|
Originally published In Press as doi:10.1074/jbc.M005166200 on July 28, 2000
J. Biol. Chem., Vol. 275, Issue 42, 32535-32542, October 20, 2000
Metabolism of D-Aminoacyl-tRNAs in
Escherichia coli and Saccharomyces
cerevisiae Cells*
Julie
Soutourina,
Pierre
Plateau , and
Sylvain
Blanquet
From the Laboratoire de Biochimie, Unité Mixte de Recherche
7654, CNRS-Ecole Polytechnique, 91128 Palaiseau Cedex, France
Received for publication, June 14, 2000, and in revised form, July 27, 2000
 |
ABSTRACT |
In Escherichia coli, tyrosyl-tRNA
synthetase is known to esterify tRNATyr with tyrosine.
Resulting D-Tyr-tRNATyr can be hydrolyzed by a
D-Tyr-tRNATyr deacylase. By monitoring E. coli growth in liquid medium, we systematically searched for
other D-amino acids, the toxicity of which might be
exacerbated by the inactivation of the gene encoding
D-Tyr-tRNATyr deacylase. In addition to the
already documented case of D-tyrosine, positive responses
were obtained with D-tryptophan, D-aspartate, D-serine, and D-glutamine. In agreement with
this observation, production of D-Asp-tRNAAsp
and D-Trp-tRNATrp by aspartyl-tRNA synthetase
and tryptophanyl-tRNA synthetase, respectively, was established
in vitro. Furthermore, the two D-aminoacylated tRNAs behaved as substrates of purified E. coli
D-Tyr-tRNATyr deacylase. These results indicate
that an unexpected high number of D-amino acids can impair
the bacterium growth through the accumulation of
D-aminoacyl-tRNA molecules and that
D-Tyr-tRNATyr deacylase has a specificity broad
enough to recycle any of these molecules. The same strategy of
screening was applied using Saccharomyces cerevisiae, the
tyrosyl-tRNA synthetase of which also produces D-Tyr-tRNATyr, and which, like E. coli, possesses a D-Tyr-tRNATyr deacylase
activity. In this case, inhibition of growth by the various 19 D-amino acids was followed on solid medium. Two isogenic strains containing or not the deacylase were compared. Toxic effects of
D-tyrosine and D-leucine were reinforced upon
deprivation of the deacylase. This observation suggests that, in yeast,
at least two D-amino acids succeed in being transferred
onto tRNAs and that, like in E. coli, the resulting two
D-aminoacyl-tRNAs are substrates of a same
D-aminoacyl-tRNA deacylase.
| |
INTRODUCTION |
D-Amino acids are widely distributed in living
organisms (1). Examples of D-amino acid natural occurrence
include D-Ala- and D-Glu-containing
peptidoglycans of bacterial cell walls, D-amino acid-containing natural peptide antibiotics (1-5), or the presence of
D-Asp and D-Ser at high concentrations in human
brain (6-10). D-Ala results from the transformation of
L-Ala by a racemase. In the case of D-Glu
production, a racemase or a transaminase is involved, depending on the
bacterium. Small amounts of D-amino acids can also possibly
appear as side products of various biosynthetic pathways. Conversion of
the L- to the D-stereoisomer of tryptophan was
observed in the presence of tryptophan synthase (11, 12). Similarly, in
the case of methionyl-tRNA synthetase, a weak catalysis of  -carbon
hydrogen-deuterium exchange of L-methionine was evidenced in vitro (13). Such an exchange suggests possible conversion of L-methionine into D-methionine at the
surface of an amino acid-binding enzyme. As discussed earlier (14),
D-tyrosine might arise at the step of the addition of an
amino group to 4-hydroxyphenylpyruvate. Moreover, D-amino
acids are likely to be nonspecifically formed as side reaction products
in the presence of pyridoxal phosphate-containing enzymes or of
pyridoxal phosphate alone (15, 16).
If externally added, D-amino acids exert toxicity toward
many organisms (1, 17-26). Possible causes of this toxicity are multiple. For instance, in the case of Escherichia coli,
D-amino acids can be lethal because they are erroneously
incorporated in peptidoglycan (23, 25, 26). In the case of
Bacillus subtilis, strains capable of efficiently pumping
D-tyrosine have been described. The growth of such strains
is decreased upon addition of this D-amino acid to the
culture medium (19). Inhibition of prephenate dehydrogenase and the
consequent curtailment of L-tyrosine biosynthesis may
account for this behavior (18). However, incorporation of D-tyrosine into proteins could be evidenced with the above
B. subtilis strains (18).
In agreement with this observation, several studies indicate
D-tyrosyl-tRNATyr formation in E. coli (27-29) as well as in Saccharomyces cerevisiae (14). Moreover, to recycle tRNATyr esterified by
D-tyrosine and/or to minimize D-tyrosine
incorporation into proteins, E. coli and S. cerevisiae harbor a gene (dtd or DTD1,
respectively) encoding a deacylase capable of hydrolyzing D-Tyr-tRNATyr into free tRNATyr and
D-tyrosine. In vitro, the E. coli
deacylase also hydrolyzes D-Phe-tRNAPhe and, to
a lesser extent, Gly-tRNAGly (28). Upon inactivation of
these genes, cell growth becomes sensitive to the presence of
D-tyrosine in the culture medium (14, 29).
In the present study, the possibility that
D-Tyr-tRNATyr deacylase could be involved in
the deacylation of other D-aminoacyl-tRNAs distinct from
D-Tyr-tRNATyr was investigated in
vivo. In a first experiment, toxicities of each
D-amino acid on the growth of E. coli
dtd+ and  dtd strains were systematically
compared. An involvement of the dtd product in the
protection of E. coli against the toxicity of
D-Trp, D-Asp, D-Ser, and
D-Gln was revealed. In the case of D-Trp and
D-Asp, in vitro studies established that
D-Trp-tRNATrp and
D-Asp-tRNAAsp could be produced by the
corresponding aminoacyl-tRNA synthetases. Moreover, the two
D-aminoacyl-tRNAs were efficient substrates of purified
D-Tyr-tRNATyr deacylase.
In the case of S. cerevisiae, comparison of the growth of
DTD1 and dtd1 strains indicated protection
against D-leucine afforded by the
D-Tyr-tRNATyr deacylase gene.
| |
MATERIALS AND METHODS |
L-[5-3H]Tryptophan (740 GBq/mmol),
L-[2,3-3H]aspartic acid (703 GBq/mmol),
D-[2,3-3H]aspartic acid (666 GBq/mmol), and
[32P]-PPi (300 GBq/mmol) were from
PerkinElmer Life Sciences. Non-radioactive D- and
L-amino acids were from Sigma. Q-Sepharose and Superose 6 were from Amersham Pharmacia Biotech. Hydroxylapatite was from Bio-Rad.
