Molecular Mechanisms of Virus BuddingEnveloped viruses acquire their lipid bilayers by budding through cellular membranes, but their mechanisms of release are unknown. The goal of our research is to determine how the structural components of these viruses move through the budding pathway, culminating in this virus-cell separation step. Our primary focus has been on the retroviral Gag protein because it is the only protein needed for the budding of these viruses; however, we have recently expanded efforts to address the more complicated herpesviruses, for which the minimal budding machinery is completely unknown.
Retrovirus budding. Gag proteins are synthesized on free ribosomes in the cytoplasm and are subsequently transported by an unknown mechanism to the inner face of the plasma membrane (see first figure, below). There, approximately 1500 copies come together to direct the emergence of a bud on the cell surface. Our studies have revealed that very large sections of the Rous sarcoma virus Gag protein (>75%) can be removed without impairing budding. The essential portions fall within three small regions, each of which has functional homologs in the Gag proteins of distantly-related retroviruses and also in unrelated enveloped viruses. Moreover, these "assembly domains" are interchangeable between viruses. The "M" domain provides specific membrane-targeting activity and is located near the amino terminus of Gag. The "I" domain mediates the primary interactions among the 1500 Gag molecules that come together during assembly to make a virus particle. The "L" domain is located at different positions in the Gag proteins of different viruses and is needed for a very late step in virus-cell separation. We are particularly interested in this late membrane-fusion event that enables the nascent particle to "pinching-off".
L domains are only 4-5 amino acids in length and work in a positionally-independent manner within Gag proteins. How can something so small have such a powerful activity? Our experiments were the first to suggest that L domains are ligands that recruit host factors to the site of budding, and numerous experiments in a variety of laboratories have proven this to be the case. In other words, the virus is not capable of separating itself from the host membrane but instead induces the cell to mediate this step for it. Our current efforts are directed towards identifying the host factors that are involved in budding. These experiments have revealed that ubiquitin is actually part of the budding machinery, and we are pursuing the role this small host protein plays. However, it is clear that the budding machinery is complex, and hence, there are numerous opportunities available in the lab.
Herpesvirus budding. Very little is known about the budding of herpesviruses. In recent years, it has become clear that these viruses (ulike retroviruses) bud into the trans Golgi network (see second figure). A great deal of genetic data suggest that there are probably several herpesvirus proteins that are needed for budding, and all of these reside in the "tegument" region of the virus (that part of the virus that is situated between the membrane and the capsid). The goal of our current experiments is to characterize the trafficking properties of tegument proteins that enable them to accumulate at the TGN compartment of the cell. We hope to eventually elucidate the minimal budding machinery of these viruses and to learn how that machinery interacts with the host cell to enable virus production. It will be interesting to see whether herpesviruses utilize some of the same host proteins that retroviruses use for budding. | 
