Cysteinyl-tRNA synthetase is not essential for viability of the archaeon Methanococcus maripaludis

Abstract

The methanogenic archaea Methanocaldococcus jannaschii and Methanothermobacter thermautotrophicus contain a dual-specificity prolyl-tRNA synthetase (ProCysRS) that accurately forms both prolyl-tRNA (Pro-tRNA) and cysteinyl-tRNA (Cys-tRNA) suitable for in vivo translation. This intriguing enzyme may even perform its dual role in organisms that possess a canonical single-specificity cysteinyl-tRNA synthetase (CysRS), raising the question as to whether this latter aminoacyl-tRNA synthetase is indeed required for cell viability. To test the postulate that all synthetase genes are essential, we disrupted the cysS gene (encoding CysRS) of Methanococcus maripaludis. The knockout strain was viable under normal growth conditions. Biochemical analysis showed that the pure M. maripaludis ProCysRS was capable of forming Cys-tRNA, implying that the dual-specificity enzyme compensates in vivo for the loss of CysRS. The canonical CysRS has a higher affinity for cysteine than ProCysRS, a reason why M. maripaludis may have acquired cysS by a late lateral gene transfer. These data challenge the notion that all twenty aminoacyl-tRNA synthetases are essential for the viability of a cell.

Full text

Cysteinyl-tRNA synthetase is not essential for viability
of the archaeon
Methanococcus maripaludis
Constantinos Stathopoulos*, Wonduck Kim
†
, Tong Li*, Iain Anderson
†
, Britta Deutsch
†
, Sotiria Palioura*,
William Whitman
†
, and Dieter So
¨
ll*
‡§
Departments of *Molecular Biophysics and Biochemistry, and
‡
Chemistry, Yale University, New Haven, CT 06520-8114; and
†
Department of Microbiology,
University of Georgia, Athens, GA 30602-2605
Contributed by Dieter So¨ ll, October 11, 2001
The methanogenic archaea Methanocaldococcus jannaschii and
Methanothermobacter thermautotrophicus contain a dual-speci-
ficity prolyl-tRNA synthetase (ProCysRS) that accurately forms both
prolyl-tRNA (Pro-tRNA) and cysteinyl-tRNA (Cys-tRNA) suitable for
in vivo translation. This intriguing enzyme may even perform its
dual role in organisms that possess a canonical single-specificity
cysteinyl-tRNA synthetase (CysRS), raising the question as to
whether this latter aminoacyl-tRNA synthetase is indeed required
for cell viability. To test the postulate that all synthetase genes are
essential, we disrupted the cysS gene (encoding CysRS) of Meth-
anococcus maripaludis. The knockout strain was viable under
normal growth conditions. Biochemical analysis showed that the
pure M. maripaludis ProCysRS was capable of forming Cys-tRNA,
implying that the dual-specificity enzyme compensates in vivo for
the loss of CysRS. The canonical CysRS has a higher affinity for
cysteine than ProCysRS, a reason why M. maripaludis may have
acquired cysS by a late lateral gene transfer. These data challenge
the notion that all twenty aminoacyl-tRNA synthetases are essen-
tial for the viability of a cell.
P
rior to the availability of genomic sequences, aminoacyl-
tRNA synthetases (AARSs) were believed to consist of a
family of twenty highly conserved enzymes found in all organ-
isms (1). They were divided into two classes (I and II) of ten
members each, based on the presence of mutually exclusive
amino acid sequence motifs that reflected structurally distinct
topologies of the active site. These conserved features allow the
facile recognition of AARS genes by sequence similarity
searches of the known organismal genomes. In addition, the two
classes differ in the way they bind their substrates. Whereas the
class I enzymes approach the minor groove side of the tRNA’s
acceptor helix, the class II enzymes bind to the major groove side
(1). Based on these idiosyncrasies, an aminoacyl-tRNA syn-
thetase of particular substrate specificity was always believed to
belong to the same class regardless of its biological source,
reflecting the ancient origin of this enzyme family. The process
of amino acid attachment to tRNA is further refined in some
synthetases by editing mechanisms that enhance amino acid
selection and contribute to the overall quality control during
protein synthesis (2). Because faithful translation is indispens-
able for viability of organisms, all twenty members of the AARS
family were thought to be essential.
