About the Lab
Resident Faculty Member, CABM
Professor, Rutgers, Department of Molecular Biology and Biochemistry
CABM Structural Bioinformatics Laboratory
Dr. Gaetano Montelione develops new NMR techniques and software for refining 3D structures of proteins and triple resonance experiments for making 1H, 13C, and 15N resonance assignments in intermediate sized proteins. His laboratory has determined 3D solution structures for several small proteins including epidermal growth factor, type-alpha transforming growth factor, RNA-binding proteins involved in cold-shock response, and immunoglobulin-binding proteins. Dr. Montelione has received the Searle Scholars Award, the Dreyfus Teacher-Scholar Award, a Johnson and Johnson Research Discovery Award, the American Cyanamid Award in Physical Chemistry, the NSF Young Investigator Award, and the Michael and Kate Bárány Award for Young Investigators of the Biophysical Society. He is an elected Fellow of the American Association for the Advancement of Science. Montelione serves as Associate Editor to the journals Proteins: Structure, Function, Genetics and the Journal of Structural and Functional Genomics, and as a member of the Executive Committee of the International Structural Genomics Organization (ISGO). He also participated in the NIH Protein Structure Initiative as Director of the Northeast Structural Genomics Research Consortium (NESG).
The general aim of Dr. Montelione's research is to implement NMR spectroscopy as an enabling technology for structural and functional genomics. Montelione and his group develop and refine methods for protein solution structure determination, and apply these techniques to proteins of pharmaceutical or medical interest. This research has important implications in the fields of protein physical chemistry, structural bioinformatics, molecular design, receptor-ligand interactions, and oncogenesis.
Description of Research
For small proteins (<25 kD), distance-geometry calculations with NMR-derived distance constraints already provide reliable, high resolution structures. Recent experience indicates that the same methods can be extended to proteins with molecular weights up to ca. 40 kD. Approximately 25% of the laboratory effort is directed toward developing new NMR pulse sequences for solving larger proteins and for determining protein structures more precisely. The major part of this effort involves heteronuclear 2D-, 3D-, and 4D- NMR experiments in which magnetization is transferred back and forth between 1H and 13C and/or 15N nuclei. These experiments are carried out with proteins biosynthetically enriched with 13C,15N, or 2H. They provide conformation-dependent NMR parameters, which in turn facilitate protein structure refinement and studies of protein dynamics. Three- or four-dimensional NMR experiments, using additional 13C or 15N frequency axes, constitute an important new approach for unraveling complex spectra of larger proteins. Additionally, Montelione's group develops artificial intelligence computer software for automated analysis of NMR spectra.
The second 25% of the group's work is focused on developing and implementing computational methods for determining and refining protein structures based on NMR data. The researchers develop methods for refining protein structures by comparing experimental and simulated NMR spectra. Borrowing from the extensive experience of workers in the energy-refinement field, they utilize molecular-dynamics and Monte Carlo sampling procedures for overcoming local minima in distance geometry calculations.
The balance of the laboratory effort centers on determining 3D protein structures and protein folding. The proteins currently under study include: growth factors, immunoglobulin-binding proteins, ribonucleases and both RNA- and DNA-binding proteins. Solution structure analysis of these proteins will promote a more complete understanding of their structure-function relationships.
The general aim of our research is to use nuclear magnetic resonance (NMR) spectroscopy for understanding the atomic basis of molecular recognition and the relationships between protein structure and function. We develop new methods for protein structure determination and apply these techniques to proteins of pharmaceutical and medical interest. A significant effort in the lab focuses on developing new spectroscopic and computational methods for determining protein structures from NMR data. These methods are used to (i) determine three-dimensional structures of proteins and characterize their overall hydrodynamics and internal molecular dynamics, (ii) determine the structures and dynamics of protein-protein and protein-nucleic acid complexes and provide information for rational drug design, (iii) study the mechanisms by which proteins fold into their biologically-active conformations, and (iv) develop the concepts of structure-based functional genomics and structural bioinformatics in which structure determination and computational bioinformatics methods are combined to discover biochemical functions of proteins identified in the human genome project. Our aim is to apply this new technology to protein sequences identified by genomics and bioinformatics methods in order to discover new functions for biomedically important genes and their gene products.
1. Solution structure and dynamics by NMR. We have determined three-dimensional structures of several small proteins by NMR, including human and mouse epidermal and human type-alpha transforming growth factors, the major cold-shock protein from E. coli, the RNA-binding domain of the NS1 protein from influenza A virus, and the IgG-binding domain of staphylococcal protein A. Most of this work is supported by grants from NSF and NIH.
One major focus of the lab involves analysis of structure-function relationships in RNA-binding proteins. The solution structure of the RNA-binding domain of non-structural protein 1 (NS1) from influenza virus, a novel RNA-binding motif, has been determined by NMR spectroscopy as part of collaboration with Prof. R. Krug of the Rutgers Dept. of Molecular Biology and Biochemistry. Crystals of this RNA-binding domain were also obtained and provided to Prof. Helen Berman of the Rutgers Dept. of Chemistry for structural analysis by X-ray crystallography.NS1 is an important target for the development of antiviral drugs for treatment of influenza infection.
In collaboration with Dr. Y. Furuichi (AGENE Corp., Nippon-Roche, Tokyo, Japan) and Prof. S. Anderson (CABM), we have undertaken structural analysis of a proposed RNA-binding domain for a gene associated with the premature aging disease Werner's Syndrome. We have used simulated annealing methods developed at CABM to homology model this protein, and a manuscript describing this predicted structure is now in press. A high-level bacterial expression system for the domain has been constructed and is now being used to produce isotope-enriched samples for NMR analysis.
