ReSeARch pApeR BioArchitecture 1:5, 250-255; September/October 2011; © 2011 Landes Bioscience Structural implications of conserved aspartate residues located in tropomyosin’s coiled-coil core Jefrey R. Moore,1 Xiaochuan (edward) Li,1 Jasmine Nirody,1 Stefan Fischer,2,* and William Lehman1,* 1 Boston University; School of Medicine; Boston, MA USA; 2University of heidelberg; heidelberg, Germany . e nc Keywords: tropomyosin, coiled-coil, flexibility, molecular dynamics, heptad repeat e i c s o i B . e s t e u d b i n r a t L s i 2 d 1 t 0 o 2 n o © D polar residues lying between adjacent α-helical chains of coiled-coils often contribute to coiled-coil curvature and lexibility, while more typical core hydrophobic residues anneal the chains together. In tropomyosins, ranging from smooth and skeletal muscle to cytoplasmic isoforms, a highly conserved Asp at residue 137 places negative charges within the tropomyosin coiled-coil core in a position which may afect the conformation needed for tropomyosin binding and regulatory movements on actin. proteolytic susceptibility suggested that substituting a canonical Leu for the naturally occurring Asp at residue 137 increases inter-chain rigidity by stabilizing the tropomyosin coiled-coil. Using molecular dynamics, we now directly assess changes in coiled-coil curvature and lexibility caused by such mutants. Although the coiled-coil lexibility is modestly diminished near the residue 137 mutation site, as expected, a delocalized increase in lexibility along the overall coiled-coil is observed. even though the average shape of the D137L tropomyosin is straighter than that of wild-type tropomyosin, it is still capable of binding actin due to this increase in lexibility. We conclude that the conserved, non-canonical Asp-137 destabilizes the local structure resulting in a local lexible region in the middle of tropomyosin that normally is important for tropomyosin steady-state equilibrium position on actin. Introduction Tropomyosin, the α-helical coiled-coil actin binding protein, associates superhelically along F-actin and polymerizes via end-to-end interactions to form a continuous cable-like strand. Tropomyosin mechanically stabilizes the thin filament and increases its flexural rigidity.1 In non-muscle cells, tropomyosin also regulates the interaction of many actin binding proteins, while in muscle, tropomyosin specifically controls the cooperative binding of myosin with actin, and hence motor activity. The α-helical coiled-coil is a common structural motif found in a wide array of structural proteins such as keratin, myosin and tropomyosin. Typically the coiled-coil is defined by the heptad amino acid repeat sequence (abcdefg)n. A 3.5 residue periodicity along the supercoil axis aligns a and d position hydrophobic residues in the adjacent α-helix, thus stabilizing the interaction interface,2,3 while residues at the e and g positions are typically oppositely charged and flank the hydrophobic core, forming salt bridges between apposed helical chains. Right-handed α-helices in coiled-coils wrap around each other to form a quaternary structure with function related to the specific ‘design’ of their coiled-coil domains.4 In tropomyosin the superhelical coiled-coil matches the contours of the F-actin helix.5,6 Interestingly, the canonical hydrophobic a/d seam along the heptad repeat of tropomyosin is disrupted by a highly conserved substitution placing an aspartic acid in the d position at residue 137 where typically a hydrophobic residue would reside. Biochemical studies have shown a decrease in proteolytic susceptibility when the Asp at 137 was replaced with a canonical Leu (D137L). A decrease in proteolytic susceptibility may result from this stabilization of the coiled-coil interaction and potentially a reduction in the flexibility of the tropomyosin molecule.7 Here we used Molecular Dynamic (MD) simulations to directly evaluate the structural consequences of the D137L mutation on coiled-coil structure, curvature, and flexibility. MD predicts that introduction of the leucine residue reduces both curvature and flexibility near the site of the mutation, which results in a straighter molecule. Although the overall structure of the molecule is disrupted, we find that tropomyosin