Department of Pharmacology, University of
Wisconsin Medical School, Madison, Wisconsin 53706
 |
INTRODUCTION |
The prevalence of His
Asp
two-component signal transduction relays and the diversity of important
pathways they regulate justify a search for effectors that act on such
systems. For reviews, see references 3,
10, and 16. Despite the
importance of these ubiquitous systems, very little is known about
specific effectors that can modulate their behavior. The two-component histidine kinase receptor VanS and its response regulator, VanR, are
the regulatory elements which control inducible vancomycin resistance
in Enterococcus faecium (1, 5, 6, 17); however, molecular details of the mechanism of VanS activation by vancomycin and
of possible inducers and modulators which act upon it remain undefined.
In vitro measurements of VanS activity (4, 6, 17) have been
made by using a chimeric protein consisting of the intracytoplasmic
domain of VanS fused at its amino terminus to Escherichia
coli maltose binding protein. VanS, however, is an integral
membrane protein and is believed to sense the presence of vancomycin,
thereby initiating a series of reactions which ultimately confer
resistance to vancomycin. On the basis of studies in the prototype
EnvZ-OmpR two-component signal transduction system of E. coli (7, 11, 14), one would also expect that the full-length VanS anchored in the cell membrane would therefore be
useful to elucidate the mechanism of VanS activation and the action of
effectors on the induction of vancomycin resistance. We therefore
describe the reconstitution of an active autokinase-phosphotransferase reaction which utilizes full-length VanS interacting with
[
-32P]ATP and purified VanR and the use of the
reconstituted system to study the action of effectors on the
VanSVanR, HisAsp phosphorelay system.
| |
MATERIALS AND METHODS |
Strains and plasmids.
Bacterial strains and plasmids which
were used in this study are described in Table
1.
Inhibitors.
Samples of compounds LY-266,400,
2-(2,3,4-trifluorophenyl)-2,3 dihydrothiazol-3-one, and LY-266,408,
2-(3-chloro, 4-fluorophenyl)-2,3 dihydrothiazol-3-one (inhibitors A and
A', respectively) (12) were gifts from Eli Lilly and Co.
(Indianapolis, Ind.). Their structures are shown in Fig.
1.
Plasmid constructions.
Gene cassettes containing the
vanR and vanS sequences were constructed by the
melt anneal method described by Ulijasz et al. (15) with PCR
primers described in Table 2. The
resultant cassettes were cloned into pET23b, as indicated, to obtain
the strains needed for overproduction of VanR and VanS proteins
(13). Overexpressed VanR-6His was purified by immobilized
metal (Ni2+) affinity chromatography as described
previously (9), whereas full-length VanS, overexpressed in
E. coli, was used in the form of washed cell membranes
containing VanS as an integral protein and prepared as described
previously with slight modifications (11, 14).
Incubation conditions.
(i) VanS autophosphorylation was
tested in a 20-µl total volume containing Tris-HCl (pH 7.4), 50 mM;
MgCl2, 5 mM; KCl, 100 mM; [-32P]ATP, 4 µCi or 200 pmol; and E. coli VanS membrane preparation, 0.5 µg of protein. Incubation was at ambient room temperature, 25°C, for 15 min. (ii) The VanRS coupled reaction showing the transfer of 32P from VanS~P to VanR took place under
conditions such as those described in item i, with a 20-µl scale,
supplemented with VanR, 1.4 µg of protein. Incubation was at ambient
room temperature, 25°C, for 10 min. (iii) To test for reversal of
inhibition, the VanRS coupled reaction (20-µl scale) was supplemented
with 100 µg of inhibitor A/ml (0.5 mM) and was added as 2 µl of a
10× stock solution which contained 1 mg/ml in 50% dimethylsulfoxide.