214TP5215 C4 column (2.1 × 150 mm, 5-µm particles)
was from Vydac. Bakerbond Wide-Pore Butyl (C4) (15 µm
particles) was from Baker. Crude E. coli tRNA was from Roche
Molecular Biochemicals.
Purification of E. coli Tryptophanyl-tRNA Synthetase--
To
overproduce tryptophanyl-tRNA synthetase, the trpS gene of
plasmid pCH17 (Table I) was inserted into pBluescript under the control
of the lacZ promoter. For this purpose, the 3.2-kilobase pair SalI-EcoRI fragment of pCH17 was subcloned
between the XhoI and EcoRI sites of
pBluescript( )KS. Cells transformed by the resulting plasmid (pBStrpS)
were used as a source of enzyme.
Cells were grown overnight at 37 °C in 4 liters of 2× TY (30)
medium containing 200 µg of ampicillin/ml. Crude extract preparation, nucleic acid precipitation with streptomycin sulfate, and ammonium sulfate precipitation of proteins were performed as described previously (29). After the ammonium sulfate precipitation, the protein
pellet was dissolved in 20 ml of 20 mM potassium phosphate (pH 6.75), containing 0.1 mM EDTA and 10 mM
2-mercaptoethanol (buffer A), and dialyzed against 2 liters of the same
buffer. The resulting sample was applied on a Q-Sepharose column
(2.6 × 16 cm), which was eluted at a flow rate of 180 ml/h with a
1.35-liter linear gradient from 0 to 300 mM KCl in buffer
A. Tryptophanyl-tRNA synthetase was recovered at 120 mM
KCl. Fractions showing tryptophan-dependent ATP-PPi exchange activity were directly applied on a
hydroxylapatite column (2.6 × 10 cm) equilibrated in buffer A. This column was eluted at a flow rate of 20 ml/h using a 2-liter linear
gradient from 20 to 500 mM potassium phosphate (pH 6.75).
Tryptophanyl-tRNA synthetase activity was recovered at 220 mM potassium phosphate and concentrated by an ammonium
sulfate precipitation (70% saturation). After centrifugation at
10,000 × g for 30 min, the pellet was dialyzed against
20 mM Tris-HCl (pH 7.8) containing 60% glycerol, 0.1 mM EDTA, and 10 mM 2-mercaptoethanol, and
stored at 20 °C. 230 mg of enzyme were recovered from 15.2 g
of cells (wet weight), with a yield of 70%.
According to SDS-polyacrylamide gel electrophoresis analysis, purified
tryptophanyl-tRNA synthetase was at least 95% homogeneous. Concentration of tryptophanyl-tRNA synthetase was calculated using a
Mr of 2 × 37,497 and a light absorption
coefficient of 0.739 A280 units
mg 1 ml, as deduced from the amino acid
sequence of the protein.
Purification of E. coli Aspartyl-tRNA
Synthetase--
Aspartyl-tRNA synthetase was purified from E. coli strain JM101TR transformed by plasmid pBluescript()KSaspS
(pBSaspS) (31). Cells were grown overnight at 37 °C, in 1 liter of
2× TY medium containing 500 µg of ampicillin/ml. Crude extract
preparation and streptomycin sulfate and ammonium sulfate
precipitations followed the protocol described previously (29). After
the ammonium sulfate precipitation, the protein pellet was dissolved in
5 ml of 20 mM Tris-HCl (pH 7.8), containing 10 mM 2-mercaptoethanol, 0.1 mM EDTA, and 50 mM KCl (buffer B), and dialyzed against 2 liters of the
same buffer for 2 h. Then, the protein sample was applied on a
Superose 6 column (2 × 50 cm) equilibrated in buffer B. The column was eluted at a flow rate of 12 ml/h with 112 ml of buffer B. Fractions showing aspartyl-tRNA synthetase tRNA aminoacylation activity
were pooled and directly applied on a Q-Sepharose Hi-Load column
(1.6 × 10 cm) equilibrated in buffer B. The column was eluted at
a flow rate of 150 ml/h with a 260-ml linear gradient from 50 to 400 mM KCl in buffer B. Enzyme activity was recovered at 275 mM KCl. After ammonium sulfate precipitation (70%
saturation) and centrifugation at 10,000 × g for 30 min, the pellet was dialyzed against 20 mM Tris-HCl (pH
7.8) containing 60% glycerol, 0.1 mM EDTA, and 10 mM 2-mercaptoethanol, and stored at 20 °C. 30 mg of
enzyme were recovered from 2.6 g of cells (wet weight) with a
yield of 45%.
Recovered aspartyl-tRNA synthetase was at least 95% homogeneous,
according to SDS-polyacrylamide gel electrophoresis analysis. Enzyme
concentration was determined using a Mr of
2 × 65,913 and a light-absorption coefficient of 0.589 A280 units mg1 ml, as
calculated from the amino acid sequence of the protein.
Purification of E. coli tRNATrp--
The E. coli tRNATrp gene (trpT) was constructed
and overexpressed in strain JM101TR (Table I) from a pBSTNAV
derivative, according to a previously described method (32).
Since E. coli tRNATrp renaturation requires
exposure to heat in the presence of Mg2+ (33, 34), the
crude tRNA extract from the above cells was dissolved in 20 mM Tris-HCl (pH 7.8) containing 10 mM
MgCl2 and 0.1 mM EDTA, and heated for 5 min at
60 °C. Upon this treatment, the L-tryptophan acceptance
of the tRNA sample increased from 370 to 600 pmol of
L-tryptophan/A260 unit.
The tRNATrp sample was further purified on a preparative
C4 column. A set of 10 samples of 70 A260 units each were successively chromatographed on a 20-ml Bakerbond C4 column
(1.6 × 10 cm) equilibrated in 1 M sodium formate
buffer (pH 5.5) containing 10 mM
NaH2PO4 and 8 mM MgCl2
(solution C). The column was eluted at a flow rate of 0.5 ml/min with
(i) a 15-ml linear gradient from 0% to 20% of a solution containing
10 mM NaH2PO4 (pH 5.5) plus 10%
methanol (solution D), (ii) a 20-ml linear gradient from 20% to 60%
of solution D, and (iii) a 10 ml linear gradient from 60% to 100% of
solution D. Three peaks of L-tryptophan acceptance activity were recovered at 640, 440, and 140 mM sodium formate
concentrations (36%, 56%, and 86% of solution D, respectively). With
a crude tRNA extract from a non-overproducing strain, one single peak was observed at 140 mM sodium formate. Therefore, the first
two peaks are likely to reflect undermodified stages of the
overproduced tRNATrp and were discarded. After dialysis
against 3 liters of a 20 mM Tris-HCl buffer (pH 7.8)
containing 1 mM MgCl2 and 0.1 mM
EDTA, the tRNA sample corresponding to 140 mM sodium
formate was precipitated with ethanol and stored. Before use, the
tRNATrp pellet was recovered by centrifugation, dissolved
in the above buffer, and heated for 5 min at 60 °C. Purified
tRNATrp accepted 1400 pmol of
L-tryptophan/A260 unit.