containing the D137L mutation spends a fraction of time in a structure similar to the one required for tropomyosin to be able to wrap natively around F-actin, thus explaining the ability of the mutant tropomyosin to still bind to actin.7 Results Biochemical studies previously indicated that mutation of the conserved Asp at position 137 to Leu resulted in a decrease in proteolytic susceptibility, suggesting a region existed in the core of the tropomyosin molecule that is accessible to enzymatic cleavage.7 To *Correspondence to: William Lehman; Email: wlehman@bu.edu Submitted: 09/06/11; Accepted: 09/15/11 http://dx.doi.org/10.4161/bioa.1.5.18117 250 BioArchitecture Volume 1 Issue 5 ReSeARch pApeR ReSeARch pApeR adopting the tropomyosin shape first proposed by Lorenz and Holmes5 and confirmed by Li et al.8 The supercoil fits to the long-pitch F-actin helix, as it is presumed to be when bound to muscle thin filaments.6 During the simulation, the coiled-coiled structures of the wild-type, D137L/C190A, D137L, and C190A tropomyosins remained completely intact during their respective 30 ns runs (the double mutant was run twice for 30 ns with no apparent distinctions between runs). For example, no unfolding of the α-helices, lengthwise separation of the two adjoining Figure 1. Average tropomyosin conformations from MD simulations. (A) Average structures helices, or fraying at their ends was observed. for the wild-type tropomyosin sequence (light blue) and the control c190A mutant sequence The bending of tropomyosin observed (gold). Note the overall curvature of the c190A tropomyosins matches closely to wild-type during the simulation always was gradual and 8 tropomyosin, which closely matches the long-pitch helical repeat of actin. (B) Average struccontinuous, without signs of localized kinktures for the D137L (red) and D137L/c190A (blue) mutant tropomyosin sequences. Overall, the ing or jointed areas (Fig. 1A, Supplementary average structure of the mutant tropomyosin deviates from wild-type and control average conformations resulting in a straighter molecule. (c) Tropomyosin curvature is assessed via the Movies), in agreement with rotary shadowed mean end-to-end bending angle, θ, which is calculated by comparing the tangent angles of electron microscopic images of purified tropothe central coiled-coil axis of the N and c-terminal 15 residues (red arrows). Since tropomyosin myosin molecules lying on mica.9 Although is a curved molecule, lexibility is determined by examining δ, the angular delection of the rod there are no distinct kinks or pivot points from its averaged curved shape. observed in the mutant molecule, the overall tropomyosin shape is altered by the mutaTable 1. Tropomyosin curvature and flexibility. tions. While the end-end bending angle, θ, of Global Properties Local Properties the wild-type and C190A tropomyosin molEnd-to-end Flexibility δ (°) Curvature Flexibility δ (°) ecules agree with that of the F-actin bound angle θ (°) θ (°/Å) model, the D137L mutant tropomyosin molWild-type 36.8 24.3 0.26 ±0.03 4.49±0.26 ecules were observed to be less bent (Table 1; Fig. 1A and B): The mean end-to-end bendC190A 36.1 28.6 0.22 ±0.02 4.50 ±0.30 ing angle during MD for the C190A/D137L D137L 2.63 32.7 0.09 ± 0.04** 4.09 ±0.38** double mutant and the D137L single mutant C190A/D137L 24.75 44.9 0.07 ± 0.02** 4.13±0.39* (24.75° and 2.63°, respectively), is considerend-to-end angle, θ, and flexibility, δ, for each tropomyosin are reported (Fig. 1C). Local ably lower than that of C190A control and curvature and flexibility for the region surrounding the D137L represent the mean +/- S.D. for wild-type tropomyosin (36.1° and 36.8°, residues 131-143, *p < 0.05, **p < 0.01 compared to wild-type. respectively). The difference in end-to-end bending angle gives the impression that the assess the structural importance of tropomyosin residue 137, we mutant molecules are much less flexible than the wild-type and studied both the single mutation, D137L, as well as the double C190A controls. However, the end-to-end bending angle also promutant, D137L/C190A, using MD simulations. D137L/C190A vides a measure of tropomyosin shape and thus the reduction likely and C190A tropomyosin were chosen for study to be consistent reflects an