Our research is focused on the mechanisms of selective protein degradation and the structure/function relationships of the ATP-dependent Lon and Clp proteases. Lon and Clp are found in all microorganisms and in the organelles of eukaryotes, including mammals, where they help regulate the levels of key proteins involved in cellular control and help maintain protein quality control by targeting misfolded proteins for degradation. Lon and Clp are multi-domain or multi-component assemblies comprising a chaperone component tightly associated on either side of a two-fold symmetric protease complex. The chaperone (ClpA or ClpX or the chaperone domain of Lon) recognizes specific motifs in selective targets for intracellular regulation and general motifs characteristic of unfolded or misfolded proteins. The chaperone unfolds proteins and thereby dissolves aggregates, disassemble complexes, or untangle misfolded proteins so they can either be released and allowed to refold or be translocated to the associated protease for degradation. To protect other proteins from unwanted damage, the active sites of the protease (ClpP or the protease domain of Lon) are sequestered in an internal aqueous chamber that is accessible by narrow axial channels through the protease, which has its subunits arranged six- or seven-membered rings. In the past year our laboratory has made progress in four major areas. First, we have (in collaboration with Dr. Di Xia, Laboratory of Cell Biology, NCI, continued to determine a high-resolution crystal structure for ClpA and the N-terminal domain of ClpA. ClpA has two ATPase domains, both of which have folds characteristic of the AAA super-family of proteins, a diverse group of energy-dependent molecular machines with various unfolding and disassembly activities in all organisms. The structure has provided insights into interactions of ligand-binding and catalytic residues and revealed a unique domain organization in which the D1 and D2 ATPase domains are interdigitated around the hexameric ring of ClpA. The positions of function impairing mutations have been mapped onto the 3D structure, which is being used as a template for the design of additional mutations to probe functional regions of ClpA and its mechanism of action. The structure of the N-terminal domain has revealed a novel fold that produces interaction surfaces for substrates and for an adaptor protein that modulates ClpA activity in vivo. The N-domain has also been found to have a binding site for a zinc ion and studies are underway to determine the function of the metal ion in ClpA substrate binding and unfolding. We have also obtained the structure of a complex between an adaptor protein, ClpS, and the N-domain, which reveals a unique role for the N-terminal 17 amino acids of ClpS in blocking substrate binding to ClpA. Second, a combination of kinetic biochemical studies and electron microscopy has shown that a rate-limiting step (substrate binding or unfolding) precedes the translocation step in which proteins are transferred from the apical surface of the complex to the degradation chamber of ClpP. Our data lead to a model in which substrates are primed on one side of the symmetrical complex while translocation occurs from the other side, resulting in an alternating mode of unfolding and translocation from the two sides of the complex. Third, we have used peptides corresponding to degradation motifs recognized by ClpA and ClpX to study the number and nature of protein recognition sites. Titration calorimetry and fluorescence titrations have shown that a single peptide binds with high affinity to a hexamer of ClpA or ClpX. This unusual stoichiometry suggests either that the binding sites in the subunits overlap so that only one can be occupied at a time or that there is strong negative cooperativity in binding peptides, such that a peptide produces a conformational change that decreases the binding affinity of sites on adjacent subunits. The peptides can be cross-linked to the chaperone and a site within the D1 domain of ClpA has been indicated as the site of binding. Specific identification of the site is underway. Global information about substrate binding and recognition has been obtained from in vivo pull-down assays. By overexpressing mutants of ClpP that do not degraded proteins, substrates can be trapped and identified by mass spectroscopy. These studies are going on in collaboration with Donald Hunt (U. Virginia) and have revealed a number of novel substrates. Identification of elements in common among the different substrates or subsets of substrates should provide data on the range of motifs revealed an unexpected effect of hClpX in promoting stability of the double ring structure of ClpP. hClpP rings readily separate in solution. In addition to conformational changes upon ClpX binding, ligands or residues at the N-terminus of ClpP and at the catalytic active site of ClpP have an influence on the stability of the tetradecamer. These data provide the first evidence that changes in ring contacts in ClpP may occur during catalysis and may provide a basis for effects substrate accumulation within the chamber and for a possible mechanism of product release from the chamber following processive degradation. We are nearing completion of an X-ray crystal structure of hClpP which is very similar to that of E. coliClpP but which has a different configuration of the N-terminus. Our structure suggests that the N-terminus may be in position to serve a gating function analogous to that described for the N-terminal extension of the proteasome beta subunits. We have developed an in vivo assay for human ClpP in E. coli cells and are exploiting it to identify mutations that affect activity and interaction with ClpX and are in the process of extending the E. coli model system to include human ClpX, which will allow the facile isolation of function altering mutations in ClpX.
