SEE COMMENTARY
The titin-telethonin complex is a directed, superstable molecular bond in the muscle Z-disk Morten Bertza, Matthias Wilmannsb, and Matthias Riefa,c,1 aPhysik
Department E22, Technische Universita¨t Mu¨nchen, James-Franck-Strasse, 85748 Garching, Germany; and cCenter for Integrated Protein Science Munich (CIPSM), 81377 Munich, Germany; bEuropean Molecular Biology Laboratory Hamburg, c/o DESY, Notkestrasse 85, 22603 Hamburg, Germany
Mechanical stability of bonds and protein interactions has recently become accessible through single molecule mechanical experiments. So far, mechanical information about molecular bond mechanics has been largely limited to a single direction of force application. However, mechanical force acts as a vector in space and hence mechanical stability should depend on the direction of force application. In skeletal muscle, the giant protein titin is anchored in the Z-disk by telethonin. Much of the structural integrity of the Z-disk hinges upon the titin-telethonin bond. In this paper we show that the complex between the muscle proteins titin and telethonin forms a highly directed molecular bond. It is designed to resist ultra-high forces if they are applied in the direction along which it is loaded under physiological conditions, while it breaks easily along other directions. Highly directed molecular bonds match in an ideal way the requirements of tissues subject to mechanical stress. atomic force microscopy 兩 force spectroscopy 兩 protein engineering 兩 protein folding
T
he giant muscle protein titin (1, 2) spans the complete half sarcomere from the Z-disk to the M-line and is involved in a multitude of functions including passive muscle elasticity (3, 4), stress sensing (5, 6), regulation (7), and muscle assembly (8). Rigid anchoring of titin within the Z-disk even under extreme mechanical loads is key for proper muscle function (9). Recently, Zou et al. have shown that 2 titin molecules are connected at their N termini in a palindromic arrangement through the muscle protein telethonin (also known as T-Cap) into a 2:1 assembly (see Fig. 1A and B) (10). Palindromic arrangements are rare in protein-protein complexes (11) and may provide a unique interaction under specific physiological conditions such as mechanical force. In this complex, 2 identical titin-telethonin-titin -sheets are formed that are in a diagonal orientation with respect to the principal axis of the complex that is defined by the 2 telethonin -turns (Fig. 1 A). The 2 C-terminal Z2 Ig-domains of titin are in distal positions of the overall complex. Since the remaining titin fiber extends from the C terminus of this complex it is expected that any mechanical forces originating from operating muscles would act on the Z2 domains, as opposed the N-terminal titin Z1 domains. Recent molecular dynamics simulations have suggested that a tight hydrogen bond network between titin and telethonin conveys a mechanical stabilization to the complex (12). Direct experimental measurement of the mechanical stability of a protein complex, however, has been missing to date. Therefore, a molecular basis of a force-resistant anchoring mechanism of the titin N terminus within the sarcomeric Z-disk remains yet elusive. Results Single molecule mechanical experiments using AFM critically rely on the identification of mechanical fingerprints that allow distinguishing single molecule events from nonspecific interactions as well as multimolecule events. The engineering of molecular fingerprints has so far involved the use of polyproteins www.pnas.org兾cgi兾doi兾10.1073兾pnas.0902312106
producing regular sawtooth patterns as the chain of identical subunits unfolds sequentially (13). The dissociation of a protein complex, which is connected by non-covalent bonds, however, will yield an unfolding fingerprint that is indistinguishable from the detachment of the polypeptide chain from the cantilever. How can we detect an unambiguous mechanical fingerprint of the rupture of the titin-telethonin complex? We used cysteine engineering to introduce disulfide bridges covalently connecting the 3 polypeptide chains of telethonin and the 2 titin fragments, respectively (for details, see Methods and Fig. S1). The covalent crosslinks will prevent early detachment of the protein after forced dissociation of the titin-telethonin assembly and allow observation of the complete sequence of unbinding/unfolding events of the whole complex. As handles for attachment of the complex to tip and surface we fused ubiquitin domains to the titin