overall straightening of the mutant molecule. with previous in vitro work,7 which utilized the C190A mutaSince tropomyosin is a curved molecule, the proper gauge of tions to avoid potential complications due to possible disulfide tropomyosin flexibility is the average deviation from the average cross-linking at residue Cys 190. The effects of the Leu substitu- structure, δ, during the simulation (Fig. 1C). As can be seen in tion were compared with those of both C190A mutants and a Table 1, both the C190A and the D137L single mutations modwild-type control, which retains the native Cys at position 190. estly increased flexibility (δ = 28.6° and δ = 32.7° for C190A and We find that substitution of Cys at position 190 with Ala has only D137L respectively). The increased flexibility conferred by the minor delocalized effects on the overall structure of the tropo- two mutations appears to be more than additive with the double myosin. However, substitution of Asp 137 with Leu resulted in a mutant yielding a much increased flexibility, δ = 44.9° (Table 1). generalized straightening of the molecule, which is stiffer in the Therefore the smaller θ reflects an overall straightening of the molregion surrounding the mutation but globally more flexible. ecule since the flexibility increases. . e nc e i c s o i B . e s t e u d b i n r a t L s i 2 d 1 t 0 o 2 n o © D Average Structure Anisotropic Bending MD simulations were begun by using mutant tropomyosin reference structures in the superhelical coiled-coil conformation, thus As we have previously shown,8 wild-type tropomyosin does not uniformly sample all bending directions. Instead, as can be seen www.landesbioscience.com BioArchitecture 251 in Figure 2 where the C-terminal position of tropomyosin is tracked during MD (while the N-terminus orientation is fixed at the origin) the wild-type tropomyosin molecule bends anisotropically and primarily samples positions in quadrants 3 and 4 (in the left planar slices). Thus while the wild-type tropomyosin molecule flexes it always comes back to the starting molecule conformation, i.e., where the superhelical shape of isolated tropomyosin matches its superhelical binding shape on F-actin. In contrast, the D137L and D137L/C190A mutants spend less time near the Lorenz-Holmes conformation (Square, Fig. 2) and are on average straighter (Fig. 1). However, they sample a larger amount of conformational space and are thus more flexible (Table 1). Importantly, the D137L mutants are able to occasionally visit the shape of the Lorenz-Holmes conformation (Fig. 2C and 2D). . e nc e i c s o i B . e s t e u d b i n r a t L s i 2 d 1 t 0 o 2 n o © D Local Structure As shown previously, the greatest degree of local curvature is observed in the middle portion of wild-type tropomyosin near residue 137; i.e., there is a modest local bend in the surrounds of this residue.8 We therefore assessed the effects of the D137L mutation Figure 2. c-terminal variation plots. (A–D) x,y coordinates of the tropomyosin on the local area proximal to the mutation (Fig. 3). c-terminus are plotted for each frame of the simulation after superposing the irst Namely, we examined the local curvature of the aver15 residues at the N-terminus, orienting them along the z-axis (out of page), and aged structures (a probe of intrinsic bending) and aligning them at the origin. each point represents the projection of the position of the local deviations from the average structures (a the c-terminal residue for each conformer observed during the MD simulation. The probe of local flexibility). As can be seen in Figure average c-terminal position of the MD structures for each tropomyosin is indicated in their respective panels (open diamond) for comparison with that of the Lorenz– 3A, there is a high degree of local curvature in the holmes model (open square). Wild-type (A) and c190A (B) tropomyosins display the residues immediately preceding Asp 137 for the wildcharacteristic anisotropic bending with the c-terminus frequently in quadrants 3 type tropomyosin. This curvature was diminished in and 4. On the other hand the straighter D137L (c) and D137L/c190A (D) spend much the mutant tropomyosins, indicating that the effects more time in quadrants 1 and 2. of substitution in the