Journal of Muscle Research and Cell Motility 25: 489–495, 2004.
2004 Kluwer Academic Publishers. Printed in the Netherlands.

489

Abstracts of Poster Presentations 3D Structure of vertebrate (fish) muscle myosin filaments by electron microscopy and single particle analysis HIND A. AL-KHAYAT1, EDWARD P. MORRIS1, ANDREW S. POWELL1, ROBERT W. KENSLER2 and JOHN M. SQUIRE1 1 Biological Structure and Function Section, Biomedical Sciences Division, Imperial College London, London SW7 2AZ, UK 2 Department of Anatomy, University of Puerto Rico Medical School, San Juan, Puerto Rico 00936-5067 In order to understand the structural changes involved in the forceproducing myosin cross-bridge cycle in vertebrate muscle it is important to know the conformation of the myosin heads at the start of the cycle (i.e. the relaxed state) on the surface of the myosin filaments. A better understanding of myosin filament structure will also help to define the effects of disease-related mutations in thick filament proteins. Myosin filaments isolated from goldfish muscle under relaxing conditions and viewed in negative stain by electron microscopy (EM) were divided into segments and subjected to 3-dimensional (3-D) single particle analysis by IMAGIC software (van Heel et al. 2000, Quart Rev Biophys 33: 307–369). This allowed the known systematic departure from helical symmetry characteristic of vertebrate striated muscle myosin filaments to be preserved and visualised. The resulting 3-D reconstruction revealed details of the myosin head arrangement in relaxed fish muscle myosin filaments to about 50 A˚ resolution. As well as the well-known axial perturbation, there appear to be substantial azimuthal perturbations within successive 143 A˚spaced myosin head ‘crowns’. The new observations are largely consistent with the head conformations previously deduced by modelling relaxed myosin filament structure from X-ray diffraction data (Hudson et al. 1997, J Mol Biol 273: 440–455). However, unlike X-ray diffraction, EM has the advantage that it provides information both on the polarity of the myosin head array relative to the neighbouring actin filaments in the A-band, important in understanding the geometry of the myosin head interaction with actin during the cross-bridge cycle, and also on what appear to be non-myosin densities on and between the head crowns. These may represent the locations of extra proteins such as titin and C-protein.

NMR studies of skeletal troponin C in the ternary complex T.M.A. BlUMENSCHEIN1, D.B. STONE2 and BRIAN D. SYKES1 1 Department of Biochemistry, University of Alberta, T6G 2H7, Canada 2 Department of Biophysics and Biochemistry, UCSF Mission Bay Campus, 94143-2240, USA We have used TROSY NMR to acquire 3D spectra of {15N, 13C and 2H}-TnC in a ternary complex with 2H-TnI and 2H-TnT-T2, in the presence of Ca2+ or EGTA. More than two thirds of the residues were assigned in each condition. Assignments in the presence of Ca2+ were based on the chemical shifts for the Ndomain of TnC bound to TnI96–148 (McKay et al., Biochemistry 37:

12419) and the C-domain of TnC bound to TnI1–40 (Mercier et al. Biochemistry 40: 10063), while for EGTA, skeletal and cardiac apo assignments (Gagne´ et al. Nat Struct Biol 2: 784; Sia et al. J Biol Chem 272: 18216) were used for the N-domain. The very good agreement between the domains and TnC in the complex indicates that the structure is very similar in both cases. In the presence of Ca2+, more residues were assigned in the C than in the N-domain. Helix D could not be assigned, since the linker region between the two domains cannot be studied when the two domains are separated. In the presence of EGTA, N-domain residues showed better agreement, while larger portions of the C-domain could not be assigned. These differences in the spectra suggest that the different domains are under different amounts of conformational exchange in each condition.

Regulation of oscillatory contraction by troponin in insect flight muscle U. KRZIC1,2, F. QIU1, B. AGIANIAN1, W.A. LINKE2, K. LEONARD1 and B. BULLARD1 1 EMBL, D-69117 Heidelberg, Germany 2 Institute of Physiology, University of Heidelberg, D69120 Heidelberg, Germany Oscillatory contraction of insect flight muscle (IFM) requires rapid activation and de-activation. This is achieved by periodic stretches and releases to fibres at constant [Ca2+]. The troponin complex responds to stretch. In Lethocerus IFM, there are two isoforms of TnC: F1 which binds one Ca2+ in the C-terminal lobe, and F2 which binds an additional Ca2+ in the N-terminal lobe, with lower affinity. By exchanging TnC isoforms in skinned fibres, we have found that F2 responds to raised Ca2+ and F1 responds to stretch. The effect of F1 and F2 on the power produced during oscillatory contraction was investigated at different frequencies of oscillation and [Ca2+]. Power was estimated from work loops. Fibres substituted with F1 produced power similar to those of native fibres; whereas fibres with F2 did not produce work loops. Maximum power in native fibres was obtained at lM [Ca2+] and 5 Hz; at higher [Ca2+], isometric tension was greater, due to full activation by F2, but force produced during oscillations was less. Therefore, a TnC with no exchangeable Ca2+ site is needed for oscillatory contraction, and maximum power is obtained when the fibres are not fully activated by Ca2+-binding to F2.

