A potential space-making role in cell wall biogenesis for SltB1and DacB revealed by a beta-lactamase induction phenotype in Pseudomonas aeruginosa

ABSTRACT Pseudomonas aeruginosa encodes the beta-lactamase AmpC, which promotes resistance to beta-lactam antibiotics. Expression of ampC is induced by anhydro-muropeptides (AMPs) released from the peptidoglycan (PG) cell wall upon beta-lactam treatment. AmpC can also be induced via genetic inactivation of PG biogenesis factors such as the endopeptidase DacB that cleaves PG crosslinks. Mutants in dacB occur in beta-lactam-resistant clinical isolates of P. aeruginosa, but it has remained unclear why DacB inactivation promotes ampC induction. Similarly, the inactivation of lytic transglycosylase (LT) enzymes such as SltB1 that cut PG glycans has also been associated with ampC induction and beta-lactam resistance. Given that LT enzymes are capable of producing AMP products that serve as ampC inducers, this latter observation has been especially difficult to explain. Here, we show that ampC induction in sltB1 or dacB mutants requires another LT enzyme called MltG. In Escherichia coli, MltG has been implicated in the degradation of nascent PG strands produced upon beta-lactam treatment. Accordingly, in P. aeruginosa sltB1 and dacB mutants, we detected the MltG-dependent production of pentapeptide-containing AMP products that are signatures of nascent PG degradation. Our results therefore support a model in which SltB1 and DacB use their PG-cleaving activity to open space in the PG matrix for the insertion of new material. Thus, their inactivation mimics low-level beta-lactam treatment by reducing the efficiency of new PG insertion into the wall, causing the degradation of some nascent PG material by MltG to produce the ampC-inducing signal. IMPORTANCE Inducible beta-lactamases like the ampC system of Pseudomonas aeruginosa are a common determinant of beta-lactam resistance among gram-negative bacteria. The regulation of ampC is elegantly tuned to detect defects in cell wall synthesis caused by beta-lactam drugs. Studies of mutations causing ampC induction in the absence of drug therefore promise to reveal new insights into the process of cell wall biogenesis in addition to aiding our understanding of how resistance to beta-lactam antibiotics arises in the clinic. In this study, the ampC induction phenotype for mutants lacking a glycan-cleaving enzyme or an enzyme that cuts cell wall crosslinks was used to uncover a potential role for these enzymes in making space in the wall matrix for the insertion of new material during cell growth.

