Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Feb 10.
Published in final edited form as: Mol Cell. 2008 Sep 26;31(6):873. doi: 10.1016/j.molcel.2008.08.006

Structure of a Protein Phosphatase 2A Holoenzyme: Insights into B55-Mediated Tau Dephosphorylation

Yanhui Xu 1,2,3, Yu Chen 1,2, Ping Zhang 1, Philip D Jeffrey 1, Yigong Shi 1,4,*
PMCID: PMC2818795  NIHMSID: NIHMS166818  PMID: 18922469

Summary

Protein phosphatase 2A (PP2A) regulates many essential aspects of cellular physiology. Members of the regulatory B/B55/PR55 family are thought to play a key role in the dephosphorylation of Tau, whose hyperphosphorylation contributes to Alzheimer's disease. The underlying mechanisms of the PP2A-Tau connection remain largely enigmatic. Here, we report the complete reconstitution of a Tau dephosphorylation assay and the crystal structure of a heterotrimeric PP2A holoenzyme involving the regulatory subunit Bα. We show that Bα specifically and markedly facilitates dephosphorylation of the phosphorylated Tau in our reconstituted assay. The Bα subunit comprises a seven-bladed β propeller, with an acidic, substrate-binding groove located in the center of the propeller. The β propeller latches onto the ridge of the PP2A scaffold subunit with the help of a protruding β hairpin arm. Structure-guided mutagenesis studies revealed the underpinnings of PP2A-mediated dephosphorylation of Tau.

Introduction

Protein phosphorylation and dephosphorylation are essential aspects of biology (Hunter, 1995). Protein phosphatase 2A (PP2A) is an important serine/threonine phosphatase that plays a critical role in cellular physiology, including cell cycle, cell proliferation, development, and regulation of multiple signal transduction pathways (Janssens and Goris, 2001; Lechward et al., 2001; Mumby and Walter, 1993; Virshup, 2000). PP2A is also an important tumor suppressor protein (Janssens et al., 2005; Mumby, 2007).

The PP2A core enzyme comprises a 65 kD scaffold subunit (known as A or PR65 subunit) and a 36 kD catalytic subunit (or C subunit). To gain full activity toward specific substrates, the PP2A core enzyme interacts with a variable regulatory subunit to form a heterotrimeric holoenzyme. The variable regulatory subunits consist of four families: B (also known as B55 or PR55), B′ (B56 or PR61), B″ (PR48/PR72/PR130), and B‴ (PR93/PR110), with at least 16 members in these families (Janssens and Goris, 2001; Lechward et al., 2001). There is no detectable sequence homology among the four families of regulatory subunits; the expression levels of various regulatory subunits are highly diverse depending upon cell types and tissues. In this regard, the regulatory subunits determine the substrate specificity as well as the spatial and temporal functions of PP2A.

PP2A is particularly abundant in the brain, accounting for up to 1% of total cellular protein mass. An important function of PP2A is to dephosphorylate the hyperphosphorylated Tau protein (Bennecib et al., 2000; Goedert et al., 1995; Gong et al., 2000; Kins et al., 2001; Sontag et al., 1996, 1999), which has a tendency to polymerize into neurofibrillary tangles, a hallmark of Alzheimer's disease (Goedert and Spillantini, 2006). The hyperphosphorylated Tau also sequesters normal Tau protein, whose function is to promote assembly and stabilization of microtubules (Weingarten et al., 1975; Witman et al., 1976), and, thus, causes damage to the microtubules (Alonso et al., 1994). PP2A-mediated dephosphorylation of Tau appears to be facilitated by the B/B55/PR55 regulatory subunit (Drewes et al., 1993; Gong et al., 1994). How PP2A specifically recognizes and dephosphorylates the phosphorylated Tau protein (pTau) remains poorly understood. There is no structural information on any of the B family subunits.

The A subunit of PP2A contains 15 tandem repeats of a conserved 39 residue sequence known as a huntingtin-elongation-A subunit-TOR (HEAT) motif (Hemmings et al., 1990; Walter et al., 1989). Structural investigation revealed that these 15 HEAT repeats are organized into an extended, L-shaped molecule (Groves et al., 1999). The C subunit recognizes the conserved ridge of HEAT repeats 11–15 (Ruediger et al., 1992, 1994; Xing et al., 2006), whereas the B′ regulatory subunit interacts with the ridge of HEAT repeats 2–8 (Cho and Xu, 2007; Xu et al., 2006). At present, it remains completely unknown how the B subunit recognizes the PP2A core enzyme to form a heterotrimeric holoenzyme.

PP2A functions by removing phosphate groups from substrate proteins; ultimately, understanding the function and mechanism of PP2A depends on elucidating the underpinnings of substrate dephosphorylation. Despite vigorous investigation, however, there is little information on how PP2A specifically facilitates dephosphorylation of target proteins such as Tau. Our study reported in this manuscript addresses these important issues.

Results

In Vitro Reconstitution of a Tau Dephosphorylation Assay

Dephosphorylation of pTau was shown to be mediated mainly by the heterotrimeric PP2A holoenzyme involving the B family of regulatory subunits (Bennecib et al., 2000; Drewes et al., 1993; Goedert et al., 1995; Gong et al., 1994, 2000; Kins et al., 2001; Sontag et al., 1996, 1999). In the past, biochemical investigation of this process relied on PP2A and pTau, both derived from animal tissues. Such an experimental setup does not allow genetic manipulation or mutagenesis of PP2A or Tau, which is crucial to deciphering the mechanisms of PP2A-mediated dephosphorylation of pTau. For example, the endogenous nature of PP2A and pTau does not allow critical assessment of the biochemical roles of individual amino acids. To remedy this problem, we sought and, after many trials, successfully reconstituted an in vitro assay for pTau dephosphorylation using highly purified, recombinant proteins (Figure 1A). A major splice variant of human Tau (4R0N), which contains four microtubule-binding repeats (Gong et al., 2005), was overexpressed in E. coli and purified to homogeneity. The purified Tau was phosphorylated in vitro using the protein kinase GSK-3β, and pTau was isolated by gel filtration. Finally, pTau was dephosphorylated by recombinant PP2A complexes, and the extent of dephosphorylation was examined by an antibody that specifically recognizes phosphorylated Ser396 (which corresponds to Ser338 in Tau-4R0N).

Figure 1. Complete Reconstitution of a Tau Dephosphorylation Assay Using Homogeneous, Recombinant Proteins.

Figure 1

Open in a new tab

(A) Scheme of the in vitro dephosphorylation assay for phosphorylated Tau (pTau). There are five major steps. Representative quality of the unphosphorylated and phosphorylated Tau is shown on SDS-PAGE gels stained by Coomassie blue (right panels).

(B) The heterotrimeric PP2A holoenzyme involving Bα exhibited an enhanced ability to dephosphorylate pTau compared to the heterodimeric PP2A core enzyme. The PP2A concentrations used in lanes 2–6 are 0.73 nM, 2.2 nM, 6.7 nM, 20 nM, and 60 nM. The quality of PP2A core enzyme and holoenzyme is shown in the right panel.

(C) The PP2A holoenzyme involving B′γ exhibited a decreased ability to dephosphorylate pTau compared to the PP2A core enzyme.