Myosin IIA is required for neuronal growth cone collapse S.R. WYLIE and P.D. CHANTLER Royal Veterinary College, University of London, UK Antisense oligonucleotides targeting isoform-specific sequences encoding myosin IIA or myosin IIB can lead to disruption of neuronal cell adhesion (Wylie and Chantler 2001, Nat Cell Biol 3: 88–92) and neurite outgrowth (Wylie et al., 1998, PNAS 95: 12967– 12972), respectively. We have shown recently that neurite retraction, elicited by myosin IIB antisense oligonucleotides following normal outgrowth, is retarded by addition of myosin IIA antisense

490 oligonucleotides, suggesting that myosin IIA could be the motor responsible for retraction (Wylie and Chantler 2003, Mol Biol Cell 14: 4654–4666). Neuronal growth cones collapse rapidly when treated with lyosphosphatidate (LPA) or thrombin, the process completing within 30 min. However, we show that if cells are pretreated with either the Rho-kinase (ROCK) inhibitor, Y27632, or with antisense oligonucleotides directed against myosin IIA sequence, their ability to retract is severely impaired. As expected, sense or scrambled oligonucleotides derived from myosin IIA sequence had no effect. Significantly, sense, scrambled or antisense oligonucleotides derived from myosin IIB sequence, also exhibited no inhibitory effect on LPA- or thrombin-induced retraction. Y27632 did not inhibit neurite outgrowth. Taken together, these results are consistent with a model for growth cone motility in which myosin IIB brings about neurite outgrowth and myosin IIA is required constitutively for cell adhesion and neurite retraction.

Time resolved X-ray diffraction studies of active bony fish muscle FELICITY EAKINS1, CARLO KNUPP1,2, CHRISTIAN PINALI2 and JOHN M. SQUIRE1 1 Biological Structure & Function Section, Biomedical Sciences Division, Imperial College London, UK 2 School of Optometry & Vision Sciences, Cardiff University, Cardiff, UK Previous X-ray studies carried out on bony fish muscle have looked at fluctuations in the X-ray pattern during contraction in conjunction with the muscle tension time course (Harford and Squire 1990). Up until now the effects of any sarcomere length change which might occur during contraction have been neglected. There is evidence for a small, fast change in lattice spacing during contraction and so previously it was argued that the sarcomere length change may be small (Harford and Squire, 1992). However, this needs to be confirmed and if the tension time-course is affected by maintaining sarcomere length, then sarcomere length will need to be controlled. This poster describes the laser and detector system which has been developed to control the sarcomere length of plaice fin muscles whilst collecting Xray patterns. It also presents some of the preliminary data obtained using this system in experiments on beamline 16.1 at the CLRC Daresbury Synchrotron Radiation Source.

In situ measurents of calpain activity in isolated FDB fibres of the mouse F. DE BACKER, PH. GAILLY and J.M. GILLIS Department of Physiology, Catholic University of Louvain, 1200 Bruxelles, Belgium Calpain activity is involved in myoblast fusion and is thought to play a role in modelling the cytoskeleton. Uncontrolled calpain activation may occur in pathological processes such as dystrophinopathies. The in situ activity of calpains in fully differentiated muscle fibres is unknown. We used the permeant and specific probe BOC-Leu-MetCMAC (Molecular probes) which becomes fluorescent after cleavage by m and l calpains. We study fibres after collagenase isolation, followed by plating on an artificial extracellular matrix. Activity (recorded as the number of photons/min, through a standard window) is increased in the presence of the Ca ionophore 4-bromo-A23187 (Alexis) and greatly reduced after incubation with the permeant EGTA-AM, or the specific inhibitor Z-Leu-Leu-CHO (Biomol Lab.). Thus, resting fibres show a detectable calpain activity which is Cadependent. Two hours after isolation, hypo-osmotic shock (0.6 · normal) increases calpain activity by 1.8 ·, but, surprisingly, is no longer effective on fibres plated for >24 h.

