Báo cáo khoa học: Regulation of the actin–myosin interaction by titin

doi:10.1111/j.1432-1033.2004.04429.x

Eur. J. Biochem. 271, 4572–4581 (2004) (cid:1) FEBS 2004

Regulation of the actin–myosin interaction by titin

Nicolas Niederla¨ nder1, Fabrice Raynaud2, Catherine Astier2 and Patrick Chaussepied1 1CRBM-CNRS, Montpellier, France; 2EPHE-UMR5539-CNRS, Montpellier, France

1

and with an inhibition of HMM binding to actin N-terminal residues as shown by chemical cross-linking. At the same time, T800 did not affect the efficiency of the Ca2+- controlled on/off switch, nor did it alter the overall binding energetics of HMM to actin, as revealed by cosedimentation experiments. These data are consistent with a competitive effect of PEVK domain-containing T800 on the electrostatic contacts at the actin–HMM interface. They also suggest that titin may participate in the regulation of the active tension generated by the actin–myosin complex.

Keywords: ATPase; chemical cross-linking; mass spectro- metry; motility assay; muscle contraction.

Titin is known to interact with actin thin filaments within the I-band region of striated muscle sarcomeres. In this study, we have used a titin fragment of 800 kDa (T800) purified from striated skeletal muscle to measure the effect of this interaction on the functional properties of the actin– myosin complex. MALDI-TOF MS revealed that T800 contains the entire titin PEVK (Pro, Glu, Val, Lys-rich) domain. In the presence of tropomyosin–troponin, T800 increased the sliding velocity (both average and maximum values) of actin filaments on heavy-meromyosin (HMM)- coated surfaces and dramatically decreased the number of stationary filaments. These results were correlated with a 30% reduction in actin-activated HMM ATPase activity

domain (Pro, Glu, Val, Lys-rich) whose size depends on the muscle fibre isotype. Specific structural properties and mechanical force/extension measurements made on muscle fibres or at the single molecule level suggest that the tandem Ig- and PEVK-domains are two elements of differential stiffness that function as a two-spring system [13–24]. This elastic system is now believed to be a major contributor to the passive tension developed in striated muscle.

Titin is the largest known protein, containing more than 38 000 residues in its longest human striated muscle isoform. It represents the third most abundant component of vertebrate striated muscle, after myosin and actin, and is also present in smooth muscle and nonmuscle cells (recently reviewed in [1,2]). The importance of intact titin for normal muscle function has been demonstrated in vitro [3–5], as well as in vivo through its implication in muscular dystrophies such as dilated cardiomyopathies and Udd’s tibial muscular dystrophy (reviewed in [6]).

that

In striated muscle, titin is involved in several fundamental processes, including sarcomere assembly, possibly in thick filament length control [4,7–9], maintenance of the sarco- meric structure, muscle elasticity and passive tension development [10–12]. These functions are related to three main structural properties of the protein: titin spans half a sarcomere, from the Z disks to the M line (connecting the Z disks to myosin thick filaments), it contains subdomains that confer unusual elastic properties, and it interacts with several protein partners such as myosin, actin, M protein, C protein, MURF-1, calpain 3, myomesin, a-actinin, nebulin, telethonin and obscurin.

The elastic domains are made of tandemly arranged immunoglobulin (Ig)-like domains and a unique PEVK

Another important feature of the I-band region was first revealed by electron microscopy images, which showed that in this region titin and actin can come close enough to associate with each other [25,26]. This association has now been confirmed by numerous in vitro experiments involving actin and the titin PEVK domain [27–33]. The dynamics of this association seem to act together with the elastic elements of titin to modulate muscle passive stiffness the PEVK [34–36]. Indeed, recent data suggest domain from cardiac muscle titin interacts with actin much more efficiently than does that from skeletal muscle titin [36,37], supporting the idea that this interaction may be correlated with passive stiffness in each muscle type. It is important to note, however, that both the size of the PEVK domain, and the difficulty involved in extracting large amounts of native titin from muscle, have restricted these studies to examining the interaction between actin and bacterially expressed recombinant PEVK titin sub- fragments. In the case of the single in vitro motility assay that has been achieved using tissue-extracted titin, the experiments were designed to favour titin binding to the coverslip, which stopped actin motion during the assay [30].

In this study, we have further investigated the interaction of titin with actin by using two new experimental tools. First, we have used a native titin fragment of 800 kDa (encompassing the entire PEVK domain) that was isolated from the muscle sarcomeric I-band region. Second, we

Correspondence to P. Chaussepied, Centre de Recherche de Biochimie Macromole´ culaire, CNRS, 1919 Route de Mende, 34293 Montpellier Cedex 5, France. Tel.: +33 467613334, Fax: +33 467521559, E-mail: patrick.chaussepied@crbm.cnrs-mop.fr Abbreviations: DTE, dithioerythritol; EDC, 1-ethyl-3-(3-dimethyl- aminopropyl)carbodiimide; NHS, N-hydroxysuccinimide; F-actin, filamentous actin; HMM, heavy meromyosin; T800, titin fragment of 800 kDa; Tm–Tn, tropomyosin–troponin complex. (Received 11 August 2004, revised 4 October 2004, accepted 11 October 2004)

Effect of titin on actin–myosin activity (Eur. J. Biochem. 271) 4573

(cid:1) FEBS 2004

centrifuged (except F-actin) at 190 000 g for 20 min prior to each experiment.

have worked with reconstituted thin filaments containing both actin and the regulatory tropomyosin–troponin (Tm– Tn) complex. Data obtained using these tools have confirmed the interaction between the PEVK domain- containing titin fragment and reconstituted thin filament. They have also shown that the titin fragment reduces the number of contacts between myosin and the N-terminal part of actin, producing significant effects on both in vitro motility and the ATPase activitiy of the actin–myosin complex.

Protein concentrations were determined spectrophoto- 280 of 5.7 cm)1, metrically assuming extinction coefficients A1% 6.5 cm)1, 11.0 cm)1, 3.3 cm)1, 4.5 cm)1 and 10.0 cm)1 for myosin (500 kDa), HMM (360 kDa), actin (42 kDa), troponin (70 kDa) and T800 tropomyosin (66 kDa), (800 kDa), respectively. The extinction coefficient for T800 was estimated experimentally using the Bradford method [44] to measure the protein concentration of the T800-containing solution, using HMM for the standard curve.

Materials and methods

MS

Reagents

1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) and N-hydroxysuccinimide (NHS) were from Sigma. a-Chymotrypsin was from Worthington. All other chemi- cals were of the highest analytical grade.

Proteins were in-gel digested by trypsin according to Rosenfeld et al. [45]. The resulting digests were cleaned using the ZipTip device (Millipore Inc) and analysed by MALDI-TOF MS (BiflexIII, Bruker). Database queries were performed using the Mascot search engine (Matrix Science at http://www.matrixscience.com/).

