For 70 years antibiotics have saved countless lives and enabled the development of modern medicine, but it is becoming clear that the success of antibiotics may have only been temporary and we now anticipate a long-term, generational and perhaps never-ending challenge to find new therapies to combat antibiotic-resistant bacteria. As the search for new conventional antibiotics has become less productive and there are no clear strategies to improve success, a broader approach to address bacterial infection is needed. This review of potential alternatives to antibiotics (A2As) was commissioned by the Wellcome Trust, jointly funded by the Department of Health, and involved scientists and physicians from academia and industry. For the purpose of this review, A2As were defined as non-compound approaches (that is, products other than classical antibacterial agents) that target bacteria or approaches that target the host. In addition, the review was limited to agents that had potential to be administered orally, by inhalation or by injection for treatment of systemic/invasive infection. Within these criteria, the review has identified 19 A2A approaches now being actively progressed. The feasibility and potential clinical impact of each approach was considered. The most advanced approaches (and the only ones likely to deliver new treatments by 2025) are antibodies, probiotics, and vaccines now in Phase II and Phase III trials. These new agents will target infections caused by P. aeruginosa, C. difficile and S. aureus. However, other than probiotics for C. difficile, this first wave will likely best serve as adjunctive or preventive therapies. This suggests that conventional antibiotics will still be needed. The economics of pathogen-specific therapies must improve to encourage innovation, and greater investment into A2As with broad-spectrum activity (e.g. antimicrobial-, host defense-and, anti-biofilm peptides) is needed. Increased funding, estimated at >£1.5 bn over 10 years is required to validate and then develop these A2As. Investment needs to be partnered with translational expertise and targeted to support the validation of these approaches at Clinical Phase II proof of concept. Such an approach could transform our understanding of A2As as effective new therapies and should provide the catalyst required for both active engagement and investment by the pharma/biotech industry. Only a sustained, concerted and coordinated international effort will provide the solutions needed for the next decade.

Antibiotics of the vancomycin group are shown to enhance their affinities for the bacterial cell wall by the devices of either dimerization (vancomycin and other glycopeptides which dimerize even more strongly) or use of a membrane anchor (teicoplanin); a chelate mechanism is suggested in both cases, as supported by antagonism experiments with the cell wall analog di-N-acetyl-L-Lys-D-Ala-D-Ala. These results may have implications for other binding processes which occur near membrane surfaces.Recent reviews of structure-activity relationships for antibiotics of the vancomycin group (10) have not considered dimerization of the antibiotics, which we now believe plays an important role in the mode of action (5,8,9,16). Here, we present results consistent with a mechanism by which dimeric (e.g., vancomycin [Fig. 1A]) and lipoylated (e.g., teicoplanin [ Fig. 1C]) glycopeptides enhance the binding affinities to their cellular targets by preferential location of the antibiotic near the site of cell wall biosynthesis.The antibacterial activities of vancomycin antibiotics result from their affinities for the Opeptidyl-D-Ala-D-Ala sequence present in the growing cell wall of gram-positive bacteria (11,14,18); bound antibiotic inhibits the transglycosylase and transpeptidase enzymes responsible for construction of the cell wall (13). While the antibiotic activity can generally be correlated with the association constant for the cell wall, as measured with the cell wall analog di-N-acetyl-L-Lys-D-Ala-D-Ala (DALAA), antibiotics which dimerize strongly show anomalously high in vitro activities (4, 6, 9).Back-to-back homodimers (Fig. 2) are formed by most antibiotics of the vancomycin group (with the exception of teicoplanin [see below]), as shown by nuclear magnetic resonance studies (4,8,16). Dimerization is promoted by structural epitopes which appear as additions to the basic heptapeptide motif, including chlorine atoms and sugars at two sites (8). Furthermore, dimerization increases the affinity for cell wall analogs by a factor of 1 to 10, as shown in nuclear magnetic resonance experiments (9); conversely, cell wall fragments enhance dimerization by factors of 2 to 100 (9).This evidence suggests that dimerization may have physiological significance in the action of antibiotics of the vancomycin group. This was investigated by determining the effects of the cell wall analog DALAA on the potencies of glycopeptide antibiotics exhibiting a range of dimerization constants by an agar diffusion assay by the methods of Rake et al. (12). Glycopeptides (1 g) and DALAA (1 to 200 g) were added as aqueous solutions to paper disks (6-mm diameter; Whatman AA), which were dried before being placed on the surfaces of agar plates (1-mm agar thickness); antibiotic medium 1 (Difco Laboratories, Detroit,) was inoculated with Bacillus subtilis ATCC 6633 (30-l spore suspension/10 ml of agar Difco Laboratories, Detroit, Mich.), which was used as the indicator organism. After incubation at 37ЊC for 18 h, inhibition zone diameters were measured. D...

