Supplementary Figure 7: Trimethylation of talin1 at Lys2454 is critical for binding to F-actin.

(a) Recombinant talin1(2300-2451) containing C-terminal actin-binding site and dimerization domain was purified and incubated with F-actin. Binding of the recombinant talin1(2300-2451) (wild-type or with mutations at position K2454) to F-actin was determined using the actin-co-sedimentation assays with 4 µM of talin1(2300-2541) and indicated amounts of F-actin; talin1 and actin (1 µM) input (In), co-sedimented talin and actin (Pellet), remaining talin1(2300-2541) in supernatant (Supernatant). Representative images from 1 of 3 independent experiments are shown. (b) Quantitative analysis of the binding of talin1 mutants to various concentrations of F-actin was determined by actin co-sedimentation assay. Means ± s.d. of 3-5 independent experiments are shown, talin1(2300-2541)-K2454 (black squares), talin1(2300-2541)-K2454Q (blue diamonds), talin1(2300-2541)-K2454F (red triangles), talin1(2300-2541)-D2447N (filled purple circle), talin1(2300-2541)-V2444D (open green circle). (Two-way ANOVA with replication). ***P=1.8x10-8 (top), 1.2x10-11 (bottom). (c) The dissociation rate of F-actin from various talin1(2300-2451) was analyzed using Surface plasmon resonance (SPR) measurement as described in materials and methods. Dissociation response for wild-type talin1(2300-2541)-K2454/F-actin at 4, 20 and 40 µM was analyzed, and globally fit rate of dissociation was kd=1.4x10-5 s-1 (second panel from top). Dissociation response for talin1(2300-2541)-K2454Q/F-actin at 4, 20 and 40 µM was analyzed, and globally fit rate of dissociation was kd=4.14x10-6 s-1 (P < 0.0001 against wild-type talin1(2300-2541)-K2454) (third panel from top). Dissociation response for talin1(2300-2541)-K2454F/F-actin complex was not detectable at 4 µM, but consistently measurable at 10, 20 and 40 µM, and globally fit rate of dissociation was kd = 1.4 x 10-5 s-1 (P=0.09 against wild-type talin1(2300-2541)-K2454) (bottom panel). A typical sensorgram showing overlay responses of complex concentrations from initial capturing followed by dissociation (top panel). The initial capturing phase of other graphs was shortened to improve clarity. (d) Biotin labeled dimeric talin1 peptides (aa2450-2461) containing tri-methylated (open bars) or un-modified K2454 (filled bars) were incubated with various concentrations of F-actin and analyzed for their co-sedimentation as described in Fig. 7d except that the co-sedimentated peptides were visualized on dot blots by streptavidin-HRP. Error bars represent the mean ± s.d. of 3 independent experiments. *P=0.026 (left), 0.013 (right). (e-g) The talin1-K2454me3 antibody was generated by immunizing rabbits with (Ac-SEAMK(Me)3RLQAAG)2-K-C-amide branched peptides conjugated to carrier protein followed by affinity depletion and purification (YenZym antibodies LLC.). The specificity of this talin1-K2454me3 antibody was verified by dot-blotting of non-methylated and tri-methylated peptides (e), immunoblotting (IB) of endogenous talin1 (Endo-talin1) using tri-methylated talin1-K2454 peptide-blocking (1 µg/ml) as negative controls (f), and recombinant GFP-talin1 fusion protein using GFP-talin1-2454 mutants as negative controls (g).