Research in the last few years revealed that, whereas trans-
lation certainly requires a full complement of AA-tRNAs, their
synthesis is not always catalyzed by the complete set of twenty
canonical AARS enzymes (3). The most frequent exception is
the formation of Asn-tRNA and Gln-tRNA; their synthesis is
accomplished in most bacteria, in all archaea, and also in some
organelles by an indirect route involving transamidation of
misacylated tRNA in contrast to the direct acylation by aspar-
aginyl-tRNA synthetase or glutaminyl-tRNA synthetase (4, 5).
tRNA-dependent transamidation may also represent in some
organisms the sole route to asparagine synthesis (6, 7). Whole
genome sequence analyses have failed to identify genes encoding
two other AARSs in some organisms. The lack of a recognizable
LysRS in most archaea was explained by the discovery of a novel
class I-type synthetase, which was unrelated to the class II-type
lysyl-tRNA synthetase previously characterized in bacteria and
eukaryotes (8). The absence of a canonical class I cysteinyl-
tRNA synthetase (CysRS) in the genomes of the thermophilic
archaea Methanocaldococcus (Methanococcus) jannaschii (9)
and Methanothermobacter (Methanobacterium) thermautotrophi-
cus (10) was an additional puzzle. Subsequent biochemical
experimentation revealed that CysRS was indeed absent and
instead, cysteinyl-tRNA (Cys-tRNA) was synthesized by a dual-
specificity prolyl-tRNA synthetase (11, 12), termed ProCysRS
(13). The presence of synthetases with relaxed substrate speci-
ficity is assumed to have been a step in AARS evolution (11, 14).
Recently, the dual-specificity ProCysRS was also shown to be
present in Giardia lamblia (15), which represents an ancient
lineage of the eukaryotes, and the extremely thermophilic
bacterium Thermus thermophilus (16). However, both of these
organisms also possess a canonical CysRS. These observations
raised the question of whether the dual-specificity enzyme was
sufficient for Cys-tRNA biosynthesis in organisms with both
enzymes.
Previous studies have suggested that the mesophilic archaeon
Methanococcus maripaludis, which contains a canonical CysRS
(17), also possessed a ProCysRS because a temperature-sensitive
Escherichia coli cysS mutant could be rescued by the methano-
coccal proS gene (11). Because M. maripaludis is a genetically
tractable archaeon (18), we wanted to test whether the ProCysRS
was sufficient for Cys-tRNA synthesis in vivo, as it appears to be
in the hyperthermophile M. jannaschii. If the dual-specificity
enzyme was sufficient, we predicted that the cysS gene, which
encodes the canonical CysRS enzyme, would not be essential for
growth. Here, we document the viability of a mutant of M.
maripaludis where the cysS gene has been disrupted, and propose
that the presence of a functional proS gene is the minimum
requirement for Cys-tRNA synthesis in this archaeon.
Materials and Methods
Construction of the Integration Vectors for
M. maripaludis
. The
integration vector pIJA03 was based on the E. coli plasmid pUC
and lacks a suitable replication origin for the methanococci. It
contains the pac cassette, which encodes puromycin resistance in
methanococci (19, 20). The pac cassette is flanked by two
multicloning regions that allow direct cloning of genomic DNA.
For construction of pIJA03-cysS, an internal part of the cysS
gene (507 bp) was PCR amplified from the plasmid pPH21310
(21) containing the M. maripaludis strain JJ1 cysS gene with the
Abbreviations: AARS, aminoacyl-tRNA synthetase; Cys-tRNA, cysteinyl-tRNA; CysRS, cystei-
nyl-tRNA synthetase; ProRS, prolyl-tRNA synthetase; ProCysRS, prolyl-cysteinyl-tRNA
synthetase.
§
To whom reprint requests should be addressed at: Department of Molecular Biophysics
and Biochemistry, Yale University, P.O. Box 208114, 266 Whitney Avenue, New Haven, CT
06520-8114. E-mail: soll@trna.chem.yale.edu.
The publication costs of this article were defrayed in part by page charge payment. This
article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C.
§1734 solely to indicate this fact.
14292–14297
兩
PNAS
兩
December 4, 2001
兩
vol. 98
兩
no. 25 www.pnas.org兾cgi兾doi兾10.1073兾pnas.201540498

primers cysSA (GGACGCGTTGCATACAAAACTGAA-
GACG) and cysSB (GCTCTAGACAATCGGGCTTCGG-
TAG), and it was subsequently cloned into the pZERO-2 vector
(Invitrogen). After digestion of the construct with the appro-
priate restriction enzymes (MluI and XbaI), the cysS fragment
was excised, purified, and cloned into the pIJA03 vector by using
the multicloning site MCS1. The orientation of the cysS fragment
was confirmed by DNA sequencing. The plasmid pIJA03-hdrA
was constructed in a similar fashion by using specific primers
(179A: GCAGATCTCTGAATTAGACGGTGTAGCC and
179B: GGTCTAGAGTCATCTGTCGGGAAAGT) and plas-
mid pPH41717f containing the gene encoding heterodisulfide
reductase as template. The PCR product was cloned in to
pZERO-2 vector after digestion with the restriction enzymes
BglII and XbaI. Finally, the plasmid pIJA03-fmdB was con-
structed as above (primers: 132A, GGAGATCTTATCGTCT-
GTCCAGTATGC; and 132B, GGTCTAGAGGAAATACTC-
CATATCTTGAC) by using the plasmid pPH22314r containing
the gene encoding the molybdenum-dependent formyl meth-
anofuran dehydrogenase. The PCR product was cloned into the
pZERO-2 vector previously digested with BglII and XbaI re-
striction enzymes. Both pIJA03-hdrA and pIJA03-fmdB plasmids
were used as control vectors to check transformation efficiencies
in M. maripaludis after replacement of an essential and a
nonessential gene, respectively.
