Role of murF in Cell Wall Biosynthesis: Isolation and Characterization of a murF Conditional Mutant of Staphylococcus aureus
http://www.100md.com
《细菌学杂志》
Molecular Genetics Laboratory, Instituto de Tecnologia Química e Biologica da Universidade Nova de Lisboa, 2780 Oeiras, Portugal,The Rockefeller University, 1230 York Ave., New York, New York 10021
ABSTRACT
The Staphylococcus aureus murF gene was placed under the control of a promoter inducible by IPTG (isopropyl--D-thiogalactopyranoside). It was demonstrated that murF is an essential gene; it is cotranscribed with ddlA and growth rate, level of beta-lactam antibiotic resistance, and rates of transcription of the mecA and pbpB genes paralleled the rates of transcription of murF. At suboptimal concentrations of the inducer, a UDP-linked muramyl tripeptide accumulated in the cytoplasm in parallel with the decline in the amounts of the normal pentapeptide cell wall precursor. The abnormal tripeptide component incorporated into the cell wall as a monomeric muropeptide, accompanied by a decrease in the oligomerization degree of the peptidoglycan. However, incorporation of the tripeptide into the cell wall was limited to a relatively low threshold value. Further reduction of the amounts of pentapeptide cell wall precursor caused a gradual decrease in the cellular amounts of peptidoglycan, the production of a thinner peripheral cell wall, aberrant septae, and an overall increase in the diameter of the cells. The observations suggest that the role of murF exceeds its primary function in peptidoglycan biosynthesis and may also be involved in the control of cell division.
INTRODUCTION
The cell wall of Staphylococcus aureus contains a highly cross-linked peptidoglycan in which most of the muropeptide units have pentaglycine branches attached to the aminogroup of the lysine residue, and virtually all monomeric and acceptor muropeptides carry a carboxy-terminal D-alany-D-alanine residue (3). The biosynthesis and attachment of this dipeptide is catalyzed by the protein product of the genes ddlA and murF, with the first, DdlA, producing the dipeptide and the second, MurF, attaching the dipeptide to the UDP-N-acetylmuramic acid (MurNAc)-tripeptide, thus completing the formation of the peptidoglycan building block, the UDP-linked MurNAc-pentapeptide.
The D-alanyl-D-alanine C-terminal residues of the peptidoglycan precursor unit are essential for important reactions which take place at the cell wall level such as peptide cross-linking, recognition by penicillin-binding proteins (PBPs) or recognition by the glycopeptide class of antibiotics (1, 2). The study of a murF insertion mutant (27) has allowed this gene to be added to the extensive list of auxiliary genes that are essential for the optimal expression of methicillin resistance in Staphylococcus aureus (4, 5).
MurF and its biochemical function are unique to bacteria; thus, such an enzyme is a potential antimicrobial target. Compounds with specific inhibitory action against MurF have been developed, but so far none with in vivo antibacterial activity has been described (9).
The purpose of the study described here was to construct a conditional mutant of murF and use it for the exploration of the physiological role of this determinant in growth, cell wall synthesis, and antibiotic susceptibility of S. aureus.
MATERIALS AND METHODS
Bacterial strains, plasmids, and growth conditions. Bacterial strains and plasmids used in this study are listed in Table 1. Staphylococcus aureus strains were grown at 37°C with aeration in tryptic soy broth (TSB) or tryptic soy agar (TSA) (Difco Laboratories, Detroit, Mich.). The conditional mutant strains, RN4220spacmurF, 27SspacmurF, COLspacmurF, and ZOX3spacmurF were grown in the presence of the respective antibiotics (Table 1), and the medium was always supplemented with 100 μM isopropyl--D-thiogalactopyranoside (ITPG; Sigma, St. Louis, MO), unless otherwise described.
Escherichia coli strains (Table 1) were grown in Luria-Bertani broth (LB; Difco Laboratories) with aeration at 37°C.
Erythromycin (10 μg/ml), chloramphenicol (10 μg/ml), and ampicillin (100 μg/ml) were used as recommended by the manufacturer (Sigma) for the selection and maintenance of S. aureus and E. coli mutants.
DNA methods. DNA manipulation was performed following standard methods (17). Restriction enzymes from New England Biolabs (Beverly, MA) were used as recommended by the manufacturer. Routine PCR amplification was performed with Tth DNA polymerase (HT Biotechnology, Cambridge, United Kingdom). The purification systems Wizard Plus Minipreps and Wizard Plus Midipreps (Promega, Madison, MA) were used for plasmid DNA extraction. PCR and digestion products were purified using Wizard PCR Preps and Wizard DNA Clean-Up systems. Ligation reactions were performed using T4 DNA ligase (New England Biolabs).
Construction of pRS10 plasmid. A 767-bp murF fragment was amplified by PCR with Pfu DNA polymerase (Stratagene, Heidelberg, Germany) using COL DNA as a template and the specific primers pmurFupSmaI and pmurFdnBglII (Table 2). The amplification conditions used were as follows: 94°C for 4 min; 30 cycles, each consisting of 94°C for 30 s, 50°C for 30 s, and 72°C for 1 min 30 s; and one final extension step of 72°C for 10 min. The amplified fragment and the integrative plasmid pMGPI (26) were both digested with SmaI and BglII and subsequently ligated, generating plasmid pRS10.
Construction of conditional mutants 27SspacmurF, COLspacmurF, and ZOX3spacmurF. Plasmid pRS10 was electroporated into electrocompetent cells of S. aureus RN4220 with a Gene Pulser apparatus (Bio-Rad, California) under conditions described previously (14). Selection of the transformants was performed using erythromycin and IPTG (500 μM). The correct insertion of pRS10 into the RN4220 chromosome was confirmed by PCR for six independent electrotransformants, with an internal murF primer chosen outside the region cloned in pRS10, p6, and a Pspac promoter primer, p5. Long-range PCR was performed to amplify the complete construct using primers p6 and p1 and Pfu Turbo DNA polymerase (Stratagene, Heidelberg, Germany) under the following conditions: 94°C for 4 min; 30 cycles, each consisting of 94°C for 30 s, 55°C for 30 s, and 72°C for 12 min; and one final extension step of 72°C for 15 min.
The conditional construct obtained and the pMGPII replicative plasmid were sequentially transduced into the background of strains 27S, COL, and ZOX3 (15) with phage 80 (21), using erythromycin, chloramphenicol, and IPTG as selection conditions. Independent transductants of 27SspacmurF, COLspacmurF, and ZOX3spacmurF were selected for further study.
Growth curves. The parental strains 27S, COL, and ZOX3 and overnight cultures of the conditional mutants 27SspacmurF, COLspacmurF, and ZOX3spacmurF were diluted 1:1,000 into 50 ml of fresh TSB supplemented with the respective antibiotics (Table 1). The conditional mutants were grown with the following IPTG concentrations: 0, 2.5, 5.0, 7.5, 10, and 100 μM. The cultures were incubated at 37°C with agitation, and the optical density at 620 nm (OD620) was monitored.
Determination of beta-lactam resistance. Overnight cultures were plated on TSA with increasing concentrations of IPTG (0, 2.5, 5.0, 7.5, 10, and 100 μM) and incubated overnight at 37°C. Oxacillin (1 mg; Sigma) and ceftizoxime (30 μg; Sigma) diffusion disks were used to measure inhibition halos.
Northern blotting analysis. Cells were grown in TSB at 37°C to an optical density at 620 nm of 0.7 to 0.8 (log-phase growth). Prior to harvesting the cells, RNAprotect Bacteria reagent (QIAGEN, Hilden, Germany) was added to the culture. The mixture was incubated for 5 min at room temperature, and RNA was extracted as previously described (27). Briefly, cells were centrifuged, frozen in dry ice, and resuspended in Trizol reagent (Gibco BRL, Maryland). The lysing procedure applied was mechanical disruption using silica beads and a FastPrep FP120 apparatus (Bio 101, La Jolla, Calif.). A chloroform extraction was performed, and the RNA was recovered by precipitation with isopropyl alcohol, washed with 80% ethanol, and resuspended in diethyl pyrocarbonate-treated water. The RNA samples were run in an agarose gel under denaturing conditions (0.66 M formaldehyde-1x morpholinepropanesulfonic acid [MOPS]; Sigma) and were blotted onto Hybond N+ membranes (Amersham, Buckinghamshire, United Kingdom). The DNA probes used for the hybridization were internal to ddlA, murF, mecA, pbpB, and pta genes and were amplified by PCR with the respective primers described in Table 2. The DNA probes were labeled with [-32P]dCTP (Amersham Life Sciences, New Jersey).
RT analysis. Reverse transcription-PCR (RT-PCR) was performed using the GeneAmp RNA PCR kit (Perkin Elmer). COL RNA treated with DNase was used as the template. Random hexamers, a primer internal to ddlA (p2), and a primer internal to murF (p7) were used for the reverse transcriptase reaction. The following conditions were applied: 94°C for 2 min; 30 cycles, each consisting of 94°C for 30 s, 53°C for 30 s, and 72°C for 2 min; and one final extension step of 72°C for 5 min.
