Molecular Biology and Evolution 18:962-974 (2001)
© 2001 Society for Molecular Biology and Evolution
ARTICLE |
Evolution of Proteasomal ATPases
Department of Genetics, North Carolina State University
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsliterature cited |
|---|
In eukaryotic cells, the majority of proteins are degraded via the ATP-dependent ubiquitin/26S proteasome pathway. The proteasome is the proteolytic component of the pathway. It is a very large complex with a mass of around 2.5 MDa, consisting of at least 62 proteins encoded by 31 genes. The eukaryotic proteasome has evolved from a simpler archaebacterial form, similar in structure but containing only three different peptides. One of these peptides is an ATPase belonging to the AAA (Triple-A) family of ATPases. Gene duplication and diversification has resulted in six paralogous ATPases being present in the eukaryotic proteasome. While sequence analysis studies clearly show that the six eukaryotic proteasomal ATPases have evolved from the single archaebacterial proteasomal ATPase, the deep node structures of the phylogenetic constructions lack resolution. Incorporating physical data to provide support for alternative phylogenetic hypotheses, we have constructed a model of a possible evolutionary history of the proteasomal ATPases.
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsliterature cited |
|---|
Proteins have been described as "molecular fossils," the study of which may unravel the history and mechanism of evolution. As such, the molecular phylogenetics of proteins and families of proteins is receiving much attention. One specific area is the coevolution of proteins within multiprotein complexes. Such studies not only promise to expand our understanding of protein evolution and coevolution, but should also prove to be powerful in the construction of the phylogenetic "Tree of Life." One suitable eukaryotic complex, whose analysis can also be extended back into the Archaea, is the proteasome.
Structure and Function of the Proteasome
Within eukaryotic cells, proteins undergoing controlled, energy-dependent proteolysis are polyubiquitinylated and targeted to the proteasome for cleavage into small peptides. This ubiquitin/proteasome pathway degrades the majority of cellular proteins, while the proteasome itself can comprise up to 1% of the total soluble protein in metabolically active cells (Hendil 1988
; Tanaka et al. 1992; Rock et al. 1994
; Lee and Goldberg 1996
; Ciechanover, Orian, and Schwartz 2000
). The 26S proteasome is a very large complex with a mass of approximately 2,600 kDa containing at least 62 protein subunits encoded by 31 genes. This "holoproteasome" can be separated into two subparticles, the 20S core and the regulatory particle (RP) (also known as PA700, 19S Cap particle, or µ) (Glickman et al. 1988b
; Voges, Zwickl, and Baumeister 1999
).
The 20S core proteasome has a hollow barrel structure consisting of four stacked rings in the order 
ßß, with each ring consisting of seven subunits each (fig. 1
) (Lowe et al. 1995
; Groll et al. 1997
; Gerards et al. 1998
). The particle is self-compartmentalized, with the catalytic sites, formed by ß-subunits, sequestered within the central cavity of the "barrel," rendering the isolated particle inactive (Lupas et al. 1997
). In vitro 20S proteasomal activity against peptide substrates and casein can be stimulated by a number of small molecules (e.g., low concentrations of SDS); however, to function in vivo, the 20S proteasome needs to associate with additional proteins (Adams et al. 1997
; Groll et al. 2000
). In higher eukaryotes, there are two activating complexes: (RP) and PA28 (also known as the 11S particle or REG). Recent investigations have indicated that in mammalian cells, the proteasomal population consists of a mix of [RP-20S-RP], [RP-20S-PA28], and [PA28-20S-PA28] (Tanahashi et al. 2000).
|
The 20S/PA28 form of the proteasome is involved in the production and presentation of immunogenic peptides in vertebrates with an adaptive immune system (immunosomes) (Dick et al. 1996
; Groetrupp et al. 1996a, 1996b
). PA28 consists of a single hexameric ring structure composed of two related proteins (PA28 and PA28ß), with a total mass of about 200 kDa. It binds to the -rings at the ends of the 20S barrel and stimulates the degradation of peptides but not the ubiquitin-dependent degradation of proteins (Chu-Ping et al. 1994
; Gray, Slaughter, and DeMartino 1994
). Homologs of the PA28 proteins have been identified neither in yeast nor in plants. In contrast, the RP is remarkably similar in all eukaryotes from yeast to humans. The yeast RP is the best characterized and consists of at least 17 proteins with masses of 22–110 kDa, giving a total mass of 900 kDa (Glickman et al. 1998b
; Rubin et al. 1998
; Verma et al. 2000
). The individual protein subunits are designated Rpn1–Rpn12 based on apparent size during denaturing polyacrylamide gel electrophoresis (SDS-PAGE). Six additional subunits are designated Rpt1–Rpt6. These proteins are a set of highly homologous ATPases that are each essential for cell viability (Rubin et al. 1998
).
While the structure of the yeast 20S proteasomal core particle is known at 2.4 Å resolution through crystallographic studies, the structure of the RP is much less clearly defined (Groll et al. 1997
). Electron microscopy has provided a low-resolution structure of the 26S proteasome but no information about the positioning of individual subunits (Walz et al. 1998
). In contrast, purification studies of mutant yeast proteasomes have localized RP subunits into two subcomplexes (Glickman et al. 1998a
). One (the base) contains the six related ATPases and two additional proteins (Rpn1 and Rpn2) and can bind independently to the 20S core. The other subcomplex (the lid) contains Rpn3–Rpn9, as well as Rpn11 and Rpn12 (fig. 1
). Rpn10 is thought to provide a "bridge" between the base and the lid. The lid is positioned distal to the 20S core.
It is hypothesized that the six regulatory ATPases associate into a heterohexameric ring that forms the interface between the RP and the 20S core particle. While no definitive evidence (e.g., crystallographic structure) exists, other supporting evidence is available: (1) The six proteasomal ATPases belong to a very large family of proteins known as the AAA proteins (see below) (Beyer 1997
; Swaffield and Purugganan 1997
). A number of the members of this family of proteins form hexameric structures, e.g., NSF/Sec18, VCP, CDC48, and ARC (Peters, Walsh, and Franke 1990
; Frohlich et al. 1995
; Lenzen et al. 1998
; Wolf et al. 1998
). (2) The RP base particle, consisting of the six ATPases, Rpn1, and Rpn2, associates with the 20S core and stimulates peptidase and caseinase activity (Glickman et al. 1998a
). (3) The best structural data available for the 26S proteasome (low-resolution scanning electron microscopy) reveals a hexameric structure at the interface of the 20S and regulatory particles (Walz et al. 1998
).
