Targeting of PBP1 by β-lactams Determines recA/SOS Response Activation in Heterogeneous MRSA Clinical Strains
Rosato AE (2013) Targeting of PBP1 by b-lactams Determines recA/SOS Response Activation in Heterogeneous
MRSA Clinical Strains. PLoS ONE 8(4): e61083. doi:10.1371/journal.pone.0061083
Targeting of PBP1 by b-lactams Determines recA /SOS Response Activation in Heterogeneous MRSA Clinical Strains
Konrad B. Plata 0
Sarah Riosa 0
Christopher R. Singh 0
Roberto R. Rosato 0
Adriana E. Rosato 0
J. Ross Fitzgerald, University of Edinburgh, United Kingdom
0 1 Department of Pathology and Genomic Medicine, The Methodist Hospital , Houston , Texas, United States of America, 2 Center for Molecular and Translational Human Infectious Diseases Research, The Methodist Hospital Research Institute , Houston, Texas , United States of America
The SOS response, a conserved regulatory network in bacteria that is induced in response to DNA damage, has been shown to be associated with the emergence of resistance to antibiotics. Previously, we demonstrated that heterogeneous (HeR) MRSA strains, when exposed to sub-inhibitory concentrations of oxacillin, were able to express a homogeneous high level of resistance (HoR). Moreover, we showed that oxacillin appeared to be the triggering factor of a b-lactam-mediated SOS response through lexA/recA regulators, responsible for an increased mutation rate and selection of a HoR derivative. In this work, we demonstrated, by selectively exposing to b-lactam and non-b-lactam cell wall inhibitors, that PBP1 plays a critical role in SOS-mediated recA activation and HeR-HoR selection. Functional analysis of PBP1 using an inducible PBP1-specific antisense construct showed that PBP1 depletion abolished both b-lactam-induced recA expression/activation and increased mutation rates during HeR/HoR selection. Furthermore, based on the observation that HeR/HoR selection is accompanied by compensatory increases in the expression of PBP1,-2, -2a, and -4, our study provides evidence that a combination of agents simultaneously targeting PBP1 and either PBP2 or PBP2a showed both in-vitro and in-vivo efficacy, thereby representing a therapeutic option for the treatment of highly resistant HoR-MRSA strains. The information gathered from these studies contributes to our understanding of b-lactam-mediated HeR/HoR selection and provides new insights, based on b-lactam synergistic combinations, that mitigate drug resistance for the treatment of MRSA infections.
Funding: This study was funded by United States National Institutes of Health grant 5R01AI080688-03 (to A. E. Rosato, principal investigator). The funders had no
role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
Staphylococcus aureus is a main pathogen responsible for a number
of diseases ranging from skin and soft tissue infections to
lifethreatening endocarditis, both in hospitals and community settings
. In S. aureus, the cell wall is a large mesh-like polymer where the
main constituent, the peptidoglycan polymer, is made of long
glycan chains cross-linked with peptide bridges . The primary
target of b-lactam antibiotics are penicillin binding proteins
(PBPs), which are involved in the last stages of peptidoglycan
biosynthesis . b-lactam resistance in methicillin-resistant S.
aureus (MRSA) involves the acquisition of PBP2a, a protein
encoded by mecA, to complement the four native staphylococcal
PBPs (PBP1-4). In MRSA strains, native PBP2 has been shown to
be required for expression of resistance, while in
methicillinsusceptible S. aureus (MSSA) strains it is essential for growth [3,4].
PBP1 localizes at the division septum which is the main site of cell
wall synthesis in S. aureus. It has also been shown to be critical for
growth and cell division and separation, but not the major
contributor of peptidoglycan cross-linking in both MRSA and
MSSA . Orthologs of S. aureus PBP1 are PBP3 in Escherichia
coli and Caulobacter crescentus and PBP2B in Bacillus subtilis; in all
cases, they have been shown to be involved in peptidoglycan
synthesis during cell division . Both PBP1 and PBP2 are the
essential PBPs in S. aureus, but only PBP1 remains active in the
presence of PBP2a, which can replace the essential transpeptidase
function of PBP2 . PBP4, although not essential for viability,
has been shown to play a key role in b-lactam resistance in
community-acquired MRSA (CA-MRSA), an effect that is linked
to its unique function in producing highly-cross linked
A feature of MRSA strains from both hospital- and
communityassociated infections is their heterogeneous expression of resistance
to b-lactam antibiotics , in which the majority of the cells
express resistance to low concentrations of oxacillin [i.e., $10 mg/
ml; heterotypic resistance (HeR)], while the minority of cells
(#0.1%) expresses resistance to a high concentration [i.e.,
$256 mg/ml; homotypic resistance (HoR)] .
In previous studies, we demonstrated that MRSA-HeR strains,
clinically misinterpreted as MSSA (MICs to oxacillin: 2 mg/ml),
were able to express a homogeneous, high level of resistance
(MICs: 256 mg/ml; HoR) when exposed to sub-inhibitory
concentrations of oxacillin (0.5 mg/ml). MRSA-HoR strains are
stable and retain their trait of high level of resistance after several
passages in drug-free media. We determined the pre-existence of a
hypermutable sub-population in MRSA-HeR strains that favored
the selection to HoR phenotype in the presence of oxacillin .
Oxacillin appeared as the triggering factor of a b-lactam-induced
SOS response through lexA/recA regulators, responsible for an
increased mutation rate and selection of the highly resistant HoR
derivative . The activated LexA/RecA complex induces
autocleavage of the repressor LexA leading to the transcription
of genes involved in DNA repair. Moreover, an error-prone
polymerase (umuC) has been identified within the S. aureus lexA
regulon as being involved in the mutation rate .
