Internal Motions in Myosin Head: Effect of ADP and ATP

JOBNAME: BBRC 219#3 PAGE: 1 SESS: 13 OUTPUT: Wed Apr 10 11:40:25 1996/xypage/worksmart/tsp000/68953f/27

Internal Motions in Myosin Head: Effect of ADP and ATP

Joseph Belagyi1 and Dénes Lo˝rinczy

Central Research Laboratory and Department of Biophysics, University Medical School, Pécs, Hungary

Received January 18, 1996

Internal flexibility of myosin heads in glycerinated muscle fibres in the presence of MgADP plus orthovana-date and after addition of Ca-ATP was studied using an isothiocyanate-based spin label attached to the reactivesulfhydryl sites of myosin. The spin labels were immobilized on the microsecond time scale and exhibitedsignificant orientational order in rigor. In AM+.ADP.Vi state a smaller fraction of ordered population was foundshowing distinct orientation from rigor; the larger population of heads was in dinamically disordered state. Thisnew ordered population of heads was detected even in contracting fibres.© 1996 Academic Press, Inc.

The sliding motion in striated muscle requires the cyclic interaction of myosin with ATP andactin, and for actomyosin ATPase the presently accepted mechanism suggests at least six inter-mediates (1,2). The study of molecular events in different states of muscle and during contractioncan be achieved by using motion sensitive probe molecules (3,4,5). In earlier investigations theefforts were concentrated to detect rotational motions of the whole myosin heads in the presenceof ATP and analogues of ATP (6,7,8). Using spin label EPR spectroscopy, maleimide spin labelseems to be one of the best choices because it attaches rigidly to the reactive sulfhydryl sites ofmyosin as revealed by ST EPR technique, and reflects the global motion of heads in the submil-lisecond time range (9). The conventional EPR spectrum reports a narrow distribution of labels inrigor that represents the end of the powerstroke, and shows large change of ordering in the presenceof ATP evidencing the detachment of myosin heads from actin in relaxed state.Recently, the modification of the atomic structure of myosin was suggested (10,11). The new

model involves structural rearrangement of myosin head, and consequently internal motions andflexibility of the major segments of the motor domain may be an integral part of the contractileprocess. Using an isothiocyanate-based spin label (TCSL) or ana-iodoketo spin label, it waspossible to detect slow segmental or domain motions in myosin heads even in glycerinated musclefibres in rigor and in MgADP state (12,13). This finding offers a possibility to characterize variouscross-bridge states and transitions between cross-bridge states.The aim of the present paper was to investigate the conformations and distributions of cross-

bridges in the presence of nucleotides in glycerinated muscle fibres with an isothiocyanate-basedspin label, and to detect internal motions correlating actomyosin ATPase. Our results show theTCSL probe molecules which exhibited in rigor two narrow distributions with respect to the longeraxis of the fibre, reflected large changes in the fractions of the populations in the presence ofCa2+-ATP or ADP plus orthovanadate. Addition of Ca2+-ATP to the fibres resulted in a complexspectra: three spectral components could be identified by spectrum manipulation.

MATERIALS AND METHODS

Preparation and spin-labelling of muscle fibre.Glycerol-extracted muscle fibre bundles were prepared from rabbit psoasmuscle as reported earlier (12). Spin-labelling of the fibres was performed in rigor buffer (100 mM KCl, 5 mM MgCl2, 1mM EGTA, pH 7.0) plus 2 mM pyrophosphate with about one mole of TCSL to one mole of myosin for 1–3 hours at 0°C.After spin-labelling the fibre bundles were washed in great amount of rigor buffer plus 25 mM K3Fe(CN)6 for 16 hours to

1 Corresponding author. Fax: +(36 72) 315 864.Abbreviations: ST EPR, saturation transfer EPR; MSL, maleimide spin label; TCSL, isothiocyanate spin label; Ap5A,

P1,P 5-di(adenosine-59) pentaphosphate; AM+.ADP.Vi, trapped intermediate state; vanadate ion, Vi.