Replacement of the
cysS
Gene in
M. maripaludis
. The pBD1 vector
for the cysS gene replacement was constructed by using the
plasmid pPH21310 (containing the complete cysS gene of M.
maripaludis strain JJ1 as well as part of the upstream ORF). The
plasmid was digested with the blunt end restriction endonucle-
ases MscI and HpaI to remove a 482-bp internal fragment of cysS,
and it was replaced with a 1351-bp EcoRV兾Eco47III fragment
from pIJA03 containing the pac cassette. After screening, a
plasmid pBD1 was found that contained the pac cassette in the
same orientation as the remaining portions of the cysS gene.
Before transformation of M. maripaludis cells by using the
polyethylene glycol (PEG) method (22), pBD1 was linearized by
digestion with HindIII and PvuII restriction enzymes. To verify
the insertion of the pac cassette into the cysS gene, total M.
maripaludis DNA was isolated and PCR analyzed with the
primers cysSF (ACTTACAACAACACTCGGGG) and cysSR
(CCTTCTTTTTGGGTTGTCCTC).
Cloning, Overexpression, and Purification of
M. maripaludis
Prolyl-
tRNA Synthetase (ProCysRS) and CysRS. The sequences of M.
maripaludis proS (accession no. AAG28517) and cysS gene
(accession no. AF163997) were used to design specific primers
for amplification of both genes from M. maripaludis genomic
DNA (wild-type strain JJ1). The primers in both cases contained
NdeI and BamHI restriction sites for subsequent cloning in to the
pET 15b expression vector (Novagen). The PCR product was
first cloned into the pCR2.1-TOPO vector (Invitrogen) and
sequenced. On digestion with NdeI and BamHI, the genes were
ligated into pET15b for expression of N-terminal His
6
-tagged
proteins in the E. coli BL21-Codon Plus(DE3)-RIL strain.
Cultures were grown at 37°C in LB medium, supplemented with
100
␮
g兾ml ampicillin and 34
␮
g兾ml chloramphenicol. Expres-
sion of the His
6
-tagged protein was induced for 4–6 h with the
addition of 1 mM isopropyl-
␤
-D-thiogalactoside before harvest-
ing the cells. The enzyme was purified by Ni-nitrilotriacetic
acid-agarose chromatography (Qiagen, Chatsworth, CA) as pre-
viously described (23). The M. maripaludis His
6
-ProRS and
His
6
-CysRS were ⬎99% pure, as judged by Coomassie Brilliant
Blue staining after SDS兾PAGE. Active fractions were pooled
and dialyzed against aminoacylation buffer (see below) contain-
ing 40% glycerol and stored at ⫺20°C.
Detection of ProCysRS and CysRS by Immunoblot Analysis. Both
recombinant proteins were used to raise polyclonal rabbit anti-
bodies (four injections, ⬇250
␮
g each) at the Yale Immunization
Service (Yale University, New Haven, CT). The polyclonal
antibodies were purified before use by Sepharose-protein A
affinity chromatography. Different dilutions of S100 extracts
from both the wild-type and the mutant strain (varying from
1:10–1:100) were analyzed on 10% polyacrylamide兾SDS gels. As
positive controls for the detection of the corresponding proteins
in the cell extracts, we included purified His
6
-tagged enzymes.
The proteins were transferred onto nitrocellulose membranes
(Nitropure, Micron Separations) by using a Bio-Rad semidry
blotter. For the immunoblot analysis, the colorimetric Opti-4CN
substrate and detection kit (Bio-Rad) was used (horseradish
peroxidase conjugate). The membranes were incubated with
different dilutions of the polyclonal antibodies (1:100–1:1000)
for optimal detection.
Aminoacylation and ATP-PP
i
Exchange Assays. Cys-tRNA or prolyl-
tRNA formation was assayed in aminoacylation buffer [50 mM
Hepes-KOH, pH 7.0兾50 mM KCl兾10 mM ATP兾15 mM MgCl
2
兾5
mM DTT兾0.05 mM [
35
S]cysteine (1,075 Ci兾mmol) or [
3
H]pro-
line (104 Ci兾mmol)] at 37°C as previously described (23) in the
presence of 50–250 nM of either M. maripaludis ProCysRS or
CysRS, by using unfractionated tRNA from M. maripaludis (40
␮
M,1mg兾ml) as a substrate (prepared with standard methods
and purified by DEAE-cellulose chromatography). Aliquots (20
␮
l) from the reaction mixture were removed periodically, spotted
on Whatman 3 MM paper filter disks and washed three times in
10% trichloroacetic acid to remove the free amino acid. After
drying, the radioactivity was measured by liquid scintillation
counting. The K
M
values were determined from the correspond-
ing reciprocal plots in the presence of limiting concentrations of
the variable substrates (0.5–500
␮
M[
35
S]cysteine or [
3
H]proline)
and saturating conditions of the fixed substrates (10–100 ⫻ K
M
).
