore viability. Thus only the partial or leu1-1 {hygR, AS4)} AS4+ strain can grow on minimal medium
diploid heterozygous strains that carry AS4+ and AS4-55 were at a speci®c rate. The ratio of the growth rate on M2 medium to that
analysed.
on M2 supplemented with50 mg/l leucine can thus be used to
Correct expression of the AS4-56, AS4-57, AS4-58, AS4-60 and quantify the level of readthrough level at the site of the leu1-1
AS4-61 mutant alleles was veri®ed as previously described by mutation.
Table 2 Characteristics of AS4 suppressor alleles
Genotypea
Mutation Phenotypic parameter
Suppressor activity
Morphology
Longevityb Fertilityc
CGd
193c
leu1-1c
Pm
resistanced
AS4+
±
9.5‹1.0 +++
16.5‹2.2 ++
15.7‹2.3 +++
9.9‹1.9 +++
9.6‹1.0 +++
9.4‹0.9 +++
10.8‹1.5 +++
No
No
Yes
No
No
No
No
No
ND
No
No
No
No
No
White
0
70%
55%
40%
70%
70%
70%
40%
70%
0%
70%
65%
70%
30%
70%
AS4+ AS4-55
AS4+ AS4-56
AS4+ AS4-57
AS4+ AS4-58
E287K
G53K
T143I
E41K
Slightly altered
WT
WT
WT
WT
WT
WT
Very altered
WT
WT
WT
WT
Light green
White to dark green
Almost white
Almost white
White
0
<0.05
0
0
0
0
0
ND
0
0
AS44+ AS4-59 E163Q
AS4+ AS4-60
AS4+ AS4-61
D AS4 AS4-56
D AS4 AS4-57
D AS4 AS4-58
D AS4 AS4-59
D AS4 AS4-60
D AS4 AS4-61
G51K
G350K
G53K
T143I
E41K
E163Q
G51K
G350K
Green
12.8‹1.7 +++
?
Almost white
White to dark green
Very light green
Very light green
White
g
±
10.1‹0.9 +++
9.8‹1.0 +++
10.0‹0.8 +++
11.1‹1.4 ++
13.1‹2.0 +++
0
0
0
Green
Very light green
WT
a The ®rst allele refers to the one present at the AS4 locus, the e Suppression at the 193 and leu1-1 sites was measured as de-
second one to the nature of the transgenic allele carried at an ec- scribed in Materials and methods
topic position
f Paromomycin resistance +Pm) is expressed as the percentage of
linear growthobserved on medium containing 500 mg/l of
paromomycin relative to that on unsupplemented medium after
3 days of culture
b Longevity is expressed in cm‹SD
c Fertility was estimated for self-crosses of the strain with the in-
dicated genotype
g See text for details
d Presence of Crippled Growth+CG) was checked as described by
Silar et al. +1999)
ND: not determined, see text

357
Paromomycin resistance
corresponding mutant allele +AS4-55) in P. anserina. A
similar situation was also reported for the TEF2-1 allele
in yeast +Dinman and Kinzy 1997). The analysis was
thus performed on heterozygous, partial diploid strains
carrying AS4-55 and AS4+. As described in Table 2,
these strains exhibit increased longevity and slightly re-
duced fertility. Clearly, the mutation promotes UGA
readthrough at the 193 mutation site but not at leu1-1.
Like the TEF2-1 mutation, it also promotes a marked
reduced sensitivity to paromomycin.
The paromomycin resistance level of a particular strain is the ratio
obtained by dividing the diameter of the thalli after 3 days of
growthon M medium supplemented with500 mg/l of paromo-
2
mycin by the diameter of the thalli after 3 days of growth on non-
supplemented M2 medium.