Incubation was at ambient room temperature, 25°C, for 10 min to allow
accumulation of VanS~P. The inhibited reaction mixture was then
supplemented with either VanR, 1.4 µg of protein, or VanS-containing
membranes, 0.5 µg of protein, as indicated. Twenty-microliter samples
were taken at 1, 3, and 10 min, by which time the transfer to VanR was
complete. The 20-µl samples were mixed with 10 µl of sodium dodecyl
sulfate (SDS) sample buffer, and 24 µl of the resulting reaction
mixture was fractionated by SDS-polyacrylamide gel electrophoresis (PAGE), followed by Coomassie blue staining. The resultant
phosphoprotein-containing bands were quantified with a PhosphorImager
(Molecular Dynamics).
| |
RESULTS |
Autophosphorylation of VanS.
Membrane preparations obtained
from E. coli EAU-112 were fractionated by SDS-PAGE, followed
by Coomassie blue staining. The results, shown in Fig.
2, lanes 1 and 2, demonstrate the
presence of a prominent 39-kDa Coomassie-staining band that was absent in the control membrane sample. When membrane preparations were incubated with [-32P]ATP, a strongly-labeled band that
was absent in the control was seen; this band comigrated with the
prominent VanS-containing 39-kDa Coomassie-staining band (Fig. 2, lanes
3 and 4). 32P incorporated into the 39-kDa band was
therefore used as a measure of VanS~P formation. An unidentified ca.
35-kDa labeled band, further characterized below, was labeled in
E. coli membrane preparations irrespective of the presence
of VanS. Based on the inhibitor data reported below, we believe that
this spurious 35-kDa band protein may have some functional similarity
with VanR.
View larger version (54K):
[in this window]
[in a new window]
|
FIG. 2.
VanS, VanR, and their respective phosphorylated forms.
Analysis by PAGE and autoradiography of washed membranes from E. coli EAU110, lacking VanS, Coomassie blue stain (lane 1); washed
membranes from E. coli EAU112 containing overexpressed VanS,
Coomassie blue stain (lane 2); washed membranes from E. coli
EAU110, lacking VanS, incubated with [-32P]ATP (lane
3); washed membranes from E. coli EAU112 containing VanS,
incubated with [-32P]ATP (lane 4); purified VanR,
Coomassie blue stain (lane 5); purified VanR plus washed membranes from
E. coli EAU110 and [-32P]ATP (lane 6); and
purified VanR plus washed membranes from E. coli EAU112 and
[-32P]ATP (lane 7). Reaction mixtures were
incubated as described in Materials and Methods. See text for a
detailed analysis.
|
|
Phosphate acceptor activity of VanR-6His.
A recombinant
C-terminal His-tagged VanR, VanR-6His, was overexpressed in E. coli EAU-111 and purified by immobilized metal (Ni2+)
affinity chromatography. Results of the analytical PAGE, Fig. 2, lane
5, showed a major band of 27 kDa corresponding to VanR and a smaller,
faster-running (ca. 25-kDa) component of unknown origin which
cofractionated with VanR through the Ni2+ affinity
chromatography step. The coupled VanSVanR reaction required the
presence of VanS; no transfer of phosphate was seen if membranes from
E. coli EAU-110 (VanS negative) were used, as shown in Fig.
2, lanes 6 and 7. Thus, the labeling of VanR requires VanS and is not
mediated by cross talk with contaminating kinases present in the
membrane preparation.
To verify the transfer of phosphate from VanS to VanR, a membrane
preparation containing full-length VanS was incubated with [-32P]ATP in a series of reaction mixtures containing
increasing VanR concentrations. Samples were collected, as indicated
previously, and fractionated by PAGE, and label in
phosphoprotein-containing bands was measured with the PhosphorImager.
The results, shown in Fig. 3, indicate that the
recombinant VanR-6His preparation was active in accepting phosphate
from VanS~P. The slower, ca. 35-kDa band appears to be present in all
samples irrespective of the presence of VanR, as shown in Fig. 3, lane
1, and is therefore associated with E. coli membranes
whether or not they contain VanS. This observation is consistent with
the findings shown in Fig. 2, lanes 3 and 4.
View larger version (28K):
[in this window]
[in a new window]
|
FIG. 3.
