The Schepartz laboratory develops chemical tools to study and manipulate protein–protein and protein–DNA interactions inside the cell. Our approach centers on the design of molecules that Nature chose not to synthesize--miniature proteins, ß-peptide foldamers, polyproline hairpins, and proto-fluorescent ligands--and the use of these molecules to answer biological questions that would otherwise be nearly impossible to address. Current topics include the use of miniature proteins to identify the functional role of discrete protein-protein interactions and rewire cellular circuits, the use of cell permeable molecules to image misfolded proteins or protein interactions in live cells, and the design of protein-like assemblies of ß-peptides that are entirely devoid of -amino acids.
For a description of early work in the Schepartz lab, click here.
Miniature Proteins. Our laboratory pioneered an approach to the design of very small yet well-folded proteins that inhibit or promote protein-protein interactions with exceptional levels of specificity. Our approach is referred to as protein grafting, and the molecules that result are called miniature proteins--miniature because they contain fewer than 40 amino acids, proteins because they often fold cooperatively.
The very first miniature protein --
an extraordinarily selective DNA binding molecule known as p007--was designed by Jason Chin and reported in 2001 . Since that time we have described miniature proteins that recognize a vast array of DNA and protein targets. These targets include homeobox DNA , the Bcl-2 protein hetero-dimerization interface , the polyproline helix binding cleft of the Mena EVH1 domain , the CREB binding site on CBP , and the p53 binding site on hDM2 . Of particular note are a pair of miniature proteins developed by Jason Chin and Anja Gemperli, respectively, that bind Bcl-2 proteins in a paralog selective manner and promote apoptosis ; a paralog- selective EVH1-domain binding miniature protein designed by Dasantila Golemi-Kotra that causes a unique defect in L. monocytogenes motility (viewable below) ; and a miniature protein-small molecule conjugate synthesized by Tanya Schneider that functions as an isoform-selective protein kinase inhibitor . Most recently, Crystal Zellefrow and Joshua Kritzer reported miniature proteins that selectively activate Scr-family kinases .
Our current work is focused on broadening our repertoire of paralog-selective miniature protein ligands for SH3 and EVH1 domains. We are especially keen on a pair of projects in which miniature proteins are substituted for promiscuous signaling domains within full-length proteins to impose specificity on protein-protein interactions (and the pathways they modulate) inside the cell. We are also actively pursuing a very exciting structure-based approach to encode the important property of cell permaeability within the miniature protein structure. Doug Daniels recently reported our first exciting contibution in this area ; Betsy Smith is about to report the second (so check back soon!).
Foldamers. The foldamers our laboratory studies are short oligomers of ß3-amino acids commonly called ß-peptides. Although previous work demonstrated that short ß-peptides could assemble into 14-helical secondary structures in organic solvents, their application as tools in chemical biology or as therapeutic leads was limited severely by low solubility and, more critically, by the general absence of a well-organized secondary structure in aqueous solution.
In 2003, we reported that the 14-helicity of a ß-peptide in aqueous solution could be enhanced dramatically by acknowledging the direction of the 14-helix macrodipole and that ß-peptide helicity would tolerate the introduction of diverse proteinogenic side chains . More recently, we reported the very first 14-helical ß-peptide inhibitor of a protein-protein interaction , determined the structure of the ß-peptide ligand using high-resolution NMR , and improved affinity using the first one-bead-one-ß-peptide (OBOß) combinatorial library . Structural analysis of these ligand indicates that high affiinity demands an imperfect 14-helix characterized by unwinding along the helix axis.
We have also applied our design strategy to generate a ß-peptide ligand for HIV gp41 that functions as a potent inhibitor of cell-cell fusion . The molecule we describe is protease-stable and one-third the size of the peptidic gp41 inhibitor Fuzeon™,
a drug currently approved in the U.S. for use against HIV.
Our newest focus in the ß-peptide area bears on the fundamental question of why natural proteins are composed solely of natural -amino acids. Jade Qiu reported that judiciously designed ß-peptides can spontaneously self-assemble in aqueous solution into cooperatively folded bundles . Most recently, Doug Daniels determined the high-resolution structure of one such molecule using x-ray crystallography . James Petersson has shown that its thermodynamic and kinetic properties are quantitatively similar to natural -proteins of comparable size . Most recently, James Petersson has demonstrated that the parallel ß-peptide dimer unit embodied in the octameric Zwit-1F bundle can be converted into a linear, antiparallel single chain peptide . The molecule James describes is an important step toward protein-like ß-peptides, and the apparent ease of flipping the 14-helical epitope may soon enable the design of even more complex geometries.
Proto-fluorescent probes: new additions to the cell biology toolkit. There is considerable current interest in the design of reagents capable of monitoring the location, activity, or associations of proteins inside the cell. The tool historically used for this purpose, the Green Fluorescent Protein (GFP) and its many derivatives, suffer from several drawbacks including bulk (GFP contains 238 amino acids), a tendency to aggregate, and unpredictable effects on protein activity or localization. Nathan Luedtke, Dan Fried, and Rachel Dexter recently reported that recombinant proteins containing two cysteine pairs located distal in primary sequence but proximal in the native folded state are labeled selectively in vitro and in live mammalian cells using the pro-fluorescent biarsenical reagents FlAsH-EDT2 and ReAsH-EDT2 . This strategy, which we refer to as bipartite tetracysteine display, enables the detection of protein-protein interactions and alternative protein conformations in live cells. Bipartite tetracysteine display may provide a means to detect early protein misfolding events associated with Alzheimer's and Parkinson's disease and cystic fibrosis, and enable high-throughput screening of compounds that stabilize discrete protein folds.