fragment (compare construct in Fig. 1C). The small protein ubiquitin (76 amino acids) can be readily distinguished from the unfolding of the larger Ig-domains (94 residues) present in the titin-telethonin complex by their contour length increase. Since in vivo, most of the long titin polypeptide chain aligns with the principal filament direction of muscle sarcomeres (compare Fig. 1B), we postulate that mechanical forces generated by the muscle contraction/relaxation cycle translate into pull/push forces that act on the C terminus of the N-terminal titin fragment (Ig domains Z1 and Z2), from where the remaining titin filament extends. Therefore, we first studied the stability of the titin-telethonin complex by applying force at the titin Ctermini. For this purpose, we designed the molecular construct shown in Fig. 1C (Z1Z2TC-C) where the ubiquitin handles were fused to the C-termini of the titin fragments. Force extension traces at low extensions show the characteristic unfolding patterns of the ubiquitin handles (14) (gray peaks in Fig. 1C). Since attachment to tip and surface occurs through non-specific adsorption not necessarily at the termini of the protein, the number of ubiquitin unfolding events can vary. After complete unfolding of all ubiquitins, a marked fingerprint of the stepwise breaking of the titin-telethonin complex can be observed (colored section of the force curves in Fig. 1C). The most striking result is the extremely high force required to initiate the unbinding process (green circle in Fig. 1C and histogram in Fig. 1D). An average force of 707 ⫾ 24 pN (average ⫾ SEM, n ⫽ 34) at a pulling speed of 1 m/s exceeds all known measured stabilities of protein structures or complexes (15). In fact, covalent bonds break at only twice this value (16, 17). We used the worm-like chain (WLC) model (18) to determine the increase in contour length that follows an unfolding event (dashed Author contributions: M.B. and M.R. designed research; M.B. performed research; M.B. analyzed data, M.W. contributed new reagents/analytic tools; and M.B., M.W., and M.R. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. See Commentary on page 13149. 1To
whom correspondence should be addressed. E-mail:
[email protected].
This article contains supporting information online at www.pnas.org/cgi/content/full/ 0902312106/DCSupplemental.
PNAS 兩 August 11, 2009 兩 vol. 106 兩 no. 32 兩 13307–13310
BIOPHYSICS AND COMPUTATIONAL BIOLOGY
Edited by James A. Spudich, Stanford University School of Medicine, Stanford, CA, and approved June 11, 2009 (received for review March 2, 2009)
Fig. 1. (A) Structure of the antiparallel titin-telethonin complex (pdb: 1YA5) (10). The first 2 N-terminal domains [Z1 (dark) and Z2 (light)] of 2 titin molecules [chain A (green) and chain B (blue)] are assembled into a palindromic complex by telethonin (pink). (B) Schematic overview of the sarcomere and its 3 major filaments: actin, myosin, and titin. Only one-half of the sarcomere is shown. The titin-telethonin complex located in the Z-disk is shown as a space-filling model. (C) Typical force-extension traces of the titin-telethonin complex pulled at its C termini (Z1Z2TC-C): The structural model illustrates the pulling geometry and the location of the cysteine crosslinks (yellow bars, compare Fig. S1) necessary to obtain a clear fingerprint of the rupture of the complex. Unfolding events of the ubiquitin handles (gray filled circles in the structural model) are colored gray in the unfolding traces. WLC curves fit to the rupture of the complex (colored part of the traces) are shown as dashed lines with contour length increases indicated above the curves. (D) Force histogram of the initial rupture event of the complex (green circle in C).
curves in Fig. 1C). Analysis of these contour length increases observed during rupture of the complex allows drawing a structural picture of the sequence of events in the rupture process (Fig. S2). The major rupture peak is associated with a length increase of 29.7 ⫾ 0.4 nm (⫾ SEM). This length increase corresponds exactly to the length expected for unfolding of the approximate 94 amino acid residues of titin domain Z2 (29.8 nm). In the subsequent peak (pink in Fig. 1C), both Z1 domains as well as telethonin unfold. It is important to note that a large part of the Z1 domains and of telethonin are shut off from force by the cysteine crosslinks and hence do not contribute to the length increase (see Fig. S2). The last peak (blue event) reflects unfolding of the now isolated remaining Z2 domain.
