Precise Binding of Tropomyosin on Actin Involves Sequence-Dependent Variance in Coiled-Coil Twisting

Abstract

Often considered an archetypal dimeric coiled coil, tropomyosin nonetheless exhibits distinctive "noncanonical" core residues located at the hydrophobic interface between its component α-helices. Notably, a charged aspartate, D137, takes the place of nonpolar residues otherwise present. Much speculation has been offered to rationalize potential local coiled-coil instability stemming from D137 and its effect on regulatory transitions of tropomyosin over actin filaments. Although experimental approaches such as electron cryomicroscopy reconstruction are optimal for defining average tropomyosin positions on actin filaments, to date, these methods have not captured the dynamics of tropomyosin residues clustered around position 137 or elsewhere. In contrast, computational biochemistry, involving molecular dynamics simulation, is a compelling choice to extend the understanding of local and global tropomyosin behavior on actin filaments at high resolution. Here, we report on molecular dynamics simulation of actin-free and actin-associated tropomyosin, showing noncanonical residue D137 as a locus for tropomyosin twist variation, with marked effects on actin-tropomyosin interactions. We conclude that D137-sponsored coiled-coil twisting is likely to optimize electrostatic side-chain contacts between tropomyosin and actin on the assembled thin filament, while offsetting disparities between tropomyosin pseudorepeat and actin subunit periodicities. We find that D137 has only minor local effects on tropomyosin coiled-coil flexibility, (i.e., on its flexural mobility). Indeed, D137-associated overtwisting may actually augment tropomyosin stiffness on actin filaments. Accordingly, such twisting-induced stiffness of tropomyosin is expected to enhance cooperative regulatory translocation of the tropomyosin cable over actin.

Figure 1

Conserved noncanonical core residue Asp137 present along the tropomyosin “hydrophobic stripe” yields atypical coiled-coil interactions. (A) A helical wheel diagram of a canonical dimeric coiled coil is shown. Heptad positions are labeled a–g and a′–g′ for the respective helices of such a dimer; the figure is adapted from (14). (B) Striated muscle tropomyosin (Tpm1.1) sequence annotation is shown; the figure is similar to (18, 19, 49) to emphasize residues within heptad repeats (a, b, c, d, e, f, g) in each chain of the tropomyosin coiled-coil homodimer, with the first residue in each heptad repeat numbered. The so-called α-zones of each pseudorepeat of tropomyosin that locate close to actin subdomains 1 and 3 are shaded in tan (β-zones found over subdomains 2 and 4 are white) (18, 19). Generally hydrophobic residues are localized at a and d residue positions (found within the black border in the diagram). d-position aspartate 137 is highlighted in red. (C) Enlargement of (B) schematically highlighting unconventional R133–D137, g-d pairing in wild-type tropomyosin and canonical R133–E138, with e-g pairing expected in the D137L mutant is shown (Results). (D) A schematic representation of tropomyosin residue 137 inducing interactions of residues 139 and 142 with actin is shown (Results). To see this figure in color, go online.

Figure 2

Local coiled-coil contacts between component α-helices in tropomyosin pseudorepeat 4. Averaged positions of selected amino acid side chains during MD are shown. (A) Wild-type tropomyosin (gold) displays atypical intrastrandg-d contact (dotted lines) between D137 and R133 but no local e-g links. Here, a-position M127 and M141 may prevent local collapse of the coiled coil. (B) Mutant D137L makes canonical interstrand e-g contact between R133 and E138 (dotted lines). Residues discussed here and in the text are color coded. To see this figure in color, go online.

Figure 3

Overtwisting of tropomyosin during MD directs charged side chains in pseudorepeat 4 toward actin. A view of averaged actin-tropomyosin structure during MD highlights electrostatic contacts between side chains of wild-type tropomyosin residues 125, 139, and 142 on pseudorepeat 4 and residues 25, 147, 328, and 326 of its neighboring actin subunit along the filament (actin subunit number 4, colored redbrick); a key to interacting actin and tropomyosin residues is located right of figure. Actin subunit 3 toward the barbed end of the filament is colored light green, and actin subunits on F-actin’s opposite long-pitch helix are colored light violet, dark cyan, and dark violet. Actin subunit and tropomyosin pseudorepeat numbering was done according to (11). Over the last 5 ns of MD, root mean-square deviation values for the actin-tropomyosin complex were 1.77 ± 0.43 Å for explicit solvent simulations and 1.53 ± 0.23 Å for implicit solvent runs. To see this figure in color, go online. To see this figure in color, go online.

Figure 4

Fitting average MD structures of tropomyosin to the Lorenz-Holmes canonical model of tropomyosin linked to F-actin. (A and B) Idealized (canonical) tropomyosin (cyan) form fitted to the F-actin helix using algorithms outlined in Lorenz et al. (33) and mounted on F-actin as in Li et al. (11) is shown; then, MD-average structures of tropomyosin are superposed. In (A), the best fitting of the averaged MD structure of actin-free tropomyosin (yellow) to canonical tropomyosin is shown, and in (B), the best fitting of the averaged MD structure of actin-bound tropomyosin (gold) to canonical tropomyosin is shown. Note that in both (A) and (B) the C-terminal zone of MD-averaged tropomyosin fits to the canonical tropomyosin well. However in (A), gradual undertwisting at the N-terminal half of MD-averaged, isolated, actin-free tropomyosin (yellow) causes it to diverge from the canonical tropomyosin model (cyan) and therefore from the F-actin surface. In (B), overtwisting of the central domain of MD-averaged actin-bound tropomyosin (gold) leaves the overall trajectory of tropomyosin well matched to the F-actin helix. The position of tropomyosin residue 137 is indicated for reference. To see this figure in color, go online.

Figure 5

Tropomyosin twisting during MD influenced by actin. The graphs show the average deviation of the cumulative twist angle around the superhelical axis from a canonical tropomyosin superhelical structure. The cumulative angle is defined starting at residue 252 and proceeding toward the termini of tropomyosin. (A) In the absence of actin (red curve), the twist of the C-terminal portion of the molecule fluctuates around the ideal superhelix; however, the N-terminus shows considerable untwisting. This untwisting is not due to the extra conformational freedom of the actin-free tropomyosin, because MD simulations, when tropomyosin is constrained to a superhelical shape mimicking that assumed on F-actin, show an almost identical pattern of untwisting (data not shown). (B) In the presence of actin (purple curve), the C-terminal portion of the molecule shows similar twisting behavior; however, the N-terminus shows overtwisting rather than untwisting. Here, the presence of actin drives the tropomyosin to specific twists to optimize its interactions with actin. This twisting behavior is observed whether the simulations use an actin model based on fiber diffraction studies (44) or on electron cryomicroscopy (45), and thus is independent of the starting actin structure. To see this figure in color, go online.