Previous works have shown that: 1)- b-lactam antibiotics that
target the transpeptidase domain of PBP3 (ceftazidime) and do not
directly damage DNA or affect replication in E.coli, trigger the
SOS response via the two-component system DpiAB ; and
2)inhibition of cell wall biosynthesis at steps other than PBP3 activity
may specifically induce DNA Pol IV expression in E.coli,
contributing to an increased mutation rate . Based on these
and our own observations that b-lactam-mediated SA13011 HeR/
HoR selection involves induction of the SOS response, we
hypothesized that the b-lactams-induced SOS response in
heterogeneous MRSA may originate under circumstances in
which cell wall integrity is compromised. The purpose of the
following studies was to test this hypothesis, and to elucidate
whether signals from cell wall components are involved in
triggering recA-mediated SOS response in SA13011-HeR, allowing
for HoR resistant phenotype selection and survival in the presence
of b-lactams. In this context, by selectively exposing MRSA
SA13011-HeR cells to b-lactams and non-b-lactam cell wall
inhibitors, we determined that PBP1 plays a critical role in recA
activation and SOS-mediated HeR/HoR selection. Functional
analysis of PBP1 with an inducible PBP1-specific antisense RNA
demonstrated that PBP1 depletion may lead to decreased recA
expression during HeR/HoR selection, causing a decrease of
mutation rate through UmuC. Furthermore, based in the
mechanistic observations established for PBP1, our study provides
evidence that combination of agents targeting PBP1 and b-lactams
targeting PBP2 and/or PBP2a showed in-vitro as well in-vivo
efficacy, representing a therapeutic option for the treatment of
highly-resistant MRSA-HoR. Our results provide an important
contribution to our understanding of b-lactam-mediated HeR/
HoR selection and new insights for the treatment of MRSA
Materials and Methods
grown in 5 ml of LB broth without antibiotic overnight. Cultures
were then back-diluted to an optical density at 600 nm (OD600) of
0.050.1 in LB broth with or without sub-inhibitory concentration
of b-lactam and non-b-lactam (Sigma-Aldrich) antibiotics, and
grown at 37uC with shaking (180 rpm).
Determination of mutation frequency
Mutation frequencies for resistance to rifampicin were
determined during the selection process for SA13011-HeR and
derivatives (2/+ OXA 0.5 mg/ml). Inoculated flasks were
incubated at 37uC with shaking at 145 rpm; aliquots of 100 ml
were taken at different time intervals, including 6, 27, and 33 h as
we previously described [15,21]. All of the variants were selected
on TSA plates containing rifampicin 200 mg/ml and TSA plates
using serial dilutions to determine CFU/ml. Mutation frequencies
were expressed as the number of rifampicin-resistant mutants
recovered as a fraction of the viable count. Three independent
cultures were sampled in triplicate to minimize error caused by
inter- and intra-sample variation.
DNA manipulation and sequencing
Chromosomal DNA was extracted using the Qiagen genomic
DNA preparation kit (Qiagen, Valencia, CA) according to the
manufacturers directions. Sequencing of all PCR amplification
products and plasmids was performed by the Nucleic Acid
Research Facility at GENEWIZ (Houston, TX). Sequence analysis
of recA from SA13011 and derivative strains was performed using a
set of primers previously described . Consensus sequences
were assembled from both orientations and DNASTAR Lasergene
(Madison, WI). S. aureus N315 (accession # BA000018) was used
as a positive control.
Construction of recA promoter reporter activity assay
A promoter fragment of recA was amplified by employing
recAF2 and recA-R2 primers (Table 1). The PCR product, covering
398 bp upstream of the recA start codon (containing regulatory
sequences of the gene, including LexA binding sites ) and
102 bp encoding for the first 34 amino acids of RecA, were ligated
in front of the promoterless lacZ gene of pMC1871 . The
plasmid was verified by restriction enzyme digestion and PCR to
determine the correct orientation, and transformed into competent
E. coli TOP10 cells (Invitrogen Life Technologies, Carlsbad, CA).
Colonies containing the recA-lacZ fusion were selected in
tetracycline agar plates (10 mg/ml) and X-gal
(5-bromo-4-chloro-3indolyl-b-D-galactopyranoside; 40 mg/ml). The construct was
digested with PstI, and ligated into the PstI restriction site of a
shuttle vector, pRB473. The pRB473-recAprom-lacZ reporter was
sequenced using recA seq and lacZ seq primers (Table 1) to verify
the absence of mutations. The final construct was electroporated
into S. aureus RN4220 and transduced into SA13011-HeR by 80a
phage-mediated transduction . The resulting strain harboring
SA13011-HeR recA-lacZ was designated LMR-21 (Table 1).
pbp1 antisense sequence construct
To construct the antisense sequence of pbp1, a 336-bp fragment
corresponding to the downstream region of pbp1 and containing
the transcriptional terminator of the gene, was amplified using
Pbp1-term-F and Pbp1-term-R primers (Table 1). The PCR
product was digested with BamHI and SacI and ligated into pCL15
; the resulting construct was designated pCL15-pbp1T. A
second fragment of pbp1 covering 475-bp upstream and 146-bp
downstream of the start codon was amplified using PBP1F-ans
(containing the XbaI restriction site) and PBP1R-ans (containing
Heterogeneous-mecA(+) oxacillin susceptible
SA13011-HeR+OXA (0.5 mg/ml); SA13011 homogeneous derivative
LMR-21+OXA (0.5 mg/ml); LMR-21 homogeneous derivative
LMR23+OXA (0.5 mg/ml); LMR-23 homogeneous derivative
LMR-25+OXA (0.5 mg/ml); LMR-25 homogeneous derivative
the HindIII restriction site) primers. The PCR product was
digested with XbaI and HindIII and cloned into the corresponding
restriction sites of pCL15-pbp1T, yielding pCL15-pbp1T-ANT
(Table 1). The pCL15-pbp1T-ANT construct was confirmed by
restriction enzyme digestion and sequencing with PBP1R-ans and
PBP1F-ans primers. The constructs pCL15-pbp1T-ANT and
pCL15-pbp1T (empty-vector) were electroporated into S. aureus
RN4220 and transduced into SA13011-HeR by 80a
phagemediated transduction. The resulting strains,
SA13011-pCL15pbp1T-ANT and SA13011-pCL15-pbp1T-EV (empty-vector), were
designated LMR23 and LMR25, respectively.