The proteasomal ATPases belong to the Triple-A family of ATPases (AAA ATPases; ATPases Associated with a variety of cellular Activities) (Confalonieri and Duguet 1995
; Patel and Latterich 1998
; Zwickl and Baumeister 1999
; Vale 2000
). These proteins are found in multiple forms in all extant life. Initial homology searches of genomic databases revealed that yeast contains 22 such proteins, with approximately 27 in Caenorhabditis elegans and possibly 30 in Drosophila melanogaster. The most recent analysis of the family using statistical criteria of alignment has expanded the family such that the "AAA+" family has approximately 50 members in yeast (Neuwald et al. 1999
). AAA+ proteins all possess a conserved ATPase domain (CAD or AAA cassette)—a Walker type ATPase motif embedded within a much larger region (220–250 amino acids) of high homology (Walker et al. 1982
; Beyer 1997
; Swaffield and Purugganan 1997
; Neuwald et al. 1999
). The exact biological role performed by the CAD has been a matter of debate because of the extremely wide range of cellular activities AAA+ proteins exhibit, but it is now thought to be that of manipulation of protein structure via nucleotide-dependent conformation changes, enabling the assembly, operation, and disassembly of numerous protein "machines." Such manipulations are the result of chaperonin-like abilities and include unfolding of proteins prior to degradation; chaparonin foldase/unfoldase activities, DNA replication, transcription, recombination, and various "activase" operations (Sollner et al. 1993
; Suzuki et al. 1997
; Golbik et al. 1999
; Leonhard et al. 1999
; Neuwald et al. 1999
; Schtilerman, Lorimer, and Englander 1999
; Weber-Ban et al. 1999
; Zwickl and Baumeister 1999
; Strickland et al. 2000
).
Evolutionary Significance of the Proteasome
The proteasome may serve as a paradigm for the study of the evolution of very large protein complexes. Proteasomes are found in all eukaryotes and archaebacteria (Dahlman et al. 1989
). The ancestral proteasome present in the archaebacteria is a much simpler structure than that found in eukaryotes. While the core 20S particle still consists of an almost identical barrel structure composed of four seven-membered rings, it is composed of only two peptides—one alpha-subunit and one beta-subunit (Lowe et al. 1995
). Associated with this core particle appears to be a homohexameric ATPase ring, known as PAN (Zwickl et al. 1999
). Thus, the archael proteasome is composed of only three different peptides (fig. 1
). In all eukaryotes for which data are available, as noted above, the single alpha- and beta-subunits have multiplied to at least seven proteins each, the single ATPase has become six, and an additional 11 proteins have been "recruited" to the 19S RP. Thus, the eukaryotic proteasome is composed of at least 62 proteins encoded by 31 genes (fig. 1
). This multiplication of subunits within a conserved structure seems to be a common theme. Archibald, Logsdon, and Doolittle (1999)
have published a recent review on the recurrent paralogy within an archael chaperonin (the thermosome, which may contain one, two, or three different subunits), while the eukaryotic chaperonin CCT is completely hetero-oligomeric (Liou and Willison 1997
).
Thus, the study of any of these three aspects of proteasomal structure (20S component divergence, RP ATPase divergence, RP lid acquisition/COP9 signalsome divergence), separately or in toto, will shed valuable light on the evolution of large protein complexes. Hughes (1997)
has conducted an initial study on the core 20S components. This study is concerned with the evolutionary divergence of the RP ATPases.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsliterature cited |
|---|
Alignment and Phylogenetic Tree Construction
We previously carried out a phylogenetic study of the proteasomal ATPases as part of a larger study of the AAA ATPase family (Swaffield and Purugganan 1997
). The previous work revealed that the proteasomal ATPases were a monophyletic group within the AAA ATPases that could be divided further into six subgroups. The Rpt1 and Rpt6 subgroups appeared more closely related to each other than to the other groups, and there was moderate support for joining the Rpt2 and Rpt3 subgroups. However, deep-node relationships were unclear. We revisited the problem because of the availability of a more complete array of protein sequences, and we included physical data in our analysis.
An initial alignment of 103 ATPase sequences (table 1
) was performed with the global alignment program CLUSTAL W (Thompson, Higgins, and Gibson 1994
). This alignment was then inspected and manually improved. At the amino terminal of the protein was a region of residues exhibiting relatively high levels of conservation within groups but very little homology between groups. This region corresponded to the nuclear localization signal and coiled-coil region of figure 2
of Fu et al. (1999)
. While an alignment could be forced on the residues in this region, it is doubtful that this alignment would be biologically meaningful. For this reason, the ATPase sequences were truncated to include only the CAD domain (domains B–H and K of Beyer [1997] and intervening residues). Following alignment of the CAD domain sequences, phylogenetic trees for the ATPase proteins were calculated. Trees were based on a representative subset of the total sequences. In addition to eight archaebacterial sequences included as an outgroup, 47 eukaryotic sequences representing the complete complement of ATPases from mammals (Homo sapiens), insects (Drosophila melanogaster), nematodes (C. elegans), fungi (Schizosaccharomyces pombe, Saccharomyces cerevisiae), and plants (Arabidopsis thaliana), together with 14 sequences from "primitive eukaryotes," were used to generate trees. Two different tree-building algorithms were used. First, a parsimony analysis was performed with the PAUP* program (Swofford 2000
). A bootstrap analysis was used to determine levels of support for the interior nodes of the parsimony tree. Two hundred fifty bootstrap replicates were performed, and a full heuristic search (using 10 random-addition replicates each) was performed on each replicate data set. The results of the bootstrap analysis were used to construct a majority-rule consensus tree (fig. 2
).
|
|
Second, a neighbor-joining tree (Saitou and Nei 1987
) was calculated based on maximum-likelihood distances calculated using the program PUZZLE (Strimmer and von Haeseler 1996
). Parameter options chosen included the BLOSUM62 substitution matrix (Henikoff and Henikoff 1992
) and a discrete gamma distribution of substitution rates with eight rate categories with the distribution shape parameter estimated from the data. Default values were used for all other options. The NEIGHBOR program of the PHYLIP package (Felsenstein 1995
) calculated a neighbor-joining tree using the JUMBLE option to randomize the input order of the sequences.
Expression of Rpt4-H6 and Purification of 26S Proteasomes
Proteasomes containing His6-tagged Rpt4 were formed in vivo after yeast strain W303 was transformed with pYES/GS::RPT4 (Invitrogen), in which the yeast RPT4 gene, tagged with six histidines, is expressed from the galactose-inducible GAL 1/10 promoter. 26S proteasomes were isolated by conventional chromatography (unpublished data).
