Nature Genetics pp 427 - 430

Cleft lip, with or without cleft palate, is one of the most frequent birth defects in humans-as many as 1 individual per 500 live births are affected. The condition can be caused by a number of factors, some environmental and others genetic. Richard Spritz (of the University of Colorado Health Sciences Center) and colleagues now report the isolation of a gene mutated in some cases of cleft lip. The researchers first noted that the condition was quite common among an inbred population on the Caribbean island of Margarita. By studying families with a high incidence of the disorder, they were able to localize the region in which the gene lay and determine its identity.

The gene, PVRL1, encodes a protein expressed on the surface of cells that is thought to function as an 'adhesion' molecule, mediating contact between the cell and its environment. The same protein was recently shown to be a major receptor for most herpes simplex viruses-perhaps the most common viruses that infect humans. Although having two mutated copies of PVRL1 leads to cleft lip and other symptoms (such as mental retardation), individual in whom only one copy of the gene is mutated have greater resistance to herpesvirus infection and only slightly unusual facial features. The authors speculate that the high representation of the gene within the population of this Caribbean island might result from increased resistance against viral infection.

Nature Genetics pp 440 - 443, pp 444 - 447 and pp 381 - 384

The completion of the human genome draft sequence is a major-but preliminary-step in understanding how genes and genomes function. One of the most important tasks that remains is to determine what every single one of the 100,000 (or so) human genes does. But one normally figures out what a gene's role is by making it non-functional, which for obvious reasons cannot be done in humans. Enter the mouse: easy to manipulate genetically (and clonable), feasible to house in large numbers and studied in great depth, they serve as accurate models for many human conditions, from dopamine deficiency to dermatitis. But inactivating each gene-by independent targeting-would be difficult as well as time-consuming.

But there's more than one way to mutate a mouse. Using chemicals, such as ethylnitrosourea (ENU), DNA can be randomly mutagenized and thousands (or more) mutant individuals can be screened for any defect of interest. This approach has long been applied to fruit flies and other organisms, but to do a systematic screen with the mouse-a much more developmentally complex organism-would require large facilities, extensive databases and a great deal of organization and cooperation. (Not to mention a good bit of funding.)

Two consortia have now assembled large-scale mouse ENU mutagenesis projects and conducted preliminary screens. One group is mainly located in Germany and headed by Martin Hrabé de Angelis (of the GSF Research Center for Environment and Health), whereas the other is based in the UK and led by Steve Brown (of the MRC Mammalian Genetics Unit). Using similar approaches, each consortium generated over 10,000 mutant mice and screened them in a variety of clinical, biochemical, immunological and behavioural assays. In each case, hundreds of new mutant lines were successfully created-a valuable resource which both consortia plan to make freely available to researchers in the academic community. As indicated by Joseph Nadeau (of Case Western Reserve University School of Medicine) and Wayne Frankel (of The Jackson Laboratory) in a related Commentary article, the value of large-scale ENU mutagenesis projects is derived not only from the creation of the mutant mice as community resources but from the relative ease of identifying their underlying gene defects compared with other approaches.

Nature Genetics pp 436 - 439 and pp 366 - 367

Viral infection is generally thought of as a bad thing-and rightly so. But as we have gained understanding of how viruses function at the molecular level, we have been able to adapt them to serve as tools. The unique features of viruses-their abilities to enter cells and insert the DNA they are carrying-give them great promise as vectors for gene therapy, which depends upon introducing normal or engineered DNA back into cells to correct an existing condition. Unfortunately, the current generation of viral vectors for gene therapy has many disadvantages: they might not infect enough or the right type of cells, they can unnecessarily stimulate the immune system, or they might not be able to carry enough of the 'right' type of DNA. And our understanding of how viruses function is still not sophisticated enough to facilitate designing a virus that's better at performing a specific function.

Nay-Wei Soong (of Maxygen, Inc.) and colleagues have now found a means to make a 'better' virus. Taking advantage of the basic genetic concepts of variation and recombination, followed by selection, they have 'evolved' a virus that can infect specific types of cells more efficiently. The researchers combined DNA from six different virus strains in various combinations-approximately one million-and infected a cell type that none of the six strains would normally infect. By repeating the infection process five times-giving the best virus ample opportunity to outcompete the others-they managed to recover a new subtype of the virus that had the ability to infect the cells efficiently. In an accompanying News & Views article, David Curiel (of the University of Alabama) opines that although the technology has some limitations, its potential in "pushing the envelope" of vector development could be profound.