Nature Biotechnology pp 37 - 41 and p 19

Cancer patients may have found an unlikely ally in the food-poisoning bacterium Salmonella typhimurium. A team of US scientists has engineered the bacterium so that it not only can shrink tumors, but also lose its toxic and life-threatening inflammatory properties. Administration of this strain to mice bearing melanoma resulted in tumors that were 6% of the size of those in melanoma mice receiving no treatment. The altered Salmonella strain also induces markedly lower levels of inflammatory proteins and is almost 10,000-fold less toxic than other Salmonella.

Apart from being a major cause of food-poisoning, Salmonella typhimurium possesses the remarkable property of accumulating in, and retarding, tumors. The down side is that this pathogenic organism also contains a cell wall component (lipopolysaccharide) that elicits a huge and potentially life-threatening inflammatory response, known as septic shock. To get around this problem, David Bermudes and collaborators have altered the lipopolysaccharide content of the Salmonella cell wall. They showed that these Salmonella strains were non toxic to mice and pigs, whereas the wild type bacterium killed all of the mice injected and 90% of the pigs. The work suggests that tumor-targeting strains of Salmonella that have lost their inflammatory properties could provide novel treatment approaches for cancer, moving the work closer to application in humans.

Nature Biotechnology pp 42 - 47 and p 20

An approach described in this issue may mean fewer visits to the dentist in the future. Researchers in the UK and the Netherlands have designed a new type of antimicrobial agent that stops plaque bacteria in their tracks before they can get a foothold on the tooth surface. The approach may be generally applicable to other bacteria, highly specific, and very effective in preventing infection by clinically important pathogens.

Tooth brushing and antiseptic mouthwash help in preventing tooth decay, but why not prevent plaque bacteria from binding to the tooth surface in the first place? With this in mind, Charles Kelly and colleagues have designed an antimicrobial peptide to mimic the tooth-binding protein of the plaque pathogen Streptococcus mutans. The protein was designed to bind to the receptors on the tooth surface occluding them from the bacteria. In a trial of 11 patients, four out of four subjects receiving the peptide mimic were protected from streptococcal recolonization of their teeth. In contrast, adhesion of another nonpathogenic bacterium, Actinomyces, was unaffected by the peptide, suggesting that the antimicrobial action is specific for S. mutans. When tested in the laboratory, the peptide was also shown to suppress adherence of the pathogen to salivary receptors. The authors speculate that such peptide inhibitors of adhesion might be of use as a general antimicrobial strategy against other microorganisms.

Nature Biotechnology pp 58 - 61 and p 21

Scientists are always on the look out for ways of making biological enzymes more efficient, whether for harsh industrial processes or for making clothes brighter in the wash. Now, a team of Japanese scientists has come up with an approach to further improve enzyme properties beyond what was previously thought possible. By adding new amino acids to one end of an existing enzyme, catalase, they have created hybrid enzymes that are significantly more stable at high temperatures than the wild type protein. The approach may prove useful when used in combination with other protein engineering strategies to design enzymes with desirable qualities.

Bacterial catalase is an industrially important enzyme involved in the catalysis of hydrogen peroxide. The wild-type enzyme is isolated from Bacillus stearothermophilus, a bacteria tolerant of very high temperatures, so improving nature's design is a difficult problem. The Japanese researchers set about their task by adding a completely new sequence of amino acids to one end (the carboxyl terminus) of the catalase enzyme and expressing the enzyme in the bacteria Escherichia coli. This amino acid tail contained regions in which the amino acids had been randomized, so that a diverse range of enzyme sequences were produced. After screening each E. coli colony for catalase activity, they isolated 15 from 58 clones that had higher temperature stability than the wild type. They also demonstrated that the approach could be used to improve the temperature stability of a mutant prepared by traditional mutagenesis approaches.