tropomyosin sequence can be both short- and long-ranged. Here, substitution of Asp with Leu at position 137 caused an overall reduction in curvature substitution with the leucines (Fig. 4B). The Leu side chains at in the surrounding region (Fig. 3B; Table 1). While the local position 137 within the hydrophobic core result in a straighter curvature decreases near the site of the mutation (Fig. 3B) other molecule in the region surrounding the mutation (Fig. 4C), in regions also behave differently from the wild-type sequence (Fig. accord with the straight shape of a canonical coiled-coil. 3A). Some regions displayed an increase in curvature while others displayed a decrease resulting in an average structure that is less Discussion curved than wild-type (Fig. 1). Local flexibility, namely the variance in curvature, is also A charged aspartic acid at residue 137 is a highly conserved substiaffected by mutation of residue 137 to Leu. Changes were tution in the d position of the canonical coiled-coil hydrophobic observed throughout the length of the molecule (Fig. 3C), sug- core of tropomyosin. The substitution stems from the constigesting some long-range effects of the mutation on supercoiling. tutive exon 4 and is found in both low molecular weight and In the region immediately surrounding the 137 mutation site high molecular weight tropomyosins.10 Here we used molecular MD predicts that the coiled-coil flexibility would be modestly dynamics simulations to characterize the structural consequences diminished (Fig. 3D; Table 1). of this unique Asp insertion in tropomyosins. The principal findThe effects of the Leu substitution on the packing of the ing of this study is that Asp at position 137 is critically important coiled-coil can be seen in Figure 4. In wild-type tropomyosin for maintaining an average tropomyosin structure that matches containing Asp 137, a “cavity” is introduced between the tropo- the superhelical shape of its actin binding site. Substitution of myosin helices as a result of the electrostatic repulsion between Asp137 with a more typical Leu hydrophobic residue causes an the two side chains. The typical “knobs-into-holes” packing overall straightening and a local stiffening of the region surroundexpected with a canonical hydrophobic side chain is noted after ing the mutation. This could be thought to reduce the ability of 252 BioArchitecture Volume 1 Issue 5 to be a preferential site for tryptic cleavage,14 and V8 protease cleaves at Glu-131.15 Substitution of Asp 137 with Leu reduces proteolytic susceptibility suggesting that the D137 could destabilize or otherwise expose the surrounding region of tropomyosin. Since the proteolytic susceptibility is related to the fraction of time that the molecule is in the cleavable state and the susceptibility to proteolysis is diminished with the Leu substitution, it was proposed that Asp 137 is responsible for conformational fluctuation in this region. This is consistent with the reduced “cavity” at position 137 observed when comparing the wild-type and the mutant MD simulations (Fig. 4A and B). In the myosin rod domain, introduction of an aspartic acid residue into the hydrophobic core at the a position of the coiledFigure 3. Local changes in tropomyosin curvature and lexibility induced by the D137L mutation. (A and B) coil produced dramatic localized The local curvature of the average structure. Local lexibility (c and D) is determined as the time average of bending points that correlated the local deviations from the average structure. Local curvature and lexibility for the region immediately with the site of the mutation.16 surrounding residue 137 (arrow) are shown in panels b and d respectively. Wild type is shown in cyan, To a degree our observation of while the c190A control, c190A/D137L and D137L mutant tropomyosins are shown in gold, blue and red modestly increased flexibility respectively. in tropomyosin with Asp-137 inserting into the d-position of the D137L mutant to bind to actin. However, the overall flexibil- the hydrophobic ridge parallels this observation; however, the ity of the D137L mutants increases significantly compared with present MD simulations indicate a gradual curvature with no the wild-type tropomyosin. The mutant molecules still