B56δ long-disordered arms form a dynamic PP2A regulation interface coupled with global allostery and Jordan's syndrome mutations

Abstract

Intrinsically disordered regions (IDR) and short linear motifs (SLiMs) play pivotal roles in the intricate signaling networks governed by phosphatases and kinases. B56δ (encoded by PPP2R5D) is a regulatory subunit of protein phosphatase 2A (PP2A) with long IDRs that harbor a substrate-mimicking SLiM and multiple phosphorylation sites. De novo missense mutations in PPP2R5D cause intellectual disabilities (ID), macrocephaly, Parkinsonism, and a broad range of neurological symptoms. Our single-particle cryo-EM structures of the PP2A-B56δ holoenzyme reveal that the long, disordered arms at the B56δ termini fold against each other and the holoenzyme core. This architecture suppresses both the phosphatase active site and the substrate-binding protein groove, thereby stabilizing the enzyme in a closed latent form with dual autoinhibition. The resulting interface spans over 190 Å and harbors unfavorable contacts, activation phosphorylation sites, and nearly all residues with ID-associated mutations. Our studies suggest that this dynamic interface is coupled to an allosteric network responsive to phosphorylation and altered globally by mutations. Furthermore, we found that ID mutations increase the holoenzyme activity and perturb the phosphorylation rates, and the severe variants significantly increase the mitotic duration and error rates compared to the normal variant.

Keywords: Jordan’s syndrome; cryogenic electron microscopy; phosphatase.

Fig. 1.

The cryo-EM structure of the closed form of the E197K PP2A-B56δ holoenzyme. (A) The overall cryo-EM structure of the E197K holoenzyme in the closed form. The A subunit, PP2Ac, B56δ core, and N/C-arms are colored green, magenta, yellow, red, and blue, respectively. The electron density map for N/C-arms is colored cyan. The N/C-arms are in sticks, and the rest of the structure is in cartoon. Manganese ions are shown in gray spheres. (B) Mapping of the contact properties along the path of the super-dynamic interface spanning over 190 Å. The presentation of the structural model and the color scheme are the same as in (A).

Fig. 2.

Mapping and closeup views of unfavorable and hydrophobic contacts along the dynamic interface. (A) Distribution of unfavorable and repulsive contacts at the dynamic interface, highlighted in dashed cycles (Left) and illustrated at the upper right panel. The closeup stereoview of the central repulsive contacts with E200 is shown at the lower right panel. (B) Patches of hydrophobic contacts are in cycles and major hydrophobic interfaces are highlighted in thick cycles (Left). The closeup stereoviews for the latter are shown (Right). For (A) and (B), the color scheme is the same as in Fig. 1. The structural models are shown in cartoon and key residues at the interfaces in sticks.

Fig. 3.

Structural mechanism of dual auto-inhibition and roles of the B56δ N/C-arms. (A) The overall structure of the PP2A-B56δ holoenzyme highlights phosphorylation sites on the N/C-arms and residues essential for suppressing the phosphatase active site and the SLiM-binding groove. The structure is shown in cartoon and colored as in Fig. 1, except that the N/C-arms are colored pink and blue, respectively. Residues with key regulation functions and manganese ions (gray) are shown in spheres. (B) The closeup view of auto-inhibition at the phosphatase active site. The active site metal ions are in spheres. Active site residues and E574 from the C-arm are shown in sticks. (C) Truncations of either N- or C-arm increase the phosphatase activity of both the WT and E200K holoenzymes. (D) The closeup view of auto-inhibition at the B56 SLiM-binding groove, buttressed by extended interactions. (E) Examples of pulldown assays of the WT and E200K PP2A-B56δ holoenzyme full length (FL), truncation of N-arm (ΔN) or C-arm (ΔC) via GST-tagged CREB (99-161) (Upper) or GST-SYT16 (132-147) (Middle). One fifth of the holoenzyme input is shown (Lower). (F) All experimental repeats from (E) are normalized to the WT holoenzyme, and the scatter plots of the normalized results, averages of all repeats, and SD are shown. The P-values for the full-length versus truncated holoenzymes are calculated using Welch’s t test. For (B) and (D), the structural models are shown in cartoon, and the color scheme is the same as in (A). Residues at the interfaces are shown in sticks. H-bond interactions are shown in cyan dashed lines.