Preparation of proteins

Invitromotility assay

et al.

buffer

containing

3.3 mgÆmL)1

F-actin (0.6 lM) was first stabilized and labelled by adding a twofold excess of tetramethyl-rhodamine phalloidin in motility buffer (50 mM KCl, 10 mM MgCl2, 40 mM DTE, 60 mM Hepes pH 7.8, 90 mM ionic strength). Labelled F-actin was then diluted (2 nM final concentration) in motility glucose, 0.37 mgÆmL)1 catalase, 0.11 mgÆmL)1 glucose oxidase, 0.5% (w/v) methylcellulose, and 0.1 mM CaCl2 or 1 mM EGTA (only when the Tm–Tn complex was present). The solution was supplemented by Tm–Tn and T800 (both at 20 nM, conditions for a saturating effect), and ATP (2 mM) was added to flow cells containing HMM-coated glass coverslips just prior to image recording. Coverslips were pretreated overnight at room temperature with 1 M HCl, rinsed with distilled water, 95% ethanol and air-dried. They were then treated with BSA/casein (10 mgÆmL)1) for 10 min at 20 (cid:2)C, air-dried, mounted on the flow cell, and coated with HMM (50 lgÆmL)1 solution containing 600 mM KCl, 10 mM Hepes, pH 7.0) for 10 min on ice prior to the addition of the actin solution. (cid:1)Dead(cid:2) HMM molecules were removed before the coating step by two consecutive ultracentrifugation steps at 190 000 g for 20 min in the presence of a threefold molar excess of F-actin-phalloidin and 2.5 mM ATP in 10 mM Hepes, 600 mM KCl pH 7.0. After each ultracentrifuga- tion step, the HMM concentration was evaluated by the Bradford method. The (cid:1)dead(cid:2) HMM eliminated in this way corresponded to 5–10% of the total HMM in the preparation.

2

(enzyme/myofibril weight

Images of the microfilaments were obtained with a DMR B microscope (Leica, Bensheim, Germany) using a PL APO 100 · objective (NA 1.40) with a 1.6 · tube factor and immersion oil Immersol 518 F (Zeiss, Go¨ ttingen, 3 Germany). Preparations were illuminated with a 100 W 4 HBO 103 W/2 light bulb (OSRAM, Regensburg, Germany) through a N 2.1 filter cube (Leica) for the visualization of rhodamine fluorescence. The microscope was equipped with a homemade heating stage. The heat regulation was

All proteins were extracted from rabbit skeletal muscle. Myosin and myosin fragments were prepared as described by Offer [38]. Heavy meromyosin (HMM) was obtained after a-chymotrypsin digestion of myosin (enzyme/substrate mass ratio of 1 : 400) for 15 min at 25 (cid:2)C in 10 mM NaH2PO4, 600 mM NaCl, 1 mM MgCl2, 1 mM dithioerythritol (DTE) pH 7.0. After the reaction was stopped by phenylmethanesulfonyl fluoride (phenyl- methanesulfonyl fluoride/substrate mass ratio of 1 : 200), the solution was dialysed overnight against 20 mM Mops, 0.2 mM DTE, pH 7.0, and centrifuged 20 min at 100 000 g. HMM was purified by ion exchange chromatography on SP-sephacryl (Pharmacia-Biotech) using a 0–200 mM NaCl gradient, drop-frozen in liquid nitrogen, and stored at )80 (cid:2)C. Filamentous actin (F-actin) was prepared from acetone powder and further purified by two cycles of polymerization-depolymerization [39]. The final polymer- ization step was performed by overnight incubation of monomeric actin (40 lM) at 4 (cid:2)C in the presence of 120 mM NaCl, 2.5 mM MgCl2. Polymerized actin was concentrated by centrifugation at 190 000 g for 20 min and kept at 4 (cid:2)C (120 lM final concentration) in 100 mM NaCl, 2.5 mM MgCl2, 50 mM Mops, pH 7.5. For the in vitro motility assay, F-actin was not concentrated but rather used directly at 40 lM for rhodamine–phalloidin labelling (see below). Tropomyosin and troponin complex (troponin I, T and C) were prepared from acetone-dried muscle powder according to Smillie [40] and Potter [41], respectively. They were stored in the lyophilized form and used as a solution containing equimolar amounts of tropomyosin and troponin (Tm–Tn). Titin fragment (T800) was obtained from rabbit back muscles (mainly trapezius and lattissimus dorsi muscles) after Staphylococcus aureus V8 protease treatment of myofibrils ratio of 1 : 200, 30 min, 25 (cid:2)C) and centrifugation at 5000 g for 5 min [42,43]. T800 was subsequently purified through gel filtra- tion S300 HR (Pharmacia-Biotech) followed by Poros HQ/H column (Boehringer) in 2 mM Tris, 1 mM DTE, 1 mM EDTA, pH 7.9. Pure T800 was eluted at 250 mM NaCl. All proteins were used within 5–6 days and ultra-

(cid:1) FEBS 2004

4574 N. Niederla¨ nder et al. (Eur. J. Biochem. 271)

addition of HMM by adding an aliquot of the reaction mixture to a boiling Laemmli solution.

PAGE

Gel electrophoresis was as described by Laemmli [47] using a 2–15% gradient acrylamide gel. Densitometric analysis of the scanned gels was performed using METAMORPH 6.1 software.

5

Results

Localization of T800 within the I-band region of skeletal titin

6

7

stabilized to prevent undesired minute up and down movements of the stage, which can upset the stability of image focusing during time-lapse recording. The stability was further enhanced by the presence of a plexiglass box that protected the front part of the microscope (objective barrel, stage, etc. down to the bench) from surrounding air movements. The front part of the box consisted of a plastic curtain that allowed easy access to the stage. Images were captured with an ORCA 100 (B/W) 10 bits cooled CCD camera (C mount 1x), C 4742-95 controller and HIPIC controller program (Hamamatsu, Shizuoka, Japan) run by a PC-compatible computer. Time-lapse recording of the images (time intervals ranging from 0.1 to 1 s) were carried out with the Camera Sequence option of the controller program, with a 2 · 2 binning of the detector and camera gain set at its maximum value. Sequences were saved as a suite of individual TIFF format images (up to 250 in one sequence). Measurements were carried out with META- MORPH 6.1 software (Universal Imaging Corporation, by following the movement of Downington, PA, USA) the leading end of the actin filament. Statistical analyses were performed using PRISM 2.2 software (GraphPad Software, Inc., San Diego, CA, USA) . Mann–Whitney test was used to compare sets of data and a P-value < 0.005 was used to determine statistical significance.

Steady-state ATPase and actin binding assays

Various mixtures containing F-actin (3 lM) alone or with T800 (0.15 lM), Tm–Tn (1.0 lM) and HMM (0.25 lM in the ATPase activity and 1.5 lM in the binding assay) were incubated for 10 min in 50 mM Hepes, 5 mM MgCl2, 50 mM KCl, 2 mM DTE (80 mM ionic strength), with or without 0.1 mM CaCl2, 1 mM EGTA or 2 mM ATP, pH 7.8 (the binding assay was also conducted in the presence of 100 mM NaCl, that is in a final 180 mM ionic strength without ATP).

