Carbohydrate Chain Terminators: Rational Design of Novel Carbohydrate-Based Antifungal Agents

As bacterial, viral and fungal drug resistance to currently administered treatments increase, the need for the development of new therapeutic strategies against infective agents becomes more urgent. Crucial to the survival of many pathogens are carbohydrate structures, which are either themselves structurally unique or specific to nonmammalian organisms. Indeed, inhibition or interference with correct biosynthesis of oligosaccharide materials represents an attractive and potentially highly selective strategy for the development of new classes of therapeutic agents. Over recent years substantial effort has been expended in the search for inhibitors of particular glycosyl transferase and glycosidase enzymes as a means of disrupting specific biosynthetic pathways. The design of such inhibitors can be problematic, since in many cases little is known about the precise enzymes involved. However, an alternative strategy, and one that does not rely on precise structural information about specific enzymes, in particular biochemical pathways, is to exploit structural analysis of the key targeted oligosaccharide structure. This strategy is to invoke chain termination of oligosaccharide biosynthesis. Chain-termination of oligonucleotide synthesis was originally developed by Sanger et al. as a means of sequencing DNA. Subsequently, chain termination strategies have found routine and widespread use in interference with oligonucleotide synthesis, and, moreover, have become the molecular basis of antiviral therapies in clinical use, perhaps most preeminently in the case of AZT. Although chain-termination processes have been implicated in the biological affects of some monosaccharide derivatives on mammalian glycoconjugate and glycosoaminoglycan biosynthesis, it is curious that a chain termination approach has not been widely promulgated as a strategy for the development of new classes of inhibitors of pathogenic oligosaccharide biosynthesis. Although there could be potential pitfalls in this approach, it is our considered opinion that this research route merits ACHTUNGTRENNUNGfurther investigation. For example, in cases in which multiplerepeat units of carbohydrate structures are essential for pathogenic survival (e.g. , in cell-wall formation), then the statistical chances of incorporations that lead to chain termination, would be more favourable. Moreover, there is now good evidence that glycosyl transferases do process activated donor substrates that are modified in a minimal way at a single hydroxyl group. Although certain of these compounds have been shown to act as enzyme inhibitors, many are good substrates that are readily processed. Indeed, as alluded to above, there is also literature precedent that chain-terminating modified carbohydrates have already been incorporated into mammalian glycoprotein oligosaccharides and glycosaminoglycans, with the net results of inhibition of their biosynthesis. As part of a program aimed at investigating the potential opportunities that chain termination of oligosaccharide biosynthesis offers to disease control, we became interested in the rational design of novel antifungal agents. Fungal infections represent a serious hazard to human and animal health, and—just as for bacterial infection—drug resistance to current therapies is increasing. The fungal cell wall consists of large sections of oligosaccharide materials including chitin, a polymer of bACHTUNGTRENNUNG(1–4)N-acetylglucosamine (GlcNAc), and also bglucan. The biosynthesis of both of these nonmammalian oligosaccharides could be potentially targeted by using a chain termination approach. The present study focuses on attempts to inhibit chitin biosynthesis. Chitin is assembled stepwise by the enzyme chitin synthase, which transfers single GlcNAc residues to a growing oligomeric chain; the donor substrate for the enzyme is UDP–GlcNAc. Potential chain terminators of this process are therefore GlcNAc residues in which the 4-hydroxyl has been modified. If such materials are processed by chitin synthase, then their transfer to the terminus of the growing chitin chain will result in a chain-termination step, since the required 4-hydroxyl at which subsequent units would be added will then be absent. Before embarking on the synthesis and testing of potential chain terminators two factors were considered. Firstly, the stability and polarity profile of potential drug compounds was borne in mind. Secondly, as it is a requirement that the potential chain terminator must be actively processed by the chitin synthase, thought was given as to what types of activated glycosyl donor should be synthesised and tested. The most obvious approach was that of synthesising and testing the modified UDP donors themselves, since these are the actual substrates processed by chitin synthase. Alternatively, a potential prodrug molecule could be synthesised, which might then be converted to the active UDP donor once inside the cell. Such a prodrug would have the added advantages of thermal stability and more appropriate polarity as compared to the UDP derivative. In line with previous literature precedent, it was reasoned that perhaps esters of GlcNAc derivatives could act as suitable prodrugs that might be able to penetrate into the cell without relying on carbohydrate-transport mechanisms. Therefore, peresters of GlcNAc derivatives in which the [a] Dr. R. Danac, Dr. T. Muller, Dr. A. J. Fairbanks Chemistry Research Laboratory, Oxford University Mansfield Road, Oxford, OX1 3TA (UK) Fax: (+44)1865-275674 E-mail : antony.fairbanks@chem.ox.ac.uk [b] L. Ball, Prof. S. J. Gurr Department of Plant Sciences, Oxford University South Parks Road, Oxford, OX1 3RB (UK) Supporting information for this article is available on the WWW under http://www.chembiochem.org or from the author.