Cell wall composition. The peptidoglycan composition was analyzed as previously described (3). Cells were harvested by centrifugation, and the cell wall-associated proteins were removed by extraction with boiling 10% sodium dodecyl sulfate. After the sodium dodecyl sulfate was washed off, the cell wall was mechanically disrupted with glass beads in the FastPrep FP120 apparatus, purified, and washed. The isolated cell wall obtained was then lyophilized and weighed; for each strain, the same amount of sample was used for analysis. The peptidoglycan fraction was next extracted with 49% hydrofluoric acid to remove teichoic acids, and the purified peptidoglycan was washed to remove all traces of the hydrofluoric acid reagents and was lyophilized. The same amounts of the dried peptidoglycan preparations isolated from the various constructs were carefully weighed, and identical amounts of material were used for hydrolysis by the M1 muramidase. The resulting muropeptides were separated by reversed-phase high-performance liquid chromatography (HPLC).
Analysis of the UDP-linked precursor pool. The UDP-linked peptidoglycan precursor cytoplasmic pool was extracted as previously described (20) and resolved by reversed-phase HPLC with an octyldecyl silane column (3 μm; particle size, 250 by 4.6 mm; pore size, 120 ) and the following linear elution gradient: 5% to 30% methanol in 100 mM sodium phosphate buffer, pH 2.5, at a flow rate of 0.5 ml/min. The sample absorbance was assayed at a wavelength of 254 nm.
Electron microscopy. Strain COL was grown in TSB, and COLspacmurF was grown in TSB supplemented with IPTG at the following concentrations: 0, 2.5, 5.0, 7.5, 10, and 100 μM. When grown in the presence of IPTG, glutaraldehyde-CaCo was added to the cultures at an OD620 of 0.7 to a final concentration of 2.5% glutaraldehyde-0.1 M CaCo. In the absence of inducer, COLspacmurF culture does not reach an OD value of 0.7; hence, glutaraldehyde-CaCo was added to a 10 times superior volume of culture when the OD620 value reached 0.07. The cultures were kept overnight at 4°C and then harvested by gentle centrifugation. The hard pellets obtained were covered by a small volume of a 2.5% glutaraldehyde-0.1 M CaCo. Preparation of the blocks, ultrathin sectioning, and electron micrography were performed at the Bio-Imaging Resource Center of The Rockefeller University.
RESULTS
Construction of murF conditional mutant. A 767-bp murF N-terminal fragment which does not contain the promoter region but includes the ribosome-binding site and the first 251 codons was cloned into the suicide vector pMGPI (26). The plasmid has inserted into the chromosomal DNA of RN4220 strain by Campbell-type recombination, resulting in a partial duplication of the murF gene in the chromosome, as shown in Fig. 1. The complete functional copy of the gene was now located immediately downstream from the inducible promoter Pspac. The second copy of murF was truncated at its C terminus by plasmid pRS10 integration and included only the first 251 codons of the 452 murF codons. The correct recombinational event was confirmed by Southern blotting, using an internal probe for murF (data not shown) and by PCR, using combinations of primers internal to murF and primers for lacI and Pspac (Fig. 2A). In both cases, fragments of the expected size were obtained (Fig. 2B). In this way, it was verified that the complete copy of the murF gene was placed under the control of the Pspac promoter.
The ddlA and murF genes are cotranscribed. The gene ddlA is located immediately upstream from murF in all S. aureus genomes sequenced, and both genes are transcribed in the same direction as represented in Fig. 1. The information that the ddlA stop codon and the murF methionine codon are separated by 14 bp only, along with the fact that they encode proteins with sequential biochemical functions, strongly suggested that they were cotranscribed. Northern blotting analysis was performed independently for both genes in COL and in RN4220, using internal labeled probes for each open reading frame (ORF). A single transcript approximately 2.5 kb long was obtained for both hybridizations, which matched the size of both genes (data not shown). The cotranscription of ddlA and murF was confirmed by RT-PCR (data not shown).
Controlling murF expression. It was previously shown that the Pspac promoter is leaky in the absence of the inducer (12). To prevent the basal transcription of murF, several copies of the repressor LacI were provided through the introduction of pMGPII (26) plasmid into the murF conditional mutant. This vector is a multicopy replicative plasmid containing the lacI gene. The PspacmurF construct and the pMGPII plasmid were subsequently transferred into strains 27S, COL, and ZOX3 by transduction, generating mutants 27SspacmurF, COLspacmurF, and ZOX3spacmurF, respectively.
The successful induction of murF gene expression in the presence of the inducer was confirmed in a COL background by Northern analysis, as shown in Fig. 3. More than one transcript could be visualized in the Northern blot from hybridization with the murF probe. The sizes of these transcripts (1.3, 2.5, and 6.7 kb) did not correspond exactly to the expected result (except for the band with approximately 1.3 kb, which should correspond to the murF copy transcribed from the spac promoter). The presence of these three different mRNAs seemed to vary with the IPTG concentration; thus, the expression of the three mRNAs seems to be under the control of the Pspac promoter. To further confirm the correct insertion of the pRS10 suicide plasmid, it was demonstrated by PCR that the 2.5-kb fragment containing the entire ddlA-murF operon could not be amplified from the mutant chromosomal DNA. Using long-range PCR conditions, a fragment of approximately 9.0 to 9.2 kb in size was amplified corresponding to the sizes of ddlA-murF operon (2.5 kb) plus the size of the pRS10 plasmid (5.9 + 0.8 = 6.7 kb) (Fig. 2).
Impact of murF on S. aureus growth. The growth of the conditional spacmurF mutant was determined in liquid medium for IPTG concentrations 0, 2.5, 5.0, 7.5, 10, and 100 μM by monitoring the culture absorbance, as shown for the case of one of these conditional mutants, 27SspacmurF. In the presence of 100 μM inducer, the mutant's growth curve was found to be very similar to that of the respective parental strain. This IPTG concentration was therefore considered to be the optimal inducer concentration. The growth rate of the conditional mutants varied with IPTG concentration in the medium. No growth was detectable in the absence of IPTG (Fig. 4A and B).
Impact of murF on -lactam resistance. COLspacmurF was plated on TSA containing various IPTG concentrations (0, 2.5, 5.0, 7.5, 10, and 100 μM) and was tested for effect on oxacillin resistance. The sizes of the inhibition zones varied with the concentration of IPTG (Fig. 5A).
The conditional mutation of murF was also introduced into strain ZOX3, which is a spontaneous mutant derivative of the fully antibiotic-susceptible S. aureus strain 27S (15). ZOX3 has an increased level of resistance to ceftizoxime, a beta-lactam antibiotic with high selective affinity for S. aureus PBP2. Strain ZOX3 owes its resistance to ceftizoxime to a single point mutation in the pbpB structural gene, causing decreased affinity of PBP2 for ceftizoxime (15). ZOX3spacmurF was plated on TSA containing 2.5, 5.0, 7.5, 10, and 100 μM concentrations of IPTG and was assayed for ceftizoxime inhibition halos, using 30-μg disks. For IPTG concentrations of 10 and 100 μM, the size of inhibition halos was the same as that of the parental strain ZOX3. However, for lower IPTG concentrations (7.5, 5.0, and 2.5 μM), resistance decreased to the level of the susceptible parental strain 27S (Fig. 5B).
Changes in the composition of cell wall precursor pool. The cytoplasmic fraction of strains COL and COLspacmurF were isolated from cultures grown at several IPTG concentrations, and the cell wall precursor pool was purified and analyzed by HPLC.
The HPLC profiles showed that as the IPTG concentration decreased, an accumulation of UDP-linked muramyl tripeptide occurred in the cytoplasm, paralleled by the gradual reduction in the amount of UPD-linked muramyl pentapeptide (Fig. 6). The relative accumulation values for the UDP-MurNAc-tripeptide and UDP-MurNAc-pentapeptide are listed in Table 3.
Changes in the composition of cell wall peptidoglycan. The peptidoglycan composition of strain COLspacmurF was determined for bacteria grown in the presence of optimal (100 μM) and suboptimal IPTG concentrations. The HPLC profiles of enzymatic hydrolysates of these peptidoglycans are shown in Fig. 7. For cells grown in 100 μM IPTG-supplemented medium, the peptidoglycan HPLC profile was identical to that of strain COL.
At all suboptimal IPTG concentrations, a new, abnormal, monomeric muropeptide appeared in the HPLC profile. This component was identified by mass spectrometry as the disaccharide tripeptide lacking the pentaglycine side chain (27). The relative amounts of this muropeptide were smallest (5%) in cells grown at 10 μM IPTG but remained constant (7%) for all other suboptimal concentrations of IPTG (Table 4).
The peptidoglycan composition of ZOX3spacmurF was also determined for the same IPTG concentrations. For the suboptimal IPTG concentrations of 10, 7.5, 5.0, and 2.5 μM, the disaccharide tripeptide structure appeared in the cell wall, but the incorporation level was much less than in the COL background. For instance, in bacteria grown in 10 μM IPTG, the tripeptide represented 2% and, for the lower concentrations, about 3% of all muropeptides (data not shown).
Decrease in the cellular amounts of peptidoglycan. Cell walls purified from COLspacmurF grown at different concentrations of IPTG were dried and weighed, and identical (5-mg) amounts were treated with hydrofluoric acid to remove wall teichoic acids. After hydrofluoric acid treatment, the peptidoglycan preparations were dried and weighed. Cells grown in the presence of 100 μM IPTG produced peptidoglycan in excess of what was found in strain COL, while all bacteria grown at suboptimal inducer concentrations contained reduced amounts of peptidoglycan (Table 4).