Purification of E. coli tRNAAsp--
The E. coli tRNAAsp gene (aspT) was constructed
and overexpressed in JM101TR cells as described above in the case of
tRNATrp. The resulting crude tRNA extract accepted 700 pmol
of L-aspartic acid/A260 unit.
The tRNAAsp sample was further purified on a 20-ml
Bakerbond C4 column, as described above for
tRNATrp. One single peak of L-Asp acceptance
activity was recovered in agreement with the already described full
post-transcriptional modification of overproduced tRNAAsp
(35). After C4 chromatography, the tRNA sample was
precipitated with ethanol and stored. Purified tRNAAsp
accepted 1000 pmol of L-aspartic
acid/A260 unit.
Aminoacyl-tRNA Synthetase Activity
Assays--
Tryptophan-dependent ATP-PPi
exchange activity was assayed during 10 min at 28 °C in 100 µl of
a reaction mixture containing 20 mM Tris-HCl (pH 7.8), 7 mM MgCl2, 2 mM ATP, 0.1 mM EDTA, 50 µg/ml bovine serum albumin, 2.5 mM 2-mercaptoethanol, 2 mM
L-tryptophan, 2 mM
[32P]PPi (20-80 MBq/mmol), and catalytic
amounts of enzyme. After quenching of the reaction, labeled ATP was
adsorbed on charcoal, filtered, and counted as described previously
(36).
Aminoacylation of tRNATrp with L- or
D-tryptophan was assayed during 10 min at 28 °C in 100 µl of a reaction mixture containing 20 mM Tris-HCl (pH
7.8), 7 mM MgCl2, 2 mM ATP, 0.1 mM EDTA, 50 µg/ml bovine serum albumin, 2.5 mM 2-mercaptoethanol, 1.8-3.6 µM
L- or D-tryptophan, 0.35 µM
purified tRNATrp, and catalytic amounts of
tryptophanyl-tRNA synthetase. The reaction was quenched by the addition
of 10 µl of 3 M sodium acetate (pH 5.5) plus 55 µl of
phenol saturated with a 20 mM sodium acetate solution (pH
5.3). After vigorous shaking and centrifugation, the aqueous phase was
precipitated with ethanol. The samples were chromatographed on a
C4 HPLC1 column
as described below. Light absorption of the elution profile was
measured at 260 nm. Concentrations of aminoacylated and
non-aminoacylated tRNATrp were deduced from the areas of
the corresponding absorption peaks.
Aminoacylation of tRNATrp with
L-[3H]tryptophan (2500 Ci/mol) was also
measured by scintillation counting. In this case, the reaction was
quenched by the successive addition of 2.5 ml of ice-cold trichloroacetic acid (5%, w/w) containing 0.5% tryptophan, and of 10 µl of carrier RNA from yeast (4 mg/ml). The precipitate was recovered
on a Whatman GF-C filter, and the retained radioactivity was measured
in a Beckman LS1801 scintillation counter.
Aminoacylation of tRNAAsp by aspartyl-tRNA synthetase was
assayed by scintillation counting, as described above for
tryptophanyl-tRNA synthetase, except that 5 µM
tRNAAsp and 1 µM either L- or
D-[3H]aspartic acid (500 Ci/mol) were used in
the assay. Ice-cold trichloroacetic acid without amino acid was added
to quench the reaction.
HPLC of Aminoacylated tRNATrp--
The tRNA samples
(0.025-2 A260 units) were dissolved in 100 µl
of a 3 M sodium formate buffer (pH 5.5) containing 10 mM NaH2PO4 and 8 mM
MgCl2. Then, each sample was loaded on C4 HPLC
column (2.1 × 150 mm) equilibrated with solution C. To recover
the tRNA and its aminoacylated derivative, the eluant of the column was progressively changed to solution D, through a 6.75-ml linear gradient
from 0% to 30% of solution D, followed by a 12-ml linear gradient
from 30% to 100% of solution D. The absorbance of the column effluent
was measured at 260 nm.
Preparation of Aminoacylated tRNAs--
For
D-Trp-tRNATrp preparation, a reaction mixture
(4 ml) containing 20 mM Tris-HCl (pH 7.8), 7 mM
MgCl2, 2 mM ATP, 0.1 mM EDTA, 50 µg/ml bovine serum albumin, 2.5 mM 2-mercaptoethanol,
0.35 µM of purified tRNATrp, 3.6 µM D-tryptophan, and 2 µM
tryptophanyl-tRNA synthetase was incubated at 28 °C for 10 min.
After phenol extraction and ethanol precipitation, the sample was
purified by C4 HPLC, as described above. Fractions
containing D-Trp-tRNATrp were pooled and
precipitated with ethanol.
L-[3H]Trp-tRNATrp was prepared
under the same conditions, except that the reaction mixture (1.2 ml)
contained 0.35 µM purified tRNATrp (0.3 A260 units), 10 µM
L-[3H]tryptophan (100 Ci/mol), and 1 µM tryptophanyl-tRNA synthetase.
For L-[3H]Asp-tRNAAsp
preparation, a reaction mixture (300 µl) containing 20 mM
Tris-HCl (pH 7.8), 7 mM MgCl2, 2 mM
ATP, 0.1 mM EDTA, 50 µg/ml of bovine serum albumin, 2.5 mM 2-mercaptoethanol, 1 µM purified
tRNAAsp, 30 µM
L-[3H]aspartic acid (500 Ci/mol), and 1.5 µM aspartyl-tRNA synthetase was incubated at 28 °C for
10 min. After phenol extraction and ethanol precipitation, the product
was purified by chromatography on a Trisacryl GF05 column, as described
previously for Tyr-tRNATyr (29).
Because the D-aspartic acid sample was contaminated by
L-aspartic acid (see "Results"), the preparation of
D-aspartyl-tRNA was achieved in two steps. First, the
reaction mixture (1 ml) containing 20 mM Tris-HCl (pH 7.8),
7 mM MgCl2, 2 mM ATP, 0.1 mM EDTA, 50 µg/ml bovine serum albumin, 2.5 mM 2-mercaptoethanol, 60 µM
D-[3H]aspartic acid (500 Ci/mol), 5 µM purified tRNAAsp, and 0.325 µM aspartyl-tRNA synthetase was incubated at 28 °C for
10 min. Under these conditions, most of the contaminating L-aspartic acid was exhausted through rapid transfer onto
the tRNA and practically no D-aspartic acid was consumed
for aminoacylation. After precipitation with ethanol, the supernatant
containing L-Asp-free D-[3H]aspartic acid was lyophilized,
suspended in water, and further used in an aminoacylation reaction at
28 °C for 20 min with 5 µM purified
tRNAAsp and 6.5 µM aspartyl-tRNA synthetase
(second step). After phenol extraction and ethanol precipitation, the
sample was purified on a Trisacryl GF05 column, as described (29).