In this in vitro assay, the heterotrimeric PP2A holoenzyme involving the Bα subunit efficiently dephosphorylated pTau (Figure 1B, top panel). The specificity of Bα was manifested by the observation that, at the same concentrations, the heterodimeric PP2A core enzyme exhibited a markedly reduced ability to dephosphorylate pTau compared to the Bα-containing holoenzyme (Figure 1B, bottom panel). The B′ subunits represent an entirely different family of regulatory subunits that are thought to have their own distinct substrate specificity. Interestingly, the heterotrimeric PP2A holoenzyme involving the B′γ subunit displayed a further decreased activity compared to the PP2A core enzyme (Figure 1C), suggesting that the presence of the B′γ subunit may limit access of the pTau substrate to the active site of the C subunit. This assay greatly expands our ability to understand the PP2A-Tau interactions and the underlying mechanisms. However, the lack of structural information on Bα and the cognate PP2A holoenzyme limits the application of this assay.

Assembly and Crystallization of the PP2A Holoenzyme

As reported recently (Ikehara et al., 2007), the in vitro assembly of a PP2A holoenzyme between the PP2A core enzyme and the regulatory B subunit does not require carboxy-methylation of the C subunit (data not shown). Nonetheless, to prepare for the possibility that the methylated carboxy-terminal residues of the C subunit may play a minor role in holoenzyme assembly, we first prepared the fully methylated PP2A core enzyme as described (Xu et al., 2006) and then assembled the heterotrimeric PP2A holoenzyme involving Bα. The methylated PP2A holoenzyme was incubated with 1.2 molar equivalence of microcystin-LR (MCLR) prior to crystallization. After experimenting with more than 150,000 crystallization hanging drops, we eventually succeeded in obtaining small crystals of the PP2A holoenzyme. These crystals had poor reproducibility and were sensitive to radiation damage at synchrotron. The structure was determined by molecular replacement, aided by a multiwavelength anomalous dispersion map. The atomic model has been refined to 2.85 Å resolution (Table 1).

Table 1. Data Collection and Statistics from Crystallographic Analysis.

Protein PP2A-Aα-Bα-Cα
Beamline/wavelength NSLS-X29/1.0809 Å
Space group C2
Resolution (outer shell) (Å) 50.0–2.85 (2.95–2.85)
Total observations 275,520
Unique observations 87,353
Redundancy (outer shell) 3.2 (3.2)
Data coverage (outer shell) 98.9% (99.9%)
Rsym (outer shell) 0.050 (0.465)
Refinement:
Resolution range (Å) 50.0–2.85 Å
Number of reflections (|F| > 0) 83,615
Data coverage 95.3%
Rworking/Rfree 0.228/0.286
Number of atoms 20,708
Number of waters 0
Rmsd bond length (Å) 0.0096
Rmsd bond angles (degree) 1.47
Ramachandran Plot:
Most favored (%) 81.8
Additionally allowed (%) 17.1
Generously allowed (%) 0.8
Disallowed (%) 0.3
Open in a new tab

Rsym = ΣhΣi | Ih,i − Ih | / ΣhΣiIh,i, where Ih is the mean intensity of the i observations of symmetry related reflections of h. R = Σ | Fobs − Fcalc | / ΣFobs, where Fobs = FP and Fcalc is the calculated protein structure factor from the atomic model (Rfree was calculated with 5% of the reflections). Rmsd in bond lengths and angles are the deviations from ideal values, and the rmsd deviation in B factors is calculated between bonded atoms.

Overall Structure of the PP2A Holoenzyme

Structure of the 155 kD PP2A holoenzyme measures 100 Å in width, 90 Å in height, and 90 Å in thickness (Figures 2A and 2B). There are 15 HEAT repeats in the A subunit, with each HEAT repeat comprising a pair of antiparallel α helices. Lateral packing among these HEAT repeats gives rise to a horseshoe-shaped structure characterized by double-layered α helices. The loop region connecting two adjacent helices within each HEAT repeat forms a contiguous, conserved ridge (Groves et al., 1999). Compared to those in the PP2A core enzyme (Xing et al., 2006), the N-terminal HEAT repeats in the Bα-containing PP2A holoenzyme pivot around repeats 10 and 11 by as much as 35° to a position that is similar to those in the holoenzyme involving the B′ subunit (Cho and Xu, 2007; Xu et al., 2006) (Figure 2C). However, the 15 HEAT repeats from the B′-containing holoenzyme exist in a more compact conformation, as the N and C termini of the A subunit are closer to each other (Figure 2C). As previously observed (Xing et al., 2006), the C subunit binds to one end of the A subunit through interactions with the ridge of HEAT repeats 11–15.

Figure 2. Overall Structure of the Heterotrimeric PP2A Holoenzyme Involving the Bα Subunit.

Figure 2

Open in a new tab

(A) Overall structure of the PP2A holoenzyme involving the Bα subunit and bound to MCLR. The scaffold (Aα), catalytic (Cα), and regulatory B (Bα) subunits are shown in yellow, green, and blue, respectively. MCLR is shown in magenta. Bα primarily interacts with Aα through an extensive interface. Cα interacts with Aα as described (Xing et al., 2006). Two views are shown here to reveal the essential features of the holoenzyme.

(B) The regulatory Bα subunit contains a highly acidic top face and a hairpin arm. The electrostatic surface potential of Bα is shown. Aα and Cα are shown in backbone worm.

(C) Comparison of the distinct conformations of the A subunit in the PP2A core enzyme and in the two holoenzymes. Figures 2B, 3C, and 5C were prepared using GRASP (Nicholls et al., 1991); all other structural figures were made using MOLSCRIPT (Kraulis, 1991).

The core of the regulatory Bα subunit forms a seven-bladed β propeller, with each blade comprising four antiparallel β strands (Figure 3). By convention of the WD40 domain structure (Wall et al., 1995), the four β strands in each blade are designated A, B, C, and D, radiating from the center of the torus-like structure. In the middle of the top face of the β propeller (convention of Wall et al., 1995), there is a highly acidic groove (Figure 2B). The location and size of the groove are reminiscent of a peptide-binding site that has been observed in other cases (Wilson et al., 2005). In addition to the canonical core structural elements of a β propeller, Bα also contains two β hairpins and two α helices, all of which are located above the top face. These additional structural elements contribute to the formation of the putative substrate-binding groove. In blade 2, β strands C and D extend out of the propeller and form a β hairpin arm that grabs onto the A subunit.

Figure 3. Structural Feature of the Regulatory B Subunit.

Figure 3

Open in a new tab

(A) Sequence alignment of the four isoforms of the regulatory B subunits from humans. Secondary structural elements are indicated above the sequences. Conserved residues are highlighted in yellow. Residues that H-bond to Aα using side chain and main chain groups are identified with red and green circles, respectively, below the sequences. Amino acids that make van der Waals interactions are indicated by blue squares. The sequences shown include all four isoforms of B subunit from humans: α (GI: 4506019), β (GI: 4758954), γ (GI: 21432089), δ1 (GI: 51093851), and δ2 (GI: 51093853).

