FEMS Microbiology Letters 163 (1998) 143^148
In vivo e¡ects of anti-inducers of the cysteine regulon in
Salmonella typhimurium
Oscar J. Oppezzo *
Comisioèn Nacional de Energ|èa Atoèmica, Departamento de Radiobiolog|èa, Av. del Libertador 8250, 1429 Buenos Aires, Argentina
Received 10 March 1998; revised 7 April 1998; accepted 14 April 1998
Abstract
Growth on readily utilizable sulfur sources reduces expression of the cysteine regulon in Salmonella typhimurium. Inhibition
of serine transacetylase by cysteine and direct actions of the anti-inducers sulfide and thiosulfate are responsible for reduction
of expression. In order to evaluate individual contributions of each mechanism, the inhibitory effects of Na2 S and Na2 S2 O3
were studied in strains with or without the capacity to synthesize cysteine from these compounds, using a transcriptional fusion
to the cysDNC operon. In a cysK cysM strain, although cysteine synthesis from sulfide and thiosulfate was blocked, Na2 S and
Na2 S2 O3 efficiently reduced expression of the cysDNC operon. The inhibitory effect observed in this mutant was equivalent to
70^100% of that found in a strain carrying the fusion in a wild-type context grown in the same conditions. The actions of
sulfide and thiosulfate as anti-inducers seem therefore to be responsible for most of the reduction of expression caused by these
agents in vivo. z 1998 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights
reserved.
Keywords : Salmonella typhimurium; Cysteine biosynthesis; Cysteine regulon
1. Introduction
The synthesis of cysteine in Salmonella typhimurium (Fig. 1) is achieved by reaction of sul¢de with
O-acetyl-L-serine (OAS), or by reaction of thiosulfate
with OAS to yield S-sulfocysteine which is then converted to cysteine [1]. The enzyme serine transacetylase, encoded by the gene cysE, catalyzes acetylation
of serine to form OAS, and cysteine is an e¤cient
inhibitor of this enzyme [2]. OAS is also converted to
N-acetyl-L-serine (NAS) by a spontaneous O- to Nacetyl migration [3]. Most of the genes involved in
* Corresponding author. Tel.: +54 (1) 7041107;
Fax: +54 (1) 7041187; E-mail: [email protected]
cysteine biosynthesis are regulated at the level of
transcription, and their expression requires NAS as
an internal inducer and the transcriptional activator
protein CysB [1,3,4]. Thus, concentrations of most
enzymes required for cysteine biosynthesis are functions of OAS concentration, and can be down-regulated by cysteine [1,3]. High level expression of the
cysteine biosynthetic pathway also requires sulfur
limitation [3,5]. In vivo experiments have shown
that 1 mM sul¢de or 0.5 mM cystine almost completely abolishes induction of the regulon by exogenous OAS [5] or NAS [6]. The e¡ect of sul¢de in
these experiments has been clari¢ed by in vitro experiments which demonstrated that sul¢de and thiosulfate are anti-inducers. These compounds reverse
0378-1097 / 98 / $19.00 ß 1998 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.
PII: S 0 3 7 8 - 1 0 9 7 ( 9 8 ) 0 0 1 6 4 - 5
FEMSLE 8174 11-6-98
144
O.J. Oppezzo / FEMS Microbiology Letters 163 (1998) 143^148
the e¡ect of the inducer on binding of CysB to several promoters of the cysteine regulon and also reduce the e¡ect of the inducer on transcription initiation from these promoters [6,7]. Cystine has no antiinducer e¡ects in vitro [6] and its ability to prevent
the action of exogenous inducers in vivo is believed
to be mediated by sul¢de generated by the inducible
enzyme cysteine desulfhydrase [3], which degrades
cysteine to pyruvate, ammonia and sul¢de [8].
In vitro experiments did not reveal the physiologic
signi¢cance of the competitive interaction of inducer
and anti-inducer with CysB protein. This subject is
also di¤cult to evaluate from the results of the in
vivo experiments mentioned above, because these assays were performed at sul¢de and cystine concentrations 10 times greater than that required for
growth of a wild-type strain of S. typhimurium in
liquid medium (estimated to be 70 WM [1]). It is
considered that both feedback inhibition of serine
transacetylase by cysteine and the direct action of
anti-inducers are involved in repression of the cysteine regulon by growth on readily utilizable sulfur
sources such as sul¢de or cysteine [3]. Nevertheless,
the contribution of each mechanism to regulation
has not yet been established. The aim of the present
study was to estimate these contributions. The in£uence of di¡erent sulfur compounds on the expression
of the regulon was studied in a mutant strain (cysK
cysM) unable to produce cysteine by reaction of
OAS with sul¢de or thiosulfate. In this mutant, feedback inhibition of serine transacetylase is not possible when thiosulfate, sul¢de or its precursors are
used as the sulfur source. The repression observed
was compared with that observed in a strain capable
of performing cysteine biosynthesis from sul¢de and
thiosulfate. Expression of the regulon was measured
using a transcriptional fusion to the cysDNC operon.
