UDP-N-acetylmuramoyl-L-alanyl-D-glutamate:meso-diaminopimelate ligase is a cytoplasmic enzyme that catalyzes the addition of meso-diaminopimelic acid to nucleotide precursor UDP-N-acetylmuramoyl-L-alanyl-D-glutamate in the biosynthesis of bacterial cell-wall peptidoglycan. The crystal structure of the Escherichia coli enzyme in the presence of the final product of the enzymatic reaction, UDP-MurNAc-L-Ala-␥-D-Glu-meso-A 2 pm, has been solved to 2.0 Å resolution. Phase information was obtained by multiwavelength anomalous dispersion using the K shell edge of selenium. The protein consists of three domains, two of which have a topology reminiscent of the equivalent domain found in the already established three-dimensional structure of the UDP-N-acetylmuramoyl-L-alanine: D-glutamateligase (MurD) ligase, which catalyzes the immediate previous step of incorporation of D-glutamic acid in the biosynthesis of the peptidoglycan precursor. The refined model reveals the binding site for UDP-MurNAc-LAla-␥-D-Glu-meso-A 2 pm, and comparison with the six known MurD structures allowed the identification of residues involved in the enzymatic mechanism. Interestingly, during refinement, an excess of electron density was observed, leading to the conclusion that, as in MurD, a carbamylated lysine residue is present in the active site. In addition, the structural determinant responsible for the selection of the amino acid to be added to the nucleotide precursor was identified.Peptidoglycan, the polymeric mesh of the bacterial cell wall, plays a critical role in protecting bacteria against osmotic lysis. It consists of linear repeating disaccharide chains cross-linked by short peptide bridges. During the cytoplasmic steps involved in the biosynthesis of the peptidoglycan precursor, four ADPforming ligases (namely the Mur ligases) catalyze the assembly of the peptide moiety by the successive addition of L-alanine, D-glutamate, diaminopimelic acid, or L-lysine, and, finally, dipeptide D-alanyl-D-alanine to UDP-N-acetylmuramic acid (1, 2). Because all these enzymes are essential for cell viability, they are attractive targets for antibacterial chemotherapy. In Escherichia coli, these ligases are the products of the murC, murD, murE, and murF genes, located in the mra region (3). Sequence comparison of the four E. coli Mur ligases shows several homologous regions, suggesting that these enzymes may be evolutionarily related and may use similar enzymatic mechanisms (4 -6). In earlier publications, we reported the structure of UDP-N-acetylmuramoyl-L-alanine:D-glutamate ligase (MurD), 1 both in the native form and complexed with substrates (7-9). MurD consists of three domains with topologies reminiscent of a nucleotide-binding fold; the N-and Cterminal domains have a dinucleotide-binding fold (the Rossmann fold), and the central domain displays a mononucleotidebinding fold, also seen in ATP-binding proteins. A comparison of six MurD structures reveals that large C-terminal rotation, loop rearrangement, and subdomain movements occur upon subs...

The cellular pool levels of most of the cytoplasmic precursors of peptidoglycan synthesis were determined for normally growing cells of Escherichia coli K-12. In particular, a convenient method for analyzing the uridine nucleotide precursor contents was developed by associating gel filtration and reverse-phase high-pressure liquid chromatography techniques. The enzymatic parameters of the four synthetases which catalyze the stepwise addition of L-alanine, D-glutamic acid, meso-diaminopimelic acid, and D-alanyl-D-alanine to uridine diphosphate-N-acetylmuramic acid were determined. It was noteworthy that the pool levels of L-alanine, D-glutamic acid, meso-diaminopimelic acid, and D-alanyl-D-alanine were much higher than the Km values determined for these substrates, whereas the molar concentrations of the uridine nucleotide precursors were lower than or about the same order of magnitude as the corresponding Km values. Taking into consideration the data obtained, an attempt was made to compare the in vitro activities of the D-glutamic acid, meso-diaminopimelic acid, and D-alanyl-D-alanine adding enzymes with their in vivo functioning, expressed by the amounts of peptidoglycan synthesized. The results also suggested that these adding activities were not in excess in the cell under normal growth conditions, but their amounts appeared adjusted to the requirements of peptidoglycan synthesis. Under the different in vitro conditions considered, only low levels of L-alanine adding activity were observed.

The 2-min region of the Escherichia coli genome contains a large cluster of genes from pbpB to envA that code for proteins involved in peptidoglycan biosynthesis and cell division. From pLC26-6 of the collection of Clarke and Carbon (L. Clarke and J. Carbon, Cell 9:91-99, 1976) plasmids carrying different fragments from the 8-kilobase-pair region downstream of pbpB were constructed and analyzed for their ability to direct protein synthesis in maxicells, to complement various thermosensitive mutations, and to overproduce enzymatic activities. We report the localization of the previously unidentified murD gene coding for the D-glutamic acid-adding enzyme within this region. Our data show that the genes are in the order pbpB-murEmurF-X-murD-Y-murG, where X and Y represent chromosomal fragments from 1 to 1.5 kilobase pairs, possibly coding for unknown proteins. Furthermore, the murE and murF genes, encoding the meso-diaminopimelic acid and D-alanyl-D-alanine-adding enzymes, respectively, may be translationally coupled when transcription is initiated upstream of murE, within the preceding structural gene pbpB coding for penicillin-binding protein 3.