(B) Structure of the B subunit. The β propeller core is shown in blue; the additional secondary structural elements above the top face are shown in yellow; and the β2C-β2D hairpin arm is highlighted in magenta. Two perpendicular views are shown.

(C) The putative substrate-binding groove on the top face of the Bα propeller is located in close proximity to the active site of the C subunit of PP2A.

Bα makes extensive interactions with the Aα subunit. The bottom face of the propeller binds to the ridge of HEAT repeats 3–7. The β2C-β2D hairpin arm reaches down to interact with HEAT repeats 1 and 2 (Figures 2 and 3). Unlike the PP2A holoenzyme involving the B′ subunit, Bα makes few interactions with the C subunit, with Leu87 from Bα making van der Waals contacts to Val126 and Tyr127 of the C subunit. The methylated carboxy-terminal tail of the C subunit does not have well-defined electron density and appears to be disordered in the crystals. This structural observation is consistent with the biochemical data that methylation is not required for the in vitro assembly of PP2A holoenzyme involving the B family subunits (Ikehara et al., 2007). These observations further suggest that the Bα subunit may form a relatively stable complex with the isolated A subunit, which has been confirmed by our glutathione S transferase (GST)-mediated pull-down assays (data not shown). This result and the structural observation do not seem to support the notion that the C subunit is required for interaction between the A and B subunits (Ruediger et al., 1994).

Interface between the Regulatory and Scaffold Subunits

The continuous interface between Bα and the A subunit can be described in two portions. One portion is mediated by the β2C-β2D hairpin arm of Bα, which makes extensive van der Waals interactions with residues in HEAT repeats 1 and 2 of the A subunit (Figure 4A). The other portion is dominated by hydrogen bonds (H-bonds) between amino acids in the bottom face of the Bα propeller and the ridge of HEAT repeats 3–7 (Figure 4B). These interactions result in the burial of 4270 Å2 exposed surface area.

Figure 4. Specific Recognition of the B Subunit for the PP2A Scaffold Subunit.

Figure 4

Open in a new tab

(A) A stereo view of the atomic interactions between the β2C-β2D hairpin arm of Bα and HEAT repeats 1 and 2 of Aα. This interface is dominated by van der Walls contacts.

(B) A stereo view of the recognition between the bottom face of Bα and HEAT repeats 3–7. This interface contains a number of hydrogen bonds, which are represented by red dashed lines.

(C) Structural comparison of the PP2A holoenzymes involving the regulatory B/B55/PR55 and B′/B56/PR61 subunits.

The hydrophobic amino acids from the β2C-β2D arm of Bα interdigitate with surrounding residues that are located in the outer layer of α helices in HEAT repeats 1 and 2 (Figure 4A). In particular, the hydrophobic side chains of Pro131 and Tyr157 of Bα make multiple van der Waals contacts to Phe54, Tyr60, and the aliphatic portion of side chains in Asp57 and Arg21. These interactions likely make a major contribution to the binding affinity between Bα and the A subunit. Supporting this analysis, deletion of the β2C-β2D arm in Bα resulted in complete loss of interaction between Bα and the A subunit (data not shown).

The specificity of the interaction appears to be provided by seven intermolecular H-bonds at the interface between the bottom face of the Bα propeller and the ridge of HEAT repeats 3–7 (Figure 4B). In particular, the guanidinium group of Arg257 from loop CD of blade 4 donates a pair of charge-stabilized H-bonds to the side chain carboxylate of Asp218 in the A subunit; these interactions are further buttressed by a main chain H-bond between carbonyl oxygen of Arg257 and amide nitrogen of Trp257.

All amino acids in Bα that H-bond to residues in the A subunit are invariant in the β, γ, and δ1 isoforms of the regulatory B family, whereas the Bα amino acids that make van der Waals contact to the A subunit are conserved (Figure 3A). This analysis suggests that Bβ, Bγ, and Bδ1 should also interact with the A subunit identically as observed in our crystal structure. Interestingly, however, the Bδ2 subunit contains a large truncation, which results in the removal of blades 1, 2, and 3 (Figure 3A). Because most PP2A-binding elements are contained within blades 2–4, the Bδ2 subunit is likely to have lost its ability to form a PP2A holoenzyme.

Comparison of Holoenzymes Involving B and B′

Comparison between structures of the holoenzyme that involve the regulatory B subunit and that involve the B′ subunit (Cho and Xu, 2007; Xu et al., 2006) revealed interesting functional similarity. In both cases, the regulatory subunit recognizes the amino-terminal HEAT repeats of the A subunit, with Bα interacting with HEAT 1–7 and B′γ binding to HEAT 2–8 (Figure 4C). The putative substrate-binding site is located on the top face of the regulatory subunit at a position that is proximal to the active site of the C subunit of PP2A. Thus, a major function of both regulatory subunits appears to facilitate the targeting of the substrate phosphoprotein to the phosphatase activity of PP2A.

Important structural differences underlie the contrasting functions of the B and B′ families of regulatory subunits. First, they share no structural similarity, as reflected by their diverging sequences. The B subunit is a seven-bladed β propeller, whereas the B′ subunit comprises eight HEAT-like repeats. Second, the B′ subunit makes significant interactions with the C subunit of PP2A, which consequently strengthens the intersubunit packing, making the resulting holoenzyme relatively compact and rigid (Figure 4C). In contrast, the B subunit makes few interactions with the C subunit, and the resulting holoenzyme appears to be relatively loose.

A previous investigation identified two conserved regions in the regulatory subunits of PP2A that were thought to be involved in binding to the A subunit (Li and Virshup, 2002). These two regions map to portions of the blades 3–6 in Bα (residues 176–274 and 306–364) (Figure 3A) and HEAT repeats 3–7 in B′γ (residues 163–268 and 304–361) (Cho and Xu, 2007; Xu et al., 2006). These regions only contain a small subset of amino acids that interact with the A subunit. Perhaps more importantly, the interactions between the regulatory B/B′ and A subunits depend on the folded structures, but these regions in isolation are unlikely to adopt the native fold.

Mapping the Tau-Binding Site on B Subunit

Reconstitution of the pTau dephosphorylation assay allowed the identification of a Tau-binding site on the B subunit through mutagenesis. Previous studies on β propeller proteins show that the central groove on the top face of the β propeller represents a candidate binding site for ligand peptide (Wilson et al., 2005). To examine this scenario, we generated seven baculoviruses, each containing a different Bα mutant for expression in insect cell. Then, we individually purified the seven Bα mutants, assembled the corresponding PP2A holoenzymes, each involving a different Bα mutant, and purified these holoenzymes to homogeneity (Figure 5A). The mutations affect amino acids that are located in or close to the central acidic groove on the top face of the β propeller. Among the seven mutants, four contain missense mutations (E27R, K48E, D197K, and K345E), each involving a change of the electrostatic charge to the opposite type. The other three are composite mutations: mutant 1 (M1) involves replacing seven residues, Phe84–Leu90, in the β1-β2 hairpin with two amino acids, Gly-Gly; M2 and M3 involve mutating Glu93-Glu94-Lys95 and Tyr178-His179 to Ala93-Ala94-Ala95 and Ala178-Ala179, respectively.