Expression of the ampC gene is regulated in response to drug treatment.In the absence of antibiotic, it is expressed at low levels.However, treatment with some beta-lac tams like cefoxitin, referred to as beta-lactamase inducers, results in potent induction of ampC expression and resistance.Other beta-lactams like piperacillin and ceftazidime (Caz) are not ampC inducers (4).These drugs therefore have anti-pseudomonal activity despite the ability of AmpC to hydrolyze them.Mutants with defects in ampC regulation causing constitutive beta-lactamase production are resistant to piperacillin and Caz.They are known to arise in the clinic and can result in treatment failures (5-9).There has thus been considerable interest in understanding ampC regulation and the mechanism by which mutations promote its aberrant overexpression.
The expression level of ampC is linked to the status of peptidoglycan (PG) synthesis, and it responds to signals produced when beta-lactam antibiotics disrupt the process (3,(10)(11)(12)(13).The PG cell wall surrounds most bacteria and is essential for maintaining cellular integrity.It is composed of glycan chains with a repeating disaccharide unit of N-ace tylmuramic acid (MurNAc) and N-acetylglucosamine (GlcNAc).A pentapeptide that in most gram-negative bacteria has the sequence L-Ala-ɣ-D-Glu-meso-diaminopimelic acid (mDAP)-D-Ala-D-Ala is attached to the MurNAc sugar (14).It is used to form crosslinks between glycan chains, generating the matrix-like structure of the wall (Fig. 1) (15).
Two different types of synthases build the PG layer.The Class A penicillin-binding proteins (aPBPs) possess both glycosyltransferase (GTase) and transpeptidase (TPase) activities in a single polypeptide for the polymerization and crosslinking of PG, respectively.The other major synthases are composed of complexes between a SEDS family protein with GTase activity and a Class B penicillin-binding protein (bPBP) with TPase activity (21)(22)(23).Because the PG matrix is continuous, the insertion of new material requires the action of PG-cleaving enzymes to make space for the incoming nascent glycans (15).Currently, endopeptidases (EPs) that cut the peptide crosslinks are thought to be the main PG-processing enzymes that function as space makers (15,(24)(25)(26)(27). to the MurNAc sugars is a pentapeptide used to form crosslinks between adjacent glycans.PG synthesis starts in the cytoplasm, followed by the generation of lipid-linked precursors.Polymerization (GTase) and crosslinking (TPase) reactions at the membrane surface are used to form and insert nascent glycans into the mature matrix.The crosslinking reaction and carboxypeptidase enzymes act to rapidly convert the pentapeptide side chains to tetrapeptides via removal of the terminal D-Ala.The mature PG is subject to degradation by LT and EPs to generate anydro-MurNAc (anh-MurNAc) containing muropeptide (AMP) turnover products, which are imported to the cytoplasm by AmpG and recycled.The tetrapeptide AMP products, which are normally converted to tripeptide species upon import, are not thought to be good inducers for the activation of ampC expression by AmpR such that ampC is repressed in the absence of drug (16).(Right) Upon beta-lactam treatment, TPases are inhibited, and uncrosslinked nascent glycans are formed.These glycans are rapidly degraded generating pentapeptide AMP products that are also transported to the cytoplasm by AmpG (17,18).These pentapeptide products are thought to be potent activators of AmpR for the induction of ampC expression (12,13,19,20).Thus, ampC expression is a sensitive reporter for problems with PG crosslinking.
Beta-lactams covalently modify the TPase active sites of PBPs and inhibit PG crosslinking (28).These drugs do not block the GTase activity of the polymerase enzymes.Thus, uncrosslinked PG glycans are produced following drug treatment (29).In the related gram-negative bacterium Escherichia coli, these uncrosslinked strands have been shown to be rapidly degraded by lytic transglycosylase (LT) enzymes (29,30).LTs cleave the glycan strand and generate disaccharide-peptide products with a 1,6-anhy dro linkage on the MurNAc sugar (31).These so-called anhydro-muropeptides (AMPs) are produced by LTs during normal growth as these enzymes help promote the high turnover of the mature PG observed per generation (ca.40%/generation) (Fig. 1) (32).In this case, the turnover products are primarily in the tetrapeptide form (33) because the terminal D-Ala of the stem peptide is either removed in the process of crosslinking or rapidly trimmed by enzymes called D-Ala-D-Ala carboxypeptidases.However, when nascent PG is processed by LTs during beta-lactam treatment, the AMPs produced are in the pentapeptide form (29).These AMPs are likely to be the preferred inducer of the AmpR regulator, converting it to an activator of ampC expression (Fig. 1) (12,13,19,20).Thus, ampC regulation is elegantly tuned to detect problems with nascent PG crosslinking as a proxy for the presence of beta-lactams.
Genetic inactivation of DacB, a PG-processing enzyme with both PG EP and carbox ypeptidase activity (34), has been known to induce ampC expression and promote beta-lactam resistance for some time (7)(8)(9).However, the mechanism by which the ampC-inducing signal is generated in dacB mutants has remained unclear.Mutations that inactivate LT enzymes have also been found to increase the expression of beta-lac tamase genes in several different gram-negative organisms (35)(36)(37)(38).Given that LTs are typically associated with the production of the ampC-inducing signal, these results have been difficult to explain.
In this report, we used a P. aeruginosa strain with a lacZ reporter gene fused to the ampC promoter (P ampC ::lacZ) (39) to find that defects in the LT enzyme SltB1 aberrantly induce ampC.In an effort to uncover the mechanism that leads to ampC induction in sltB1 mutants, we found that the activation of ampC expression in cells lacking SltB1 or DacB requires MltG, an LT enzyme previously implicated in the turnover of nascent PG in E. coli and the induction of ampC following beta-lactam treatment in P. aeruginosa (30,40,41).Accordingly, in sltB1 and dacB mutants, we detected the MltG-dependent production of pentapeptide-containing AMP products that are signatures of nascent PG degradation.Our results therefore support a model in which SltB1 and DacB use their PG-cleaving activity to open space in the PG matrix for the insertion of new material.Thus, in the absence of these enzymes, the efficiency of incorporation of new PG strands into the wall is reduced, mimicking low-level beta-lactam treatment to cause a fraction of nascent PG material to be degraded by MltG to produce the ampC-inducing signal.

