In eukaryotic cells, the nuclear genome and its transcriptional apparatus is separated from the site of protein synthesis by the nuclear envelope. Thus, a constant flow of proteins and nucleic acids has to cross the nuclear envelope in both directions. This transport in and out of the nucleus is mediated by nuclear pore complexes (NPCs) and occurs in an energy and signal-dependent manner. Thus, nucleocytoplasmic translocation of macro molecules across the nuclear envelope appears to be a highly specific and regulated process. Viruses that replicate their genome in the cell nucleus are therefore forced to develop efficient ways to deal with the intracellulZlr host cell transport machinery. Historically, investigation of Polyomavirus replication allowed identification ofsequences that mediate nuclear import, which led subsequently to our detailed understanding of the cellular factors that are involved in nuclear import. Transport ofmacromolecules in the opposite direction, however, is less well understood. The investigation of retroviral gene expression in recent years pro vided the first insights into the cellular mechanisms that regulate nuclear export. In particular, the detailed dissection of the function of the human immunodeficiency virus type I (HIV-I) Rev trans-activator protein identified CRMI, as a hona fide nuclear export receptor. CRM I appears to be involved in the nucleocytoplasmic translocation of the vast majority of viral and cellular proteins that have subsequently been found to contain a Rev-type leucine-rich nuclear export signal (NES)."

The genome of retroviruses contains three major coding regions for virion proteins, gag, pol and env. Gag encompasses information for nonglycosylated viral proteins that form the matrix, the capsid and the nucleoprotein structures. From pol derive reverse transcriptase and integrase, and env codes for the surface glycoproteins of the virion which consist of a transmembrane and a surface domain, linked by disulfide bonds. A viral protease is derived eitherfrom the gagorfrom the pol coding region, depending on the virus. Simple retroviruses contain only this elementary gag, pol, and env coding information. Once integrated, they are able to multiply efficiently, using the cellular transcriptional and replication machineries without intervention of viral transacting factors. Most oncogenic retroviruses belong in this category. Complex retroviruses, on the other hand, encode additional nonstructural proteins from multiply spliced messages. These proteins play important regulatory roles in the life cycle of the virus. They function as transacting factors that, in concert with cellular regulatory proteins, control viral gene expression and function and are essential components in the replication of complex retroviruses. To this category belong the lentiviruses, the spumaviruses and a group of oncogenic retroviruses that includes human T cell leukemia virus (HTLV) and bovine leukosis virus(BLV).

In eukaryotic cells, the nuclear genome and its transcriptional apparatus is separated from the site of protein synthesis by the nuclear envelope. Thus, a constant flow of proteins and nucleic acids has to cross the nuclear envelope in both directions. This transport in and out of the nucleus is mediated by nuclear pore complexes (NPCs) and occurs in an energy and signal-dependent manner. Thus, nucleocytoplasmic translocation of macro molecules across the nuclear envelope appears to be a highly specific and regulated process. Viruses that replicate their genome in the cell nucleus are therefore forced to develop efficient ways to deal with the intracellulZlr host cell transport machinery. Historically, investigation of Polyomavirus replication allowed identification ofsequences that mediate nuclear import, which led subsequently to our detailed understanding of the cellular factors that are involved in nuclear import. Transport ofmacromolecules in the opposite direction, however, is less well understood. The investigation of retroviral gene expression in recent years pro vided the first insights into the cellular mechanisms that regulate nuclear export. In particular, the detailed dissection of the function of the human immunodeficiency virus type I (HIV-I) Rev trans-activator protein identified CRMI, as a hona fide nuclear export receptor. CRM I appears to be involved in the nucleocytoplasmic translocation of the vast majority of viral and cellular proteins that have subsequently been found to contain a Rev-type leucine-rich nuclear export signal (NES)."