Figure 5. Identification of Tau-Binding Elements in Bα.

Figure 5

Open in a new tab

(A) Representative quality of the PP2A holoenzymes involving seven different mutants of the Bα subunit. The holoenzymes were visualized on SGS-PAGE by Coomassie blue staining.

(B) Effect of various mutations in the Bα subunit on PP2A-mediated dephosphorylation of pTau. Bα-K345E functioned similarly as the WT protein. Other mutations affected PP2A-mediated dephosphorylation of pTau to varying degrees.

(C) A close-up view of the amino acids that are implicated in binding to pTau. These amino acids, shown in yellow, define one side of the surface groove in the top face of the β propeller.

These mutations exhibited different effects on Bα-mediated dephosphorylation of pTau (Figure 5B). The missense mutant K345E displayed a similar activity as the WT Bα, suggesting that Lys345 may be dispensable for binding to pTau. In contrast, all other Bα mutants showed varying degrees of compromised ability to facilitate the dephosphorylation of pTau. For example, the ability of the PP2A holoenzyme involving Bα-E27R or Bα-D197K to dephosphorylate pTau was even worse than the heterodimeric PP2A core enzyme. These results suggest that the central groove on the top face of Bα is the likely binding site for Tau and that a cluster of amino acids on one side of the groove may play a critical role in binding (Figure 5C).

Identification of Bα-Binding Sequences in Tau

Next, we sought to identify the Bα-binding sequences in Tau. The primary sequences of all isoforms of Tau contain an unusually high percentage of hydrophilic amino acids and many proline residues. This sequence feature, as well as computer-based sequence analysis, suggested that Tau is unlikely to be a folded protein. Consistent with this analysis, the Tau protein, derived from either bovine brain or recombinant expression in E. coli, was previously shown to contain little or no secondary structure (Cleveland et al., 1977; Wille et al., 1992). We confirmed this conclusion by performing circular dichroism measurement on the full-length, unphosphorylated splice variant 4R0N of Tau (data not shown). The lack of stably folded structure in Tau justified the strategy of dividing the full-length Tau proteins into overlapping peptide fragments, which are subsequently evaluated for their ability to interact with the PP2A holoenzyme involving Bα.

We generated and purified 18 overlapping Tau fragments (Figure 6A and data not shown). Two different binding assays were used to assess the interaction between each of the 18 Tau fragments and the PP2A holoenzyme involving Bα: polyacrylamide gel electrophoresis (PAGE) under nondenaturing conditions (Figure 6B) and gel filtration (Figure 6C). Because of the sensitive nature for the detection of protein-protein interaction, native PAGE was first employed to assess binding of the various Tau fragments to PP2A. The results were further confirmed by gel filtration chromatography. Analysis of the native PAGE results revealed that the full-length Tau binds to the PP2A holoenzyme with an affinity of ∼3 μM (Figure 6B). These experiments identified two nonoverlapping peptide segments of Tau that are capable of binding to the PP2A holoenzymes: residues 197–259 and residues 265–328 (Figure 6A). The presence of more than one PP2A-binding site in Tau may greatly facilitate the dephosphorylation of hyperphosphorylated Tau (Figure 6D), because hyperphosphorylated Tau is thought to contain multiple phosphorylated Ser/Thr residues that are spread throughout the sequences.

Figure 6. Identification of Peptide Fragments in Tau that Are Critical for Binding to Bα.

Figure 6

Open in a new tab

(A) A summary of the binding assays between various Tau fragments and the PP2A holoenzyme involving Bα. Potential phosphorylation sites in Tau are indicated by asterisks. Due to the mutations, the positions of the Bα variants appear a bit different on the SDS-PAGE gel shown here.

(B) A representative native PAGE gel showing interaction between the full-length Tau and the PP2A holoenzyme involving Bα. The free PP2A holoenzyme involving Bα migrated in two discrete bands (lane 2). This result was confirmed by western blot using antibodies specific for Cα and Bα. Binding of the PP2A holoenzyme by Tau resulted in two slower-migrating species.

(C) A representative example of the result from gel filtration chromatography. In this experiment, an excess amount of the Tau fragment (residues 197–259) was incubated with the PP2A holoenzyme involving Bα and applied to gel filtration. Relevant peak fractions from gel filtration were visualized on SDS-PAGE by Coomassie blue staining. The apparent comigration of Tau (197–259) with PP2A indicates interaction. The control (free Tau fragment on gel filtration) is shown in the lower panel.

(D) A proposed model of PP2A-mediated dephosphorylation of pTau. In this model, pTau binds to the acidic groove on the top face of the B subunit, which presumably facilities access of the nearby phosphorylated serine and threonine residues to the active site of the C subunit of PP2A. Tau contains at least two binding elements for the B subunit, which likely maximize the efficiency of dephosphorylation by enhanced presentation of phosphor-amino acids to PP2A.

Discussion

In the last 2 years, there has been a rapid accumulation of structural information on PP2A and related proteins, including the PP2A phosphatase activator (Chao et al., 2006; Leulliot et al., 2006; Magnusdottir et al., 2006), the PP2A core enzyme (Xing et al., 2006), the PP2A holoenzyme involving B′ subunit (Cho and Xu, 2007; Xu et al., 2006), PP2A-binding protein Tap42/alpha4 (Yang et al., 2007), the PP2A scaffold subunit bound to small t antigen of SV40 (Chen et al., 2007; Cho et al., 2007), and the PP2A core enzyme bound to PME-1 (Xing et al., 2008). The structural information improved our understanding of some aspects of PP2A assembly, function, and regulation. However, mechanistic understanding of PP2A is far from complete. It is fair to say that what we know today represents a small proportion of what is required to have a comprehensive understanding of the function and mechanisms of PP2A. In particular, there is no published structural information on two families of PP2A holoenzyme: families involving the B/B55/PR55 or B″/PR72 regulatory subunits. There is a serious lack of structural information on how LCMT1 regulates the reversible methylation of PP2A and how methylation impacts the assembly of the holoenzymes in vitro. More importantly, despite the fact that PP2A functions through dephosphorylation of substrate phosphoproteins, how PP2A recognizes substrate proteins and mediates this activity remains largely unexplored. The major obstacles for solving these problems appear to be technical challenges in dealing with what is now known as a very tough protein complex.

In this study, we report two major advances. First, we determined the crystal structure of the PP2A holoenzyme involving a B/B55/PR55 regulatory subunit. This structure reveals how the Bα subunit specifically recognizes the PP2A core enzyme and how Bα may facilitate substrate dephosphorylation. This structure also represents structural information on the B family of regulatory subunits, which contains seven WD40 repeats rather than five as previously thought (Janssens and Goris, 2001). Second, we reconstituted a Tau dephosphorylation assay and applied this assay to characterize the interaction between Tau and Bα in the context of PP2A holoenzyme. Our assay relies entirely on recombinant components rather than endogenous materials and, thus, allows us to manipulate each component through mutagenesis—a strategy required for mechanistic understanding of PP2A function. Using this strategy, we mapped the respective binding epitopes on Bα and on Tau.