Cell morphology and cell wall thickness. Mid-exponential-phase cells of strain COL and its murF conditional mutant grown with several IPTG concentrations were analyzed by electron microscopy (Fig. 8). For each inducer concentration tested, the average cell wall thickness value was calculated, based on measurement of >100 cross-sectioned cells. The cell wall thickness values for all mutants were expressed relative to that of the parental strain COL, which was set to 100% (Table 4). When grown in the presence of 100 μM IPTG, the mutant cells showed a thicker and less sharply defined cell wall, indicating that this inducer concentration did not exactly mimic the native expression of murF, despite the unaltered antibiotic resistance, growth rate, and HPLC profile. A cell wall thickness similar to the parental cell was obtained for bacteria grown in 10 μM IPTG. Mutants cultivated at suboptimal concentrations (7.5, 5.0, 2.5, and 0 μM) showed a 20 to 30% reduction in cell wall thickness.
In mutant cultures grown at or below 5.0 μM of IPTG, several morphological abnormalities appeared in most (>90%) of the bacteria. These abnormalities included asymmetrically distributed and thickened septae, which contrasted sharply with the thinner peripheral cell wall. Some cells also showed multiple septae and hollow cells (ghost cells) with multiple vesicles in the cytoplasm. Accumulation of cellular debris in the medium (lysis) was observed in cells cultivated without IPTG. An overall increase of the cell diameter was also apparent and the appearance of the chromosomal area was also abnormal (more condensed) in many cells (Fig. 8).
Altered transcription of mecA and pbpB genes. The expression of mecA and pbpB genes, which encode PBP2A and PBP2, respectively, was analyzed by Northern blotting. The hybridization performed with a pbpB internal probe showed two bands. In fact, pbpB gene can be transcribed either alone or together with the upstream gene recU (23), resulting in two independent mRNAs species, the pbpB transcript (2.1 kb) and the transcript from pbpB and recU (2.9 kb).
Figure 9 shows that the transcription of mecA and pbpB was reduced in COLspacmurF (Fig. 9A) grown at the suboptimal IPTG concentrations, compared to that for the parental strain COL. For IPTG concentrations of 100 μM and 10 μM, both transcript signal intensities were stronger than in the parental strain. In the murF conditional mutant derivatives of both 27S and ZOX3, the transcription of pbpB varied in parallel with the transcription of murF (Fig. 9B). As an internal control, the transcription of the housekeeping gene pta (expressing phosphate acetyltransferase) was also determined at the same IPTG concentrations. No significant differences were detected (Fig. 9C).
DISCUSSION
The MurF enzyme catalyzes a critical step in the biosynthesis of the bacterial cell wall: the addition of the C-terminal D-alanyl-D-alanine dipeptide to the cell wall precursor muropeptide. The construction of a conditional mutant of murF in the background of both methicillin-resistant S. aureus (MRSA) and methicillin-susceptible S. aureus (MSSA) strains has allowed us to explore the physiological functions of this gene in S. aureus in more detail than was possible from the study of a murF insertional mutant (27).
S. aureus strains carrying the spacmurF construct showed an absolute dependence on the IPTG inducer for growth, confirming the essential nature of the murF gene in Staphylococcus aureus. Omission of the IPTG inducer from the medium prevented growth in both MRSA and MSSA strains, and suboptimal concentrations of IPTG caused parallel decreases in growth rate and murF expression levels. At the inducer concentration of 100 μM IPTG, the growth rate and yield were indistinguishable from those of parental strains carrying the native promoters. The essentiality of the murF gene had already been reported in the gram-negative bacteria Escherichia coli through the study of temperature-sensitive mutants, which would lyse when grown at restrictive temperatures (16).
Open reading frame SACOL0054 was identified in staphylococcal cassette chromosome mec type I (11), present in strain COL. This ORF was identified as a member of the Mur ligase family of proteins by homology (29.6%) with the murF homologue from Mesorhizobium loti. It has 25.3% homology and 49.2% similarity to the COL murF gene. The essentiality of the S. aureus murF gene shown by the study of COL conditional mutant suggests that ORF SACOL0054 cannot replace the chromosomal murF. A frameshift mutation was described in ORF SACOL0054, which may be responsible for the synthesis of a nonfunctional protein.
An analysis of the extracts of the mutants growing in the presence of different IPTG concentrations demonstrated the sensitive dependence of the composition of the cell wall precursor pool on the expression of murF. In COLspacmurF growing in the presence of 100 μM IPTG, the relative amount of the UDP-MurNAc-pentapeptide in the cell wall precursor pool was 64.1%, virtually the same as in strain COL (64.8%) growing with its native murF promoter. There was no detectable tripeptide in the wall precursor pools of these bacteria. However, reduction of the concentration of IPTG in the growth medium caused striking changes in the composition of cell wall precursors. Decrease in the IPTG concentration from 100 to 10, 7.5, 5.0, and 2.5 μM caused a stepwise decrease in the amount of pentapeptides from 64.1 to 11.1% and a parallel increase in the abnormal tripeptide component from undetectable to 59.3% (Table 3). The progressive accumulation of tripeptide and concomitant shortage of pentapeptide precursors indicate that maintenance of normal levels of a catalytically active MurF depends on a steady expression of the murF gene.
Analysis of the peptidoglycan composition of bacteria grown at different IPTG concentrations demonstrated that the tripeptides were also able to incorporate into the polymerized cell wall. However, the maximum amounts of disaccharide tripeptides in the cell wall had an upper limit, from 6.7 to 7.1% of all muropeptides in the murF mutant of MRSA strain COL and up to 2 to 3% in the MSSA strains (Table 4). This observation was in sharp contrast to the more extensive variations in the composition of the cell wall precursor pool, in which the relative amounts of tripeptides varied through the entire range of IPTG concentrations used.
The abnormal disaccharide tripeptides lacking the C-terminal D-alanyl-D-alanine residues cannot participate as donors in the transpeptidation reaction, but they could, in principle, serve as acceptors. However, careful examination of the HPLC profiles showed no evidence for muropeptide oligomers of such structures. Thus, the disaccharide tripeptides appearing in the cell walls of the murF mutants were only present as monomers, which must have been incorporated into the peptidoglycan through the activity of transglycosylases such as the native PBP2 or one of the monofunctional transglycosylases present in S. aureus (6, 28).
The drastic alteration in the composition of the cell wall precursor pool in the murF mutants growing at suboptimal concentrations of IPTG presents a serious dilemma for cell wall biosynthesis. The utilization of the tripeptides appears to be limited, presumably because enrichment of the peptidoglycan in monomeric components above these threshold levels may jeopardize the structural integrity of the cell wall and is lethal for the bacteria. This suggests that the S. aureus cell requires a minimal level of peptidoglycan cross-linking, below which it is not able to grow and divide. On the other hand, the limited expression of murF also leads to a greatly diminished pool of the normal pentapeptide components. Under these conditions, the only option for the cells appears to be to produce less peptidoglycan. This was actually observed in the murF mutants growing at suboptimal IPTG concentrations: electron microscopic analysis showed a striking decrease in the thickness of peripheral cell wall, and direct determination of the yield of peptidoglycan isolated from such bacteria showed a gradual decrease in the cellular amounts of peptidoglycan, which represented about half (55%) of the dry weight of purified cell walls in the wild-type strain COL but dropped to about 34 to 36% in the mutant cells grown at suboptimal concentrations of IPTG (Table 4).
The 100 μM IPTG concentration was considered optimal for murF expression, since it allowed normal growth rate and maximum expression of beta-lactam resistance and normal composition of the cell wall precursor pool, as well as peptidoglycan. Nevertheless, to our surprise, we found that the cell walls of bacteria grown under these conditions showed profound alterations: there was significant thickening (by about 20%) of the cell wall and there was an increase in the cellular amounts of peptidoglycan as well (from 55% to 60%). These findings suggested that in bacteria growing with 100 μM IPTG, the expression of murF might be increased over the level of the parental strain using its native murF promoter. Comparison of murF transcription levels confirmed this (Fig. 3).
At 10 μM IPTG, the cell wall thickness and the peptidoglycan recovery yield were very similar to the parental values. However, the tripeptide precursor synthesis and its incorporation (5%) into the cell wall still occurred in very significant amounts. The fact that the oxacillin resistance level was visibly affected for this IPTG concentration suggests that the resistance decrease is not related to alterations in the cell wall thickness but may be related to changes in its composition and/or the composition of the cell wall precursor pool. A decrease in oxacillin resistance has frequently been observed, along with changes in the peptidoglycan composition of mutants of other cell wall-related genes, such as murE (20), glnR/A (10, 19), and femA and femB (13).
The availability of the murF conditional mutants has allowed us to produce S. aureus strains in which the expression of murF depended on the concentration of the IPTG inducer added to the medium. Most interestingly, ddlA, supplying the D-alanyl-D-alanine substrate for the MurF ligase, appears to be cotranscribed with murF. This and the profound changes observed in the conditional mutants in composition of cell wall precursors, cellular amounts of peptidoglycan, and the fine structure of the cell wall indicate that the addition of the C-terminal dipeptide to the cell wall precursor represents an important control point in the peptidoglycan synthesis of S. aureus. The greatly increased thickness of septal areas in the same cells in which the thickness of the peripheral cell wall was decreased, the substantial increase in cell diameters, and the other morphological abnormalities observed in the cells with the suboptimal murF function indicate that undisturbed functioning of the ddlA-murF system is essential for normal cell division in these bacteria.