Inactivation of sltB1 induces ampC expression and promotes beta-lactam resistance
A wild-type (WT) P. aeruginosa strain (PAO1) carrying a P ampC ::lacZ reporter (39) inserted at the attB locus was mutagenized with a transposon.The resulting mutant library was plated on agar containing X-gal to identify insertions causing ampC expression in the absence of beta-lactam treatment as a mean to identify factors required for proper PG biogenesis.Prior screens using this library identified transposon mutants in the mupP gene, which led to its functional characterization as an enzyme important for the recycling of PG fragments (42).We continued to screen the library and identified an additional mutant forming solid blue colonies on X-gal agar, a phenotype that is indicative of aberrant ampC induction.PCR-based mapping revealed that this isolate had a transposon inserted in the sltB1 (PA4001) gene (Fig. 2A), encoding the LT enzyme SltB1 that is related to E. coli MltB (43).
A deletion mutation of sltB1 was previously found to promote beta-lactam resist ance (35,36).However, in these studies, the authors did not detect elevated AmpC production in an sltB1 mutant by nitrocefin hydrolysis assays or immunoblotting despite observing an ampC requirement for the resistance phenotype and an increased AmpC level when the sltB1 mutation was combined with deletion of dacB (35,36).It was therefore concluded that beta-lactam resistance upon SltB1 inactivation was not due to ampC induction but instead was likely to result from the inactivation of a lysis pathway involving cell wall damage caused by SltB1 (35,36).The identification of the transposon insertion in sltB1 in our screen argues against this interpretation and for a more direct role of SltB1 inactivation in ampC induction.
To validate the results from the screen, an in-frame deletion of sltB1 similar to the previously published deletion (35,36) was generated in the reporter strain.An aliquot of the mutant culture was spotted on agar containing X-gal alongside cultures of WT, a ∆dacB mutant known to promote high-level ampC induction, and a mutant lacking ampR that is defective for ampC expression (8,46).As expected, the spots from the WT and genes encode the RodA-PBP2 complex that forms the essential PG synthase of the cell elongation system (15).The rlpA gene encodes an LT enzyme that functions in cell division by promoting daughter cell separation (44), and dacC encodes a carboxypeptidase that trims PG peptides (45).(B) Culture aliquots (5 µL) of strains CF262 (PAO1 WT), CF268 (∆dacB), CF1143 (∆sltB1), and CF604 (∆ampR) containing the P ampC ::lacZ reporter were spotted onto lysogeny broth (LB) agar containing X-gal (50 µg/mL), grown overnight at 30°C, and photographed.(C) Beta-galactosidase activity in Miller units was measured in liquid cultures of the indicated strains.Results shown are the average of three assays with two or three biological replicates per strain, except for the sltB1::Tn strain (no biological replicate), and the error bars represent the standard deviation.A one-way analysis of variance revealed that there was a statistically significant difference in ampC expression level between at least two strains [F(106610.925,6211.075)= 188.81114,P < 0.00001].Tukey's honestly significant difference test for multiple comparisons found that the mean value of ampC expression was significantly different between WT and ∆sltB1, WT and sltB1::Tn, and ∆sltB1 and sltB1::Tn (P = 0.01).(D) Cultures of strains PAO1 (WT), CF155 (∆dacB), and CF1105 (∆sltB1) were serially diluted, and 5 µL of each dilution was spotted onto LB agar supplemented with Caz (10 µg/mL) or piperacillin (10 µg/mL) as indicated.
∆ampR culture remained white, whereas that from the ∆dacB culture turned blue (Fig. 2B).The spot from the ∆sltB1 culture also turned blue, indicating that SltB1 inactivation induces the ampC promoter in the absence of beta-lactams (Fig. 2B).Quantification of beta-galactosidase activity confirmed that the P ampC ::lacZ reporter was induced in the ∆sltB1 mutant and the original sltB1::Tn isolate relative to WT (Fig. 2C).Notably, the transposon insertion allele led to a greater induction of the reporter than the deletion, which is potentially due to effects of the insertion on the expression of nearby genes encoding other PG biogenesis proteins (Fig. 2A).We conclude that inactivation of SltB1 induces expression from the ampC promoter.
To monitor the effect of SltB1 inactivation on native P ampC induction, we tested the beta-lactam resistance of mutant cells and measured the production of AmpC using a nitrocefin hydrolysis assay.Consistent with prior results, the ∆sltB1 strain was resistant to the antipseudomonal beta-lactams Caz and piperacillin, as was a control ∆dacB strain (Fig. 2D).Similarly, and in contrast to previous work (35), a ∆sltB1 mutant with an empty vector showed elevated AmpC activity in the nitrocefin assay relative to WT cells (Fig. 3A).Deletion of sltB1 did not strongly affect the induction of AmpC activity by the beta-lactam cefoxitin (Fig. 3B).The sltB1 gene is in a putative operon that includes several genes encoding PG synthesis and remodeling proteins, including the downstream rlpA and dacC genes that encode an LT enzyme and a D-Ala-D-Ala carboxypeptidase, respectively (Fig. 2A) (44,45).Notably, normal beta-lactam sensitivity and AmpC activity were restored to ∆sltB1 cells upon expression of sltB1 from a plasmid (Fig. S1), indicating that the phenotype of the deletion allele was caused by the inactivation of SltB1 and not an effect of the deletion on the expression of the nearby genes encoding PG biogenesis factors.Expression of a catalytic mutant of SltB1, SltB1(E135A), failed to complement the beta-lactam resistance phenotype of the sltB1 deletion (Fig. S1), indicating that it is the loss of SltB1 activity that results in ampC induction.

AmpC production in ∆sltB1 cells is induced via the canonical mechanism
The results presented thus far suggest that mutants lacking SltB1 are resistant to betalactam treatment via the induction of ampC expression as opposed to alternative mechanisms proposed previously (35,36).To further test this possibility, we investigated the requirements for resistance in the ∆sltB1 background.Deletion of ampC or the ampR gene encoding the ampC transcriptional activator eliminated the beta-lactam resistance phenotype of ∆sltB1 cells and resulted in the loss of AmpC accumulation as determined by immunoblot (Fig. 4A and B).Beta-lactam resistance and AmpC accumulation were also blocked in the ∆sltB1 mutant by inactivation of AmpG, the transporter that imports the AMP products from cell wall degradation that are sensed by AmpR to activate ampC expression (Fig. 1, 4A and B) (17,18).We therefore conclude that ampC is being induced in the ∆sltB1 cells by the canonical induction mechanism.