Nuclear relaxation time measurements are used to characterize intramolecular motion in these small proteins. Working together with Prof. R. Levy of the Rutgers Dept. of Chemistry, we have published a detailed study of molecular dynamics and internal motions in human type-alpha transforming growth factor (hTGFa) based on nitrogen-15 relaxation studies, and have compared these experimental data with molecular dynamics simulations that provide a motion picture of the structural dynamics of hTGFa in solution. Together with Dr. Levy, we have also developed a novel method for graphical analysis of nuclear relaxation data. Nuclear relaxation studies have also been used to characterize dynamic changes associated with disulfide deletion in the proteins bovine pancreatic trypsin inhibitor (BPTI) and bovine ribonuclease A (RNase A). This research has important implications in the fields of protein physical chemistry, molecular design, receptor-ligand interactions, protein folding, and oncogenesis.
2. Structures and dynamics of protein-protein and protein-nucleic acid complexes. In collaboration with Prof. M. Inouye of the UMDNJ Dept. of Biochemistry, we have used NMR to identify the single-strand RNA-binding site on the surface of the major cold shock automated methods and computer software for determining protein structures from NMR data. The combined techniques of NMR spectroscopy and conformational energy calculations are being used to (i) determine three-dimensional structures and dynamics of small proteins in solution, protein (CspA) from E. coli. This protein-RNA interaction appears to be critical in allowing bacterial cells to acclimate to cold stress. We recently completed a refinement of the solution NMR structure of CspA. Homologous "cold-shock" proteins occur in eukaryotic organisms, including humans. These proteins appear to function as "RNA chaperones" to maintain RNA in a single-stranded form even under cold conditions where internal hydrogen-bonded duplex formation can occur.
NMR methods have also been used to characterize the structure and dynamics of a serine protease inhibitor (BPTI) bound to its serine protease target (trypsin). Serine proteases play critical roles in the physiology of blood. A paper describing this work has been submitted for publication.
3. Mechanisms by which proteins fold into their biologically-active conformations. A crucial area in the field of molecular biophysics involves determining how protein sequences fold into unique three-dimensional structures. We have made some efforts to understand protein folding mechanisms by analysis of three-dimensional structures of protein folding intermediates. In collaboration with Prof. H. Scheraga (Dept. of Chemistry, Cornell University), we have completed an NMR structural analysis of two disulfide-deletion mutants of the protein ribonuclease A (RNase A) that are analogs of protein folding intermediates identified from kinetic folding experiments. These analyses provide important information about the sequential order of native structure formation in this folding process and about the role of structural dynamics in the energetics of folding. Efforts have also been initiated to develop new methods for characterizing kinetic folding intermediates directly by time-resolved NMR spectroscopic methods. Information gained from these techniques will allow us to understand better the rules that determine the folding mechanisms of proteins.
4. Structure-based Functional Genomics. A major focus in the lab involves the development of new NMR pulse sequences and computer software for the rapid and automatic analysis of protein structures from NMR. Significant progress in this area includes the development of computer programs for automated analysis of protein NMR data, and the development of improved NMR pulse sequences for protein structural analysis. The software development is part of an ongoing collaboration with Prof. C. Kulikowski in the Rutgers Dept. of Computer Science. NSF largely supports these projects, but the software development work also receives some support from the Rutgers University Strategic Resource (SROA) Program.
The goal of this part of the project is to develop high throughput technologies suitable for determining many new protein structures from the human genome project. These structures provide important insights into the functions of novel gene products identified by genomic and/or bioinformatic analysis. The resulting knowledge of structure and biochemical function provides the basis for collaboration with pharmaceutical companies to develop drugs useful in treating human diseases that are targeted to these newly discovered functions. The approach we are taking is opportunistic in the sense that only proteins which express well in bacterial expression systems will be screened for their abilities to provide high quality NMR spectra, and only those that provide good NMR data are subjected to automated analysis methods for structure determination. The success of our approach relies on our abilities to identify, clone, express, and analyze by NMR many biologically-interesting proteins per year; only a fraction of the initial sequences chosen for cloning and analysis will result in useful structures. However, this "funnel" process can yield new functions for tens of new structures per year, and can thus have tremendous scientific impact. A New Jersey Commission on Science and Technology Excellence Award has been made to a research team organized jointly by Montelione and Anderson to pursue an Initiative in Structural Bioinformatics based on these ideas.
We have established a highly productive laboratory for protein structural analysis by NMR. A new 600 MHz NMR spectrometer has recently been installed in CABM, and the existing 500 MHz NMR has been upgraded. The mission of the laboratory is both to develop new technologies for structural analysis and to apply these technologies to important biological and biomedical systems. NMR pulse sequences and software have been developed which reduce the time required for analysis of a protein structure, under favorable circumstances, from several months to just a few weeks. This technology is being applied in characterizing the structures and dynamics of several protein-protein and protein-nucleic acid complexes, and to provide new insights into the roles of internal dynamics in macromolecular recognition and protein folding. The technology and knowledge generated in this work has broad values in the areas of biotechnology, bioinformatics, and pharmaceutical discovery.
We will continue our efforts to automate the NMR analysis procedure using artificial intelligence computer algorithms. The structural changes of several small proteins as they bind to receptor and polynucleic acid molecules will also be characterized. Over the coming months, particular emphasis will be placed on small proteins as they bind to receptor and polynucleic acid molecules. Efforts will also be made to develop technologies for high throughput structure analysis and structural genomics.