To assess the amount of stabilization conveyed onto the titin domains by forming the titin-telethonin complex, we now investigated the stability of the Z1 and Z2 domains in the absence of telethonin. Ubiquitin handles were attached to the titin fragment at the N terminus of Z1 and at the C terminus of Z2 (compare construct in Fig. 2A). The unfolding events of the titin domains (green events in Fig. 2 A) can be clearly distinguished from the ubiquitin handles (gray in Fig. 2 A) by their larger contour length increase. We find that the 2 Ig-domains Z1 and Z2 unfold with an average contour length increase of 29.4 ⫾ 0.2 nm (⫾ SEM, n ⫽ 205) at an average force of 168 ⫾ 2 pN (⫾ SEM) (compare histogram in Fig. 2B). Interestingly, most titin Ig domains investigated so far exhibit larger unfolding forces (4) and hence
Fig. 2. (A) Force-extension traces of the Z1Z2 fragment from titin flanked by ubiquitin domains (gray filled circles). Unfolding events of the ubiqutin handles are colored in gray in the traces. Unfolding events of the titin fragment are colored in green. Dashed lines indicate WLC fits with the contour length increases shown above the curves. (B) Force histogram of the mechanical unfolding of the Z1Z2 domain pair. (C) Force-extension traces of the inverted pulling geometry of the titin-telethonin complex (Z1Z2TN-N): The structural model illustrates the architecture of the construct. Cysteine crosslinks are shown as yellow bars (compare Fig. S1), ubiquitin handles as gray filled circles. The rupture of Z1Z2TN-N is indicated by the various colors in the traces. WLC fits to the data and contour length increases are show in the left panel. (D) Force histogram of the initial rupture event of Z1Z2TN-N. 13308 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0902312106
Bertz et al.
SEE COMMENTARY
the domain pair Z1Z2 alone is a comparatively labile structure in vitro. The formation of the titin-telethonin sandwich complex increases the stability of this structure 4-fold, far beyond all other Ig domains present in the titin molecule examined so far by mechanical experiments. How is the mechanical superstability of the titin-telethonin complex tuned to its physiological function of forming a durable link in the mechanically strained Z-disk? To answer this question, we applied mechanical load to the titin-telethonin complex in a geometry reverse to the physiological C-terminal direction. We fused the ubiquitin handle domains to the N terminus of the titin Z1 domain (Z1Z2TN-N) and applied load in the single molecule experiment (compare structural model in Fig. 2C). Sample traces and a force histogram of this inverted pulling experiment are shown in Fig. 2C and D. The characteristic unfolding events of ubiquitin are shown in gray whereas the colored part of the curve corresponds to the rupture of the titin-telethonin complex when load is applied to the N termini. The difference in rupture force (green circle Fig. 2C) in the inverted pulling geometry Z1Z2TN-N is indeed drastic if compared to the physiological pulling geometry Z1Z2TC-C: In the inverted pulling direction, the stability is almost indistinguishable from the isolated domains Z1Z2 with only a slightly elevated average stability of 237 ⫾ 5 pN (⫾ SEM, n ⫽ 47). The sequence of events leading to breakage of Z1Z2TN-N is illustrated in Fig. S3. The initial event is detachment and unfolding of a Z1 domain (dark green peak in Fig. 2C) from the complex. In the next step, telethonin detaches and unfolds (red peak). Finally, the remaining 3 Ig domains unfold subsequently (light green and blue peaks).
Bertz et al.