Protective effect on fatigue development and twitch potentiation by LASSBIO-294 in mammalian skeletal muscle H. GONZALEZ-SERRATOS1, E. PEREIRA2, R. CHANG2, E. BARREIRO3 and E.X. ALBUQUERQUE2,4 1 Department of Physiology 2 Department of Pharmacology and Experimental Therapeutics, University of Maryland School of Medicine, Baltimore, MD, USA 3 Faculdade de Farmacia, 4Depto Farmacol Bas Clin, ICB, CCS, Univ Fed Rio de Janeiro, Rio de Janeiro, RJ, Brazil The extensor digitorum longus (EDL) muscles of new-born (NB) rats have slow Type I contractile and fatigability characteristics but fast Type II myosin. Type I muscles have a poor T-system, are fatigue resistant and the SR Ca2+ ATPase is strongly stimulated by cyclic AMP. Thus, in NB rat EDL muscles these characteristics are of the slow type. If LASSBio-294, a phosphodiesterase inhibitor, does not affect acto–myosin interaction and early e–c coupling steps, then it should have a profound effect on NB EDL muscles. We tested this hypothesis by studying the effect of LASSBio-294 in EDL muscles dissected from 2 to 10 days old post-partum rats. Fatigue was induced with 20 Hz 3 s stimulations, repeated every 6.9 s. LASSBio294 (50 lM) prolonged dramatically the time of fatigue development. Compared with controls, it took 3.4–9.7 times longer for tetanic force to decline to 50% of pre-fatigue values, representing a 70–89% fatigue protection. Twitch tension was potentiated by 54%. Thus, LASSBio-294 protects fatigue development by stimulating the SR ATPase.

Myosin heavy chain isoforms and isometric cross-bridge cycling kinetics of human ventricular myocytes TODD J. HERRON1, RAJIV R. CHATURVEDI2, MAX BAGHAI1, DARRYL F. SHORE3, CONAL B. AUSTIN1, DAVID R. ANDERSON1 and JONATHAN C. KENTISH1 1 Centre for Cardiovascular Biology & Medicine, King’s College London, London, UK SE17EH 2 The Hospital for Sick Children, Toronto, Canada, M5G1X8 3 Royal Brompton Hospital, London SW36NP Cardiac myosin heavy chain (MyHC) is expressed as two distinct isoforms, namely, a- and b-MyHC. We examined the relative expression of these MyHC isoforms in non-failing and end-stage failing ventricles. Non-failing human ventricular tissue was obtained from transplant donor hearts or from patients undergoing corrective heart surgery. End-stage failing ventricular tissue was obtained from heart transplant recipients at the time of transplant. a-MyHC protein represented 63% of total MyHC protein in non-failing ventricular tissue (both sets, n = 7), while there was effectively no detectable a-MyHC protein in the ventricles of end-stage failing human hearts (n = 4). The rate constant for tension redevelopment, ktr, was also measured in maximal Ca2+-activated skinned ventricular myocytes taken from these groups. Skinned myocytes from non-failing ventricles (both sets) exhibited significantly faster ktr (1.34 ± 0.08 s)1, n = 14) than ventricular myocytes from failing human hearts (0.86 ± 0.09 s)1, n = 5; P = 0.05). These data show that the loss of a-MyHC slows cross-bridge cycling kinetics of human ventricular myocytes, and may contribute to ventricular contractile dysfunction in heart failure.

491 Probing the strain dependence of attached smooth muscle myosin II states N.M. KAD, J.B. PATLAK, P.M. FAGNANT, K.M. TRYBUS and D.M. WARSHAW Department of Molecular Physiology & Biophysics, University of Vermont, Burlington VT 05401 USA We have employed a force-clamp laser trap technique to investigate the strain dependence of attached myosin states within the crossbridge cycle in both expressed smooth muscle S1 and a G709V point mutant. Uyeda et al. (2002) showed that the equivalent mutation in Dicty myosin II (G680V) significantly slowed crossbridge kinetics and appeared to populate an attached pre-powerstroke state. As expected, the G709V mutant slowed motility over 100-fold. At the single-molecule level, the slowing of kinetics was found to be due to a 10-fold slower ATP-binding rate and a 100fold slower ADP release rate compared to control S1. These kinetic estimates were confirmed by stopped-flow solution studies. For control S1, rapid force-clamping, within 10 ms of myosin attachment, revealed a slowing or acceleration at positive or negative loads, respectively, in both the rate of ADP release and ATPbinding. However, the G709V mutant exhibits unusual load dependence, perhaps implying the presence of a force labile state that is not the ADP bound or rigor states.

Orientation and motility of actin in different intermediates of the ATP cycle S.S. KHAIMINA1, A. WRZOSEC2, R. DABROWSKA2 and YU.S. BOROVIKOV1 1 Institute of Cytology RAS, Russia 2 Nencki Institute of Experimental Biology PAS, Poland The conformational changes in FITC-phalloidin-labelled F-actin of ghost muscle fibers induced by myosin subfragment-1 (S1) in the absence or presence of MgADP, MgAMP-PNP, MgATPyS or MgATP, were investigated by polarized fluorescence technique. The fluorescent label was bound rigidly with actin monomers, making absorption and emission dipoles sensible to any changes in orientation and mobility of actin monomers in the thin filaments of muscle fibers. It was shown that the orientation and mobility of dye dipoles change discretely when modeling different intermediate states of actomyosin, indicating multi-stage changes of the actin monomers orientation and mobility in ATP cycle. The most appreciable differences in orientation (by 4) and in mobility (by 43%) of actin were revealed between actomyosin states induced by MgADP and MgATP, respectively. It is assumed that in ATP cycle there takes place a multi-stage alteration of rotation and mobility of the actin monomers in thin filaments. Actin monomers and the head of myosin molecule move like a solid body, whose elasticity depends on actin–myosin binding.