Values of k
cat
were determined by using saturating substrate
concentrations and 100–250 nM of either ProCysRS or CysRS.
Pro-AMP and Cys-AMP formation were determined in the
presence or absence of unfractionated M. maripaludis tRNA (40
␮
M) by using [
32
P]PP
i
with a specific activity of 2000 cpm兾nmol.
The reaction mixture also contained 50–500 nM ProCysRS, 1
mM ATP, 1 mM KF, and 2 mM proline or 1–10 mM cysteine in
aminoacylation buffer. Aliquots (40
␮
l) were removed period-
ically, and the reaction was quenched by the addition of 1%
activated carbon in the presence of 0.4 M sodium pyrophosphate
and 15% perchloric acid. After filtration of the mixture through
glass microfiber filter disks (GF兾C, Whatman), the amount of
32
P-labeled ATP was measured by liquid scintillation counting.
Results
M. maripaludis
Cells Are Viable in the Absence of
cysS
. To determine
whether cysS was indeed essential for cell viability, M. maripalu-
dis cells were transformed with pIJA03-cysS,a‘‘suicide’’ vector
containing 507 bp of an internal portion of cysS. This vector
contains the pac cassette, which encodes puromycin resistance in
methanococci (19). The transformants were expected to acquire
puromycin resistance after a single homologous recombination
between the cysS fragment on the circular plasmid and the cysS
gene encoded in the genomic DNA of the organism (24). The
resulting merodiploid would then contain two truncated copies
of the cysS gene and would be viable only if cysS was not an
essential gene. Thus, the ability to recover transformants would
be a first test of whether or not cysS was essential. Because the
transformation frequency depends to a great extent on the size
of the fragment used to achieve homologous recombination,
controls included fragments of the same size from essential and
nonessential genes. Moreover, even for an essential gene, a low
transformation frequency was expected because of rearrange-
Stathopoulos et al. PNAS
兩
December 4, 2001
兩
vol. 98
兩
no. 25
兩
14293
BIOCHEMISTRY

ments that might restore a wild-type copy of the gene or
recombination at other sites. The number of transformants
obtained with pIJA03-cysS was comparable to that found with
pIJA03-fmdB, a plasmid that integrated at a nonessential gene
encoding for the molybdenum-dependent formyl methanofuran
dehydrogenase (Table 1). In contrast, the number of transfor-
mants obtained with pIJA03-hdrA, a plasmid that integrated at
an essential gene encoding for the heterodisulfide reductase, was
much lower. Therefore, by this test, cysS did not appear to be an
essential gene in M. maripaludis.
To confirm this result, cells were transformed with the lin-
earized plasmid pBD1 (Fig. 1A). In this plasmid, the pac cassette
was incorporated in the middle of the cysS gene. Puromycin
resistance would then be acquired by homologous recombination
at two sites, so that an internal portion of cysS would be replaced
by the pac cassette (Fig. 1 A). The number of transformants
obtained with pBD1 was comparable to that of other linear
plasmids containing similar size fragments of genomic DNA
(data not shown), supporting the conclusion that integration of
the pac cassette did not require rare rearrangements or other
genetic events. To determine whether the cysS gene was in fact
disrupted, the DNA near the cysS locus and the levels of CysRS
were examined in a representative mutant strain JJ200. Ampli-
fication by PCR of the cysS locus in strain JJ200 indicated an
increase in the size of the DNA from 1.9 kb to 2.8 kb, as expected
if a ⬇500-bp fragment of cysS was replaced with the 1.35-kb pac
cassette (Fig. 1B). Similarly, immunoblot analysis with polyclonal
antibodies raised against either M. maripaludis ProCysRS or
CysRS also verified that CysRS was not present in cell extracts
of the mutant strain (Fig. 1C Lower). On the contrary, both
synthetases were present in cell extracts of the wild type strain
JJ1 (Fig. 1C Upper). The inability to detect CysRS in the mutant
strain confirms that the cysS gene is not functional and that this
gene is not essential for growth.
ProCysRS Is Sufficient for Cys-tRNA Formation in the
M. maripaludis
cysS
Mutant. To elucidate which enzyme was responsible for
Cys-tRNA formation in the mutant lacking the canonical CysRS,
Table 1. Transformation of M. maripaludis by integration vectors
at essential and nonessential genes*
Plasmid Size of insert, bp
†
Total number of transformants
pIJA03-cysS 507 840
pIJA03-fmdB 488 470
pIJA03-hrdA 495 4
*Transformations by the PEG method (22) with 1
␮
g of supercoiled plasmid
DNA.
†
Size of the genomic DNA cloned into the insertion plasmid.
