Involvement of the relA gene in the autolysis of Escherichia coli induced by inhibitors of peptidoglycan biosynthesis

It is generally assumed that inhibitors of peptidoglycan biosynthesis do not kill nongrowing bacteria. An exceptional case is reported here. The addition of chloramphenicol to amino acid-deprived cultures of relA+ strains of Escherichia coli which were treated with beta-lactam antibiotics, D-cycloserine, or moenomycin resulted in lysis. This phenomenon is termed chloramphenicol-dependent lysis. To be effective, chloramphenicol had to be present at its minimum growth-inhibitory concentration (or higher). Analogs of chloramphenicol which did not bind to ribosomes were completely ineffective. Amino acid deprivation was actually not required to demonstrate chloramphenicol-dependent lysis, and cultures treated with growth-inhibitory levels of chloramphenicol alone were lysed when challenged with inhibitors of peptidoglycan synthesis. Peptidoglycan synthesis has been shown previously to be under stringent (relA+) control, and chloramphenicol is known to be an antagonist of stringent control. Thus, it is proposed that the mechanism of chloramphenicol-dependent lysis is based on the ability of chloramphenicol to relax peptidoglycan synthesis in nongrowing relA+ bacteria. This is also consistent with the observation that treatment of amino acid-deprived relA mutants with inhibitors of peptidoglycan synthesis resulted in lysis, i.e., without the mediation of chloramphenicol.     

Alteration of Escherichia coli murein during amino acid starvation.

We have studied the mechanisms by which amino acid starvation of Escherichia coli induces resistance against the lytic and bactericidal effects of penicillin. Starvation of E. coli strain W7 of the amino acids lysine or methionine resulted in the rapid development of resistance to autolytic cell wall degradation, which may be effectively triggered in growing bacteria by a number of chemical or physical treatments. The mechanism of this effect in the amino acid-starved cells involved the production of a murein relatively resistant to the hydrolytic action of crude murein hydrolase extracts prepared from normally growing E. coli. Resistance to the autolysins was not due to the covalently linked lipoprotein. Resistance to murein hydrolase developed most rapidly and most extensively in the portion of cell wall synthesized after the onset of amino acid starvation. Lysozymes digests of the autolysin-resistant murein synthesized during the first 10 min of lysine starvation yielded (in addition to the characteristic degradation products) a high-molecular-weight material that was absent from the lysozyme-digests of control cell wall preparations. It is proposed that inhibition of protein synthesis causes a rapid modification of murein structure at the cell wall growth zone in such a manner that attachment of murein hydrolase molecules is inhibited. The mechanism may involve some aspects of the relaxed control system since protection against penicillin-induced lysis developed much slower in amino acid-starved relaxed controlled (relA) cells than in isogenic stringently controlled (relA+) bacteria.     

Inhibition of uridine 5'-diphosphate-N-acetylmuramyl-L-alanine synthetase by beta-chloro-L-alanine in Escherichia coli

The synthesis of the nucleotide precursors for peptidoglycan is regulated by the relA gene in Escherichia coli. Thus, nucleotide precursors labeled with [3H]diaminopimelic acid accumulated in a relA strain but not in an isogenic relA+ strain during amino acid deprivation. Furthermore, nucleotide precursor synthesis was relaxed in the amino acid deprived relA+ strain by treatment with chloramphenicol. Uridine diphosphate-N-acetylmuramyl-pentapeptide (UDP-MurNAc-pentapeptide) was the major component accumulated during the relaxed synthesis of nucleotide precursors in both relA+ and relA strains. The effect of beta-chloro-L-alanine (CLA) on the relaxed synthesis of nucleotide precursors for peptidoglycan was determined. At a low concentration (0.0625 mM) CLA inhibited the synthesis of UDP-MurNAc-pentapeptide and caused the accumulation of UDP-MurNAc-tripeptide. Thus, low concentrations of CLA probably inhibited alanine racemase, as reported previously. Higher concentrations of CLA also inhibited an earlier step in nucleotide precursor synthesis. This was shown to be due to the inhibition of UDP-MurNAc-L-alanine synthetase by CLA. CLA inhibited the activity of this enzyme in cell-free extracts as well as in intact cells.     