LMR-21 [SA13011-HeR(recAP::lacZ)] was grown as we
previously described  during the process of HeR/HoR selection, in
the absence and presence of sub-inhibitory concentrations
(Table 2) of various b-lactam antibiotics, including OXA
(0.5 mg/ml), CLOX (0.25 mg/ml), BAL (0.25 mg/ml), ZOX
(32 mg/ml), FOX (2 mg/ml), and IMP (0.012 mg/ml); and
nonb-lactam cell wall inhibitors including, BAC (32 mg/ml), DCS
(8 mg/ml), and VAN (0.5 mg/ml). An equal number of bacteria
were centrifuged, re-suspended in 750 ml of buffer Z (0.06M
Na2HPO4, 0.04M NaH2PO4, 0.01M KCl, 1 mM MgSO4, 0.05M
b-mercaptoethanol, pH 7.0), transferred into Lysing Matrix B
tubes with silica spheres (MP Biomedicals, Santa Ana, CA), and
disrupted using a FastPrep FP120 high-speed homogenizer
(Thermo Fisher Scientific, Rockford, IL) for 40 sec at a speed
setting of 6.0. Protein concentrations were determined using
Protein Assay Reagent (Pierce, Thermo Fisher Scientific) per the
manufacturers recommendations.b-galactosidase activity was
measured by using total protein extracted from cultures and
assessed by 2-nitrophenyl b-D-galactopyranoside degradation
(ONPG). As a positive control of the recA promoter activity, we
used a protein sample extracted from LMR-21 grown in the
presence of 0.005 mg/ml of mitomycin C, a well-known inducer of
the SOS system [17,29].
Real-time Reverse Transcriptase -PCR
RNA extractions for real-time reverse transcriptase (RT)-PCR
were performed as previously described [15,30,31]. Total RNA
was extracted using a RNeasy Mini Kit (Qiagen, Valencia, CA); all
RNA samples were analyzed by A260/A280 spectrophotometry and
gel electrophoresis to estimate concentration and asses integrity,
and cleaned of potential DNA contamination by treating them
with DNAse per the manufacturers recommendations (Ambion,
Life Technologies, Austin, TX). Real time RT-PCR analysis was
done using the SensiMix SYBR One-Step kit (Quantace/Bioline,
Taunton, MA) according to the manufacturers protocol. Gene
expression was evaluated using CT values converted to fold change
and compared with a sample used as a reference (value = 1) using
log2 2(DDCt). The change (n-fold) in the transcript level was
calculated using the following equations: DCT = CT(test DNA)2
CT(reference 2cDDNDAC), DDCT = DCT(target gene)2DCT(16S rRNA), and
ratio = 2 T . The quantity of cDNA for each
experimental gene was normalized to the quantity of 16S
cDNA in each sample as determined in a separate reaction.
Each RNA sample was run in triplicate. Values represent the
means of at least three biological replicates 6 standard error
of the mean (SEM), sampled in triplicate to minimize error by
inter- and intra-samples. Differences between the mean
values were analyzed using a one-way analysis of variance
(ANOVA). A P value of ,0.01 was considered statistically
significant (*). Oligonucleotide primers are shown in Table 1.
Measurement of DNA synthesis during HeR-HoR
blactam mediated selection
The rate of DNA synthesis in cells initiating growth was
measured by following the incorporation of methyl-[3H]-thymine
into cells uniformly labeled with the same precursor. Cultures of
SA13011 cells were grown in the absence and presence of OXA
(0.5 mg/ml) during the selection process. When SA13011-HoR
reached OD600 nm: 0.13, cultures of SA13011-HeR were
backdiluted and adjusted to the same OD with pre-warmed medium.
Both SA13011-HeR/-HoR cultures were supplemented with the
radioactive precursor (1 mCi/ml) at the same specific activity and
concentration. Triplicate samples were removed at different
timeintervals and collected for DNA extraction (Qiagen, Valencia,
CA); eluted, labeled DNAs were transferred to a Whatman filter
paper and, after drying, radioactivity was measured by liquid
scintillation. An unlabeled set of inoculated flasks (SA13011-HeR
6 OXA 0.5 mg/ml) were grown as controls to monitor growth and
for measurement of CFU/ml at the same time intervals.
Time kill analyses
Bactericidal synergy assays for CTX, BAL, FOX, and IMP
were performed using MH broth with an initial inoculum of
16106 CFU/ml at K MICs (based on individual strain E-test data
shown in Table 2), as previously described . Aliquots of each
culture were serially diluted and plated for CFU determination at
time intervals 0, 2, 4, 6, 8, and 24 h. A minimum of two
independent experiments were run for each combination.
Wax worm infection
Galleria mellonella larvae possess an immune system with
reasonable homology to vertebrates containing hemolymph
analogous to blood, transports nutrients, hemocytes and immune
molecules [34,35]. These tissue types are similar to those
encountered by S. aureus during invasive infections in humans.
Groups of larvae of G. mellonella (10/group) were inoculated with
10 ml of the bacterial suspension of SA13011-HoR strain (1.56106
CFU) as previously described . All larvae were confirmed to be
alive at 2 h post-inoculation (here designated 0 h). Then, The first
clinically-recommended treatment doses [36,37] of IMP (10 mg/
kg), NAF (5 mg/kg), CTX (10 mg/kg), and BAL (5 mg/kg), or the
combinations IMP/NAF, IMP/CTX, and IMP/BAL, were
administered in PBS into the right hind most proleg, and
reincubated for 24 h at 37uC. In addition, control groups included
one group of larvae that had been inoculated with live bacteria
that received PBS only as treatment, and the uninfected control
group, which received PBS treatments to control for multiple
injections. Repeat treatment doses were given at 24 and 48 h.
Worms were checked daily, and deaths were recorded for a total of
10 days. A minimum of three independent experimental runs were
performed for each IMP/b-lactam combination. The survival data
were plotted using the KaplanMeier method.
Activation of recA-mediated SOS induction during HeR/
In our previous studies, we determined that b-lactam-mediated
activation of the SOS response during HeR/HoR selection
involved increased expression of recA . As mentioned above,
the present study was designed to test the hypothesis that
b-lactaminduced SOS response in heterogeneous MRSA may originate
when the cell wall integrity is compromised. To determine the
impact of b-lactam-mediated HeR/HoR selection on the
expression of PBPs, we analyzed their expression profile by real-time
RT-PCR analysis. RNA samples from SA13011-HeR/HoR cells
were collected at the exponential phase of growth (OD600 = 0.6
0.7). As shown in Figure 1, selection of the HoR derivative resulted
in increased expression of most of the pbp genes, including pbp1,
pbp2, pbp2a, and pbp4; levels of pbp3 remained almost undetectable
under any condition and displayed no changes in expression.
Thus, this analysis showed that OXA-mediated HeR/HoR
selection is characterized by a generalized increase in the
expression of most of the genes coding for PBPs.