Our biochemical characterization suggests that at least two separate peptide fragments of Tau have the ability to interact with the acidic groove of Bα. The presence of more than one PP2A-binding site may allow Tau to “slide” on Bα so as to more efficiently target nearby phosphoserine/phosphothreonine residues to the C subunit of PP2A for dephosphorylation. Interestingly, the two putative Bα-binding elements fall within the microtubule-binding repeats of Tau (Figure 6A). This result is in excellent agreement with a previous study (Sontag et al., 1999). These two Bα-binding repeats are characterized by an enrichment of positively charged amino acids. For example, the Tau fragment 197–259 is highly basic, with 11 lysine residues. This sequence feature agrees well with the acidic nature of the putative substrate-binding groove on the Bα subunit (Figure 2B). The minimal or consensus peptide that retains binding to the B subunit remains to be identified.

It remains to be investigated whether other substrates of the B-containing PP2A holoenzyme contain similar sequence features as the positively charged motifs of Tau. Although Tau has been extensively studied, most other substrate proteins are poorly characterized. The relatively weak binding affinity between Tau and Bα likely ensures that the binding is dynamic and the dephosphorylated Tau can be readily released from the B subunit. Furthermore, our biochemical and structural analysis suggests that the isolated Bα is likely to retain the ability to interact with Tau.

The total reconstitution of Tau dephosphorylation in vitro using homogeneous, recombinant proteins represents an important step toward deciphering the underpinnings of PP2A-mediated regulation of Tau. GSK-3β, which potently phosphorylates Tau at multiple sites in vitro (Gong et al., 2005), was used as the kinase for Tau in our assay. Compared to the unphosphorylated Tau, pTau exhibited a retarded mobility on SDS-PAGE (Figure 1A). Consistent with published reports, Ser396 was among the Ser/Thr residues in Tau that were phosphorylated by GSK-3β and was recognized by a specific antibody (Figures 1B and 1C). The extent of dephosphorylation of pSer396 was used as a direct readout of PP2A activity.

Previous studies suggested that carboxy-methylation of the C subunit was important for the assembly of PP2A holoenzymes involving the B subunits in cells (Bryant et al., 1999; Gentry et al., 2005; Kloeker et al., 1997; Longin et al., 2007; Tolstykh et al., 2000; Wei et al., 2001; Wu et al., 2000; Yu et al., 2001). A common feature of these studies is that the assembly of PP2A holoenzymes was investigated in cells, rather than in vitro using purified recombinant proteins. In contrast, a recent study using purified proteins showed that the methylation status of the C subunit had little impact on the in vitro assembly of PP2A holoenzyme involving the B subunit (Ikehara et al., 2007). Our structural analysis supports this conclusion. In fact, the carboxy-terminal 14 amino acids of the C subunit are disordered in the crystals and are dispensable for formation of the PP2A holoenzyme in vitro. Nonetheless, these observations do not rule out the possibility that carboxy-methylation of the C subunit may facilitate the assembly of holoenzymes, perhaps through enhanced binding affinity of the PP2A core enzyme for the B subunit. Supporting this hypothesis, competition experiments suggested that the methylated PP2A core enzyme exhibited a higher binding affinity for Bα than the unmethylated core enzyme (data not shown).

Methylation of the C subunit was shown to have little impact on the in vitro assembly of the PP2A holoenzyme involving the B′ subunit (Gentry et al., 2005; Longin et al., 2007; Xu et al., 2006). These observations, together with those listed above, argue that the carboxy-methylation of the C subunit is not required for the in vitro assembly of PP2A holoenzymes involving the B and B′ regulatory subunits. If methylation is not required for PP2A holoenzyme assembly in vitro, why does it appear to play an important role in cells (Bryant et al., 1999; Gentry et al., 2005; Kloeker et al., 1997; Longin et al., 2007; Tolstykh et al., 2000; Wei et al., 2001; Wu et al., 2000; Yu et al., 2001)? One possibility is that methylation gives a slight advantage in terms of binding affinity between the PP2A core enzyme and the regulatory subunit such as Bα; this advantage may be sufficient to tip the balance for holoenzyme assembly in cells. Another possibility is that the carboxy-methylation mainly serves as a signal for assembly of the PP2A holoenzyme. For example, the regulatory subunits may be sequestered in a specific cellular compartment, and the methylated carboxyl terminus of the C subunit may allow its targeting to this location for holoenzyme assembly. In addition, the methylated carboxyl terminus might help recruit assembly factors that actively promote assembly of the PP2A holoenzymes in cells. Examination of these hypotheses awaits future experimental investigation.

Experimental Procedures

Protein Preparation and Assembly of PP2A Holoenzyme

All constructs and point mutations were generated using a standard PCR-based cloning strategy. Aα (residues 1–589) was overexpressed in E. coli as a fusion protein with GST and purified as described (Xu et al., 2006). Full-length His6-tagged Cα (residues 1–309) and Bα (residues 1–447) were coexpressed in baculovirus-infected insect cells. The PP2A holoenzyme was purified to homogeneity first by glutathione sepharose 4B resin, using GST-Aα to pull out Bα and Cα, followed by anion exchange and gel filtration chromatography. We also assembled the holoenzyme by first reconstituting the PP2A core enzyme, which was methylated by a PP2A-specific leucine carboxyl methyltransferase (LCMT) in the presence of S-adenosyl methionine (SAM), and then incubating the homogeneously methylated PP2A core enzyme with purified Bα subunit. Both assembly protocols gave rise to identical holoenzymes as examined by phosphatase assays and identical crystals. To facilitate structure determination, we also prepared the PP2A holoenzyme complex using seleno-methionine-substituted Aα, Cα, and Bα proteins using a published protocol (Cronin et al., 2007).

Crystallization and Data Collection

Diffracting crystals were obtained for the PP2A holoenzyme described above, which was incubated with 1.2 molar equivalence of MCLR prior to crystallization. We also generated crystals of the holoenzyme using selenomethionine-substituted holoenzyme. Crystals were grown by the hanging-drop vapor-diffusion method by mixing the protein (∼8 mg/ml) with an equal volume of reservoir solution containing 7%–10% PEG35000 and 0.1–0.15 M sodium citrate (pH 5.5). Small crystals appeared within a few days. The crystals were in three closely related crystal forms: P1 with a = 124 Å, b = 141 Å, c = 141 Å, α = 79, β = 64, and γ = 64 with four complexes in the asymmetric unit (AU); C2 with a = 247 Å, b = 121 Å, c = 172 Å, and β = 133 with two complexes per AU; I4 with a = b = 182 Å and c = 124 Å with one complex per AU. Most of the structural work and the definitive refinement were done with the C2 form. Crystals were slowly equilibrated in a cryoprotectant buffer containing reservoir buffer plus 20% glycerol (v/v) and were flash-frozen in a cold nitrogen stream at −170°C. The native and selenium MAD data sets were collected at NSLS beamline X29 and processed using the software Denzo and Scalepack (Otwinowski and Minor, 1997).