Fig. 1. Replacement of the M. maripaludis cysS gene with the pac cassette. (A) Strategy for construction of the gene replacement. pBD1 was constructed by
replacement of the MscI (M)-HpaI (H) fragment of cysS (solid arrow) with the pac cassette (arrow with vertical bars). Before transformation, pBD1 was digested
with PvuII (P) and HindIII (I) to form a linear plasmid. Puromycin resistance can then be acquired through two homologous recombination events leading to the
replacement of the internal portion of cysS with the pac cassette on the genome. Neighboring ORFs are indicated by open and stippled arrows. (B) PCR
amplification of the cysS locus in the mutant and wild type. The amplification was performed with the primers cysSF and cysSR (A). The templates were: 1,
pPH21310, which contained the wild-type cysS gene; 2, pBD1, which contained the pac cassette inserted into the cysS gene; 3, genomic DNA of the wild-type
strain JJ1; 4, genomic DNA of the M. maripaludis mutant strain JJ200 (see Materials and Methods). (C) Immunoblot analysis of M. maripaludis ProCysRS and CysRS
expression in wild-type and mutant strains. (Upper) S100 extracts (wild-type and mutant) and purified recombinant M. maripaludis CysRS and ProCysRS in the
presence of polyclonal anti-ProCysRS antibodies. (Lower) The same as above in the presence of polyclonal anti-CysRS antibodies.
14294
兩
www.pnas.org兾cgi兾doi兾10.1073兾pnas.201540498 Stathopoulos et al.

cell lysates were prepared from both the wild-type strain JJ1 and
the mutant strain JJ200. Thiaproline, which is a specific inhibitor
of the aminoacylation reactions of ProRS, also inhibits the
Cys-tRNA synthetase activity of the ProCysRS from M. jann-
aschii and other organisms (11, 15). For an S100 extract of the
wild-type strain of M. maripaludis, 1 mM of the inhibitor caused
an almost 50% inhibition of the trichloroacetic acid-precipitable
counts (Fig. 2A). The residual activity was attributed to the
canonical CysRS. On the other hand, when S100 extracts from
the mutant strain were tested for Cys-tRNA formation, only
background activity was detected in the presence of 1 mM of the
inhibitor (Fig. 2B). Thus, the only Cys-tRNA synthetase activity
detectable in the mutant was thiaproline sensitive. This result
supports the conclusion that M. maripaludis possesses a Pro-
CysRS capable of forming Cys-tRNA efficiently.
Genes encoding ProCysRS from three methanogenic archaea
have been previously shown to rescue the growth of a temper-
ature-sensitive E. coli cysS strain that also contained a gene
encoding the M. jannaschii tRNA
Cys
. To examine the dual
specificity of M. maripaludis ProCysRS in vitro, we cloned and
expressed the corresponding gene in E. coli, and purified the
enzyme by affinity chromatography. When M. maripaludis Pro-
CysRS was tested in the presence of homologous tRNA and
cysteine in the reaction mixture, we observed direct attachment
of cysteine on the tRNA (Fig. 3B). The enzyme was also able to
use proline (like any other ProRS, Fig. 3A). A conserved
property of all currently characterized dual-specificity ProCysRS
enzymes is the tRNA-dependent activation of cysteine (11, 15,
16, 23). This result is also the case with the M. maripaludis
enzyme, because Cys-AMP was formed only in the presence of
unfractionated homologous tRNA (Fig. 3C).
The coexistence of two pathways of Cys-tRNA formation in
M. maripaludis, one employing the canonical CysRS and one
based on ProCysRS, led to the examination of the kinetic
parameters that govern both aminoacylation reactions in vitro.
As determined in the aminoacylation reaction, the K
M
for
cysteine of the ProCysRS was almost eight times higher than that
of the canonical CysRS (Table 2). However, the k
cat
values for
the formation of Cys-tRNA by both enzymes were much closer
(2.2 s
⫺1
for CysRS and 0.8 s
⫺1
for ProCysRS). The canonical
activities of both enzymes exhibit almost the same affinities for
cysteine and proline, respectively, and the k
cat
values are com-
parable. From all of the above, it appears that, although Pro-
CysRS in this organism has a lower affinity for cysteine than the
single-specificity canonical enzyme, its catalytic efficiency as
measured in vitro was sufficient to support Cys-tRNA formation
in vivo in the absence of CysRS.
Growth Phenotype of
M. maripaludis cysS
Mutant. The growth of the
M. maripaludis cysS deletion mutant strain JJ200 was very similar
to that of the wild-type strain under a variety of conditions. M.
maripaludis is a facultative autotroph (25). Although it is capable
of autotrophic growth in mineral medium, it readily assimilates
acetate and a variety of amino acids, including proline, when they
are present (26, 27). Both the mutant and the wild-type strains
exhibited similar growth rates under autotrophic growth condi-
tions as well as in rich medium containing organic carbon sources
(Fig. 4 A and B, and data not shown). Low concentrations of
Fig. 2. Inhibition of Cys-tRNA formation in M. maripaludis S100 extract in the
presence of 1 mM thiaproline. (A) Wild-type S100 extract in the absence (
F
)or
presence (
E
) of thiaproline. The residual activity is attributed to the canonical
CysRS present in the extract. (B) Mutant JJ200 S100 extract under the same
conditions. The level of inhibition (⬇95%) indicates that only ProCysRS is
responsible for Cys-tRNA formation in this strain. Closed squares represent the
background level.