D-Tyr-tRNATyr Deacylase Activity
Measurements--
Initial rates of D- or
L-Trp-tRNATrp deacylation were measured for 5 min at 28 °C in 100-µl assays containing 0.1-0.3 µM
D-Trp-tRNATrp or
L-[3H]Trp-tRNATrp, 20 mM Tris-HCl (pH 7.8), 5 mM MgCl2,
0.1 mM EDTA, and catalytic amounts of purified E. coli D-Tyr-tRNATyr deacylase (29). The
reaction was quenched by addition of phenol. Then, the aqueous phase
was precipitated with ethanol and analyzed by C4 HPLC.
Initial rates of
L-[3H]Trp-tRNATrp hydrolysis were
also measured by scintillation counting, as described previously for
tyrosyl-tRNA hydrolysis (29).
Initial rates of D- or
L-[3H]-Asp-tRNAAsp hydrolysis by
the deacylase were measured by scintillation counting (29).
| |
RESULTS |
Toxicity of D-Tryptophan Depends on the Presence or
Absence of D-Tyr-tRNATyr
Deacylase--
Preliminary experiments suggested that the growth of a
dtd strain exhibited enhanced sensitivity to the presence
of D-tryptophan, when compared with an isogenic
dtd+ strain (29). This behavior was confirmed by
plating strains K37TyrH
(dtd::kan, Table
I) and K37 (dtd+)
on M9 minimal medium agar plates (30) supplemented with 0.2% glucose
and 5 mM D-tryptophan. In these conditions, the
growth of the dtd strain was almost completely abolished,
while the dtd+ strain still grew (Fig.
1).
View larger version (151K):
[in this window]
[in a new window]
|
Fig. 1.
Effect of D-tryptophan on the
growth of E. coli strains K37
(dtd+) and
K37TyrH
(dtd). Cells were left to grow
for 40 h at 37 °C on M9 minimal medium agar plates supplemented
with the indicated concentrations of D-tryptophan (0, 0.2, 0.8, and 5 mM).
|
|
Therefore, it could be suspected that D-tryptophan,
similarly to D-tyrosine (29), was transferred onto its
specific tRNA in vivo, and that the resulting
D-Trp-tRNATrp was deacylated by
D-tyrosyl-tRNA deacylase.
Toxicity Associated with Several Other D-Amino Acids
Also Depends on the Presence of the Deacylase--
To determine
whether the D-tyrosyl-tRNA deacylase protected E. coli cells against D-amino acids other than
D-tyrosine or D-tryptophan, we studied the
effect of 18 D-amino acids on the growth of strains K37
(dtd+) and K37TyrH
(dtd::kan) (Table
II). Because of its limited solubility,
D-tyrosine could not be included in this toxicity test.
Bacteria were inoculated at a final OD650 of 0.003 in
liquid M9 minimal medium supplemented with 0.2% glucose and various
concentrations of the D-amino acid under study (0-25
mM). After an 8-h incubation at 37 °C, the optical
density of the culture was compared with that of a control culture
without added D-amino acid.
View this table:
[in this window]
[in a new window]
|
Table II
Effect of various D-amino acids on the growth of E. coli strains K37 and K37TyrH
Cells were pre-grown to mid-exponential phase (0.2-0.3 OD650)
in M9 minimal medium without D-amino acid. Then, they were
inoculated at an optical density of 0.003 at 650 nm in liquid M9
minimal medium containing or not different D-amino acid
concentrations. After an 8-h incubation at 37 °C, the optical
densities of the cultures were measured at 650 nm. In the control
culture without D-amino acid, the optical density amounted
to 0.3-0.4. D-amino acid concentrations were varied from 1 µM to 512 µM in the case of
D-Val, D-Cys, D-Trp, and from 0.05 mM to 25 mM with the other D-amino
acids. Successive concentrations differed by a factor of 2. Because of
its low solubility, D-tyrosine was omitted from this test.
Indicated in the table are the D-amino acid concentrations
causing a 2-fold decrease in the optical density at the end of
incubation, as compared to the control culture without
D-amino acid.
|
|
Under these conditions, no inhibition of the bacterial growth was
observed upon addition of D-Ala, D-Pro, or
D-Arg. With D-Asn, D-His,
D-Ile, D-Leu, D-Met, and
D-Phe, the toxic effect was small and no difference could
be noted between strains K37 and K37TyrH. Several
D-amino acids (D-Glu, D-Thr,
D-Lys, D-Val, and D-Cys) markedly
inhibited E. coli growth (Table II). However, the toxicity of these amino acids was the same with the two strains.
Finally, a difference in the behaviors of dtd and
dtd+ strains was observed in the presence of
D-Asp, D-Ser, D-Gln, or, as expected, of D-Trp (Table II). The toxicity of these
D-amino acids was significantly more pronounced when
the dtd gene was lacking. Consequently, we suspected that
these D-amino acids are transferred onto their specific
tRNA in vivo and that the resulting aminoacyl-tRNAs are
deacylated by D-Tyr-tRNATyr deacylase.
These two hypotheses were verified in the case of D-tryptophan, a hydrophobic amino acid larger than
D-tyrosine, and in the case of the small ionizable
D-aspartic acid.
E. coli Tryptophanyl-tRNA Synthetase Catalyzes Aminoacylation of
tRNATrp with D-Tryptophan--
Because
radioactive D-tryptophan was not commercially available, we
monitored aminoacylation of tRNATrp by
D-tryptophan with the help of HPLC. Enzymatic assays were performed with purified tRNATrp and purified
tryptophanyl-tRNA synthetase. The chromatogram of a control reaction
without added synthetase is shown in Fig.
2A. Assignation of the peak at
52 min to tRNATrp was obtained by collecting fractions and
assaying them for L-tryptophan acceptance. Upon incubation
of the tRNA sample in the presence of 1.8 µM
L-[3H]tryptophan and 1 nM
tryptophanyl-tRNA synthetase prior to the chromatography, an additional
peak became visible at 68 min on the chromatogram (Fig. 2B).
This peak was assigned to Trp-tRNATrp because (i) the
fractions corresponding to this peak contained radioactivity and (ii)
the quantity of L-Trp-tRNA as calculated from the peak area
corresponded to the quantity measured by trichloroacetic acid
precipitation and counting before application on the column.