This is a suitable tool, since CysB and the inducer
regulate transcription of this operon [3] and the
products of cysD, cysN and cysC (ATP sulfurylase
and adenosine 5P-phosphosulfate kinase), required
for the sulfate activation, are involved neither in uptake of sul¢de, thiosulfate or cysteine, nor in reactions related to the use of these compounds [1,3]
(Fig. 1).
2. Materials and methods
2.1. Bacterial strains and media
Strains used in this study were derivatives of S.
typhimurium LT2. Their origins and relevant genetic
properties are shown in Table 1. Transducing bacter-
Fig. 1. Pathway of cysteine biosynthesis in S. typhimurium. The genes encoding the involved enzymes are indicated for each reaction. Except for cysE and cysG, expression of these genes requires CysB protein and acetylserine [3]. Serine transacetylase is inhibited by cysteine
[2]. Cysteine desulfhydrase is induced by cysteine [5].
FEMSLE 8174 11-6-98
O.J. Oppezzo / FEMS Microbiology Letters 163 (1998) 143^148
iophage P22 HT105/1 int-201 was used for all transductional crosses. Minimal medium was modi¢ed
Vogel-Bonner medium E, in which MgSO4 was replaced by an equimolar amount of MgCl2 , with 0.5%
glucose as the carbon source. It was prepared as
described by Kredich [5], except that pH adjustment
with HCl was omitted. This medium was supplemented with di¡erent sulfur sources: 1 mM L-djenkolic acid, 1 mM Na2 SO4 , 1 mM Na2 S2 O3 , 1 mM
Na2 S, 1 mM Na2 SO3 , 0.5 mM L-cysteine, or 1 mM
L-djenkolic acid in addition to Na2 S, Na2 S2 O3 , or Lcysteine. In the latter case, both L-djenkolic acid and
the additional sulfur compound were added at 1 mM
and the concentration of Na2 S, Na2 S2 O3 , or L-cysteine was adjusted by suitable dilution with medium
containing 1 mM L-djenkolic acid. Arginine and histidine were added at 40 Wg ml31 . Complete medium
was LB [9] with the addition of ampicillin 30 Wg
ml31 , kanamycin 20 Wg ml31 , tetracycline 50 Wg
ml31 (with MgCl2 1 mM), 5-bromo-4-chloro-3-indolyl-L-D-galactopyranoside (X-gal) 25 Wg ml31 , and/or
OAS 1 mM when required. Solid media were prepared by the addition of agar to a ¢nal concentration of 1.5%.
2.2. Strain construction
A transcriptional fusion to the cysDNC operon
was obtained by the method described by Hughes
and Roth [10]. The cysE2 mutation was introduced
into strain TT7610 by transduction with phage
grown on strain DA1716, using resistance to kana-
145
mycin linked to cysE for positive selection. The
strain obtained was transduced with phage grown
on strain TT7674, which carries an insertion of Mu
d1-8 (Tpn[Am] Amp Lac c62ts) [10]. Ampicillin-resistant transductants from this cross were selected
and screened for increased production of L-galactosidase in the presence of exogenous OAS, by replica
on plates containing ampicillin and X-gal with or
without OAS. A colony showing increased production of L-galactosidase in the presence of exogenous
OAS was found within about 1.5U104 ampicillin-resistant transductants.
The isolated Mu d1-8 insertion was transferred by
transduction to strain DA1468 (sup‡ cys‡ ). The
strain obtained was incapable of growing with
Na2 SO4 as the sole sulfur source, although it was
capable of growing in the presence of Na2 S2 O3 ,
Na2 SO3 , Na2 S, or L-cysteine. This result showed
that the conversion of sulfate to sul¢te, catalyzed
by the products of the genes cysC, cysN, cysD and
cysH [1] (Fig. 1), was impaired. Cotransduction of
the Mu d1-8 gene responsible for ampicillin resistance with markers linked to these loci was studied.