Fig. 3. Aminoacyl-tRNA and Cys-AMP formation by M. maripaludis ProCysRS. Aminoacylation was performed as described in Materials and Methods in the
presence of 0.05 mM [
3
H]proline or 0.05 mM [
35
S]cysteine (A and B;
F
). No activity was observed in the absence of either enzyme or tRNA (A and B;
■
). (C)
tRNA-dependent Cys-AMP synthesis as measured in the ATP-PPi exchange reaction (see Materials and Methods). The cysteine concentration used was 1–10 mM.
Formation of radiolabeled ATP was observed only in the presence of 1 mg兾ml total M. maripaludis tRNA (
F
) and not in the absence of tRNA (
E
). Closed squares
represent the background level (absence of either enzyme or cysteine from the reaction mixture).
Table 2. Kinetic constants of M. maripaludis ProCysRS and CysRS
in tRNA aminoacylation
Enzyme Amino acid K
M
,
␮
M k
cat
,s
⫺1
k
cat
兾K
M
,
␮
M
⫺1
䡠s
⫺1
ProCysRS Proline 4.6 ⫾ 1.3 0.9 0.19
Cysteine 74.5 ⫾ 4.7 0.8 0.01
CysRS Cysteine 9.7 ⫾ 2.3 2.2 0.22
Stathopoulos et al. PNAS
兩
December 4, 2001
兩
vol. 98
兩
no. 25
兩
14295
BIOCHEMISTRY

proline (ⱕ0.05 M) had little effect on the growth of the wild-type
or mutant strains (data not shown). Higher concentrations were
inhibitory for both strains, and, finally, no growth was observed
above 0.6 M proline (data not shown). M. maripaludis as a marine
archaeon exhibits optimum growth in medium containing a high
salt concentration (0.4 M NaCl). Both the mutant and wild-type
strains grew similarly throughout a concentration range of 0.05–1.2
M NaCl (data not shown). However, the mutant strain was some-
what less sensitive to a rapid downshift of NaCl concentration than
the wild-type (Fig. 4A). When cultures were shifted from the
optimum NaCl concentration (0.4 M) to 0.1 M, the wild-type strain
underwent a prolonged lag of ⬇10 h (Fig. 4C). In contrast, the lag
for the mutant was only 5 h, and was comparable to the lag observed
in the absence of a change in salt concentrations. This difference
was observed only over a narrow salt concentration range. A shift
to 0.05 M NaCl caused a prolonged lag for both the mutant and
wild-type strains (Fig. 4C). Similarly, when the shift to 0.1 M NaCl
was performed in mineral medium, both strains exhibited pro-
longed lags (data not shown).
Discussion
The growth rate of prokaryotes depends on the complex inter-
action of the rates of small molecule and macromolecule bio-
synthesis and is tightly regulated. Not surprisingly, the growth of
the M. maripaludis mutant that lacks a canonical CysRS was
nearly identical to that of wild type over a range of growth
conditions. Thus, the rate of Cys-tRNA formation was not
growth limiting under the conditions tested even in the absence
of the canonical CysRS. The Cys-tRNA-forming activity of the
ProCysRS is inhibited by high concentrations of proline. There-
fore, the failure of high concentrations of extracellular proline to
inhibit growth was presumably due to an inability of the cells to
accumulate high intracellular concentrations. The major growth
difference between the mutant and wild type was a 2-fold
reduction in the growth lag of the mutant during the shift to low
salt in rich medium. This effect did not appear to be a general
stress response because no difference was observed during the
shift from rich to mineral media. Salt downshifts in prokaryotes
are usually accompanied by rapid efflux of potassium ions and
other small molecules (28), and for some reason the mutant
strain was able to recover more quickly. The very narrow range
of conditions where this effect was observed is consistent with
the hypothesis that Cys-tRNA formation is not growth-limiting
under the described conditions.
The set of twenty different single-specificity aminoacyl-tRNA
synthetases has been regarded to be essential for translation in
every cell. Whereas other pathways exist for the generation of
amide aminoacyl-tRNAs (5), until recently, Cys-tRNA forma-
tion was considered to be carried out only by a canonical CysRS.
Our data clearly show that CysRS is not required for M.
maripaludis to be viable. Instead, a dual-specificity ProCysRS
appears to take over the task of Cys-tRNA synthesis in this
organism. Thus, one wonders how many other single-specificity
AARS enzymes may be dispensable in organisms that contain
the full complement of twenty AARSs. The intriguing discovery
that ProRS from several bacteria (e.g., T. thermophilus; ref. 16)
is able to catalyze in vitro the formation of both prolyl-tRNA and
Cys-tRNA suggests that such a dual-specificity enzyme may exist
even in bacteria. This remains to be tested in the future. Are
there any candidates for other dual-specificity synthetases?
