Hardened by natural selection in the face of decades of overuse of antibiotics, some strains of bacteria are now resistant to most of the standard antibiotic agents used to combat infection. In 1992 alone, over 13,000 people in the USA died in hospitals as a result of infections caused by resistant bacteria.
Now researchers seeking new, more potent antibiotics are hoping to capitalize on nature's own inventiveness. As a team of US chemists reports in the
Many organisms generate their own antibiotics as a defence against infection. Natural antibiotics have a wide range of molecular structures. 'Macrolides' or 'polyketides', which include the drug erythromycin, contain large rings of carbon and oxygen atoms. Like penicillin, erythromycin is produced by a mould; it is used to treat a variety of infections including bronchitis. Like any other natural antibiotic, it is made by the action of biological catalysts, or 'enzymes'.
Polyketide antibiotics are unusual in that a single enzyme, a 'polyketide synthase', conducts several successive steps in the synthetic pathway -- normally a different enzyme facilitates each step. Polyketide synthase consists of several modules, each overseeing a different step. It puts together a polyketide antibiotic as a chain, then links the chain end to end, making a ring. Each of the enzyme's modules adds a few atoms to the chain.
Now Chaitan Khosla of Stanford University in California and co-workers are redesigning these molecular manufacturers in the hope of constructing a broad range of new and potent polyketide antibiotics. By altering the products of the enzyme's multistep synthesis, they aim to generate libraries of polyketide molecules, all differing slightly in structure, which could then be tested to see whether some act as antibiotics. These molecules are hard to build from scratch in the laboratory; making them using modified enzymes speeds up the search for new antibacterial agents.
The product of polyketide synthase can be altered in two ways: by changing the order of the steps, and by altering the starting material. Experiments hint that the modules can be reshuffled without disrupting their activity.
What Khosla's team has now investigated is: do the modules care about what they are given? Will they work when presented with substances other than their customary 'natural' molecules?
To find out, the researchers studied the modular enzyme 'DEBS', which constructs the molecule from which erythromycin is ultimately made. They generated enzymes that each comprised a single module from DEBS, and presented them with a variety of starting materials to see what they achieved.
Surprisingly, the lone modules preferred non-natural substances to the natural one that they transform when operating together in the full DEBS molecule. This, say Khosla's group, indicates that the most important factor driving the evolution of the modules is not that they work as efficiently as possible but that they make the most biologically effective product -the best antibiotic agent.
So redesigned polyketide synthases may indeed be able to operate -- within limits -- on non-natural starting materials, but their ability to do so might be hard to predict on the basis of what they normally do in the cell.