An extra protective layer primes this virus to do its harm, mainly in children in the developing world.

For a virus to infect the cells of its host, it must first find a way in. For some viruses, the strategy is to fuse membranes with the target cell, often after hitching a ride in a vesicle with an entry ticket. In the case of viruses with no membranes of their own, such as rotavirus, the story is more complicated. Using two complementary techniques of structural biology—x-ray crystallography and electron cryomicroscopy—HHMI researchers have uncovered some important details about how rotavirus manages to make its way into a cell.

Rotavirus particles surround their RNA genomes with three protein layers. Understanding the structure and behavior of the proteins that constitute those layers could help scientists design better vaccines for a disease that yearly kills half a million people, most of them children in the developing world. Current vaccines, based on a live, attenuated form of the virus, may be impractical in the world's poorest regions. A protein-based vaccine would be easier to ship, store, and combine with other vaccines, says HHMI investigator Stephen Harrison of Harvard Medical School and Children's Hospital, Boston. With this goal in mind, Harrison studies the virus's outermost coat proteins, which he says are appropriate candidates for a vaccine because immune system antibodies recognize them.

Despite its multilayered configuration, rotavirus lacks a lipid membrane, or envelope, that can fuse with lipid-containing membranes of host cells. Entry of the so-called nonenveloped viruses is not as well understood as that of their enveloped counterparts. They use a diverse, and still largely unexplored, range of strategies to infiltrate host cells.

Unlike the genomes of simpler viruses, rotavirus RNA always remains enclosed within two of the three protein layers that surround it in the infectious virus particle. Enzymes packaged with the genome make and export new RNA for incorporation into progeny particles. The outermost layer, acquired before the virus emerges from one infected cell and searches for another, consists of two proteins—VP4 and VP7. The spike-shaped VP4 is thought to perforate the membranes of host cells, but the role of VP7 in penetration has been less clear.

In work published in the June 12, 2009, issue of Science, Harrison and his team (which included student Scott Aoki, HHMI postdoctoral fellow Ethan Settembre, and collaborator Philip Dormitzer) crystallized VP7 in the clutch of an antibody and, using x-ray crystallography, determined the molecular structure of the complex. It showed both how calcium ions hold VP7 together as a trimer of three identical molecules and how an antibody can prevent it from coming apart. The investigators concluded that the VP7 trimer must come apart during viral entry and that a loss of calcium ions in the host environment might trigger this process.

Illustration: Ian Wright

	PAGE 1 OF 2	Continue