Some of us have previously demonstrated that mild treatment of myofibrils with S. aureus V8 protease releases a soluble titin fragment of 800 kDa (T800) that can be purified to homogeneity [42]. In order to localize T800 within titin, we performed MALDI-TOF MS following in-gel digestion of T800 by trypsin. The set of molecular weights corresponding to the resulting tryptic peptides was then examined by a search in the NCBI nonredundant protein database using the search engine (cid:1)Mascot(cid:2) without any manual interpretation [48]. The results of this search are summarized in Fig. 1A in the form of a graph showing scores reflecting the probability that an observed match is a random event. A score higher than 65 indicates identity or extensive homology with theoretical sequences in the database. Significant scores of 98 and 84 were obtained for a human skeletal titin fragment (correspond- ing to residues 4262–12 392) and full-length human skeletal titin (residues 1–26 926), respectively. Of 79 peptides analysed, 22 matched with the two proteins, with the difference between calculated and experimental molecular weights being lower than 0.1 Da. These 22 peptides were located between residues 4670 and 9070 of full-length human titin, within the I-band region of the skeletal muscle sarcomere and encompassing the entire PEVK domain (amino acid segment 5618–7792; Fig. 1B). Based on these experimentally determined boundaries, and considering that T800 contains approximately 7200 resi- dues, we estimate that the extreme borders of T800 could lie between residue 1870 (lower value) and residue 1–11 500 (higher value). These data demonstrated that T800 contains the PEVK domain and falls entirely within the I-band region of skeletal titin.

The Mg.ATPase activities were measured at 25 (cid:2)C. The reaction was started by the addition of 2 mM ATP and stopped after 10 min by 5% trichloroacetic acid. The amount of Pi liberated was evaluated colorimetrically [46]. The actin binding assay was carried out by ultracentri- fugation of the reaction mixtures at 190 000 g for 20 min. An aliquot of each supernatant was removed after centri- fugation and mixed with Laemmli solution [50 mM Hepes, 2% (w/v) NaDodSO4, 1% 2-mercaptoethanol and 50% (v/v) glycerol, pH 8.0]. Air-dried pellets were homo- genized in Laemmli solution and aliquots of both the supernatant and the resuspended pellets were analysed by PAGE after boiling the samples for 3 min.

T800 accelerates invitro motility of the reconstituted thin filament

Two-step cross-linking experiments

Because the titin PEVK domain is known to interact with actin, we studied the effects of T800 on the movement of reconstituted actin filaments on HMM coated coverslips, using the in vitro motility assay.

Figure 2A depicts a typical velocity–time pattern for one actin filament. Such a pattern was representative of the results obtained, regardless of the experimental conditions or of the presence of T800 and the regulatory proteins Tm–Tn. The filament motion displayed acceleration/decel- eration phases throughout the entire time course of the movement. This periodicity, which has been reported earlier

During the activating step, 80 lM F-actin was treated for 10 min at 20 (cid:2)C with 50 mM NHS and 25 mM EDC in buffer C (50 mM NaCl, 5 mM MgCl2, 50 mM Mops stopped with pH 7.0). The activating reaction was 100 mM b-mercaptoethanol. During the condensation step, an aliquot of activated F-actin (3 lM final concentration) was mixed with 0.15 lM T800 with or without 1.0 lM Tm–Tn and 1.5 lM HMM in buffer C in the presence of 0.1 mM CaCl2. Reactions were terminated 30 min after the

Effect of titin on actin–myosin activity (Eur. J. Biochem. 271) 4575

(cid:1) FEBS 2004

[49,50], is probably due to the heterogeneity of the HMM molecules coated on the glass surface, although other explanations such as intra-actin cooperativity have also been proposed.

filaments was also found to be altered by the addition of T800. While this number was slightly increased in the absence of Tm–Tn (21.6% vs. 16.0%), it was dramatically reduced in the presence of reconstituted thin filaments, containing Tm–Tn (5.6% vs. 21.0%; Table 1, Fig. 2C). Note that the mean filament length was not significantly affected by T800 in the absence of Tm–Tn (2.3 vs. 2.1 lm), and was slightly decreased in its presence (1.5 vs. 1.1 lm). Note also that the presence of Tm–Tn on its own decreased the mean filament length and increased sliding velocity, in good accordance with previously published data [52–54]. Finally, as expected for filaments that are normally regulated by Tm–Tn, we did not observe any movement in the absence of Ca2+, inde- pendent of the addition of T800. This result, together with the fact that T800 increased the average and maximum velocity values of moving filaments, both in the absence and in the presence of Tm–Tn–Ca2+, argues against a simple effect of T800 on the calcium sensitivity (pCa curve) of the movement and for an effect involving the actin–HMM interaction.

The recording time varied from 50 to 150 s and was generally limited by the loss of focus. Due to data scattering, we favoured a global analysis of the entire set of velocity values recorded for all the moving filaments (without stop events), rather than an analysis of the average values for each filament. Depending on the experimental conditions, 850–1700 data points were collected. The data obtained for four different experi- mental conditions (actin alone, actin in the presence of T800, actin with Tm–Tn, and actin with Tm–Tn in the presence of T800) are presented in Fig. 2B and C and in Table 1. The average velocity obtained for actin alone (2.5 lmÆs)1) was lower than the values generally obtained with nitrocellulose pretreated coverslips, but it was very comparable to the value (about 3 lmÆs)1) obtained with untreated coverslips under very similar conditions, using HMM frozen in liquid nitrogen [51]. The most significant result is that the average velocity was increased by the addition of T800, from 2.5 to 3.4 lmÆs)1 and from 3.9 to 4.3 lmÆs)1 in the absence and the presence of Tm–Tn, respectively (Table 1). This increase in the average velocity was accompanied by an increase in the maximum velocity (Fig. 2B). Statistical analysis revealed that these differences were significant, with a P-value < 0.0001. stationary actin More

the number of

importantly,

Interestingly, the order of addition of the various actin- bound components turned out to be essential in these experiments, as mixing T800 with actin prior to the addition of Tm–Tn resulted in the immobilization of the thin filaments, even in the presence of Ca2+. This result demonstrated that T800 binds to actin filaments differently in the absence and in the presence of Tm–Tn, and can promote, when added prior to Tm–Tn, an unproductive

Fig. 1. Identification of T800. (A) Mascot search result for T800 after its run in SDS gel (inset), in-gel digestion with trypsin, and ana- lysis with automated MALDI-TOF MS, fol- lowed by a search in the NCBR nonredundant protein database. (B) Schematic representa- tion of human skeletal muscle titin (gi|17066105; score 84) and a human skeletal muscle titin fragment (gi|7512404; score 98). The location of matching peptides around the PEVK domain and the two predicted extreme boundaries (residues 1870–9070 and 4670– 11500) of T800 are also shown.

(cid:1) FEBS 2004

4576 N. Niederla¨ nder et al. (Eur. J. Biochem. 271)

Table 1. In vitro motility assay analysis. Analyses were performed on three slides containing 81–91% moving filaments. Velocities were estimated on approximately 857–1709 points (without stops) for each experiment; Mann–Whitney test showed a P-value < 0.0001 com- paring either actin alone and actin + T800 or actin + Tm–Tn and actin + Tm–Tn + T800. STOPS correspond to the time that fila- ments were stationary, expressed as a percentage of total time of analysis for all moving filaments.

