James E. Darnell, Jr., M.D. - US grants

This is a broadly based program in which gene expression and genome replication in both prokaryotes and eukaryotes is studied. Cells infected with RNA and DNA tumor viruses are used frequently throughout the work. In addition the biosynthesis of proteins in cell membranes is studied. All of the studies are carried by the molecular level involving DNA, RNA and protein sequencing and levels of interactions and regulations between various virus specific and cell macromolecules explored. The implications for all of these results in cancer biology underlies the design of many experiments.

The aims of this proposal are to complete a description of transcription unit design in eukaryotic cells by determining how and where RNA polymerase II (pol II) transcription ceases. The laboratory is in the process of shifting mainly to transcriptional control studies of pol II transcription units during differentiation. The material to be used is chiefly differentiating erythroblastic (erythroleukemia) cells in culture and normal fetal and adult liver cells. A program is described for the isolation and characterization of genes that are expressed primarily or exclusively in the liver followed by testing which of the sequences confer liver specificity when these genes are reintroduced into animals. Finally, various techniques for following liver specific characteristics will be applied to developing mice to track at a molecular level the earliest detectable markers for liver differentiation. The health relevance of basic studies of this type lies in the confidence that many diseases result from defects in gene regulation both in adult and developing tissues. (G)

The experiments described in this proposal aim to contribute specifically to an understanding of hepatocyte specific gene control. Experiments are described that are aimed both at the signals that trigger and maintain liver-specific expression and the nuclear factors that play a direct role in coordinated hepatocyte expression of a number of genes (al anti- trypsin, transthyretin, albumin and others). In addition to specific expression of some genes in all hepatocytes we will study how certain hepatocytes that are geographically localized for example around the central vein produce particular mRNAs that other hepatocytes do not produce. This kind of information should have wide application in questions related to specific gene expression in other tissues and to specific gene expression during development as well. A fundamental understanding of the mechanisms of normal gene regulation in growing, developing and differentiated cells is widely believed, a belief we share, to be necessary to understand many medical problems including cancer.

Interferons (IFNs) are small polypeptides that not only cause cells to become resistant to viral infection but are among the most powerful natural growth regulatory substances that we know. They act by binding to specific cell surface receptors. The first effects of ligand- receptor interaction are changes in the rate of transcription of a limited set of genes in the nucleus of the cell. The major aims of this research are to understand the mechanisms by which these changes in gene transcription are brought about after IFN attachment to cells. The specific aims are to purify the proteins and then clone the genes encoding the proteins which we can demonstrate bind to DNA sequences required for IFN induction of transcription. In the case of IFN-alpha gene stimulation we have identified a factor, ISGF-3, composed of four proteins that reside in the cytoplasm of the untreated cell awaiting activation to become a positive acting nuclear transcription factor. The amino acid sequence of each of four proteins that make up ISGF-3 is being obtained now and the genes for all four proteins (113, 91, 84 and 48KD) must be obtained, sequenced and used in recombinant DNA form to map functional domains and to produce antibodies to study protein association. In the case of IFN-gamma stimulation of gene transcription a new immediately responsive, IFN-gamma dependent positive acting factor, GAF (gamma activity factor), and its IFN-gamma responsive DNA element have been defined. This will allow the purification of GAF and the cloning of genes encoding its proteins. In this way we can compare proteins whose transcriptional stimulation depends on activation by two different specific ligands acting on two different receptors, the overall outcome of which is to induce the antiviral state or slow cell growth. The health relatedness of understanding details of action of these two cytokines are quite direct: not only do both IFN-alpha and -gamma cause growth inhibition in many different cell types in culture, there is often a potentiation with both ligands. Knowing which intracellular molecules are primarily affected by each and by both ligands may help target IFN therapy more precisely both in establishing growth restraint and in inducing the anti-viral state.

Different polypeptide ligands can bind to specific cell surface receptors on the same cell and initiate different intracellular events including the immediate (non-protein synthesis requiring) activation of different sets of genes. We have discovered latent cytoplasmic transcription factors that are activated by such ligands. Those latent cytoplasmic proteins, termed STAT proteins for signal transducers and activators of transcription are phosphorylated on tyrosine in the cytoplasm before translocating to the nucleus to direct transcription. They were first discovered in cells treated with IFN-alpha or IFN-gamma. In this proposal we describe experiments to define the functional domains of one of these proteins, STAT 91, and the functional domains of the kinases (Jak1 and Jak2) that have been shown to be involved in the STAT activation pathway. A most important thrust of future work will be discover other proteins in this same family that serve in response to other ligands. Because the genes encoding the presently known STAT proteins have been found to have many (20) exons we will study the STAT mRNAs in different mouse tissues and in cells treated in a variety of ways to search for differently spliced STAT mRNAs that might function in different ligand-dependent pathways. Several newly discovered STAT protein family members with mRNAs that are present in the thymus are also described that have high homology to but are distinctly different from the already described STAT 91 and 113 proteins. Characterization of these proteins with particular attention to their possible tyrosine phosphorylation in response to other ligands is an important part of this proposal. Finally, collaboration is planned to study the three- dimensional structure of important domains of the STAT proteins and the kinases with which they interact.