Cross-Linking of Proteasomal ATPase Subunits
Purified 26S proteasomes containing Rpt4-H6 were dialyzed against PBS containing 10% glycerol and 100 µM ATP. Cross-linking reactions with 3,3'-dithio bis(sulfosuccinimidylpropionate) (DTSSP) (Pierce) were carried out in a total volume of 80 µl containing 80 µg purified proteasome (0.38 µM) and 8 µl DMSO containing 0 µM, 380 µM, 1.9 mM, or 9.5 mM DTSSP. This gave molar proteasome : DTSSP ratios of 0, 100x, 500x, and 2,500x. Reactions were incubated at 30°C for 30 min and stopped by the addition of 4.2 µl 1 M Tris (pH 8.0) (final 50 mM Tris) and a further 15 min incubation at 30°C. Rpt4-H6 and covalently cross-linked proteins were bound to 100 µl (50% slurry) Talon Beads (Clontech) in the presence of 650 µl 8 M urea, 0.1% NP40 for 2 h at room temperature with nutation. The beads were washed three times in 1 ml 8 M urea, 0.1% NP40, then resuspended in 130 µl SDS PAGE loading buffer containing 5% (v/v) ß-mercaptoethanol. Twenty-microliter aliquots were loaded onto six identical 10% PAGE SDS gels and run with proteasome standards. Separated proteins were transferred to PVDF membrane in Towbin buffer (25 mM Tris, 192 mM glycine, 0.01% SDS [un pH'd]) and blocked for 1 h with 10% Blotto (10% Carnation nonfat dried milk in TBST—100 mM Tris [pH 7.5], 150 mM NaCl, 0.2% Tween 20). Immunoblots were performed with Rpt subunit–specific antibodies and HRP conjugated secondary antibodies (Pierce) in 1% Blotto and developed with SuperSignal West Pico solution (Pierce).
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsliterature cited |
|---|
Analysis of Phylogenetic Trees
The parsimony analysis returned 72 equally parsimonious trees. The majority-rule consensus (fig. 2 ) suggests that the deep structures of these equally parsimonious trees are identical. Thus, the among-trees variation occurs in the clustering of the terminal branches. The topology of the parsimony consensus tree was not congruent with the topology of the neighbor-joining tree (fig. 3 ). Parsimony resolved two groups, one of four clades and one of two clades. Neighbor-joining placed the Rpt4 clade basal to all of the other ATPases, while in the parsimony trees this clade occupied a descendant position in a sibling relationship with the Rpt1 clade. The remaining sequences in the neighbor-joining tree grouped into two clades, one containing three groups and the other containing two. The groups within these two clades did not match the groups in the two major clades in the parsimony trees. Neither of these topologies matched the tree derived by Fu et al. (1999)
.
|
The parsimony bootstrap analysis indicated that there was no significant support for the internal nodes grouping the major clades of ATPases (fig. 4 ). This is consistent with the neighbor-joining tree, where the internal branches connecting the major clades have relatively short lengths. Similarly, none of the internal nodes connecting the major clades in the analysis of Fu et al. (1999)
had bootstrap support >50%. This pattern of short internal branches supported by relatively few characters is consistent with a scenario of ancient diversification of the six proteasome regulatory ATPase groups. If the six independent clades were established long ago (e.g., before diversification of a eukaryotic ancestor into plants, fungi, and animals), the process of substitution should leave few residues to accurately estimate the true phylogenetic relationships of the six ATPase clades. This problem would only be compounded by structural constraints imposed by interactions between subcomponents, reducing the sequence space available to these genes, and gene correction events, possibly erasing the evolutionary history of the divergence between the duplicated genes. Thus, we cannot determine the true phylogeny of the proteasomal ATPases by analysis of currently available sequences.
|
Physical Relationships Between the Proteasomal ATPases
Next, we considered the physical relationships of the ATPases within the regulatory particle of the proteasome. Assuming the ATPases form a heterohexameric ring at the interface between the 20S core and RP, what is their relative order within the ring? Studies have addressed the question of ATPase/ATPase interaction (reviewed in Ferrell et al. 2000
) with a number of reports of homo- and heterodimerization and oligomerization among the six proteins. Early reports consisted of genetic and two-hybrid data. Gordon et al. (1993)
presented genetic data to suggest an interaction between Rpt1 and Rpt2 in S. pombe and have unpublished data suggesting an interaction between Rpt2 and Rpt6 (quoted in Richmond, Gorbea, and Rechsteiner 1997). Two-hybrid data have indicated interactions between Rpt4 and Rpt6 and Rpt3 and Rpt5 (Ohana et al. 1993
; Russell, Sathyanarayana, and Johnston 1996
). The interaction between Rpt3 and Rpt5 was dependent on N-terminal regions of the proteins containing leucine zipper domains. Five of the six ATPases contain predicted leucine Zipper coiled coil motifs within this region, while the sixth subunit (Rpt2) contains an atypical predicted coiled coil (Richmond, Gorbea, and Rechsteiner 1997
). A more recent analysis of proteasomal ATPase interactions consisted of experiments based on the coexpression of subsets of the ATPases in vitro (Richmond, Gorbea, and Rechsteiner 1997
). These experiments revealed strong pairwise interactions between Rpt2 and Rpt1, Rpt3 and Rpt6, and Rpt4 and Rpt5. In addition, a tetrameric complex consisting of Rpt1–Rpt3 and Rpt6 was also detected. These interactions again were dependant on the N-terminal regions of the ATPases.
This dependence of interaction on the N-terminal coiled-coil regions resulted in the hypothesis that these regions formed the interaction surfaces between these proteins. Recent determination of the crystal structure of part of the NSF protein, however, revealed that in this case, the CAD itself forms the hexamerization domain (Lenzen et al. 1998
). In addition, the yeast endosomal AAA protein Vps4 has an N-terminal coiled-coil essential for interaction with the membrane, while the CAD drives oligomerization (Babst et al. 1998
). If this arrangement is retained by the RP ATPases (as seems likely given the high degree of homology of the CAD and the ability of many AAA proteins to hexamerize), this would suggest that the N-terminal coiled-coil regions are probably involved in interactions with other RP subunits or substrate proteins (Gorbea, Taillandier, and Rechsteiner 2000
). This scenario raises doubts about the biological relevance of some of the ATPase interactions noted above, especially if the N-terminal coiled-coil regions are designed to interact with a large number of substrate proteins and thus may be "sticky." The two-hybrid data are particularly uncertain because of the bias inherent in the system caused by the overproduction of the interacting proteins driving weak, possibly non-biologically-relevant, interactions. One way these results may be resolved, however, is that while the interactive surfaces between the ATPases are formed by the CAD, the N-terminal regions may be involved in initial partner selection.
The possibility of the CADs forming a hexameric ring of RP ATPases is further supported by the presence of a central opening 1.8 nm in diameter in the NSF complex. This hole would lie in register with, and be slightly larger than, the presumed opening in the apical rings of the 20S core, thus providing a continuous route of entry of substrate proteins into the central cavity of the 20S particle for degradation (Groll et al. 2000
). In addition, the diameter of the NSF complex (9.7–11.6 nm) is similar to that of the yeast 20S complex (11.3 nm), consistent with the electronmicrograph images of the 26S proteasome (Groll et al. 1997
).