can spend distinct pivot points or kinks for both mutant and wild-type a fraction of time in a conformation near to that competent for tropomyosin molecules (Fig. 1; Movies 1–3). The distinct actin binding, hence explaining the experimentally shown ability structural geometry17 and the amino acid selection at the a and of the mutant molecules to retain binding affinity for actin.7 d positions of coiled-coils18 likely accounts for distinct effects In close proximity to Asp137, Glu139 and Glu142 are among of substitution with a charged side chain at the respective posithe few amino acids on tropomyosin involved in binding to basic tions in the two proteins. amino acids on F-actin.11,12 Thus the role of these negatively Several other non-canonical residues are present in the charged residues differ: Glu 139 and Glu 142 are responsible for tropomyosin heptad repeat. For example, Gly 126 in the g posifine tuning the binding precision of tropomyosin to actin while tion might perturb coiled-coil structure by disrupting α-helical the curvature conferred by Asp 137 facilitates the overall binding structure and inter-chain salt bridges. Recently it was shown ´ association of tropomyosin and F-actin.13 Numerous mutations that, similar to the Leu substitution at Asp 137, mutation of in tropomyosin have been reported to be associated with various Gly 126 to Ala or Arg also dramatically reduced proteolytic cardiomyopathies and other muscle disorders. We note that none susceptibility at Arg-133 in both smooth and skeletal muscle of these are at residue 137, highlighting the essential nature of tropomyosin,19 suggesting that both D137 and G126 destabilize this conserved feature of the protein’s architecture; mutation at the middle region of the tropomyosin molecule, consistent with Asp137 presumably would be lethal. Moreover, some degree of the decreased stiffness seen in the region between residues 130 conformational flexibility at this site might be needed to allow and 140 of wild-type tropomyosin (Fig. 3). accurate movement of tropomyosin on actin between regulatory When bound to actin, the D137L tropomyosin signifistates of the thin filament. cantly increases the ATPase activity of myosin in the presence The region surrounding Asp 137 has been shown to be par- of thin filaments at maximal calcium activation.7 This effect can ticularly susceptible to proteolysis. Arg 133, has been shown be explained by considering the consequences of the three state . e nc e i c s o i B . e s t e u d b i n r a t L s i 2 d 1 t 0 o 2 n o © D www.landesbioscience.com BioArchitecture 253 model of thin filament regulation. Tropomyosin molecules are known to undergo regulatory movement azimuthally over the surface of the actin filament. In the so-called “blocked” position, tropomyosin lies over successive actin subunits on their outer domains while in the “closed” position, tropomyosin lies on the actin inner domain near the junction of the inner and outer domains.20 Myosin binding results in a further movement to the “open” position. Movement between these three discrete positions on actin affect myosin binding, actomyosin ATPase, and actin filament velocity. In the absence of calcium, myosinbinding sites on actin are covered by tropomyosin constrained in the blocking position. The transition from the blocked state to the closed state is regulated by Ca 2+ binding to troponin while full activation requires the myosin induced movement of tropomyosin from the closed position to the open state of the thin filament. This latter component is cooperative in nature and results in the rapid switching on of muscle during activation. The lack of effect on calcium sensitivity7 suggests that the structural alterations caused by the mutation do not affect the blocked-open equilibrium. Instead the increased activity observed for the mutation could be explained by a shift in either the open-closed equilibrium or the cooperative unit size. It seems reasonable that the introduction of a flexible region near Asp 137 in a molecule less well designed to match the F-actin helical contour could affect the ability of tropomyosin position switching to be propagated along the thin filament and decrease cooperative unit size. In contrast, the D137L mutation may disrupt