To investigate b-lactam-mediated recA activation, we generated
a SA13011-HeR derivative strain harboring a recA-lacZ reporter
gene [SA13011-HeR(recAP::lacZ), LMR21]. The construct
contains the upstream recA promoter followed by a portion of the 59
recA coding region ligated to a lacZ gene as described in Materials
and Methods. Phenotypic analysis of LMR21 is shown in Table 2.
To establish the role that different b-lactam antibiotics may play in
activating the recA promoter, recA promoter-mediated
b-galactosidase activity was determined in LMR21 cells exposed to sub-lethal
concentrations (1/4 MIC; Table 2) of different antibiotics
harvested at the exponential phase of growth (OD: 0.6). We
based our choice on their specific PBP/s target affinity, i.e.
nonselective: OXA and CLOX (PBP1-PBP3); selective: CTX
(targets PBP2); BAL (PBP2a), CEC (PBP3), and FOX (PBP4). The
other antibiotics used were based on the fact that they are directed
against different steps of cell wall synthesis, although no PBPs
mediated, including VAN (binding to C-terminal of D-Ala- D-Ala
residues of the externally oriented lipid-linked peptidoglycan
precursor), BAC (binds to undecaprenyl pyrosphospate [UPP], the
lipid carrier for translocation of cell envelope precursors from
cytosol to the extracellular surface of the cytoplasmic membrane),
and DCS (interferes with D-alanine incorporation into
peptidoglycan). As shown in Table 3, increased b-galactosidase activity
(approx. 23-fold change) was observed in the presence of
subinhibitory concentrations of both OXA (0.5 mg/ml) and CLOX
(0.25 mg/ml), both agents targeting PBP1 and PBP3. By contrast,
no changes in b-galactosidase activity were observed in the
presence of FOX (PBP4), ZOX (PBP2), or BAL (PBP2a).
Similarly, no recA promoter induction was observed with other
cell wall targeting antibiotics, including CEC, CTX, BAC, VAN,
or DCS (Table 3). These results suggested that inhibition of PBP1
and/or PBP3 by OXA and CLOX may be involved in recA
activation occurring during HeR/HoR selection. To determine
the specific contribution of each of these PBPs, we used IMP,
which targets PBP1 , and cefaclor, a b-lactam specific for
PBP3 . While exposure of LMR21 cells to IMP resulted in a
2.7-fold increase in b-galactosidase activity, no induction of recA
promoter activity was observed in cells exposed to cefaclor
(Table 3). Furthermore, and in agreement with a role of PBP1 on
recA promoter activation, LMR21 cells exposed simultaneously to
IMP and OXA (0.06 mg/ml and 0.5 mg/ml, respectively)
displayed a 5.3-fold increase in recA promoter/b-galactosidase activity
compared to a ,2.5-fold increase produced by either IMP or
OXA administered individually (Table 3). Together, these
observations led us to postulate that PBP1, a common target of
OXA and IMP , may play an important role in recA-mediated
activation of the SOS system during b-lactam-mediated HeR/
PBP1 is required for b-lactam-mediated recA increased
The observations that OXA and IMP, by interfering with PBP1,
were the two b-lactams increasing recA activity support the notion
that, during b-lactam-induced HeR/HoR selection, PBP1 may
have a critical role in the activation of the RecA-mediated SOS
response. To test this hypothesis, functional assays where
conducted by using an inducible PBP1-specific antisense RNA
system. The choice of an IPTG-inducible PBP1 antisense
construct to reduce pbp1expression was based on the fact that
PBP1 is an essential enzyme and, as shown by Pereira et al. , its
constitutive down-regulation may result in cell death [6,7]. The
resulting strains, SA13011-HeR+pCL15-pbp1T-ANT (LMR23)
and the corresponding empty-vector control LMR25
(SA13011HeR+pCL15-pbp1T), were first evaluated for their capacity to
grow 2/+ IPTG (1 mM) (Figure 2A). Both LMR23/25 strains
were grown overnight in the absence of IPTG and then
backMitomycin C (+control)
Fold change (Miller units)
The number of Miller units is expressed in fold change. Values represent the
mean 6 SD of three independent replicates.
diluted to an OD600 nm: 0.020; after 2 h, IPTG (1 mM) was added
to the cultures and cells were collected at the indicated time
intervals for OD measurement. Interestingly, inducible expression
of the pbp1 antisense (LMR23+IPTG 1 mM) significantly impaired
and delayed the growth of the strain as compared to the
corresponding empty vector-expressing cells (LMR25 6 IPTG),
or LMR23 growing in the absence of the inducer IPTG
The expression of pbp1 was monitored by real-time RT-PCR.
RNAs were extracted from cells grown 2/+ IPTG (1 mM) in the
absence or presence of OXA (0.5 mg/ml) and collected at the
exponential phase during OXA-mediated HeR/HoR selection. As
shown in Figure 2B, IPTG-induced pbp1-antisense expression
abolished the OXA-induced increase in pbp1 expression.
Importantly, inactivation of the pbp1 gene also abolished OXA-induced
increased recA expression (Figure 2B).
In previous studies, we demonstrated that b-lactam-mediated
triggering of the SOS response was associated with an increase in
the mutation rates . To determine whether inactivation of
PBP1, which blocked OXA-induced recA expression, may affect
the b-lactam-mediated change in mutation rate, the number of
mutants in strains LMR23 and corresponding controls
[SA13011HeR and LMR25 (pCL15-T-empty-vector)], were measured.
Frequency of mutation was determined by growing the strains in
the absence and presence of sub-inhibitory concentrations of OXA
(0.5 mg/ml) at 6-, 27-, and 33-h time intervals during the HeR/
HoR selection process. We used the events of occurrence of
rifampicin-resistant mutants as a marker for mutation frequency
expressed as the ratio of rifampicin-resistant mutants recovered as
a fraction of the viable count (Table 4) [15,21,41,42]. While
exposure to sub-inhibitory concentrations of OXA determined a
,4-log increase in mutation rates in both SA13011-HoR and
LMR26 (LMR25-empty vector+OXA) control strains at the 33 h
interval (i.e., 1.461029/2.861025, 1.461029/2.361025,
SA13011-HeR/HoR, LMR25/26, respectively), these changes
were suppressed in cells expressing the inducible pbp1-antisense
(i.e., 1.861029/2.561028, LMR23/LMR24+IPTG/OXA,
respectively) (Table 4).