The transcriptional analysis by Northern blotting showed that mecA (PBP2A) and pbpB (PBP2) transcription is also increased in the murF conditional mutants at IPTG concentrations of 10 and 100 μM, a fact that may explain the mechanism of overproduction of peptidoglycan. Altered transcription of mecA and pbpB genes has already been observed with a murF insertion mutant of strain COL (27) and was also noted with a murE conditional mutant (8). The presence of abnormal precursors may play a role in the altered transcription rates but cannot alone account for the regulation of these transcripts, since both mecA and pbpB transcripts are overexpressed in the presence of both 100 μM and 10 μM IPTG.
Regarding beta-lactam resistance, the spacmurF construct was analyzed in two different backgrounds: COL, a MRSA strain with constitutive expression of the mecA gene, and ZOX3, a MSSA strain which lacks the mecA gene and which is resistant to the beta-lactam cefizoxime. It was shown that in the absence of murF, both oxacillin (in COL) and ceftizoxime (in ZOX3) resistance decreases. These results suggest that murF affects beta-lactam resistance even in the absence of the mecA gene.
The transcription of pbpB was found to parallel the decrease of murF, not only in COL but also in the MSSA strains 27S and ZOX3. Thus, the postulated coregulatory mechanism between murF and pbpB transcription must be independent of mecA.
The level of incorporation of the abnormal disaccharide tripeptide was different in the background of the MRSA strain COL (7%) compared to that in the MSSA strain ZOX3 (2 to 3%). We propose that these differences may be related to a recent observation concerning the cooperative functioning of PBP2 and PBP2A. It has been shown that in MSSA strains, PBP2 normally localizes to the division septum, where cell wall synthesis is known to take place in S. aureus (22, 24). Pinho and colleagues have also shown that in the presence of modified precursors, specifically in the presence of tripeptide precursors accumulating in cells exposed to D-cycloserine, PBP2 loses the capacity to correctly localize at the septum and delocalizes along the peripheral cell wall. However, in an MRSA strain a functional PBP2A is able to maintain the correct localization of PBP2 at the septum, in spite of the presence of the inhibitor (25). The higher level of incorporation of the tripeptide into the peptidoglycan of the MRSA strain COL compared to that of the MSSA strain ZOX3 may then be explained by this putative "helper" function of PBP2A, present only in the antibiotic-resistant strain COL.
ACKNOWLEDGMENTS
Partial support for this study was provided by a grant (2 RO1 A1045738-06) from the National Institutes of Health, the U.S. Public Health Service, and contract POCTI/BIA-MIC/58416/2004 (Fundao para a Ciência e a Tecnologia, Portugal). R.G.S. was supported by grants SFRH/BD/3138/2000 and 022/BI/2005 (Fundao para a Ciência e a Tecnologia, Portugal).
REFERENCES
Barna, J. C., and D. H. Williams. 1984. The structure and mode of action of glycopeptide antibiotics of the vancomycin group. Annu. Rev. Microbiol. 38:339-357.
Beauregard, D. A., A. J. Maguire, D. H. Williams, and P. E. Reynolds. 1997. Semiquantitation of cooperativity in binding of vancomycin-group antibiotics to vancomycin-susceptible and -resistant organisms. Antimicrob. Agents Chemother. 41:2418-2423.
de Jonge, B. L., Y. S. Chang, D. Gage, and A. Tomasz. 1992. Peptidoglycan composition of a highly methicillin-resistant Staphylococcus aureus strain. The role of penicillin binding protein 2A. J. Biol. Chem. 267:11248-11254.
de Lencastre, H., and A. Tomasz. 1994. Reassessment of the number of auxiliary genes essential for expression of high-level methicillin resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 38:2590-2598.
de Lencastre, H., S. W. Wu, M. G. Pinho, A. M. Ludovice, S. Filipe, S. Gardete, R. Sobral, S. Gill, M. Chung, and A. Tomasz. 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.
Di Berardino, M., A. Dijkstra, D. Stuber, W. Keck, and M. Gubler. 1996. The monofunctional glycosyltransferase of Escherichia coli is a member of a new class of peptidoglycan-synthesising enzymes. FEBS Lett. 392:184-188.
Enright, M. C., N. P. Day, C. E. Davies, S. J. Peacock, and B. G. Spratt. 2000. Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. J. Clin. Microbiol. 38:1008-1015.
Gardete, S., A. M. Ludovice, R. G. Sobral, S. R. Filipe, H. de Lencastre, and A. Tomasz. 2004. Role of murE in the expression of -lactam antibiotic resistance in Staphylococcus aureus. J. Bacteriol. 186:1705-1713.
Gu, Y. G., A. S. Florjancic, R. F. Clark, T. Zhang, C. S. Cooper, D. D. Anderson, C. G. Lerner, J. O. McCall, Y. Cai, C. L. Black-Schaefer, G. F. Stamper, P. J. Hajduk, and B. A. Beutel. 2004. Structure-activity relationships of novel potent MurF inhibitors. Bioorg. Med. Chem. Lett. 14:267-270.
Gustafson, J., A. Strassle, H. Hachler, F. H. Kayser, and B. Berger-Bachi. 1994. The femC locus of Staphylococcus aureus required for methicillin resistance includes the glutamine synthetase operon. J. Bacteriol. 176:1460-1467.
Ito, T., Y. Katayama, K. Asada, N. Mori, K. Tsutsumimoto, C. Tiensasitorn, and K. Hiramatsu. 2001. Structural comparison of three types of staphylococcal cassette chromosome mec integrated in the chromosome in methicillin-resistant Staphylococcus aureus. Antimicrob. Agents Chemother. 45:1323-1336.
Jana, M., T. T. Luong, H. Komatsuzawa, M. Shigeta, and C. Y. Lee. 2000. A method for demonstrating gene essentiality in Staphylococcus aureus. Plasmid 44:100-104.
Kopp, U., M. Roos, J. Wecke, and H. Labischinski. 1996. Staphylococcal peptidoglycan interpeptide bridge biosynthesis: a novel antistaphylococcal target Microb. Drug Resist. 2:29-41.
Kraemer, G. R., and J. J. Iandolo. 1990. High-frequency transformation of Staphylococcus aureus by electroporation. Curr. Microbiol. 21:373-376.
Leski, T. A., and A. Tomasz. 2005. Role of penicillin-binding protein 2 (PBP2) in the antibiotic susceptibility and cell wall cross-linking of Staphylococcus aureus: evidence for the cooperative functioning of PBP2, PBP4, and PBP2A. J. Bacteriol. 187:1815-1824.
Lugtenberg, E. J. J., and A. Van Schijndel-van Dam. 1972. Temperature-sensitive mutants of Escherichia coli K-12 with low activities of the L-alanine adding enzyme and the D-alanyl-D-alanine adding enzyme. J. Bacteriol. 110:35-40.
Maniatis, V., E. F. Fritsch, and J. Sambrook. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Miragaia, M., I. Couto, and H. de Lencastre. 2005. Genetic diversity among methicillin-resistant Staphylococcus epidermidis (MRSE). Microb. Drug Resist. 11:83-93.
Ornelas-Soares, A., H. de Lencastre, B. de Jonge, D. Gage, Y. S. Chang, and A. Tomasz. 1993. The peptidoglycan composition of a Staphylococcus aureus mutant selected for reduced methicillin resistance. J. Biol. Chem. 268:26268-26272.
Ornelas-Soares, A., H. de Lencastre, B. L. de Jonge, and A. Tomasz. 1994. Reduced methicillin resistance in a new Staphylococcus aureus transposon mutant that incorporates muramyl dipeptides into the cell wall peptidoglycan. J. Biol. Chem. 269:27246-27250.
Oshida, T., and A. Tomasz. 1992. Isolation and characterization of a Tn551-autolysis mutant of Staphylococcus aureus. J. Bacteriol. 174:4952-4959.
Pinho, M. G., H. de Lencastre, and A. Tomasz. 2001. An acquired and a native penicillin-binding protein cooperate in building the cell wall of drug-resistant staphylococci. Proc. Natl. Acad. Sci. USA 98:10886-10891.
Pinho, M. G., H. de Lencastre, and A. Tomasz. 1998. Transcriptional analysis of the Staphylococcus aureus penicillin binding protein 2 gene. J. Bacteriol. 180:6077-6081.
Pinho, M. G., and J. Errington. 2003. Dispersed mode of Staphylococcus aureus cell wall synthesis in the absence of the division machinery. Mol. Microbiol. 50:871-881.
Pinho, M. G., and J. Errington. 2005. Recruitment of penicillin-binding protein PBP2 to the division site of Staphylococcus aureus is dependent on its transpeptidation substrates. Mol. Microbiol. 55:799-807.
Pinho, M. G., S. R. Filipe, H. de Lencastre, and A. Tomasz. 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.
Sobral, R. G., A. M. Ludovice, S. Gardete, K. Tabei, H. de Lencastre, and A. Tomasz. 2003. Normally functioning murF is essential for the optimal expression of methicillin resistance in Staphylococcus aureus. Microb. Drug Resist. 9:231-241.