MltG is required for ampC induction in ∆sltB1 and ∆dacB cells
In E. coli, uncrosslinked PG strands are generated upon beta-lactam treatment, and these strands are rapidly degraded by the LT enzymes Slt and MltG, with Slt playing the predominant role (29,30).The action of these LTs generates pentapeptide-containing AMP products that in P. aeruginosa would serve as inducers of ampC expression.We therefore investigated whether Slt or MltG is required for ampC induction in ∆sltB1.Deletion of slt did not have a strong effect on the piperacillin resistance of ∆sltB1 cells and actually appeared to increase AmpC production based on immunoblotting (Fig. 4A  and B).By contrast, inactivation of MltG restored piperacillin and Caz sensitivity to the ∆sltB1 mutant and resulted in AmpC being undetectable in these cells (Fig. 4A and B).We also tested the requirement for Slt or MltG for ampC induction in ∆dacB cells.As with ∆sltB1 cells, AmpC protein accumulation was abolished by MltG inactivation but not the loss of Slt function (Fig. 4C).Additionally, AmpC activity in ∆dacB cells as assessed by nitrocefin hydrolysis was also found to be dependent on MltG but not Slt (Fig. S2).Notably, the piperacillin and Caz sensitivity of ∆sltB1 ∆mltG and ∆dacB ∆mltG cells was test for multiple comparisons found that the mean value of nitrocefin hydrolysis level was significantly different between the ∆sltB1 + vector and all the other strains grown in presence of IPTG (WT + vector, WT + sltB1, and ∆sltB1 + sltB1), as indicated by the asterisks (P = 0.01).(B) Cells were grown in presence of the ampC inducer cefoxitin (50 µg/mL).A one-way analysis of variance revealed that there was a statistically significant difference in nitrocefin hydrolysis level between at least two strains [F(3, 33) = 10.66,P < 0.000076].Tukey's honestly significant difference test for multiple comparisons found that the mean value of nitrocefin hydrolysis level was significantly different between the ∆sltB1 (+vector or +sltB1) and WT (+vector or +sltB1) strains for the early time points, as indicated by the asterisk (P = 0.05).converted back to resistance upon expression of mltG(WT) from a plasmid but not by mltG(E217Q) encoding a catalytically dead MltG enzyme (Fig. 5; Fig. S3).Similarly, the induction of the P ampC ::lacZ reporter in ∆sltB1 ∆mltG and ∆dacB ∆mltG cells was restored by plasmid-based production of MltG(WT) but not MltG(E217Q) (Fig. S3).Thus, mutants lacking SltB1 or DacB similarly require MltG activity for ampC induction and beta-lactam resistance.

MltG-dependent accumulation of pentapeptide-containing AMPs in ∆sltB1 and ∆dacB cells
The results thus far are consistent with the production of uncrosslinked PG strands in ∆sltB1 and ∆dacB mutants that are turned over by MltG to produce the AMP products that induce ampC expression.To investigate this possibility, we first performed muropep tide analysis to determine whether the overall structure of PG was altered in cells lacking SltB1 or DacB.No major changes in PG composition were detected whether or not the ∆sltB1 or ∆dacB mutant cells possessed functional MltG (Fig. 6A through C).Thus, the bulk of PG synthesis appears to be proceeding normally in cells lacking SltB1 or DacB as expected based on the robust growth of the mutants.However, based on their ampC induction phenotypes, we suspected that in cells inactivated for DacB or SltB1, a subset of PG synthetic complexes may encounter problems with PG crosslinking such that uncrosslinked material is generated and rapidly degraded to produce pentapeptidecontaining AMP products.We therefore measured the levels of AMP products produced in WT and mutant cells.In cells with a functional PG recycling system, equivalent levels of tripeptide-containing AMPs were detected in all strains (Fig. 6D).Pentapeptide-contain ing AMPs were undetectable presumably due to their low steady-state levels in cells capable of recycling the AMPs (Fig. 6E).We therefore transitioned to monitoring AMPs in cells defective for the AmpD recycling amidase (Fig. 1), which would prevent AMP recycling and raise their steady-state levels to improve detection of the pentapeptidecontaining species.Cells lacking AmpD showed an elevated level of tripeptide-contain ing AMPs as expected, but there were no major increases in the levels of these species detected in ∆ampD cells inactivated for SltB1 or DacB whether or not they had an additional defect in MltG (Fig. 6D).However, in the ∆ampD background, cells defective for SltB1 or DacB showed a significant increase in the level of pentapeptide-containing AMPs that was MltG-dependent (Fig. 6E).This result indicates that there is an elevated level of nascent PG degradation in cells inactivated for SltB1 or DacB and that this degradation is most likely performed by MltG.We therefore infer that SltB1 and DacB are functioning as space-making enzymes to promote the efficient incorporation of new PG strands into the matrix for its expansion (see Discussion).