Results
We have previously shown that antisuppressor muta-
In yeast, TEF2-3 causes a dominant increase in
tions in the AS4 gene that encodes eEF1A in P. anserina frameshifting but has no e€ect on readthrough +Sand-
increase longevity, impair ascospore formation and baken and Culberston 1988; Dinman and Kinzy 1997).
permit the development of Crippled Growth +Silar and It does not promote any modi®cation of sensitivity to
Picard 1994; Silar et al. 1999). It was thus of interest to paromomycin +Sandbaken and Culberston 1988; Din-
test the e€ect on these phenotypes of suppressor muta- man and Kinzy 1997). The corresponding mutation in
tions also located in AS4. Unlike the antisuppressor P. anserina is called AS4-58. Homozygous or heterozy-
mutation, these would decrease translational accuracy. gous strains carrying this mutation do not present any
Mutations with this latter property have already been phenotypic modi®cation. They display a very small in-
described in Saccharomyces cerevisiae +Sandbaken and crease in UGA readthrough, which is more pronounced
Culberston 1988).
in the homozygous strain. Note that, unlike the yeast
Suppressor alleles that were designed to mimic the strain carrying TEF2-3, strains homozygous for AS4-58
S. cerevisiae ones could perhaps display slightly di€erent show a small decrease in paromomycin resistance. This
properties compared to the yeast alleles and may modify is clearly visible on medium containing 750 mg/l pa-
life span.
romomycin, since at this concentration the wild-type S
In order to perform site-directed mutagenesis on the strain grows at 10% of the rate on medium without
AS4 gene, three of the known suppressor alleles of paromomycin, whereas the AS4-58 homozygous strains
TEF2, one of the two genes that encode eEF1A in do not grow at all.
S. cerevisiae, were chosen, TEF2-1, TEF2-3 and TEF2-7.
The TEF2-7 mutation causes a dominant increase in
These alleles display various e€ects on frameshifting, frameshifting and UAG readthrough +Sandbaken and
readthrough and sensitivity to inhibitors of translation Culberston 1988; Dinman and Kinzy 1997). It is also
elongation +Sandbaken and Culberston 1988; Dinman associated withincreased resistance to paromomycin
and Kinzy 1997; Farabaughand Vimaladithan 1998). In
eachcase, the corresponding mutation was constructed
when homozygous +Dinman and Kinzy 1997). The cor-
responding mutation in P. anserina, AS4-57, does not
in vitro in AS4 +see Table 1) to yield the alleles AS4-55, promote any phenotypic alteration. As observed for
AS4-57 and AS4-58, respectively. These alleles were then AS4-58, it promotes, at the 193 locus only, a very small
transformed into the P. anserina S strain and two in- increase in readthrough, that becomes more pronounced
dependent transformants were selected for further when the mutation is homozygous. This mutation does
analysis +see Materials and methods), in order to take not modify the resistance/sensitivity of the strains to
position e€ects into account. Transformants were cros- paromomycin.
sed with the heterokaryotic strains carrying a deletion of
An additional allele was designed following a similar
AS4 +Silar et al. 2000), allowing the recovery of strains trans-species rationale. Comparison of the two mammal
carrying the ectopically inserted mutant alleles in asso- isoforms of eEF1A reveals several polymorphic posi-
ciation witheitehr
AS4+ or DAS4. Phenotypic and tions. Some of these could a€ect the degree of transla-
translational parameters were then measured +Table 2). tional ®delity +Silar 1994). We thus modi®ed AS4 at one
In no instance were di€erences detected between the of these positions; the glutamic acid at position 163 was
two integrated copies; similarly no di€erence due to the changed into a glutamine, to obtain AS4-59. Analysis of
mating type was seen. For eachallele, the data described
this mutant allele revealed that it does not promote any
below are thus a compilation of the results observed for modi®cation in the parameters of translation or any
the four di€erent combinations, e.g., the two integrated obvious phenotypic alteration.
copies associated with either of the mating types.
In yeast, TEF2-1 promotes a strong dominant in-
crease in frameshifting and moderate UAG/UAA Electrostatic interactions may be involved
readthrough; no e€ect is detected on UGA readthrough in the control of translational accuracy
+Sandbaken and Culberston 1988; Farabaughand
Vimaladithan 1998). The mutant allele also increases Because AS4-55 displays increased life span, we decided
misreading and causes a strong reduction in resistance to to test additional suppressor alleles to con®rm this
paromomycin +Sandbaken and Culberston 1988). We phenotype. However, we changed the rationale for the
could not obtain a homozygous strain carrying the design of the mutations. There is a striking correlation

358
between the net charge of eEF1A and translational ac-
Strikingly, the AS4-56 allele is associated withvery
curacy. In the yeast mutants, eight out of the nine yeast unusual phenotypes that are described in the next sec-
suppressor alleles change a negatively charged amino tion.
