Stephen P. Goff

Scholar: 1982

Awarded Institution
Higgins Professor of Biochemistry & Molecular Biophysics
Columbia University
Department of Biochemistry and Molecular Biophysics


Research Interests

Areas of research by the Goff laboratory

The Goff laboratory is interested in the study of retrovirus replication and retroviral oncogenes. There are currently three major areas of investigation: mutational studies of the replication of the murine leukemia and human immunodeficiency viruses; biochemical studies of retroviral enzymes, including reverse transcriptase; and genetic studies of the tyrosine kinase oncogenes in mice.

In the area of retrovirus replication, the laboratory is interested in determining the roles of the various viral proteins in executing each step in the complex life cycle. Retroviral genomes are compact, and only carry three major structural genes: gag, encoding virion structural proteins; pol, encoding the reverse transcriptase and integrase enzymes; and env, encoding the virion envelope protein. The product of each of these genes is a polyprotein that is processed to several mature products; in many cases, the role of each domain is unknown. The major approach being taken in the lab to determine these roles is to generate mutations in cloned proviral DNAS, to recover virus after transformation of mammalian cells with the altered DNA, and to analyze the resulting virus for its ability to replicate in culture. Major efforts are focussed on the functions of the Gag polyprotein in retrovirus assembly and disassembly; and on reverse transcriptase and integrase.

Another very active area of research is centered around the use of the yeast two-hybrid system to detect and characterize protein-protein interactions between viral proteins, and also between viral and host gene products. The oligomerization of Gag proteins can be readily monitored using the system, permitting a rapid assay for interactions that can reflect virion assembly. The system thus provides a genetic screen for mutants affected in Gag multimerization. Similarly, dimerization of the HIV-1 integrase can be detected, and has been used to permit isolation of interactionnegative mutants. Mammalian CDNA libraries have been screened for clones whose products interact with the viral Gag proteins; these studies lead to the discovery that cyclophilin A, a host prolyl isomerase, binds specifically to the HIV-1 Gag, is incorporated into the virion particle, and is required for infectivity. A number of other host genes have been identified through similar searches with the Gag proteins of other retroviruses. A screen for interacting partners of the HIV-1 integrase uncovered a human homologue of the Snf5 transcriptional activator; the protein acts to stimulate the integrase function in vitro, and may serve to target retroviral integration into selected sites in the host genome.

Substantial efforts are focussed on the biochemistry of the retroviral reverse trasncriptase enzyme. The lab has developed plasmids that express enzymatically active reverse transcriptase of both mouse and human viruses; refinements of these clones have resulted in the production of stable, soluble, and highly active protein.

Fine-structure mutational analysis of these expression constructs proved that reverse transcriptase contains separate, and indeed separable, DNA polymerase and RNAse H domains. Analysis of mutant viruses specifically lacking the RNAse H function has shown the importance of this activity in virus replication, and study of mutant enzyme has lead to the suggestion that the RNase H domain may be required for proper formation of the polymerase-substrate complex. These mutants have also been helpful in defining the substrate specificity of the RNase H enzyme. Future efforts will address the determinants of DNA polymerase processivity, fidelity, and specificity for RNA and DNA templates, and for ribose- and deoxyribose-containing substrates.

In the area of retroviral oncogenes, the laboratory is interested in understanding the determinants of -tissue tropism manifested by several activated tyrosine kinases. The v-abl gene of the Abelson murine leukemia virus, for example, is narrowly restricted to the transformation of cells in a particular stage of the developmental pathway of B-lymphocytes. Similar viruses carrying the v-src oncogene transform a much broader spectrum of cell types, inducing sarcomas and several lymphoid tumors. To identify the regions of the viral oncogenes responsible for these specificities, the lab has generated large libraries of mutants with alterations in the oncogenes and determined the tumor types induced in mice. In addition, viruses have been constructed expressing hybrid oncogenes containing portions of the v-abl and v-src genes, and similarly tested in animals. The results suggest that there is a portion of the C-terminus of the v-abl protein required to target the Abelson virus to transform B lineage cells. Work is in progress to identify new host proteins that bind to this specific region of the v-abl protein; one such protein, dubbed Abi-1 ' is an SH3 protein that can modulate v-abl transforming activity. Dr. Goff is also extending these efforts by making viruses with.novel tumor specificities.A new transforming virus containing the axl gene has recently been generated, and work is in progress'to determine its tumorigenicity in animals.

In the third area, we are involved in makingmutations in the mouse germ line. Several years ago we successfully generated a targeted disruption of the murine c-abl locus on chromosome 2 in embryonic stem cells, and were able to introduce the mutant allele into the mouse germ line. The mice carrying homozygous mutations in the c-abl gene show particular defects in the development of early Band T-lymphocytes, and usually succumb to fatal infections early in life. Particular compartments of cells in the bone marrow are profoundly underpopulated in the mutant animals. In addition, those B and T cells which do mature and move into the peripheral circulation are defective in their response to various mitogens such as concanavalin A, phytohemagglutinin and lipopolysaccharide. Very recently, in a collaboration with Dr. Ihor Lemischka at Princeton University, we have disrupted theflk-2 gene, encoding a receptor kinase expressed on the surface of fetal liver cells. The mutant allele has been introduced into the germ line, and analysis of these homozygous mutant mice suggests that lymphoid cells early in the developmental pathway are affected. Disruptions of genes encoding other tyrosine kinases, including axl, and other proteins important in signal transduction, are underway.