Overexpression of catalytically inactivated SltB1 induces ampC expression
PG-cleaving enzymes have been proposed to function as part of multi-protein com plexes that help coordinate their activities with those of PG synthases (14).Notably, SltB1 has been found to interact with PBP2 (43,47), a bPBP that is involved in cell elongation and shape determination (23).We therefore wondered whether SltB1 might have a limited number of binding sites within the PG layer and/or with protein-binding partners in the cell, which are required for its function.To test this possibility, we overproduced a FLAG-tagged variant of either SltB1(WT) or the catalytically inactive variant SltB1(E135A)  in WT cells and monitored beta-lactam resistance and AmpC production.Overexpression of sltB1(WT)-FLAG did not alter the Caz or piperacillin sensitivity of the WT strain, nor did it lead to the detectable induction of AmpC production (Fig. 7).However, overproduction of SltB1(E135A)-FLAG to levels equivalent to that of the WT protein conferred a Caz and piperacillin resistance phenotype to otherwise WT cells and led to AmpC production (Fig. 7).This resistance phenotype required ampC, ampR, and mltG similar to that of the sltB1 deletion (Fig. S4).Thus, overproduction of SltB1(E135A)-FLAG is likely to saturate a critical protein partner or PG binding site used by SltB1(WT) to perform its function, leading to an SltB1-defective phenotype.The binding partner may be a PG synthase like PBP2, and SltB1 may work with this and other PG synthases to promote the efficient incorporation of nascent PG strands into the matrix for PG expansion.Accordingly, when ampC is deleted to remove its influence on beta-lactam resistance, cells lacking SltB1 display increased sensitivity to a range of different beta-lactams but not the aminoglycoside tobramycin or the outer membrane-impermeable PG synthesis inhibitor vancomycin (Table 1).Notably, SltB1 inactivation also resulted in a mild increase in sensitivity to the MreB antagonist A22 that disrupts the function of the PBP2-containing cell elongation system (Table 1) (48).We therefore conclude that SltB1 is likely to participate in cell wall synthesis and expansion, potentially as a component of a PG synthesis complex, a role not typically associated with LT enzymes capable of cleaving PG strands.