acid to a positively charged one +Sandbaken and Cul-
berston 1988). To con®rm this, the nine available
P. anserina antisuppressor mutant alleles were cloned AS4-56 strains display a variegated suppressive e€ect,
and sequenced +Table 1). Four out of the nine P. anse- a new growtharrest phenotype
rina antisuppressor alleles +AS4-4, AS4-29, AS4-30 and and show Crippled Growth
AS4-44) display the same correlation between charge
and translational ®delity +Table 1). To ascertain whether We observed that the AS4-56 allele promotes pro-
charge controls accuracy, we designed three more alleles nounced and dominant suppression of 193. Most sur-
that change the amino acid glycine to lysine at the three prisingly, suppression is highly variable, since ascospore
positions where it is changed to aspartic acid in AS4-29, colour ranges from white to dark green, most being
AS4-30 and AS4-44, yielding AS4-60, AS4-61 and AS4- green. This is not due to the presence of another, cryptic,
56, respectively +Table 1). These were then analysed as mutation but is likely to be caused by a novel epigenetic
described for the previous alleles +Table 2). In all three e€ect, akin to variegation. +1) No Mendelian segregation
cases, we were able to obtain the alleles associated with of the ascospore colour is observed. +2) 193 DAS4 AS4-
the AS4+ or the DAS4 allele, showing that these alleles 56 or 193 AS4+ AS4-56 mycelium derived from white
are viable. In all three cases, the mutations caused a ascospores yields ascospores withthe full spectrum of
suppressor phenotype, as measured by the level of colour when crossed with 193. This was observed in two
readthrough at the 193 locus +Table 2). Note that the consecutive generations. +3) The same phenomenon is
presence of a wild-type copy of AS4 in association with observed withthe ascospores derived from dark green
the mutant copy reduces the suppression eciency in the ascospores. +4) The positions of transgene integration is
three cases. This con®rms the role of electrostatic in- unlikely to play a role, since the two copies tested map at
teractions in the control of translational accuracy.
For AS4-60, suppression is moderate at the 193 locus
di€erent locations and both display the phenomenon.
Unlike the AS4 suppressor alleles previously exam-
but none is detected at the leu1-1 mutation site. ined, suppression by AS4-56 is not restricted to the 193
Paromomycin sensitivity is strongly increased. Longev- site, since leu1-1 is also suppressed in the AS4+ AS4-56
ity and mycelium morphology are the same as in wild background +we were not successful in constructing a
type. However, fertility is slightly diminished.
leu1-1 DAS4 strain, which prevents us from testing the
For AS4-61, suppression is barely detectable at the suppression eciency when AS4-56 is homozygous).
193 locus and none is detected at the leu1-1 locus. The mutation also promotes a dominant increase in
Paromomycin sensitivity is not a€ected. It is noteworthy sensitivity to paromomycin.
that the corresponding antisuppressor mutation +AS4-
AS4+ AS4-56 strains live longer than wild type but
30), from which AS4-61 was derived, surprisingly shows have wild-type morphology and fertility. In contrast,
increased sensitivity to this antibiotic +Coppin-Raynal DAS4 AS4-56 strains present a very altered mycelium
1981). Longevity is slightly increased ±to the same extent morphology, grow very slowly, and are unable to form
in the homozygous and heterozygous strains. Otherwise, perithecia +Fig. 1).
AS4-61 mutant strains are indistinguishable from wild
type.
DAS4 AS4-56 ascospores present an interesting pat-
tern of germination and growth. A few form a germi-
Fig. 1 A The three AS4 DAS4-
56 cultures depicted in the Fig-
ure were obtained from three
independent ascospores and
were inoculated onto M2 me-
dium on the same day. The
resulting cultures showed a
range of growthpotential de-
spite having the same genotype.