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS219,936–940 (1996)ARTICLE NO. 0336

9360006-291X/96 $18.00Copyright © 1996 by Academic Press, Inc.All rights of reproduction in any form reserved.


remove the unreacted labels and to reduce labels bound to weakly immobilizing sites (14). The sarcomere length of thefibres was measured as reported earlier (12). The number of spin labels bound to myosin was determined from the EPRspectra of muscle fibres by comparing the double integrals of the spectra with known concentration of MSL in aqueoussolution in the same sample cell.ATPase activity.The K+-EDTA ATPase and Ca2+-ATPase activity of myosin was measured by determining the rate of

release of inorganic phosphate as described earlier (15).EPRmeasurements.Conventional and ST EPR spectra were taken with ESP 300 E (Bruker, Germany) spectrometer. First

harmonic, in-phase, absorption spectra were obtained using 20 mW microwave power and 100 kHz field modulation withamplitude of 0.1–0.2 mT. Second harmonic, 90° out-of-phase, absorption spectra were recorded with 63 mW and 50 kHzfield modulation of 0.5 mT amplitude detecting the signals at 100 kHz out-of-phase. The 63 mW microwave powercorresponds to an average microwave field amplitude of 0.025 mT in the center region of the standard tissue cell of Zeiss(Carl Zeiss, Germany) (16). In this region of the tissue cell, small segments of the muscle fibres (6–7 mm long) weremounted parallel to each other. The spectra were recorded in two positions at temperature of (22 ± 1) °C, where the longeraxis of the fibres was oriented parallel and perpendicular to the laboratory field. In order to get reasonable rotationalcorrelation time of the attached labels in the ST EPR time domain, the muscle fibres were homogenized before measure-ments to avoid orientational effect in the spectra.Analysis of spectra.The derivatives of the EPR absorption spectra were digitized into 1024 data points at equal intervals

over 10 mT sweep width of the magnetic field, and the spectra were normalized to the same number of unpaired electronscalculating the double integral of the derived spectra. In order to have separately the populations of spin labels fromTCSL-fibres in different states, the manipulations were performed on spectra by digital subtraction. By subsequent sub-tractions it was possible to obtain the component spectra in rigor, in AM.ADP and in AM+.ADP.Vi state.

RESULTS AND DISCUSSION

Characterization of labelled muscle fibres.The fibre bundles could produce tension in activatingsolution (rigor buffer plus 0.1 mM CaCl2 and 5 mM ATP) after spin labelling. The maximumtension of the labelled fibres was in an average about 10–15% smaller than the maximum tensionproduced by the untreated fibres. This agrees with the earlier observations (17). From the extent ofthe reduction of the K+-EDTA ATPase activity and from the number of spin labels bound to myosinwe estimated that about 90% of spin labels was located on the reactive sulfhydryl sites.Orientational order of probe molecules in rigor.In rigor, the EPR spectra of labelled fibres

showed strong dependence on orientation with respect to the fibre axis indicating that the z-axis ofthe magnetic tensor was oriented nearly perpendicular to the longer axis of the muscle fibres.Rigorous analysis of spectra gave evidence that the labels had two distinct orientations, the meanangles between the principal axis (z) of the spin label and the fibre axis were 75° (R1) and 56° (R2),the fractions were R1 4 0.76 and R2 4 0.24, respectively (12).Orientational order of spin labels in presence of MgADP and Vi.In solution containing 100 mM

KCl, 5 mM MgCl2, 5 mM ADP, 1 mM EGTA, 100mM Ap5A, in the absence or presence of 2.0mM NaV3O5, in 10 mM histidine. HCl buffer, pH 7.0, large changes were detected in the con-ventional EPR spectra of TCSL-fibres (Fig. 1). Moreover, in the presence of MgADP and ortho-vanadate the orientation dependence measured in parallel and perpendicular orientations wassignificantly reduced indicating that the binding of substrates to the nucleotide-binding domaininfluenced significantly the orientation of the segment that held the probe molecule. Recently, itwas shown that in MgADP state only little changes were detected in the azimuthal and torsionalangles of the cross-bridges using multiple probes attached to myosin (8). This statement is inaccordance with the low-angle x-ray diffraction studies and fluorescence-quenching experimentsthat the angle of attachment for the myosin head in the ternary actin-myosin-ADP complex wasabout the same as in rigor muscle (18,19). ST EPR measurements indicated no increased rotationalmotion in MgADP state in comparison to rigor. This supports the view the TCSL reports segmentalrearrangement without significant head reorientation.Orthovanadate ion binds stoichiometrically with ADP to the nucleotide- binding site of myosin,

and according to the kinetics of actomyosin ATPase the initial complex undergoes to a stablecomplex which is believed to be an analogue of the AM.ADP.Pi state (20,21). The addition oforthovanadate ions with ADP led to a drop in tension, therefore it was proposed that the myosin