DISCUSSION
Mutations that inactivate LT enzymes have been found to increase the expression of beta-lactamase genes in several different gram-negative organisms (35)(36)(37)(38).However, the mechanism behind this phenomenon and how it relates to the role(s) of LTs in cell wall biogenesis has remained unclear.It was originally reported that mutants defective for SltB1 in P. aeruginosa confer elevated beta-lactam resistance via the inactivation of a cell death pathway involving autolysis by this PG-cleaving enzyme (35,36).This mechanism was proposed because the authors did not detect elevated AmpC activity or protein levels in extracts from ∆sltB1 cells despite the resistance phenotype being dependent on ampC and the detection of elevated AmpC in ∆dacB cells inactivated for SltB1 (35,36).By contrast, our results indicate that ∆sltB1 cells are induced for ampC expression, which was shown in several ways: (i) via a P ampC ::lacZ reporter gene, (ii) AmpC activity assays, and (iii) immunoblotting for AmpC.Furthermore, induction was shown to depend on the importer AmpG that transports AMP products into the cytoplasm and the AmpR activator that stimulates ampC expression when it senses these molecules (17,18,46).Why the previous reports did not also observe ampC induction in their sltB1 mutant is not known (35,36).Nevertheless, the results presented here strongly support the conclusion that SltB1 inactivation confers beta-lactam resistance by activating ampC expression via the canonical mechanism involving the sensing of AMP turnover products.
Inactivation of DacB has also been associated with ampC induction through the canonical sensing of AMP turnover products (7)(8)(9).Pentapeptide-containing AMP molecules were previously shown to be produced in ∆dacB cells, providing evidence that the pentapeptide species is the most potent inducing molecule for the activation of ampC expression (49).DacB is known to have both PG carboxypeptidase and PG EP activity in vitro (34).However, cells lacking DacB were not found to have a significant increase in pentapeptides in their cell wall, indicating that it is not a major carboxypepti dase in vivo (34,49).Thus, the pentapeptide-containing AMPs detected in ∆dacB cells are not generated via turnover of the mature PG sacculus.How they are produced has remained unclear.
An important clue to the mechanism by which DacB or SltB1 inactivation causes ampC induction came from the observation that it requires the LT enzyme called MltG.In E. coli, this inner membrane-anchored LT has been implicated in the turnover of nascent PG glycans following cefsulodin treatment of E. coli (30).Additionally, LT activity has been shown to degrade uncrosslinked PG strands in the periplasm of other gram-negative bacteria (50).In P. aeruginosa, MltG was found to be a key target of bulgecin A, an LT inhibitor that sensitizes P. aeruginosa and other gram-negative bacteria to beta-lactams (41).The implication of this finding is that MltG is likely involved in the turnover of uncrosslinked strands produced upon beta-lactam treatment, preventing the toxic side effects of these glycans (29) and providing the pentapeptide-containing AMP products for ampC induction.Accordingly, we showed that pentapeptide-containing AMPs are produced in ∆dacB and ∆sltB1 cells and that their production is MltG-dependent.Notably, MltG was also recently shown to be required for ampC induction and beta-lac tam resistance in a clinical isolate of P. aeruginosa (40).Therefore, based on the MltG requirement for ampC activation in ∆sltB1 and ∆dacB cells, we infer that the inactivation of these enzymes causes problems with the incorporation of nascent PG strands into the cell wall matrix such that uncrosslinked nascent PG strands are produced and rapidly degraded by MltG to generate the AMP signals that induce beta-lactamase expression.Given that ∆sltB1 and ∆dacB mutants grow normally and do not show a decrease in overall PG crosslinks in their mature sacculi, the defect in nascent PG incorporation in these cells is likely to be relatively minor.It is only through the high sensitivity of the ampC regulatory system for pentapeptide-containing AMPs that the roles for these enzymes in nascent PG incorporation are revealed.
How might DacB and SltB1 be promoting the insertion of nascent PG strands into the mature wall?The simplest explanation is that they are functioning as space-making enzymes that cleave linkages in the cell wall to make room for the insertion of new material to expand the PG matrix during growth.PG EPs have previously been implicated as space makers (15,(24)(25)(26)(27). DacB is therefore likely to similarly use its PG EP activity to provide sites for nascent PG incorporation.Unlike EPs, LTs have not traditionally been thought to function in a space-making capacity, but a recent report suggests the LT enzyme MltD in E. coli performs such a role (51).Based on the ampC induction phenotype upon SltB1 inactivation in P. aeruginosa, we propose that this LT enzyme also functions as a space maker.
A major outstanding question in the field is how the activity of space-making enzymes is coordinated with PG synthesis to prevent imbalanced PG cleavage and cell lysis.Notably, DacB has been shown to interact with the aPBP-type PG synthase PBP1a in E. coli (52), raising the possibility that it may work in complex with PBP1a to promote the insertion of new material made by this enzyme.SltB1, on the other hand, has been found to interact with PBP2 in P. aeruginosa (43,47).PBP2 is a bPBP with TPase activity that together with the SEDS GTase RodA forms the essential cell wall synthase of the Rod system (elongasome) that is responsible for cell elongation and shape determination (15,(21)(22)(23).Although the physiological relevance of the SltB1-PBP2 interaction has yet to be demonstrated, the location of the sltB1 gene just downstream of the genes encoding RodA and PBP2 (Fig. 2A) suggests that SltB1 may be a non-essential component of the Rod system that cleaves glycan strands in the mature PG matrix that may interfere with the insertion of new PG material during cell wall expansion.Consistent with this model, catalytically inactive SltB1 exerts a dominant-negative, ∆sltB1-like phenotype indicating that SltB1 is likely to have a limited number of protein partner binding sites in the cell, which are required for its function.Cells lacking SltB1 show increased sensitivity to the Rod system antagonist A22 (Table 1) (48), further supporting an auxiliary role for SltB1 in the function of the cell elongation system.
In conclusion, we have used an ampC induction phenotype in P. aeruginosa to reveal a potential function for the DacB EP and the SltB1 LT enzyme as space makers promoting the insertion of new PG strands into the mature wall.Such a function is only commonly ascribed to EPs that cut PG crosslinks (15,(24)(25)(26)(27).However, the PG matrix is unlikely to be so neatly arranged that crosslinks are the only impediment to the insertion of nascent glycans.Active synthases are also likely to encounter mature glycans that cross their path and require removal for the new strand to be effectively incorporated.Thus, based on our findings for SltB1 and recent work on MltD in E. coli (51), it is likely that a subset of the many LTs encoded by gram-negative bacteria function in a space-making capacity similar to that of the EPs.

P. aeruginosa viability assay
For viability assays, overnight cell cultures were normalized to an OD 600 of 1 and subjected to serial 10-fold dilutions with LB.Five microliters of each dilution was then spotted onto the indicated agar, and plates were incubated at 30°C for 24 h prior to imaging.

P. aeruginosa electroporation
P. aeruginosa strains were made competent using previously described methods (53).Briefly, 4 mL of overnight cultures grown at 37°C were centrifuged and washed twice with 1-mL 300-mM sucrose.Cell pellets were resuspended in 500 µL of 300-mM sucrose, and 100 µL was used for electroporation.One microliter of replicative plasmid was used for the electroporation, using the following settings: 25 mF, 200 O, 2.5 kV.LB medium (1 mL) was added, and the cells were incubated with shaking (200 rpm) for 1 h at 37°C.Cells were then plated on the appropriate selective medium.