Fig. 4.

Responses of the PP2A-B56δ holoenzyme to phosphorylation of the arms. (A) The most frequently detected phosphorylation sites on the N/C-arms are highlighted (as summarized from https://www.phosphosite.org) (Left). The closeup views of these serine residues on the N/C-arms and their interactions at the dynamic regulation interface (Right). The color scheme is the same as in Fig. 3A. Phosphorylation at these sites is expected to create repulsive contacts. (B) Time-dependent changes in pS573 of the PP2A-B56δ holoenzyme by PKA in vitro. (C) Time-dependent increase of B56δ pS573 upon cellular activation of cAMP/PKA. (D) Time-dependent conformational changes of the PP2A-B56δ holoenzyme (WT and E198K) in mammalian cells upon cellular activation of cAMP/PKA by forskolin and rolipram. The non-phosphorylatable mutation of B56δ, S573Q, abolishes this response. (E) The model of phosphorylation-induced loosening of the N/C-arms. The color scheme is the same as in Fig. 1.

Fig. 5.

The allosteric network of the PP2A-B56δ holoenzyme, perturbation by B56δ ID mutations, and effects on holoenzyme functions. (A) The overall structure of the B56δ holoenzyme highlights the B56δ ID residues predominantly located at the dynamic regulation interface. (B) Illustration of the allosteric network of the PP2A-B56δ holoenzyme and its relationship to ID residues (green ball-and-stick) and activation phosphorylation sites on N/C-arms (orange ball-and-stick). The allosteric weights are shown as the thickness of blue wires. They are estimated from the enhanced-sampling simulations of the WT holoenzyme, modified from the cryo-EM structure of the E197K holoenzyme in the closed form. (C and D) The global perturbation of residue weights of ID residues and activation phosphorylation sites on the allosteric network by E198K and E200K. The results in (B–D) are the average of 20 enhanced-sampling simulation trajectories. (E) Pulldown of WT and mutant holoenzymes by GST-tagged CREB (99-161) assessed the effects of ID mutations on substrate SLiM binding. The data is normalized with the binding intensity of WT. The number of repeats, their scatter plots, averages, and SD are shown. The P values for comparison of WT and disease variants are calculated using Welch’s t test. (F) FRET assay measuring the distance (energy transferring efficiency) between the N terminus and C terminus of the B56δ subunit in WT and E198K PP2A-B56δ holoenzyme. Means ± SD are shown (N = 15). (G) The normalized Pair-distance function P(r) plot generated from SAXS analysis on the WT and E198K PP2A-B56δ holoenzymes. (H) Time-dependent phosphorylation of S573 in HEK293 cells expressing WT and mutant B56δ upon cellular cAMP/PKA stimulation. The experiments were repeated six times. Means ± SD was calculated for WT for comparison to disease variants. The P-values for time-dependent changes were calculated using the Jonckheere–Terpstra test.

Fig. 6.

Effects of B56δ ID mutations on cellular PP2A-B56δ activity, mitotic progression, and mitotic errors. (A) Illustration of the cellular PP2A-B56 activity assay. Phosphorylation-dependent B56 SLiMs in varied binding affinities (color coded) were fused to CDK5 and expressed from the same plasmid as the CDK5 activator p35. (B) Representative HA-immunoblot of HEK293T whole cell lysates expressing a control protein (CTRL, monomeric YE289RR-mutant B56δ), FLAG-tagged B56δ, or the B56δ-SmBiT/LgBiT conformation sensors (WT or E198K) after treatment with 2 µM of the CDK5 inhibitor dinaciclib for 0 to 15 min. Electrophoretic mobility of the CDK5-B56 SLiM substrate by SDS-PAGE ranges between ~55 and 75 kD and is influenced by cellular CDK5 and PP2A activity. (C) Densitometric quantification of the experiment in (B) (Means ± SD from duplicate blots). (D) Area-under-the-curve (AUC) quantification of dephosphorylation time courses (individual AUCs and Means ± SD from N = 3 independent transfections are plotted). (E) Mitotic durations from NEBD to anaphase onset (Left) and the frequency of mitotic errors (Right) in the cells shown in SI Appendix, Fig. S14 are plotted. Means ± SD was calculated and shown. The P values for comparison of WT and disease variants are calculated using the one-way ANOVA (Left) and two-tailed t test (Right).