Conformations of myosin heads in muscle studied with low-angle X-ray diffraction and direct modelling N. KOUBASSOVA1, S.Y. BERSHITSKY2, M.A. FERENCZI3 and A.K. TSATURYAN1 1 Institute of Mechanics, Moscow University, Russia; 2 Institute of Immunology and Physiology, Yekaterinburg, Russia 3 Imperial College, London, UK The structure of myosin heads in contracting muscle was studied with time-resolved low-angle X-ray diffraction. Data were collected

from bundles of permeabilised rabbit fibres subjected to Joule temperature jump (T-jump, from 5 to 30 C completed in 1 ms) using the RAPID X-ray detector at the ESRF ID02 beam line. A 3D structural model of overlap zone of the sarcomere was developed for the interpretation of diffraction patterns. Dependence of the calculated diffraction pattern on model parameters was systematically studied. The model demonstrates that observed changes in the diffraction pattern are explained by stereo-specific locking of catalytic domains of the myosin heads on actin. Comparison of calculated and observed diffraction patterns show that at 30 C about 40% of the heads are stereo-specifically attached to actin with their light chain domains oriented nearly perpendicular to the fibre axis. Supported by DL, ESRF, EMBL, HHMI, INTAS, MRC, NATO, RFBR.

Defining the positions of tropomyosin and troponin on thin filaments W. LEHMAN1, R. CRAIG2, L. TOBACMAN3, A. PIRANI1, V. HATCH1 and C. XU4 1 Boston University School of Medicine, USA 2 University of Massachusetts Medical School, USA 3 University of Illinois College of Medicine, USA 4 Brandeis University, USA EM and helical reconstruction (HR) shows that Ca2+ causes tropomyosin (Tm) to move on actin (Ac) in troponin (Tn)-regulated thin filaments (TF) from the outer domain (‘B’-position) to the inner domain (‘C’-position), consistent with the ‘steric model’ of muscle regulation. HR of TFs (~150–300 nm long stretches) defines the average position of Tm on Ac but not potentially different local Tm positions. IHRSR (‘‘Iterative Helical Real Space Reconstruction’’, Egelman, 2000) of short TF segments (30–40 nm) is better suited to determine this. IHRSR reconstructions of Ca2+-free and Ca2+-treated TF were identical to those first obtained by HR. When each of the two data sets was sorted into classes by cross-correlation to B- or C-state models, the majority of the respective segments produced reconstructions indistinguishable from those above. However, a subpopulation of Ca2+-free segments (15–20% of the total) clearly fitted best to the Cstate model and yielded a reconstruction with Tm in the C-state position; conversely, a subpopulation of the Ca2+-treated segments fitted best to the B-state model, yielding this type reconstruction. Our results are consistent with tropomyosin being in equilibrium between distinct positions, with the B- or C-states favored depending on Ca2+ levels (cf. Lehrer and Geeves, 1998). Tn density also could be detected in each data set adjacent to tropomyosin on the extreme outer edge of actin subdomain 1.

The effects of troponin C isoform on the action of the cardiotonic potentiator, EMD 57033 SIMON LIPSCOMB1, LAURA PRESTON1, PAUL ROBINSON2, CHARLES REDWOOD2 and CHRIS ASHLEY1 1 Department of Physiology, University of Oxford, UK 2 Department of Cardiovascular Medicine, University of Oxford, UK The endogenous troponin C (TnC) or whole troponin (Tn) of glycerinated rabbit psoas fibres was replaced with the relevant human cardiac isoforms. The cardiotonic potentiator EMD 57033 caused a significant increase in maximal calcium-activated force in all preparations (native 140.7 ± 5.5%, skeletal WT replacement 141.4 ± 2.3%, cardiac Tn 146 ± 10.8%, cardiac TnC 228 ± 2.3%). The pCa50 of

492 normalized pCa-force curves was not significantly altered. Stiffness was measured by applying a rapid stretch of 1% initial length (Lo). EMD 57033 did not alter the stiffness/force ratio in any preparation, hence force per crossbridge is not altered. Preliminary studies on ktr (the time of force redevelopment following a rapid shortening) shows that EMD 57033 slows the rate of crossbridge transitions, suggesting that EMD acts via rate-modulation. The mechanism of action appears to be independent of the TnC isoform, but the magnitude of effect is slightly greater with the cardiac TnC isoform.