the open-closed equilibrium via an altered interaction with troponin. Consistent with this notion, recent experiments that show cross-linking of tropomyosin residue 146 to the C-terminus of troponin I.21 indicate that a direct interaction between tropomyosin and troponin I may modulate tropomyosin position. Alterations in the interaction of the D137L mutant tropomyosin with the C-terminus of troponin I could result in a perturbation of the tropomyosin with an equilibration toward an open or near-open state as has been proposed from related studies on a the effect of a troponin I truncation mutant.22 . e nc e i c s o i B . e s t e u d b i n r a t L s i 2 d 1 t 0 o 2 n o © D Conclusions 1. 254 Greenberg MJ, Wang CL, Lehman W, Moore JR. Modulation of actin mechanics by caldesmon and tropomyosin. Cell Motil Cytoskeleton 2008; 65:15664; PMID:18000881; http://dx.doi.org/10.1002/ cm.20251 Materials and Methods Molecular Dynamics: Molecular Dynamics simulations were run on homology models of tropomyosin “single mutants” (D137L or C190A) and a D137L/C190A “double mutant.” The models were constructed to otherwise match the structure of a previously determined atomic model of αα-cardiac tropomyosin; 8 in each case, the Asp137 and/or the Cys190 on each chain of the αα-cardiac tropomyosin template was replaced with a corresponding Leu for Asp or Ala for Cys. The starting structure was first energy optimized and MD run for 25 to 50 ns at 300° K with Langevin dynamics and an implicit solvent model using the program CHARMM c33b223,24 as described in reference 8. Previously developed methods8,25 to determine tropomyosin local curvature and flexibility along the length of tropomyosin were implemented. Disclosure of Potential Conflicts of Interest No potential conflicts of interest were disclosed. Tropomyosins possess a conserved Asp at amino acid position 137. Here MD simulations indicate that substitution of Asp 137 with a hydrophobic residue in the canonical d position of the tropomyosin coiled-coil decreases overall curvature, yet increases flexibility. The strong conservation of Asp 137 in tropomyosin isoforms suggests that this non-canonical residue, is required for proper actin binding and modulation of tropomyosin equilibrium position, and consequently is necessary for fine-tuning physiological responses. References Figure 4. comparing the local packing around position 137 (A) packing of Asp 137 at the d-position residue (B) packing of Leu 137 at the d-position. Note that the Leu side chains of the D137L mutant stably pack into the hydrophobic core of the α-helical coiled-coil while the Asp side chains are pushed outwards. The average structure of the region surrounding Asp 137 (residues 131–142) for the wild-type and D137L mutant tropomyosin (c) show that the D137L substitution yields a straighter molecule. 2. 3. Acknowledgments We would like to thank Anastasia Karabina for help with figures. This work was supported by NIH-HL077280 to J.M. and NIH-HL086655 and HL036153 to W.L. Note Supplemental material can be found at: w w w.la nde sbioscienc e.c om /jou r na l s / bioa rc h itec t u re / article/18117 Hodges RS, Saund AK, Chong PC, St-Pierre SA, Reid RE. Synthetic model for two-stranded α-helical coiledcoils. Design, synthesis, and characterization of an 86-residue analog of tropomyosin. J Biol Chem 1981; 256:1214-24; PMID:7451500 Hodges RS, Zhou NE, Kay CM, Semchuk PD. Synthetic model proteins: contribution of hydrophobic residues and disulfide bonds to protein stability. Pept Res 1990; 3:123-37; PMID:2134057 BioArchitecture 4. 5. Burkhard P, Stetefeld J, Strelkov SV. Coiled coils: a highly versatile protein folding motif. Trends Cell Biol 2001; 11:82-8; PMID:11166216; http://dx.doi. org/10.1016/S0962-8924(00)01898-5 Lorenz M, Poole KJ, Popp D, Rosenbaum G, Holmes KC. An atomic model of the unregulated thin filament obtained by X-ray fiber diffraction on oriented actin-tropomyosin gels. J Mol Biol 1995; 246:10819; PMID:7853391; http://dx.doi.org/10.1006/ jmbi.1994.0070 Volume 1 Issue 5 6. Poole KJ, Lorenz M, Evans G, Rosenbaum G, Pirani A, Craig R, et al. A comparison of muscle thin filament models obtained from electron microscopy reconstructions and low-angle X-ray fibre diagrams from non-overlap muscle. J