Transfer of 32P from VanS~P to VanR. VanS,
VanR, and [-32P]ATP were incubated, and the amount
of 32P contained in VanS and in VanR was measured as a
function of increasing VanR concentration. Lanes 1 to 6, respectively,
contained no VanR and 0.3, 0.6, 1.2, 3.0, and 6.0 µg of added VanR. A
ca. 35-kDa labeled band, labeled Unk and present in all samples, was
due to a component present in the E. coli membrane
preparation. The Unk component cannot mediate the transfer of
32P to VanR, since the transfer does not occur with
membranes which contain this component but lack VanS.
|
|
Inhibition of the coupled VanRS system in vitro.
Roychoudhury et al. (12) discovered lead
compounds, A and an analog, inhibitor A', that inhibited several
HisAsp phosphorelay signal transduction systems. Inhibitors A and A'
can be classified as halophenyl isothiazolones, and their structures
are shown in Fig. 1. The effects of inhibitors A and A' were tested in
the coupled VanRS system. Two effects were noted, shown in Fig.
4. First, an inhibition of VanR
phosphorylation, 50% effective dose (ED50), 70 µg/ml
(0.35 mM), and second, an apparent parallel accumulation of VanS~P,
suggested that 32P was incorporated into VanS but was not
transferred to VanR in the presence of inhibitor A. Inhibitor A' had an
effect similar to that of inhibitor A (data not shown) and was not
examined further.
View larger version (29K):
[in this window]
[in a new window]
|
FIG. 4.
Inhibition of 32P transfer from VanS~P to
VanR. The coupled VanRS system was used as described in Materials and
Methods. Lanes 1 to 6, respectively, contained 0, 10, 25, 55, 77.5, and
100 µg/ml. A concentration of 100 µg of inhibitor A/ml corresponds
to 0.5 mM. Concentrations of inhibitor A above 60 µg/ml (0.3 mM)
completely inhibited phosphoryl transfer from VanS~P to VanR, leading
to an apparent accumulation of the former. (Upper panel) VanS~P and
VanR~P quantified by PhosphorImager analysis. (Lower panel)
Autoradiogram corresponding to upper panel. Two components in addition
to VanS~P and VanR~P were seen in the autoradiogram one larger
than VanS~P (ca. 45 kDa), which accumulated in response to inhibitor
A, and one larger than VanR~P (ca. 35 kDa), labeled Unk, whose
concentration, like that of VanR~P, was reduced by inhibitor A. The
Unk component was shown to be derived from the E. coli
membrane preparation and does not play a significant role in the
phosphorylation of VanR, as shown in Fig. 2, lanes 6 and 7.
|
|
Two other phosphoprotein bands with lower intensity, whose
behaviors, respectively, parallel those of VanS and VanR, are
also present. One of these, a minor band with mobility corresponding to
43 kDa, accumulates in the presence of inhibitor A in parallel with
VanS. The other corresponds to the unidentified 35-kDa band described
previously in Fig. 3, labeled Unk, and its presence is inhibited in
parallel with VanR. It is possible that the 35- and 43-kDa proteins
correspond to contaminating kinase and response regulator proteins,
suggesting that inhibitor A may affect other two-component signal
transduction systems in a similar way.
Site of inhibitor action in vitro.
Reduced phosphate transfer
from VanS~P to VanR, shown in Fig. 4, could be due either to
inhibition of the donor activity of VanS~P or to inhibition of the
acceptor activity of VanR. To distinguish between these two
possibilities, VanS-containing membranes, VanR, and
[-32P]ATP were incubated in the presence of
inhibitor A to accumulate VanS~P maximally, as described above. The
reaction mixture was then supplemented with either VanR or VanS and
sampled at 1, 3, and 10 min for determination of the relative amounts
of VanS~P and VanR~P. Results are shown in Fig.
5. The relative amounts of VanS~P and
VanR~P remained unchanged in the control reaction mixture over the
sampling period in the absence of supplementation, as shown in the
control sample (Fig. 5a). The addition of VanR to the reaction mixture
resulted in a transfer of label from VanS~P to VanR, as shown in Fig.