Structure Determination

The structure was determined by molecular replacement, using the PP2A core enzyme (Xing et al., 2006) and various WD40 repeats as a model, against an initial 3.5 Å native dataset in the C2 form. Molecular replacement solutions of the P1 and I4 form confirmed the close relationships between these crystal forms. Calculations were performed with the program PHASER (McCoy et al., 2005). Structure determination was complicated by the apparent flexibility of the complexes with the carboxy-terminal end of the Aα subunit and the Cα subunit displaying elevated B factors. Two AC complexes were assembled based on molecular replacement solutions of the Cα domain and three fragments of the Aα domain. Based on this solution, it was not possible to build the B subunit. A 5.5 Å resolution Ta6Br12 MAD map, calculated using SHELX (Sheldrick, 2008) and SHARP, in the P1 crystal form confirmed the presence of the B subunit and the packing arrangement. In the absence of an available homologous structure for the B subunit, an ensemble of five superimposed WD40 domains with trimmed loops was used to find a single B subunit in the P1 crystal form, and the position was confirmed by reference to the Ta6Br12 MAD map. The second B subunit was generated using the known noncrystallographic symmetry relationship. Superimposition of the heterotrimeric complex in the C2 form showed that it was compatible with existing maps and packing in that form.

Despite low homology between the initial model and the B subunit sequence, model-phased 2-fold averaged 2Fo − Fc, α-calc maps were sufficient to make modifications to the poly-Ala backbone, and successive improvements to the model led to the appearance of interpretable side chain density. Addition of a 2.9 Å native data set and use of the model-phased SeMet anomalous difference map from a 3.8 Å SeMet MAD data set enabled definitive interpretation of the sequence for the B subunit. The structure was refined at 2.85 Å resolution using the program CNS (Brunger et al., 1998), incorporating noncrystallographic symmetry restraints between the two heterotrimeric complexes. The final atomic model contains amino acids 6–293 for Cα, residues 9–589 for Aα, and residues 8–137 and 146–446 for Bα. There is no electron density for residues 294–309 of Cα and residues 138–145 of Bα; we presume these regions are disordered in the crystals. The few amino acids in each holoenzyme that are in the disallowed region of the Ramachandran plot are located in the active site of the C subunit, involved in metal binding, and consistent with their functional roles as has been observed in other PP2A and PP1 structures (Goldberg et al., 1995; Xing et al., 2006).

Methylation of PP2A Core Enzyme by LCMT

This assay was performed as previously described (Xu et al., 2006).

Native PAGE and Gel Filtration

These assays were performed as previously described (Xing et al., 2006).

Phosphorylation and Dephosphorylation Assays of Tau Protein

Bacterially expressed Tau was purified by ion-exchange chromatography and gel filtration to homogeneity. The phosphorylation reaction was carried out by mixing purified Tau with GSK3β (Upstate Biotechnology) in the presence of 2 mM ATP and 10 mM MgCl2 in phosphorylation buffer (8 mM Tris·Cl buffer [pH 7.5], 0.2 mM EDTA) at 37°C for 16 hr. Phosphorylated Tau (pTau) was further purified by gel filtration.

In dephosphorylation assay of pTau, 0.36 uM pTau was incubated with PP2A samples in dephosphorylation buffer (20 mM Tris [pH 7.5], 1 mM DTT) at 30°C for 30 min. The reaction was stopped by adding SDS loading buffer, and the samples were loaded onto SDS-PAGE. The phosphorylation status of Tau was examined by western blot using an antibody (Biosource) that specifically recognizes phosphorylated Ser396 of Tau. Antibody recognizing both the phosphorylated and nonphosphorylated Tau (Invitrogen) was used as control.

Accession Numbers

The atomic coordinates of the PP2A holoenzyme have been deposited in the Protein Data Bank with the accession codes 3DW8.

Acknowledgments

We thank Dr. Bing Zhou at Tsinghua University (Beijing, China) for the cDNA of Tau, Dr. Zhu Li for technical assistance, and Dr. Anand Saxena at BNL NSLS beamlines for help. This work was supported by NIH grant CA123155 (to Y.S.).