Fig. 3. (A) Top: Structure of the titin-telethonin complex with part of the hydrogen bond network (black lines) stabilizing the distal Z2 domains when load is applied at the C termini. Bottom: Back view of the titin-telethonin complex. The N termini of the proximal Z1 domains do not form a tight hydrogen bond network with telethonin. (B) The titin-telethonin complex is a directed, highly stable molecular bond. Load applied to the C termini, as is the case in the muscle during passive stretching, requires high forces to dissociate the complex. If loaded at the N termini, the complex unravels readily.
function beyond mechanosensing. Our results support this: With a rupture force of 700 pN, exceeding all Ig domains from titin measured so far (4), the titin-telethonin complex is perfect for firmly anchoring the giant muscle protein titin in the Z-disk. Such a rigid molecular complex, however, is rather ill suited for contributing to a strain sensor in muscle. Our data may thus provide a molecular rationale of an N-terminal titin assembly complex that allows stable anchoring of this filament system in the sarcomeric Z-disk, while being exposed to large mechanical forces during the muscle contraction/relaxation cycle. We anticipate that directedness of molecular bonds will be a concept important for a variety of other molecular complexes subject to mechanical strain in living organisms. Methods Cloning and Protein Expression. The human titin N terminus covering domains Z1 and Z2 (residues 1–195) was fused to the C terminus of 3 ubiquitin domains carrying an N-terminal His-Tag (3U-Z1Z2) or to the N terminus of 3 ubiquitin subunits carrying a C-terminal His-tag (Z1Z2–3U) in pET28a (Novagen) by standard molecular biology techniques. Cysteine residues for disulfide crosslinks (S86C in Z1Z2–3U and T188C in 3U-Z1Z2) were introduced using the Quikchange site-directed mutagenesis kit (Stratagene). Furthermore, Z1Z2 was sandwiched between subunits 3 and 4 in a construct containing 6 concatenated ubiquitin domains in pET28a (3U-Z1Z2–3U,). The N-terminal truncation of telethonin (residues 1–90) with all cysteines mutated to serines has been described previously (21). We constructed 2 variants for cysteine crosslinking (TeleAT: A20C/T50C and TeleEQ: E16C/Q46C) using the Quikchange Multi Site directed mutagenesis kit (Stratagene). PNAS 兩 August 11, 2009 兩 vol. 106 兩 no. 32 兩 13309
BIOPHYSICS AND COMPUTATIONAL BIOLOGY
Discussion Our results reveal the titin-telethonin interaction as a strongly directed molecular bond, optimized to resist loads applied in the C-terminal direction (see Fig. 3B, bottom), which presents the direction of the remaining titin fiber. While the titin-telethonin complex strongly resists unbinding when loaded in the naturally occurring pulling direction, it slips apart relatively easily when pulled in the reverse geometry. Consistent with the strong mechanical directedness of the titin-telethonin complex, pulling the complex in an N-C-terminal direction (Z1Z2TN-C) (see Fig. S4) leads to only slight stabilization compared to the isolated domains. A likely molecular explanation for this directedness is the tight network of hydrogen bonds established by telethonin binding and stabilizing predominantly the C terminus of the distal Z2 domains (see Fig. 3A, top) (12), where load is applied during passive muscle stretching. In contrast, the N terminus of the proximal Z1 domains does not interact with telethonin and only forms hydrogen bridges to a neighboring strand from the same Ig domain (Fig. 3A, bottom); hence, a significant stabilization of the complex if pulled from the N terminus cannot occur. Apparently, nature uses selective clustering of H-bonds to reinforce the C terminus of Z2 and thus prevent breaking of the titin-telethonin complex under mechanical loads. A macroscopic mechanical analog of the directed titin-telethonin bond is a hook connecting the 2 Z2 domains thus preventing unbinding of the complex in C-terminal direction but letting the