Vol. 219, No. 3, 1996 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

937


FIG. 1. Conventional EPR spectra of glycerol-extracted muscle fibres labelled with isothiocyanate spin label. Fibre axiswas oriented parallel to the laboratory magnetic field. The fibres were incubated in rigor buffer plus 4 mM MgADP, 100mM Ap5A (spectrum B) and plus 2 mM Na3VO5 (spectrum A) for 15 minutes at 0 °C before spectra were taken. Bottom:difference spectrum (spectrum C). Scan width is 100 G.

FIG. 2. Conventional EPR spectra of glycerinated muscle fibres labelled with TCSL. Fibre axis was oriented parallel tothe laboratory magnetic field. The fibres were incubated in rigor buffer, 5 mM ATP and 0.1 mM Ca2+ without EGTA, plus20 mM creatine phosphate and 200 units/ml creatine phosphokinase as ATP regenerating system for 30 minutes at 20 °Cduring EPR measurement.


938


heads with MgADP and Vi are in weakly binding state. The lineshape of the EPR spectrum (upperspectrum in Fig. 1) suggests that the spectrum is superposition of two spectra; spectrum manipu-lation confirmed that the first state corresponds to the strongly binding state (AM.ADP state) withwell-defined axial orientation, distinct from rigor (middle spectrum in Fig. 1), whereas the secondstate can be characterized by a population of disordered probe molecules. The probability for thelabels being in the first rotational state is about 0.34, and about 0.66 in second state. InAM+.ADP.Vi state, the second component might correspond to either the disorder of a fraction ofheads or the internal motion of a larger segment in the cross-bridge. The experiments support thesecond possibility, the medium MgADP and Vi induce intrinsic changes in the multisubunitstructure of the subfragment-1, but these changes do not lead to the reorientation of the entire headas shown by measurements on MSL-fibres (figure is not shown). The apparent rotational correla-tion time of the TCSL measured by ST EPR decreased from 70ms to 20ms evidencing largermotion at the reactive sulfhydryl sites.Orientational order in the presence of ATP and Ca2`. In activating solution (rigor buffer without

EGTA plus 5 mM MgATP and 0.1 mM Ca2+; 20 mM creatine phosphate and 200 units/ml creatinephosphokinase as ATP regenerating system), the spectral changes at H || k orientation were morepronounced (Fig. 2). In these samples the spectrum changed rapidly during the first five minutesbecause of the low level of ATP in the sample cell, it has a capacity of 120ml, therefore only thefirst 50 or 100 scans (scan time: 2 seconds) were accumulated to obtain relevant spectrum onworking muscle. For accurate data analysis in the presence of ATP, more precise experimentalconditions are required that ensure the constant level of ATP during the EPR acquisition time (22).

FIG. 3. Decomposition of a conventional EPR spectrum obtained on spin-labelled glycerol-extracted muscle fibres in thepresence of Ca2+-ATP. Symbols: ATP: isometric contraction; ATP + Vi: fibres in buffer containing ATP and orthovanadate;R2: composite spectrum of ADP state; R1: residual spectrum. R1 spectrum was obtained after successive subtraction ofspectrum ATP + Vi and spectrum R2 from spectrum ATP. R1 spectrum is composite spectrum of the rigor state. Scan widthis 100 G.


939


Carefull manipulation of spectra showed that during isometric contraction of the fibres at leastthree components could be derived. Using orthovanadate to stabilize the AM.ADP.Pi state, wefound that 38% of the total EPR absorption arised from labels with isotropic distribution(AM+.ADP.Vi state), and 62% of labels could be assigned to a mixed population with high ordering(AM and AM.ADP states). Subtracting one of the components of rigor spectrum (R2) from thedifference spectrum (Ca2+-ATP spectrum-(orthovanadate plus Ca2+-ATP spectrum)), the residualcomponent approximated quite well the R1 component of the rigor state (12). Therefore, from theanalysis of spectra we can conclude that at least three main spectral components (AM, AM.ADPand very likely AM.ADP.Pi) characterize the state of muscle during isometric contraction (Fig. 3).Recently a revised model for the molecular basis of contraction was proposed (11). This model