Site of Inhibition of Peptidoglycan Biosynthesis During the Stringent Response in Escherichia coli

The site of inhibition of peptidoglycan synthesis during the stringent response in Escherichia coli was determined in strains which were auxotrophic for both lysine and diaminopimelic acid (DAP). Cells were labeled with [(3)H]DAP for 30 to 60 min in the presence and absence of required amino acids, and the cellular distribution of [(3)H]DAP was determined. In both stringent (rel(+)) and relaxed (relA) strains, amino acid deprivation did not inhibit the incorporation of [(3)H]DAP into the nucleotide precursor and lipid intermediate fractions. The amount of [(3)H]DAP incorporated into the peptidoglycan fraction by the amino acid-deprived relA strain was over 70% of the amount incorporated in the presence of required amino acids. In contrast, the amounts of labeled peptidoglycan in amino acid-deprived rel(+) strains were only 20 to 44% of the amounts synthesized in the presence of amino acids. These results indicate that a late step in peptidoglycan synthesis is inhibited during the stringent response. The components of the lipid intermediate fraction synthesized by rel(+) strains in the presence and absence of required amino acids were quantitated. Amino acid deprivation did not inhibit the synthesis of either the monosaccharide-pentapeptide or the disaccharide-pentapeptide derivatives of the lipid intermediate. Thus, the reaction which is most likely inhibited during the stringent response is the terminal one involving the incorporation of the disaccharide-pentapeptide into peptidoglycan.     

Control of lipopolysaccharide biosynthesis and release by Escherichia coli and Salmonella typhimurium.

The influence of the relA gene on lipopolysaccharide (LPS) biosynthesis and release by Escherichia coli and Salmonella typhimurium was investigated. Similar results were obtained with both species. The incorporation of [3H]galactose into LPS by galE mutants was inhibited by at least 50% (as compared with normal growing controls) during amino acid deprivation of relA+ strains. This inhibition could be prevented by the treatment of the amino acid-deprived relA+ bacteria with chloramphenicol, a known antagonist of the stringent control mechanism. Furthermore, LPS biosynthesis was not inhibited during amino acid deprivation of isogenic relA mutant strains. These results indicate that LPS synthesis is regulated by the stringent control mechanism. Normal growing cells of both relA+ and relA strains released LPS into the culture fluid at low rates. Amino acid deprivation stimulated the rate of LPS release by relA mutants but not by relA+ bacteria. Chloramphenicol treatment markedly stimulated the release of cell-bound LPS by amino acid-deprived relA+ cells. Thus, a low rate of LPS release was characteristic of normal growth and could be increased in nongrowing cells by relaxing the control of LPS synthesis.     

Regulation of D-alanine carboxypeptidase and peptidoglycan cross-linkage in amino acid-deprived Escherichia coli.

The effect of amino acid deprivation on the activities of D-alanine carboxypeptidase (CPase) and peptidoglycan transpeptidase in Escherichia coli was determined. Enzymes were assayed in ether-treated bacteria (ETB) which were permeable to peptidoglycan nucleotide precursors. ETB were prepared at intervals from cultures grown in the presence and absence of a required amino acid. The specific activity of CPase in ETB decreased 50 to 85% during amino acid deprivation. This was paralleled by a 60 to 70% decrease in the specific activity of peptidoglycan transpeptidase. Both enzymes reached their lowest level of activity about 40 min after the onset of amino acid deprivation. The decrease in CPase activity apparently was not due to degradation of the enzyme, since full activity was restored after disruption of ETB by sonication. A decrease in CPase activity was associated with an enhancement of transpeptidation. The peptidoglycan synthesized in vitro by amino acid-deprived ETB was 1.7 times more cross-linked than the peptidoglycan synthesized by control ETB These results support the proposal that CPase may be involved in regulating transpeptidation in E. coli.     

Autolysis-resistant peptidoglycan of anomalous composition in amino-acid-starved Escherichia coli.

