PerspectiveMicrobiology

Deconstructing Vancomycin

See allHide authors and affiliations

Science  16 Apr 1999:
Vol. 284, Issue 5413, pp. 442-443
DOI: 10.1126/science.284.5413.442

Vancomycin is an antibiotic that occupies an important niche in the treatment of life-threatening infections caused by Gram-positive bacteria such as Staphylococcus aureus and Enterococcus faecalis. This glycopeptide is crucial for treating infections in, for example, cancer patients undergoing chemotherapy and renal patients on dialysis. The widespread prevalence of methicillin-resistant S.aureus has made vancomycin the antibiotic of last resort for eliminating multi-drug-resistant Gram-positive bacteria. But recent reports of patients infected with methicillin-resistant S.aureus that also proved to be resistant to vancomycin (1, 2) raise the specter of the worst kind of antibiotic-resistant superbug (3). The arrival of vancomycin-resistant S. aureus follows the emergence of vancomycin-resistant enterococcus, which has plagued hospitals for the last 10 years.

But all may not be lost in the fight against multi-drug-resistant superbugs, at least not according to the report by Ge and colleagues on page 507 of this issue (4). By modifying the sugar groups attached to vancomycin's peptide backbone, these investigators synthesized analogs that were not only more efficient than the parent compound at dispatching vancomycin-resistant bacteria but were also better at killing vancomycin-sensitive organisms. The authors showed that the vancomycin analogs blocked an earlier step (transglycosylation) in bacterial cell wall synthesis than did vancomycin, and that the modified sugar groups themselves had substantial antibacterial activity. These results should redirect attention to designing new, sugar-based antibiotics that inhibit the transglycosylase step in bacterial cell wall synthesis.

Vancomycin interdicts bacterial growth primarily by blocking the cross-linking of adjacent peptidoglycan strands by peptide bonds during synthesis of the bacterial cell wall (see the figure, top). Without sufficient cross-linking, the cell wall becomes mechanically fragile and the bacteria lyse when subjected to changes in osmotic pressure. Vancomycin binds to the D-alanine-D-alanine (D-Ala-D-Ala) terminus of the pentapeptide portion of the peptidoglycan precursor before cross-linking. The D-Ala-D-Ala dipeptide forms complementary hydrogen bonds with the peptide backbone of vancomycin. It is thought that the vancomycin-peptidoglycan complex physically occludes the subsequent action of transpeptidase enzymes (the targets of penicillins) and in so doing blocks formation of the peptide cross-bridges that confer strength on the peptidoglycan (see the figure, bottom). As a secondary event, the transglycosylation step that connects the disaccharide unit of the pentapeptide precursor to existing glycan strands also appears to be inhibited by vancomycin, although to a lesser extent. The two most prevalent clinical isolates of vancomycin-resistant enterococci, Van A and Van B, reprogram the peptidoglycan synthetic machinery, replacing the D-Ala-D-Ala dipeptide with D-Ala-D-lactate. The loss of a crucial hydrogen bond between vancomycin and the terminal dipeptide results in a decrease (by three orders of magnitude) in the binding affinity of the antibiotic for the peptidoglycan (5, 6).

Hitting bacteria where it hurts.

The antibiotic vancomycin (yellow and red) blocks synthesis of peptide cross-links between peptidoglycan strands, resulting in a weakening of the bacterial cell wall. Vancomycin primarily blocks the transpeptidation step of peptide cross-link formation by binding to the D-Ala-D-Ala terminus of peptidoglycan precursor molecules (which are composed of pentapeptide-disaccharides anchored by lipid to the cytoplasmic membrane). Analogs in which the two sugar groups (red) of vancomycin have been modified block transglycosylation, the first step in peptidoglycan synthesis. Vancomycin analogs and the free modified disaccharides alone are more effective than the parent compound against vancomycin-resistant bacteria.

Much interest surrounds methods to modify vancomycin to combat resistance, but the heptapeptide backbone of vancomycin has a rigid cup-shaped architecture that prevents it from binding to the D-Ala-D-lactate dipeptide in the peptidoglycans of resistant bacteria. However, modifying the disaccharide sugar of vancomycin rather than the peptide backbone proved to be a novel approach to making effective analogs. The synthesis of the vancomycin peptide backbone was recently accomplished by the research groups of Evans (7) and Nicolaou (8). Kahne's team then succeeded in attaching the disaccharide vancosamine-glucose to the peptide backbone, thus completing the synthesis of the vancomycin molecule (9). It proved possible to modify the free amino group in the vancosamine sugar moiety with hydrophobic substituents, such as biphenyls, and to show that this modification increased activity against vancomycin-resistant enterococci (10). Now Ge et al. have gone further, using their expertise in carbohydrate chemistry to obtain unanticipated findings about vancomycin analogs.

Their first intriguing discovery was the retention of substantial antibacterial activity in a chlorobiphenyl vancomycin derivative that was missing the first (leucyl) residue. This meant that recognition of the D-Ala-D-Ala terminus by the analog was abrogated, ruling out the conventional high-affinity interaction between the antibiotic and the peptidoglycan as the mode of drug action. The investigators further pruned the vancomycin skeleton down to the modified vancosamine-glucose disaccharide and found, to their amazement, that the disaccharide alone retained powerful antibiotic activity. They used permeabilized Escherichia coli bacteria to determine which step in the peptidoglycan synthesis pathway was blocked by the disaccharide. In contrast to vancomycin, which primarily abrogates transpeptidation, the vancomycin analog and its disaccharide fragment alone selectively blocked the transglycosylation step of peptidoglycan synthesis.

It is thought that vancomycin binds to D-Ala-D-Ala termini in the non-cross-linked mature peptidoglycan, and also in the lipid-disaccharide pentapeptide precursor molecules that are substrates for incorporation into expanding peptidoglycan chains. It now appears that the vancomycin analog and its disaccharide fragment directly interact with one or more of the transglycosylases involved in oligomerization of the glycan strands. The discovery that these enzymes are targets for the modified sugars of vancomycin reveals simple strategies for defining which transglycosylases are involved in bacterial cell wall synthesis and for designing a new class of antibiotics by manipulation of vancosamine and glucose rings. To what extent the dimethoxyphenyl ring component (mimicking residue 4 of vancomycin) of the disaccharide fragment is crucial for anchoring the disaccharide in the bacterial cell membrane and for the fragment's antibacterial activity will also need to be explored.

All in all, the deconstructionist approach to vancomycin appears to have struck gold with the discovery of a disaccharide fragment of a vancomycin analog that is more powerful than vancomycin itself. These results may accelerate the discovery and development of simple sugar-based fragments that are equally adept at killing vancomycin-sensitive and vancomycin-resistant pathogenic bacteria by targeting not the transpeptidation step but the transglycosylation step of peptidoglycan synthesis.

References

View Abstract

Navigate This Article