ReviewProteasomes and their associated ATPases: A destructive combination
Introduction
A fundamental feature of protein breakdown in eukaryotic and prokaryotic cells is its requirement for ATP (Goldberg and St. John, 1976). Much of our current knowledge about intracellular proteolysis came from studies seeking to understand the biochemical basis of this surprising requirement (Ciechanover, 2005, Goldberg, 2005). The key early developments were the discovery of a soluble (nonlysosomal) ATP-dependent proteolytic system in reticulocytes (Etlinger and Goldberg, 1977) followed by the establishment of similar energy-dependent proteolytic systems in extracts of Escherichia coli (Murakami et al., 1979). Analysis of these bacterial systems led to the discovery of large ATP-dependent proteolytic complexes that degrade proteins and ATP in linked processes (Chung and Goldberg, 1981, Gottesman, 1996).
In eukaryotes, ATP is required both for ubiquitin conjugation to substrates and for the function of the 26S proteasome, the ATP-dependent complex that catalyzes the breakdown of ubiquitinated and certain non-ubiquitinated polypeptides (Ciechanover, 2005, Goldberg, 2005, Voges et al., 1999). The discovery of the first ATP-dependent protease in bacteria (lon/La) (Chung and Goldberg, 1981) was made about the same time as the classic discovery of the role of ubiquitin in protein breakdown in the reticulocyte system by Hershko, Ciechanover, and Rose (Ciechanover, 2005, Glickman and Ciechanover, 2002). The energy-requirement for ubiquitin conjugation in eukaryotes was thought to explain the ATP requirements for intracellular proteolysis in eukaryotes. Thus, initially it was believed that there are two very different explanations for the ATP requirements for proteolysis in prokaryotes and eukaryotes. However, after further study, it became clear that after ubiquitination, ATP was still required for breakdown of the protein (Tanaka et al., 1983), and by the late 1980s, the 26S proteasome was identified as the ATP-dependent proteolytic complex that degrades ubiquitinated proteins (Hough et al., 1987, Waxman et al., 1987).
As interest in the eukaryotic 20S proteasome developed, archaea were found to contain a simpler, but structurally markedly similar proteolytic complex in Thermoplasma acidophilum (Baumeister et al., 1998, Dahlmann et al., 1989, Voges et al., 1999). Further work also uncovered the existence of an ATPase complex, PAN, which functions together with the archaeal 20S proteasome (Benaroudj and Goldberg, 2000, Smith et al., 2005, Zwickl et al., 1999). Thus, protein breakdown in archaea, bacteria and eukaryotes is catalyzed by large proteolytic complexes that hydrolyze ATP and protein in linked reactions. Interestingly, PAN is not found in all archaea (e.g., T. acidophilum). However, it appears likely that all archaea contain ATPase ring complexes of the AAA family that may also function in protein degradation by the proteasome. For example, VAT, which is found in T. acidophilum, seems likely to function in substrate recognition, unfolding, and translocation of substrates into the 20S proteasome (Gerega et al., 2005).
The ATP-dependent 26S proteasome is composed of one or two 19S regulatory complexes and the central 20S particle (Voges et al., 1999, Zwickl et al., 1999), which is a hollow cylinder, within which proteolysis occurs. The two outer α rings and two inner β rings of the 20S particles are each composed of seven distinct but homologous subunits. In eukaryotes, three of the β subunits contain proteolytic sites, which are sequestered in the hollow interior of the 20S particle (Groll et al., 1997). Substrates enter the 20S through a narrow channel formed by the α subunits, whose N-termini, depending on their conformation, can either obstruct or allow substrate entry and thus function as a gate (Groll et al., 2000, Groll and Huber, 2003). This entry channel is narrow and only permits passage of unfolded, linearized polypeptides (Groll et al., 1997). The 19S regulatory complex is composed of two subcomplexes, the lid, which seems to bind and disassemble the ubiquitin-conjugated substrate, and the base, which contains six homologous ATPase subunits (termed Rpt1–6 in yeast) plus two non-ATPases, Rpn 1 and 2 (Voges et al., 1999). These ATPases are members of the AAA family of ATPases (Patel and Latterich, 1998). For a globular protein to be degraded, it must associate with the 19S ATPases and undergo ATP-dependent unfolding followed by translocation into the 20S particle, which requires opening of the gate in the α ring (Kohler et al., 2001). Each of these steps is regulated in some way by the ATPase complex.
The ATPase complexes that regulate protein degradation in eukaryotes, bacteria and archaea are all members of the AAA+ (ATPases Associated with various cellular Activities) ATPase superfamily (for review see Ogura and Tanaka, 2003). The AAA+ family of ATPases are found in all living organisms and in all cell compartments, where they participate in a variety of essential cellular processes such as mitosis, protein folding and translocation, DNA replication and repair, membrane fusion and proteolysis. They are characterized by the presence of one or two conserved ATP-binding domains (200–250 residues), called the AAA motif, consisting of a Walker A and a Walker B motif (Confalonieri and Duguet, 1995). The eukaryotic and archaeal (PAN) proteasomal ATPases belong to a subfamily of AAA+ ATPases (AAA family) that contains an additional motif called the second region of homology (SRH) (Lupas and Martin, 2002). Despite the large variety of cellular processes in which AAA+ ATPases participate, they have some common features. A recurrent structural feature of most AAA+ ATPases is their assembly into oligomeric (generally hexameric) ring-shaped structures with a central pore. In addition, most appear to be involved in protein folding or unfolding, and assembly or disassembly of protein complexes through nucleotide-dependent conformational changes. Thus, recent insights into the functioning of the archaeal PAN complex and the 19S proteasomal regulatory ATPase may illuminate the functioning of these other AAA Family members (and vise versa).