Nongrowing Escherichia coli deprived of an essential amino acid continued to produce peptidoglycan at a rate approximately 30% of that of growing cells. The composition of this peptidoglycan was very different from that of growing cells and resembled that of peptidoglycan left undegraded during partial autolysis of the bacteria. Synthesis of this peptidoglycan of anomalous composition began at once upon the removal of the amino acid from the medium. Fifteen minutes of amino acid deprivation was sufficient to virtually completely prevent penicillin-induced autolytic wall degradation in vivo. During this time, although the specific activities of soluble and membrane-bound hydrolytic transglycosylases and endopeptidases remained high, the peptidoglycan produced showed decreased sensitivity to degradation in vitro. After more extensive (2-h) starvation, triggering of autolysis by chaotropic agents was also blocked. Autolysis in growing cells may be selective for peptidoglycan representing the cylindrical portion of the sacculus. It is suggested that at least part of the mechanism of the well-known lysis resistance of nongrowing E. coli is related to the deposition of structurally anomalous and relatively autolysin-resistant peptidoglycan at some strategically located sites on the bacterial surface.     

In situ assay for identifying inhibitors of bacterial transglycosylase

An in situ transglycosylase assay has been developed using endogenously synthesized lipid II. The assay involves the preferential synthesis and accumulation of lipid II in a reaction mixture containing the cell wall membrane material isolated from Escherichia coli, exogenously supplied UDP-MurNAc-pentapeptide, and radiolabeled UDP-GlcNAc. In the presence of Triton X-100, the radiolabeled product formed is almost exclusively lipid II, while the subsequent formation of peptidoglycan is inhibited. Removal of the detergent resulted in the synthesis of peptidoglycan (25% incorporation of radiolabeled material) from the accumulated lipid II. This reaction was inhibited by moenomycin, a known transglycosylase inhibitor. In addition, tunicamycin, which affects an earlier step of the pathway by inhibiting MraY, had no effect on the formation of peptidoglycan in this assay, as expected. Similarly, ampicillin and bacitracin did not inhibit the formation of peptidoglycan under the conditions established.     

In vivo and in vitro action of new antibiotics interfering with the utilization of N-acetyl-glucosamine-N-acetyl-muramyl-pentapeptide

Recent literature on the antibiotics enduracidin, moenomycin, prasinomycin, and 11.837 RP suggested an interaction with murein synthesis. Incubation of sensitive strains from Bacillus cereus and Staphylococcus aureus in a "wall medium" containing labeled l-alanine showed that all four antibiotics inhibited the incorporation of alanine into murein and gave rise to accumulation of radioactive uridine diphosphate-N-acetyl-muramyl (UDP-MurNAc)-pentapeptide. Peptidoglycan was synthesized when the particulate enzyme of B. stearothermophilus was incubated with the murein precursors UDP-N-acetyl-glucosamine (UDP-GlcNAc) and UDP-MurNAc-pentapeptide. The newly formed polymer was less accessible for lysozyme and more strongly bound to the acceptor than the same product from the Escherichia coli particulate enzyme. After incubation in the presence of penicillin, a greater part of the peptidoglycan was lysozyme sensitive and more loosely bound to the acceptor. The antibiotics enduracidin, moenomycin, prasinomycin, and 11.837 RP inhibited peptidoglycan synthesis by the B. stearothermophilus particulate enzyme. The rate of synthesis of GlcNAc-MurNAc(-pentapeptide)-P-P-phospholipid was independent from the addition of these antibiotics, but its utilization was strongly inhibited. With the present results, it is not possible to distinguish the mechanisms of action of enduracidin, moenomycin, prasinomycin, and 11.837 RP from the mechanisms of action of vancomycin and ristocetin.     

Penicillin-insensitive incorporation of D-amino acids into cell wall peptidoglycan influences the amount of bound lipoprotein in Escherichia coli.