Wang, Q. M., R. B. Peery, R. B. Johnson, W. E. Alborn, W. K. Yeh, and P. L. Skatrud. 2001. Identification and characterization of a monofunctional glycosyltransferase from Staphylococcus aureus. J. Bacteriol. 183:4779-4785.(R. G. Sobral, A. M. Ludov)
ABSTRACT
The Staphylococcus aureus murF gene was placed under the control of a promoter inducible by IPTG (isopropyl--D-thiogalactopyranoside). It was demonstrated that murF is an essential gene; it is cotranscribed with ddlA and growth rate, level of beta-lactam antibiotic resistance, and rates of transcription of the mecA and pbpB genes paralleled the rates of transcription of murF. At suboptimal concentrations of the inducer, a UDP-linked muramyl tripeptide accumulated in the cytoplasm in parallel with the decline in the amounts of the normal pentapeptide cell wall precursor. The abnormal tripeptide component incorporated into the cell wall as a monomeric muropeptide, accompanied by a decrease in the oligomerization degree of the peptidoglycan. However, incorporation of the tripeptide into the cell wall was limited to a relatively low threshold value. Further reduction of the amounts of pentapeptide cell wall precursor caused a gradual decrease in the cellular amounts of peptidoglycan, the production of a thinner peripheral cell wall, aberrant septae, and an overall increase in the diameter of the cells. The observations suggest that the role of murF exceeds its primary function in peptidoglycan biosynthesis and may also be involved in the control of cell division.
INTRODUCTION
The cell wall of Staphylococcus aureus contains a highly cross-linked peptidoglycan in which most of the muropeptide units have pentaglycine branches attached to the aminogroup of the lysine residue, and virtually all monomeric and acceptor muropeptides carry a carboxy-terminal D-alany-D-alanine residue (3). The biosynthesis and attachment of this dipeptide is catalyzed by the protein product of the genes ddlA and murF, with the first, DdlA, producing the dipeptide and the second, MurF, attaching the dipeptide to the UDP-N-acetylmuramic acid (MurNAc)-tripeptide, thus completing the formation of the peptidoglycan building block, the UDP-linked MurNAc-pentapeptide.
The D-alanyl-D-alanine C-terminal residues of the peptidoglycan precursor unit are essential for important reactions which take place at the cell wall level such as peptide cross-linking, recognition by penicillin-binding proteins (PBPs) or recognition by the glycopeptide class of antibiotics (1, 2). The study of a murF insertion mutant (27) has allowed this gene to be added to the extensive list of auxiliary genes that are essential for the optimal expression of methicillin resistance in Staphylococcus aureus (4, 5).
MurF and its biochemical function are unique to bacteria; thus, such an enzyme is a potential antimicrobial target. Compounds with specific inhibitory action against MurF have been developed, but so far none with in vivo antibacterial activity has been described (9).
The purpose of the study described here was to construct a conditional mutant of murF and use it for the exploration of the physiological role of this determinant in growth, cell wall synthesis, and antibiotic susceptibility of S. aureus.
MATERIALS AND METHODS
Bacterial strains, plasmids, and growth conditions. Bacterial strains and plasmids used in this study are listed in Table 1. Staphylococcus aureus strains were grown at 37°C with aeration in tryptic soy broth (TSB) or tryptic soy agar (TSA) (Difco Laboratories, Detroit, Mich.). The conditional mutant strains, RN4220spacmurF, 27SspacmurF, COLspacmurF, and ZOX3spacmurF were grown in the presence of the respective antibiotics (Table 1), and the medium was always supplemented with 100 μM isopropyl--D-thiogalactopyranoside (ITPG; Sigma, St. Louis, MO), unless otherwise described.
Escherichia coli strains (Table 1) were grown in Luria-Bertani broth (LB; Difco Laboratories) with aeration at 37°C.
Erythromycin (10 μg/ml), chloramphenicol (10 μg/ml), and ampicillin (100 μg/ml) were used as recommended by the manufacturer (Sigma) for the selection and maintenance of S. aureus and E. coli mutants.
DNA methods. DNA manipulation was performed following standard methods (17). Restriction enzymes from New England Biolabs (Beverly, MA) were used as recommended by the manufacturer. Routine PCR amplification was performed with Tth DNA polymerase (HT Biotechnology, Cambridge, United Kingdom). The purification systems Wizard Plus Minipreps and Wizard Plus Midipreps (Promega, Madison, MA) were used for plasmid DNA extraction. PCR and digestion products were purified using Wizard PCR Preps and Wizard DNA Clean-Up systems. Ligation reactions were performed using T4 DNA ligase (New England Biolabs).
Construction of pRS10 plasmid. A 767-bp murF fragment was amplified by PCR with Pfu DNA polymerase (Stratagene, Heidelberg, Germany) using COL DNA as a template and the specific primers pmurFupSmaI and pmurFdnBglII (Table 2). The amplification conditions used were as follows: 94°C for 4 min; 30 cycles, each consisting of 94°C for 30 s, 50°C for 30 s, and 72°C for 1 min 30 s; and one final extension step of 72°C for 10 min. The amplified fragment and the integrative plasmid pMGPI (26) were both digested with SmaI and BglII and subsequently ligated, generating plasmid pRS10.
Construction of conditional mutants 27SspacmurF, COLspacmurF, and ZOX3spacmurF. Plasmid pRS10 was electroporated into electrocompetent cells of S. aureus RN4220 with a Gene Pulser apparatus (Bio-Rad, California) under conditions described previously (14). Selection of the transformants was performed using erythromycin and IPTG (500 μM). The correct insertion of pRS10 into the RN4220 chromosome was confirmed by PCR for six independent electrotransformants, with an internal murF primer chosen outside the region cloned in pRS10, p6, and a Pspac promoter primer, p5. Long-range PCR was performed to amplify the complete construct using primers p6 and p1 and Pfu Turbo DNA polymerase (Stratagene, Heidelberg, Germany) under the following conditions: 94°C for 4 min; 30 cycles, each consisting of 94°C for 30 s, 55°C for 30 s, and 72°C for 12 min; and one final extension step of 72°C for 15 min.
The conditional construct obtained and the pMGPII replicative plasmid were sequentially transduced into the background of strains 27S, COL, and ZOX3 (15) with phage 80 (21), using erythromycin, chloramphenicol, and IPTG as selection conditions. Independent transductants of 27SspacmurF, COLspacmurF, and ZOX3spacmurF were selected for further study.
Growth curves. The parental strains 27S, COL, and ZOX3 and overnight cultures of the conditional mutants 27SspacmurF, COLspacmurF, and ZOX3spacmurF were diluted 1:1,000 into 50 ml of fresh TSB supplemented with the respective antibiotics (Table 1). The conditional mutants were grown with the following IPTG concentrations: 0, 2.5, 5.0, 7.5, 10, and 100 μM. The cultures were incubated at 37°C with agitation, and the optical density at 620 nm (OD620) was monitored.
Determination of beta-lactam resistance. Overnight cultures were plated on TSA with increasing concentrations of IPTG (0, 2.5, 5.0, 7.5, 10, and 100 μM) and incubated overnight at 37°C. Oxacillin (1 mg; Sigma) and ceftizoxime (30 μg; Sigma) diffusion disks were used to measure inhibition halos.
Northern blotting analysis. Cells were grown in TSB at 37°C to an optical density at 620 nm of 0.7 to 0.8 (log-phase growth). Prior to harvesting the cells, RNAprotect Bacteria reagent (QIAGEN, Hilden, Germany) was added to the culture. The mixture was incubated for 5 min at room temperature, and RNA was extracted as previously described (27). Briefly, cells were centrifuged, frozen in dry ice, and resuspended in Trizol reagent (Gibco BRL, Maryland). The lysing procedure applied was mechanical disruption using silica beads and a FastPrep FP120 apparatus (Bio 101, La Jolla, Calif.). A chloroform extraction was performed, and the RNA was recovered by precipitation with isopropyl alcohol, washed with 80% ethanol, and resuspended in diethyl pyrocarbonate-treated water. The RNA samples were run in an agarose gel under denaturing conditions (0.66 M formaldehyde-1x morpholinepropanesulfonic acid [MOPS]; Sigma) and were blotted onto Hybond N+ membranes (Amersham, Buckinghamshire, United Kingdom). The DNA probes used for the hybridization were internal to ddlA, murF, mecA, pbpB, and pta genes and were amplified by PCR with the respective primers described in Table 2. The DNA probes were labeled with [-32P]dCTP (Amersham Life Sciences, New Jersey).
RT analysis. Reverse transcription-PCR (RT-PCR) was performed using the GeneAmp RNA PCR kit (Perkin Elmer). COL RNA treated with DNase was used as the template. Random hexamers, a primer internal to ddlA (p2), and a primer internal to murF (p7) were used for the reverse transcriptase reaction. The following conditions were applied: 94°C for 2 min; 30 cycles, each consisting of 94°C for 30 s, 53°C for 30 s, and 72°C for 2 min; and one final extension step of 72°C for 5 min.