Several discoveries are noteworthy for allowing us to probe the recesses of the virus infected cell and to search for cryptic viral genomes which might provide clues in our studies of cancer etiology or developmental biology. One of the most notable was the dis covery of reverse transcriptase. This marked a momentous occasion in the history of molecular biology. Not only did it provide insight into the mechanism of persistence of retroviruses but it also provided us with an enzyme that could synthesize a DNA copy of any RNA. This DNA copy could then be used as a hybridization reagent to search for both complementary DNA and viral-specific RNA. Thus one could follow the course of any viral infection or probe in tumor cells for hidden viral genomes. Second, a great deal of credit must be given to the geneticists who isolated the various deletion mutants in the 'avian retrovirus system and thus provided us with the frrst means of isolating gene-spe cific probes. Finally, the laboratories which have mapped the genome have provided us with the framework in which to ask very specific questions with our gene-specific probes. Recently, numerous excellent reviews concerning various aspects of the retroviruses have appeared. In this review I shall not even attempt to present a comprehensive review of retroviruses."

1. 1 Scope of the Review This review was intended initially as a reference source for those interested in the origins and fITst descriptions ofthe defective avian sarcoma viruses. Quite a few of these viruses have been characterized in the past few years and their varied nomenclature according to source, discoverer, date of isolation or biological properties could result in some con fusion among those attempting to follow the literature. Information will be included on the molecular biology of the sarcoma viruses, rather more of which is available than when the review was fITst conceived, although in this respect the review will inevitably be out of date by the time of publication. If any bias of content is introduced, this will be towards a more detailed coverage of the author's own area of interest, the gene products of the defective sarcoma viruses. Rous sarcoma virus (RSV) serves as the model for much of this work and will frequently be referred to for comparative purposes, as will the mammalian defective transforming viruses. Recent reviews provide more complete coverage of these topics (l-4a). This review is complemented by a discussion ofthe avian acute leukaemia viruses, which appears elsewhere in this volume 5]. As will be seen, concentrating primarily on the defective sarcoma viruses and comparing them to RSV can be justified in terms of their biochemical properties as well as their similar biology."

Binding of various ligands (hormones, neurotransmitters, immunological stimuli) to membrane receptors induces the following changes: 1. Receptor redistribution (clustering, "capping") 2. Conformational changes that can be detected by fluorescent probes 3. Alteration in membrane fluidity (spin label and fluorescence polarization probes) 4. Changes in fluxes of ions and metabolites 5. Increased phospholipid turnover (especially of phosphatidyl inositol) 6. Activation of membrane-bound enzymes (adenyl cyclase, ATPase, transmethylases). Some of the early changes resulting from or associated with the binding (adsorption) of virions to the host cell membrane are of the same type. Adsorption of animal viruses to cells is the ftrst step in a chain of events resulting in the production of progeny virus on the one hand and in damage to cells and tissues on the other. In the classical studies of viral infection, cells are adsorbed with virus, usually for 60 min, and the changes induced by the virus in the host cell are recorded thereafter. In the past decade, more and more studies have been aimed at the events occurring in these ftrst 60 min of the so-called adsorption period. These studies deal with the nature of adsorption, e. g. , the ligand-receptor type of interaction between the virus and the cell membrane. Many receptors for viruses were identifted and so were the viral proteins which take part in adsorption.

Many of the fundamental concepts of animal virology originated from the study of the variola-cowpox-vaccinia virus system with vaccinia virus serving as the type species (Fen- nerand Burnet 1957; Burnet 1959; Fenner 1976a, b). The importance of the Poxviridae(Fen- ner 1979) for the study of viruses as biologic entities and in defIning the events which occur in virus-infected cells are exemplifIed by investigations which: (a) described the epidemiology of a virus disease in an animal population (Fenner1949, 1959b); (b) em- ployed electron microscopy to study virion structure (Peters 1956, Nagington and Home 1962, Dales and Siminovitch 1961) and to derme the morphologic stages of virion develop- ment in infected cells (Morgan et al. 1954, Dales 1963); (c) dermed and elaborated on the mechanism of nongenetic reactivation for an animal virus (Joklik et al. 1960a, Fenner and Woodroofe 1960, Hanafusa 1960); (d) described the intracellular uncoating of a viral genome (Joklik 1964a, b); (e) studied the antigenic structure and complexity of poxvirions (Loh and Riggs 1961, Woodroofe and Fenner 1962, Appleyard et al. 1964, Appleyard and Westwood 1964); (1) described the use of chemotherapy to treat viral infec- tions (Bauer et al. 1963); (g) fIrst demonstrated the presence of virion-coded enzymes encapsulated within virions (Kates and McAuslan 1967, Munyon et al. 1967); and (h) established the H -2 restriction of cytotoxic T-cell killing of virus-infected cells in the murine system (Doherty et al. 1976).