Using the data of Gordon et al. (1993)
and Richmond, Gorbea, and Rechsteiner (1997)
, a model of proteasomal ATPase organization can be derived (fig. 5a
). As can be seen, uncertainty is focused on the orientation of the Rpt4/5 dimer with respect to the rest of the ATPases. If the two hybrid-data are included, the model depicted in figure 5b
can be derived as put forward by Richmond, Gorbea, and Rechsteiner (1997)
. This is very similar to the organization depicted in a recent review (Ferrell et al. 2000
). However, because of the dependence on N-terminal regions for some of these interactions, we feel that these models should be treated with caution—especially the second, incorporating the two-hybrid data.
|
To try and resolve this problem, we adopted a different strategy. Because of the uncertainties noted above inherent in the in vitro and two-hybrid techniques used, we started to analyze the structure of proteasomes assembled in vivo. Briefly, individual RP subunits were expressed in yeast (strain W303) as His6 tagged versions. The yeast were harvested, protein extracts were made, and the 26S proteasome was purified. The purified proteasomes were then cross-linked with a reversible reagent, DTSSP. Following cross-linking, noncovalent interactions within the proteasome were disrupted with 8 M urea and the His6 tagged subunit (plus covalently cross-linked subunits) absorbed onto nickel beads. The beads were extensively washed, and the bound subunit and cross-linked proteins were eluted. After cleavage of the cross-link by ß-mercaptoethanol, the proteins were detected by immunoblot following SDS PAGE. The first ATPase chosen for cross-linking was Rpt4, as the results were expected to resolve the ambiguity of the orientation of the Rpt4/5 pair noted above in figure 5a. The results obtained, combined with the results detailed above, enabled us to make a putative determination of the molecular organization of the proteasomal ATPases.
Employing this protocol, two of the five remaining ATPases were cross-linked to Rpt4: Rpt5 and Rpt3 (fig. 6 ). Very small amounts of all ATPases appeared to be retained under the highest concentration of cross-linker used. This probably represented some nonspecific carry-over of huge aggregations of cross-linked proteasomes. The same weak band was also observed with anti-Rpt4 antibodies, although it is not apparent in the figure because of the relatively greater intensity of the surrounding bands (compare the amounts of proteasome loaded as controls). These immunoblots also showed the decrease in capacity of binding with increasing cross-linking. This may be due to steric problems associated with binding extremely large complexes to the Ni beads. It is a common finding in affinity chromatography that increasing the protein size from 50 kDa to 100 kDa greatly reduces the amount of protein bound. Here, with highly cross-linked proteasomes, the size may be approaching 2,500 kDa. Thus, optimum cross-linking occurs within a window of cross-linker concentration.
|
The interaction of Rpt4 and Rpt5 agrees with the data of Richmond, Gorbea, and Rechsteiner (1997)
, while the interaction of Rtp4 and Rpt3 resolves the ambiguity of the orientation of the Rpt4/5 pair. Including this result with the data described above (fig. 5a
) produced the probable molecular organization of the proteasomal ATPases depicted in figure 7
.
|
Reconciliation of Physical and Phylogenetic Data
The two interdependent questions that now arise are: (1) Does this model make sense for a heterohexameric ring derived from a homohexameric ring? and (2) Does this model help us choose between the possible phylogenies estimated above?
As a first step, we must work from first principles. With the increase in subunit number arising from gene duplication followed by diversification, what pattern(s) of relatedness with regard to position within a six-membered ring would we expect?
Immediately following duplication of the ancestral gene, the identical products of the two genes (A and B) will be randomly positioned within the ring, with the overall composition ranging from A6 through A3B3 to B6 (fig. 8A ). As soon as diversification of the interactive surfaces occurs, however, the subunits will assume an A3B3 alternating pattern. This is because as soon as complementary mutations occur in the interactive surfaces, they will become inviolate partners (evolutionary ratchet). That is, surface A1 will now interact only with surface B2, and A2 will interact only with B1 (fig. 8B ). As further diversification then occurs within the A and B lineages, A and B descendants will always alternate around the ring. Furthermore, as the current ring is heterohexameric with six different subunits, three will be descended from one lineage and three from the other. None of the trees calculated above possess this structure; however, the parsimony analysis (figure 2 ) is very close. Feeling that the inclusion of the more divergent "primitive" eukaryotic sequences may be complicating matters, the tree was recalculated using only the representative crown eukaryotic sequences (fig. 9 ). Now, the majority-rule consensus of the 32 most-parsimonious trees groups Rpt6 with Rpt1 and Rpt4 within a tree that contains an initial bifurcation.
|
|
Can we now reconcile this predicted pattern with the physical structure? Recall that Richmond, Gorbea, and Rechsteiner (1997)
found strong interactions between pairs of ATPases—Rpt1 and Rpt2, Rpt6 and Rpt3, and Rpt4 and Rpt5, with the interaction between Rpt4 and Rpt5 being confirmed in the present study. Note how, based on the parsimony tree (fig. 2
), each pair contains one member of each lineage (Note also how the two-hybrid data pair subunits within the same lineage—Rpt4 and Rpt6, and Rpt3 and Rpt5 [Ohana et al. 1993
; Russell, Sathyanarayana, and Johnston 1996
]—supporting our contention that these data are unreliable for determining a physical order of subunits around the ring.) Arranging the pairs so that A and B lineage members alternate around the ring results in two possible arrangements (fig. 10
), of which one (a) corresponds to our proposed structure (fig. 7
).
|
Thus, the parsimony majority-rule consensus tree and the experimentally determined structure are consistent with each other and the generic, logically expected structure. We can now propose a possible evolutionary history of the proteasomal ATPases (fig. 11 ).
|
Following the divergence of the archaebacterial and eukaryotic lineages, the first duplication of the ancestral ATPase gene occurred and was followed by a period of diversification resulting in a fixed alternating pattern of the two subunits. Two further duplications then occurred in each lineage, resulting in a total of six genes, three in each lineage.
We cannot tell whether the "final" duplication occurred in both genes of each lineage and one duplicate of each was lost, or whether only a single gene in each lineage underwent duplication. Nor can we determine the arrangement of the various ATPases at intermediate stages of diversification. (fig. 10 , stages 4–6). If the position within the ring is determined solely by the ATPases, then with three, four, or five ATPases, fixed positions cannot be determined within a six-membered ring. For intermediate numbers of ATPases to be present within fixed positions in the hexameric ring, positional information would have to be supplied by other proteins—the apical ring of the 20S core and/or the other components of the RP. If fixed positions were determined, then unequal numbers of ATPases would be present in each ring.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsliterature cited |
|---|
When Did the Diversification of the Proteasomal ATPases Occur?
Based on the phylogenetic and physical relationships between the proteasomal ATPases, we have derived a model for the diversification of the single ancestral ATPase into the minimum of six paralogs extant today. Sometime during the interval between the split between archaebacteria and eukaryotes and the split between fungi and the crown eukaryotes, five gene duplication events occurred. When these events occurred is currently impossible to estimate, but it was possibly relatively soon after the divergence of the archaebacteria and eukaryotes, as indicated by the difficulty in obtaining resolution in the deep nodes—indicative of long periods of evolutionary change. Likewise, it is impossible to determine whether these duplications occurred sequentially or simultaneously or whether or not they represent whole-genome duplications.