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 2.
C4-HPLC analysis of
tRNATrp aminoacylation (A-C) and of
D-Trp-tRNATrp hydrolysis
(D-F). The absorbance of the column effluent was
measured at 260 nm. A, tRNATrp (0.35 µM) incubated without tryptophanyl-tRNA synthetase and
without tryptophan. B, tRNATrp incubated for 10 min in the presence of 1.8 µM
L-[3H]tryptophan and 1 nM
synthetase. C, tRNATrp incubated for 10 min in
the presence of 1.8 µM D-tryptophan and 300 nM synthetase. The minor peak at 46 min corresponds to
non-aminoacylatable material in the sample. D,
D-Trp-tRNATrp (0.3 µM) before
incubation. E, D-Trp-tRNATrp
incubated for 5 min in the presence of 200 pM
D-tyrosyl-tRNATyr deacylase. F,
D-Trp-tRNATrp incubated for 5 min in the
presence of 30 nM deacylase.
|
|
When D-tryptophan was used as substrate, a peak migrating
at 68 min was also obtained on the chromatogram (Fig. 2C).
The area of this peak was used to calculate the concentration of the
D-Trp-tRNATrp product. Initial rate of
D-Trp-tRNATrp formation in the presence of 1.8 µM D-tryptophan and 0.35 µM tRNATrp was 1.1 × 103
s1. In the presence of excess enzyme, maximum
esterification of D-tryptophan to tRNA reached 0.18 µM. Since the optical purity of the D-amino
acid sample is >98%, this value guaranteed that more than 80% of the
esterified amino acid corresponded to the D-stereoisomer.
Under the same assay conditions, initial rate of tRNA aminoacylation
with L-tryptophan was 0.16 s1
(Table III).
D-Trp-tRNATrp Is Hydrolyzed by
D-Tyr-tRNATyr Deacylase--
As explained
above, the enhanced sensitivity of a dtd strain to the
presence of D-tryptophan suggested that
D-Tyr-tRNATyr deacylase can hydrolyze
D-Trp-tRNATrp. To explore this possibility,
D-Trp-tRNATrp was purified by C4
HPLC and assayed as a substrate of the deacylase. After incubation in
the presence of the enzyme, aminoacylated and non-aminoacylated tRNAs
were separated by HPLC and quantitated. Typical chromatograms are shown
in Fig. 2 (D-F). Incubation for 5 min in the presence of a
200 pM enzyme concentration resulted in the hydrolysis of
28% of the initially added D-Trp-tRNATrp (0.3 µM) (Fig. 2E). At 30 nM enzyme
(Fig. 2F), deacylation was practically complete (91%) in 5 min. When the incubation was carried out without enzyme, only 10% of
the substrate were deacylated. From such experiments, it could be
concluded that D-Tyr-tRNATyr deacylase promotes
the hydrolysis of D-Trp-tRNATrp, with a
catalytic efficiency
(kcat/Km) equal to 2.8 µM1
s1 (Table III). Calculated rate of
spontaneous chemical hydrolysis was 3 × 104 s1.
The deacylase did not hydrolyze L-Trp-tRNATrp
at a significant rate
(kcat/Km<103
µM1
s1) (Table III). The different behaviors of
the deacylase in the presence of D- or
L-Trp-tRNATrp further show that the tRNA
aminoacylation reaction obtained with D-tryptophan did not
result from an L-tryptophan contamination in the
D-amino acid sample.
E. coli Aspartyl-tRNA Synthetase Catalyzes the Aminoacylation of
tRNAAsp with D-Aspartic Acid--
The
formation of D-Asp-tRNAAsp was investigated in
the presence of 1 µM [3H]-radiolabeled
D-aspartic acid (500 Ci/mol), 5 µM purified
tRNAAsp, and purified aspartyl-tRNA synthetase. In these
conditions, the kinetics of incorporation of radioactivity into tRNA
were biphasic. First, 1.4% of total 3H in the sample was
rapidly transferred onto tRNA. Then, incorporation of 3H
continued at a much slower rate. The first part of the kinetic was
likely to correspond to the rapid esterification of tRNAAsp
by L-aspartic acid contaminating the D-aspartic
acid. The second phase was attributed to the formation of
D-Asp-tRNAAsp. A 1.4% contamination was
compatible with the given optical purity of
D-[3H]aspartic acid (>97%). In agreement
with this hypothesis, only the first phase of the kinetics was still
observed if 150 nM E. coli
D-Tyr-tRNATyr deacylase was added to the tRNA
aminoacylation assay mixture. Indeed, since the deacylase is expected
to hydrolyze specifically D-Asp-tRNAAsp, its
addition in the aminoacylation assay prevents the accumulation of the
D-Asp-tRNAAsp product.
From the above experiments, we measured an initial rate of
esterification of tRNAAsp (5 µM) by
D-Asp (1 µM) equal to 0.8 × 105 s1. Under the
same experimental conditions, L-[3H]aspartic
acid was transferred onto tRNAAsp at an initial rate of
3.3 × 102 s1
(Table III).
D-Asp-tRNAAsp Is Hydrolyzed by
D-Tyr-tRNATyr Deacylase--
To measure the
initial rate of D-Asp-tRNA hydrolysis by the deacylase,
D-[3H]Asp-tRNA was prepared and assayed as a
substrate of this enzyme. As explained under "Materials and
Methods," care was taken to minimize contamination of
D-[3H]Asp-tRNAAsp by
L-[3H]Asp-tRNAAsp.
A value of 12 µM1
s1 was obtained for the catalytic efficiency
(kcat/Km) of
D-Asp-tRNAAsp hydrolysis by the deacylase
(Table III). This value is 4-fold greater than the catalytic efficiency
measured with the D-Trp-tRNATrp substrate (2.8 µM1
s1), and 2-fold higher than that obtained in
the presence of D-Tyr-tRNATyr (6 µM1
s1) (29).
L-Asp-tRNAAsp was not a substrate of
D-Tyr-tRNATyr deacylase
(kcat/Km<103
µM1
s1). In the incubation mixture, spontaneous
chemical hydrolysis of either D- or
L-Asp-tRNAAsp occurred at a same rate equal to
2 × 104
s1.
Effect of D-Amino Acids on the Growth of S. cerevisiae
DTD1 or dtd1 Strains--
In S. cerevisiae, the
DTD1 gene product is structurally and functionally
homologous to E. coli D-Tyr-tRNATyr
deacylase (14). Disruption of the DTD1 gene markedly
increases the sensitivity of the growth of yeast to the presence of
external D-tyrosine.