B A AS4 DAS4-56 culture was
inoculated on M2 medium and
incubated until it stopped
growing. Mycelium explants
from a wild-type S mycelium
were then added and the culture
was incubated for another
4 days. Explants placed near the
AS4 DAS4-56 culture showed
an inhibition of growth, where-
as those located farther away
did not

359
nation tube, but do not grow further. The others stop exit from stationary phase these strains exhibit the
growing at various times, generating a range of cultures Crippled Growthalteration. The alteration is not stable
withvarious diameters +see Fig. 1 for typical examples). and the cultures revert to Normal Growth after a few
Although reminiscent of the Senescence exhibited by centimetres of expansion. Because their growth is very
P. anserina, this phenomenon is not due to the same altered, we could not reliably test this phenotype in
cause. Indeed, several lines of evidence suggest that this is DAS4 AS4-56 strains. Note that none of the AS4 sup-
a novel growthtype of growtharrest. +1) Upon reinoc-
ulation, all the explants taken either from the centre or
from the edge of the arrested cultures regenerate a short
pressor mutations exhibits this e€ect.
culture. +2) This occurs with both short-grown and long- Discussion
grown cultures. +3) Because mtDNA is rearranged during
Senescence in P. anserina +Osiewacz and Kimple 1999 for The complex phenotypes associated
review), we puri®ed mtDNA from arrested AS4-56 cul- witheEF1A mutations in P. anserina
tures and subjected it to molecular analysis. The analysis
of arrested cultures showed no obvious mtDNA rear- Our previous analysis of eEF1A antisuppressor muta-
rangements +in contrast, mtDNA modi®cations classical tions showed that these types of mutations cause three
of Senescence are detected in senescent, heterozygous, main phenotypes: an increase in life span, impairment of
AS4+ AS4-56 cultures, as well as in all strains carrying sporulation and the potential for the Crippled Growth
any of the alleles AS4-55, AS4-57, AS4-58, AS4-59, AS4- phenomenon to develop +Silar and Picard 1994; Silar
60 or AS4-61). +4) Arrested cultures are not able to et al. 1999). Note that, in the original screen, numerous
transmit Senescence to young wild-type mycelia, under additional mutations were described as lethal +Picard-
conditions where wild-type senescent cultures do so.
Bennoun 1976). These mutations are recessive with re-
Growtharrest of the DAS4 AS4-56 strains could be spect to the three phenotypes +Silar and Picard 1994;
due to several factors; among these are exhaustion of Silar et al. 1999).
some nutrient in the growth medium or modi®cation of Here, we have described the e€ects of six suppressor
this medium in such a way that it can no longer sustain mutations located in AS4. These display a set of phe-
growth. In order to test the ®rst of these possibilities, we notypes that vary depending on the mutation considered.
First, at the translational level, all promote suppres-
increases the concentration of the nutrients up to 1.2
times that in the classical medium, while maintaining the sion of 193, although with di€erent eciencies. Indeed,
volume constant. This results in a uniform set of the only with three of them is moderate suppression of 193
cultures that display a short life span. Clearly, nutrient achieved. Note that these same mutations also trigger an
de®ciency is not the cause of the growth arrest and in- increase in sensitivity to paromomycin. AS4-56 deserves
creasing the concentration of nutrients actually de- special mention since +1) it presents a variegation for
creases the growth potential. Secondly, we increased the suppression of 193; +2) it is the only one of the six sup-
volume of the medium in the Petri dishes without in- pressors that is also able to suppress the leu1-1 mutation;
creasing the amounts of nutrients. This results in a set of and +3) the AS4+ AS4-56 strains present the Crippled
cultures witha long life span, as if dilution of some
Growth alteration. This latter phenotype is quite unex-
substance allows the hyphae to grow longer. Strikingly, pected for a suppressor mutation, because we have
we observe inhibition of the growth of explants of wild- presented convincing evidence +Silar et al. 1999) to show
type mycelium that are inoculated close to an arrested that this phenotype is caused by an increase in transla-
DAS4 AS4-56 mycelium +see Fig. 1 for a typical result). tional accuracy. We thus favour the idea that AS4-56
In contrast, explants located further away from such an may display very unusual properties withrespect to
arrested mycelium do grow. These data indicate that suppression, in the sense that it may decrease or increase
dying AS4 AS4-56 mycelia release some poisonous accuracy depending on the nucleotide context that sur-
substance that may cause their growth arrest.
rounds the decoded codon in the mRNA. Such opposite
Because the AS4-56 mutation causes a variegated e€ects on accuracy at di€erent codons have already been
suppression e€ect and the mutant strains show a range observed when the level of eEF1A is varied +Farabaugh
of life span, we investigated the relationship between and Vimaladithan 1998).