A number of genome duplications have been proposed to have occurred in eukaryotic evolutionary history with dates of 600–375 MYA (Ohno 1973
; Skrabanek and Wolfe 1998
). Genome duplications are also proposed to have occurred in fish, maize, Arabidopsis, S. cerevisiae, and Xenopus (Bailey, Poulter, and Stockwell 1978
; Helentjaris, Weber, and Wright 1988
; Ahn and Tanksley 1993
; Hughes and Hughes 1993
; Kowalski et al. 1994
; Wolf and Shields 1997
). All these events occurred long after the divergence of fungi from the main eukaryotic line, by which time the diversification of the ancestral single ATPase into six had already occurred.
Has Further Diversification of the Proteasomal ATPases Occurred?
Multicellular eukaryotes contain more than six proteasomal ATPase genes: Arabidopsis contains at least 9, C. elegans 9, and humans 7, while Drosophila contains 11 such genes (table 1
). This further diversification could allow increased regulation of proteasomal activity through tissue or developmental specific expression of the different ATPase genes. This is well documented with the active-site 20S core ß-subunits. Vertebrates with adaptive immune systems contain more than the seven ß-subunits of the 20S core. Three additional active site ß-subunit genes (MECL-1, LMP2, and LMP7) are present in humans, with two (LMP2 and LMP7) being found within the major histocompatability (MHC) region, a locus that arose by gene duplication prior to the divergence of jawed and jawless vertebrates around 600 MYA (Hughes 1997
). These proteins play a specific role in the generation of immunogenic peptides by replacing their paralogous 20S core subunits in cells stimulated by interferon-
and generating more hydrophobic peptides, as preferred by the TAP transporter/presentation proteins (Groettrup et al. 1996a
). In addition, Arabidopsis contains at least 23 core 20S genes, rather than the "standard" 14 (Fu et al. 1998
).
Despite this proliferation of regulatory ATPases, there is no evidence to suggest that a complimentary change in structure has also occurred. Electron micrographs and SDS gels of purified 26S proteasomes from a number of eukaryotes are remarkably similar. It is much more likely that these additional subunits allow increased levels of tissue-specific and/or developmental regulation (note that the two unicellular eukaryotes for which data are available (S. pombe and S. cerevisiae) contain only six ATPase genes.
What Was the Relative Positioning of the ATPases During Intermediate Stages of Their Evolution and Diversification?
As noted above, unless determined by interactions with other proteasomal subunits, the ATPases within the ring could not maintain specific relationships with each other between the stages of having two ATPases and having six, implying that functional diversification, rather than structural diversification, may have been responsible for the maintenance of the duplicated genes. This same problem of specificity of position is apparent in the 20S proteasome found in the actinomycete Rhodococcus erythropolis. Rhodococcus contains two -subunit genes and two ß-subunit genes. Four possible arrangements can be envisioned (Zuhl et al. 1997b
): (1) two distinct populations consisting of homomeric rings (e.g., 1 ß1 ß1 1 and 2 ß2 ß2 2); (2) rings with alternating subunits (which would require hexameric or even numbered rings); (3) four homoseptameric rings (1 ß1 ß2 2); (4) random subunit distribution. Zuhl et al. (1977a)
were able to show that the Rhodococcus subunits are stochastically arranged. Thus, functional diversification must account for the retention of both genes.
If one sets aside the addition of the Rpn subunits to the RP, the structures of the archael and eukaryotic proteasomes are nearly identical, despite the increase in subunit numbers from 3 (1 -subunit, 1 ß-subunit, 1 ATPase) to 22 (7 -subunits, 7 ß-subunits, 6 ATPases). Thus, the ancestral generalized subunits have been replaced by more specialized components. This scenario fits not with the most widely cited hypothesis of the evolution of protein function (Ohno 1970, 1973
), but with the duplication-degeneration-complementation (DDC) hypothesis put forward by Force et al. (1999)
.
However, not all complexes are so constrained, as illustrated by the evolution of the archael thermosome—yet another chaperonin composed of multiple rings of subunits (Archibald, Logsdon, and Doolittle 1999
). The last common ancestor of the two main divisions of archaebacteria—euryarchaeotes and crenarchaeotes—contained a single thermosomal subunit that assembled into two stacked, octameric rings. In each lineage, after divergence, multiple duplications occurred (followed in some cases by gene loss or conversion such that only one subunit now remains). These subunit gene duplications occurred within the original octameric ring structure of the complex and resulted in alternating subunits within the rings (Nitsch et al. 1997
). In one case, however, a further duplication occurred such that there are now three different subunits within the organism. This additional duplication appears also to have coincided with a change in structure from octameric to nonameric rings. This is possible in the thermosome because its constituent rings interact only with themselves. In the proteasome, while diversification of the single ATPase has occurred, the structure has been constrained to being hexameric, presumably by the requirement to maintain interaction with the septameric 20S rings.
Study of Primitive Eukaryotes
This paper proposes a hypothetical evolutionary history of the proteasomal ATPases that fits the current data. It is possible that very primitive eukaryotes may preserve intermediate stages of this evolution by possessing intermediate numbers of proteasomal ATPases. A few sequences from such organisms are available in the databases (six from Trypanasoma brucei, three from Dictostelyium, and one each from Plasmodium, Naegleria, Giardia, and Leishmania—see table 1
), but so far, these proteins align as specific subunits rather than as ancestral, intermediate sequences (fig. 3
). To test this possibility more thoroughly will require the sampling of additional primitive eukaryotes (Heterokonts, Alveolates, Diplomonads, Microsporidia, Metamonads, etc.) for their complete complement of proteasomal ATPases. Currently, T. brucei is the subject of a genome sequencing project (TIGR and the Sanger Center). While not completed, six proteasomal regulatory ATPase sequences have been determined. Although annotated as subunits 1–6, our analysis suggests that they represent only five of the six subunits (no Rpt1, but two paralogs of Rpt4) (fig. 3 ). This can be interpreted in at least two ways. The first assumes that upon completion of the genomic sequence of T. brucei, a gene encoding Rpt1 will be found. Thus, additional paralogy is occurring in primitive eukaryotes, strengthening the hypothesis that this is a common event in all or most multicellular organisms but not in single-celled eukaryotes (e.g., the yeasts). The second interpretation is that T. brucei does not contain a Rpt1 subunit. Thus, the duplication of Rpt4 to generate Rpt1 would be the final such duplication in generating six individual RPT genes and would have occurred after trypanosomes split from the main eukaryotic lineage. This duplication then independently occurred within trypanosomes, resulting in the two forms of Rpt4 seen. Although the neighbor-joining tree places Rpt4 as the most basal eukaryotic ATPase (fig. 3
), this arrangement has no bootstrap support (fig. 4
). The possible primitive, intermediate nature of the trypanosomal proteasome is supported by the apparent lack of RP lid subunits. Blast searches with the sequences for yeast Rpn3, Rpn5–Rpn9, Rpn11, and Rpn12 against the nonredundant trypansomal protein database revealed no significant matches. In contrast, Trypanosoma do contain a homolog of yeast Rpn1—a non-ATPase proteasomal subunit found in the base of the RP.
