http://www.pubmedcentral.gov/articlerender.fcgi?tool=pubmed&pubmedid=4562386

Repression of Enzymes of Arginine Biosynthesis by l-Canavanine in
Arginyl-Transfer Ribonucleic Acid Synthetase Mutants of Escherichia
coli
Ronald Faanes and Palmer Rogers

Department of Microbiology, University of Minnesota, Minneapolis,
Minnesota 55455

This article has been cited by other articles in PMC.
Abstract
We show that the arginine analogue, l-canavanine, repressed the
accumulation of translatable messenger ribonucleic acid (RNA) for three
arginine biosynthetic enzymes in Escherichia coli. The method used to
determine the level of translatable messenger RNA depended upon
measurement of a burst of enzyme synthesis as described previously. E.
coli strains with defective arginyltransfer ribonucleic acid (tRNA)
synthetase (argS mutants) were insensitive to canavanine repression.
When deprived of leucine, a leu argS strain regained normal sensitivity
to canavanine repression. The level of in vivo canavanyl-tRNAarg was
determined for a normal strain and an argS mutant. After 20 min of
growth with canavanine only 9% of tRNAarg from the argS strain was
protected from periodate oxidation, while 42% of the tRNAarg from an
argS+ strain was charged. When deprived of leucine, leu argS or leu
argS+ strains grown with canavanine contained more than 60% charged
tRNAarg. Reverse phase column chromatography of periodate-oxidized tRNA
from canavanine-grown argS and argS+ strains showed no preferential
charging of any isoaccepting species of tRNAarg. Therefore, we failed
to detect a specific arginyl-tRNA species that might be involved in
repression by canavanine. However, the data suggest that canavanine
repression of the arginine pathway occurs only when high levels of
canavanyl-tRNA are present, and thus support the notion that
arginyl-tRNA synthetase plays a role in generating a repression signal.
Full text

RESULTS
Canavanine repression of the arginine
biosynthetic enzymes. After a short period of
arginine deprivation, addition of arginine
elicits a burst of omithine carbamoyltransferase
synthesis lasting 2 to 4 min. We have
shown that this burst of enzyme synthesis
is a measure of translatable messenger RNA
for this enzyme in E. coli (12). Arginine-deprived
E. coli cells exhibited a similar burst of
acetylornithine deacetylase and arginosuccinate
lyase synthesis upon addition of L-arginine
as shown in Fig. 1. As found previously
for ornithine carbamoyltransferase, the burst
of synthesis for these enzymes lasted 2 to 4
min owing to repression by L-arginine, and no
burst was observed when L-arginine was added
at 0 min. We studied repression of the burst of
ornithine carbamoyltransferase synthesis by
adding a number of analogues of arginine to E.
coli strain 254 cultures at 0 min followed by
arginine 10 min later (Table 2). Only L-canavanine
effectively prevented the build-up of
the capacity to form the burst of enzyme synthesis.
Thus, apparently L-canavanine represses
accumulation of translatable messenger
RNA for this enzyme (12) and similarly
effects acetylornithine deacetylase and arginosuccinate
lyase synthesis as well (Table 3).
Loss of canavanine repression in arginyltRNA
synthetase mutants. When argS mutants
were transferred from AF medium containing
arginine to AF medium containing Lcanavanine
(200 ,g/ml) and incubated for 20
min, a burst of ornithine carbamoyltransferase
synthesis was observed upon addition of L-arginine
that was the same as that observed after5003) were reported by
Hirshfield and Bloemers
(14). The Km values for E. coli strains 254
and 90-9 (Table 4) were determined on partially
purified nuclease-free synthetase preparations
(Fig. 2). Since L-canavanine can be activated
and esterified to tRNA (see below and