Temperature dependence of the working stroke of muscle myosin V. LOMBARDI, V. DECOSTRE, P. BIANCO and G. PIAZZESI Universita` di Firenze and Istituto Nazionale di Fisica della Materia, Italy In intact fibres from frog skeletal muscle (sarcomere length 2.1 lm, temperature 4 C) the size of the working stroke in the myosin heads attached to actin can be estimated by recording the filament sliding (LT) during the early phase of the isotonic shortening (phase 2 of the velocity transient) following a step reduction in force from the isometric tetanic force (T0) to a fraction (T) of T0 (Piazzesi et al., 2002, J Physiol 545:145; Reconditi et al., 2004, Nature 428: 578). In this way we showed that the energy of the myosin working stroke (ET = LT T) increases with T reaching a maximum at 0.7T0. Here we investigate how ET changes with the rise of temperature from 2 to 17 C, that increases T0 by 50% without affecting the fibre stiffness. At low loads LT was reduced by 30% by the rise in temperature, so that ET was only slightly increased; at high load LT remained almost constant so that the maximum value of ET was increased by 40%. In conclusion, in the range of temperature studied, the endothermic process responsible for force generation is able to increase by 40% the efficiency of energy transduction. Supported by EU and MIUR (Italy).

Functional characterisation of a mutant (Met132Val) in the Acta1 gene from a patient with nemaline myopathy JUAN-JUAN FENG, STEVEN MARSTON1, MAHMOODA MIRZA, HASSAN ABDULRAZZAK and CAROLINE SEWRY2 1 Imperial College London, National Heart and Lung Institute, SW3 6LY, UK 2 Department of Paediatrics, London W12 0NN, UK We examined actin mRNA and protein from biopsy samples and compared with a normal human muscle control. The stretch of mRNA containing the mutation site was amplified by PCR and digested with BstX1: the restriction site is absent in the mutant. An average 61% of the mRNA from the patient biopsy was not cleaved, indicating the presence of roughly equal amounts of wild-type and mutant messenger in vivo. Actin was extracted from 2.5 mg of biopsy material. In the mutant sample a proportion of the actin did not polymerise in 50 mM KCl, 2.5 mM MgCl2 but all the wild-type actin did (visualisation of rhodamine–phalloidin labelled filaments by fluorescence microscopy). We prepared a second batch of actin (from 0.5 mg of tissue) and polymerised actin with 50 mM KCl, 2.5 mM MgCl2, 150 nM rhodamine–phalloidin which produced polymerisation of all the actin. This actin produces faster sliding of thin filaments at pCa5, higher relative isometric force and lower Ca2+-sensitivity of activation. We conclude that the mutant mRNA and protein was expressed and that the mutation reduced polymerisability and, when copolymerised with wild-type actin, altered thin filament function.

On the mechanism of Ca2+-regulation of smooth muscle thin filaments SAIRA ANSARI, MOHAMMED El-MEZGUELDI and STEVEN MARSTON Imperial College London, NHLI, UK The thin filaments of vascular smooth muscle are Ca2+ regulated by tropomyosin, caldesmon (CaD) and a Ca2+-binding protein. We have proposed that the regulatory mechanism of smooth muscle thin filaments is closely analogous to Ca2+-regulation of skeletal muscle thin filaments. (1) Inhibitory concentrations of caldesmon inhibit S1ADP (strong) binding to ATm and make it cooperative. (2) If ATm is switched to the ON state using NEM S-1 (1:10 actin), then CaD binding is weakened up to 20-fold and becomes cooperative. (3) If ATm is switched to the OFF state with troponin I (1:10 actin) CaD binding to ATm does not change. We conclude that CaD acts as an allosteric inhibitor by preferentially binding to ATm in the OFF state (4) Thin filaments are activated by Ca2+calmodulin with no dissociation of caldesmon from actin. (5) Further addition of NEM S-1 activated ATm ATPase but had no effect on CaD-CaM-Ca2+ binding to ATm. (6) In contrast, an inhibitory concentration of troponin I substantially reduced the initial affinity of CaD-CaM-Ca2+ for actin and made the binding curve cooperative. We conclude that thin filaments are activated in the presence of Ca2+ because CaD-CaM-Ca2+ preferentially binds to the ON state of ATm.

An ultra short yeast tropomyosin shows novel myosin regulation R. MAYTUM1, M.A. GEEVES2 and M. KONRAD2 1 MPI for Biophysical Chemistry, Go¨ttingen, Germany 2 Department of Biosciences, University of Kent, Canterbury, UK Saccharomyces cerevisiae has two single exon genes TPM1 and TPM2 that produce the 199 and 161 amino acid tropomyosins (Tm) yTm1 and yTm2. These correspond in length to 5 and 4 actin binding sites, respectively. The two genes appear to have evolved from a common ancestor, the larger having an internal 38 amino duplication, corresponding to a single actin-binding site. We have produced an ultra-short 3 actin-spanning Tm by removing both of the repeats from yTm1. This short Tm binds actin with an affinity typical for other Tms (K50% = 0.5 lM). However, an equilibrium binding assay shows much greater inhibition of myosin binding to actin than other Tms previously studied. Further analysis by actin cosedimentation assays shows the Tm to binds to actin in direct competition with myosin, myosin binding excluding Tm from the actin filament. This is unlike any Tms previously characterised, which partially compete with myosin according to the steric blocking model, but can bind simultaneously with myosin. This new Tm provides direct evidence for the steric blocking model showing that Tm can compete directly with myosin for the same binding site. It also hints at a possible evolution of Tm function from direct competition to co-operative regulation.