To assess the specificity of the yeast
D-Tyr-tRNATyr deacylase, the effects of 19 D-amino acids on the growth of the S. cerevisiae DTD1 (DBY2057) and dtd1 (DBY2057DTD1) strains were
compared (Table IV). Because limiting
nitrogen conditions enhance the sensitivity of yeast growth to various
D-amino acids (37-39), L-proline was used as
nitrogen source in our experiments. Yeast cells were left to grow at
30 °C on minimal medium agar plates (yeast nitrogen base without
amino acids and ammonium sulfate, from Difco) supplemented with uracil,
glucose, L-proline, and various concentrations of D-amino acids (0-10 mM).
View this table:
[in this window]
[in a new window]
|
Table IV
Effect of various D-amino acids on the growth of S. cerevisiae strains DBY2057 and DBY2057DTD1
Cells were grown at 30 °C on minimal medium agar plates supplemented
with uracil, glucose, L-proline, and D-amino
acids at different concentrations. Explored D-amino acid
concentrations were 0.01, 0.03, 0.1, 0.3, 1, 3, 5, and 10 mM. Because of its low solubility, D-tyrosine
could not be tested beyond 0.3 mM. Indicated in the table
are the minimal D-amino acid concentrations resulting in
the complete absence of growth after a 6-day incubation.
|
|
The addition of up to 10 mM D-Pro,
D-Asn, or D-Ile to the minimal medium did not
modify the growth of the cells. Little growth inhibition was observed
with D-Cys, D-Thr, D-Glu,
D-Asp, D-Val, D-Lys, and
D-Gln. The inhibition was the same with wild type or dtd1 cells (Table IV).
Cell growth was clearly sensitive to the presence of D-Ser,
D-Ala, D-Arg, D-Trp,
D-Met, D-Phe, and D-His, in order
of increasing toxicity. However, again, the DTD1 and
dtd1 strains were affected to similar extents.
Finally, upon addition of D-leucine to the growth medium, a
marked difference could be observed between the two strains. Although growth of the wild-type strain was insensitive to the presence of up to
10 mM D-leucine, that of the dtd1
strain was stopped by the addition of 0.3 mM
D-leucine.
As a control, the DTD1 and dtd1 strains were
also exposed to D-tyrosine. As shown in Table IV, full
inhibition of the growth of the dtd1 strain was reached
at 0.1 mM D-tyrosine. With the DTD1
strain, identical growth inhibition required 0.3 mM
D-amino acid.
| |
DISCUSSION |
This study enables us to draw two main conclusions. First,
aminoacyl-tRNA synthetases are less specific than generally believed, since D-amino acids succeed in being transferred onto
tRNAs. Second, the specificity of the previously characterized
D-tyrosyl-tRNATyr deacylase is much broader
than expected.
By analogy with the case of D-tyrosine, the screening for
D-amino acid toxicity, as shown here, suggests
aminoacylation by their cognate D-amino acid of
tRNATrp, tRNAAsp, tRNASer, and
tRNAGln in E. coli, and of tRNALeu
in yeast. In the cases of tRNATrp and tRNAAsp,
the transfer reaction could be demonstrated in vitro.
However, the measured rates of D-aminoacylation are slow.
They are smaller than the rates of L-aminoacylation by 1-3
orders of magnitude, depending on the considered aminoacyl-tRNA
synthetase. Such low rates are likely to reflect relatively high
Km and small kcat values
associated with D-amino acids in the formation of D-aminoacyl-adenylates, as well as possible impairments at
the level of the transfer reaction itself. On the other hand, the D-amino acids are not likely to affect tRNA binding.
D-Tryptophan and D-Phenylalanine--
The
observed toxic effects of D-amino acids on E. coli growth may result from several mechanisms in addition to a
mischarging of tRNA. For instance, the toxicity of
D-tryptophan toward E. coli growth is acute.
Possibly, part of this toxicity is contributed for by misincorporation
of D-tryptophan in the bacterium cell walls (23, 25, 26).
The mechanism of this incorporation does not imply tRNATrp
esterification. It is independent of the normal biosynthetic pathway
and apparently involves a penicillin-insensitive LD-transpeptidase enzyme (25, 26). The harmful effect of D-tryptophan
incorporation into cell walls is explained by an increased
susceptibility of modified peptidoglycan to the action of lytic
transglycosylases, the main autolytic enzymes in E. coli
(25). D-Tryptophan can also inhibit the formation of
lipoprotein-peptidoglycan linkage (23).
Similarly to D-tryptophan, D-phenylalanine
exerts toxicity through misincorporation in cell walls (25, 26). In our
experiments, the effect of D-phenylalanine does not depend
on the presence or absence of the dtd allele. However, this
D-amino acid was shown early to be esterified to tRNA by
phenylalanyl-tRNA synthetase in vitro (28). Hence, we
conclude that, even if Phe-tRNAPhe was produced in
vivo, the inhibitory effect of this molecule on E. coli
growth is negligible as compared with the perturbation of cell wall
synthesis induced by the D-amino acid.
D-Serine Toxicity--
The present study indicates
that the toxicity of D-serine is at least partially
contributed for by the formation of a D-Ser-tRNA. Nevertheless, D-serine may also behave as a powerful
feedback inhibitor of L-serine and pantothenate
biosyntheses (40). It is noteworthy that the bacterium is able to
counter-react against the in vivo accumulation of
D-serine through the action of D-serine deaminase. Moreover, in the presence of D-serine or of its
analog D-threonine, D-serine deaminase
expression is induced via a specific activator protein (dsdC
gene product) and the catabolite gene activator protein-cAMP
complex. Under conditions of full induction, the
D-serine deaminase activity becomes high enough to allow
growth in the presence of D-serine as the only nitrogen and
carbon source (40, 41). The efficiency of this enzyme may explain why
addition of millimolar concentrations of D-serine are
required to evidence toxicity against E. coli.
Other D-Amino Acids--
The screening also shows an
effect of relatively small concentrations of D-Glu,
D-Thr, D-Lys, D-Val, and
D-Cys on the growth of E. coli. The toxicity of
the above five D-amino acids may originate from mechanisms
that do not involve tRNAs. For instance, in Bacillus sphaericus, D-cysteine as well as
L-cysteine behave as substrates of cystathionine
 -synthase (42). As a result, excess D- or L-cysteine triggers consumption of homoserine at the
expense of threonine biosynthesis. Cysteine growth-inhibitory effects
on E. coli cells were also reported to be mediated by an
inhibition of threonine deaminase and the resulting starvation for
isoleucine, leucine, and valine (22).