Cell wall composition. The peptidoglycan composition was analyzed as previously described (3). Cells were harvested by centrifugation, and the cell wall-associated proteins were removed by extraction with boiling 10% sodium dodecyl sulfate. After the sodium dodecyl sulfate was washed off, the cell wall was mechanically disrupted with glass beads in the FastPrep FP120 apparatus, purified, and washed. The isolated cell wall obtained was then lyophilized and weighed; for each strain, the same amount of sample was used for analysis. The peptidoglycan fraction was next extracted with 49% hydrofluoric acid to remove teichoic acids, and the purified peptidoglycan was washed to remove all traces of the hydrofluoric acid reagents and was lyophilized. The same amounts of the dried peptidoglycan preparations isolated from the various constructs were carefully weighed, and identical amounts of material were used for hydrolysis by the M1 muramidase. The resulting muropeptides were separated by reversed-phase high-performance liquid chromatography (HPLC).
Analysis of the UDP-linked precursor pool. The UDP-linked peptidoglycan precursor cytoplasmic pool was extracted as previously described (20) and resolved by reversed-phase HPLC with an octyldecyl silane column (3 μm; particle size, 250 by 4.6 mm; pore size, 120 ) and the following linear elution gradient: 5% to 30% methanol in 100 mM sodium phosphate buffer, pH 2.5, at a flow rate of 0.5 ml/min. The sample absorbance was assayed at a wavelength of 254 nm.
Electron microscopy. Strain COL was grown in TSB, and COLspacmurF was grown in TSB supplemented with IPTG at the following concentrations: 0, 2.5, 5.0, 7.5, 10, and 100 μM. When grown in the presence of IPTG, glutaraldehyde-CaCo was added to the cultures at an OD620 of 0.7 to a final concentration of 2.5% glutaraldehyde-0.1 M CaCo. In the absence of inducer, COLspacmurF culture does not reach an OD value of 0.7; hence, glutaraldehyde-CaCo was added to a 10 times superior volume of culture when the OD620 value reached 0.07. The cultures were kept overnight at 4°C and then harvested by gentle centrifugation. The hard pellets obtained were covered by a small volume of a 2.5% glutaraldehyde-0.1 M CaCo. Preparation of the blocks, ultrathin sectioning, and electron micrography were performed at the Bio-Imaging Resource Center of The Rockefeller University.
RESULTS
Construction of murF conditional mutant. A 767-bp murF N-terminal fragment which does not contain the promoter region but includes the ribosome-binding site and the first 251 codons was cloned into the suicide vector pMGPI (26). The plasmid has inserted into the chromosomal DNA of RN4220 strain by Campbell-type recombination, resulting in a partial duplication of the murF gene in the chromosome, as shown in Fig. 1. The complete functional copy of the gene was now located immediately downstream from the inducible promoter Pspac. The second copy of murF was truncated at its C terminus by plasmid pRS10 integration and included only the first 251 codons of the 452 murF codons. The correct recombinational event was confirmed by Southern blotting, using an internal probe for murF (data not shown) and by PCR, using combinations of primers internal to murF and primers for lacI and Pspac (Fig. 2A). In both cases, fragments of the expected size were obtained (Fig. 2B). In this way, it was verified that the complete copy of the murF gene was placed under the control of the Pspac promoter.
The ddlA and murF genes are cotranscribed. The gene ddlA is located immediately upstream from murF in all S. aureus genomes sequenced, and both genes are transcribed in the same direction as represented in Fig. 1. The information that the ddlA stop codon and the murF methionine codon are separated by 14 bp only, along with the fact that they encode proteins with sequential biochemical functions, strongly suggested that they were cotranscribed. Northern blotting analysis was performed independently for both genes in COL and in RN4220, using internal labeled probes for each open reading frame (ORF). A single transcript approximately 2.5 kb long was obtained for both hybridizations, which matched the size of both genes (data not shown). The cotranscription of ddlA and murF was confirmed by RT-PCR (data not shown).
Controlling murF expression. It was previously shown that the Pspac promoter is leaky in the absence of the inducer (12). To prevent the basal transcription of murF, several copies of the repressor LacI were provided through the introduction of pMGPII (26) plasmid into the murF conditional mutant. This vector is a multicopy replicative plasmid containing the lacI gene. The PspacmurF construct and the pMGPII plasmid were subsequently transferred into strains 27S, COL, and ZOX3 by transduction, generating mutants 27SspacmurF, COLspacmurF, and ZOX3spacmurF, respectively.
The successful induction of murF gene expression in the presence of the inducer was confirmed in a COL background by Northern analysis, as shown in Fig. 3. More than one transcript could be visualized in the Northern blot from hybridization with the murF probe. The sizes of these transcripts (1.3, 2.5, and 6.7 kb) did not correspond exactly to the expected result (except for the band with approximately 1.3 kb, which should correspond to the murF copy transcribed from the spac promoter). The presence of these three different mRNAs seemed to vary with the IPTG concentration; thus, the expression of the three mRNAs seems to be under the control of the Pspac promoter. To further confirm the correct insertion of the pRS10 suicide plasmid, it was demonstrated by PCR that the 2.5-kb fragment containing the entire ddlA-murF operon could not be amplified from the mutant chromosomal DNA. Using long-range PCR conditions, a fragment of approximately 9.0 to 9.2 kb in size was amplified corresponding to the sizes of ddlA-murF operon (2.5 kb) plus the size of the pRS10 plasmid (5.9 + 0.8 = 6.7 kb) (Fig. 2).
Impact of murF on S. aureus growth. The growth of the conditional spacmurF mutant was determined in liquid medium for IPTG concentrations 0, 2.5, 5.0, 7.5, 10, and 100 μM by monitoring the culture absorbance, as shown for the case of one of these conditional mutants, 27SspacmurF. In the presence of 100 μM inducer, the mutant's growth curve was found to be very similar to that of the respective parental strain. This IPTG concentration was therefore considered to be the optimal inducer concentration. The growth rate of the conditional mutants varied with IPTG concentration in the medium. No growth was detectable in the absence of IPTG (Fig. 4A and B).
Impact of murF on -lactam resistance. COLspacmurF was plated on TSA containing various IPTG concentrations (0, 2.5, 5.0, 7.5, 10, and 100 μM) and was tested for effect on oxacillin resistance. The sizes of the inhibition zones varied with the concentration of IPTG (Fig. 5A).
The conditional mutation of murF was also introduced into strain ZOX3, which is a spontaneous mutant derivative of the fully antibiotic-susceptible S. aureus strain 27S (15). ZOX3 has an increased level of resistance to ceftizoxime, a beta-lactam antibiotic with high selective affinity for S. aureus PBP2. Strain ZOX3 owes its resistance to ceftizoxime to a single point mutation in the pbpB structural gene, causing decreased affinity of PBP2 for ceftizoxime (15). ZOX3spacmurF was plated on TSA containing 2.5, 5.0, 7.5, 10, and 100 μM concentrations of IPTG and was assayed for ceftizoxime inhibition halos, using 30-μg disks. For IPTG concentrations of 10 and 100 μM, the size of inhibition halos was the same as that of the parental strain ZOX3. However, for lower IPTG concentrations (7.5, 5.0, and 2.5 μM), resistance decreased to the level of the susceptible parental strain 27S (Fig. 5B).
Changes in the composition of cell wall precursor pool. The cytoplasmic fraction of strains COL and COLspacmurF were isolated from cultures grown at several IPTG concentrations, and the cell wall precursor pool was purified and analyzed by HPLC.
The HPLC profiles showed that as the IPTG concentration decreased, an accumulation of UDP-linked muramyl tripeptide occurred in the cytoplasm, paralleled by the gradual reduction in the amount of UPD-linked muramyl pentapeptide (Fig. 6). The relative accumulation values for the UDP-MurNAc-tripeptide and UDP-MurNAc-pentapeptide are listed in Table 3.
Changes in the composition of cell wall peptidoglycan. The peptidoglycan composition of strain COLspacmurF was determined for bacteria grown in the presence of optimal (100 μM) and suboptimal IPTG concentrations. The HPLC profiles of enzymatic hydrolysates of these peptidoglycans are shown in Fig. 7. For cells grown in 100 μM IPTG-supplemented medium, the peptidoglycan HPLC profile was identical to that of strain COL.
At all suboptimal IPTG concentrations, a new, abnormal, monomeric muropeptide appeared in the HPLC profile. This component was identified by mass spectrometry as the disaccharide tripeptide lacking the pentaglycine side chain (27). The relative amounts of this muropeptide were smallest (5%) in cells grown at 10 μM IPTG but remained constant (7%) for all other suboptimal concentrations of IPTG (Table 4).
The peptidoglycan composition of ZOX3spacmurF was also determined for the same IPTG concentrations. For the suboptimal IPTG concentrations of 10, 7.5, 5.0, and 2.5 μM, the disaccharide tripeptide structure appeared in the cell wall, but the incorporation level was much less than in the COL background. For instance, in bacteria grown in 10 μM IPTG, the tripeptide represented 2% and, for the lower concentrations, about 3% of all muropeptides (data not shown).
Decrease in the cellular amounts of peptidoglycan. Cell walls purified from COLspacmurF grown at different concentrations of IPTG were dried and weighed, and identical (5-mg) amounts were treated with hydrofluoric acid to remove wall teichoic acids. After hydrofluoric acid treatment, the peptidoglycan preparations were dried and weighed. Cells grown in the presence of 100 μM IPTG produced peptidoglycan in excess of what was found in strain COL, while all bacteria grown at suboptimal inducer concentrations contained reduced amounts of peptidoglycan (Table 4).
