Our research is focused on the mechanisms of selective protein degradation and the structure/function relationships of the ATP-dependent proteases responsible for intracellular protein degradation. The Lon and Clp proteases are found in all organisms, where they help regulate the levels of important proteins and contribute to protein quality control pathways. ATP-dependent proteases are high molecular weight complexes of a molecular chaperone tightly associated with a protease. Electron microscopy of ClpAP and ClpXP has provided a structural model that serves as a paradigm for other ATP-dependent proteases. Clp proteases have 2 seven-membered rings of ClpP flanked on each side by a six-membered ring of either ClpA or ClpX. The proteolytic active sites are located in a large aqueous chamber enclosed by the rings of ClpP. The ClpA subunits enclose another aqueous chamber which may be the site where unfolded proteins are sequestered prior to translocation to the proteolytic chamber of ClpP. Electron microscopic images of substrate complexes during translocation confirm the model derived from kinetic studies. Substrates migrate from a binding site on the apical surface of the ATPase to a position over an axial channel, and thereafter are transferred to the interior of the complex. For ClpAP, some substrate can be seen within the interior chamber of ClpA, and the remainder accumulates within ClpP. For ClpX, substrate is seen either at the apical surface or within ClpP, implying that translocation is a rapid and concerted process. By slowing translocation, we were able to show that proteins are translocated from only one side of the ClpXP complex at a time, indicating that the ends of the complex are in communication and regulate a reciprocating mechanism of translocation. Limited proteolysis has shown that ClpX folds into three domains, an N-terminal domain that , which can be removed from the protein without disrupting the remainder of the holoenzyme complex, and two sub-sections of the ATPase domain analogous to those found in all AAA family members. The N-terminus may have a role in promoting ATP-dependent unfolding or translocation substrate but is not required for activating degradation of small proteins or peptides. Limited proteolysis also identified a removable N-terminal domain in Lon protease and similar residual non-ATP-dependent activity after truncation. Over-expression of specific N-terminal fragments of Lon interferes with Lon-dependent degradation in vivo, implying that the N-terminus may have a role in binding of substrates. Biochemical studies have shown that ClpX can unfold a stable protein as long as the protein contains an accessible motif recognized by ClpX. Proteins recognized by ClpX bind more tightly when they are unfolded, implying that ClpX can also interact with unfolded regions of proteins, but in general ClpX does not have high affinity for unfolded proteins without some recognition motif. Binding of unfolded proteins to both ClpX and ClpA occurs when a non-hydrolyzable analog of ATP is present, but ATP hydrolysis promotes the release of bound proteins (studies conducted in collaboration with S. Wickner, NCI). Human ClpP has been expressed and purified. The crystal structure of the protein shows that hClpP folds in a virtually identical manner as E. coli ClpP. The human enzyme shows different specificity towards peptide substrates. Interesting, the human ClpP can be activated by and can target specific substrates recognized by E. coli ClpX, providing a clear demonstration that the specificity of degradation by ATP-dependent proteases resides in the associated ATPase. Studies are underway to isolate the human ClpX protein and to obtain mutants of ClpP that can be used to inhibit endogenous activity in vivo.
