PAN complex

Abstract: Energy-dependent proteolysis is not only vital to elimination of defective cellular proteins but is also central to regulation of cell division, metabolism, transcription, and other essential cellular functions (12, 22). A small group of related energy-dependent proteases have been identified from several diverse organisms; these proteases include Lon, FtsH (HflB), ClpAP, ClpXP, HslUV (ClpYQ), and the 26S proteasome (4, 39). Although this group of proteins shares limited primary sequence identity, the proteases have converged to a universal structure in which relatively nonspecific proteolytic active sites are compartmentalized from the cell by narrow openings (37, 67). The current model is that the degradation of biological substrates by this group of proteases requires additional energy-dependent proteins or protein domains for the recognition and/or unfolding of substrate proteins (21, 29, 65). Some of these energy-dependent components may not only participate in proteolysis but also have independent roles as chaperones (60). Thus, the energy-dependent component of these proteases may provide a proofreading step following the initial binding of protein substrates and enable the cell to distinguish between proteins destined for refolding, disaggregation, or destruction (23). Some energy-dependent proteases are organized in a symmetry mismatch, including ClpAP and ClpXP which are complexes of the hexameric ClpA and ClpX ATPases interfaced with the ClpP protease of two heptameric rings (5). It is postulated that hydrolysis of nucleotides by the energy-dependent component results in rotation at the ATPase-protease interface (5) and that this rotation facilitates the processive degradation of substrate proteins (63). This mismatch, however, does not appear to be universal, since the fully functional Lon and FtsH proteases assemble into homomultimeric complexes with the ATPase and proteolytic components encoded by a single gene (10, 47). Energy-dependent proteases, including Lon and proteasome complexes, are predicted from the genome sequences of archaea (8, 32, 34, 57). Little is known, however, about the biochemistry or biological significance of these energy-dependent proteolytic pathways in this unusual group of organisms. 20S proteasomes have been purified from diverse archaea (3, 13, 42, 74) and appear to have a self-compartmentalized structure which requires polypeptides to be at least partially unfolded prior to hydrolysis (72). This suggests that the archaeal 20S component alone, as purified, has little biological significance in native protein hydrolysis and suggests that additional components are necessary for protein degradation at the optimal growth temperature of these organisms. The related eucaryal 20S proteasome is fully functional in degrading proteins but only when associated with an energy-dependent 19S cap regulatory complex or an eight-subunit derivative of this complex, designated the base domain (21). Six of these base domain subunits are ATPases which are members of the AAA family (named AAA for ATPases associated with a variety of cellular activities) (46) and have recently been named Rpt proteins (for regulatory particle triple-A) (16). It is postulated that these energy-dependent subunits are responsible for the unfolding and/or translocation of substrate proteins into the central proteolytic chamber of the 20S proteasome. Closely related homologs of the eucaryal ATPase proteasome subunits are predicted from the genome sequences of the archaea (8, 32, 34, 57). One homolog from Methanococcus jannaschii (MJ1176) has recently been purified from recombinant Escherichia coli and named PAN for proteasome-activating nucleotidase (84). This protein was purified as a 650-kDa complex of an N-terminal polyhistidine-tagged form of PAN in association with a derivative of PAN which had a deletion of residues 1 to 73 [PAN(Δ1–73)]. The purified PAN complex had ATPase activity and activated the energy-dependent degradation of proteins by 20S proteasomes from Methanosarcina thermophila and Thermoplasma acidophilum by four- to ninefold (84). Characteristic of AAA proteins, the deduced sequence of PAN reveals a highly conserved Walker A motif (211-GPPGTGKT-218) which is predicted to be involved in coordination of Mg2+ and formation of hydrogen bonds with nucleotide triphosphates including the β- and γ-phosphates (54, 66). A modified Walker B motif or “DEAD” box which is also predicted to be involved in Mg2+ binding and ATP hydrolysis is conserved and includes residues 270-DEID-273 of PAN (9). Additionally, a SRH or second region of homology motif [(T/S)-(N/S)-X5-D-X-A-X2-R-X2-R-X-(D/E)] which distinguishes AAA proteins from the broader family of Walker-type ATPases is also conserved and spans residues 315 to 333 of PAN. The SRH motif appears to be important in ATP hydrolysis but not ATP binding (31). The PAN sequence also has a highly charged N-terminal coiled-coil spanning residues 49 to 83 which is conserved in other AAA proteins (84). This coiled-coil may play an important role in regulating nucleotide hydrolysis, protein-protein interaction, and/or other activities as proposed for other AAA proteins including the ATPase subunits of the 26S proteasome as well as ClpA and ClpB (38, 51, 56, 68). In this communication, the biochemical and physical properties of the 20S proteasome and PAN proteins of M. jannaschii are presented. Our results demonstrate that this 20S proteasome requires a nucleotidase complex such as PAN to mediate the energy-dependent hydrolysis of folded-substrate proteins. The N-terminal coiled-coil region of PAN is not required for this reaction and is not needed for association of the PAN protein into an ∼12-subunit complex. However, deletion of the N-terminal 73 residues does appear to influence the biochemical properties of nucleotide hydrolysis mediated by the PAN protein.