Cell morphology and cell wall thickness. Mid-exponential-phase cells of strain COL and its murF conditional mutant grown with several IPTG concentrations were analyzed by electron microscopy (Fig. 8). For each inducer concentration tested, the average cell wall thickness value was calculated, based on measurement of >100 cross-sectioned cells. The cell wall thickness values for all mutants were expressed relative to that of the parental strain COL, which was set to 100% (Table 4). When grown in the presence of 100 μM IPTG, the mutant cells showed a thicker and less sharply defined cell wall, indicating that this inducer concentration did not exactly mimic the native expression of murF, despite the unaltered antibiotic resistance, growth rate, and HPLC profile. A cell wall thickness similar to the parental cell was obtained for bacteria grown in 10 μM IPTG. Mutants cultivated at suboptimal concentrations (7.5, 5.0, 2.5, and 0 μM) showed a 20 to 30% reduction in cell wall thickness.
In mutant cultures grown at or below 5.0 μM of IPTG, several morphological abnormalities appeared in most (>90%) of the bacteria. These abnormalities included asymmetrically distributed and thickened septae, which contrasted sharply with the thinner peripheral cell wall. Some cells also showed multiple septae and hollow cells (ghost cells) with multiple vesicles in the cytoplasm. Accumulation of cellular debris in the medium (lysis) was observed in cells cultivated without IPTG. An overall increase of the cell diameter was also apparent and the appearance of the chromosomal area was also abnormal (more condensed) in many cells (Fig. 8).
Altered transcription of mecA and pbpB genes. The expression of mecA and pbpB genes, which encode PBP2A and PBP2, respectively, was analyzed by Northern blotting. The hybridization performed with a pbpB internal probe showed two bands. In fact, pbpB gene can be transcribed either alone or together with the upstream gene recU (23), resulting in two independent mRNAs species, the pbpB transcript (2.1 kb) and the transcript from pbpB and recU (2.9 kb).
Figure 9 shows that the transcription of mecA and pbpB was reduced in COLspacmurF (Fig. 9A) grown at the suboptimal IPTG concentrations, compared to that for the parental strain COL. For IPTG concentrations of 100 μM and 10 μM, both transcript signal intensities were stronger than in the parental strain. In the murF conditional mutant derivatives of both 27S and ZOX3, the transcription of pbpB varied in parallel with the transcription of murF (Fig. 9B). As an internal control, the transcription of the housekeeping gene pta (expressing phosphate acetyltransferase) was also determined at the same IPTG concentrations. No significant differences were detected (Fig. 9C).
DISCUSSION
The MurF enzyme catalyzes a critical step in the biosynthesis of the bacterial cell wall: the addition of the C-terminal D-alanyl-D-alanine dipeptide to the cell wall precursor muropeptide. The construction of a conditional mutant of murF in the background of both methicillin-resistant S. aureus (MRSA) and methicillin-susceptible S. aureus (MSSA) strains has allowed us to explore the physiological functions of this gene in S. aureus in more detail than was possible from the study of a murF insertional mutant (27).
S. aureus strains carrying the spacmurF construct showed an absolute dependence on the IPTG inducer for growth, confirming the essential nature of the murF gene in Staphylococcus aureus. Omission of the IPTG inducer from the medium prevented growth in both MRSA and MSSA strains, and suboptimal concentrations of IPTG caused parallel decreases in growth rate and murF expression levels. At the inducer concentration of 100 μM IPTG, the growth rate and yield were indistinguishable from those of parental strains carrying the native promoters. The essentiality of the murF gene had already been reported in the gram-negative bacteria Escherichia coli through the study of temperature-sensitive mutants, which would lyse when grown at restrictive temperatures (16).
Open reading frame SACOL0054 was identified in staphylococcal cassette chromosome mec type I (11), present in strain COL. This ORF was identified as a member of the Mur ligase family of proteins by homology (29.6%) with the murF homologue from Mesorhizobium loti. It has 25.3% homology and 49.2% similarity to the COL murF gene. The essentiality of the S. aureus murF gene shown by the study of COL conditional mutant suggests that ORF SACOL0054 cannot replace the chromosomal murF. A frameshift mutation was described in ORF SACOL0054, which may be responsible for the synthesis of a nonfunctional protein.
An analysis of the extracts of the mutants growing in the presence of different IPTG concentrations demonstrated the sensitive dependence of the composition of the cell wall precursor pool on the expression of murF. In COLspacmurF growing in the presence of 100 μM IPTG, the relative amount of the UDP-MurNAc-pentapeptide in the cell wall precursor pool was 64.1%, virtually the same as in strain COL (64.8%) growing with its native murF promoter. There was no detectable tripeptide in the wall precursor pools of these bacteria. However, reduction of the concentration of IPTG in the growth medium caused striking changes in the composition of cell wall precursors. Decrease in the IPTG concentration from 100 to 10, 7.5, 5.0, and 2.5 μM caused a stepwise decrease in the amount of pentapeptides from 64.1 to 11.1% and a parallel increase in the abnormal tripeptide component from undetectable to 59.3% (Table 3). The progressive accumulation of tripeptide and concomitant shortage of pentapeptide precursors indicate that maintenance of normal levels of a catalytically active MurF depends on a steady expression of the murF gene.
Analysis of the peptidoglycan composition of bacteria grown at different IPTG concentrations demonstrated that the tripeptides were also able to incorporate into the polymerized cell wall. However, the maximum amounts of disaccharide tripeptides in the cell wall had an upper limit, from 6.7 to 7.1% of all muropeptides in the murF mutant of MRSA strain COL and up to 2 to 3% in the MSSA strains (Table 4). This observation was in sharp contrast to the more extensive variations in the composition of the cell wall precursor pool, in which the relative amounts of tripeptides varied through the entire range of IPTG concentrations used.
The abnormal disaccharide tripeptides lacking the C-terminal D-alanyl-D-alanine residues cannot participate as donors in the transpeptidation reaction, but they could, in principle, serve as acceptors. However, careful examination of the HPLC profiles showed no evidence for muropeptide oligomers of such structures. Thus, the disaccharide tripeptides appearing in the cell walls of the murF mutants were only present as monomers, which must have been incorporated into the peptidoglycan through the activity of transglycosylases such as the native PBP2 or one of the monofunctional transglycosylases present in S. aureus (6, 28).
The drastic alteration in the composition of the cell wall precursor pool in the murF mutants growing at suboptimal concentrations of IPTG presents a serious dilemma for cell wall biosynthesis. The utilization of the tripeptides appears to be limited, presumably because enrichment of the peptidoglycan in monomeric components above these threshold levels may jeopardize the structural integrity of the cell wall and is lethal for the bacteria. This suggests that the S. aureus cell requires a minimal level of peptidoglycan cross-linking, below which it is not able to grow and divide. On the other hand, the limited expression of murF also leads to a greatly diminished pool of the normal pentapeptide components. Under these conditions, the only option for the cells appears to be to produce less peptidoglycan. This was actually observed in the murF mutants growing at suboptimal IPTG concentrations: electron microscopic analysis showed a striking decrease in the thickness of peripheral cell wall, and direct determination of the yield of peptidoglycan isolated from such bacteria showed a gradual decrease in the cellular amounts of peptidoglycan, which represented about half (55%) of the dry weight of purified cell walls in the wild-type strain COL but dropped to about 34 to 36% in the mutant cells grown at suboptimal concentrations of IPTG (Table 4).
The 100 μM IPTG concentration was considered optimal for murF expression, since it allowed normal growth rate and maximum expression of beta-lactam resistance and normal composition of the cell wall precursor pool, as well as peptidoglycan. Nevertheless, to our surprise, we found that the cell walls of bacteria grown under these conditions showed profound alterations: there was significant thickening (by about 20%) of the cell wall and there was an increase in the cellular amounts of peptidoglycan as well (from 55% to 60%). These findings suggested that in bacteria growing with 100 μM IPTG, the expression of murF might be increased over the level of the parental strain using its native murF promoter. Comparison of murF transcription levels confirmed this (Fig. 3).
At 10 μM IPTG, the cell wall thickness and the peptidoglycan recovery yield were very similar to the parental values. However, the tripeptide precursor synthesis and its incorporation (5%) into the cell wall still occurred in very significant amounts. The fact that the oxacillin resistance level was visibly affected for this IPTG concentration suggests that the resistance decrease is not related to alterations in the cell wall thickness but may be related to changes in its composition and/or the composition of the cell wall precursor pool. A decrease in oxacillin resistance has frequently been observed, along with changes in the peptidoglycan composition of mutants of other cell wall-related genes, such as murE (20), glnR/A (10, 19), and femA and femB (13).
The availability of the murF conditional mutants has allowed us to produce S. aureus strains in which the expression of murF depended on the concentration of the IPTG inducer added to the medium. Most interestingly, ddlA, supplying the D-alanyl-D-alanine substrate for the MurF ligase, appears to be cotranscribed with murF. This and the profound changes observed in the conditional mutants in composition of cell wall precursors, cellular amounts of peptidoglycan, and the fine structure of the cell wall indicate that the addition of the C-terminal dipeptide to the cell wall precursor represents an important control point in the peptidoglycan synthesis of S. aureus. The greatly increased thickness of septal areas in the same cells in which the thickness of the peripheral cell wall was decreased, the substantial increase in cell diameters, and the other morphological abnormalities observed in the cells with the suboptimal murF function indicate that undisturbed functioning of the ddlA-murF system is essential for normal cell division in these bacteria.