Certain D-amino acids, such as D-methionine and D-cystine, were incorporated into cells of Escherichia coli under conditions inhibiting protein and cell wall synthesis. Part of the radioactivity of D-14C-amino acids incorporated into the cells was found in the isolated cell wall peptidoglycan. A covalent linkage between the amino group of the D-amino acids and the peptidoglycan was presumed to be the main cause of the binding of the D-amino acids to peptidoglycan, because the amino group of the D-amino acids in the incorporation product was substituted. Whether the carboxyl terminus was substituted was unknown. The formation of the D-amino acid-peptidoglycan linkage was insensitive to beta-lactam antibiotics such as benzylpenicillin and ampicillin (500 micrograms/ml) and therefore was not due to the reaction of DD-transpeptidation which is involved in the biosynthesis of peptidoglycan. The D-amino acids also strongly inhibited the formation of peptidoglycan-bound lipoprotein in the E. coli cells. The results may suggest the correlation between binding of D-amino acid to peptidoglycan and inhibition of formation of the bound form of lipoprotein.     

Anchor structure of staphylococcal surface proteins. IV. Inhibitors of the cell wall sorting reaction.

Surface proteins of Staphylococcus aureus are covalently linked to the bacterial cell wall by a mechanism requiring a COOH-terminal sorting signal with a conserved LPXTG motif. Cleavage between the threonine and the glycine of the LPXTG motif liberates the carboxyl of threonine to form an amide bond with the amino of the pentaglycine cross-bridge in the staphylococcal peptidoglycan. We asked whether antibiotic cell wall synthesis inhibitors interfere with the anchoring of surface proteins. Penicillin G, a transpeptidation inhibitor, had no effect on surface protein anchoring, whereas vancomycin and moenomycin, inhibitors of cell wall polymerization into peptidoglycan strands, slowed the sorting reaction. Cleavage of surface protein precursors did not require a mature assembled cell wall and was observed in staphylococcal protoplasts. A search for chemical inhibitors of the sorting reaction identified methanethiosulfonates and p-hydroxymercuribenzoic acid. Thus, sortase, the enzyme proposed to cleave surface proteins at the LPXTG motif, appears to be a sulfhydryl-containing enzyme that utilizes peptidoglycan precursors but not an assembled cell wall as a substrate for the anchoring of surface protein.     

Peptidoglycan synthesis in the absence of class A penicillin-binding proteins in Bacillus subtilis

Penicillin-binding proteins (PBPs) catalyze the final, essential reactions of peptidoglycan synthesis. Three classes of PBPs catalyze either trans-, endo-, or carboxypeptidase activities on the peptidoglycan peptide side chains. Only the class A high-molecular-weight PBPs have clearly demonstrated glycosyltransferase activities that polymerize the glycan strands, and in some species these proteins have been shown to be essential. The Bacillus subtilis genome sequence contains four genes encoding class A PBPs and no other genes with similarity to their glycosyltransferase domain. A strain lacking all four class A PBPs has been constructed and produces a peptidoglycan wall with only small structural differences from that of the wild type. The growth rate of the quadruple mutant is much lower than those of strains lacking only three of the class A PBPs, and increases in cell length and frequencies of wall abnormalities were noticeable. The viability and wall production of the quadruple-mutant strain indicate that a novel enzyme can perform the glycosyltransferase activity required for peptidoglycan synthesis. This activity was demonstrated in vitro and shown to be sensitive to the glycosyltransferase inhibitor moenomycin. In contrast, the quadruple-mutant strain was resistant to moenomycin in vivo. Exposure of the wild-type strain to moenomycin resulted in production of a phenotype similar to that of the quadruple mutant.     

