A peek into tropomyosin binding and unfolding on the actin filament

Abstract

Background: Tropomyosin is a prototypical coiled coil along its length with subtle variations in structure that allow interactions with actin and other proteins. Actin binding globally stabilizes tropomyosin. Tropomyosin-actin interaction occurs periodically along the length of tropomyosin. However, it is not well understood how tropomyosin binds actin.

Principal findings: Tropomyosin's periodic binding sites make differential contributions to two components of actin binding, cooperativity and affinity, and can be classified as primary or secondary sites. We show through mutagenesis and analysis of recombinant striated muscle alpha-tropomyosins that primary actin binding sites have a destabilizing coiled-coil interface, typically alanine-rich, embedded within a non-interface recognition sequence. Introduction of an Ala cluster in place of the native, more stable interface in period 2 and/or period 3 sites (of seven) increased the affinity or cooperativity of actin binding, analysed by cosedimentation and differential scanning calorimetry. Replacement of period 3 with period 5 sequence, an unstable region of known importance for cooperative actin binding, increased the cooperativity of binding. Introduction of the fluorescent probe, pyrene, near the mutation sites in periods 2 and 3 reported local instability, stabilization by actin binding, and local unfolding before or coincident with dissociation from actin (measured using light scattering), and chain dissociation (analyzed using circular dichroism).

Conclusions: This, and previous work, suggests that regions of tropomyosin involved in binding actin have non-interface residues specific for interaction with actin and an unstable interface that is locally stabilized upon binding. The destabilized interface allows residues on the coiled-coil surface to obtain an optimal conformation for interaction with actin by increasing the number of local substates that the side chains can sample. We suggest that local disorder is a property typical of coiled coil binding sites and proteins that have multiple binding partners, of which tropomyosin is one type.

Figure 1. Models illustrating the P2 and P3 regions in wildtype and mutant tropomyosins.

The side chains of the alanine clusters (magenta) and consensus actin binding sites (cyan) are illustrated on a ribbon model of the 7 Å structure (pdb ID: 1C1G; [60]). Period 2, residues 46–69, and period 3, residues 88–111, are enlarged with the side chains of interface Ala residues (in magenta, space filling), canonical interface residues (green) and consensus residues (cyan, C_β in spacefill). The colors used are the same as in Table 1.

Figure 2. The effect on the thermal stability of the cluster and period replacement mutants.

Fraction folded as measured by relative ellipticity at 222 nm as a function of temperature in 500 mM NaCl, 10 mM sodium phosphate pH 7.5, 1 mM EDTA, 0.5 mM DTT. The tropomyosin concentration was 0.01 mg/ml (1.6 µM). The T_Ms are reported in Tables 2 and 3. The fraction folded is relative to the mean residue ellipticity at 0°C where the proteins were fully folded. A. Symbols: •, wildtype; ○, P3Shift, ▾, P2Shift. B. Symbols: •, wildtype; ○, P2P3Shift. C. Symbols: •, wildtype; ○, P5→P3.

Figure 3. DSC scans of single cluster shift mutants with TnT_70–170 in the presence and absence of F-actin.

Tropomyosin (15 µM) and TnT_70–170 (18 µM) were mixed with phalloidin (36 µM) stabilized F-actin (24 µM) in 100 mM NaCl, 10 mM Hepes pH 7.0, 2 mM MgCl₂, 1 mM DTT and heated as described in Materials and Methods. The second (with F-actin-solid lines) and third (post-F-actin denaturation-dotted lines, TM- TnT_70–170) scans are shown. A. wildtype (black); B. P2Shift (green); C. P3Shift (red); D. Combination of A-C in the presence of F-actin with the color scheme as indicated in A-C. The P2Shift and P3Shift mutants bind F-actin with higher affinity than wildtype. E. Experimental procedure: An excess of unacetylated recombinant tropomyosins and the TnT_70–170 fragment was added to phalloidin-stabilized F-actin and the sample was heated a total of three times. An excess was added to ensure maximal binding of tropomyosin. The experiments were done in the presence of TnT_70–170 because unacetylated tropomyosin binds poorly to F-actin. First, the sample was heated from 0–70°C (black line) and then cooled. This curve shows the dissociation of tropomyosin from F-actin (main peak) and the unfolding of TM-TnT_70–170 . Second, the sample was heated to 90°C (green line) to denature the F-actin and subsequently cooled. In the third scan (magenta line), post-F-actin denaturation, the sample was heated to 70°C and only the TM+TnT_70–170 signal is present. The first two scans are similar, and mark the tropomyosin unfolding/dissociation thermal peak, consistent with previously published results , , . The thermal unfolding of tropomyosin is almost completely reversible, while F-actin denaturation is irreversible.

Figure 4. Excimer fluorescence of PIA-labeled (at Cys190) and light scattering of unlabeled wildtype, P2Shift, and P3Shift mutants.

A. PIA wt; B. PIA P2Shift; C. PIA P3Shift. Excimer fluorescence and light scattering experiments: Tropomyosin (6 µM) and TnT_70–170 (9 µM) were mixed with phalloidin (27 µM) stabilized F-actin (18 µM). Light scattering (red solid line and left axis with red label) was conducted with unlabeled tropomyosin due to interference from excimer peak in light scattering experiments with labeled tropomyosin. Excimer fluorescence (right axis) in the presence of actin (black solid line) and absence of actin (black dotted lines). Buffer conditions for all of the experiments: 100 mM NaCl, 10 mM sodium phosphate pH 7.5, 2 mM MgCl₂, 1 mM DTT. Experimental Procedure: The excimer temperature scans were repeated three times at a 1°C/min scan rate . First, the sample was heated from 0–70°C and then cooled again. Second, the sample was heated to 90°C to denature the F-actin and subsequently cooled. In the third scan, post F-actin denaturation, the sample was heated to 70°C and the excimer peak is detected in the absence of actin. The first two scans, in the presence of actin, were marked with appearance of the excimer peak at higher temperatures, indicating actin-induced stabilization.

Figure 5. Excimer fluorescence of PIA-labeled and light scattering of unlabeled P2 and P3 mutants.

A. P2 control (C190S/L71C) B. P2Shift C190S/L71C C. P3 control (C190S/L113C). D. P3Shift C190S/L113C. Light scattering (red solid line and left axis with red label) was conducted with unlabeled tropomyosin due to interference from excimer peak in light scattering experiments with labeled tropomyosin. Excimer fluorescence (right axis) in the presence of actin (black solid line) and absence of actin (black dotted lines). Buffer conditions and the experimental procedure are as described for Figure 4. The P2Shift and P3Shift mutations locally stabilize the interaction with actin.

Figure 6. The effect the mutations on the actin affinity measured by cosedimentation with F-actin.

Binding to filamentous actin. Tropomyosin (0.1–10 µM, depending on the tropomyosin) and 0.12–12 µM troponin T_70–170 were combined with 5 µM actin and sedimented at 20°C in 250 mM NaCl, 10 mM TrisHCl, pH 7.5, 2 mM MgCl₂, and 0.5 mM DTT. Stoichiometric binding of 1 tropomyosin: 7 actins is represented by the 1.0 fraction of maximal binding. The apparent K_apps are reported in Table 2. A. Symbols: •, wildtype; ○, P3Shift, ▾, P2Shift. B. Symbols: •, wildtype; ○, P2P3Shift. C. Symbols: •, wildtype; ○, P5→P3. The mutations affect the affinity and cooperativity of binding.