New molecules involved in physiological hypertrophy in mouse heart? C. MUNDING1, F. BAUMGARTNER2, M. LEU2, J.-C. PERRIARD2, M. GAUTEL1, A. SHAH3 and E. EHLER1 1 Cardiovascular Division and the Randall Division of Cellular and Molecular Biophysics, King’s College London, London SE1 1UL, UK

493 2

Institute of Cell Biology, ETH Ho¨nggerberg, Zurich, Switzerland 3 Cardiovascular Division, King’s College London, London SE5 9BJ, UK Hypertrophic cardiomyopathy is the most common form of heart disease and will lead to heart failure. Since it is likely that similar pathways are involved in physiological and pathological hypertrophy, we searched for molecules that are upregulated during the phase of normal heart development using the representational difference analysis method. Based on the results from volume measurements of freshly isolated cardiomyocytes after birth, it appeared promising to compare the mRNA from P0 with P5, when the exponential phase of the hypertrophic growth starts. Several upregulated molecules were identified. Among the known sequences there were genes which have already been implicated in the process of pathological hypertrophy. Full length cDNAs of some of the unknown clones were generated. In particular, we analysed clone D4.1, which encodes a zinc finger protein. Transient transfections of this clone into neonatal rat cardiomyocytes show a variable localisation pattern, sometimes in the nucleus and sometimes in the cytoplasm, indicating a shuttling of this molecule. Antibodies generated against this protein suggest the existence of two different isoforms, which might be expressed in a developmentally regulated fashion. Further investigations with these antibodies as well as using inhibitors or activators of signalling cascades will show what the relation of clone D4.1 to hypertrophy might be. Supported by the BHF, the SNSF and the Fondation G. & S. Pre´vot.

Single particle analysis of muscle regulatory proteins on actin D. PAUL, J.M. SQUIRE and E.P. MORRIS Biological Structure and Function Section Biomedical Sciences Division, Imperial College London, UK The application of single particle techniques to the three-dimensional analysis of electron microscope images of the muscle thin filament is considered. The troponin complex on the thin filament plays a crucial role in regulation of muscle contraction. However, the precise location of troponin relative to actin and tropomyosin and how its structure changes with Ca2+ have not been determined. Problems occur when using the traditional form of helical reconstruction; the perturbations and asymmetric components on the filaments, such as the troponin complex, get averaged out. This has led us to develop a method of reconstructing filamentous proteins that does not impose any helical symmetry. Modifications were made to the single particle techniques of Euler angle assignment and back projection. A reference model is used in the initial angle assignment, allowing us to orient the filament and our adaptation of the exact filtered back projection method allows the correct weighting of projection images. This results in significant improvements in the reconstruction. Our initial class-averages clearly show the position of the troponin complex.

A component of stretch induced force in active muscle is not inhibited by BTS (a Myosin-II Inhibitor) G.J. PINNIGER and K.W. RANATUNGA Department of Physiology, University of Bristol, Bristol BS8 1TD, UK Stretch of an active muscle induces a transient increase in force that decays, after the stretch, to a level that exceeds the steady state

isometric force. The mechanism underlying this residual force enhancement is unknown. We examined the stretch-induced force responses when the active force was significantly reduced by using N-benzyl-p-toluene sulphonamide (BTS), a compound known to inhibit myosin-II ATPase (Cheung et al. 2002, Nat Cell Biol, 4: 83– 88). Experiments were done on bundles of 5 intact (fast) fibers (fiber length, Lo, 2 mm) isolated from the flexor hallucis brevis of the rat (at 20 C, sarcomere length 2.5 lm). Ramp stretches of 5% Lo were applied on the tetanus plateau and stimulation was maintained for a further 500 ms after the stretch where the force enhancement was measured. In the presence of 10 lM BTS, the tetanic force decreased significantly from 209 ± 23.4 to 42.7 ± 6.5 kN m)2 (80%; student’s t-test, P < 0.05, n = 5); whereas, the stretch-induced residual force was not significantly different (control = 26.6 ± 4.9, BTS = 35.8 ± 7.3 kN m)2; P > 0.05). These findings indicate that residual force enhancement is not due to actively cycling crossbridges. Supported by The Wellcome Trust.