characteristic features in common with the genome of other retroviruses: long terminal repeats (L TR), and coding regions for internal proteins (gag), for re verse transcriptase (pol), and for glycosylated virion surface proteins (env), ar ranged in the sequence gag, pol, env from the 5' to the 3' end of the genome. However, the HTL V genome also contains some specific features not shared with all other retroviruses: the LTR regions are unusually long (745 base pairs, with 298 base pairs constituting the R region), but unlike the long L TRs of mouse mammary tumor viruses, they do not contain open reading frames. A stretch of noncoding sequences separates the gag and the pol genes. Most interestingly, the HTLV genome contains a region between the 3' end of the env gene and the L TR, called the pX region, that encompasses four open reading frames. Leukemic T cells freshly obtained from patients contain the HTL V provirus but usually do not express it. However, once established in culture, these cells produce viral proteins and release type C particles. Likewise, T cells infected and transformed by HTL V in vitro synthesize virus. Such producing cell lines have been widely used in seroepidemiological surveys and continue to be of importance for detailed studies of viral proteins and nucleic acids."

The technique of microinjection along with viral genetics and molecular biology has proven useful in the correlation of retroviral polynucleotide structure with function. The advantage of this technique is the involvement of living cells where rare activities may be observed and where properties of living cells can be assayed. Future studies involving recombinant DNA molecules and the asso- ciation of proteins with nucleic acids promise to yield additional insight into the nucleotide sequences involved in the expression of viral activities. References Anderson SM, Chen JH (1981) In vitro translation of avian myeloblastosis virus RNA. J Virol 40: 107-117 Berget SM, Moore C, Sharp PA (1977) Spliced segments of the 5' terminus of adenovirus 2 late mRNA. Proc Nat! Acad Sci USA 74:3171-3175 Bishop JM (1978) Retroviruses. Annu Rev Biochem 47:35-88 Capecchi MR (1980) High efficiency transformation by direct microinjection of DNA into cultured mammalian cells. Cell 22:479-488 Chien, Y-H, Junghans RP, Davidson W (1980) Electron microscopic analysis of the structure of RNA tumor virus nucleic acids. In: Stephenson JR (ed) Molecular biology of RNA tumor viruses.

and how the known vertebrate homologues of these genes are expressed normally in differentiation and proliferation pathways as well as abnormal ly in well-defined lymphomagenic and other oncogenic pathways. What emerged from this meeting are a better understanding of the evolution of these gene systems themselves and an elucidation of simpler systems open to more rapid genetic and molecular genetic analysis to reveal the normal functions of these genes and their gene products. Thus we sought new answers to several old questions concerning differenti ation, proliferation, and neoplastic transformation. We gathered together in an unusual format - that of the unique Dahlem Workshops - not just to reiterate data which has recently emerged but to think about how these findings might lead to new approaches for the understanding and therapy of the leukemias and lymphomas. We deliberately chose experts from several different disciplines, ranging from the clinicians who diag nose, describe, and treat these maladies, to the molecular geneticists trying to reduce the analysis of the problem to its simplest variables in the simplest systems possible."

This book is a comprehensive survey of new and exciting developments regarding the role of DNA methylation in human cancer. Issues related to the mutagenicity of 5-methylcytosine and the increase in the interaction of chemical and physical carcinogens with these residues is discussed. The book summarizes the modulation of viral gene expression and the silencing of tumor suppressor genes and illustrates mechanisms by which the methylation signal is translated into altered chromatin structure. The relationship between DNA methylation and genomic imprinting and cancer, and changes in CpG island methylation which occur in aging are discussed. Mouse model systems have played a key role in our dissection of the relationship between methylation and cancer, and these are also portrayed together with descriptions of new clinical trials in which methylation inhibitors are being used to treat leukemia, myeloid dysplastic syndromes and hemoglobinopathies.