complex slide apart if pulled from the N terminus (see Fig. 3B). It is important to note, that the concept of directedness of a molecular bond cannot be understood if considering only thermodynamic parameters like free energy of complex formation or off-rates of the complex, since those parameters do not contain directional information. A complex between telethonin and the muscle LIM protein (MLP) has been suggested to be a key component of the muscle stretch sensing machinery (19). Recent findings, however, have shown that MLP is a highly mobile component of muscle cells and not anchored to sarcomeric structures (20) suggesting a
The telethonin and titin constructs were expressed and purified as described previously (21). The titin-telethonin complexes were formed by adding the appropriate purified titin-ubiquitin chimeras to telethonin solutions (3UZ1Z2(T188C) ⫹ TeleAT forming Z1Z2TN-N and Z1Z2(S86C)-3U ⫹ TeleEQ forming Z1Z2TC-C, compare Fig. S1) at a final concentration of 4 M urea with excess telethonin followed by dialysis over night against PBS containing 10 mM DTT. After dialysis, the complexes were further purified by size-exclusion chromatography using a Superdex 200 column (GE Healthcare) in PBS. 3U-Z1Z2–3U was purified by size exclusion chromatography after elution from the Ni-NTA matrix and used directly for measurements. Single-Molecule Force Spectroscopy. Single-molecule force spectroscopy was performed on a custom-built atomic force microscope at ambient temperature. Gold-coated cantilevers (Biolever Type A, Olympus) with a spring constant of 30 pN/nm were used in all experiments. In a typical experiment, protein solution was applied to a freshly activated Ni-NTA glass slide and force traces were recorded at a pulling velocity of 1 m/s. All traces were inspected upon recording and traces that showed no clear single molecule events (at least 2 ubiquitin unfolding peaks) were discarded immediately. For Z1Z2 the probability to pick up a single molecule was approximately 5%, for Z1Z2TN-N and Z1Z2TC-C the probability was ⬍2%. These probabilities are typical for single-molecule force spectroscopy experiments (22). All traces were then screened visually and by fitting WLC-curves as described below in Igor Pro (Wavemetrics) for reproducible features that appeared in addition to the regular sawtooth pattern of the ubiquitin handles. For Z1Z2, a total of 173 single molecule traces was recorded, 102 of which showed unfolding events of both the Z1 and Z2 domains. For Z1Z2TN-N, 594 single-molecule traces were
1. Tskhovrebova L, Trinick J (2003) Titin: Properties and family relationships. Nat Rev Mol Cell Biol 4:679 – 689. 2. Labeit S, Kolmerer B (1995) Titins: Giant proteins in charge of muscle ultrastructure and elasticity. Science 270:293–296. 3. Maruyama K (1997) Connectin/titin, giant elastic protein of muscle. FASEB J 11:341– 345. 4. Li H, et al. (2002) Reverse engineering of the giant muscle protein titin. Nature 418:998 –1002. 5. Mayans O, et al. (1998) Structural basis for activation of the titin kinase domain during myofibrillogenesis. Nature 395:863– 869. 6. Puchner EM, et al. (2008) Mechanoenzymatics of titin kinase. Proc Natl Acad Sci USA 105:13385–13390. 7. Trinick J, Tskhovrebova L (1999) Titin: A molecular control freak. Trends Cell Biol 9:377–380. 8. Gregorio CC, Granzier H, Sorimachi H, Labeit S (1999) Muscle assembly: A titanic achievement? Curr Opin Cell Biol 11:18 –25. 9. Gregorio CC, et al. (1998) The NH2 terminus of titin spans the Z-disc: Its interaction with a novel 19-kD ligand (T-cap) is required for sarcomeric integrity. J Cell Biol 143:1013– 1027. 10. Zou P, et al. (2006) Palindromic assembly of the giant muscle protein titin in the sarcomeric Z-disk. Nature 439:229 –233. 11. Pinotsis N, Wilmanns M (2008) Protein assemblies with palindromic structure motifs. Cell Mol Life Sci 65:2953–2956. 12. Lee EH, Gao M, Pinotsis N, Wilmanns M, Schulten K (2006) Mechanical strength of the titin Z1Z2-telethonin complex. Structure 14:497–509.