Our findings show that no one species of
tRNAarg is selectively left uncharged by canavanine
in argS strain 90-9 even when only 9%
of the total tRNAarg is protected (Fig. 5). Celis
and Maas (4) reported similar results upon
chromatographic comparison of arginyl-tRNA
from wild-type and argS strains grown with
and without arginine. Most likely then, neither
canavanine nor arginine repression is caused
by a peculiar species of tRNAar .
The observed recovery of canavanine repression
by leucine-deprived argS mutants (Table
5) coupled with the observed rise of the in vivo
levels of canavanyl-tRNAarg (Table 8) suggest
that the repression signal is in some manner
related to the concentration of canavanyltRNAarg
or some other intermediate involving
the arginyl-tRNA synthetase. During leucine
deprivation, the level of translatable messenger
RNA for ornithine carbamoyltransferase
decayed 10 min after canavanine was
added in both argS+ and argS E. coli (Tables 5
and 6). Also, 2 gsg of arginine promoted the
same rate of decay of translatable message inboth strains under these
conditions (Fig. 3).
These results show that the repression system
can function normally in argS mutants when
protein synthesis is reduced below the point
where arginyl-tRNA synthetase is limiting.
Probably all experimental values for in vivo
charged tRNA must be viewed with caution,
simply because the pool of charged tRNA may
very well change radically during manipulation
of cells prior to inactivation of cellular enzymes.
In specific cases, unexpectedly high or
low levels of in vivo charged tRNA were obtained
from cultures chilled before harvesting
when compared to more reasonable levels obtained
from cultures first treated with 5%
trichloroacetic acid before manipulation (Neidhardt,
personal communication). In our experiments,
cultures were simply centrifuged for 5
min without chilling, the pellets were rapidly
suspended in acetate buffer containing detergent,
and the cells were lysed by freezing and
thawing. Although we did not test the acid fixation
method of Neidhardt, our data on levels
of in vivo charged tRNAarg (Tables 7 and 8)
appear to follow the pattem expected from the
properties of the arginyl-tRNA synthetases of
the strains used (Table 4) and the conditions
employed.
Since uptake of canavanine into E. coli is
about one-tenth the uptake of arginine (reference
41 and unpublished data), part of the requirement
for a high exogenous concentration
of canavanine for repression may be due to
this fact. However, it is unlikely that the free
canavanine pool is responsible for repression of
translatable messenger RNA, since, when protein
synthesis is permitted, canavanine (200
,ug/ml) represses argS+ cells but does not repress
argS cells. In this situation, argS bacteria
probably contain an even higher pool of free
canavanine, since conversion of canavanine to
canavanyl-tRNAarg and thence to protein is
apparently limited at the synthetase step
(Table 7).
Recently, we reported that L-arginine repressed
the level of hybridizable argECBH
messenger RNA in E. coli to an amount 7- to
24-fold lower than that found in derepressed
cells (30). Our data indicate that arginine acts
at the transcriptional level (19). However, we
found that added canavanine (200 ug/ml) had
little or no effect on reducing-the level of hybridizable
argECBH messenger RNA in either
argS or argS+ strains (30). The repression of
the level of translatable messenger RNA for
the arginine enzymes by canavanine shown in
this report appears to contradict our hybridizablehybridizable
data. Also, as observed by others (5, 15),
arginine apparently signals repression normally
in all argS strains tested so far, even
when the level of charged tRNAarg remains
constant at 20% with or without added arginine
in the growth medium. However, these
discrepancies would be predicted by a dual
model in which both canavanine and arginine
exert control upon the translation of arginine
messenger RNA, mediated by some function of
arginyl-tRNA synthetase, while arginine but
not canavanine represses transcription of new
messenger RNA. Lavelle presented evidence
suggesting that the arginine regulon (20) and
the tryptophan operon (21) are under dual control
by both a transcriptional and a translational
mechanism. Indeed, Vogel et al. (38)
have reviewed the indirect evidence that led
them to propose a translational control model
for the arginine regulon, that they suggest may
operate in conjunction with transcriptional
control.

Tim, thanks for answering, however, at this point I need to study more
in order to understand this complex issues.