The aims in this proposal are to continue fundamental studies that have provided an initial understanding of endodermal choices in development through the study of transcription factors, particularly hepatocyte nuclear factors, HNF-3alpha, beta and gamma and HNF-4 which has been the focus of our laboratory. Adult liver contains transcription factors that are also prominent in the GI tract and its accessory tissues but scarce or not present in most other adults tissues. After cloning the genes for HNF-3alpha, Beta and gamma and HNF-4, we have shown that N+HNF-3alpha and beta are expressed very early in embryogenesis, e.g. in Henson's node, even before germ layer separation. In addition HNF-4, shown earlier to be an activator of HNF-1 in cultured hepatoma cells, was found to be expressed in the very earliest liver bud. An additional critical finding that instills confidence that the HNF-3 and HNF-4 genes are important in development is the occurrence of very similar genes in Drosophila, mutations in which prevent normal gut development in the fly. The proposal for the next period of study includes molecular genetics experiments to examine differential promoter site and/or splice site utilization in the HNF-3 family and cell biology experiments to examine the signalling required to trigger HNF-4 and HNF-3 synthesis in embryonic cells and mouse genetic experiments including transgenic experiments with HNF-3 promoters. Gene inactivation by homologous recombination has already been achieved for HNF-4 and has produced embryonic lethal which are under further study. Gene knockouts for HNF-3Beta and alpha are also underway as a first test of a required role in early embryogenesis. The correct differentiation during embryogenesis of the specialized epithelial cell types is crucial to avoid developmental during embryogenesis of the specialized epithelial cell types is crucial to avoid developmental defects of the gastrointestinal tract. Furthermore, the maintenance of specialized epithelial cells through regeneration is crucial in adult life to avoid or combat many different gastrointestinal diseases including hepatitis, toxic liver damage, pancreatitis and various intestinal abnormalities such as inflammatory large and small bowel disease and many other conditions.

This ongoing program involves the collaboration of eight laboratory units of The Rockefeller University in predoctoral and postdoctoral training in virology, viral oncology, and cell regulation. Each laboratory has one or two faculty members who serve as preceptors for 1 to 20 postdoctoral fellows and 0 to 5 graduate students. With the addition of laboratories in which paraneoplastic neurologic diseases, HTLV and HIV, and phage recombination are investigated, the program includes a broad spectrum of studies designed to understand the basic mechanisms of virus replication, virus-induced transformation, cytokine function, signal transduction, and the regulation of transcription and other cellular processes. The combined faculty has expertise in the biochemistry of nucleic acids and proteins, the genetics of viruses and cells, and immunology. Research at the laboratory bench will be the primary emphasis of the training program. Courses and discussion groups in virology viral oncology, cell biology, biochemistry, and gene expression will be conducted by the staffs of the participating laboratories. In addition to a series of seminars given by invited speakers who will also meet with the trainees and discuss their research, there will also be laboratory research meetings and journal clubs that will provide important training opportunities. The University's excellent academic program and frequent lectures by distinguished visiting scientists in a wide variety of fields will also be available to the trainees. Predoctoral trainees will be drawn from students admitted to the University Ph.D. program. Selection of 5 trainees will be made jointly by the program director and the Dean of Graduate Studies. Postdoctoral trainees will be recent Ph.D.s in the biomedical sciences or M.D.s interested in research. The primary training facilities will be the laboratories of the eight participating units, each of which is well designed and equipped for the proposed research training. Other special facilities, such as for electron microscopy, synthesis of oligonucleotides and polypeptides, micro-sequencing of proteins, biohazard containment, and animal research, are available.