Although the toxicity indices associated with D-Glu,
D-Thr, D-Lys, D-Val, and
D-Cys did not vary upon inactivation of the dtd
gene, the possibility that these D-amino acids can also be esterified to their corresponding tRNAs by E. coli
synthetases cannot be completely excluded. Indeed, as discussed above
in the case of D-phenylalanine, the inhibitory effect on
E. coli growth of a given D-aminoacyl-tRNA may
be dominated by other toxic mechanisms and, consequently, be not
detectable in our experiments. Another possibility would be that tRNAs
esterified with D-Glu, D-Thr, D-Lys, D-Val, and D-Cys are not
substrates of D-Tyr-tRNATyr deacylase.
Actually, D-valine binding to valyl-tRNA synthetase was
reported by Owens and Bell (43, 44). However, the same authors observed
stereospecificity in valyl-adenylate formation. Additional evidence
against an esterification of the D-enantiomer to
tRNAVal was gained by Calendar and Berg (28) and
Hélène et al. (45). At this stage, one must be
aware of the difficulty to conclusively measure very slow rates of
D-aminoacylation (103 to
104 s1) without
available radiolabeled D-amino acid. Moreover, a few aminoacyl-tRNA synthetases, including valyl-tRNA synthetase, have the
capacity to promote hydrolysis of their misproducts through efficient
proofreading mechanisms. Therefore, edition of
D-aminoacyl-tRNAs by the synthetases themselves might also
account for the difficulty to evidence the mischarged products. In the
case of E. coli lysyl-tRNA synthetase, non-radioactive
D-lysine was shown to significantly compete with
L-[14C]lysine incorporation onto tRNA.
Nevertheless, this inhibition can equally well be explained by
non-radioactive L-lysine contaminating the
D-lysine sample (46).
The screening enables us to also detect toxicity of D-Asn,
D-His, D-Ile, D-Leu, and
D-Met. D-Met can be suspected to be
incorporated in cell walls, like D-Phe (23, 25, 26). The
origin of the toxicity of D-histidine is unknown. However,
it should not involve D-His-tRNAHis formation
since histidyl-tRNA synthetase activity from Salmonella typhimurium was shown insensitive by close to 100% to the
addition of up to 5 mM concentration of the
D-enantiomer of histidine (47). In the cases of
D-isoleucine and D-leucine,
D-aminoacyl-adenylate formation by the corresponding
E. coli synthetases was <1%, as compared with the
reactions with the L-enantiomers (48). Moreover, E. coli isoleucyl-tRNA synthetase was shown to be specific for the
L-enantiomer in ultracentrifuge studies on the binding of aliphatic amino acids (49, 50).
Comparison of E. coli and S. cerevisiae Sensitivities to
D-Amino Acids--
As already shown (29), the toxicity of
D-tyrosine toward yeast depends on the presence of a
functional DTD1 allele. In the present study, the toxicity
of the other D-amino acids was explored and inactivation of
DTD1 was found to also exacerbate the sensitivity to
D-leucine. This behavior may reflect the capacity of
leucyl-tRNA synthetase to accept D-leucine as substrate.
However, the sensitivity of E. coli growth to
D-leucine did not vary upon disruption of the
dtd gene. Such a difference between the two cells suggests a
distinct specificity of the eubacterial leucyl-tRNA synthetase as
compared with that of the fungal one. In a same manner, variations between the specificities of the bacterial and fungal synthetases may
be invoked to explain the different behaviors of serine, tryptophan, glutamine, and aspartate in the absence of either the dtd or
the DTD1 genes.
However, differences between the two cells may result from many other
causes, as we shall now discuss. E. coli growth as well as
that of yeast were not affected by the presence of
D-proline. On the other hand, D-alanine and
D-arginine slowed down yeast growth, not that of E. coli. D-Asparagine and D-isoleucine
slightly affected E. coli growth but not that of yeast. Such
apparent discrepancies in behaviors between the two cells may reflect
the different metabolic pathways susceptible to play with
D-amino acids in each case. For instance, in E. coli, D-alanine and D-glutamate naturally occur as products of specific racemases and contribute to cell wall
biosynthesis. In E. coli K12 cells grown in minimal glucose medium, concentrations of these two D-amino acids are 0.5 and 1 mM, respectively (51). However, when added to the
growth medium at a 25 mM concentration,
D-glutamate slows down E. coli growth, while
D-alanine does not. The lower sensitivity to
D-alanine is possibly explained for by the occurrence of a
bacterial D-amino acid dehydrogenase, which favors the
transformation of amino acids with a hydrophobic character like
D-Ala, D-Met, or D-Phe
(52-54).
Transport systems of D-amino acids also are different in
E. coli and in yeast. In E. coli, the permease
produced by the aroP gene favors the import of aromatic
amino acids. Upon deprivation of L-amino acids in the
growth medium, this permease actively transports aromatic
D-amino acids (55). In yeast, the import of both
L- and D-amino acids involves the product of
GAP1, a general permease with a broad specificity (21,
56).
Finally, in the yeast cytoplasm, D-amino acids are
transformed by an -N-acetyltransferase (57).
N-Acetylated amino acids are then believed to counter-react
against further uptake of D-amino acids (21). The
specificity of this acetylase is broad. However, its efficiency in the
transformation of D-Pro, D-Asp, and
D-Glu is relatively low (57, 58).
Broad Specificity of E. coli D-Tyr-tRNA
Deacylase--
The broad specificity of E. coli
D-Tyr-tRNA deacylase was early observed in vitro
by Calendar and Berg (28). According to the present study, the
deacylase recognizes very different D-aminoacyl moieties
like the acidic aspartate or the bulky aromatic tryptophan. However,
the deacylase appears not to simply behave as a D-amino acid esterase. Calendar and Berg showed that
D-Tyr-adenosine was not hydrolyzed by the deacylase (28).
In contrast, a D-Tyr-esterified oligonucleotide produced by
RNase T1 digestion was a substrate. In the 19-mer oligonucleotide
produced from tRNATyr, the deacylase possibly recognizes a
part of the acceptor stem of tRNA, for instance the 5'-CCA-3' triplet
common to all tRNAs. On the side of the amino acid moiety, the
deacylase would only distinguish the stereoisomeric character of the
C. In agreement with this idea, glycyl-tRNA obtained in
vitro was a substrate of the E. coli deacylase
(28).
General Remarks--
As already discussed (29),
dtd-like genes are widely distributed in the living world
and deacylase homologs are expected to occur in many bacteria, as well
as in yeasts, nematodes, higher plants, mouse, and man. The only
exceptions are archaebacteria, parasitic auxotrophic bacteria, and
prototrophic cyanobacteria.
In the case of auxotrophic bacteria, the lack of many biosynthetic
pathways possibly results in especially low levels of endogenous D-amino acids. As a consequence, deacylase activity would
be useless in such bacteria. In other cells,
D-aminoacyl-tRNA deacylase activity would be necessary to
counter-react against the harmful effect of D-amino acid
transfer onto tRNA. The set of incorporated D-amino acids
may vary from one type of cell to another. However, in each cell, one
single species of deacylase should be enough to hydrolyze any
D-aminoacyl-tRNA molecule.