Within the SOS-mediated response, the UmuC error-prone
polymerase has been identified in S. aureus as one of the enzymes
involved in the increased mutation rate observed after SOS
induction by ciprofloxacin . The expression levels of umuC
were determined in the same set of samples described above. As
shown in Figure 2B, and consistent with the mutation rates shown
in Table 4, expression of umuC, which underwent a significant
increase in SA13011+OXA, remained unchanged in the absence
SA13011-HeR+OXA * (HoR) 4.061027
LMR26 (LMR25+IPTG+OXA*) 1.761027
LMR24 (LMR23+OXA)+IPTG 2.361028
Mutation frequency was determined using rifampicin selection and expressed
as a ratio of rifampicin resistance as a fraction of viable cells, during time points
at 6, 27, and 33 h of the selection process, 2/+OXA 0.5 mg/ml (*).
of pbp1 induction (i.e. LMR23 +OXA/IPTG, Figure 2B).
Together, these results highlight the central role played by PBP1
in b-lactam-mediated RecA activation and subsequent triggering
of the SOS response and increased mutation rate.
Effect of oxacillin in DNA synthesis during
b-lactammediated HeR/HoR selection
There is evidence that PBP1 is a major player involved in cell
division and separation in S. aureus [6,7]. Furthermore, in E.coli,
studies have shown that DNA replication and cell division are
tightly coupled and synchronized processes [43,44]. Taking into
account the observations showing a marked delay in growth
provoked by the inducible reduction of pbp1 gene expression
(Figure 2A), we investigated the status of DNA replication in
SA13011 undergoing OXA-mediated selection. DNA synthesis
was measured by following the incorporation of
methyl-[3H]thymine into SA13011 cells grown in the absence and presence of
OXA (0.5 mg/ml) during the selection process, as described in
Methods. Consistent with previous studies investigating the
phenomenon of OXA-mediated HeR/HoR selection , the
selected homotypic derivatives displayed slow kinetic of growth
when compared to the heterotypic variant (e.g., OD600 nm 8 h of
culture: 1.051/0.2; CFU/ml: 7.56108/7.26105,
SA13011-HeR/HoR, respectively; Figure 3A). OXA-mediated effects on cell
growth were also reflected in the rate of DNA synthesis as shown
by a significant reduction in the rate of methyl-[3H]-thymine
incorporation into DNA in cells growing in the presence of OXA
[e.g., cpm (6103) 8 h: 10.9/0.9, SA13011-HeR/-HoR,
respectively; Figure 3B].
Antibacterial efficacy of combining PBP1 and PBP2/
PBP2a inhibitors against HoR-MRSA strains
The results showing that PBP1 plays a pivotal role in
determining the signaling responsible of increasing the mutation
rate, a critical step in b-lactam-mediated HeR/HoR selection, plus
the fact that this process is accompanied by a compensatory
increase in the expression of not only pbp1 but also pbp2, mecA, and
pbp4 (Figure 1), led us to hypothesize that simultaneous
inactivation of PBP1 (because of its effects on the recA/SOS
response) and one of the other PBPs (e.g., PBP2, MecA, or PBP4)
may result in synergistic antibacterial effects and provide an
effective therapy against HoR-MRSA strains. To test this concept,
we used both in-vitro synergy time-kill curves and an in-vivo wax
worm model, as previously described [34,45]. In-vitro synergy-kill
analysis of interactions between IMP and b-lactams currently used
in clinical therapeutics (CTX [PBP2], BAL [PBP2a], and FOX
[PBP4]) against SA13011-HoR strain were performed at 0, 2, 4, 8,
and 24 h using MH broth; an initial inoculum of 16106 CFU/ml
in the presence of K MICs (Table 1) of IMP and each of the
blactams were tested. The size of the inoculum was determined by
matching bacterial counts commonly achieved in all target tissues
of animals with experimental infective endocarditis [46,47]. At
these concentrations, neither antibiotic alone displayed significant
bactericidal effects (Figure 4AC). While IMP and FOX
administered together have no bactericidal effect (Figure 3A),
the combination of IMP with either BAL (PBP2a) or CTX (PBP2)
were highly synergistic as demonstrated by cell killing at 24 h
$4log CFU vs. single agents and the initial inoculum (Figure 4BC).
The population remaining after exposure to either IMP/BAL or
IMP/CTX does not represent a subpopulation of resistant cells
and correspond to metabolic inert cells, as we verified by E-Test
(data not shown). These results suggested that the combination
between IMP (PBP1) and antibiotics blocking either PBP2 or
PBP2a may have a major impact as an anti-infective alternative in
To investigate whether regimens identified by in-vitro time-kill
analysis combining IMP with b-lactams targeting either PBP2a
(BAL) or PBP2 (CTX) may enhance the in-vivo therapeutic efficacy
over that of each single agent against HoR-MRSA strains, we used
a model of wax worms [34,45,48]. Larvae of the greater wax moth
(Galleria mellonella) were shown recently to represent an alternative
to vertebrates as a host model for studying pathogenic microbes,
virulence, and therapeutic regimens [45,48,49], including the
efficacy of antistaphylococcal agents . Groups of larvae (10/
group) were inoculated with a bacterial suspension containing
SA13011-HoR (105 bacteria/worm) as we previously described
, incubated for 2 h at 37uC, after which IMP (10 mg/kg), BAL
(5 mg/kg), CTX (10 mg/kg), or the corresponding combinations
IMP/BAL, and IMP/CTX were administrated (0 h), and
reincubated for 24 h at 37uC. Additionally, a group was treated
with NAF (5 mg/kg; targets both PBP1 and PBP3; OXA or
CLOX were replaced by NAF because of its clinical use
recommendation) and IMP/NAF. An uninfected control group
received PBS treatment to control for multiple injections. After the
first 24 h incubation, treatment was repeated. Worms were
checked daily, and recorded for any deaths for a total of 10 days.
Uninfected worms treated with PBS, single drugs or the
combinations IMP/NAF, IMP/CTX, or IMP/BAL showed
100/90% survival at day 10 of treatment (Figure 5A). On the
other hand, groups of HoR-MRSA-injected worms untreated
(PBS), or treated with single drugs, displayed low survival rates
(#5010%, day 10; Figure 5B). By contrast, groups of worms
treated with the combinations IMP/CTX and IMP/BAL resulted
in survival rates of 90% at day 10 (Figure 5B). Interestingly, the
combination of IMP/NAF, in which both antibiotics target PBP1
(PBP3 expression levels were very low and not affected in the
SA13011-HoR strain; Figure 1) was ineffective. Thus, these results
were in agreement with data from the in-vitro time-kill experiments
(Figure 4) showing significant synergistic interactions between
IMP, which targets PBP1, and b-lactams targeting PBPs whose
expression levels, as a compensatory mechanism, appeared
increased, i.e. PBP2 (CTX) or PBP2a (BAL).