Actin alone Actin + T800 Actin + Tm–Tn Actin + Tm–Tn + T800

2.5 ± 1.3 3.4 ± 1.6 3.9 ± 2.0 4.3 ± 2.2

16.0 21.6 21.0 5.6

a Average length of more than 200 filaments for each experimental condition; values under 0.2 lm were excluded from all analysis.

interaction between HMM and actin. It is likely that in adult native striated muscle, titin interacts with a preformed thin filament containing bound Tm–Tn, similar to the interactions described in the present study.

T800 decreases actin-HMM ATPase activity

In order to understand the effect of T800 on thin filament sliding velocity, we measured the Mg2+-ATPase activity of HMM and various actin–HMM complexes in the presence or absence of T800. The ATPase activity of HMM alone was not changed in the presence of T800 (varying from 0.17 to 0.19 s)1; Table 2). This result was in accordance with the lack of interaction between the two proteins as revealed by the absence of cosedimentation of T800 with myosin during a low speed centrifugation experiment (data not shown), and by the absence of interaction between titin and HMM-coated coverslips during the in vitro motility assays. In contrast, T800 lowered the actin-activated HMM Mg2+- ATPase activity at a saturating T800/actin molar ratio of 1 : 20, regardless of whether or not Tm–Tn was bound to actin. This inhibition was of 39% and 31% in the absence and presence of Tm–Tn (with Ca2+), respectively. In addition, T800 did not significantly alter the EGTA-induced reduction of HMM ATPase activity, measured in the presence of Tm–Tn (62% vs. 69% reduction without and with T800, respectively). This small or nonexistent effect of T800 on the Ca2+-linked regulation of the actin–HMM ATPase is entirely consistent with the lack of effect on the Ca2+-controlled on/off switch of thin filament motion.

T800 specifically reduces HMM binding to the N-terminal part of actin

2.3 ± 2.3 2.4 ± 2.0 1.5 ± 1.7 1.1 ± 1.4 Velocity (lmÆs)1) Stops (% time) Filament length (lm)a

We studied in greater detail the simultaneous binding of T800 and HMM to reconstituted thin filaments containing the Tm–Tn complex at two ionic strengths (80 mM and 180 mM). As shown in Fig. 3, the presence of T800 did not have much effect on HMM binding to actin as judged by the constant amount of HMM in the pellet of ultracentrifuga-

Fig. 2. In vitro motility data. (A) Typical velocity vs. time trace obtained from the analysis of the movement of a single filament during the in vitro motility assay. (B and C) Box representation of the velo- cities (B) and the percentile of STOPS (C) obtained under four different experimental conditions: actin alone (Actin); actin + T800 (Actin + T800); actin + Tm–Tn + CaCl2 (Actin + Tm-Tn); actin + Tm– Tn + T800 + CaCl2 (Actin + Tm–Tn + T800). STOPS corres- pond to the time filaments were stationary, expressed as a percentage of total time of analysis for each moving filament. Boxes extend from the 25th percentile to the 75th percentile of each data set with the horizontal line at the median. Whiskers show the range of the data. Detailed numbers and experimental conditions are reported in Table 1 and in Materials and methods.

Effect of titin on actin–myosin activity (Eur. J. Biochem. 271) 4577

(cid:1) FEBS 2004

Table 2. Effect of T800 on HMM ATPase activity. ATPase activities are the average values of three experiments performed as described in Materials and methods.

+Actin +Tm–Tn +T800 (EGTA) Proteins (in order of assembly) HMM alone +T800 +Actin +Actin +T800 +Actin +Tm–Tn (Ca2+) +Actin +Tm–Tn +T800 (Ca2+) +Actin +Tm–Tn (EGTA)

tion experiments. Under rigor conditions, all HMM was bound to actin and remained in the pellet independently of the other components and the ionic strength of the mixture. In the presence of ATP, the amount of bound HMM was very similar (average value of 52.5 ± 1.1%) in all experi- ments except in the presence of Ca2+ and high ionic strength (average value of 37.1 ± 4.0% for panels Fs + ATP and Gs + ATP; see figure legend for the detailed quantitative data). On the other hand, the percent- age of T800 bound to actin was not affected by ATP and/or by CaCl2 but it was decreased by elevating ionic strength from 80 to 180 mM with average values of 62.3 ± 2.2% and 45.1 ± 3.5%, respectively (compare detailed values in figure legend, panels B and D vs. panels F and H).

and the positively charged segment (also called loop 2) of HMM using EDC-induced cross-linking experiments [55]. We used a two-step cross-linking reaction which has the property of only modifying reactive acidic residues on actin, thereby reducing the number of nonspecific cross-linking reactions. As previously described, the effect of EDC on the actin–HMM complex results in a covalent actin–HMM adduct that migrates as a double band (Fig. 4, [56]). This double band corresponds to two cross-linked products, which are each known to contain an equimolar actin– HMM complex, but involving different cross-linked resi- dues within the actin–HMM interface [55,57]. These cross- linked products were observed in the absence of the lane b), or when regulatory proteins, Tm–Tn (Fig. 4, T800 was added to actin prior to Tm–Tn (Fig. 4, lane c), but they were almost totally absent under physiological

Concerning the actin–HMM interface, we explored the electrostatic contacts between the N-terminal part of actin

ATPase (s)1) 0.17 ± 0.02 0.19 ± 0.02 2.8 ± 0.7 1.7 ± 0.3 2.0 ± 0.5 1.4 ± 0.1 0.6 ± 0.1 0.5 ± 0.1

Fig. 3. T800 and HMM binding to F-actin. Gel electrophoresis analysis of cosedimentation experiments performed as described in Materials and methods. In all experiments, T800 was added to the preformed actin–Tm–Tn complex and HMM was added last. A mixture of all the proteins used is shown in (A). Proteins were preincubated in the presence of CaCl2 (B,C,F,G) or EGTA (D,E,H,I) with or without 2 mM ATP as indicated. After ultracentrifugation, supernatants (s) and pellets (p) were analysed. The percentages of HMM in the pellets were 52.8 (Bs + ATP), 50.9 (Cs + ATP), 52.4 (Ds + ATP), 54.3 (Es + ATP), 39.9 (Fs + ATP), 34.2 (Gs + ATP), 52.6 (Hs + ATP) and 51.7 (Is + ATP). The per- centages of T800 in the pellets were 62.1 (Bs), 61.4 (Bs + ATP), 65.4 (Ds), 60.2 (Ds + ATP), 50.2 (Fs), 44.3 (Fs + ATP), 43.6 (Hs) and 42.3 (Hs + ATP).

(cid:1) FEBS 2004

4578 N. Niederla¨ nder et al. (Eur. J. Biochem. 271)

proteolytic treatment of skeletal myofibrils [42]. It is also noteworthy that we used in this study rabbit back muscles which are heterogeneous in their fibre-type content [59]. Nevertheless, both the homogeneity of the T800 preparation and the results of the mass peptide analysis suggest that the proteolysis and purification protocols selected preferentially the longest skeletal muscle titin isoform.