The strain carrying the fusion was transduced with
phage grown on strain TT173 (cysC1511: :Tn10) and
95% of the tetracycline-resistant transductants obtained lost resistance to ampicillin. When the marker
zga-7119: :Tn10v16v17, linked 60% with cysH363,
50% with cysI23, and 15% with cysD125 [11], was
introduced by transduction with phage grown on
strain DA1640, 14% of the tetracycline-resistant
transductants lost resistance to ampicillin. These re-
Table 1
Bacterial strains
Strain
Pertinent genotype
Source or reference
NK186
TT173
TT7610
TT7674
TT10286
DA1468
DA1640
DA1642
DA1716
DA1889
DA1889a
DA1889b
cysA1637 : :Tn10
cysC1511 : :Tn10
supD10 zeb609: :Tn10
pncA212 : :Mud1-8
hisD9953: :MudJ
argC95
cysD125 zga-7119: :Tn10v16v17 argC95
cysK1772 cysM1770 zej-7118: :Tn10v16v17 argC95
cysE2 zhj-7122: :Tn10v16v17 argC95
cys-1865 : :MudJ argC95
cys-1865 : :MudJ cysA1637: :Tn10 argC95
cys-1865 : :MudJ cysK1772 cysM1770 zej-7118: :Tn10v16v17 argC95
Laboratory collection
J.R. Roth via SGSCa
[10]
[10]
[13]
[12]
[12]
[12]
Laboratory collection
This study
This study
This study
a
SGSC, Salmonella Genetic Stock Center.
FEMSLE 8174 11-6-98
146
O.J. Oppezzo / FEMS Microbiology Letters 163 (1998) 143^148
sults indicate that insertion of Mu d1-8 took place in
cysC, cysD or cysN rather than in cysH. Variation in
the level of L-galactosidase activity in strains carrying this fusion re£ects the level of transcription from
the promoter of the cysDNC operon.
For further stabilization of the insertion, the strain
derived from DA1468 carrying the fusion was transduced using phage grown on strain TT10286, which
carries MudJ (the MudI1734 (Kan, Lac) element)
[12,13]. A kanamycin-resistant histidine-prototroph
transductant from this cross, which lost ampicillin
resistance, was obtained and designated DA1889.
This transductant was a recombinant whose Mu
d1-8 has inherited the transposition defect and kanamycin resistance determinant of MudJ by recombination events between the lac and Mu c regions at
the ends of the Mud prophages. The method used
was described by Hughes and Roth [12] in S. typhimurium and by Castilho et al. [13] in Escherichia coli.
The fusion, designated cys-1865: :MudJ, was introduced into two strains derived from DA1468. One
of them, carrying cysA1367: :Tn10, was obtained by
transduction of DA1468 with phage grown on strain
NK186 and selection by tetracycline resistance. Construction of the other strain, carrying cysK1772 and
cysM1770 (DA1642), has been described elsewhere
[11]. Both derivatives of DA1468 were used as recipient strains in transductions with phage grown on
strain DA1889. Using kanamycin resistance, a transductant carrying cys-1865: :MudJ was selected from
each cross. The strains obtained from these crosses
(DA1889a and DA1889b) and DA1889 were used in
the enzyme assays described below to study the expression of the cysDNC operon in di¡erent genetic
backgrounds.
incubation at 37³C with shaking during 4 h, L-galactosidase activity was measured as described by
Miller [14].
3. Results and discussion
3.1. E¡ect of di¡erent sulfur sources on the expression
of the cysDNC operon
A transcriptional fusion to the cysDNC operon
was introduced in a cysteine prototroph and in two
mutant strains. One of them (cysK cysM) was unable
to synthesize cysteine using sul¢de or thiosulfate. In
the other mutant (cysA) the sulfate and thiosulfate
uptake was impaired. Using the fusion, expression of
the operon was measured in media containing increasing concentrations of Na2 S, L-cysteine or
Na2 S2 O3 . Comparison of the e¡ects of Na2 S and
Na2 S2 O3 in the wild-type context and the cysK
cysM mutant allowed for the evaluation of the con-
2.3. Growth conditions and enzyme assays
Overnight cultures of strains under study were
grown at 37³C with shaking in minimal medium
with L-djenkolic acid as the sole sulfur source. Aliquots of these cultures were inoculated in di¡erent
fresh media, containing an alternative sulfur source
in addition to L-djenkolic acid. Na2 S, L-cysteine and
Na2 S2 O3 , at several concentrations, were assayed as
alternative sulfur sources. Aliquots of the overnight
cultures were inoculated in medium containing Ldjenkolic acid alone, and used as a reference. After
Fig. 2. E¡ects of Na2 S and L-cysteine on the expression of a
transcriptional fusion to the cysDNC operon. Cells were grown
in minimal medium containing 1 mM L-djenkolic acid and increasing concentrations of: Na2 S (closed symbols, solid lines) or
L-cysteine (open symbols, dashed lines). Strains used were derivatives of DA1468 carrying cys-1865: :MudJ in addition to: none
(DA1889; squares), or cysK1772 cysM1770 (DA1889b; circles).