Transposon mutagenesis studies to create a minimal Myco-
plasma genitalium genome suggested that the genes encoding
isoleucyl- (IleRS) and tyrosyl-tRNA synthetase were dispens-
able (29). However, such a procedure could give rise to truncated
versions of AARS enzymes, some of which have been shown to
be active as well as pairs of noncontiguous fragments with
synthetase activity (30). There is also no compelling reason why
an AARS should be required to compensate for the loss of
another synthetase. Possibly, synthetase activity could also be
provided by proteins that have other functions. For instance, a
recent report (31) demonstrated CysRS activity by a M. jann-
aschii ORF (MJ1477) that was assigned a polysaccharide hydro-
lase function (32). Because this M. jannaschii protein has ho-
mologs only in a small number of organisms with known
genomes (Thermotoga maritima and Deinococcus radiodurans),
it is not a general CysRS enzyme that provides Cys-tRNA in
the genomes that lack the canonical cysS gene such as M.
thermautotrophicus.
Aminoacyl-tRNA synthetase evolution may have involved en-
zymes that could specify more than one amino acid in a primitive
protein synthesis machinery (33). Duplication and diversification of
these primitive synthetases would then form the basis for the
evolutionary radiation that gave rise to the contemporary AARSs.
However, the high specificity of the ProCysRS for each of its
substrates is not a property expected of a primitive synthetase.
Therefore, we propose that this enzyme has evolved to function
optimally in certain types of cells. Because of its lower affinity for
cysteine, these cells presumably contain higher cytoplasmic con-
centrations of this amino acid. Whereas further proposals must be
speculative given the limited data on the dual-specificity enzyme’s
distribution, one can imagine that many anaerobes or organisms
that do not make glutathione might possess high cytoplasmic
concentrations of cysteine. The presence of a canonical CysRS in
many of these same organisms is consistent with proposals that this
gene was acquired late in evolution by horizontal gene transfer (34).
Because of the higher affinity of the CysRS for cysteine, the
acquisition of this enzyme could have been selected for because it
enabled the organism to grow with lower cysteine levels in the
cytoplasm.
We thank Carla Polycarpo, Gregory Razcniak, and Anselm Sauerwald
for advice, and Michael Ibba for critical comments. We also thank Gary
Olsen and Paul Haney for sharing sequences of M. maripaludis genes
before their publication. This work was supported by grants from the
National Institute for General Medical Sciences (D.S.), the National
Aeronautics and Space Administration (D.S.), and the Office of Science
of the Department of Energy (W.W. and D.S.).
Fig. 4. Growth response of the cysS mutant strain JJ200 and wild-type strain
JJ1ofM. maripaludis to a down shift in NaCl concentrations. The inoculum was
grown in rich medium with acetate and casamino acids at 37°C containing 0.4
M NaCl (24). Wild-type cells (
F
and
E
) and cysS mutant JJ200 (
Œ
and
‚
). (A)At
zero time, 5 ⫻ 10
7
cells were inoculated into prewarmed medium containing
the same concentration of NaCl (
F
and
Œ
) or 0.1 M NaCl (
E
and
‚
). (B) Specific
growth rates after the shift from 0.4 M NaCl into media of lower NaCl
concentrations. (C) Growth lags after the same shift into media of lower NaCl
concentrations.
14296
兩
www.pnas.org兾cgi兾doi兾10.1073兾pnas.201540498 Stathopoulos et al.

1. Ibba, M. & So¨ll, D. (2000) Annu. Rev. Biochem. 69, 617–650.
2. Ibba, M. & So¨ll, D. (1999) Science 286, 1893–1897.
3. Ibba, M. Becker, H. D., Stathopoulos, C., Tumbula, D. L. & So¨ll, D. (2000)
Trends Biochem. Sci. 25, 311–316.
4. Curnow, A. W., Hong, K.-W., Yuan, R., Kim, S.-I., Martins, O., Winkler, W.,
Henkin, T. M. & So¨ll, D. (1997) Proc. Natl. Acad. Sci. USA 94, 11819–11826.
5. Tumbula, D. L., Becker, H. D., Chang, W.-Z. & So¨ll, D. (2000) Nature
(London) 407, 106–110.
6. Curnow, A. W., Tumbula, D. L., Pelaschier, J. T., Min, B. & So¨ll, D. (1998)
Proc. Natl. Acad. Sci. USA 95, 12838–12843.
7. Becker, H. D. & Kern, D. (1998) Proc. Natl. Acad. Sci. USA 95, 12832–12837.
8. Ibba, M., Morgan, S., Curnow, A. W., Pridmore, D. R., Vothknecht, U. C.,
Gardner, W., Lin, W., Woese, C. R. & So¨ll, D. (1997) Science 278,
1119–1122.