13310 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0902312106
recorded, 47 of which contained the unfolding fingerprint of the titintelethonin in the N-terminal pulling geometry. For Z1Z2TC-C, 939 single molecule traces were recorded, 34 of which showed the rupture of the titintelethonin complex. The low probability of obtaining single molecule events for Z1Z2TC-C is likely due to detachment of the protein from the cantilever before the high forces necessary to dissociate the complex could be reached or incomplete disulfide bond formation. Contour length increases (⌬L) were determined by fitting worm-like chain curves calculated using the interpolation formula by Bustamante et al. (18) with a fixed persistence length of 0.5 nm for forces up to 250 pN. For the higher force regime in Z1Z2TC-C, 0.3 nm was used. Contour length increases determined at a persistence length of 0.3 nm were corrected to a persistence length of 0.5 nm by ⌬L0.5 ⫽ ⌬L0.3䡠␥ with ␥ ⫽ 0.964 (23). Contour length increases at a persistence length of 0.5 nm were used to calculate the number of amino acids n involved in an unfolding event via ⌬L ⫽ n䡠daa ⫺ dintermediate⫹ dfolded. Here, dintermediate and dfolded denote the distance between the points of force application in the intermediate and the native state, respectively, as determined from the crystal structure of the titin-telethonin complex (1YA5) (10). daa is the contour length increase per amino acid, which has been determined to be daa ⫽ 0.365 ⫾ 0.002 nm for our instrument at a persistence length of 0.5 nm (23). ACKNOWLEDGMENTS. We thank Jan-Philipp Junker for discussions and assistance with cloning, Michael Schlierf for discussions, and Anja Gieseke for assistance with cloning. Financial support of Deutsche Forschungsgemeinschaft Grant RI 990/3-1 (to M.R.) is gratefully acknowledged. M.W. acknowledges funding from Fonds zur Fo¨rderung der wissenschaftlichen Forschung/ Deutsche Forschungsgemeinschaft (P1906).
13. Marszalek PE, et al. (1999) Mechanical unfolding intermediates in titin modules. Nature 402:100 –103. 14. Carrion-Vazquez M, et al. (2003) The mechanical stability of ubiquitin is linkage dependent. Nat Struct Biol 10:738 –743. 15. Dietz H, Berkemeier F, Bertz M, Rief M (2006) Anisotropic deformation response of single protein molecules. Proc Natl Acad Sci USA 103:12724 –12728. 16. Grandbois M, Beyer M, Rief M, Clausen-Schaumann H, Gaub HE (1999) How strong is a covalent bond? Science 283:1727–1730. 17. Schmidt SW, Beyer MK, Clausen-Schaumann H (2008) Dynamic strength of the siliconcarbon bond observed over three decades of force-loading rates. J Am Chem Soc 130:3664 –3668. 18. Bustamante C, Marko JF, Siggia ED, Smith S (1994) Entropic elasticity of lambda-phage DNA. Science 265:1599 –1600. 19. Knoll R, et al. (2002) The cardiac mechanical stretch sensor machinery involves a Z disc complex that is defective in a subset of human dilated cardiomyopathy. Cell 111:943– 955. 20. Geier C, et al. (2008) Beyond the sarcomere: CSRP3 mutations cause hypertrophic cardiomyopathy. Hum Mol Genet 17:2753–2765. 21. Zou P, et al. (2003) Solution scattering suggests cross-linking function of telethonin in the complex with titin. J Biol Chem 278:2636 –2644. 22. Rounsevell R, Forman JR, Clarke J (2004) Atomic force microscopy: Mechanical unfolding of proteins. Methods 34:100 –111. 23. Dietz H, Rief M (2006) Protein structure by mechanical triangulation. Proc Natl Acad Sci USA 103:1244 –1247.
Bertz et al.