L-Galactosidase activities were measured, and the results are expressed as fractions of the values obtained for cells grown with
1 mM L-djenkolic as the sole sulfur source (1803 and 1230 Miller
units for the fusion in a wild-type context and in a cysK cysM
mutant, respectively).
FEMSLE 8174 11-6-98
O.J. Oppezzo / FEMS Microbiology Letters 163 (1998) 143^148
tribution of cysteine accumulation to the regulation.
The e¡ect of Na2 S2 O3 on the cysA mutant showed
the in£uence of uptake on the repression produced
by this compound.
Results of L-galactosidase assays performed with
cells grown in the presence of Na2 S or L-cysteine are
shown in Fig. 2. At concentrations around 10 WM
both L-cysteine and Na2 S reduced the L-galactosidase activity to half of its maximum value, measured
in cells grown with L-djenkolic acid as the sole sulfur
source. The inhibitory e¡ect of L-cysteine on expression was similar in strains carrying either wild-type
or mutant alleles of cysK and cysM. At high concentrations, the e¡ect of Na2 S on the strain carrying
cysK1772 cysM1770 seemed slightly weaker than
that observed in the control strain.
When thiosulfate was introduced as an alternative
sulfur source, results shown in Fig. 3 were obtained.
L-Galactosidase activity was reduced by Na2 S2 O3 to
half of its maximum value at concentrations around
1 WM, not only in the strain carrying wild-type alleles
of cysK and cysM but also in the strain carrying
cysK1772 cysM1770. The in£uence of these mutations was not evident at concentrations lower than
10 WM. Unexpectedly, Na2 S2 O3 concentrations higher than 10 WM did not produce further reduction of
L-galactosidase activity in the strains under study.
Moreover, in the cysK cysM mutant the inhibition
produced by Na2 S2 O3 at concentrations above 1 WM
was 70^80% of that found in the control strain. The
reasons for the limited e¡ect of thiosulfate at high
concentrations are unknown. Probably, thiosulfate
uptake is limited when the intracellular concentration is increased, since it requires transcription of
the cysPTWA operon [1,3] and it has been shown
that thiosulfate e¤ciently inhibits transcription
from the cysP promoter in vitro [7]. Reduction of
L-galactosidase activity by Na2 S2 O3 was blocked in
the strain carrying cysA1637: :Tn10 (Fig. 3), illustrating the in£uence of thiosulfate transport on the inhibitory e¡ect of this agent. The signi¢cance of the
di¡erence between the minimum levels of expression
found in strains capable or not of synthesizing cysteine from thiosulfate is unclear. It has been shown
that thiosulfate is more potent than sul¢de as antiinducer of the cysteine regulon in vitro [7]. Comparison of Figs. 2 and 3 suggests that at concentrations
lower than 100 WM, Na2 S2 O3 is more e¡ective than
147
Fig. 3. E¡ect of Na2 S2 O3 on the expression of a transcriptional
fusion to the cysDNC operon. Cells were grown in minimal medium containing 1 mM L-djenkolic acid and increasing concentrations of Na2 S2 O3 . Strains used were derivatives of DA1468 carrying cys-1865: :MudJ in addition to: none (DA1889; squares),
cysK1772 cysM1770 (DA1889b; circles), or cysA1637 : :Tn10
(DA1889a; triangles). L-Galactosidase activities were measured,
and the results are expressed as fractions of the values obtained
for cells grown with 1 mM L-djenkolic as the sole sulfur source
(1948 Miller units for the fusion in the cysA mutant and the values indicated in Fig. 2 for the other strains).
Na2 S in reducing the expression of the regulon in
vivo. Except for Na2 S2 O3 acting on the cysA mutant,
the minimal concentrations of Na2 S2 O3 and Na2 S
that support the growth of a wild-type strain (10^
100 WM) produced signi¢cant repression. Thus, the
action of anti-inducers is probably useful to reduce
the expression of the regulon during rapid changes of
sulfur availability, as has been suggested [3].
The results obtained show that when sul¢de or
thiosulfate is available, the expression of the cysDNC
operon is reduced to a considerable extent, even if
the synthesis of cysteine from these compounds is
impaired. Since the direct action of sul¢de and thiosulfate as anti-inducers is independent of cysteine
synthesis and accumulation, these results may be explained assuming that anti-inducers are responsible
for most of the reduction of expression. Thus, the
contribution to repression of serine transacetylase
feedback inhibition would be relatively small.