In the absence of complete genome data (or ongoing sequencing projects) a survey of proteasomal ATPases is feasible by PCR. Degenerate oligonucleotides complementary to the most highly conserved portions of the CAD could be designed and used to amplify CAD fragments for sequence analysis. This method was employed in the early identification of many of the yeast AAA ATPases (Schnall et al. 1994
). These data may refine the phylogenetic trees calculated above, removing the uncertainty now present. In addition to sequence data, biochemical data could be obtained by the purification (where feasible) of the proteasome from these organisms—it should be noted, however, that the archaebacterial 20S and PAN ATPase complex were biochemically isolated not from the native organisms, but only after the genes were overexpressed in Escherichia coli. In addition to testing our model of proteasomal evolution, these data may also help clear up the positions of these primitive eukaryotes on the Tree of Life. Currently, there is uncertainty in their relative positions, with different phylogenetic arrangements being derived from different data sets (e.g., srRNA and amino acid sequences) (Baldauf and Palmer 1993
; Nikoh et al. 1994
; Kumar and Rzhetsky 1996
). A robust phylogeny of the proteasomal proteins may clarify some relationships.
As noted above, studies of primitive eukaryotes may also shed light on when the regulatory particle of the proteasome recruited the extra components present in addition to the ATPases. These include Rpn1, Rpn2, and Rpn10, as well as the eight proteins related to the signalsome.
|
Supplementary Material |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsliterature cited |
|---|
See the accompanying file for the ATPase alignment on which this analysis is based.
|
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsliterature cited |
|---|
We thank Drs. Michael D. Purruganan and William R. Atchley for helpful discussions and constructive comments on the manuscript. This work was supported by the NSF (J.C.S.).
|
Footnotes |
|---|
Antony M. Dean, Reviewing Editor
1 Present address: New England Eye Center, Tufts University School of Medicine, Boston, Massachusetts. 
1 Abbreviations: CAD, conserved ATPase domain; RP, regulatory particle.
2 Address for correspondence and reprints: Jonathan C. Swaffield, Department of Genetics, North Carolina State University, Raleigh, North Carolina 27695-7614. jon_swaffield{at}ncsu.edu
.
3 Keywords: proteasome
proteasomal ATPases
multiprotein complex
cross-linking
AAA ATPase
evolution
|
literature cited |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsliterature cited |
|---|
Adams, G. M., S. Falke, A. L. Goldberg, C. A. Slaughter, G. N. DeMartino, E. P. Gogol. 1997. Structural and functional effects of RP and modulator protein on proteasomes. J. Mol. Biol. 273:646–657[Web of Science][Medline]
Ahn, S., S. D. Tanksley. 1993. Cereal genome evolution: grasses, line up and form a circle. Proc. Natl. Acad. Sci. USA. 90:7980–7984
Archibald, J. M., J. M. Logsdon Jr., W. F. Doolittle. 1999. Recurrent paralogy in the evolution of archaeal chaperonins. Curr. Biol. 9:1053–1056[Web of Science][Medline]
Babst, M., B. Wendland, E. J. Estepa, S. D. Emr. 1998. The Vps4p AAA ATPase regulates membrane association of a Vps protein complex required for normal endosome function. EMBO J. 17:2982–2993[Web of Science][Medline]
Bailey, G. S., R. T. Poulter, P. A. Stockwell. 1978. Gene duplication in tetraploid fish: model for gene silencing at unlinked duplicated loci. Proc. Natl. Acad. Sci. USA. 75:5575–5579
Baldauf, S. L., J. D. Palmer. 1993. Animals and fungi are each others closest relatives: congruent evidence from multiple proteins. Proc. Natl. Acad. Sci. USA. 90:11558–11562
Beyer, A.. 1997. Sequence analysis of the AAA protein family. Protein Sci. 6:2043–2058[Web of Science][Medline]
Chu-Ping, M., J. H. Vu, R. J. Proske, C. A. Slaughter, G. N. DeMartino. 1994. Identification, purification, and characterization of a high molecular weight, ATP-dependant activator (RP) of the 20S proteasome. J. Biol. Chem. 269:3539–3547
Ciechanover, A., A. Orian, A. L. Schwartz. 2000. Ubiquitin-mediated proteolysis: biological regulation via destruction. BioEssays. 22:442–451[Web of Science][Medline]
Confalonieri, F., M. Duguet. 1995. A 200-amino acid ATPase module in search of a basic function. Bioessays. 17:639–650[Web of Science][Medline]
Dahlman, B., F. Kopp, L. Kuehn, B. Niedel, G. Pfeifer, R. Hegerl, W. Baumeister. 1989. The multicatalytic proteinase (prosome) is ubiquitous from eukaryotes to archaebacteria. FEBS Lett. 251:125–131[Web of Science][Medline]
Dick, L., T. Ruppert, M. Groettrup, P. M. Kloetzel, L. Kuehn, U. H. Koszinowski, S. Stevanovic, H. Schild, H. G. Rammensee. 1996. Coordinated dual-cleavage induced by the proteasome regulator PA28 leads to dominant MHC ligands. Cell. 86:253–262[Web of Science][Medline]
Felsenstein, J.. 1995. PHYLIP (phylogeny inference package)Version 3.57c. Distributed by the author, Department of Genetics, University of Washington, Seattle
Ferrell, K., C. R. M. Wilkinson, W. Dubiel, C. Gordon. 2000. Regulatory subunit interactions of the 26S proteasome, a complex problem. Trends Biochem. Sci. 25:83–88[Web of Science][Medline]
Force, A., M. Lynch, F. B. Pickett, A. Amores, Y. Yan, J. Postlethwait. 1999. Preservation of duplicate genes by complementary, degenerative mutations. Genetics. 151:1531–1545
Frohlich, K. U., H. W. Fries, J. M. Peters, D. Mecke. 1995. The ATPase activity of purified CDC48p from Saccharomyces cerevisiae shows complex dependence on ATP-, ADP-, and NADH-concentrations and is completely inhibited by NEM. Biochim. Biophys. Acta. 1253:23–32
Fu, H., J. H. Doelling, C. S. Arendt, M. Hochstrasser, R. A. Vierstra. 1998. Molecular organization of the 20S proteasome gene family from Arabidopsis thaliana.. Genetics. 149:677–692