these two phenomena. The longevity of white or dark
Second, at the physiological level, four out of the six
green 193 AS4-56 ascospores was measured. No corre- suppressor mutations +AS4-57, AS4-58, AS4-60 and
lation was observed between the two phenotypes: both AS4-61) cause no or only very limited extension of life
white and dark green ascospores yielded strains with the span, and wild-type or near-wild-type fertility. AS4-55
full range of longevity. Because there is a strong e€ect of proved to be lethal, and to cause a signi®cant, dominant,
culture conditions +see above) on the growth potential, increase in life span when heterozygous. Strikingly,
the most likely explanation to account for the range of AS5-56 displays a very complex set of phenotypes that
life span is some variability in medium composition di€er depending on whether a wild-type copy of the gene
between the individual petri plates.
One ®nal striking phenotype caused by the AS4-56
is present or not.
When heterozygous, its main phenotype is increased
mutation is observed in the AS4+ AS4-56 strains. Upon life span. Note that the same phenotype is promoted by

360
AS4-55 and, to a lesser extent, by AS4-61. The increase mutation may indeed speci®cally perturb some but not
in longevity associated with AS4-56 and other suppres- all of these processes, leading to a complex e€ect on
sor alleles is dominant, unlike that promoted by the AS4 physiology. Note that the fact that increased longevity
antisuppressor alleles. We thus propose that increase in may be caused by di€erent mechanisms in the antisup-
longevity may be caused by di€erent mechanisms in the pressor and suppressor strains agrees well withour es-
suppressor and antisuppressor strains. This is supported timate of the number of genes involved in longevity
by the following observation. When the same amino acid control +Rossignol and Silar 1996), which suggests that
is changed to lysine in the suppressor or to aspartic acid many mechanisms are involved in de®ning longevity.
in the antisuppressor mutations the results on longevity
can be very di€erent; for example, AS4-29 doubles life
span +Silar and Picard 1994) whereas AS4-60 does not Role of electrostatic interactions
+Table 2). In the suppressor, it is unlikely that the error in de®ning translational accuracy
rate per se is involved because the level of suppression of
193 associated withthe di€erent alleles is not correlated
withextension of life span +see Table 2). We have ob-
We have detected an e€ect of the net charge of the
eEF1A protein on translational ®delity: the more nega-
tained data +Silar et al., unpublished results) that dem- tively charged the protein is, the more accurate is
onstrate that the error rate per se is also not involved in translation. It is noteworthy that the mutations con-
the life span extension associated with AS4 antisup- cerned are scattered all over the protein. This relation-
pressor alleles. This is consistent with previous ®ndings ship has been explored and con®rmed by changing three
that the error level is not involved in de®ning life span positions in the P. anserina eEF1A protein. The same
+Belcour et al. 1991; Silar et al. 1997).
charge relation has already been observed for the S7
When associated with DAS4, AS4-56 determines ribosomal protein of P. anserina +Dequard-Chablat
drastic alterations in growth, which stops at various 1986). This protein is encoded by the su12 gene in
times. This growth arrest syndrome is di€erent from the P. anserina +Silar et al. 1997). We propose that a
classical Senescence presented by P. anserina +Rizet mechanism that involves electrostatic interactions is
1953). Our data strongly suggest that the DAS4 AS4-56 implicated at the decoding center in checking the ®delity
strains release into the medium a toxic substance that of the translation process. One possibility is that elec-
may be responsible for cessation of their own growth. trostatic interactions participate in the binding of the
Several compounds are good candidates for this poison. eEF1A:GTP:aa-tRNA complex to the ribosome. Re-
Indeed, some P. anserina are known to produce anseri- jection of the incoming complex could be in part based
nones, which are antifungal, antibacterial and cytotoxic on repulsion of negative charges carried both by the ri-
compounds +Wang et al. 1997). Like its close relative bosome and the elongation complex. If these repulsive
Sordaria araneosa, P. anserina may also produce +a) forces prove to be stronger than the codon/anticodon
Sordarin-like toxin+s), which are speci®c antifungal interactions, the complex could be rejected. Under these
agents that act on the translational apparatus +Hauser conditions, increasing the negative charge of the ribo-
and Sigg 1971).