Resistance in S. aureus to widely used antibiotics can be achieved
through the selection of single mutations or through gene
acquisition by horizontal transfer [17,50]. Both of these processes
have been linked to the RecA-mediated SOS response [17,50].
The bacterial SOS response is a conserved regulatory network that
is induced in response to DNA damage . Its activation in vitro
results in the emergence of resistance to antibiotics, an observation
that has led to the speculation that this is an important clinical
mechanism of resistance acquisition . In previous studies, we
demonstrated that, in clinical heterogeneous MRSA strains, the
SOS response is activated in response to exposure tob-lactam
antibiotics, a mechanism associated with the selection of highly
resistant derivative variants . Taking into account that
blactams are not DNA damage agents but cell wall inhibitors, we
hypothesized that inhibition of the cell wall machinery may be
involved in the triggering of SOS-mediated recA activation. The
present results showing that recA was activated in cultures treated
only with antibiotics targeting PBP1 (i.e., IMP or OXA), but not
other PBPs (i.e., PBP2, -2a, -3, or 4), suggest that interference with
this particular PBP plays a key role in b-lactam-mediated
triggering of the SOS response. These results are consistent with
evidence showing the role that PBP1 plays in cell division and
separation in S. aureus [6,7], highlighting the extent to which
targeting it with b-lactams affects the growth of cells under
antibiotic pressure. In the context of PBP1 inactivation by
exposure to b-lactams, studies in E.coli have shown the
synchronization existing between DNA replication and cell division
[43,44]. As mentioned above, despite that b-lactams are not
considered DNA damaging agents, and as our results suggest, and
it may be plausible to postulate that by interfering with PBP1
functions, b-lactam antibiotics may lead to cell division arrest and,
in turn, could affect replication and generate the presence of
single-stranded DNA, which may then contribute to the signals
involved in the triggering of the SOS response and hypermutable
state. Further studies are needed to test this hypothesis and are
currently planned in our laboratory.
In the present work, we demonstrated that inhibition of PBP1
was linked to both recA activation and the concomitant increase in
mutation rate, a critical step in b-lactam-mediated selection of the
highly resistant HoR phenotype. Moreover, following PBP1
inhibition and recA activation, compensatory up-regulation of
pbp2, pbp2a, and pbp4 were observed. This observation raised the
question of whether simultaneous targeting of PBP1 and
compensatory PBP2, PBP2a, or PBP4 may provide a therapeutic
advantage to limit the growth of HoR-MRSA, both in-vitro and in
vivo. Our results indicate that, in fact, combining IMP (PBP1) with
either CTX (PBP2) or BAL (PBP2a) results in a synergistic activity
against HoR-MRSA. By contrast, simultaneous inhibition of PBP1
(IMP) and PBP4 (FOX), or using a combination of IMP/NAF,
where both compounds target PBP1, did not result in increased
antibiotic activity. Together, these observations were consistent
with the notion that PBP1, which determines the signaling that
leads to RecA activation and increased mutation rate, requires the
cooperation of both PBP2 and MecA in order to sustain the highly
resistant HoR-MRSA phenotype. Furthermore, this rationale is
consistent with the idea that PBP2A is not the sole determinant for
b-lactam resistance in MRSA, as previously suggested [11,12,52].
In addition, these results also show that, although PBP4 has been
shown to be essential for expression of resistance in CA-MRSA, it
does not play a significant role in heterogeneous MRSA strains
which are mostly linked to HA-MRSA infections .
It is well known that b-lactam antibiotics are inducers of the
SOS response; [53,54] however, less is known about the
relationship between SOS response and cell wall-mediated
response in MRSA strains. The present work provides evidence
that b-lactam-induced impairment of PBP1 function is responsible
for generating the signal and triggering SOS response-mediated
RecA activation, which, in turn, leads to an increase in the
mutation rate. In summary, the present results offer new insights
into the complex biology underlying b-lactam-induced HeR/HoR
selection and acquisition of the highly resistant MRSA phenotype.
The results from this study have not been presented at any scientific
meetings. Special thanks to Philip Randall, from the Office of Academic
Development, Department of Pathology and Genomic Medicine, The
Methodist Hospital, for his assistance with manuscript editing.
Conceived and designed the experiments: RRR AER. Performed the
experiments: KBP SR CRS AER RRR. Analyzed the data: KBP CRS
RRR AER. Contributed reagents/materials/analysis tools: KBP CRS
RRR AER. Wrote the paper: KBP RRR AER.
1. Herold BC , Immergluck LC , Maranan MC , Lauderdale DS , Gaskin RE , et al. ( 1998 ) Community-acquired methicillin-resistant Staphylococcus aureus in children with no identified predisposing risk . JAMA 279 : 593 - 598 .
2. Holtje JV ( 1998 ) Growth of the stress-bearing and shape-maintaining murein sacculus of Escherichia coli . Microbiol Mol Biol Rev 62 : 181 - 203 .
3. Pinho MG , Ludovice AM , Wu S , de Lencastre H ( 1997 ) Massive reduction in methicillin resistance by transposon inactivation of the normal PBP2 in a methicillin-resistant strain of Staphylococcus aureus . Microb Drug Resist 3 : 409 - 413 .
4. Pinho MG , de Lencastre H , Tomasz A ( 2001 ) An acquired and a native penicillin-binding protein cooperate in building the cell wall of drug-resistant staphylococci . Proc Natl Acad Sci U S A 98 : 10886 - 10891 .
5. Wada A , Watanabe H ( 1998 ) Penicillin-binding protein 1 of Staphylococcus aureus is essential for growth . J Bacteriol 180 : 2759 - 2765 .
6. Pereira SF , Henriques AO , Pinho MG , de Lencastre H , Tomasz A ( 2007 ) Role of PBP1 in cell division of Staphylococcus aureus . J Bacteriol 189 : 3525 - 3531 .
7. Pereira SF , Henriques AO , Pinho MG , de LH, Tomasz A ( 2009 ) Evidence for a dual role of PBP1 in the cell division and cell separation of Staphylococcus aureus . Mol Microbiol 72 : 895 - 904 .
