Richard W. Roberts, Ph.D. - US grants

DESCRIPTION (provided by applicant): One (1) of the most challenging problems in molecular design is developing molecules that can bind protein surfaces and block protein-protein interactions. The ability to modulate protein-protein interactions in a directed fashion opens the possibility of probing and controlling biological systems through design. The long-term objective of this proposal is to use in vitro selection experiments to target protein surfaces, exploring fundamental aspects of protein recognition, structure, function, specificity, and catalysis. In vitro genetic approaches currently represent a powerful operational solution to the protein design problem. Previously, the PI has conceived, developed, and implemented mRNA-peptide and protein fusions (hereafter "mRNA display"), to design proteins using in vitro selection experiments. mRNA display provides important advantages relative to other in vitro and in vivo protein design strategies, such as the ability to examine very large libraries (>1013 individual sequences) in the absence of a living cell, with tight experimental control over binding and stringency. In this proposal, we will use G-protein linked signaling (specifically G protein a subunits and G protein-coupled receptors) as targets for ligand design. Our goal is to explore the biophysical chemistry of protein recognition, as it pertains to this important signaling pathway. Our specific aims are: 1) To develop class- and state-specific G protein-directed ligands. 2) To test the function of our G protein-directed ligands. 3) To explore the structure and function of peptide ligands targeting the Methuselah G protein-coupled receptor (GPCR). 4) To develop novel antibody mimetic ligands targeting chemokine GPCRs.

[unreadable] DESCRIPTION (provided by applicant): There is a pressing need to develop new molecular tools that recognize protein surfaces for uses ranging from imaging probes to systems biology (blocking or modulating protein-protein interactions, serving as sensors) to proteomics (affinity reagents for mass spectrometry or chip-based diagnostics) to novel therapeutic leads. Here, we propose to implement and test a new combinatorial approach for constructing both natural and unnatural mRNA display libraries-PURE mRNA display (PURE = Protein synthesis Using Recombinant Elements). In this approach, the protein synthesis machinery will be assembled from purified components, providing complete experimental control of both concentrations and constituents. PURE mRNA display thus has several potential advantages relative to the classical mRNA display, most importantly in providing a facile route to construct designer display libraries bearing unnatural amino acids. Our specific aims are: 1. To test if it is possible to construct mRNA-peptide fusions using the PURE protein synthesis system. 2. To Compare and contrast PURE vs. retic-based mRNA display for in vitro directed evolution experiments. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Studying the localization of proteins with conventional antibodies has greatly contributed to our understanding of the structure and function of neurons. However, conventional antibodies have several limitations that drastically limit their utility. Tissue must be fixed and permeabilized prior to staining and often the overlapping expression patterns of adjacent neurons are difficult to interpret because of the lack of contextual information. For these reasons, the precise subcellular localization patterns in vivo of the majority of neuronal proteins have not been well characterized. The purpose of the studies proposed in this grant is to develop genetically encoded probes that will allow the subcellular localization of neuronal proteins to be mapped in vivo and in real time with high fidelity. These probes consist of genetically encoded aptamers (intrabodies) that bind to endogenous neuronal proteins and are generated using the mRNA display system. Three different types of intrabodies will be generated: 1. Binders to individual cytoskeletal proteins that mark neuronal structures such as pre- and postsynaptic sites. 2. Binders to transmembrane proteins. These intrabodies will be modified to enable them to label either total protein or only protein that is present on the plasma membrane of the cell. 3. Binders to activated G-proteins. Intrabodies will be used to attach three types of molecules to endogenous target proteins: 1. Fluorescent molecules that can be used to report the localization of the protein. 2. proteins for measuring Ca++ concentration in the region around the protein. 3. proteins that are activated by light to produce depolarizing currents. Subcellular trafficking of proteins is crucial to virtually all neuronal functions, including establishment of synaptic connections, axon guidance and synaptic plasticity. Disruption of protein trafficking has been linked to such diseases as Alzheimer's disease and Parkinson's disease. Protein trafficking also plays a critical role in drug addiction. Intrabodies generated through RNA display will provide tools to map the subcellular localization of endogenous proteins with high fidelity, in vivo and in real time, which is not possible with current technology.

A grant has been awarded under the NSF's MRI program to the University of Southern California under the supervision of Dr. Richard W. Roberts and Dr. Mark Thompson. Funds from this grant will be used to purchase a 600 MHz NMR spectrometer that will be accessible to all departments at USC. Nuclear Magnetic Resonance spectroscopy (also known as NMR) is arguably the most versatile technique available for studying molecules. This instrument will allow scientists at USC to determine the structures and movements of molecules with detail down to less than 1 billionth of a meter, a nanometer. As such, this instrument will have a major positive impact on USC's scientific goals in the areas of 1. Biomedical Nanoscience, 2. Imaging, and 3. Energy. In particular, the instrument will be used 1) to determine the structures of biological molecules (DNA, RNA and proteins) needed for basic research, new therapies, and diagnostics, 2) to examine the motions of biological molecules, 3) to measure the interactions of drugs and drug-like molecules with their therapeutic targets, 4) to study chemical reactions in great detail, and 5) to explore new chemical methods and materials relevant to energy production.

Prior to this grant, USC lacked a modern biomolecular NMR in an open user facility. The first impact of this instrument will thus be to provide a new training and research tool for user groups involved in this project. This includes 18 faculty, 14 undergraduates, 104 graduate students and 54 postdocs across 6 academic departments and 5 schools within USC. Second, this grant will enable establishing a general structural biology colloquia at USC integrating this instrument (NMR) with other methods used to determine the structures of molecules, such as X-ray crystallography. Finally, this instrument will be used as a recruiting and training tool and for the USC TRIO and USC MESA programs aimed at serving disadvantaged, science-oriented high school juniors and seniors.