A question to eventually address is why cells have maintained
dtd-like genes rather than evolved through the selection of more specific aminoacyl-tRNA synthetases. Possibly, the primitive cell
indifferently incorporated L- and D-amino acids
into polypeptides. At this stage, acquisition of a general
D-aminoacyl-tRNA deacylase activity could have helped the
cell to shift protein synthesis in the L-amino acid world.
According to this scenario, further evolution of cells implies an
improvement of the specificity of the synthetases toward
L-amino acids, and the progressive loss of a no more useful
deacylase gene. Cyanobacteria and archaebacteria may correspond to such cells.
| |
ACKNOWLEDGEMENTS |
We are grateful to Dr. J. Chen and Dr.
A. Brevet for the construction of the tRNAAsp gene, to
Prof. C. Yanofsky for the gift of plasmid pCH17, and to Dr. G. Eriani
for providing us with plasmid pBSaspS.
| |
FOOTNOTES |
*
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
To whom correspondence should be addressed. Tel.:
33-1-69-33-41-81; Fax: 33-1-69-33-30-13; E-mail:
plateau@coli.polytechnique.fr.
Published, JBC Papers in Press, July 28, 2000, DOI 10.1074/jbc.M005166200
| |
ABBREVIATIONS |
The abbreviation used is:
HPLC, high performance
liquid chromatography.
| |
REFERENCES |
| 1.
|
Friedman, M.
(1999)
J. Agric. Food Chem.
47,
3457-3479
|
| 2.
|
Shoji, J.,
and Kato, T.
(1976)
J. Antibiot.
29,
1275-1280
|
| 3.
|
Peypoux, F.,
Marion, D.,
Maget-Dana, R.,
Ptak, M.,
Das, B.,
and Michel, G.
(1985)
Eur. J. Biochem.
153,
335-340
|
| 4.
|
Sivonen, K.,
Namikoshi, M.,
Evans, W.,
Fardig, M.,
Carmichael, W.,
and Rinehart, K.
(1992)
Chem. Res. Toxicol.
5,
464-469
|
| 5.
|
Jack, R.,
and Jung, G.
(1998)
Chimia
52,
48-55
|
| 6.
|
Man, E.,
Fisher, G.,
Payan, I.,
Cadilla-Perezrios, R.,
Garcia, N.,
Chemburkar, R.,
Arends, G.,
and Frey, W., II
(1987)
J. Neurochem.
48,
510-515
|
| 7.
|
Fisher, G.,
D'Aniello, A.,
Vetere, A.,
Cusano, G.,
Chavez, M.,
and Petrucelli, L.
(1992)
Neurosci. Lett.
143,
215-218
|
| 8.
|
Schell, M.,
Cooper, O.,
and Snyder, S.
(1997)
Proc. Natl. Acad. Sci. U. S. A.
94,
2013-2018
|
| 9.
|
Wolosker, H.,
Blackshaw, S.,
and Snyder, S.
(1999)
Proc. Natl. Acad. Sci. U. S. A.
96,
13409-13414
|
| 10.
|
Wolosker, H.,
Sheth, K.,
Takahashi, M.,
Mothet, J.,
Brady, R., Jr.,
Ferris, C.,
and Snyder, S.
(1999)
Proc. Natl. Acad. Sci. U. S. A.
96,
721-725
|
| 11.
|
Miles, E. W.,
Phillips, R. S.,
Yeh, H. J. C.,
and Cohen, L. A.
(1986)
Biochemistry
25,
4240-4249
|
| 12.
|
Ahmed, S. A.,
Martin, B.,
and Miles, E. W.
(1986)
Biochemistry
25,
4233-4240
|
| 13.
|
Williams, J. S.,
and Rosevear, P. R.
(1991)
Biochemistry
30,
6412-6416
|
| 14.
|
Soutourina, J.,
Blanquet, S.,
and Plateau, P.
(2000)
J. Biol. Chem.
275,
11626-11630
|
| 15.
|
Kumagai, H.,
Kashima, N.,
and Yamada, H.
(1970)
Biochem. Biophys. Res. Commmun.
39,
796-801
|
| 16.
|
Metzler, D. E.,
Ihawa, M.,
and Snell, E. E.
(1954)
J. Am. Chem. Soc.
76,
648-652
|
| 17.
|
Ohnishi, E.,
Macleod, H.,
and Horowitz, N. H.
(1962)
J. Biol. Chem.
237,
138-142
|
| 18.
|
Champney, W. S.,
and Jensen, R. A.
(1970)
J. Bacteriol.
104,
107-116
|
| 19.
|
Jensen, R. A.,
Stenmark, S. L.,
and Champney, W. S.
(1972)
Arch. Mikrobiol.
87,
173-180
|
| 20.
|
Cosloy, S. D.,
and McFall, E.
(1973)
J. Bacteriol.
114,
685-694
|
| 21.
|
Rytka, J.
(1975)
J. Bacteriol.
121,
562-570
|
| 22.
|
Harris, C. L.
(1981)
J. Bacteriol.
145,
1031-1035
|
| 23.
|
Tsuruoka, T.,
Tamura, A.,
Miyata, A.,
Takei, T.,
Iwamatsu, K.,
Inouye, S.,
and Matsuhashi, M.
(1984)
J. Bacteriol.
160,
889-894
|
| 24.
|
Friedman, M.
(1991)
Adv. Exp. Med. Biol.
289,
447-481
|
| 25.
|
Caparros, M.,
Torrecuadrada, J. L. M.,
and de Pedro, M. A.
(1991)
Res. Microbiol.
142,
345-350
|
| 26.
|
Capparos, M.,
Pisabarro, A. G.,
and de Pedro, M. A.
(1992)
J. Bacteriol.
174,
5549-5559
|
| 27.
|
Calendar, R.,
and Berg, P.
(1966)
Biochemistry
5,
1690-1695
|
| 28.
|
Calendar, R.,
and Berg, P.
(1967)
J. Mol. Biol.
26,
39-54
|
| 29.
|
Soutourina, J.,
Plateau, P.,
Delort, F.,
Peirotes, A.,
and Blanquet, S.
(1999)
J. Biol. Chem.
274,
19109-19114
|
| 30.
|
Miller, J. H.
(1992)
A Short Course in Bacterial Genetics
, Cold Spring Harbor Laboratory, NY
|
| 31.
|
Eriani, G.,
Dirheimer, G.,
and Gangloff, J.
(1990)
Nucleic Acids Res.
18,
7109-7118 | |
TOP
ABSTRACT
INTRODUCTION