8. Spratt BG ( 1975 ) Distinct penicillin binding proteins involved in the division, elongation, and shape of Escherichia coli K12 . Proc Natl Acad Sci U S A 72 : 2999 - 3003 .
9. Botta GA , Park JT ( 1981 ) Evidence for involvement of penicillin-binding protein 3 in murein synthesis during septation but not during cell elongation . J Bacteriol 145 : 333 - 340 .
10. Costa T , Priyadarshini R , Jacobs-Wagner C ( 2008 ) Localization of PBP3 in Caulobacter crescentus is highly dynamic and largely relies on its functional transpeptidase domain . Mol Microbiol 70 : 634 - 651 .
11. Pinho MG , Filipe SR , de Lencastre H , Tomasz A ( 2001 ) Complementation of the essential peptidoglycan transpeptidase function of penicillin-binding protein 2 (PBP2) by the drug resistance protein PBP2A in Staphylococcus aureus . J Bacteriol 183 : 6525 - 6531 .
12. Memmi G , Filipe SR , Pinho MG , Fu Z , Cheung A ( 2008 ) Staphylococcus aureus PBP4 is essential for beta-lactam resistance in community-acquired methicillinresistant strains . Antimicrob Agents Chemother 52 : 3955 - 3966 .
13. Finan JE , Rosato AE , Dickinson TM , Ko D , Archer GL ( 2002 ) Conversion of oxacillin-resistant staphylococci from heterotypic to homotypic resistance expression . Antimicrob Agents Chemother 46 : 24 - 30 .
14. Forbes BA , Bombicino K , Plata K , Cuirolo A , Webber D , et al. ( 2008 ) Unusual form of oxacillin resistance in methicillin-resistant Staphylococcus aureus clinical strains . Diagn Microbiol Infect Dis 61 : 387 - 395 .
15. Cuirolo A , Plata K , Rosato AE ( 2009 ) Development of homogeneous expression of resistance in methicillin-resistant Staphylococcus aureus clinical strains is functionally associated with a beta-lactam-mediated SOS response . J Antimicrob Chemother 64 : 37 - 45 .
16. Tomasz A , Nachman S , Leaf H ( 1991 ) Stable classes of phenotypic expression in methicillin-resistant clinical isolates of staphylococci . Antimicrob Agents Chemother 35 : 124 - 129 .
17. Cirz RT , Jones MB , Gingles NA , Minogue TD , Jarrahi B , et al. ( 2007 ) Complete and SOS-mediated response of Staphylococcus aureus to the antibiotic ciprofloxacin . J Bacteriol 189 : 531 - 539 .
18. Miller C , Thomsen LE , Gaggero C , Mosseri R , Ingmer H , et al. ( 2004 ) SOS response induction by beta-lactams and bacterial defense against antibiotic lethality . Science 305 : 1629 - 1631 .
19. Perez-Capilla T , Baquero MR , Gomez-Gomez JM , Ionel A , Martin S , et al. ( 2005 ) SOS-independent induction of dinB transcription by beta-lactammediated inhibition of cell wall synthesis in Escherichia coli . J Bacteriol 187 : 1515 - 1518 .
20. National Commitee for Clinical Laboratory Standards . ( 2007 ) Performance standards for antimicrobial disk susceptibility tests . Approved standard M2-A8 . 8th.
21. Plata KB , Rosato RR , Rosato AE ( 2011 ) Fate of mutation rate depends on agr locus expression during oxacillin-mediated heterogeneous-homogeneous selection in methicillin-resistant Staphylococcus aureus clinical strains . Antimicrob Agents Chemother 55 : 3176 - 3186 .
22. Friedman L , Alder JD , Silverman JA ( 2006 ) Genetic changes that correlate with reduced susceptibility to daptomycin in Staphylococcus aureus . Antimicrob Agents Chemother 50 : 2137 - 2145 .
23. Julian K , Kosowska-Shick K , Whitener C , Roos M , Labischinski H , et al. ( 2007 ) Characterization of a daptomycin-nonsusceptible vancomycin-intermediate Staphylococcus aureus strain in a patient with endocarditis . Antimicrob Agents Chemother 51 : 3445 - 3448 .
24. Pillai SK , Gold HS , Sakoulas G , Wennersten C , Moellering RC , Jr. , et al. ( 2007 ) Daptomycin nonsusceptibility in Staphylococcus aureus with reduced vancomycin susceptibility is independent of alterations in MprF . Antimicrob Agents Chemother 51 : 2223 - 2225 .
25. Bisognano C , Kelley WL , Estoppey T , Francois P , Schrenzel J , et al. ( 2004 ) A recA-LexA-dependent pathway mediates ciprofloxacin-induced fibronectin binding in Staphylococcus aureus . J Biol Chem 279 : 9064 - 9071 .
26. Rosato A , Vicarini H , Bonnefoy A , Chantot JF , Leclercq R ( 1998 ) A new ketolide, HMR 3004, active against streptococci inducibly resistant to erythromycin . Antimicrob Agents Chemother 42 : 1392 - 1396 .
27. Novick R ( 1967 ) Properties of a cryptic high-frequency transducing phage in Staphylococcus aureus . Virology 33 : 155 - 166 .
28. Luong TT , Lee CY ( 2006 ) The arl locus positively regulates Staphylococcus aureus type 5 capsule via an mgrA-dependent pathway . Microbiology 152 : 3123 - 3131 .
29. Cirz RT , Romesberg FE ( 2006 ) Induction and inhibition of ciprofloxacin resistance-conferring mutations in hypermutator bacteria . Antimicrob Agents Chemother 50 : 220 - 225 .
30. Rosato AE , Craig WA , Archer GL ( 2003 ) Quantitation of mecA transcription in oxacillin-resistant Staphylococcus aureus clinical isolates . J Bacteriol 185 : 3446 - 3452 .
31. Goldstein F , Perutka J , Cuirolo A , Plata K , Faccone D , et al. ( 2007 ) Identification and phenotypic characterization of a beta-lactam-dependent, methicillin-resistant Staphylococcus aureus strain . Antimicrob Agents Chemother 51 : 2514 - 2522 .
32. Livak KJ , Schmittgen TD ( 2001 ) Analysis of relative gene expression data using real-time quantitative PCR and the 2(- Delta Delta C(T)) Method. Methods 25 : 402 - 408 .