Our data clearly indicate that T800 binds to actin thin filaments, in good agreement with numerous works previ- ously published on PEVK domain-containing titin frag- ments (see Introduction for references). The PEVK domain remains the main actin-binding candidate identified in the I-band titin region and we propose, without totally exclu- ding other possibilities, that the interaction of T800 with actin is primarily mediated by the PEVK domain. Another actin-binding site candidate was proposed within stretch of residues 1791–2126 of cardiac titin [31], but we are still not certain whether this stretch of residues belongs to T800 as the corresponding residues 1870–2205 of skeletal titin are located close to the hypothetic extreme N-terminal end of T800 (Fig. 1). The interaction between actin and T800 is characterized by an apparent saturating T800/actin molar ratio of 1 : 20 as determined by centrifugation experiments with increasing amounts of T800. This ratio suggests that T800 covers a rather long segment of actin filament and that either it sterically protects part of actin filament region around the interaction site or it contains multiple actin binding sites. This last suggestion is compatible with the presence of repeated stretches of charged/uncharged resi- dues along the PEVK domain [13,60] and with the ionic strength dependence of T800 binding to actin.

conditions, when T800 was added to reconstituted thin filaments (Fig. 4, lane d). An additional faint band was observed at about 70 kDa when Tm–Tn was present. This latter product corresponded presumably to a cross-linking reaction between actin and the Troponin I subunit (Fig. 4, lanes c and d). Finally, the absence of bands above T800 in all experiments argue against a cross-linking reaction of T800 to actin N-terminal segment.

Fig. 4. EDC-induced cross-linking at the actin–HMM interface. Gel electrophoresis analysis of the cross-linking experiments performed on mixtures composed of (in the order of addition): F-actin + T800 (a), F-actin + T800 + HMM (b), F-actin + T800 + Tm–Tn + HMM (c) and F-actin + Tm–Tn + T800 + HMM (d).

Discussion

T800 represents the first native titin fragment containing the entire PEVK domain to be directly extracted from skeletal muscle myofibrils. This titin fragment has the ability to interact with reconstituted thin filaments, and has unex- pected effects on HMM sliding velocity and on HMM binding to actin filaments. These results provide an experi- mental basis to investigate a possible role for titin in the regulation of energetics and force generation in the actomyosin system.

Identification of T800 was performed by MALDI-TOF spectrometry. In fact, the 800 kDa titin fragment represents the largest protein fragment so far identified using MS and in-gel tryptic digestion approaches. T800 contains the titin PEVK domain and is entirely located within the I-band region of the skeletal muscle sarcomere. Its boundaries are estimated to be at the most around residues 1870 and 11 500 of skeletal muscle titin, two loci where titin is free of interaction with its protein partners and could easily be attacked by V8 protease [58]. Outside the I-band region, it is likely that the interactions of titin with myosin (in the A-band region) and actin or a-actinin (in Z-disk and its periphery) are strong enough to protect it against the formation (or prevent the release) of other proteolytic fragments. Such protection could actually explain why, besides T800, only one additional 150 kDa fragment, also belonging to the I-band region, was generated by the

T800 increases the velocity of moving actin filaments. This acceleration is observed both in the absence and in the presence of Tm–Tn (with Ca2+). However, the molecular explanations seem different in the two cases as the number of stationary filaments decreases in the absence of Tm–Tn and increases in its presence, and also because the reduction in actin–HMM cross-linking occurs only in the presence of Tm–Tn. Note that in both cases, the acceleration observed in the motility assay excludes a direct interaction between T800 and the myosin motor domain, as reported for the fibronectin-like domains of the A band part of titin [61]. No attempt was made to further explain the changes observed in the absence of Tm–Tn, as this situation is highly unlikely to occur under physiological conditions. In the presence of Tm–Tn, it is very tempting to correlate the functional changes with the structural modification of the actin–HMM interface, which results in the inhibition of HMM cross- linking to the N-terminal part of actin. This change of the actin–HMM ionic interface would be in contrast to the lack of effect on the HMM binding to actin observed in cosedimentation experiments. Such a discrepancy has been previously related to the fact that the electrostatic contacts taking place at the N-terminal part of actin represent only a very weak ) sometimes considered non-specific ) compo- nent of the actin–myosin interface [57,62–66]. Moreover, it should be mentioned that a reduction in these ionic contacts is compatible with the high efficiency of the Tm–Tn–Ca2+- linked regulation observed in the presence of T800, as a recent report demonstrated that removing the negative charges in this region of actin does not affect the pCa curves of the motion of thin filaments [67].

Effect of titin on actin–myosin activity (Eur. J. Biochem. 271) 4579

(cid:1) FEBS 2004

T800 binding to actin nor the functional effect of titin on the actin–HMM complex.

This reduction in actin–HMM contacts could be due to a direct or indirect competition between T800 and HMM for binding to the negatively charged residues of the N-terminal part of actin. Our data suggest that T800 acts indirectly as T800 alone or added before Tm–Tn on actin filaments cannot displace HMM. This indirect effect could, for example, be mediated through interactions with Tm–Tn, as proposed recently by solid phase experiments [68], or following structural rearrangements within actin as sugges- ted by the decrease in filament length that is observed in the presence of T800. Interestingly, both T800 and Tm–Tn binding to actin induces a shortening of filament length and an increase in myosin sliding velocity, suggesting a strong synergy between these two actin binding components in their functional and molecular effects on actin.

the thin filament

Can the diminution of HMM contacts with the N-terminal part of actin account for the decrease in actin–HMM ATPase activity and for the increase in thin filament sliding velocity while these two activities are supposed to be correlated? Reduction of these contacts was found to induce the loss of correlation between the ATPase and sliding activities in numerous examples [69–73] and it was reported to inhibit the actin-activated myosin ATPase activity in the same way as T800 [64,69]. Inhibition of the ATPase activity was then explained by a slowing down of the formation of active complex in solution. On the other hand, the acceleration of the sliding velocity (and the lower number of stops) of thin filaments could be related to a diminution of the load in the actin–myosin interface.

How should we interpret the effects of T800 observed in vitro with respect to the in vivo functional properties of the actin–myosin complex? Reducing the ATPase activity of the actin–myosin complex could have important effects on the energetic balance during muscle activity, and speeding up the movement of actin filaments could have conse- quences for the generation of active tension. In resting and stretching conditions, there is no overlap between myosin cross-bridges and the titin PEVK domain. Therefore, the effects described in this work are unlikely to take place under these conditions, unless it is demonstrated that they propagate over long distances on actin filaments. In contrast, during muscle shortening, such an overlap, and its functional consequences for the actin-myosin complex, may occur. The facts that T800 interacts with actin in the presence of Tm–Tn, HMM, ATP and up to 0.1 mM CaCl2 and that at 180 mM ionic strength more than 45% of T800 remains bound to actin, support the in vivo extrapolation of titin binding to actin in the sarcomeric I-band region and its functional consequences in striated muscle. However, before extrapolating our results to any physiological environment, one should consider the properties of an additional natural component of framework, nebulin. Nebulin spans the entire length of thin filaments and is capable of modulating the rate of formation of the actin– myosin complex [74,75]. Interestingly, nebulin also interacts with the titin PEVK domain in a calcium/calmodulin and calcium/S100 dependent manner [76]. These data pose questions regarding the precise functional properties of the entirely reconstructed thin filament (actin–Tm–Tn–nebulin) as they relate to myosin binding and activation, and concerning how these properties are regulated by titin. These two missing but essential pieces of information will have to be addressed experimentally both in vitro and in vivo before conclusions can be drawn about the functional consequences of titin binding to thin filaments.