References

  1. Alonso AC, Zaidi T, Grundke-Iqbal I, Iqbal K. Role of abnormally phosphorylated tau in the breakdown of microtubules in Alzheimer disease. Proc Natl Acad Sci USA. 1994;91:5562–5566. doi: 10.1073/pnas.91.12.5562. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bennecib M, Gong CX, Grundke-Iqbal I, Iqbal K. Role of protein phosphatase-2A and -1 in the regulation of GSK-3, cdk5 and cdc2 and the phosphorylation of tau in rat forebrain. FEBS Lett. 2000;485:87–93. doi: 10.1016/s0014-5793(00)02203-1. [DOI] [PubMed] [Google Scholar]
  3. Brunger AT, Adams PD, Clore GM, DeLano WL, Gros P, Grosse-Kunstleve RW, Jiang JS, Kuszewski J, Nilges M, Pannu NS, et al. Crystallography and NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr D Biol Crystallogr. 1998;54:905–921. doi: 10.1107/s0907444998003254. [DOI] [PubMed] [Google Scholar]
  4. Bryant JC, Westphal RS, Wadzinski BE. Methylated C-terminal leucine residue of PP2A catalytic subunit is important for binding of regulatory Balpha subunit. Biochem J. 1999;339:241–246. [PMC free article] [PubMed] [Google Scholar]
  5. Chao Y, Xing Y, Chen Y, Xu Y, Lin Z, Li Z, Jeffrey PD, Stock JB, Shi Y. Structure and Mechanism of the Phosphotyrosyl Phosphatase Activator. Mol Cell. 2006;23:535–546. doi: 10.1016/j.molcel.2006.07.027. [DOI] [PubMed] [Google Scholar]
  6. Chen Y, Xu Y, Bao Q, Xing Y, Li Z, Lin Z, Stock JB, Jeffrey PD, Shi Y. Structural and biochemical insights into the regulation of protein phosphatase 2A by small t antigen of SV40. Nat Struct Mol Biol. 2007;14:527–534. doi: 10.1038/nsmb1254. [DOI] [PubMed] [Google Scholar]
  7. Cho US, Xu W. Crystal structure of a protein phosphatase 2A heterotrimeric holoenzyme. Nature. 2007;445:53–57. doi: 10.1038/nature05351. [DOI] [PubMed] [Google Scholar]
  8. Cho US, Morrone S, Sablina AA, Arroyo JD, Hahn WC, Xu W. Structural basis of PP2A inhibition by small t antigen. PLoS Biol. 2007;5:e202. doi: 10.1371/journal.pbio.0050202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cleveland DW, Hwo SY, Kirschner MW. Physical and chemical properties of purified tau factor and the role of tau in microtubule assembly. J Mol Biol. 1977;116:227–247. doi: 10.1016/0022-2836(77)90214-5. [DOI] [PubMed] [Google Scholar]
  10. Cronin CN, Lim KB, Rogers J. Production of selenomethionyl-derivatized proteins in baculovirus-infected insect cells. Protein Sci. 2007;16:2023–2029. doi: 10.1110/ps.072931407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Drewes G, Mandelkow EM, Baumann K, Goris J, Merlevede W, Mandelkow E. Dephosphorylation of tau protein and Alzheimer paired helical filaments by calcineurin and phosphatase-2A. FEBS Lett. 1993;336:425–432. doi: 10.1016/0014-5793(93)80850-t. [DOI] [PubMed] [Google Scholar]
  12. Gentry MS, Li Y, Wei H, Syed FF, Patel SH, Hallberg RL, Pallas DC. A novel assay for protein phosphatase 2A (PP2A) complexes in vivo reveals differential effects of covalent modifications on different Saccharomyces cerevisiae PP2A heterotrimers. Eukaryot Cell. 2005;4:1029–1040. doi: 10.1128/EC.4.6.1029-1040.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Goedert M, Spillantini MG. A century of Alzheimer's disease. Science. 2006;314:777–781. doi: 10.1126/science.1132814. [DOI] [PubMed] [Google Scholar]
  14. Goedert M, Jakes R, Qi Z, Wang JH, Cohen P. Protein phosphatase 2A is the major enzyme in brain that dephosphorylates tau protein phosphorylated by proline-directed protein kinases or cyclic AMP-dependent protein kinase. J Neurochem. 1995;65:2804–2807. doi: 10.1046/j.1471-4159.1995.65062804.x. [DOI] [PubMed] [Google Scholar]
  15. Goldberg J, Huang HB, Kwon YG, Greengard P, Nairn AC, Kuriyan J. Three-dimensional structure of the catalytic subunit of protein serine/threonine phosphatase-1. Nature. 1995;376:745–753. doi: 10.1038/376745a0. [DOI] [PubMed] [Google Scholar]
  16. Gong CX, Grundke-Iqbal I, Iqbal K. Dephosphorylation of Alzheimer's disease abnormally phosphorylated tau by protein phosphatase-2. A Neuroscience. 1994;61:765–772. doi: 10.1016/0306-4522(94)90400-6. [DOI] [PubMed] [Google Scholar]
  17. Gong CX, Lidsky T, Wegiel J, Zuck L, Grundke-Iqbal I, Iqbal K. Phosphorylation of microtubule-associated protein tau is regulated by protein phosphatase 2A in mammalian brain. Implications for neurofibrillary degeneration in Alzheimer's disease. J Biol Chem. 2000;275:5535–5544. doi: 10.1074/jbc.275.8.5535. [DOI] [PubMed] [Google Scholar]
  18. Gong CX, Liu F, Grundke-Iqbal I, Iqbal K. Post-translational modifications of tau protein in Alzheimer's disease. J Neural Transm. 2005;112:813–838. doi: 10.1007/s00702-004-0221-0. [DOI] [PubMed] [Google Scholar]
  19. Groves MR, Hanlon N, Turowski P, Hemmings BA, Barford D. The structure of the protein phosphatase 2A PR65/A subunit reveals the conformation of its 15 tandemly repeated HEAT motifs. Cell. 1999;96:99–110. doi: 10.1016/s0092-8674(00)80963-0. [DOI] [PubMed] [Google Scholar]
  20. Hemmings BA, Adams-Pearson C, Maurer F, Muller P, Goris J, Merlevede W, Hofsteenge J, Stone SR. alpha- and beta-forms of the 65-kDa subunit of protein phosphatase 2A have a similar 39 amino acid repeating structure. Biochemistry. 1990;29:3166–3173. doi: 10.1021/bi00465a002. [DOI] [PubMed] [Google Scholar]
  21. Hunter T. Protein kinases and phosphatases: the yin and yang of protein phosphorylation and signaling. Cell. 1995;80:225–236. doi: 10.1016/0092-8674(95)90405-0. [DOI] [PubMed] [Google Scholar]
  22. Ikehara T, Ikehara S, Imamura S, Shinjo F, Yasumoto T. Methylation of the C-terminal leucine residue of the PP2A catalytic subunit is unnecessary for the catalytic activity and the binding of regulatory subunit (PR55/B) Biochem Biophys Res Commun. 2007;354:1052–1057. doi: 10.1016/j.bbrc.2007.01.085. [DOI] [PubMed] [Google Scholar]
  23. Janssens V, Goris J. Protein phosphatase 2A: a highly regulated family of serine/threonine phosphatases implicated in cell growth and signalling. Biochem J. 2001;353:417–439. doi: 10.1042/0264-6021:3530417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Janssens V, Goris J, Van Hoof C. PP2A: the expected tumor suppressor. Curr Opin Genet Dev. 2005;15:34–41. doi: 10.1016/j.gde.2004.12.004. [DOI] [PubMed] [Google Scholar]
  25. Kins S, Crameri A, Evans DR, Hemmings BA, Nitsch RM, Gotz J. Reduced protein phosphatase 2A activity induces hyperphosphorylation and altered compartmentalization of tau in transgenic mice. J Biol Chem. 2001;276:38193–38200. doi: 10.1074/jbc.M102621200. [DOI] [PubMed] [Google Scholar]
  26. Kloeker S, Bryant JC, Strack S, Colbran RJ, Wadzinski BE. Carboxymethylation of nuclear protein serine/threonine phosphatase X. Biochem J. 1997;327:481–486. doi: 10.1042/bj3270481. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kraulis PJ. Molscript: a program to produce both detailed and schematic plots of protein structures. J Appl Cryst. 1991;24:946–950. [Google Scholar]
  28. Lechward K, Awotunde OS, Swiatek W, Muszynska G. Protein phosphatase 2A: variety of forms and diversity of functions. Acta Biochim Pol. 2001;48:921–933. [PubMed] [Google Scholar]
  29. Leulliot N, Vicentini G, Jordens J, Quevillon-Cheruel S, Schiltz M, Barford D, van Tilbeurgh H, Goris J. Crystal structure of the PP2A phosphatase activator: implications for its PP2A-specific PPIase activity. Mol Cell. 2006;23:413–424. doi: 10.1016/j.molcel.2006.07.008. [DOI] [PubMed] [Google Scholar]