It is widely believed that a biological system centered on RNA or RNA-like biopolymer, preceded the DNA-, RNA-, and protein-containing system that now dominates. However, this "RNA world" is now disjointed from modern biochemistry and current understanding does not make clear how it could have given rise to the present world through a continuous set of evolutionary steps. The purpose of this research is to explore the hypothesis that the modern ribosome is a molecular fossil of the primordial RNA replicase. In a broader sense, the research aims to address the transition from a hypothetical RNA organism to the Last Universal Common Ancestor (LUCA). To do this, the research group of the Principal Investigator will explore whether the ribosome that currently performs carbon-centered amide and ester bond formation can also perform phosphate-centered transesterification reactions indicative of ribozyme-like chemistry in the same active site. The research group will also investigate the structural plausibility of adaptor-mediated replication in the context of the modern translation apparatus rather than the template-directed replication model that has dominated since the early work of Watson and Crick. The broader impact of this project lies in its effort to enhance both university and pre-university science education related to the themes of the research, through formal coursework, mentoring, and science demonstrations.

DESCRIPTION (provided by applicant): In this project, we propose to combine mRNA display (a protein design/evolution method) and high efficiency microfluidic sorting to create a new technology-microfluidic mRNA display-for the purpose of enabling design of peptides and proteins that can be used as protein capture reagents. We will develop and apply this powerful new technology toward creating a comprehensive reagent set aimed at the Hepatitis C virus (HCV) proteome. In this section, we begin by describing the existing state-of-the-art in 1) mRNA display-based peptide and protein design, 2) bead-based micromagnetic separations, and 3) give an introduction to the proteins expressed by the HCV that will be the targets of this work. This is followed by our description of how we will integrate these technologies to achieve our goal of high-throughput development of new protein capture reagents.

DESCRIPTION (provided by applicant): This application aims to demonstrate that Scanning Unnatural Protease Resistant (SUPR) peptides provide a general solution to the problem of targeting traditionally undruggable proteins. To do this, we will use mRNA display with an expanded genetic code to create a new class highly stabilized, membrane-permeant peptides that can block or modulate protein-protein interactions for two of the most important intracellular proteins conferring the oncogenic phenotype-the activated form of Ras and the Stat3 protein. Our three Specific Aims are: 1) To design stabilized SUPR peptides targeting intracellular undruggable proteins involved in cancer transformation or maintenance, 2) To characterize and enhance selected SUPR peptide functions towards cancer drug applications, and 3) To evaluate in vivo characteristics and assess the therapeutic potential of optimized SUPR peptide drug candidates for cancer treatment in mice. Overall, this project is intended to develop novel molecules as well as a general approach to target cancer-relevant proteins that have proved challenging up to this point-so much so that the proteins may be called undruggable. PUBLIC HEALTH RELEVANCE: The development of novel technologies to inhibit undruggable therapeutic cancer targets is an important public health priority. It can expand our abilities to discover new cancer drugs and improve our abilities to manage cancer. Successful completion of the proposed studies will not only offer many new treatment opportunities for cancer, but also provide new tools for cancer drug discovery.

Diagnosing and treating cancer requires having a reliable set of affinity reagents that can specifically and strongly bind to cancer-related protein targets. These reagents are the necessary molecular tools that will enable next-generation technologies for studying, diagnosing, and fighting cancer. Current approaches to producing such reagents, however, are unreliable, expensive, and slow. Existing reagent generation methods (e.g., hybridoma technology, phage display, yeast display) are not readily adapted to leverage high throughput library sequencing. Where reagents do exist, they are often both extremely expensive and poorly characterized. It is these important shortfalls that this research project aims to address. This project combines two powerful existing technologies: 1) mRNA display and 2) Modular Microfluidic and Instrumentation Components (MFIC) to dramatically speed target-directed, renewable reagent development. mRNA display is a molecular selection technology that is uniquely capable of searching libraries of more than a trillion unique compounds to develop ultrahigh affinity reagents against cancer-relevant targets. While this technology has an impressive demonstrated track record of producing such reagents, it has so far been limited to the laboratory scale. This project will adapt it to a true high-throughput format by integrating it into a continous flow microfluidic system based on MFIC technology. By automating mRNA display, this project will make it broadly accessible to the research community while decreasing its cost and increasing its throughput. Once the automated system is developed, we will use it to produce affinity reagents that target two key cancer screening protein markers?PSA and CA125. These markers are broadly used (they were the primary tools used in 154,000 patient Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO)) and renewable, inexpensive reagents that recognize these biomarkers would clearly be of utility. An automated microfluidic mRNA display system will be used to develop novel, specific, and high affinity reagents for these targets. Developing a microfluidic approach for mRNA display selection will involve implementing a magnetophoretic separation system that can both perform affinity selections and purify the products of preparative biochemical reactions. An automated system for preparing mRNA reagents will be combined with an automated target selection system. The DNA-encoded products of this selection will be amplified in a microfluidic PCR system. The entire process workflow will be implemented in a closed loop to enable multiple rounds of selection and amplication, producing an optimal high affinity binding reagent. Each step of the automated mRNA display will be benchmarked for quality assurance by developing affinity reagents for the cancer marker Bcl-xL; standard manual mRNA display has proved resoundingly successful at producing such reagents and there is plentiful existing data against which to benchmark.