Fu, H., J. H. Doelling, D. M. Rubin, R. D. Vierstra. 1999. Structural and functional analysis of the six regulatory particle triple-A ATPase subunits from the Arabidopsis 26S proteasome. Plant J. 18:529–539[Web of Science][Medline]
Gerards, W. L. W., W. W. de Jong, W. Boelens, H. Bloemendal. 1998. Structure and assembly of the 20S proteasome. Cell. Mol. Life Sci. 54:253–262[Web of Science][Medline]
Glickman, M. H., D. M. Rubin, O. Coux, I. Wefes, G. Pfeifer, Z. Cjeka, W. Baumeister, V. A. Fried, D. Finley. 1998a.. A sub complex of the proteasome regulatory particle required for ubiquitin-conjugate degradation and related to the COP9-signalsome and eIF3. Cell. 94:615–623
Glickman, M. H., D. M. Rubin, V. A. Fried, D. Finley. 1998b.. The regulatory particle of the Saccharomyces cerevisiae proteasome. Mol. Cell. Biol. 18:3149–3162
Golbik, R., A. N. Lupas, K. K. Koretke, W. Baumeister, J. Peters. 1999. The Janus face of the archaeal Cdc48/p97 homologous VAT: protein folding versus unfolding. Biol. Chem. 380:1049–1062[Web of Science][Medline]
Gorbea, C., D. Taillandier, M. Rechsteiner. 2000. Mapping subunit contacts in the regulatory complex of the 26S proteasome. J. Biol. Chem. 275:872–882
Gordon, C., G. McGurk, P. Dillon, C. Rosen, N. D. Hastie. 1993. Defective mitosis due to a mutation in the gene for a fission yeast 26S protease subunit. Nature. 366:355–357[Medline]
Gray, C. W., C. A. Slaughter, G. N. DeMartino. 1994. PA28 activator protein forms regulatory caps on proteasomal stacked rings. J. Mol. Biol. 236:7–15[Web of Science][Medline]
Groetrupp, M., A. Soza, M. Eggers, L. Kuehn, T. P. Dick, H. Schild, H. G. Rammensee, U. H. Koszinowski, P. M. Kloetzel. 1996a.. A role for the proteasome regulator PA28a in antigen presentation. Nature. 381:166–168
Groettrup, M., A. Soza, U. Kuckelkorn, P. M. Kloetzel. 1996b.. Peptide antigen production by the proteasome: complexity provides efficiency. Immunol. Today. 17:429–435
Groll, M., M. Bajorek, A. Kohler, L. Moroder, D. M. Rubin, R. Huber, M. H. Glickman, D. Finley. 2000. A gated channel into the proteasome core particle. Nat. Struct. Biol. 7:1062–1067[Web of Science][Medline]
Groll, M., L. Ditzel, J. Lowe, D. Stock, M. Bochtler, H. D. Bartunik, R. Huber. 1997. Structure of 20S proteasome from yeast at 2. 4 A resolution. Nature. 386:463–471[Medline]
Helentjaris, T., D. Weber, S. Wright. 1988. Identification of the genomic locations of duplicate nucleotide sequences in maize by analysis of restriction fragment length polymorphisms. Genetics. 118:353–363
Hendil, K. B.. 1998. The 19S multicatalytic "prosome" proteinase is a constitutive enzyme in HeLa cells. Biochem. Int. 17:471–477
Henikoff, S., J. G. Henikoff. 1992. Amino acid substitution matrices from protein blocks. Proc. Natl. Acad. Sci. USA. 89:10915–10919
Hughes, A. L.. 1997. Evolution of the proteasome components. Immunogenetics. 46:82–92[Web of Science][Medline]
Hughes, M. K., A. L. Hughes. 1993. Evolution of duplicate genes in a tetraploid animal, Xenopus laevis.. Mol. Biol. Evol. 10:1360–1369[Abstract]
Kowalski, S. P., T. H. Lan, K. A. Feldmann, A. H. Paterson. 1994. Comparative mapping of Arabidopsis thaliana and Brassica oleracea chromosomes reveals islands of conserved organization. Genetics. 138:499–510[Abstract]
Kumar, S., A. Rzhetsky. 1996. Evolutionary relationships of eukaryotic kingdoms. J. Mol. Evol. 42:183–193[Web of Science][Medline]
Lee, D. H., A. L. Goldberg. 1996. Selective inhibitors of the proteasome-dependent and vacuolar pathways of protein degradation in Saccharomyces cerevisiae.. J. Biol. Chem. 271:27280–27284
Lenzen, C. U., D. Steinmann, S. W. Whiteheart, W. I. Weis. 1998. Crystal structure of the hexamerization domain of N-ethylmaleimide-sensitive fusion protein. Cell. 95:525–536
Leonhard, K., A. Steigler, W. Neupert, T. Langer. 1999. Chaperone-like activity of the AAA domain of the yeast Yme1 AAA Protease. Nature. 398:348–351[Medline]
Liou, A. K. F., K. R. Willison. 1997. Elucidation of the subunit orientation in CCT (chaperonin containing TCP1) from the subunit composition of CCT micro-complexes. EMBO J. 16:4311–4316[Web of Science][Medline]
Lowe, J., D. Stock, B. Jap, P. Zwickl, W. Baumeister, R. Huber.1995
. Crystal Structure of the 20S proteasome from the
archaeon T. acidophilum at 3.4 A resolution. Science. 268:533–539
Lupas, A., J. M. Flanagan, T. Tamura, W. Baumeister. 1997. Self-compartmentalizing proteases. Trends Biochem. Sci. 22:399–404[Web of Science][Medline]
Neuwald, A. F., L. Aravind, J. L. Spouge, E. V. Koonin. 1999. AAA+: a class of chaperone-like ATPases associated with the assembly, operation and disassembly of protein complexes. Genome Res. 9:27–43
Nikoh, N., N. Hayase, N. Iwabe, K. Kuma, T. Miyata. 1994. Phylogenetic relationships of the kingdoms Animalia, Plantae and Fungi as inferred from 23 different protein species. Mol. Biol. Evol. 11:762–768[Abstract]
Nitsch, M., M. Klumpp, A. Lupas, W. Baumeister. 1997. The thermosome: alternating and ß-subunits within the chaperonin of the archaeon Thermoplasma acidophilum.. J. Mol. Biol. 267:142–149[Web of Science][Medline]
Ohana, B., P. A. Moore, S. M. Ruben, C. D. Southgate, M. R. Green, C. A. Rosen. 1993. The type 1 human immunodeficiency virus tat binding protein is a transcriptional activator belonging to an additional family of evolutionary conserved genes. Proc. Natl. Acad. Sci. USA. 90:138–142
Ohno, S.. 1970. Evolution by gene duplicationGeorge Allen and Unwin, London
———.1973. Ancient linkage groups and frozen accidents. Nature. 244:259–262
Patel, S., M. Latterich. 1998. The AAA team: related ATPases with diverse functions. Trends Cell Biol. 8:65–71[Web of Science][Medline]
Peters, J. M., M. J. Walsh, W. W. Franke. 1990. An abundant and ubiquitous homo-oligomeric ring-shaped ATPase particle related to the putative vesicle fusion proteins Sec18p and NSF. EMBO J. 9:1757–1767[Web of Science][Medline]