some or eEF1A would result in a higher rejection rate
AS4-56 thus displays very peculiar phenotype both at and hence in increased accuracy. By combining appro-
the translational and physiological levels. AS4-60, which priate pairs of AS4 and su12 alleles, we can now test this
involves the same amino acid change two amino acids proposal.
away +Gly51®Lys in AS4-60 and Gly53®Lys in AS4-
56), or AS4-44 +Gly53®Asp), from which AS4-56 was
derived, does not cause sucha drastic phenotype. Note
Relationship between translational accuracy
that position 53 is in a region of the eEF1A protein that and sensitivity to paromomycin
does not align well withthe EF-Tu eubacterial protein
+Cousineau et al. 1992). This region is very conserved in We introduced three mutations into the P. anserina
eukaryotes and may thus have evolved to ful®l a func- eEF1A gene by using data obtained withteh yeast
tion speci®c of the eukaryotic cell. At the present time it S. cerevisiae. While the overall e€ect of the mutations is
is premature to speculate upon the nature of this func- conserved between Podospora and yeast, we did detect
tion, but as stated in the Introduction many eEF1A some di€erences in the sensitivity to paromomycin.
functions are potential candidates.
Paromomycin is known to bind to the 16S rRNA in
Thus, overall, our data show that modi®cations in E. coli +Blanchard et al. 1998). However, its potential
eEF1A may perturb many cellular processes, resulting in interactions withelongation factor eEF1A are unknown.
a large array of phenotypes. Two hypotheses may be Classically, increased paromomycin sensitivity is asso-
proposed. First, all the e€ects could be due to speci®c ciated withan increase in translation error rate +Palmer
e€ect on translation of some key proteins. The di€erent and Wilhelm 1978; Singh et al. 1979). In P. anserina, it
mutations in eEF1A may modify the translation of these has been known for a long time that this relationship is
proteins, which would result in the observed phenotype. not absolute +Coppin-Raynal 1981). Recent data ob-
Alternatively, perturbation of some of the several known tained withthe TEF2-7 allele in yeast also show that
functions of eEF1A may cause the phenotypes. One decreased accuracy may be associated withincreased

361
Dinman JD, Kinzy TG +1997) Translation misreading: Mutations
in translation elongation factor 1a di€erentially a€ect
programmed ribosomal frameshifting and drug sensitivity.
RNA 3: 870±881
Durso NA, Cyr RJ +1994) Beyond translation: elongation factor 1a
and the cytoskeleton. Protoplasma 180: 99±105
Esser K +1974) Podospora anserina. In: King RC +ed) Handbook of
genetics. Plenum, New York, pp 531±551
Farabaugh PJ, Vimaladithan A +12998) E€ect of frameshift-
inducing mutant of elongation factor 1a on programmed +1
frameshifting in yeast. RNA 4: 38±46
Gonen H, Smith CE, Siegel NR, Kahana C, Merrick WC, Cha-
kraburtty K, Schwartz AL, Ciechanover A +1994) Protein
synthesis elongation factor EF-1alpha is essential for ubiquitin-
dependent degradation of certain Nalpha-acetylated proteins
and may be substituted for by the bacterial elongation factor
Ef-Tu. Proc Natl Acad Sci USA 91: 7648±7652
Gopalkrishnan RV, Su ZZ, Goldstein NI, Fisher PB +1999)
Translational in®delity and human cancer: role of the PTI-1
oncogene. Int J Biochem Cell Biol 31: 151±162
paromomycin resistance +Dinman and Kinzy 1997). In
P. anserina, the corresponding mutation +AS4-57) is not
associated withany modi®cation of paromomycin sen-
sitivity. However, there is a situation in P. anserina that
is similar to that seen with TEF2-7, since the AS4-30
alleles decrease both the error rate and the paromomycin
resistance +Coppin-Raynal 1981). The TEF2-7 and AS4-
30 mutations are located in di€erent regions of the
protein. It is thus likely that complex interactions are
involved between the ribosome, the elongation complex
and paromomycin that de®ne translational accuracy and
paromomycin resistance. Our data suggest that the de-
tails of these interactions may not be conserved during
evolution.