Screen for mutants that induce ampC expression
The screening procedure was described previously (42).Briefly, P. aeruginosa strain CF263 [PAO1 (P ampC ::lacZ)] was transposon-mutagenized by mating with the E. coli donor SM10(λpir) harboring a mariner transposon delivery vector pIT2 (54).The transposon confers Tet resistance.Mating mixtures were plated on LB agar supplemented with Tet (50 µg/mL) to select for transposon mutants and nalidixic acid (25 µg/mL) to select against the E. coli donor.The resulting collection of colonies was resuspended in LB broth and stored at −80°C.Dilutions of the library were plated on LB containing X-gal (50 µg/mL) to identify mutants with a constitutively active P ampC ::lacZ reporter.

AmpC beta-lactamase activity assay
AmpC activity was assessed using nitrocefin hydrolysis.Overnight bacterial cultures were subcultured 1:20 in 3-mL LB or 6-mL LB supplemented with 1-mM IPTG and grown for 2 h at 30°C and 200 rpm.The cultures were split 1:1 in 2-mL LB with or without 50-µg/mL cefoxitin (final concentration), and all cultures were incubated for an additional 1.5 h at 30°C and 200 rpm.Following incubation, 1 mL of culture was pelleted at 2,300 × g for 5 min, washed once with 1 mL of 50-mM sodium phosphate buffer (pH 7.0), and resuspended in 1 mL of the same cold buffer.Samples were placed on ice and lysed at 4°C by sonication with a microprobe (Q800R2, QSonica, Newtown, Connecticut, USA).Sonicated samples were centrifuged at 12,000 × g for 5 min at 4°C, and supernatants were collected.The protein concentration was determined using a Bradford assay (57) with bovine serum albumin as the standard (G-Biosciences, Geno technology Inc., St. Louis, Missouri, USA).Nitrocefin hydrolysis assays were performed in 96-well plates.Each reaction had a final volume of 250 µL of 50-mM sodium phosphate buffer (pH 7.0) containing 10 µg of protein and 20 µg of nitrocefin (Thermo Fisher Scientific Oxoid, Waltham, Massachusetts, USA).Nitrocefin hydrolysis was monitored by measuring the absorbance at 486 nm every 5 min for 35 or 120 min at 30°C.

Antibiotic sensitivity assays
Antibiotic sensitivity assays were performed using E test strips and broth microdilutions.For E test assays, 100 µL of bacterial cultures in exponential phase were evenly spread on LB agar plates and allowed to dry.E test strips were overlaid on the agar, and plates were incubated for 24 h at 37°C.The concentration at which the zone of inhibition intersected the E test strip was used to determine the minimum inhibitory concentrations (MICs).E test assays were performed three times independently, and the MIC values were listed in Table 1.For the antibiotic MIC assays, overnight cell cultures were normalized to OD 600 of 0.0005 in LB and the indicated concentrations of A22, Carb, cefoxitin, mecillinam, or piperacillin and grown for 24 h at 30°C prior to taking optical density readings (Biotek Epoch 2, Agilent, Santa Clara, California, USA).Broth microdilution MIC assays were performed three or four times independently, each with two technical and two biological replicates.

Intracellular soluble muropeptide analysis
To determine the presence and levels of intracellular soluble muropeptides, bacteria were grown until late exponential phase (roughly OD 600 0.7) in LB media before being cooled on ice for 10 min and normalized to the same OD 600 .Cells were then harvested by centrifugation at 10,000 × g for 10 min.The supernatant was discarded, and the cell pellet was washed three times in ice-cold 0.9% NaCl, resuspended in 0.9% NaCl so that the cells are 20 times concentrated and boiled for 10 min before centrifugation at maximum speed in a benchtop centrifuge for 10 min to remove the proteins and insoluble fraction.The supernatant was used for further analysis by liquid chromatogra phy-mass spectrometry (LC-MS).

PG isolation
Cells from 0.2-L cultures of overnight stationary phase were pelleted at 5,250 × g and resuspended in 5 mL of phosphate-buffered saline, added to an equal volume of 10% SDS in a boiling water bath and vigorously stirred for 3 h, and then stirred overnight at room temperature.The insoluble fraction (PG) was pelleted at 400,000 × g, 15 min, 30°C (TLA-100.3rotor; Optima Max ultracentrifuge, Beckman) and resuspended in Milli-Q water.This step was repeated four to five times until the SDS was washed out.Next, PG was treated with Pronase E (0.1 mg/mL) at 60°C for 1 h and then boiled in 1% SDS for 2 h to stop the reaction.After SDS was removed as described previously, PG samples were resuspended in 200 µL of 50-mM sodium phosphate buffer pH 4.9 and digested overnight with 30-µg/mL muramidase (from Streptomyces albus) at 37°C.Muramidase digestion was stopped by heat inactivation (boiling for 5 min).Coagulated protein was removed by centrifugation (20,000 × g, 15 min).The supernatants (soluble muropepti des) were subjected to sample reduction.First, pH was adjusted to 8.5-9 by addition of borate buffer (0.5 M pH 9), and then muramic acid residues were reduced by sodium borohydride treatment (NaBH 4 10-mg/mL final concentration) during 30 min at room temperature.Finally, pH was adjusted to 2.0-4.0 with 25% orthophosphoric acid prior to analysis by LC.