suggests a larger role of the cleft that splits the 50 kDa domain of the myosin head. The keyassumption is the nucleotide-dependent movement of the lower domain of the 50 kDa segment.Three-dimensional structural analysis reveals that the cleft separating the central 50 kDa segmentopens when ATP binds to myosin. The binding of ATP reduces the affinity of myosin to actin, andthis allows the rotational motion of the myosin head. When myosin rebinds to actin, the stronginteraction generates translational and rotational motion of the lower domain of the 50 kDa segmentthat induces the release of the phosphate and closes the cleft. The effect is transmitted to the COOHterminus via the 20 kD region. The conformational change might be associated with a change inthe orientational distribution of the probe molecules and could result in increased rotational free-dom. The EPR results seem to support the suggestion about the communication between nucleo-tide-binding domain and COOH-terminus via the 20 kDa segment containing the reactive sulfhy-dryl sites.

ACKNOWLEDGMENTSThis work was supported by research grants from the National Research Foundation (OTKA T 017099) and Ministry of

Welfare (ETT 737/1993). The Bruker ESP 300 E spectrometer used in the experiments was purchased with funds providedby the National Research Foundation Grant C-123.

REFERENCES

1. Huxley, H. E. (1969)Science202,1356–1366.2. Brenner, B. (1990)inMolecular Mechanisms in Muscle Contraction (Squire, J. M., Ed.), pp. 77–149, Macmillan Press,

London.3. Belagyi, J., Schwarz, D., and Damerau, W. (1979)Studia Biophys.77, 77–83.4. Barnett, V. A., and Thomas, D. D. (1984)J. Mol. Biol. 179,83–102.5. Burghardt, T. P., and Ajtai, K. (1990)in Molecular Mechanism in Muscular Contraction (Squire, J. M., Ed.), pp.

211–239, Macmillan Press, London.6. Fajer, P. G., Fajer, E. A., Matta, J. M., and Thomas, D. D. (1990)Biochemistry29, 5865–5871.7. Fajer, P. G., Fajer, E. A., Brunsvold, N. J., and Thomas, D. D. (1988)Biophys. J.53, 513–524.8. Ajtai, K., Toft, D. J., and Burghardt, T. P. (1994)Biochemistry33, 5382–5391.9. Barnett, V. A., and Thomas, D. D. (1989)Biophys. J.56, 517–523.10. Rayment, I., Rypniewski, W. R., Schmidt-Bäse, K., Smith, R., Tomchick, D. R., Benning, M. M., Winkelmann, D. A.,

Wesenberg, G., and Holden, H. M. (1993)Science261,50–58.11. Fisher, A. J., Smith, C. A., Thoden, J., Smith, R., Sutoh, K., Holden, H. M., and Rayment, I. (1995)Biophys. J.68,

19s–28s.12. Belagyi, J., Frey, I., and Pótó, L. (1994)Eur. J. Biochem.224,215–222.13. Raucher, D., Sár, C. P., Hideg, K., and Fajer, P. G. (1994)Biochemistry33, 14317–14323.14. Graceffa, P., and Seidel, J. C. (1980)Biochemistry19, 33–39.15. Lanzetta, P. A., Alvarez, L. J., Reinach, P. S., and Candia, O. A. (1979)Anal. Biochem.100,95–97.16. Fajer, P., and Marsch, D. (1982)J. Mag. Res.49, 212–224.17. Crowder, M. S., and Cooke, R. (1984)J. Mus. Res. Cell Motil.5, 131–146.18. Rodger, C. D., and Treager, R. T. (1974)J. Mol. Biol. 86, 495–497.19. Franks-Skiba, K., Hwang, T., and Cook, R. (1994)Biochemistry33, 12720–112728.20. Goodno, C. C. (1982)Meth. in Enzymol.85, 116–123.21. Wells, Ch., and Bagshaw, C. R. (1984)J. Mus. Res. Cell Motil.5, 97–112.22. Fajer, P. G., Fajer, E. A., and Thomas, D. D. (1990)Proc. Natl. Acad. Sci. USA87, 5538–5542.


940

Documents

Internal Motions in Myosin Head: Effect of ADP and ATP