Mechanical effects of human cardiac troponin C mutation GLY159Asp in exchanged rabbit psoas fibres LAURA PRESTON1, SIMON LIPSCOMB1, PAUL ROBINSON2, HUGH WATKINS2, CHARLES REDWOOD2, JENS MOGENSEN3 and CHRIS ASHLEY1 1 Department of Physiology, University of Oxford, UK 2 Department of Cardiovascular Medicine, University of Oxford, UK 3 St. Georges Hospital, London, UK The endogenous fast skeletal troponin in rabbit psoas fibres was partially replaced with human whole cardiac troponin (hcTn) by the method of Brenner (Brenner et al. 1999). Analysis of the psoas fibres with SDSPAGE gels showed that two-hour incubations in 2 mg/ml troponin, at 22 C, resulted in a 35% exchange. The dummy exchange process has no effect on the isometric force produced by the fibre, but exchange with hcTn wild type (WT) produces a significant difference to the pCa50 (fsTn = pCa5.7 vs. WT hcTn = pCa5.55, P < 0.001 n = 12), and the co-operativity (nH) (fsTn = 3.6 vs. WThcTn = 2.1, P < 0.005, n = 12). The novel mutation TnC Gly159Asp has been shown to cause dilated cardiomyopathy. Exchange with hcTn containing Gly159Asp had no effect on the maximal force, pCa50, or co-operativity when compared to WT hcTn. The activation kinetics of the exchanged fibres were also investigated using flash photolysis of the caged compound NP-EGTA. When compared to WT hcTn, the half-time of activation for the Gly159Asp mutation was increased by nearly 50% (37.0 ms ± 3.5, n = 6 for Gly159Asp hcTn, compared to 20.3 ms ± 3.27, n = 6 for WT hcTn). Gly159Asp also decreased the rate constant (k1) of activation by over 50% (WT hcTn k1 = 25.15 s)1, n = 6, Gly159Asp hcTn k1 = 12.4 s)1, n = 6).

Orientation and mobility of intermediates of 1.5-iaedanslabeled subfragment-1 in ghost muscle fibers DABO.E. PRONINA1, A. WRZOSEC2, R. ROWSKA2 and YU.S. BOROVIKOV1 1 Institute of Cytology RAS, Russia 2 Nencki Institute of Experimental Biology, PAS, Poland The orientation and mobility of subfragment-1 (S1) of myosin modified by 1,5-IAEDANS were studied in a ghost muscle fiber in the absence or presence of MgADP, MgAMP-PNP, MgATPcS and MgATP by polarized fluorimetry. Since the fluorescent probe was covalently bound to catalytic domain of the myosin head, the dipoles of its absorption and emission were sensitive to the changes in the

494 orientation and mobility of S1. At different intermediate states of ATPase cycle the orientation and mobility of probe oscillators were shown to change discretely indicating multistep alterations of these characteristics of myosin heads during ATP hydrolysis. Maximum orientation changes of the myosin head by 4 and its mobility by 30% were found between AM ÆADPÆPi and A`MÆADP actomyosin states. It was supposed, that the multistep change of tilt and mobility of crossbridge occurred during ATP hydrolysis, the myosin head and the actin monomers tilting as a solid body whose elasticity is determined by interaction between myosin and actin.

Effects of DCM mutations in thin filament regulatory proteins on contractile function P. ROBINSON1, M. MIRZA2, S. MARSTON2, R. WILLOTT1, C. ASHLEY1, J. MOGENSEN3, W. MCKENNA3, H. WATKINS1 and C. REDWOOD1 1 University of Oxford, UK 2 Imperial College London, UK 3 St. George’s Hospital, London, UK Inherited dilated cardiomyopathy (DCM) can be caused by mutations in the genes encoding the cardiac thin filament regulatory proteins TnT, TnC and a-TM. We have tested how these mutations affect contractile function by comparing wild type and mutant recombinant proteins in a range of in vitro assays. We have analysed five TnT, one TnC and two a-TM mutants. Thin filaments, reconstituted using actin and recombinant wild type Tn and a-TM, activated myosin ATPase activity in a Ca2+-dependent, cooperative manner. Thin filaments containing either a Tn or a-TM DCM mutant at 1:1 ratio with wildtype (the likely in vivo ratio) all gave reduced Ca2+-sensitivity of activation. Furthermore, in all but one case (a-TM mutant E54K), thin filaments reconstituted with a DCM mutant component produced significantly lower maximum Ca2+-activation of myosin ATPase rate. Analyses of mutant thin filaments using the in vitro motility assay revealed similar abnormalities in the Ca2+-regulation of filament sliding and in the maximum activation of filament velocity. Recombinant Tn was also used to displace endogenous complex in skinned guinea pig trabeculae (to c.58% of total). Incorporation of either R141W or R205L TnT resulted in decreased Ca2+-sensitivity of force generation compared with wild type. These data contrast sharply with analyses of HCM-causing mutations in thin filaments proteins (which cause increased Ca2+-sensitivity) and suggest that the DCM mutations change contractility in a fundamentally different manner. Supported by the British Heart Foundation.

