Viral replication is most conveniently studied in uniform, continuously dividing cell cultures, and most of the information presented in the first part of this book has been derived from such studies. Given a reasonable understanding of the intimate molecular details of the virus-cell interaction, the challenge addressed in the remaining chapters is to relate this understanding of the interactions of the virus and its natural host. How does the virus survive, evolve, and cause disease, and how can it be prevented from causing the disease and even modified (as a vector) to treat disease?

Unlike most cells in in vitro cultures, cells in the body are present in a great variety of states of differentiation, functional activation, and replicative activity. Retroviruses must conform and adapt to, and may even exploit, these differences. At any one time, only a small fraction of the cells in an animal is in the process of division: the vast majority is in either a quiescent (G₀) or a postmitotic state. This is a serious limitation for viruses whose replication involves DNA synthesis, including retroviruses. In such cells, deoxynucleotide concentrations are very low—well below the K_m value of retroviral RTs—and viral DNA synthesis is likely to be inefficient, at best. In addition, in such cells, the nuclear membrane presents a barrier that blocks the access of viral preintegration complexes to their chromosomal DNA targets. For some simple retroviruses (MLV is the clearest example), this barrier appears to be insurmountable. MLV DNA integration can occur only after the nuclear membrane breaks down during mitosis. For some complex retroviruses—of which HIV-1 is the best studied—the preintegration complex can transit an intact nuclear membrane; for these viruses, DNA integration can take place even in nondividing cells. The mechanism by which HIV (and probably other lentiviruses as well) can accomplish this feat remains controversial: The virion proteins MA, Vpr, and IN have been implicated, and the viral proteins presumably interact with cellular systems involved in import into the nucleus. This ability extends the range of target cells available to lentiviruses and is being exploited in the development of vectors that can infect nondividing cells.

The expression of critical cellular components, including cell surface receptors, transcription factors necessary for the expression of the viral genome, and other factors necessary for viral replication, can be affected both by the state of differentiation of the host cell and by position of the host cell in the cell cycle. With HIV-1, for example, expression of both the coreceptor necessary for viral entry and an active form of NF-κB, a key transcription factor necessary for efficient viral genome expression, is dependent on cycling of the host cell. These requirements limit the infection of lymphocytes to activated CD4⁺ T cells. As might be expected, viruses can turn such specificities to their advantage. MMTV, for example, interacts with differentiation-specific transcription factors to limit its expression to cells (B cells and lactating mammary epithelium) that favor its transmission to new hosts. Presumably, this limitation reduces the risk of collateral damage in an infected animal; such damage could reduce the host's viability and consequently its ability to transmit the virus to its offspring.