|With a map of the human genome now complete, the next challenge is to understand how the functional interaction of gene products gives rise to cell physiology. This step is critical to translate the promise of genomics into advances in molecular medicine, but there is no clear consensus on how to achieve this objective. My laboratory is pursuing a chemical genetic approach that takes advantage of a novel intein-based technology called SICLOPPS (Scott et al., Chemistry & Biology 8: 801-15), which exploits the power of molecular biology to teach cells how to make cyclic peptides biosynthetically. |
We use cyclic peptides generated by SICLOPPS in two ways. We design high affinity antagonists of protein-protein interactions by adopting elements from physiological interfaces, stabilizing them against catabolism through backbone cyclization and then improving availability to intracellular targets through the introduction of protein transduction domains. This approach exploits the major advantage of large, polypeptides over small molecules (avidity) while simultaneously addressing their major liabilities (serum stability, bioavailability). The permuted intein that we use for cyclization is particularly well suited for making large cyclic peptides because the intein constrains the reactive ends in close proximity in the active site. Such polypeptides are an enormous challenge to traditional solid-phase or solution cyclization protocols. We are currently generating several large cyclic peptides designed to antagonize a variety of interfacial structures. one added advantage of our biosynthetic method is compatibility with biological selection and screening for rapid optimization of designed properties. When our designed constructs antagonize poorly, we perform random or targeted mutagenesis on the genetic construct encoding the cyclic peptide antagonist then screen the resulting mutant library for variants with improved function using genetic selection and screening methods (for example, two-hybrid systems).
In addition to biosynthesis of otherwise inaccessible large cyclic peptide antagonists, another significant advantage of SICLOPPS is as a method to produce molecular diversity for chemical genetics. To do this, we make a genetic construct encoding millions of different cyclic peptides, then transform or transfect the DNA into a host cell of interest. After transformation or transfection, each cell becomes a unique reactor, and its phenotype reports on the activity of the library member it encodes. Since our biosynthetic constructs are encoded genetically, the number of molecules we can screen is limited only by the amount of recombinant DNA we can introduce into a cell of interest, and the efficiency of our genetic screens. For mammalian cells, the transfection limit is somewhere between millions and billions of unique constructs. Our cyclization chemistry fixes one position of our cyclic construct (a serine, cysteine or threonine side-chain nucleophile needed for a transesterification step that affects cyclization), but the reaction has otherwise proven to be almost insensitive to the amino acid composition of the cyclic product. As a result, we can generate libraries with low molecular weight and high chemical diversity (eg cyclo-[SXXXXX] where S=serine and X=any amino acid; number of compounds = 3.2 million; average molecular weight ~650 Daltons). one nice thing about this library is that it is focused in a functionally rich region of chemical descriptor space. After all, peptides are the currency of biology. Peptides haven't been used that much for chemical genetic studies because the half-lives of linear peptides are usually too short to produce stable phenotypes. Cyclization goes a long way towards remedying that shortcoming. Biosynthesis of molecular libraries also overcomes the throughput problems that plague chemical genetic studies relying on synthetic small molecules. The key to any chemical genetic screen is to be able to attribute cell phenotype to a particular member of a library. The only way to address synthetic molecules in a cell-based chemical genetic screen is by spatial array, which dramatically reduces the number of compounds that can be screened, even in cases where an investigator or a university invests heavily in robotics, liquid handling, etc. In our method, since each cell encodes one library member, the identity of functional library members is easy to determine since the DNA encoding each library member is trivial to isolate and sequence from cells displaying diagnostic phenotypes. The bottom line is that, even without specialized equipment, our screening efforts are limited only by our transformation/transfection efficiencies and the cleverness of our genetic screens. We are currently screening libraries with one hundred million members in target-directed discovery programs, which are laying the groundwork for chemical genetic screening programs on a pathway to proteome scale. Realizing the potential of these cyclic peptide libraries as tools for chemical genetic dissection of metabolic and signaling pathways relevant to the pathophysiology of infectious diseases and cancer is the major goal of my laboratory.
Click here for a description of post-doctoral positions in my laboratory.
Keywords: chemical biology, chemical genetics, combinatorial, cyclic peptide, drug discovery, functional genomics, intein, library, SICLOPPS
|Selected Publications |
Scott, C. P., Abel-Santos, E., Wall, M., Wahnon, D. C., Benkovic, S. J. "The production of cyclic peptides and proteins in vivo." Proceedings of the National Academy of Sciences (USA) 96 (24), 13638-43, (1999).
Kashlan OB, Scott CP, Lear JD, Cooperman BS. "A Comprehensive Model for the Allosteric Regulation of Mammalian Ribonucleotide Reductase. Functional Consequences of ATP- and dATP-Induced Oligomerization of the Large Subunit." Biochemistry. 2002 Jan 15;41(2):462-74.
Maintained by the Informatics Shared Resources of the Sidney Kimmel Cancer Center at Jefferson
Copyright © Thomas Jefferson University. All Rights Reserved.
The Thomas Jefferson University web site, its contents and programs, is provided for informational and educational purposes only and is not intended as medical advice nor is it intended to create any physician-patient relationship. Please remember that this information should not substitute for a visit or a consultation with a health care provider. The views or opinions expressed in the resources provided do not necessarily reflect those of Thomas Jefferson University, Thomas Jefferson University Hospital, or the Jefferson Health System or staff.
Please read our Privacy Statement