33. Howden BP , McEvoy CR , Allen DL , Chua K , Gao W , et al. ( 2011 ) Evolution of Multidrug Resistance during Staphylococcus aureus Infection Involves Mutation of the Essential Two Component Regulator WalKR . PLoS Pathog 7 : e1002359 .
34. Desbois AP , Coote PJ ( 2011 ) Wax moth larva (Galleria mellonella): an in vivo model for assessing the efficacy of antistaphylococcal agents . J Antimicrob Chemother 66 : 1785 - 1790 .
35. Olsen RJ , Watkins ME , Cantu CC , Beres SB , Musser JM ( 2011 ) Virulence of serotype M3 Group A Streptococcus strains in wax worms (Galleria mellonella larvae) . Virulence 2: 111 - 119 . 14338
36. Akins RL , Rybak MJ ( 2001 ) Bactericidal activities of two daptomycin regimens against clinical strains of glycopeptide intermediate-resistant Staphylococcus aureus, vancomycin-resistant Enterococcus faecium, and methicillin-resistant Staphylococcus aureus isolates in an in vitro pharmacodynamic model with simulated endocardial vegetations . Antimicrob Agents Chemother 45 : 454 - 459 .
37. Rose WE , Rybak MJ , Kaatz GW ( 2007 ) Evaluation of daptomycin treatment of Staphylococcus aureus bacterial endocarditis: an in vitro and in vivo simulation using historical and current dosing strategies . J Antimicrob Chemother 60 : 334 - 340 .
38. Yang Y , Bhachech N , Bush K ( 1995 ) Biochemical comparison of imipenem, meropenem and biapenem: permeability, binding to penicillin-binding proteins, and stability to hydrolysis by beta-lactamases . J Antimicrob Chemother 35 : 75 - 84 .
39. Georgopapadakou NH , Smith SA , Bonner DP ( 1982 ) Penicillin-binding proteins in a Staphylococcus aureus strain resistant to specific beta-lactam antibiotics . Antimicrob Agents Chemother 22 : 172 - 175 .
40. Dumitrescu O , Choudhury P , Boisset S , Badiou C , Bes M , et al. ( 2011 ) Betalactams interfering with PBP1 induce Panton-Valentine leukocidin expression by triggering sarA and rot global regulators of Staphylococcus aureus . Antimicrob Agents Chemother 55 : 3261 - 3271 .
41. Aubry-Damon H , Soussy CJ , Courvalin P ( 1998 ) Characterization of mutations in the rpoB gene that confer rifampin resistance in Staphylococcus aureus . Antimicrob Agents Chemother 42 : 2590 - 2594 .
42. Trong HN , Prunier AL , Leclercq R ( 2005 ) Hypermutable and fluoroquinoloneresistant clinical isolates of Staphylococcus aureus . Antimicrob Agents Chemother 49 : 2098 - 2101 .
43. Garcia-del PF , Pisabarro AG , de la Rosa EJ , de Pedro MA ( 1987 ) Modulation of cell wall synthesis by DNA replication in Escherichia coli during initiation of cell growth . J Bacteriol 169 : 2410 - 2416 .
44. Thanbichler M ( 2010 ) Synchronization of chromosome dynamics and cell division in bacteria . Cold Spring Harb Perspect Biol 2 : Jan; 2 ( 1 )
45. Peleg AY , Jara S , Monga D , Eliopoulos GM , Moellering RC , Jr. , et al. ( 2009 ) Galleria mellonella as a model system to study Acinetobacter baumannii pathogenesis and therapeutics . Antimicrob Agents Chemother 53 : 2605 - 2609 . AAC. 01533 - 08
46. Weidenmaier C , Peschel A , Kempf VA , Lucindo N , Yeaman MR , et al. ( 2005 ) Dlt. Infect Immun 73 : 8033 - 8038 .
47. Xiong YQ , Kupferwasser LI , Zack PM , Bayer AS ( 1999 ) Comparative efficacies of liposomal amikacin (MiKasome) plus oxacillin versus conventional amikacin plus oxacillin in experimental endocarditis induced by Staphylococcus aureus: microbiological and echocardiographic analyses . Antimicrob Agents Chemother 43 : 1737 - 1742 .
48. Mehta S , Singh C , Plata KB , Chanda PK , Paul A , et al. ( 2012 ) beta-lactams increase the antibacterial activity of daptomycin against clinical MRSA strains and prevent selection of DAP-resistant derivatives . Antimicrob Agents Chemother 56 : 92 - 102
49. Kavanagh K , Reeves EP ( 2004 ) Exploiting the potential of insects for in vivo pathogenicity testing of microbial pathogens . FEMS Microbiol Rev 28 : 101 - 112 .
50. Ubeda C , Maiques E , Knecht E , Lasa I , Novick RP , et al. ( 2005 ) Antibioticinduced SOS response promotes horizontal dissemination of pathogenicity island-encoded virulence factors in staphylococci . Mol Microbiol 56 : 836 - 844 .
51. Hocquet D , Llanes C , Thouverez M , Kulasekara HD , Bertrand X , et al. ( 2012 ) Evidence for induction of integron-based antibiotic resistance by the SOS response in a clinical setting . PLoS Pathog 8: e1002778. 10.1371/journal. ppat.1002778. Epub 2012 Jun 2012
52. de Lencastre H , Wu SW , Pinho MG , Ludovice AM , Filipe S , et al. ( 1999 ) Antibiotic resistance as a stress response: complete sequencing of a large number of chromosomal loci in Staphylococcus aureus strain COL that impact on the expression of resistance to methicillin . Microb Drug Resist 5 : 163 - 175 .
53. Maiques E , Ubeda C , Campoy S , Salvador N , Lasa I , et al. ( 2006 ) beta-lactam antibiotics induce the SOS response and horizontal transfer of virulence factors in Staphylococcus aureus . J Bacteriol 188 : 2726 - 2729 .
54. Miller C , Ingmer H , Thomsen LE , Skarstad K , Cohen SN ( 2003 ) DpiA binding to the replication origin of Escherichia coli plasmids and chromosomes destabilizes plasmid inheritance and induces the bacterial SOS response . J Bacteriol 185 : 6025 - 6031 .
55. Bruckner R , Wagner E , Gotz F ( 1993 ) Characterization of a sucrase gene from Staphylococcus xylosus . J Bacteriol 175 : 851 - 857 .