5b, whereas the addition of VanS-containing membranes resulted in the
accumulation of additional VanS~P, as shown in Fig. 5c. These
observations suggest that inhibitor A affects the ability of VanR to
accept Pi rather than the ability of VanS~P to transfer
Pi.
View larger version (30K):
[in this window]
[in a new window]
|
FIG. 5.
Site of action of inhibitor A. The coupled VanRS
reaction was supplemented with 100 µg of inhibitor A per ml and
incubated for 15 min to allow maximal accumulation of VanS~P, as
shown in Fig. 4. The reaction mixture was then supplemented either with
buffer (a); VanR, 1.4 µg (b); or VanS, 0.5 µg (c). Lanes 1 to 4, respectively, correspond to samples taken at 0, 1, 3, and 10 min, as
indicated, and analyzed by PAGE, phosphorimaging, and autoradiography.
(Upper panel) Quantification of VanS~P and VanR~P as a function of
time after addition of buffer control (a), VanR (b), or VanS (c).
Phosphoproteins formed in the timed samples were measured as a function
of time after supplement was added to a preincubated mixture containing
32P~VanS, VanR, and inhibitor A. (Lower panel)
Autoradiographic analysis of phosphoproteins. The autoradiograms
present an image of the phosphoproteins that were quantified above and
graphed. Both the unknown band and VanR~P appear to increase as a
function of time. See the text for a discussion of this observation.
|
|
The appearance of label in the 35-kDa Unk component present in the
VanS-containing membrane preparation is also seen upon the addition of
VanR to the inhibited reaction (Fig. 5b). Whether the source of 35-kDa
phosphorylation is cross talk with VanS (4, 16) or with some
other kinase present in either the VanS or VanR preparation is
undetermined. This observation suggests that the action of inhibitor A
may affect the ability of other response regulators to accept
Pi from VanS or from their respective cognate or noncognate
kinases. These results also suggest that inhibitor A has a high
dissociation constant and therefore relatively low affinity in its
interaction with VanR.
| |
DISCUSSION |
We have shown the transfer of Pi from full-length
VanS~P to VanR in vitro and characterized the action of a compound,
inhibitor A, which prevents phosphoryl transfer by its action on VanR
rather than on VanS. In the course of these studies, two spurious
labeled phosphoprotein bands of interest appeared, with 43- and 35-kDa mobility (Fig. 4). In the presence of inhibitor, the
phosphorylated 43-kDa component accumulated in parallel with VanS~P,
whereas the phosphorylated 35-kDa component was reduced in parallel
with VanR~P. Moreover, the addition of VanR (but not VanS) to the
inhibited reaction restored both VanR~P and the 35-kDa component.
Constituents of other two-component systems present in the E. coli membrane preparations cannot account for the transfer
of label from VanS~P to VanR because the transfer was seen only
if the E. coli membranes were prepared from EAU-112,
carrying VanS, and not if the E. coli membranes were
prepared from EAU-110, which lacks VanS. Thus, cross talk between VanR
and an E. coli kinase cannot account for the observed
phosphorylation of VanR.
The 35- and 43-kDa spurious bands are of interest because their
presence suggests more general action by inhibitor A on other bacterial
signal transduction systems. Thus, inhibitor A might recognize a common
structural motif present in other response elements, leading to
accumulation of the kinases which phosphorylate them. Assuming that the
35-kDa component which originates in the E. coli membrane
preparation is a response regulator, the reappearance of both VanR~P
and the 35-kDa band in the inhibited reaction supplemented with VanR
(Fig. 5b) suggests that inhibitor A is dissociable from its target and
that its inhibitory activity is reversible. The presence of
constituents belonging to two-component signal transduction systems
other than VanRS thus serves a useful function by providing a parallel
test reaction. It would be desirable to reconstitute a more
defined system lacking spurious components, and indeed, Jung et al.