Myosin V moves via a hand-over-hand lever arm mechanism T.T. SAKAMOTO, A. YILDIZ, P.R. SELVIN and J.R. SELLERS NIH, Bethesda, MD and University of Illinois, Urbana, IL, USA Myosin V is a two-headed myosin with a long neck that has six IQ motifs complexed with calmodulin. It moves processively along an actin filament. We expressed HMM mutants in which the neck was either lengthened (8IQ) or shortened (4IQ) by altering the number of IQ motifs. The working stroke of single interactions was measured using optical trapping and the speed and processivity was determined by observing the movement of single fluorescently-labeled molecules on actin filament using total internal reflection fluorescent (TIRF) microscopy. The working stroke and the speed at low KCl (25 mM) of individual myosin V IQ mutants are proportional to neck length of myosin V. To address the step size during processive runs, we

measured the step size of all IQ mutants using FIONA (Fluorescent Imaging at One Nanometer Accuracy), which can determine the position of an individual myosin V fluorescently labeled at a single calmodulin residue with nanometer (Yildiz et al. 2003 Science 300: 2061). The 4IQ mutant has a shorter step size and the 8IQ has a longer step size that obtained with wild type myosin V. These results strongly support a hand-over-hand lever arm model in which the step size is predominantly determined by the length of the IQ domain.

Orientation changes of troponin C domains on activation of skeletal muscle Y.-B. SUN1, R.E. FERGUSON2, P. MERCIER3, A.S. BRACK1, B.D. SYKES3, J.E.T. CORRIE2, D.R. TRENTHAM2 and M. IRVING1 1 Randall Division, King’s College London, UK 2 NIMR, London, UK 3 Department of Biochemistry University of Alberta, Canada A novel approach for mapping protein-domain orientations and their responses to regulatory signals in a cellular environment was used to investigate the structural basis of Ca2+ regulation of skeletal muscle contraction. Polarized fluorescence from bifunctional rhodamine probes attached along four helices (N, A, C and E) of troponin C (TnC) was measured in permeabilised muscle fibres. The in situ orientation of the N-terminal domain of TnC was determined by combining polarized fluorescence intensities from fibres and in vitro structures. The D helix of this domain is approximately perpendicular to the actin filament axis. On muscle activation, the N-lobe undergoes a major reorientation. Force and orientation changes of C and E helices of TnC have similar Ca2+-sensitivity. However, force has a steeper dependence on [Ca2+]. Complete inhibition of force with 0.1 mM Nbenzyl-p-toluenesulfonamide (BTS) has no effect on the changes in orientation of the C helix of TnC on Ca2+ activation, but that of the E helix decreases by 70%. Thus the change in orientation of the N-lobe of TnC on muscle activation depends solely on [Ca2+], whereas that of the C-lobe depends on both [Ca2+] and actively cycling cross-bridges.

Mutations in the motor domain of myosin VIIa and their implications in deafness RICHARD HUGHES1, STEVE D.M. BROWN2 and JENNIFER C. PINDER 1 Randall Division for Cell and Molecular Biophysics, New Hunt’s House, King’s College London, SE1 1UL, UK 2 MRC Mammalian Genetics Unit and UK Mouse Genome Centre, Harwell, OX11 ORD, UK Myosin VIIa is found in the stereocilia of the inner ear, where it plays a critical role in hearing and balance. The likelihood that myosin VIIa is causing the Usher’s syndrome deafness phenotype by the loss of an interaction due to mutations in its motor domain is being investigated. In the Myo7a6J and Myo7a816SB mutations, there is disorganisation of the stereocilia, reduced protein levels and little cochlea response. Interestingly the Myo7ash1 phenotype exhibits nearly normal protein levels and stereocilia development, but reduced physiological response. The Myo7ash1 mutation lies in a relatively unconserved loop, so we asked ‘why does this mutation cause deafness?’ Yeast two-hybrid cDNA libraries were screened for interacting partners using the myosin VIIa motor domain as ‘bait’ protein. Two proteins were identified as potential partners for myosin VIIa, based on positive interaction in yeast. Further experiments may

495 validate an interaction in vivo. Additionally, ‘HMM’ fragments of wild-type and mutant myosin VIIa have been co-expressed with calmodulin using the baculovirus expression system. These are being used to compare the biomechanics of the molecules, and to determine whether the mutations are preventing myosin VIIa from functioning as an actin-activated ATPase.

Impacts from Hooke beyond Hanson in the zero length theory of muscle contraction FRANS A.M. VAN KAAM Mijlberg 38, 5508EN Veldhoven, The Netherlands Any body is passive elastic i.e. has the power to restore its natural length when its external load is released (Hooke R (1678) De

Potentia Restitutiva). The Zero Length is defined as the maximum, externally un-loaded, natural length of a body. In addition, an active body can change its zero length from within itself if the appropriate energy is made available.(v Kaam (1998) JMRCM ,jan). An active body which is relaxed, has finished its energy transfer or is not capable to do so anymore behaves passive elastic. In the unit sarcomere filaments can slide (Huxley and Hanson (1954), Nature; Huxley and Niedergerke (1954), Nature) during any contraction dependent upon both the elasticity of cross bridges and filaments as well as the external load. Theoretical mechanical analysis of an elastic unit sarcomere structure reveal a non-linear T1 curve which stems from the strain difference between 2 heads in the same or from different cross bridges. Concomitant contractile relaxation takes place with efficient use of the energy released as defined by Hooke De Potentia Restitutiva (1678) during a double headed contraction of a cross bridge and during contraction within an elastic half sarcomere.