Clavulanic acid, sulbactam, and tazobactam are inhibitors of a variety of plasmid-mediated Il-lactamases. However, inhibition data for these three inhibitors with a wide range of different plasmid-mediated P-lactamases have not yet been compared under the same experimental conditions. A number of groups have inferred that clavulanic acid inhibits extended-spectrum TEM and SHV P-lactamases, but inhibition data have rarely been published. In this study, the 50%o inhibitory concentrations of these three ,-lactamase inhibitors for 35 plasmid-mediated P-lactamases have been determined. Of these 35 P-lactamases, 20 were extendedspectrum TEM-or SHV-derived I-lactamases. The other 15 enzymes were conventional-spectrum P-lactamases such as TEM-1 and SHV-1. Clavulanic acid was a more potent inhibitor than sulbactam for 32 of the 35 plasmid-mediated I-lactamases tested. In particular, clavulanic acid was 60 and 580 times more potent than sulbactam against TEM-1 and SHV-1, respectively, currently the two most clinically prevalent gram-negative plasmid-mediated I-lactamases. Statistical analysis of the data of the 50% inhibitory concentrations showed that clavulanic acid was 20 times more active overall than sulbactam against the conventional-spectrum enzymes. In addition, clavulanic acid was 14 times more potent than sulbactam at inhibiting the extendedspectrum enzymes. Tazobactam also showed significantly greater activity than sulbactam against the two groups of P-lactamases. There were no significant differences between the overall activities of tazobactam and clavulanic acid against the extended-spectrum TEM and SHV enzymes and conventional-spectrum enzymes, although differences in their inhibition profiles were observed.I-lactamases are plasmid-encoded or chromosomally encoded bacterial enzymes which hydrolyze 3-lactam antibiotics.Plasmid-mediated ,-lactamases can transfer rapidly between bacterial genera and consequently pose a major threat to the successful use of ,-lactam agents. More than 60 different types of plasmid-encoded ,-lactamases have been characterized, and for the purpose of this work, the enzymes have been classified as either conventional-spectrum or TEM-and SHVderived extended-spectrum enzymes. The conventional-spectrum ,-lactamases are from Bush groups 2a, 2b, 2c, and 2d, and extended-spectrum TEM and SHV enzymes are all from group 2b' (4).The conventional-spectrum enzymes include enzymes such as TEM-1 and SHV-1, which do not confer resistance to cephalosporins such as ceftazidime and cefotaxime. A recent survey of 802 gram-negative clinical isolates showed that TEM-1 and SHV-1 were responsible for mediating P-lactam resistance in 17% of clinical isolates (37). Other conventionalspectrum enzymes which are often found in clinical isolates include the OXA and PSE P-lactamases (1). The most common conventional-spectrum plasmid-mediated 13-lactamase found in gram-positive bacteria is the penicillinase produced by the majority of Staphylococcus aureus clinical isolates (26).Most of the extended-spectrum pl...

A new technique, the double strip method, for studying the chemotaxis of myxomycete plasmodia is described. Physarum polycephalum was attracted by the aldohexoses D-glucose, D-galactose and D-mannose and their derivatives 2-deoxy-~-glucose and maltose, thresholds ranging from 0.25 mM (D-glucose) to 5 mM (D-mannose). These sugars competed with each other, a uniform background of one of them inhibiting taxis to the others. Other attractants were N-acetyl-D-glucosamine and mannitol, with thresholds at 1 mM, and fucose (6-deoxy-~-galactose). Although in general only those carbohydrates which could support growth were attractants, there were exceptions such as 2-deoxy-~-glucose ; hence metabolism of a compound was not necessary for attraction. In addition, some compounds, such as fructose, could be metabolized but did not attract. At high concentrations (about 100 mM) all the compounds tested, including attractants, could under appropriate conditions cause repulsion, probably through osmotic effects. I N T R O D U C T I O NCarlile (1970) found that plasmodia of the myxomycete Physarum polycephalum Schweinitz were attracted by (i.e. showed positive chemotaxis to) glucose, maltose, mannose and galactose but not by sucrose, fructose or ribose. Each of the former group of sugars supported growth in shaken liquid culture when present as the sole carbohydrate, but the latter group did not. I t was tentatively concluded that with P. polycephalurn there was a parallel between the chemotactic effectiveness of sugars and their ability to support growth .However, studies on chemotaxis with Escherichiu coli have established that although most nutrients are attractants, some compounds are metabolized but do not attract, and others are not metabolized but are strong attractants (Adler, 1975). This conclusion suggests that the chemotactic responses of P. polycephalum to a wider range of sugars should be examined, and compared with the ability of the sugars to support growth and influence migration. Studies on growth and migration were reported in the preceding paper (Knowles & Carlile, 1978); studies on chemotaxis are described here.

No abstract