  30. Li X, Virshup DM. Two conserved domains in regulatory B subunits mediate binding to the A subunit of protein phosphatase 2A. Eur J Biochem. 2002;269:546–552. doi: 10.1046/j.0014-2956.2001.02680.x. [DOI] [PubMed] [Google Scholar]
  31. Longin S, Zwaenepoel K, Louis JV, Dilworth S, Goris J, Janssens V. Selection of protein phosphatase 2A regulatory subunits is mediated by the C terminus of the catalytic Subunit. J Biol Chem. 2007;282:26971–26980. doi: 10.1074/jbc.M704059200. [DOI] [PubMed] [Google Scholar]
  32. Magnusdottir A, Stenmark P, Flodin S, Nyman T, Hammarstrom M, Ehn M, Bakali HMA, Berglund H, Nordlund P. The crystal structure of a human PP2A phosphatase activator reveals a novel fold and highly conserved cleft implicated in protein-protein interactions. J Biol Chem. 2006;281:22434–22438. doi: 10.1074/jbc.C600100200. [DOI] [PubMed] [Google Scholar]
  33. McCoy AJ, Grosse-Kunstleve RW, Storoni LC, Read RJ. Likelihood-enhanced fast translation functions. Acta Crystallogr D Biol Crystallogr. 2005;61:458–464. doi: 10.1107/S0907444905001617. [DOI] [PubMed] [Google Scholar]
  34. Mumby M. PP2A: unveiling a reluctant tumor suppressor. Cell. 2007;130:21–24. doi: 10.1016/j.cell.2007.06.034. [DOI] [PubMed] [Google Scholar]
  35. Mumby MC, Walter G. Protein serine/threonine phosphatases: structure, regulation, and functions in cell growth. Physiol Rev. 1993;73:673–699. doi: 10.1152/physrev.1993.73.4.673. [DOI] [PubMed] [Google Scholar]
  36. Nicholls A, Sharp KA, Honig B. Protein folding and association: insights from the interfacial and thermodynamic properties of hydrocarbons. Proteins Struct Funct Genet. 1991;11:281–296. doi: 10.1002/prot.340110407. [DOI] [PubMed] [Google Scholar]
  37. Otwinowski Z, Minor W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 1997;276:307–326. doi: 10.1016/S0076-6879(97)76066-X. [DOI] [PubMed] [Google Scholar]
  38. Ruediger R, Roeckel D, Fait J, Bergqvist A, Magnusson G, Walter G. Identification of binding sites on the regulatory A subunit of protein phosphatase 2A for the catalytic C subunit and for tumor antigens of simian virus 40 and polyomavirus. Mol Cell Biol. 1992;12:4872–4882. doi: 10.1128/mcb.12.11.4872. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Ruediger R, Hentz M, Fait J, Mumby M, Walter G. Molecular model of the A subunit of protein phosphatase 2A: interaction with other subunits and tumor antigens. J Virol. 1994;68:123–129. doi: 10.1128/jvi.68.1.123-129.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Sheldrick GM. A short history of SHELX. Acta Crystallogr A. 2008;64:112–122. doi: 10.1107/S0108767307043930. [DOI] [PubMed] [Google Scholar]
  41. Sontag E, Nunbhakdi-Craig V, Lee G, Bloom GS, Mumby MC. Regulation of the phosphorylation state and microtubule-binding activity of Tau by protein phosphatase 2. A Neuron. 1996;17:1201–1207. doi: 10.1016/s0896-6273(00)80250-0. [DOI] [PubMed] [Google Scholar]
  42. Sontag E, Nunbhakdi-Craig V, Lee G, Brandt R, Kamibayashi C, Kuret J, White CL, 3rd, Mumby MC, Bloom GS. Molecular interactions among protein phosphatase 2A, tau, and microtubules. Implications for the regulation of tau phosphorylation and the development of tauopathies. J Biol Chem. 1999;274:25490–25498. doi: 10.1074/jbc.274.36.25490. [DOI] [PubMed] [Google Scholar]
  43. Tolstykh T, Lee J, Vafai S, Stock JB. Carboxyl methylation regulates phosphoprotein phosphatase 2A by controlling the association of regulatory B subunits. EMBO J. 2000;19:5682–5691. doi: 10.1093/emboj/19.21.5682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Virshup DM. Protein phosphatase 2A: a panoply of enzymes. Curr Opin Cell Biol. 2000;12:180–185. doi: 10.1016/s0955-0674(99)00074-5. [DOI] [PubMed] [Google Scholar]
  45. Wall MA, Coleman DE, Lee E, Iñ iguez-Lluhi JA, Posner BA, Gilman AG, Sprang SR. The structure of the G protein heterotrimer Gi alpha 1 beta 1 gamma 2. Cell. 1995;83:1047–1058. doi: 10.1016/0092-8674(95)90220-1. [DOI] [PubMed] [Google Scholar]
  46. Walter G, Ferre F, Espiritu O, Carbone-Wiley A. Molecular cloning and sequence of cDNA encoding polyoma medium tumor antigen-associated 61-kDa protein. Proc Natl Acad Sci USA. 1989;86:8669–8672. doi: 10.1073/pnas.86.22.8669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Wei H, Ashby DG, Moreno CS, Ogris E, Yeong FM, Corbett AH, Pallas DC. Carboxymethylation of the PP2A catalytic subunit in Saccharomyces cerevisiae is required for efficient interaction with the B-type subunits Cdc55p and Rts1p. J Biol Chem. 2001;276:1570–1577. doi: 10.1074/jbc.M008694200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Weingarten MD, Lockwood AH, Hwo SY, Kirschner MW. A protein factor essential for microtubule assembly. Proc Natl Acad Sci USA. 1975;72:1858–1862. doi: 10.1073/pnas.72.5.1858. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Wille H, Drewes G, Biernat J, Mandelkow EM, Mandelkow E. Alzheimer-like paired helical filaments and antiparallel dimers formed from microtubule-associated protein tau in vitro. J Cell Biol. 1992;118:573–584. doi: 10.1083/jcb.118.3.573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wilson DK, Cerna D, Chew E. The 1.1-angstrom structure of the spindle checkpoint protein Bub3p reveals functional regions. J Biol Chem. 2005;280:13944–13951. doi: 10.1074/jbc.M412919200. [DOI] [PubMed] [Google Scholar]
  51. Witman GB, Cleveland DW, Weingarten MD, Kirschner MW. Tubulin requires tau for growth onto microtubule initiating sites. Proc Natl Acad Sci USA. 1976;73:4070–4074. doi: 10.1073/pnas.73.11.4070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Wu J, Tolstykh T, Lee J, Boyd K, Stock JB, Broach JR. Carboxyl methylation of the phosphoprotein phosphatase 2A catalytic subunit promotes its functional association with regulatory subunits in vivo. EMBO J. 2000;19:5672–5681. doi: 10.1093/emboj/19.21.5672. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Xing Y, Xu Y, Chen Y, Jeffrey PD, Chao Y, Zheng L, Li Z, Strack S, Stock JB, Shi Y. Structure of protein phosphatase 2A core enzyme bound to tumor-inducing toxins. Cell. 2006;127:341–352. doi: 10.1016/j.cell.2006.09.025. [DOI] [PubMed] [Google Scholar]
  54. Xing Y, Li Z, Chen Y, Stock JB, Jeffrey PD, Shi Y. Structural mechanism of demethylation and inactivation of protein phosphatase 2. A Cell. 2008;133:154–163. doi: 10.1016/j.cell.2008.02.041. [DOI] [PubMed] [Google Scholar]
  55. Xu Y, Xing Y, Chen Y, Chao Y, Lin Z, Fan E, Yu JW, Strack S, Jeffrey PD, Shi Y. Structure of the protein phosphatase 2A holoenzyme. Cell. 2006;127:1239–1251. doi: 10.1016/j.cell.2006.11.033. [DOI] [PubMed] [Google Scholar]
  56. Yang J, Roe SM, Prickett TD, Brautigan DL, Barford D. The structure of Tap42/alpha4 reveals a tetratricopeptide repeat-like fold and provides insights into PP2A regulation. Biochemistry. 2007;46:8807–8815. doi: 10.1021/bi7007118. [DOI] [PubMed] [Google Scholar]
  57. Yu XX, Du X, Moreno CS, Green RE, Ogris E, Feng Q, Chou L, McQuoid MJ, Pallas DC. Methylation of the protein phosphatase 2A catalytic subunit is essential for association of Balpha regulatory subunit but not SG2NA, striatin, or polyomavirus middle tumor antigen. Mol Biol Cell. 2001;12:185–199. doi: 10.1091/mbc.12.1.185. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES