Overall structure of LCMT-1. (A) Schematic showing the topology of the LCMT-1 structure. The canonical SAM-MT domain is shown in grey, the unique motifs inserted to different locations of the SAM-MT domain are in color, three of which (red, yellow and green) form the lid domain as indicated. Helices are shown as cylinders and strands as arrows. The N and C termini are labeled. (B) Overall structure of LCMT-1 bound to SAH in ribbon. The color scheme for MT and unique motifs is the same as in (A). BL indicates binding loop. As shown later, BL plays an important role in PP2A binding. (C) A slice of LCMT-1 surface showing the binding pockets for PP2A tail and the cofactor. SAH is bound and shown in spacefill. See also Figure S1.

Overall structure of the PP2A-LCMT-1 complex. (A) The level of methylation of PP2A core enzyme by LCMT-1 in vitro was determined by dot blot using an antibody that specifically recognizes the unmethylated PP2A. Three different amounts (24, 12, 6 ng) were spotted and PP2A without methylation was used as standard. The methylation level is around 50% or lower. (B) Illustration of LCMT-1 catalyzed covalent crosslink of the synthesized SAM mimic to PP2A tail. (C) Overall structure of the PP2A-LCMT-1 complex in ribbon diagram. The MT domain is colored magenta, unique motifs colored as in Figure 1A, C in blue, SAM mimic and L309 in spacefill. (D) The 2Fo-Fc electron density map at 1.8σ (cyan) for peptide “TPDYFL” in PP2A tail and the covalently bound SAM mimic (cylinder, colored by atom type). (E) A slice of surface showing that LCMT-1 binds to PP2A active site. Two catalytic metal ions are shown in red spheres.

Binding and methylation of PP2A tail by LCMT-1. (A) A slice of surface showing the binding pocket for PP2A tail, divided into the upper entry and the deep pocket. Residues R303, F308 and L309 and SAM mimic are labelled. (B) A stereo view of the upper entry. Residues from LCMT-1, PP2A tail and the PP2A core are shown in green, grey and cyan, respectively. (C) A stereo view of the deep pocket. The color scheme is the same as in (B). The H-bond between T29 in α1 of LCMT-1 and the sulfur atom of SAH is highlighted (inlet).

The contacts to PP2A active site are essential for PP2A methylation. (A) Interface of LCMT-1 to PP2A active site. The lid domain of LCMT-1, the protein core and the tail of the C subunit, are shown in ribbon, surface, and cylinder, respectively. (B) A close-up stereo view of the interface to PP2A active site. Residues from LCMT-1 and the C subunit of PP2A are shown in green and purple, respectively. (C) Pull-down assay using GST-tagged PP2A core enzyme immobilized on GS4B resin, and equal amount of LCMT-1 wild type (WT) or mutants in the mobile phase. Mutations to residues that contact PP2A active site or PP2A tail but do not participate in catalysis abolished binding to the PP2A core enzyme. Same amount of LCMT-1 WT or mutants was shown on SDS-PAGE (lower panel). (D) Methylation of PP2A core enzyme by WT LCMT-1 and LCMT-1 mutants. Mutations to residues that either contact PP2A active site, or bind PP2A tail, as well as mutations to catalytic residues abolished the methyltransferase activity of LCMT-1. (E) ITC measuring the binding affinity of the PP2A core enzyme to LCMT-1 and LCMT-1 T29V. T29V mutation increases the binding affinity by 4-fold. Note that PP2A-LMCT-1 interaction is endothermic. (F) The conformation of LCMT-1 active site pocket changes considerably to allow binding of PP2A tail. Based on structural alignment, PP2A tail overlaps with the LCMT-1 apo structure (the one most similar to the complex is shown). PP2A tail is shown in cylinder and the rest of the model in surface. See also Figure S2.

Precise control of PP2A methylation by the phosphatase activity of PP2A. (A) Simultaneous loss of LCMT-1 methylation activity and the phosphatase activity of PP2A core enzyme by incubation with PPi for 50 minutes. Neither activity could be recovered by addition of Mn⁺⁺. (B) PTPA stimulates the phosphatase activity and LCMT-1-mediated methylation of the C subunit. (C) Ceramide stimulates methylation of the C subunit. (D) Polycation peptides stimulate methylation of the PP2A core enzyme. (E) Simultaneous loss of LCMT-1 methylation activity and the phosphatase activity of the PP2A core enzyme caused by okadaic acid, and mild oxidation upon removal of DTT or in the presence of 20µM H₂O₂. For all panels, the results are representative of three independent experiments. Data values are the average of two independent assays. Error bars show standard deviation. See also Figure S3.

LCMT-1 Dominant-Negative Mutant T29V Attenuated Cell Proliferation. (A) Effects of recombinant expression of the human LCMT-1 (hLCMT-1) dominant-negative mutant, T29V, hLCMT-1 wild-type (WT), and the loss-of-function mutant F237D on PP2A methylation in the rat C6 glioma cell line. Expression of hLCMT-1 T29V reduced PP2A methylation without affecting the endogenous level of rat LCMT-1. (B) Images of C6 cells that stably express hLCMT-1 or its mutants cultured for 40 hr after stress. Dividing cells are highlighted by yellow dash cycles. Cells expressing hLCMT-1 T29V had much fewer dividing cells and much lower cell density. (C) The increase of cell numbers of C6 cell cultures expressing hLCMT-1 WT or mutants after stress. (D) The increase of viable cell signals of C6 cell cultures expressing hLCMT-1 WT or mutants after stress. (E) Knockdown of endogenous LCMT-1 by siRNA-induced cell death of C6 glioma cells. Knockdown was confirmed by western blot with an antibody that specifically recognizes rat LCMT-1. (F) LCMT-1 T29V inhibited the phosphatase activity of the PP2A core enzyme with a 4-fold higher efficacy than WT LCMT-1. (G) Structural alignment of the ACB0 g1 holoenzyme and the PP2A-C-LCMT-1 complex. A is shown in green, and C in the holoenzyme and the complex with LCMT-1 are in cyan and blue, respectively. B0 g1 is in yellow and LCMT-1 in magenta. A and C are in worm, B0 g1 in sphere, and LCMT-1 in tube. (H) The effect of PP2A phosphatase activity on cell fitness. The phosphatase activity of free C or the core enzyme is less controlled and might be detrimental to the cell; substrate-specific phosphatase activity of holoenzymes is essential for proper cellular function. Normal cells expressing WT LCMT-1 facilitate efficient transition of activated PP2A to holoenzymes, thus minimizing uncontrolled phosphatase activity. The effect of LCMT-1 knockdown or expression of LCMT-1 T29V is illustrated. For (C), (D), and (F), representative results of three independent experiments are shown. Data values are the average of three (C and D) and two (F) independent assays, respectively. Error bars represent the standard deviation. See also Figure S4.