(8) have reconstituted a purified homogeneous preparation of
KdpD, the high-affinity K+-translocating ATPase of E. coli, into liposomes with recovery of autokinase activity. The
method of reconstitution appears to be generally applicable.
Inhibitors A and A' were discovered by Roychoudhury et al.
(12) as part of a systematic search for inhibitors of signal transduction that would target aliginate biosynthesis in
Pseudomonas aeruginosa. Inhibitor A at 0.2 µg/ml (1 µM)
was shown to inactivate the algD promoter linked to an
xylE reporter. When the phosphorylation of AlgR1-AlgR2 in
vitro was tested, it was noted that inhibitors A and A' inactivated the
autokinase activity of AlgR2 (ED50, ca. 2 µg/ml [10
µM]). In contrast, our studies showed that inhibitor A had no effect
on the kinase activity of VanS in vitro at these low concentrations,
while at higher concentrations (ED50, 70 µg of
inhibitor A/ml [350 µM]) it appeared to stimulate
phosphorylation of VanS. Differences in ED50 could be due a
lack of standardization in the test reactions used to measure the
action of inhibitors and the variations in component concentrations;
however, we cannot reconcile the effectiveness of inhibitor A in
reactions which involve chemically different mechanisms.
The apparent stimulation of VanS phosphorylation was, in fact, an
accumulation of VanS~P owing to the inhibition of VanR to accept
Pi. The autoradiogram shown in Fig. 4 suggests that another component, possibly a histidine kinase, shows a similar response to
inhibitor A. This observation suggests a possible way to label, selectively, a functionally related set of histidine kinases.
A systematic search for inhibitors of signal transduction reported by
Barrett et al. (2) led to a series of antimicrobial tyramine
derivatives which inhibited the autokinase activity of a KinA model
system. Barrett and Hoch (3) have recently reviewed two-component signal transduction as a target for the discovery of new
antiinfective agents. This review includes a list of newer agents which
were not covered in their earlier publication (2).
The demonstration of a phosphoryltransferase inhibitor which
modifies the ability of VanR to accept Pi from VanS
complements the discovery of kinase inhibitors reported in other
studies. It will be interesting to learn the structural basis for the
interaction of two-component signal transduction systems with the
growing list of small-molecule ligands which modulate their activity.
We thank Rob Giannattasio for expertly generating the figures and
Eli Lilly and Co. for compounds LY-266,400 (inhibitor A), LY-266,408
(inhibitor A'), and LY-000051 (inhibitor B).
| 1.
|
Arthur, M.,
C. Molinas, and P. Courvalin.
1992.
The VanS-VanR two-component regulatory system controls synthesis of depsipeptide peptidoglycan precursors in Enterococcus faecium BM4147.
J. Bacteriol.
174:2582-2591[Abstract/Free Full Text].
|
| 2.
|
Barrett, J. F.,
R. M. Goldschmidt,
L. E. Lawrence,
B. Foleno,
R. Chen,
J. P. Demers,
S. Johnson,
R. Kanojia,
J. Fernandez,
J. Bernstein,
L. Licata,
A. Donetz,
S. Huang,
D. J. Hlasta,
M. J. Macielag,
K. Ohemeng,
R. Frechette,
M. B. Frosco,
D. H. Klaubert,
J. M. Whiteley,
L. Wang, and J. A. Hoch.
1998.
Antibacterial agents that inhibit two-component signal transduction systems.
Proc. Natl. Acad. Sci. USA
95:5317-5322[Abstract/Free Full Text].
|
| 3.
|
Barrett, J. F., and J. A. Hoch.
1998.
Two-component signal transduction as a target for microbial anti-infective therapy.
Antimicrob. Agents Chemother.
42:1529-1536[Free Full Text].
|
| 4.
|
Fisher, S. L.,
S.-K. Kim,
B. L. Wanner, and C. T. Walsh.
1996.
Kinetic comparison of the specificity of the vancomycin resistance kinase VanS for two response regulators, VanR amd PhoB.