Correlation between RNA synthesis and basal level guanosine 5'-diphosphate 3'-diphosphate in relaxed mutants of Escherichia coli

Through the use of a new nucleotide extraction procedure, we had previously shown that relaxed mutants of Escherichia coli exhibit a unique response to amino acid starvation (Lagosky, P. A., and Chang, F. N. (1980) J. Bacteriol. 144, 499-508). The basal level amounts of guanosine 5'-diphosphate 3'-diphosphate (ppGpp) in both relA and phenotypically relaxed relA+ rplK (relC) strains were shown to decrease at the onset of amino acid limitation and to remain severely depressed throughout the course of the starvation. Upon resupplementation of amino acid-starved relaxed mutants, the production of ppGpp resumes and results in the temporary overaccumulation of this nucleotide beyond its original basal level amount. We now show that the basal level ppGpp content of relaxed bacteria, as well as its subsequent fluctuations in response to amino acid starvation, is inversely correlated with the initial rates of RNA synthesis in these strains. The ability of ppGpp to control the rate of protein synthesis in relA mutants was also examined. It was observed that ppGpp had no apparent direct effect on the initial rates of protein synthesis in relA mutants. The constant inverse correlation which exists between ppGpp content in relA mutants, and their rates of RNa synthesis provide evidence which indicates that basal level ppGpp synthesis has definite physiological significance. It also suggests that the synthesis of basal level ppGpp might be an absolute requirement needed for normal bacterial growth.     

Attachment of lipoprotein to murein (peptidoglycan) of Escherichia coli in the presence and absence of penicillin FL 1060.

In vivo studies on the attachment of lipoprotein to the murein (peptidoglycan) of Escherichia coli showed that it takes several generations of growth until the amount of lipoprotein on newly made murein is equilibrated. The technique used involves degradation of the sodium dodecyl sulfate-insoluble murein-lipoprotein complex (sacculus, rigid layer) with lysozyme and separation of the labeled products on paper. No lipoprotein was found on murein subunits incorporated during a pulse of [3H]diaminopimelate for 1 min in logarithmically growing cells at 37 C. Even after one doubling of the cell mass, only 4 to 8% of the labeled murein was isolated as bound to lipoprotein. With uniformly labeled murein, 30% remains bound to lipoprotein after lysozyme treatment, corresponding to three murein subunits. Therefore it can be concluded that during pulse labeling either no lipoprotein is incorporated into the newly synthesized murein or no murein subunits are inserted into existing murein around lipoprotein attachment sites. Longer pulse and pulse-chase experiments argue for the latter interpretation. It is therefore concluded that incorporation of murein subunits into the growing murein polymer is not at all a random process. Instead, quite large areas of murein, on which lipoprotein is situated, seem to be preserved. Under the influence of penicillin FL 1060 murein synthesis is 50% inhibited. The rate of lipoprotein attachment is less affected so that increasing amounts of lipoprotein become attached during spheroplast formation. By the time the stationary growth phase has been reached, the lipoprotein content of the murein has doubled. Diaminopimelate auxotrophic mutants require, in the presence of penicillin FL 1060, more diaminopimelate for full growth than in the absence of penicillin FL 1060. This finding and the fact that murein synthesis is always inhibited by 50% over a wide range of penicillin concentration (1 to 1,000 mug/ml) point to the inhibition of an enzymatic step of murein synthesis which can be partially bypassed by a second enzyme, less efficient but resistant to penicillin FL 1060.     

Stringent control of peptidoglycan biosynthesis in Escherichia coli K-12.

[3H]Diaminopimelic acid (Dap) was incorporated exclusively into peptidoglycan by Escherichia coli strains auxotrophic for both lysine and Dap. The rate of [3H]Dap incorporation by stringent (rel+) strains was significantly decreased when cells were deprived of required amino acids. The addition of chloramphenicol to amino acid-starved rel+ cultured stimulated both peptidoglycan and ribonucleic acid synthesis. In contrast, a relaxed (relA) derivative incorporated [3H]Dap at comparable rates in the presence or absence of required amino acids. Physiologically significant concentrations of guanosine 5'-diphosphate 3'-diphosphate (ppGpp) inhibited the in vitro synthesis of both carrier lipid-linked intermediate and peptidoglycan catalyzed by a particulate enzyme system. The degree of inhibition was dependent on the concentration of ppGpp in the reaction mixture. Thus, the results of in vivo and in vitro studies indicate that peptidoglycan synthesis is stringently controlled in E. coli.     

Beta-lactam-induced bacteriolysis of amino acid-deprived Escherichia coli is dependent on phospholipid synthesis.