Acknowledgments We thank Corinne Jamet-Vierny and Herve
Lalucque for useful discussion. We thank Marie-Christine Scherr-
man for a tentative identi®cation of the poisonous compound. This
work was supported by Grant No. 5388 from the Association pour
la Recherche sur le Cancer +ARC). The work was done in com-
pliance withthe current laws governing genetic experimentation in
France. P. Silar is a professor at the University of Paris VII, Denis
Diderot.
Hauser D, Sigg HP +1971) Isolation and decomposition of sorda-
rin. Helv Chim Acta 54: 1178±1190
Jamet-Vierny C, Boulay J, Begel O, Silar P +1997) Contribution of
various classes of defective mitochonrial DNA molecules to
senescence in Podospora anserina. Curr Genet 31: 171±178
Jamet-Vierny C, Rossignol M, Haedens V, Silar P +1999) What
triggers Senescence in Podospora anserina? Fung Genet Biol 27:
26±35
Kielbassa K, Muller HJ, Meyer HE, Marks F, Gschwendt M
+1995) Protein kinase C delta-speci®c phosphorylation of the
elongation factor eEF-1 alpha and an eEF-1 alpha peptide at
threonine 431. J Biol Chem 270: 6156±6162
References
Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Lecellier G, Silar P +1994) Rapid methods for nucleic acids
SmithJA, Strulh K +1987) Current protocols in molecular
extraction from Petri dish-grown mycelia. Curr Genet 25:
biology. Wiley Interscience, New York
122±123
Belcour L, Begel O +1980) Life-span and senescence in Podospora Merrick WC +1992) Mechanism and regulation of eukaryotic
anserina: e€ect of mitochondrial genes and functions. J Gen
Microbiol 119: 505±515
Belcour L, Begel O, Picard M +1991) A site-speci®c deletion in the
mitochondrial DNA of Podospora is under the control of nu-
clear genes. Proc Natl Acad Sci USA 88: 3579±3583
protein synthesis. Microbiol Rev 56: 291±315
Negrutskii BS, Elskaya AV +1998) Eukaryotic translation elonga-
tion factor 1a: structure, expression, functions and possible role
in aminoacyl-tRNA channeling. Prog Nucleic Acid Res Mol
Biol 60: 47±78
Blanchard SC, Fourmy D, Eason RG, Puglisi JD +1998) rRNA OrbachMJ +1994) A cosmid witha HygR marker for fungal library
chemical groups required for aminoglycoside binding.
Biochemistry 37: 7716±24
Cavallius J, Rattan SIS, Clark BFC +1986) Changes in activity and
construction and screening. Gene 150: 159±162
Osiewacz HG, Kimple E +1999) Mitochondrial-nuclear interactions
and lifespan control in fungi. Exp Gerontol 34: 901±909
amount of active elongation factor 1a in aging and immortal Palmer E, Wilhelm JM +1978) Mistranslation in a eukaryotic
human ®broblast cultures. Exp Gerontol 21: 149±157
organism. Cell 13: 329±334
Chambers DM, Peters J, Abbott CM +1998) The lethal mutation of Picard M +1973) Genetic evidences for a polycistronic unit of
the mouse wasted +wst) is a deletion that abolishes expression of
a tissue-speci®c isoform of translation elongation factor 1a,
encoded by the Eef1a2 gene. Proc Natl Acad Sci USA 95:
4463±4468
transcription in the complex locus ``14'' in Podospora anserina.