LC-MS analysis
Chromatographic analyses of muropeptides were performed by Ultra Performance Liquid Chromatography (UPLC) on an UPLC system (Waters) equipped with a trapping cartridge precolumn (SecurityGuard ULTRA Cartridge UHPLC C18 2.1 mm, Phenomenex) and an analytical column BEH C18 column (130 Å, 1.7 µm, 2.1 mm, Waters) maintained at 45°C.Muropeptides were detected by measuring the absorbance at 204 nm using an ACQUITY UPLC UV-visible detector.Muropeptides were separated using a linear gradient from Buffer A (water + 0.1% (vol/vol) formic acid) to Buffer B (acetonitrile 100% (vol/ vol) + 0.1% (vol/vol) formic acid) over 15 min with a flow rate of 0.25 mL/min.The QTOF instrument was operated in positive ion mode, with data collection performed in untargeted MS e mode.The parameters were set as follows: capillary voltage 3.0 kV, source temperature 120°C, desolvation temperature 350°C, sample cone voltage 40 V, cone gas flow 100 L h −1 , and desolvation gas flow 500 L h −1 .Mass spectra were acquired at a speed of 0.25 s/scan.The scan was in a range of 100-2,000 m/z.Data acquisition and processing were performed using MassLynx or UNIFI software package (Waters Corp.).The quantification of muropeptides was based on their relative abundances (relative area of the corresponding peak) and relative molar abundances.A table of all the identified muropeptides and the observed ions is provided (Table S4).

Immunoblotting
Overnight bacterial cultures were subcultured 1:20 in 5-mL LB and grown for 4 h.Bacteria were collected by centrifugation, washed once with Tris-HCl buffer (pH = 8, 10 mM) and resuspended in 500 µL of the same cold buffer.The samples were then lysed at 4°C for 12 min at 60% amplitude with a pulse rate of 10 s ON/10 s OFF using a Qsonica sonicator.The samples were then centrifuged at 4°C at maximal speed to remove the cell debris.Supernatants were collected and a Bradford assay performed to measure the protein concentration.A total of 100-µg protein in a volume of 100 µL was mixed with 100 µL of 2× Laemmli buffer.Immunoblotting was performed by first separating 15 µL of each sample on 12% SDS-PAGE (polyacrylamide gel electrophoresis) gels at 90 V for 15 min and 120 V for an hour.Proteins were transferred at 90 V for an hour at 4°C to a 0.2-µm polyvinylidene difluoride membranes (Whatman) previously soaked in methanol and rinsed with transfer buffer.Membranes were blocked using 5% (wt/vol) skim milk in Tris-buffered saline (10-mM Tris-HCl pH 7.5, 150-mM NaCl) supplemented with 0.1% (vol/vol) Tween-20 (TBS-T) for 1 h.Membranes were incubated for 1 h with α-AmpC primary antibody (1:1,000 dilution in 5% skim milk in TBS-T, MyBioSource, MBS1493275, San Diego, USA) or α-FLAG primary antibody (1:1,000 dilution in 5% skim milk in TBS-T, F7425, Sigma-Aldrich) at 4°C.The membranes were washed four times in TBS-T for 5 min each before incubation for 1 h with secondary antibody (anti-rabbit IgG HRP, 1:5,000 dilution, Rockland 18-8816-33) in TBS-T with 5% (wt/vol) skim milk powder.The membranes were then washed four times with TBS-T for 5 min each before developing using SuperSignal West Pico PLUS Chemiluminescent Substrate (Thermo Fisher Scientific cat#34577) and imaged using the c600 Azure Biosystems platform.

FIG 1
FIG 1 Overview of PG synthesis and ampC regulation.(Left) The PG matrix consists of glycan chains with the repeating unit of MurNAc and GlcNAc.Attached

FIG 2
FIG 2 Inactivation of SltB1 promotes P ampC ::lacZ induction and beta-lactam resistance.(A) Diagram of the genetic locus harboring sltB1.The pbpA and rodA

FIG 3
FIG 3 AmpC activity is elevated upon inactivation of SltB1.Assays of nitrocefin hydrolysis by cell lysates of PAO1 (WT) or CF1105 (∆sltB1) with plasmids pKHT103 (vector control) or pCF533 (P lac ::sltB1) as indicated.Data are the mean of three independent assays each for two biological replicates with the error bars indicating the standard error.(A) Cells were grown in LB with IPTG (1 mM).A one-way analysis of variance revealed that there was a statistically significant difference in nitrocefin hydrolysis level between at least two strains [F(3, 25) = 72.23,P < 0.00001].Tukey's honestly significant difference

TABLE 1
MIC of selected antibiotics a MIC determined by broth microdilution.b MIC determined by E test strip.