Struct Biol 2006; 155:27384; PMID:16793285; http://dx.doi.org/10.1016/j. jsb.2006.02.020 7. Sumida JP, Wu E, Lehrer SS. Conserved Asp-137 imparts flexibility to tropomyosin and affects function. J Biol Chem 2008; 283:6728-34; PMID:18165684; http://dx.doi.org/10.1074/jbc.M707485200 8. Li XE, Holmes KC, Lehman W, Jung H, Fischer S. The shape and flexibility of tropomyosin coiled coils: implications for actin filament assembly and regulation. J Mol Biol 2010; 395:327-39; PMID:19883661; http://dx.doi.org/10.1016/j.jmb.2009.10.060 9. Sousa D, Cammarato A, Jang K, Graceffa P, Tobacman LS, Li XE, et al. Electron microscopy and persistence length analysis of semi-rigid smooth muscle tropomyosin strands. Biophys J 2010; 99:862-8; PMID:20682264; http://dx.doi.org/10.1016/j. bpj.2010.05.004 10. Gimona M. Dimerization of tropomyosins. Adv Exp Med Biol 2008; 644:73-84; PMID:19209814; http:// dx.doi.org/10.1007/978-0-387-85766-4_6 11. Li XE, Tobacman LS, Mun JY, Craig R, Fischer S, Lehman W. Tropomyosin position on F-actin revealed by EM reconstruction and computational chemistry. Biophys J 2011; 100:1005-13; PMID:21320445; http://dx.doi.org/10.1016/j.bpj.2010.12.3697 12. Dominguez R. Tropomyosin: the gatekeeper’s view of the actin filament revealed. Biophys J 2011; 100:7978; PMID:21320421; http://dx.doi.org/10.1016/j. bpj.2011.01.018 13. Holmes KC, Lehman W. Gestalt-binding of tropomyosin to actin filaments. J Muscle Res Cell Motil 2008; 29:213-9; PMID:19116763; http://dx.doi. org/10.1007/s10974-008-9157-6 14. Pato MD, Mak AS, Smillie LB. Fragments of rabbit striated muscle alpha-tropomyosin. I. Preparation and characterization. J Biol Chem 1981; 256:593-601; PMID:7451462 15. Lees-Miller JP, Helfman DM. The molecular basis for tropomyosin isoform diversity. Bioessays 1991; 13:42937; PMID:1796905; http://dx.doi.org/10.1002/ bies.950130902 16. Straussman R, Ben-Ya’acov A, Woolfson DN, Ravid S. Kinking the coiled coil–negatively charged residues at the coiled-coil interface. J Mol Biol 2007; 366:123242; PMID:17207815; http://dx.doi.org/10.1016/j. jmb.2006.11.083 17. O’Shea EK, Klemm JD, Kim PS, Alber T. X-ray structure of the GCN4 leucine zipper, a two-stranded, parallel coiled coil. Science 1991; 254:539-44; PMID: 1948029; http://dx.doi.org/10.1126/science.1948029 18. Woolfson DN, Alber T. Predicting oligomerization states of coiled coils. Protein Sci 1995; 4:1596607; PMID:8520486; http://dx.doi.org/10.1002/ pro.5560040818 19. Nevzorov IA, Nikolaeva OP, Kainov YA, Redwood CS, Levitsky DI. Conserved noncanonical residue gly-126 confers instability to the middle part of the tropomyosin molecule. J Biol Chem 2011; 286:1576672; PMID:21454502; http://dx.doi.org/10.1074/jbc. M110.209353 20. Lehman W, Hatch V, Korman V, Rosol M, Thomas L, Maytum R, et al. Tropomyosin and actin isoforms modulate the localization of tropomyosin strands on actin filaments. J Mol Biol 2000; 302:593606; PMID:10986121; http://dx.doi.org/10.1006/ jmbi.2000.4080 21. Mudalige WA, Tao TC, Lehrer SS. Ca2+-dependent photocrosslinking of tropomyosin residue 146 to residues 157-163 in the C-terminal domain of troponin I in reconstituted skeletal muscle thin filaments. J Mol Biol 2009; 389:575-83; PMID:19379756; http:// dx.doi.org/10.1016/j.jmb.2009.04.027 22. Galinska A, Hatch V, Craig R, Murphy AM, Van Eyk JE, Wang CL, et al. The C terminus of cardiac troponin I stabilizes the Ca2+-activated state of tropomyosin on actin filaments. Circ Res 2010; 106:70511; PMID:20035081; http://dx.doi.org/10.1161/ CIRCRESAHA.109.210047 23. Brooks BR, Brooks CL 3rd, Mackerell AD Jr., Nilsson L, Petrella RJ, Roux B, et al. CHARMM: the biomolecular simulation program. J Comput Chem 2009; 30:1545-614; PMID:19444816; http://dx.doi. org/10.1002/jcc.21287 24. Brooks BR, Bruccoleri RE, Olafson BD, States DJ, Swaminathan S, Karplus M. CHARMM: a program for macromolecular energy minimization, and dynamica calculations. J Comput Chem 1983; 30:118-217 25. Li XE, Lehman W, Fischer S. The relationship between curvature, flexibility and persistence length in the tropomyosin coiled-coil. J Struct Biol 2010; 170:3138; PMID:20117217; http://dx.doi.org/10.1016/j. jsb.2010.01.016 . e nc e i c s o i B . e s t e u d b i n r a t L s i 2 d 1 t 0 o 2 n o © D www.landesbioscience.com BioArchitecture 255