Richmond, C., C. Gorbea, M. Rechsteiner. 1997. Specific interactions between ATPase subunits of the 26S protease. J. Biol. Chem. 272:13403–13411
Rock, K. L., C. Gramm, L. Rothstein, K. Clark, R. Stein, L. Dick, D. Hwang, A. L. Goldberg. 1994. Inhibitors of the proteasome block the degradation of most cell proteins and the generation of peptides presented on MHC class 1 molecules. Cell. 78:761–771[Web of Science][Medline]
Rubin, D. M., M. H. Glickman, C. N. Larsen, S. Dhruvakumar, D. Finley. 1998. Active site mutants in the six regulatory particle ATPases reveal multiple roles for ATP in the proteasome. EMBO J. 17:4909–4919[Web of Science][Medline]
Russell, S. J., U. G. Sathyanarayana, S. A. Johnston. 1996. Isolation and characterization of SUG2. J. Biol. Chem. 271:32810–32817
Saitou, N., M. Nei. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406–425[Abstract]
Schnall, R., G. Mannhaupt, R. Stucka, R. Tauer, S. Ehnle, C. Schwarzlose, I. Vetter, H. Feldmann. 1994. Identification of a set of yeast genes coding for a novel family of putative ATPases with high similarity to constituents of the 26S proteasome complex. Yeast. 10:1141–1155[Web of Science][Medline]
Schtilerman, M., G. H. Lorimer, S. W. Englander. 1999. Chaperonin function: folding by forced unfolding. Science. 284:822–825
Skrabanek, L., K. H. Wolfe. 1998. Eukaryote genome duplication—where's the evidence?. Curr. Opin. Genet. Dev. 8:694–700[Web of Science][Medline]
Sollner, T., M. K. Bennett, S. W. Whiteheart, R. H. Scheller, J. E. Rothman. 1993. A protein assembly-disassembly pathway in vitro that may correspond to sequential steps of synaptic vesicle docking, activation and fusion. Cell. 75:409–418[Web of Science][Medline]
Strickland, E., K. Hakala, P. J. Thomas, G. N. DeMartino. 2000. Recognition of misfolding proteins by PA700, the regulatory subcomplex of the 26S proteasome. J. Biol. Chem. 275:5565–5572
Strimmer, K., A. von Haesseler. 1996. Quartet puzzling: a quartet maximum likelihood method for reconstructing tree topologies. Mol. Biol. Evol. 13:964–969[Web of Science]
Suzuki, C. K., M. Rep, J. M. van Dijl, K. Suda, L. A. Grivell, G. Schatz. 1997. ATP-dependent proteases that also chaperone protein biogenesis. Trends Biochem. Sci. 22:118–123[Web of Science][Medline]
Swaffield, J. C., M. D. Purugganan. 1997. The evolution of CAD's: reconstructing the history of an ancient protein module. J. Mol. Evol. 45:549–563[Web of Science][Medline]
Swofford, D.. 2000. PAUP*Phylogenetic analysis using parsimony (*and other methods). Version 4. Sinauer, Sunderland, Mass
Tanaka, K., T. Tamura, T. Yoshimura, A. Ichihara. 1992. Proteasomes: protein and gene structures. New Biol. 4:173–187[Web of Science][Medline]
Tanashi, N., Y. Murakami, Y. Minami, N Shimbara, K. B. Hendil, K. Tanaka. 2000. Hybrid proteasomes. J. Biol. Chem. 275:14336–14345
Thompson, J. D., D. G. Higgins, T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673–4680
Vale, R. D.. 2000. AAA proteins: lords of the ring. J. Cell Biol. 150:F13–F19
Verma, R., S. Chen, R. Feldman, D. Schietz, J. Yates, J. Dohman, R. J. Deshaies. 2000. Proteasomal proteomics: identification of nucleotide-sensitive proteasome-interacting proteins by mass spectrometric analysis of affinity-purified proteasomes. Mol. Biol. Cell. 11:3425–3439
Voges, D., P. Zwickl, W. Baumeister. 1999. The 26S proteasome: a molecular machine designed for controlled proteolysis. Annu. Rev. Biochem. 68:1015–1068[Web of Science][Medline]
Walker, J. E., M. J. Saraste, J. J. Runswick, N. J. Gay. 1982. Distantly related sequences in the a- and b- subunits of ATPase, myosin, kinases and other ATP-requiring enzymes and a common nucleotide binding fold. EMBO J. 1:945–951[Web of Science][Medline]
Walz, J., A. Erdmann, M. Kania, D. Typke, A. J. Koster, W. Baumeister. 1998. 26S proteasome Structure revealed by three-dimensional electron microscopy. J. Struct. Biol. 121:19–29[Web of Science][Medline]
Weber-Ban, E. U., B. G. Reid, A. D. Miranker, A. L. Horwich. 1999. Global unfolding of a substrate protein by the Hsp100 chaperone ClpA. Nature. 401:90–93[Medline]
Wolf, K. H., D. C. Shields. 1997. Molecular evidence for an ancient duplication of the entire yeast genome. Nature. 387:708–713[Medline]
Wolf, S., I. Nagy, A. Lupas, G. Pfeifer, Z. Cejka, S. A. Muller, A. Engel, R. De Mot, W. Baumeister. 1998. Characterization of ARC, a divergent member of the AAA ATPase family from Rhodococcus erythropolis.. J. Mol. Biol. 277:13–25[Web of Science][Medline]
Zuhl, F., E. Seemuller, R. Golbik, W. Baumeister. 1997a.. Dissecting the assembly pathway of the 20S proteasome. FEBS Lett. 418:189–194
Zuhl, F., T. Tamura, I. Dolenc, Z. Cejka, I. Nagy, R. De Mot, W. Baumeister. 1997b.. Subunit topology of the Rhodococcus proteasome. FEBS Lett. 400:83–90
Zwickl, P., D. Ng, K. M. Woo, H.-P. Klenk, A. L. Goldberg. 1999. An archaebacterial ATPase, homologous to ATPases in the eukaryotic 26S proteasome, activates protein breakdown by 20S proteasomes. J. Biol. Chem. 274:26008–26014
Zwickl, P., W. Baumeister. 1999. AAA-ATPase at the crossroads of protein life and death. Nat. Cell Biol. 1:E97–E98
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Muller, A. C. Nagel, D. Maier, and A. Preiss A molecular link A molecular link between Hairless and Pros26.4, a member of the AAA-ATPase subunits of the proteasome 19S regulatory particle in Drosophila J. Cell Sci., January 15, 2006; 119(2): 250 - 258. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Reuter, S. J. Kaczowka, and J. A. Maupin-Furlow Differential Regulation of the PanA and PanB Proteasome-Activating Nucleotidase and 20S Proteasomal Proteins of the Haloarchaeon Haloferax volcanii J. Bacteriol., November 15, 2004; 186(22): 7763 - 7772. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. D. Ulrich Degradation or Maintenance: Actions of the Ubiquitin System on Eukaryotic Chromatin Eukaryot. Cell, February 1, 2002; 1(1): 1 - 10. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||