Finally, it will be important to investigate the effects of titin on the actin–myosin complex using titin fragments extracted from other striated muscles, such as cardiac muscle. Titin isoforms from cardiac muscle have been shown to interact more strongly with actin than does the skeletal isoform, and titin is thought to be the main contributor to passive tension development in cardiac muscle [36,37,77]. Studying titin from smooth muscle or nonmuscle tissues will also be of particular interest for at least two reasons: the PEVK content of titin in these isoforms is not well characterized and the structural constraints in these tissues could conceivably allow the PEVK domain to control myosin binding to actin, and to play an even more crucial role in the energetics and the generation of active tension within smooth muscle or nonmuscle stress fibers.

Acknowledgements

The properties of T800 described in this paper diverge significantly from previous reports that supported the idea that binding of the PEVK domain to actin slows down or totally inhibits actin motion over the myosin motor domain [29,30,35,36]. An important issue to consider here is that most of these studies were performed with bacterially expressed titin fragments, and not with muscle-extracted native fragments. The muscle extracted titin fragment, T800, did not tend to aggregate as it remained soluble (in the supernatant after high speed centrifugation) for several days after its purification, nor did it interact in a nonspecific way with myosin or with the glass support during the motility assay. Moreover, T800 interacted very efficiently with actin filaments during cosedimentation experiments and never induced the formation of actin bundles. Note also that T800 did not perturb the Tm–Tn–Ca2+-linked regulation of the reconstituted thin filaments, neither in the motility assay nor in the ATPase experiments, further underscoring its very specific effect on thin filaments. On the other hand, the facts that the recombinant fragments may have interacted with myosin or the coverslips during the motility assay, and that in some cases they induced actin bundles, could easily explain the observed differences between their functional properties and those characterizing T800. But this is not the only parameter that one should consider, as we found for example that T800 had a different effect on the actin–HMM complex depending on whether or not Tm–Tn was present (see above). Two previous studies on actin binding to titin or recombinant titin fragments also used tropomyosin [31] or tropomyosin–troponin [30]. However, their results were controversial as the first one found an inhibition while the second one reported an increase of actin binding in the presence of calcium. In this work, calcium did not change

We are grateful to Jean Derancourt for his help in the mass spectrometry analysis of the T800 fragment (Montpellier Genopole Proteome facilities, http://genopole.igh.cnrs.fr/), Pierre Travo (CRBM imaging facilities, http://www.crbm.cnrs-mop.fr/Imagcell.html) for advice and help setting up the in vitro motility assay, and Juliette

(cid:1) FEBS 2004

4580 N. Niederla¨ nder et al. (Eur. J. Biochem. 271)

VanDijk for her critical reading of the manuscript. This work was supported by the French Centre National de la Recherche Scientifi- que. 20. Trombita´ s, K., Greaser, M., Labeit, S., Jin, J.P., Kellermayer, M., Helmes, M. & Granzier, H.L. (1998) Titin extensibility in situ: entropic elasticity of permanently folded and permanently un- folded molecular segments. J. Cell Biol. 140, 853–859.

References

1. Tskhovrebova, L. & Trinick, J. (2003) Titin: properties and family 21. Trombita´ s, K., Rodkan, A., Freiburg, A., Centner, T., Wu, Y., Labeit, S. & Granzier, H.L. (2000) Extensibility of isoforms of cardiac titin: variation in fractional extension provides a basis for diastolic stiffness diversity. Biophys. J. 79, 3226–3234. relationships. Nat. Rev. Mol. Cell Biol. 4, 679–689.

22. Helmes, M., Trombita´ s, K., Centner, T., Kellermayer, M., Labeit, S., Linke, W.A. & Granzier, H.L. (1999) Mechanically driven contour-length adjustment in rat cardiac titin’s unique N2B sequence: titin is an adjustable spring. Circ. Res. 84, 1339–1352. 8 2. Keller,T.C.S., III, Higginbotham, M., Kazmierski, S., Kim, K.T. & Velichkova, M. (2000) Role of titin in non muscle and smooth muscle cells. In Elastic Filaments of the Cell (Granzier, H.L. & Pollack, G.H., eds), pp.265–281. Kluwer Academic/Plenum Pub- lishers, Dordrecht, the Netherlands.

23. Li, H., Oberhauser, A.F., Redick, S.D., Carrion-Vazquez, M., Herickson, H.P. & Fernandez, J.M. (2001) Multiple conforma- tions of PEVK proteins detected by single-molecule techniques. Proc. Natl. Acad. Sci. USA 98, 10682–10686. 3. Peckham, M., Young, P. & Gautel, M. (1997) Constitutive and variable regions of Z-disk titin/connectin in myofibril formation: a dominant-negative screen. Cell Struct. Funct. 22, 95–101.

4. Van der Ven, P.F., Bartsch, J.W., Gautel, M., Jockusch, H. & Furst, D.O. (2000) A functional knock-out of titin results in defective myofibrils assembly. J. Cell Sci. 113, 1405–1414. 24. Watanabe, K., Nair, P., Labeit, D., Kellermayer, M.S.Z., Greaser, M., Labeit, S. & Granzier, H.L. (2002) Molecular mechanics of cardiac titin’s PEVK and N2B spring elements. J. Biol. Chem. 277, 11549–11558.

25. Kimura, S., Maruyama, K. & Huang, Y.P. (1984) Interactions of muscle beta-connecting with myosin, actin, and actomyosin at low ionic strengths. J. Biochem. 96, 499–506. 5. Gotthardt, M., Hammer, R.E., Hubner, N., Monti, J., Witt, C.C., McNabb, M., Richardson, J.A., Granzier, H., Labeit, S. & Herz, J. (2003) Conditional expression of mutant M-line titins results in cardiomyopathiy with altered sarcomere. J. Biol. Chem. 278, 6059–6065. 6. Nishino, I. & Ozawa, E. (2002) Muscular dystrophies. Curr. Opin. Neurol. 15, 539–544.