The transcriptional analysis by Northern blotting showed that mecA (PBP2A) and pbpB (PBP2) transcription is also increased in the murF conditional mutants at IPTG concentrations of 10 and 100 μM, a fact that may explain the mechanism of overproduction of peptidoglycan. Altered transcription of mecA and pbpB genes has already been observed with a murF insertion mutant of strain COL (27) and was also noted with a murE conditional mutant (8). The presence of abnormal precursors may play a role in the altered transcription rates but cannot alone account for the regulation of these transcripts, since both mecA and pbpB transcripts are overexpressed in the presence of both 100 μM and 10 μM IPTG.
Regarding beta-lactam resistance, the spacmurF construct was analyzed in two different backgrounds: COL, a MRSA strain with constitutive expression of the mecA gene, and ZOX3, a MSSA strain which lacks the mecA gene and which is resistant to the beta-lactam cefizoxime. It was shown that in the absence of murF, both oxacillin (in COL) and ceftizoxime (in ZOX3) resistance decreases. These results suggest that murF affects beta-lactam resistance even in the absence of the mecA gene.
The transcription of pbpB was found to parallel the decrease of murF, not only in COL but also in the MSSA strains 27S and ZOX3. Thus, the postulated coregulatory mechanism between murF and pbpB transcription must be independent of mecA.
The level of incorporation of the abnormal disaccharide tripeptide was different in the background of the MRSA strain COL (7%) compared to that in the MSSA strain ZOX3 (2 to 3%). We propose that these differences may be related to a recent observation concerning the cooperative functioning of PBP2 and PBP2A. It has been shown that in MSSA strains, PBP2 normally localizes to the division septum, where cell wall synthesis is known to take place in S. aureus (22, 24). Pinho and colleagues have also shown that in the presence of modified precursors, specifically in the presence of tripeptide precursors accumulating in cells exposed to D-cycloserine, PBP2 loses the capacity to correctly localize at the septum and delocalizes along the peripheral cell wall. However, in an MRSA strain a functional PBP2A is able to maintain the correct localization of PBP2 at the septum, in spite of the presence of the inhibitor (25). The higher level of incorporation of the tripeptide into the peptidoglycan of the MRSA strain COL compared to that of the MSSA strain ZOX3 may then be explained by this putative "helper" function of PBP2A, present only in the antibiotic-resistant strain COL.
ACKNOWLEDGMENTS
Partial support for this study was provided by a grant (2 RO1 A1045738-06) from the National Institutes of Health, the U.S. Public Health Service, and contract POCTI/BIA-MIC/58416/2004 (Fundao para a Ciência e a Tecnologia, Portugal). R.G.S. was supported by grants SFRH/BD/3138/2000 and 022/BI/2005 (Fundao para a Ciência e a Tecnologia, Portugal).
REFERENCES
Barna, J. C., and D. H. Williams. 1984. The structure and mode of action of glycopeptide antibiotics of the vancomycin group. Annu. Rev. Microbiol. 38:339-357.
Beauregard, D. A., A. J. Maguire, D. H. Williams, and P. E. Reynolds. 1997. Semiquantitation of cooperativity in binding of vancomycin-group antibiotics to vancomycin-susceptible and -resistant organisms. Antimicrob. Agents Chemother. 41:2418-2423.
de Jonge, B. L., Y. S. Chang, D. Gage, and A. Tomasz. 1992. Peptidoglycan composition of a highly methicillin-resistant Staphylococcus aureus strain. The role of penicillin binding protein 2A. J. Biol. Chem. 267:11248-11254.
de Lencastre, H., and A. Tomasz. 1994. Reassessment of the number of auxiliary genes essential for expression of high-level methicillin resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 38:2590-2598.
de Lencastre, H., S. W. Wu, M. G. Pinho, A. M. Ludovice, S. Filipe, S. Gardete, R. Sobral, S. Gill, M. Chung, and A. Tomasz. 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.
Di Berardino, M., A. Dijkstra, D. Stuber, W. Keck, and M. Gubler. 1996. The monofunctional glycosyltransferase of Escherichia coli is a member of a new class of peptidoglycan-synthesising enzymes. FEBS Lett. 392:184-188.
Enright, M. C., N. P. Day, C. E. Davies, S. J. Peacock, and B. G. Spratt. 2000. Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. J. Clin. Microbiol. 38:1008-1015.
Gardete, S., A. M. Ludovice, R. G. Sobral, S. R. Filipe, H. de Lencastre, and A. Tomasz. 2004. Role of murE in the expression of -lactam antibiotic resistance in Staphylococcus aureus. J. Bacteriol. 186:1705-1713.
Gu, Y. G., A. S. Florjancic, R. F. Clark, T. Zhang, C. S. Cooper, D. D. Anderson, C. G. Lerner, J. O. McCall, Y. Cai, C. L. Black-Schaefer, G. F. Stamper, P. J. Hajduk, and B. A. Beutel. 2004. Structure-activity relationships of novel potent MurF inhibitors. Bioorg. Med. Chem. Lett. 14:267-270.
Gustafson, J., A. Strassle, H. Hachler, F. H. Kayser, and B. Berger-Bachi. 1994. The femC locus of Staphylococcus aureus required for methicillin resistance includes the glutamine synthetase operon. J. Bacteriol. 176:1460-1467.
Ito, T., Y. Katayama, K. Asada, N. Mori, K. Tsutsumimoto, C. Tiensasitorn, and K. Hiramatsu. 2001. Structural comparison of three types of staphylococcal cassette chromosome mec integrated in the chromosome in methicillin-resistant Staphylococcus aureus. Antimicrob. Agents Chemother. 45:1323-1336.
Jana, M., T. T. Luong, H. Komatsuzawa, M. Shigeta, and C. Y. Lee. 2000. A method for demonstrating gene essentiality in Staphylococcus aureus. Plasmid 44:100-104.
Kopp, U., M. Roos, J. Wecke, and H. Labischinski. 1996. Staphylococcal peptidoglycan interpeptide bridge biosynthesis: a novel antistaphylococcal target Microb. Drug Resist. 2:29-41.
Kraemer, G. R., and J. J. Iandolo. 1990. High-frequency transformation of Staphylococcus aureus by electroporation. Curr. Microbiol. 21:373-376.
Leski, T. A., and A. Tomasz. 2005. Role of penicillin-binding protein 2 (PBP2) in the antibiotic susceptibility and cell wall cross-linking of Staphylococcus aureus: evidence for the cooperative functioning of PBP2, PBP4, and PBP2A. J. Bacteriol. 187:1815-1824.
Lugtenberg, E. J. J., and A. Van Schijndel-van Dam. 1972. Temperature-sensitive mutants of Escherichia coli K-12 with low activities of the L-alanine adding enzyme and the D-alanyl-D-alanine adding enzyme. J. Bacteriol. 110:35-40.
Maniatis, V., E. F. Fritsch, and J. Sambrook. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Miragaia, M., I. Couto, and H. de Lencastre. 2005. Genetic diversity among methicillin-resistant Staphylococcus epidermidis (MRSE). Microb. Drug Resist. 11:83-93.
Ornelas-Soares, A., H. de Lencastre, B. de Jonge, D. Gage, Y. S. Chang, and A. Tomasz. 1993. The peptidoglycan composition of a Staphylococcus aureus mutant selected for reduced methicillin resistance. J. Biol. Chem. 268:26268-26272.
Ornelas-Soares, A., H. de Lencastre, B. L. de Jonge, and A. Tomasz. 1994. Reduced methicillin resistance in a new Staphylococcus aureus transposon mutant that incorporates muramyl dipeptides into the cell wall peptidoglycan. J. Biol. Chem. 269:27246-27250.
Oshida, T., and A. Tomasz. 1992. Isolation and characterization of a Tn551-autolysis mutant of Staphylococcus aureus. J. Bacteriol. 174:4952-4959.
Pinho, M. G., H. de Lencastre, and A. Tomasz. 2001. An acquired and a native penicillin-binding protein cooperate in building the cell wall of drug-resistant staphylococci. Proc. Natl. Acad. Sci. USA 98:10886-10891.
Pinho, M. G., H. de Lencastre, and A. Tomasz. 1998. Transcriptional analysis of the Staphylococcus aureus penicillin binding protein 2 gene. J. Bacteriol. 180:6077-6081.
Pinho, M. G., and J. Errington. 2003. Dispersed mode of Staphylococcus aureus cell wall synthesis in the absence of the division machinery. Mol. Microbiol. 50:871-881.
Pinho, M. G., and J. Errington. 2005. Recruitment of penicillin-binding protein PBP2 to the division site of Staphylococcus aureus is dependent on its transpeptidation substrates. Mol. Microbiol. 55:799-807.
Pinho, M. G., S. R. Filipe, H. de Lencastre, and A. Tomasz. 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.
Sobral, R. G., A. M. Ludovice, S. Gardete, K. Tabei, H. de Lencastre, and A. Tomasz. 2003. Normally functioning murF is essential for the optimal expression of methicillin resistance in Staphylococcus aureus. Microb. Drug Resist. 9:231-241.
Wang, Q. M., R. B. Peery, R. B. Johnson, W. E. Alborn, W. K. Yeh, and P. L. Skatrud. 2001. Identification and characterization of a monofunctional glycosyltransferase from Staphylococcus aureus. J. Bacteriol. 183:4779-4785.(R. G. Sobral, A. M. Ludov)