Determination of the extent and location of mercury derivatization of a recombinant form of STATI protein for X-ray crystallographic phasing purposes using MALDI- and ESI-MS methods. Further biochemical characterization of the protein using crosslinking and MALDI-MS peptide mapping methods.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Interferons (IFNs) and other cytokines and growth factors activate the JAK (Janus kinase)/STAT (signal transducer and activator of transcription) pathway. In the IFN signaling, type I IFNs (IFN-a/b) upon binding to their receptor, activate intracellular, receptor-associated, tyrosine kinases JAK1 and Tyk2. These activated JAKs, in turn, phosphorylate specific tyrosine residues on latent cytoplasmic transcription factors which subsequently assemble into a complex called ISGF3 (IFN stimulated gene factor 3). This complex is composed of either STAT1a (91kD) or STAT1b (84kD) and STAT2 (113kD), which together constitute ISGF3-a, and a non-STAT protein called ISGF3-g (48kD) which is a member of the interferon regulatory factor (IRF) family. This complex accumulates in the nucleus, binds to a DNA element, ISRE (IFN-a stimulated response element, a 15 base pair non-dyad symmetrical DNA element), and activates transcription of target genes. ISGF3 is not activated by type II IFN, IFN-g. Instead, IFN-g, upon binding to its receptor and consequent activation of JAK1 and JAK2, induces the phosphorylation of STAT1a (or STAT1b), but not STAT2. The phosphorylated STAT1 then dimerizes and binds to a GAS (IFN-g activated sequence), DNA element that is dyad symmetrical 5[unreadable]TTN5AA3[unreadable]. Although both STAT1a and STAT1b can bind to GAS, only STAT1a forms functional GAF (IFN-g activated factor) that induces transcription from the GAS elements. Ultimately, one of the most crucial determinants affecting inherent transactivation potential of induced STATs may be a particular nucleoprotein microenvironment. Our objective is to discern patterns of cooperativity between activated STATs and other transcription factors, coactivator, and corepressor complexes within the context of the native promoter sequences. Such interactions are probably necessary to explain the role of STATs in gene activation at different IFN inducible genes. We have approached this problem by comparative study of the complex binding sites in several STAT-responsive genes. While STAT-binding sites exist and are likely required in chromosomal regulatory regions, single GAS elements give very little or no induction on their own in transient transfection assays. Therefore, we started with known native GAS elements and by adding adjacent native nucleotides determined the minimal size sequence that was IFN-responsive when cloned into a reporter gene. Then we determined whether this reporter inducibility correlated with an appearance of a novel band shift using a corresponding oligonucleotide in electrophoretic mobility shift assays (EMSA). We have detected in EMSA constitutive, low-mobility, protein complex that by mutational analysis is shown to be required for optimal STAT-mediated promoter activation. In particular, GBP promoter sequences that contain intact GAS and ISRE that bind STAT1 homodimer and IRF1, respectively, but that cannot bind constitutive low-mobility complex, when cloned in front of the heterologous reporter gene are inactive. Only sequences that bind STAT1, IRF1 and the constitutive low-mobility complex are able to activate a reporter gene upon IFN-g induction. Studies of the DNA affinity and specificity of this complex revealed that its DNA-binding may be affected by mutations within GAS as well as GAS-like site of the GBP promoter, suggesting a possible physical interaction with STATs or occupation of STAT sites that is relieved after appearance of activated STAT1 dimer. DNA affinity of this complex, observed in EMSA, is completely correlated to the transcriptional activation potential of the corresponding reporter constructs in transfection experiments. Currently we are engaged in obtaining larger quantities of partly purified preparation of constitutive low mobility complex which we will further purify in order to identify constituent subunits.

DESCRIPTION (provided by applicant): The STAT proteins are latent cytoplasmic transcription factors that are activated by tyrosine phosphorylation through cytokine and growth factor interaction with their specific cell surface receptors. STATS regulate many genes that are crucial to innate immunity (mediated through STATs 1 and 2 and interferons), adaptive immunity (mediated by IL-12, IL-4 through STATs 4 and 6 and a host of other cytokines) as well as growth, development and differentiation (mediated through dozens of receptor tyrosine kinases activating particularly STATS). Not only the activation but the regulated inactivation of the STATs are vital to proper functioning since over-activity particularly of STAT3 in cancer is widespread. Drawing on new structural and cellular biochemical data we have proposed a new model of inactivation. This process of dephosphorylation appears to require intra-protein contacts and extensive mutagenesis of the known and proposed contact interfaces in both STAT1 and STAT4 is described to study both in vitro and in vivo interactions and resulting effects of such interactions. In many human cancers STAT3 is persistently activated not due to STAT3 mutations but to overactivity of kinases that activate STAT3 or loss of negative acting proteins, leading to a resistance to apoptosis. We believe that the best way to intervene in STATS dependent cancer cells is direct inhibition of STATS in the act of stimulating transcription. To this end we study the interaction of STATS with other nuclear proteins whose interactions are required for STATS dependent transcription. We have found and will continue to study a STATS/cJun/cFos interaction on a model enhanceosome and extend these studies to genes (e.g. Bcl-xL) that establish an anti-apoptotic shield in the presence of persistently active STATS. Details of persistently active STATS in anti-apoptosis will be studied by abruptly removing active STATS by treatment of cells with anti IL-6 antibodies and then determining the balance of mRNAs and proteins related to apoptosis. From the studies of STAT1 dephosphorylation and of STATS protein contacts will come discreet targets for development of small molecule inhibitors. Inhibition of dephosphorylation of STAT1 would prolong interferon action;inhibition of STATS activity would induce apoptosis in cancer cells. These basic studies coupled with drug discovery programs, for example in industry, should definitely lead to new drugs in the treatment of cancer and of viral infections.