II Genetic analysis of informational suppressors. Genet Res
Camb 21: 1±15
Picard-Bennoun M +1976) Genetic evidence for ribosomal anti-
suppressors in Podospora anserina. Mol Gen Genet 147: 299±
306
Condeelis J +1995) Elongation factor 1a, translation and the cyto-
skeleton. Trends Biochem Sci 20: 169±170
Coppin E, Arnaise S, Contamine C, Picard M +1993) Deletion of Rizet G +1953) Sur l'impossibilite d'obtenir la multiplication
the mating type sequences in Podospora anserina abolishes
mating without a€ecting vegetative functions and sexual
di€erentiation. Mol Gen Genet 241: 409±414
Á
vegetative ininterrompue et illimitee de l'ascomycete Podospora
anserina. CR Acad Sci Paris 237: 838±840
Rossignol M, Silar P +1996) Genes that control longevity in
Coppin-Raynal E +1981) Ribosomal suppressors and antisuppres-
Podospora anserina. MechAgeing Dev 90: 183±193
sors in Podospora anserina: altered susceptibility to paromo- Sandbaken MG, Culberston MR +1988) Mutations in elongation
mycin and relationships between genetic and phenotypic
suppression. Biochem Genet 19: 729±740
Cottrelle P, Cool M, Thuriaux P, Price VL, Thiele D, Buhler JM,
factor 1 alpha a€ect the frequency of frameshifting and amino
acid misincorporation in Saccharomyces cerevisiae. Genetics
120: 923±934
Fromageot P +1985) Either one of the two yeast EF-1a genes is Shepherd JC, Walldorf U, Hug P, Gehring WG +1989) Fruit ¯ies
required for cell viability. Curr Genet 9: 693±697
withadditional expression of the elongation factor EF-1 a live
Cousineau B, Cerpa C, Lefebvre J, Cedergren R +1992) The
longer. Proc Natl Acad sci USA 86: 7520±7524
sequence of the gene encoding elongation factor Tu from Shikima N, Brack C +1996) Changes in the expression of genes
Chlamydia trachomatis compared with those of other organ-
involved in protein synthesis during Drosophila ageing.
isms. Gene 120: 33±41
Gerontology 42: 123±136
Dequard-Chablat M +1986) Di€erent alterations of the ribosomal Shikima N, Ackerman R, Brack C +1994) Protein synthesis elon-
protein S7 lead to opposite e€ects on translational ®delity in the
fungus Podospora anserina. J Biol Chem 261: 4117±4121
gation factor EF-1a expression and longevity in Drosophila
melanogaster. Proc Natl Acad Sci USA 91: 4199±4203

362
Silar P +1994) Is translational accuracy an out-dated topic? Trends Stearns SC, Kaiser M +1993) The e€ect of enhanced expression of
Genet 10: 71±72
Silar P +1995) Two new easy to use vectors for transformations.
Fungal Genet Newslett 42: 73
Silar P, Picard M +1994) Increased longevity of the EF-1 alpha high-
®delity mutants in Podospora anserina. J Mol Biol 235: 231±236
elongation factor EF-1a on lifespan in Drosophila melanogaster.
IV. A summary of three experiments. Genetica 91: 167±182
Tatsuka M, Mitui H, Wada M, Nagata H, Nojima H, Okayama H
+1992) Elongation factor 1 alpha gene determines susceptibility
to transformation. Nature 359: 333±336
Silar P, Koll F, Rossignol M +1997) Cytosolic ribosomal mutations Tuhackova Z, Ullrichova J, Hradec J +1985) Regulation of the
that abolish accumulation of circular intron in the mitochon-
dria without preventing senescence of Podospora anserina.
Genetics 145: 697±705
activity of eukaryotic peptide elongation factor 1 by autocata-
lytic phosphorylation. Eur J Biochem 146: 161±166
Wang H, Gloer WH, Gloer JB, Scott JA, MallochD +1997)
Silar P, Haedens V, Rossignol M, Lalucque H +1999) Propagation
of a novel cytoplasmic, infectious and deleterious determinant is
controlled by translational accuracy in Podospora anserina.
Genetics 151: 87±95
Silar P, Rossignol M, Haedens V, Derhy Z, Mazabraud A +2000)
Deletion and dosage modulation of the eEF1A gene in Podos-
pora anserina: e€ect on the life cycle. Biogerontology 1: 47±54
SinghA, Ursic D, Davies J +1979) Phenotypic suppression and
misreading in Saccharomyces cerevisiae. Nature 277: 146±148
Anserinones A and B: new antifungal and antibacterial
benzoquinones from the coprophilous fungus Podospora anse-
rina. J Nat Prod 60: 629±631
Webster GC +1985) Protein synthesis and aging organisms. In:
Sohal RS +ed) Molecular biology of aging: gene stability and
gene expression. Raven Press, New York, pp 263±289
Yang W, Burkhart W, Cavallius J, Merrick WC, Boss WF +1993)
Puri®cation and characterization of a phosphatidylinositol 4-
kinase activator in carrot cells. J Biol Chem 268: 392±398