Biochemistry
35:4732-4740[Medline].
|
| 5.
|
Haldimann, A.,
S. L. Fisher,
L. L. Daniels,
C. T. Walsh, and B. L. Wanner.
1997.
Transcriptional regulation of the Enterococcus faecium BM4147 vancomycin resistance gene cluster by the VanS-VanR two-component regulatory system in Escherichia coli K-12.
J. Bacteriol.
179:5903-5913[Abstract/Free Full Text].
|
| 6.
|
Holman, T. R.,
Z. Wu,
B. L. Wanner, and C. T. Walsh.
1994.
Identification of the DNA-binding site for the phosphorylated VanR protein required for vancomycin resistance in Enterococcus faecium.
Biochemistry
33:4625-4631[Medline].
|
| 7.
|
Igo, M. M.,
A. J. Ninfa,
J. B. Stock, and T. J. Silhavy.
1989.
Phosphorylation and dephosphorylation of a bacterial transcriptional activator by a transmembrane receptor.
Genes Dev.
3:1725-1734[Abstract/Free Full Text].
|
| 8.
|
Jung, K.,
B. Tjaden, and K. Altendorf.
1997.
Purification, reconstitution, and characterization of KdpD, the turgor sensor of Escherichia coli.
J. Biol. Chem.
272:10847-10852[Abstract/Free Full Text].
|
| 9.
|
Kovalic, D.,
R. B. Giannattasio,
H. J. Jin, and B. Weisblum.
1994.
23S rRNA domain V, a fragment that can be specifically methylated in vitro by the ErmSF (TlrA) methyltransferase.
J. Bacteriol.
176:6992-6998[Abstract/Free Full Text].
|
| 10.
|
Mizuno, T.
1998.
His-Asp phosphotransfer signal transduction.
J. Biochem.
123:555-563[Abstract/Free Full Text].
|
| 11.
|
Rampersaud, A.,
R. Utsumi,
J. Delgado,
S. A. Forst, and M. Inouye.
1991.
Ca2(+)-enhanced phosphorylation of a chimeric protein kinase involved with bacterial signal transduction.
J. Biol. Chem.
266:7633-7637[Abstract/Free Full Text].
|
| 12.
|
Roychoudhury, S.,
N. A. Zielinski,
A. J. Ninfa,
N. E. Allen,
L. N. Jungheim,
T. I. Nicas, and A. M. Chakrabarty.
1993.
Inhibitors of two-component signal transduction systems: inhibition of alginate gene activation in Pseudomonas aeruginosa.
Proc. Natl. Acad. Sci. USA
90:965-969[Abstract/Free Full Text].
|
| 13.
|
Studier, F. W.,
A. H. Rosenberg,
J. J. Dunn, and J. W. Dubendorff.
1990.
Use of T7 RNA polymerase to direct expression of cloned genes.
Methods Enzymol.
185:60-89[Medline].
|
| 14.
|
Tokishita, S.,
A. Kojima,
H. Aiba, and T. Mizuno.
1991.
Transmembrane signal transduction and osmoregulation in Escherichia coli. Functional significance of the periplasmic domain of the membrane-located protein kinase, EnvZ.
J. Biol. Chem.
266:6780-6785[Abstract/Free Full Text].
|
| 15.
|
Ulijasz, A. T.,
A. Grenader, and B. Weisblum.
1996.
A vancomycin-inducible LacZ reporter system in Bacillus subtilis: induction by antibiotics that affect cell wall synthesis.
J. Bacteriol.
178:6305-6309[Abstract/Free Full Text].
|
| 16.
|
Wanner, B. L.
1992.
Is cross regulation by phosphorylation of two-component response regulator proteins important in bacteria?
J. Bacteriol.
174:2053-2058[Free Full Text].
|
| 17.
|
Wright, G. D.,
T. R. Holman, and C. T. Walsh.
1993.
Purification and characterization of VanR and the cytosolic domain of VanS: a two-component regulatory system required for vancomycin resistance in Enterococcus faecium BM4147.
Biochemistry
32:5057-5063[Medline].
|
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