26. Funatsu, T., Kono, E., Higuchi, H., Kimura, S., Ishiwata, S., Yoshioka, T., Maruyama, K. & Tsukita, S. (1993) Elastic fila- ments in situ in cardiac muscle: deep-etch replica analysis in combination with selective removal of actin and myosin filaments. J. Cell Biol. 120, 711–724. 27. Soteriou, A., Gamage, M. & Trinick, J. (1993) A survey of the 7. Whiting, A., Wardale, J. & Trinick, J. (1989) Does titin reg- ulate the length of muscle thick filament? J. Mol. Biol. 205, 263– 268. interactions made by titin. J. Cell Sci. 104, 119–123. 28. Jin, J.-P. (1995) Cloned rat cardiac titin class I and II motifs. J. Biol. Chem. 270, 6908–6916. 8. Van der Ven, P.F., Ehler, E., Perriard, J.C. & Furst, D.O. (1999) Thick filament assembly occurs after the formation of a cytoske- letal scaffold. J. Muscle Res. Cell Motil. 20, 569–579.

29. Li, Q., Jin, J.-P. & Granzier, H.L. (1995) The effect of genetically expressed cardiac titin fragments on in vitro actin motility. Biophys. J. 69, 1508–1518. 9. Miller, G., Musa, H., Gautel, M. & Peckham, M. (2003) A tar- geted deletion of the C-terminal end of titin, including the titin kinase domain, impairs myofibrillogenesis. J. Cell Sci. 116, 4811– 4819.

30. Kellermayer, M.S. & Granzier, H.L. (1996) Calcium-dependent inhibition of in vitro thin-filament motility by native titin. FEBS Lett. 380, 281–286. 10. Horowits, R., Kempner, E.S., Bisher, M.E. & Podolsky, R.J. (1986) A physiological role for titin and nebulin in skeletal muscle. Nature 323, 160–164. 11. Magid, A. & Law, D.J. (1985) Myofibrils bear most of the resting 31. Linke, W.A., Ivemeyer, M., Labeit, S., Hinssen, H., Ruegg, J.C. & Gautel, M. (1997) Actin–titin interaction in cardiac myofibrils: probing a physiological role. Biophys. J. 73, 905–919. tension in frog skeletal muscle. Science 230, 1280–1282.

32. Granzier, H.L., Kellermayer, M., Helmes, M. & Trombita´ s, K. (1997) Titin elasticity and mechanism of passive force develop- ment in rat cardiac myocytes probed by thin-filament extraction. Biophys. J. 73, 2043–2053. 12. Wang, K., McCarter, R., Wright, J., Beverly, J. & Ramirez- Mitchell, R. (1991) Regulation of skeletal muscle stiffness and elasticity by titin isoforms: a test of the segmental extension model of resting tension. Proc. Natl Acad. Sci. USA 88, 7101–7105. 13. Labeit, S. & Kolmerer, B. (1995) Titins: giant proteins in charge of muscle ultrastructure and elasticity. Science 270, 293–296. 33. Trombita´ s, K. & Granzier, H.L. (1997) Actin removal from car- diac myocytes shows that near Z. line titin attaches to actin while under tension. Am. J. Physiol. Cell Physiol. 273, 662–670. 34. Jin, J.P. (2000) Titin thin filament interaction and potential role in muscle function. Adv. Exp. Med. Biol. 481, 319–333. 14. Gautel, M. & Goulding, D. (1996) A molecular map of titin/ connecting elasticity reveals two different mechanisms acting in series. FEBS Lett. 385, 11–14.

15. Tskhovrebova, L., Trinick, J., Sleep, J.A. & Simmons, R.M. (1997) Elasticity and unfolding of single molecules of the giant muscle protein titin. Nature 387, 308–312. 35. Kulke, M., Fujika-Becker, S., Rostokova, E., Neagoe, C., Labeit, D., Manstein, D.J., Gautel, M. & Linke, W.A. (2001) Interaction between PEVK-titin and actin filaments. Circ. Res. 89, 874–881.

16. Kellermayer, M.S.Z., Smith, S.B., Granzier, H.L. & Bustamante, C. (1997) Folding-unfolding transitions in single titin molecules characterized with laser tweezers. Science 276, 1112–1116.

36. Yamasaki, R., Berri, M., Wu, Y., Trombita´ s, K., McNabb, M., Kellermayer, M.S.Z., Witt, C., Labeit, D., Labeit, S., Greaser, M. & Granzier, H.L. (2001) Titin–actin interaction in mouse myo- cardium: passive tension modulation and its regulation by cal- cium/S100A1. Biophys. J. 81, 2297–2313. 17. Linke, W.A., Ivemeyer, M., Mundel, P., Stockmeier, M.R. & Kolmerer, B. (1998) Nature of PEVK-titin elasticity in skeletal muscle. Proc. Natl Acad. Sci. USA 95, 8052–8057.

18. Linke, W.A., Stockmeier, M.R., Ivemeyer, M., Hosser, H. & Mundel, P. (1998) Characterizing titin’s I-band Ig domain region as an entropic spring. J. Cell Sci. 111, 1567–1574. 37. Linke, W.A., Kulke, M., Li, H., Fujita-Becker, S., Neagoe, C., Manstein, D.J., Gautel, M. & Fernandez, J.M. (2002) PEVK domain of titin: an entropic spring with actin-binding properties. J. Struct. Biol. 137, 194–205.

38. Offer, G., Baker, H. & Baker, L. (1972) Interaction of monomeric and polymeric actin with myosin subfragment 1. J. Mol. Biol. 66, 435–444. 19. Linke, W.A., Rudy, D.E., Centner, T., Gautel, M., Witt, C., Labeit, S. & Gregorio, C.C. (1999) I-band titin in cardiac muscle is a three-element molecular spring and is critical for maintaining thin filament structure. J. Cell Biol. 146, 631–644.

Effect of titin on actin–myosin activity (Eur. J. Biochem. 271) 4581

(cid:1) FEBS 2004

9

10

60. 39.

Eisenberg, E. & Kielley, W.W. (1974) Troponin-tropomyosin complex: column chromatographic separation and activity of the three, active troponin components with and without tropomyosin present. J. Biol. Chem. 249, 4742–4748. 40. Smillie, L.B. (1982) Preparation and identification of alpha- and Freiburg, A., Trombitas, K., Hell, W., Cazorla, O., Fougerousse, F., Centner, T., Kolmerer, B., Witt, C., Beckmann, J.S., Gegorio, C.C., Granzier, H. & Labeit, S. (2000) Series of exon-skipping events in the elastic spring region of titin as the structural basis for myofibrillar elastic diversity. Circ. Res. 86, 1114–1121. beta-tropomyosins. Methods Enzymol. 85, 234–241. 41. Potter, J.D. (1982) Preparation of troponin and its subunits. Methods Enzymol. 85, 241–263.

61. Muhle-Goll, C., Habeck, M., Cazorla, O., Nilges, M., Labeit, S. & Granzier, H.L. (2001) Structural and functional studies of titin’s fn3 modules reveal conserved surface patterns and binding to myosin S1 – a possible role in the Frank-Starling mechanism of the heart. J. Mol Biol. 313, 431–447.

42. Astier, C., Labbe´ , J.-P., Roustan, C. & Benyamin, Y. (1993) Effects of different enzymatic treatments on the release of titin fragments from rabbit skeletal myofibrils. Biochem. J. 290, 731– 734.

