Our work can be divided into three interrelated parts: We are studying the general problem of retroviral replication, developing retroviral vectors that can be used in cells in culture and in intact animals, and focusing on one of the enzymes essential for the replication of HIV-1, reverse transcriptase (RT). We study the structure and function of HIV-1 RT because it is an important target for anti-AIDS drugs (for example, AZT, 3TC, and ddI are RT inhibitors). We believe that the development of more effective anti-RT drugs will depend on a better understanding of not only wild-type HIV-1 RT but also the drug-resistant variants that arise in response to drug therapy. This project is part of a large collaboration, a critical part of which involves Dr. Edward Arnold and his colleagues, who have used X-ray crystallography to solve a number of different structures of HIV-1 RT (both wild type and mutant). Our biochemical and genetic experiments on HIV-1 RT are, in many cases, inspired by this structural analysis. In turn, the biochemical analyses have provided guidance for some of the crystallographic experiments. Recent structural and biochemical analyses have shed new light on the mechanisms of HIV-1 RT drug resistance. For example, resistance to the nucleoside analog 3TC appears to involve steric hindrance. Replacement of the methionine normally found at position 184 of RT, which is part of the polymerase active site, with either isoleucine or valine creates a steric gate. A beta-branched amino acid at position 184 blocks the appropriate binding of 3TCTP but still permits the incorporation of normal dNTPs. In contrast, AZT resistance involves an enhanced excision of AZTMP after it has been incorporated into the growing DNA strand. The excision reaction is essentially the normal polymerization reaction run in reverse, except that the beta and gamma phosphates of ATP serve as the pyrophosphate donor. The mutations that confer resistance to AZT do not interact directly with AZTMP, but instead serve to enhance the ability of the mutant enzyme to bind ATP, which increases the rate of excision. The specificity of the excision mechanism for AZT is inherent in the structure of HIV-1 RT. Steric constraints involving the azido group cause the end of an AZTMP-terminated primer to preferentially reside in the nucleotide-binding site, which favors excision. We have spent more than 15 years developing a series of retroviral vectors (the RCAS vectors) based on the avian sarcoma/leukosis virus (ASLV) family of retroviruses. Unlike most other retroviral vectors, the RCAS vectors are replication competent. Until recently, however, the RCAS vectors would efficiently infect only avian cells. In order for a retrovirus to infect a cell, a specific protein on the surface of the virus (the envelope or Env protein) must bind to its cognate receptor on the surface of the host cell. Mammalian cells lack the receptors recognized by the RCAS vectors. We solved this problem in two ways: by creating a version of RCAS vectors that uses an Env protein from a mammalian retrovirus and by modifying mammalian cells so they will express a receptor from avian cells. Both of these methods work well; each has particular advantages. One of the advantages of using the cloned avian receptor is that we can create transgenic mice expressing the receptor in a subset of their cells or tissues; in this way, we can control the expression of genes carried by the RCAS vectors in the mouse model. There is an interesting difference in the behavior of RCAS vectors in avian and mammalian cells: Even though these viruses are replication competent in avian cells, they are replication defective in mammalian cells. Although the version of RCAS vectors that uses an env gene from a mammalian virus can infect mammalian cells, these cells do not produce infectious virus. As a result, this vector system represents a substantial improvement over previously developed retroviral vector systems: It is both simple to use (because the virus replicates efficiently in avian cells, a high-titer stock can be obtained quickly and easily) and quite safe (neither complementation nor recombination will allow these viruses to replicate in mammalian cells). The development of better vectors, and more effective AIDS treatments, rests ultimately on a better understanding of retroviruses and their life cycle. We are engaged in several projects designed to elucidate how retroviruses replicate. To give one example, the inability of avian retroviruses such as the RCAS vectors to replicate in mammalian cells raises an interesting question: How is replication blocked in these viruses? Although the analyses are not yet complete, there is good evidence that unspliced and partially spliced viral RNAs are not appropriately handled in mammalian cells. By studying this problem, we hope to learn more about the rules (and machinery) that control the splicing and transport of viral RNAs, which could provide useful information about cellular and viral processes. In addition, we hope to obtain information that would be useful in the design and development of new RCAS vectors.