Anti-inducers were e¡ective in vivo at low concentrations without inhibition of OAS synthesis. It
FEMSLE 8174 11-6-98
148
O.J. Oppezzo / FEMS Microbiology Letters 163 (1998) 143^148
might be proposed that feedback inhibition of serine
transacetylase by cysteine and direct action of antiinducers regulate the expression of the cysDNC operon in di¡erent situations. If sul¢de and/or thiosulfate are available, sulfate activation would not be
necessary, and anti-inducers would inhibit production of enzymes required for this process independently of the OAS concentration. This concentration
may £uctuate to supply substrate for the cysteine
synthesis, but an eventual rise in the level of OAS
would have no e¡ect on regulation. Conversely, if
sul¢de and thiosulfate are not available, expression
of the genes encoding the enzymes responsible for
sulfate activation is required, but it will be useful
only if OAS is available. Positive transcriptional control by CysB and NAS would provide information
on substrate concentration. Considering such a possibility, mechanisms involved in the reduction of the
expression of the cysteine regulon seem to be complementary rather than redundant.
Acknowledgments
I thank Ms. Susana N. Boretto for technical assistance and Drs. A.E. Schwint and R.A. Pizarro for
helping to correct the manuscript.
References
[1] Kredich, N.M. (1987) Biosynthesis of cysteine. In: Escherichia
coli and Salmonella typhimurium Cellular and Molecular Biology (Neidhardt, F.C., Ingraham, J.L., Low, B.K., Magasanik,
B., Schaechter, M. and Umbarger, H.E., Eds.), Vol. 1,
pp. 419^428. American Society for Microbiology, Washington
DC.
[2] Kredich, N.M. and Tomkins, G.M. (1966) The enzymic synthesis of L-cysteine in Escherichia coli and Salmonella typhimurium. J. Biol. Chem. 241, 4955^4965.
[3] Kredich, N.M. (1992) The molecular basis for positive regulation of cys promoters in Salmonella typhimurium and Escherichia coli. Mol. Microbiol. 6, 2747^2753.
[4] Lynch, A.S., Tyrrell, R., Smerdon, S.J., Briggs, G.S. and Wilkinson, A.J. (1994) Characterization of the CysB protein of
Klebsiella aerogenes: direct evidence that N-acetylserine rather
than O-acetylserine serves as the inducer of the cysteine regulon. Biochem. J. 299, 129^136.
[5] Kredich, N.M. (1971) Regulation of the L-cysteine biosynthesis in Salmonella typhimurium. J. Biol. Chem. 246, 3474^3484.
[6] Ostrowski, J. and Kredich, N.M. (1990) In vitro interactions
of CysB protein with cysJIH promoter of Salmonella typhimurium: inhibitory e¡ects of sul¢de. J. Bacteriol. 172, 779^785.
[7] Hryniewicz, M.N. and Kredich, N.M. (1991) The cysP promoter of Salmonella typhimurium: characterization of two
binding sites for CysB protein. Studies of in vivo transcription
initiation, and demonstration of the anti-inducer e¡ects of
thiosulfate. J. Bacteriol. 173, 5876^5886.
[8] Kredich, N.M., Keenan, B.S. and Foote, L.J. (1972) The puri¢cation and subunit structure of cysteine desulfhydrase from
Salmonella typhimurium. J. Biol. Chem. 247, 7157^7162.
[9] Davis, R.W., Botstein, D. and Roth J.R. (1980) Advanced
Bacterial Genetics. Cold Spring Harbor Laboratory Press,
Cold Spring Harbor, NY.
[10] Hughes, K.T. and Roth, J.R. (1984) Conditionally transposition-defective derivative of Mu d1(Amp Lac). J. Bacteriol.
159, 130^137.
[11] Oppezzo, O.J. and Antoèn, D.N. (1995) Involvement of cysB
and cysE genes in the sensitivity of Salmonella typhimurium to
mecillinam. J. Bacteriol. 177, 4524^4527.
[12] Hughes, K.T. and Roth, J.R. (1988) Transitory cis complementation : A method for providing transposition functions to
defective transposons. Genetics 119, 9^12.
[13] Castilho, B.A., Olfson, P. and Casadaban, M.J. (1984) Plasmid insertion mutagenesis and lac gene fusions with mini-Mu
bacteriophage transposons. J. Bacteriol. 158, 488^495.
[14] Miller, J.H. (1972) Experiments in Molecular Genetics. Cold
Spring Harbor Laboratory, Cold Spring Harbor, NY.
FEMSLE 8174 11-6-98