9. Bult, C. J., White, O., Olsen, G. J., Zhou, L., Fleischmann, R. D., Sutton G. G.,
Blake, J. A., FitzGerald, L. M., Clayton, R. A., Gocayne, J. D., et al. (1996)
Science 273, 1058–1073.
10. Smith, D. R., Doucette-Stamm, L. A., Deloughery, C., Lee, H., Dubois, J.,
Aldredge, T., Bashirzadeh, R., Blakely, D., Cook, R., Gilbert, J. D., et al. (1997)
J. Bacteriol. 179, 7135–7155.
11. Stathopoulos, C., Li, T., Longman, R., Vothknecht, U. C., Becker, H. D., Ibba,
M.&So¨ll, D. (2000) Science 287, 479–482.
12. Lipman, R. S., Sowers, K. & Hou, Y. M. (2000) Biochemistry 39, 7792–7798.
13. Yarus, M. (2000) Science 287, 440–441.
14. Nagel, G. M. & Doolittle, R. F. (1995) J. Mol. Evol. 40, 487– 498.
15. Bunjun, S., Stathopoulos, C., Graham, D., Min, B., Kitabatake, M., Wang A. L.,
Wang, C. C., Vivares, C. P., Weiss, L. M. & So¨ll, D. (2000) Proc. Natl. Acad.
Sci. USA 97, 12997–13002. (First Published November 14, 2000; 10.1073兾
pnas.230444397)
16. Feng, L., Stathopoulos, C., Ahel, I., Mitra, A., Tumbula-Hansen, D. L., Hartsch
T.&So¨ll D. (2001) Extremophiles, in press.
17. Li, T., Graham, D., Stathopoulos, C., Haney, P., Kim, H.-S., Vothknecht, U.,
Kitabatake, M., Hong, K.-W., Eggertsson, G., Curnow, A. W., Lin, W., Celic,
I., Whitman, W. & So¨ll, D. (1999) FEBS Lett. 462, 302–306.
18. Tumbula, D. L. & Whitman, W. B. (1999) Mol. Microbiol. 33, 1–7.
19. Gernhardt, P., Possot, O., Foglino, M., Sibold, L. & Klein, A. (1990) Mol. Gen.
Genet. 221, 273–279.
20. Metcalf, W. W., Zhang, J. K., Apolinario, E., Sowers, K. R. & Wolfe, R. S.
(1997) Proc. Natl. Acad. Sci. USA 94, 2626–2631.
21. Haney, P. J., Badger, J. H., Buldak, G. L., Reich, C. I., Woese, C. R. & Olsen,
G. J. (1999) Proc. Natl. Acad. Sci. USA 96, 3578–3583.
22. Tumbula, D. L., Makula, R. A. & Whitman, W. B. (1994) FEMS Microbiol. Lett.
121, 309–314.
23. Stathopoulos, C., Jacquin-Becker, C., Becker, H. D., Li, T., Ambrogelly, A.,
Longman, R. & So¨ll, D. (2001) Biochemistry 40, 46–52.
24. Kim, W. & Whitman, W. B. (1999) Genetics 152, 1429–1437.
25. Jones, W. J., Paynter, M. J. B. & Gupta, R. (1983) Arch. Microbiol. 135, 91–97.
26. Whitman, W. B., Sohn, S. H., Kuk, S. U. & Xing, R. Y. (1987) Appl. Environ.
Microbiol. 53, 2373–2378.
27. Whitman, W. B., Shieh, J. S., Sohn, S. H., Caras, D. S. & Premachandran, U.
(1986) System. Appl. Microbiol. 7, 235–240.
28. Csonka, L. N. & Epstein, W. (1996) Escherichia coli and Salmonella: Cellular
and Molecular Biology (Am. Soc. Microbiol., Washington, DC), pp. 1210–1231.
29. Hutchison, C. A., Peterson, S. N., Gill, S. R., Cline, R. T., White, O., Fraser,
C. M., Smith, H. O. & Venter J. C. (1999) Science 286, 2165–2169.
30. Shiba, K. & Schimmel, P. (1992) Proc. Natl. Acad. Sci. USA 89, 1880–1884.
31. Fabrega, C., Farrow, M. A., Mukhopadhyay, B., de Crecy-Lagard, V., Ortiz,
A. R. & Schimmel, P. (2001) Nature (London) 411, 110–114.
32. Iyer, L. M., Aravind, L., Bork, P., Hofmann, K., Mushegian, A. R., Zhulin, I. B.
& Koonin, E. V. (2001) Genome Biol., in press.
33. Delarue, M. (1995) J. Mol. Evol. 41, 703–711.
34. Woese, C. R., Olsen, G. J., Ibba, M. & So¨ll, D. (2000) Microbiol. Mol. Biol. Rev.
64, 202–236.
Stathopoulos et al. PNAS
兩
December 4, 2001
兩
vol. 98
兩
no. 25
兩
14297
BIOCHEMISTRY