The penicillin tolerance of amino acid-deprived relA+ Escherichia coli is attributed to the stringent response; i.e., relaxation of the stringent response suppresses penicillin tolerance. The beta-lactam-induced lysis of amino acid-deprived bacteria resulting from relaxation of the stringent response was inhibited by cerulenin, or by glycerol deprivation in the case of a gpsA mutant (defective in the biosynthetic sn-glycerol 3-phosphate dehydrogenase). Therefore, beta-lactam-induced lysis of amino acid-deprived cells was dependent on phospholipid synthesis. The lysis process during amino acid deprivation can be experimentally dissociated into two stages designated the priming stage (during which the interaction between the beta-lactam and the penicillin-binding proteins occurs) and the beta-lactam-independent lysis induction stage. Both stages were shown to require phospholipid synthesis. It has been known for some time that the inhibition of phospholipid synthesis is among the plethora of physiological changes resulting from the stringent response. These results indicate that the inhibition of peptidoglycan synthesis and the penicillin tolerance associated with the stringent response are both secondary consequences of the inhibition of phospholipid synthesis.     

Relation between cell wall turnover and cell growth in Bacillus subtilis.

The kinetics of cell wall turnover in Bacillus subtilis have been examined in detail. After pulse labeling of the peptidoglycan with N-acetylglucosamine, the newly formed peptidoglycan is stable for approximately three-quarters of a generation and is then degraded by a process that follows first-order kinetics. Deprivation of an auxotroph of amino acids required for protein synthesis results in a cessation of turnover. If a period of amino acid starvation occurs during the lag phase of turnover, then the initiation of turnover is delayed for a period of time equivalent to the starvation period. During amino acid starvation, new cell wall peptidoglycan is synthesized and added to preexisting cell wall. This peptidoglycan after resumption of growth is also subject to degradation (turnover). It is suggested that cell wall turnover is dependent on cell growth and elongation. Several possible control mechanisms for cell wall autolytic enzymes are discussed in light of these observations.     

Differential antibacterial activity of moenomycin analogues on gram-positive bacteria

The moenomycin trisaccharide degradation product and synthetic disaccharide analogues based on the disaccharide core were bactericidal to gram-positive bacteria, inhibited lipid II polymerization, and inhibited cell wall synthesis in Enterococcus faecalis. Truncating moenomycin to the trisaccharide, and building upon the core disaccharide have both led to molecules possessing properties not shared with their respective parent structures.     

Anchoring of surface proteins to the cell wall of Staphylococcus aureus. III. Lipid II is an in vivo peptidoglycan substrate for sortase-catalyzed surface protein anchoring

Surface proteins of Staphylococcus aureus are anchored to the cell wall peptidoglycan by a mechanism requiring a C-terminal sorting signal with an LPXTG motif. Surface proteins are first synthesized in the bacterial cytoplasm and then transported across the cytoplasmic membrane. Cleavage of the N-terminal signal peptide of the cytoplasmic surface protein P1 precursor generates the extracellular P2 species, which is the substrate for the cell wall anchoring reaction. Sortase, a membrane-anchored transpeptidase, cleaves P2 between the threonine (T) and the glycine (G) of the LPXTG motif and catalyzes the formation of an amide bond between the carboxyl group of threonine and the amino group of cell wall cross-bridges. We have used metabolic labeling of staphylococcal cultures with [(32)P]phosphoric acid to reveal a P3 intermediate. The (32)P-label of immunoprecipitated surface protein is removed by treatment with lysostaphin, a glycyl-glycine endopeptidase that separates the cell wall anchor structure. Furthermore, the appearance of P3 is prevented in the absence of sortase or by the inhibition of cell wall synthesis. (32)P-Labeled cell wall anchor species bind to nisin, an antibiotic that is known to form a complex with lipid II. Thus, it appears that the P3 intermediate represents surface protein linked to the lipid II peptidoglycan precursor. The data support a model whereby lipid II-linked polypeptides are incorporated into the growing peptidoglycan via the transpeptidation and transglycosylation reactions of cell wall synthesis, generating mature cell wall-linked surface protein.