Though bacteria do not contain 20S proteasomes, like those in eukaryotes, they do contain several large compartmentalized protease complexes that associate with AAA ATPase complexes such as HslUV and ClpAP. HslV is a two-ring peptidase complex which shares homology with the beta subunits of the 20S proteasome (Bochtler et al., 2000), and forms a six-membered ring (Rohrwild et al., 1997) rather than the seven-membered ring, which is characteristic of the 20S proteasome. HslU, the ATPase complex, associates with HslV to stimulate protein degradation, and is homologous to PAN. X-ray diffraction studies established that HslU induces conformational changes in the peptidase active site of HslV upon association and increases the pore size of HslV. Thus, HslU increases the peptidase activity of HslV by allosteric activation and probably also by promoting substrate unfolding for peptide entry (Huang and Goldberg, 1997, Sousa et al., 2000, Wang et al., 2001, Yoo et al., 1997). Facilitating peptide entry thus appears to be a common property shared by HslU, PAN, and the 19S ATPases, although HslV does not contain an outer α ring or gating termini like those in the 20S proteasome.
Section snippets
The function of the 26S ATPases
Studying the ATP-dependent processes and the mechanisms of protein breakdown within the 26S proteasome has proven difficult because of its structural complexity, multiple enzymatic activities and ubiquitin requirement. Nevertheless, several important discoveries about these ATPases have been made using genetic tools in yeast. Through systematic mutagenesis of the ATP binding sites in each of the six different ATPases, Finley and co-workers showed that these 19S ATPases (Rpt1–6) perform distinct
The PAN ATPase complex from archaea
The first complete genome sequence in the domain of archaea was from Methanococcus jannaschii. This sequencing revealed a gene (S4) which was highly homologous to the genes encoding the 19S ATPases (Bult et al., 1996). To test if this gene product might regulate the 20S proteasome, the S4 gene was expressed in E. coli, and the 50kDa product, named PAN (proteasome-activating nucleotidase), was purified and characterized by Zwickl et al. (1999). PAN’s sequence contains several hallmarks of the
Structure and association of PAN with the 20S particle
Although the association of an ATPase chaperone-like complex with a proteolytic particle appears to be a common feature of several ATP-dependent systems for intracellular protein degradation (the 26S proteasome and the bacterial ClpAP, ClpXP and HslUV complexes), an association between PAN and the 20S particle was difficult to observe by typical biochemical approaches, even when PAN and the 20S came from the same species. However, when PAN is mixed with archaeal 20S proteasomes and ATP, it
PAN regulates gate opening
Because of the tight interaction between the 20S proteasomes α and β subunits, substrates can enter only through the 20S pore at either end of the particle (Baumeister et al., 1998). The elegant X-ray analysis of M. Groll et al. showed that this channel is gated by the N-termini of the α subunits (Groll et al., 1997). These N-termini in eukaryotic proteasomes can assume either of two ordered structures, an open conformation and a closed one, both of which require the YDR motif for stabilization
The mechanism of the PAN–20S association and gate opening
These studies have established that ATP-binding to PAN is essential for its association with the 20S particle and for triggering gate opening in the 20S (Smith et al., 2005). A very different (non-homologous) type of proteasomal regulator, the 11S, PA28 (REG) complex and its invertebrate homolog, PA26, also bind to the α-ring (independent of ATP) and induce gate opening. However, these complexes stimulate peptide but not protein entry. Their association with the 20S requires their extreme
The energy requirements for protein unfolding and translocation
It has been clear since the early seventies that protein degradation in prokaryotes and eukaryotes requires ATP (Ciechanover, 2005, Goldberg, 2005, Goldberg and St. John, 1976). However, it is still unclear which of the multiple steps in the process of protein degradation requires ATP and how nucleotide binding or hydrolysis enhance these steps. In particular, determining the mechanisms whereby the 19S regulatory particle unfolds substrates and facilitates their entry into the 20S proteolytic
Conclusion
A more complete understanding of the molecular mechanisms involved in protein degradation by the proteasome will be aided by detailed structural information about the ATP and ADP-bound forms of ATPase complex, as well as the delineation of its ATP hydrolysis cycle. Presumably X-ray crystallography will first be achieved for PAN, whose many benefits for study have been summarized here. Already however, study of the PAN:20S complex has allowed us to learn much concerning the multiple steps in
References (78)
- et al.
Processive degradation of proteins and other catalytic properties of the proteasome from Thermoplasma acidophilum
J. Biol. Chem.
(1997) - et al.
Proteasome disassembly and downregulation is correlated with viability during stationary phase
Curr. Biol.
(2003) - et al.
A ratchet mechanism of transcription elongation and its control
Cell
(2005) - et al.
The proteasome: paradigm of a self-compartmentalizing protease
Cell
(1998) - et al.
The unfolding of substrates and ubiquitin-independent protein degradation by proteasomes
Biochimie
(2001) - et al.
ATP hydrolysis by the proteasome regulatory complex PAN serves multiple functions in protein degradation
Mol. Cell
(2003) - et al.
The multicatalytic proteinase (prosome) is ubiquitous from eukaryotes to archaebacteria
FEBS Lett.
(1989) - et al.
Regulatory subunit interactions of the 26S proteasome, a complex problem [Review]
Trends Biochem. Sci.
(2000) - et al.
The 1.9 Å structure of a proteasome-11S activator complex and implications for proteasome-PAN/PA700 interactions
Mol. Cell
(2005) - et al.
VAT, the thermoplasma homolog of mammalian p97/VCP, is an N domain-regulated protein unfoldase
J. Biol. Chem.
(2005)