43. Astier, C., Raynaud, F., Lebart, M.C., Roustan, C. & Benyamin, Y. (1998) Binding of a native titin fragment to actin is regulated by PiP2. FEBS Lett. 429, 95–98. 62. Miller, C.J., Wong, W.W., Bobkova, E., Rubenstein, P.A. & Reisler, E. (1996) Mutational analysis of the role of the N-termi- nus of actin in actomyosin interactions. Comparison with other mutant actins and implications for the cross-bridge cycle. Bio- chemistry 35, 16557–16565.

63. Wong, W.W., Doyle, T.C. & Reisler, E. (1999) Nonspecific weak actomyosin interactions: relocation of charged residues in sub- domain 1 of actin does not alter actomyosin function. Biochem- istry 38, 1365–1370.

44. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. 45. Rosenfeld, J., Capdevielle, J., Guillemot, J.-C. & Ferrara, P. (1992) In-gel digestion of proteins for internal sequence analysis after one- or two-dimensional gel electrophoresis. Anal. Biochem. 203, 173–179. 64. Van Dijk, J., Furch, M., Lafont, C., Manstein, D.J. & Chausse- pied, P. (1999) Functional characterization of the secondary actin binding site of myosin II. Biochemistry 38, 15078–15085.

65. Van Dijk, J., Furch, M., Derancourt, J., Batra, R., Knescht, M.L., Manstein, D.J. & Chaussepied, P. (1999) Differences in the ionic interaction of actin with the motor domains of nonmuscle and muscle myosin II. Eur. J. Biochem. 260, 672–683. 66. Chaussepied, P. & Van Dijk, J. (2002) Role of charges in acto- 46. Mornet, D., Bertrand, R., Pantel, P., Audemard, E. & Kassab, R. (1981) Proteolytic approach to structure and function of actin recognition site in myosin heads, Biochemistry 20, 2110–2120. 47. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriolphage T4. Nature 227, 680– 685. myosin interactions. Results Probl Cell Differ. 36, 51–64.

67. Wong, W.W., Gerson, J.H., Rubenstein, P.A. & Reisler, E. (2002) Thin filament regulation and ionic interactions between the N-terminal region in actin and troponin. Biophys. J. 83, 2726– 2732. 48. Perkins, D.N., Pappin, D.J.C., Creasy, D.M. & Cottrell, J.S. (1999) Probability-based protein identification by searching se- quence databases using mass spectrometry data. Electrophoresis 20, 3551–3567.

49. Marston, S.B., Fraser, E.D., Bing, W. & Roper, G. (1996) A simple method for automatic tracking of actin filaments in the motility assay. J. Muscle Res. Cell Motil. 17, 497–506. 68. Raynaud, F., Astier, C. & Benyamin, Y. (2004) Evidence for a direct but sequential binding of titin to tropomyosin and actin filaments. Biochim. Biophys. Acta 1700, 171–178.

69. Uyeda, T.Q., Ruppel, K.M. & Spudich, J.A. (1994) Enzymatic activities correlate with chimaeric substitutions at the actin-bind- ing face of myosin. Nature 368, 567–569. 50. DeBeer, E.L., Sontrop, A.M.A., Kellermayer, M.S., Galambos, C. & Pollack, G.H. (1997) Actin-filament motion in the in vitro motility assay has a periodic component. Cell Motil. Cytoskeleton 38, 341–350.

70. Umemoto, S. & Sellers, J.R. (1990) Characterization of in vitro motility assays using smooth muscle and cytoplasmic myosins. J. Biol. Chem. 265, 14864–14869. 51. Hamelik, W., Zegers, J.G., Treijtel, B.W. & Blange´ , T. (1999) Path reconstruction as a tool for actin filament speed determination in the in vitro motility assay. Anal. Biochem. 273, 12–19.

71. Anson, M. (1992) Temperature dependence and arrhenius acti- vation energy of F-actin velocity generated in vitro by skeletal myosin. J. Mol. Biol. 224, 1029–1038. 52. Fraser, I.D.C. & Marston, S.B. (1995) In vitro motility analysis of actin-tropomyosin regulation by troponin and calcium. J. Biol. Chem. 270, 7836–7841.

72. Bobkov, A.A., Bobkova, E.A., Lin, S.H. & Reisler, E. (1996) The role of surface loops (residues 204–216 and 627–646) in the motor function of the myosin head. Proc. Natl Acad. Sci. USA 93, 2285– 2289. 53. Gordon, A.M., Chen, Y., Liang, B., LaMadrid, M., Luo, Z. & Chase, P.B. (1998) Skeletal muscle regulatory proteins enhance F-actin in vitro motility. Adv. Exp. Med. Biol. 453, 187–196. 54. Gordon, A.M., Homsher, E. & Regnier, M. (2000) Regulation of contraction in striated muscle. Physiol. Rev. 80, 853–924. 73. Bobkova, E.A., Bobkov, A.A., Levidsky, D.I. & Reisler, E. (1999) Effects of SH1 and SH2 modifications on myosin: similarities and differences. Biophys. J. 76, 1001–1009.

55. Van Dijk, J., Fernandez, C. & Chaussepied, P. (1998) Effect of ATP analogues on the actin–myosin interface. Biochemistry 37, 8385–8394.

74. Kruger, M., Wright, J. & Wang, K. (1991) Nebulin as a length regulator of thin filaments of vertebrate skeletal muscles: correla- tion of thin filament length, nebulin size, and epitope profile. J. Cell Biol. 115, 97–105. 56. Bonafe, N. & Chaussepied, P. (1995) A single myosin head can be cross-linked to the N termini of two adjacent actin monomers. Biophys. J. 68, 35S–43S.

75. Root, D.D. & Wang, K. (2001) High-affinity actin binding nebulin fragments influence the actoS1 complex. Biochemistry 40, 1171– 1186. 57. Van Dijk, J., Celine, F., Barman, T. & Chaussepied, P. (2000) Interaction of myosin with F-actin: time-dependent changes at the interface are not slow. Biophys. J. 78, 3093–3102.

76. Gutierrez-Cruz, G., van Heerden, A.H. & Wang, K. (2001) Modular motif, structural folds and affinity profiles of the PEVK segment of human fetal skeletal muscle titin. J. Biol. Chem. 276, 7442–7449. 58. Clark, K.A., McElhinny, A.S., Beckerle, M.C. & Gregorio, C.C. (2002) Striated muscle cytoarchitecture: an intricate web of form and function. Annu. Rev. Cell Dev. Biol. 18, 637–706.

77. Leguennec, J.Y., Cazorla, O., Lacampagne, O. & Vassort, G. (2000) Is titin the length sensor in cardiac muscle: Physiological and pathophysiological perspectives. Adv. Exp. Mol. Biol. 481, 387–348. 59. Barron, D.J., Etherington, P.J., Winlove, C.P. & Pepper, J.R. (2001) Scapular insertion of the rabbit latissimus dorsi muscle: gross anatomy and fibre-type composition. Cell Tissues Organs 168, 312–318.