THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1985 by The American Society of Biological Chemists, Inc.

Vol. 260, No. 28, Issue of December 5, pp. 15156-15162,1985 Prhted in U. S. A.

Kinetic Mechanism of 1-N6-Etheno-2-aza-ATP and l-N'-Etheno- 2-aza-ADP Binding to Bovine Ventricular Actomyosin-S1 and Myofibrils*

(Received for publication, Aprii 1, 1985)

Susan J. Smith$ and Howard D. White From the Department of Biochemistry, University of Arizona, Tucson, Arizona 85721

The fluorescence emission of 1-N6-etheno-2-aza- ATP (e-aza-ATP) at 410-460 nm is enhanced approx- imately 8-fold upon mixing substoichiometric concen- trations of e-aza-ATP with bovine cardiac actomyosin- S1 or myofibrils. The time course of nucleotide fluo- rescence measured in a front face stopped flow cell upon mixing t-aza-ATP with bovine cardiac myofibrils ([Ca"'] < 10" M) is essentially the same as that with bovine cardiac actomyosin subfragment- 1. In single turnover experiments, the fluorescence rapidly rises to a maximum value, then decreases with a rate con- stant of 0.04 s" at 0 "C to a final value that is approx- imately twice the level of the unbound nucleotide. At concentrations of e-aza-ATP > 40 BM the kinetics of t-

aza-ATP binding is clearly biphasic for both acto- myosin-Sl and myofibrils. At 0 "C, the rate of the more rapid phase is proportional to nucleotide concentration and has a second order rate constant of 1.7 x 10' M" s-'; the rate of the slower phase extrapolates to a maximum of 4-5 s-' at high nucleotide concentration. The rate constants for dissociation of t-aza-ADP from bovine cardiac actomyosin-S 1 and myofibrils were measured from the decrease in e-aza-ADP fluorescence enhancement observed upon displacement by ATP to be 20 and 18 s-', respectively, at 0 "C. These results indicate that most of the cross-bridges in cardiac my- ofibrils are bound to actin and that the geometric con- straints imposed upon the interaction of actin and myosin by the three-dimensional structure of the my- ofibril do not modify the kinetics of 6-aza-ATP binding or e-aza-ADP dissociation.

Biochemical studies of the actomyosin ATPase cycle have provided a description of how the chemical reaction of ATP hydrolysis is coupled to the interaction of actin and myosin in solution (1-3). The rationale for such biochemical studies is that the chemical intermediates of the actomyosin ATPase cycle in solution are related to the states of the cross-bridges in the contractile cycle in muscle. However, in muscle, the

* This work was supported by research grants from the Arizona Heart Association, to S. S., and the Muscular Dystrophy Association of America and HL 20984 from the National Institutes of Health, to H. W. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

4 Performed during the tenure of a British-American Research Fellowship of the American Heart Association and the British Heart Foundation.

interaction of actin and myosin is subject to steric and me- chanical constraints because of the geometry and filamentous nature of the contractile apparatus (4). Myofibrils are a useful intermediate between soluble actomyosin-S1 and organized muscle fibers; they retain the filament lattice structure of muscle but can be investigated by modification of the methods of solution chemistry. The kinetics of ATP hydrolysis by skeletal myofibrils have been studied by monitoring the hy- drolysis of [y3*P]ATP during single turnovers (5); the data allow estimation of the equilibrium constant of the hydrolytic step. Another approach has been to observe the time course of the enhancement of the intrinsic tryptophan fluorescence of the myosin head, during hydrolysis of ATP by myosin-S1.l The tryptophan fluorescence is enhanced -20% providing a convenient optical method to study the reaction. However, the enhancement of myofibrillar tryptophan fluorescence is only 4-6% during steady state ATP hydrolysis (6). Although this work suggests that the overall mechanism of ATP hy- drolysis by myofibrils is similar to that of actomyosin, the small fluorescence signal makes detailed kinetic measure- ments of the hydrolytic pathway extremely difficult. Here we have used the fluorescent analogues of ATP and ADP, t-aza- ATP and e-aza-ADP for which the respective enhancement of nucleotide fluorescence can be as large as 800 and 300% when bound to myosin.

The fundamental features of the hydrolytic pathway are unchanged, although the rate constants for the mechanism of eaza-ATP hydrolysis by bovine cardiac actomyosin-S1 differ by up to 10-fold from the values measured for ATP (7). Moreover, t-aza-ATP can replace ATP and support the con- traction of muscle fibers; e-aza-ATP produces 75% of the isometric tension obtained with ATP and gives a similar maximum velocity of shortening as that of ATP in glyceri- nated rabbit psoas muscle fibers (8).

On the basis of these observations, t-aza-ATP seemed an ideal substrate analogue with which to study the mechanism of the-hydrolytic pathway of nucleotide triphosphate hydrol- ysis by myofibrils. The basic mechanochemical mechanism of coupling the hydrolysis of e-aza-ATP to work production in muscle is likely to be the same as that for ATP. Therefore, understanding the mechanism by which the free energy of 6 -

aza-ATP hydrolysis is coupled to work production in muscle will tell us how ATP hydrolysis is coupled to work production in muscle. The following simplified version of the nucleotide

' The abbreviations used are: S1, subfragment 1 of myosin; MOPS, 4-morpholinepropanesulfonic acid; e-aza-ATP, 1-N6-etheno-2-aza- adenosine-5' triphosphate; 6-aza-ADP, 1-N6-etheno-2-aza-adeno- sine-5' diphosphate; A,A, P',P5-di(adenosine 5')-pentaphosphate; EGTA, ethylene glycol bis(6-aminoethyl ether)-N,N,N',N'-tetraac- etic acid.

15156

Etheno-Aza Nucleotide Binding to Myofibrils and Actomyosin-S1 15157

hydrolysis mechanism will be used as a working model for transient kinetic studies of myofibril^.'.^ AM + AM-T + [ AM-T - p;-p) l ;A"D=AM

(1)

(M-D-P)

The states [AM-T and M-TI or [AM-D-P and M-D-PI are considered as single intermediates, (A)-M-T or (A)-M-D-P, because these states are in rapid equilibrium at least in solution (9, 10) and the extent of the association between actin and myosin is not readily determined in myofibrils. The experiments described here compare the kinetics of c-aza- ATP binding to, and e-aza-ADP dissociation from, bovine ventricular actomyosin-S1 and myofibrils.

MATERIALS AND METHODS

Protein and Myofibril Preparation-Cardiac actin, myosin, and myosin-S1 were purified from the left ventricles of bovine hearts as described by Siemankowski and White (11). Crude cardiac myofibrils were obtained using the same procedure aa that for the preparation of cardiac myosin. After 3 washes with 1% Triton X-100, the fluffy white superficial layer of the pellet was washed a further 3 times and stored in 100 mM KCl, 10 mM MOPS, 5 mM MgC12, 0.02% sodium azide, 0.2 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluoride, pH 7.0. Myofibrils were used within a few days or were stored at -20 "C in 50% glycerol. Examination of the preparation under a phase-contrast microscope showed single myofibrils that had fairly uniform dimensions of 1-2 p~ in diameter and 10-40 p~ in length. The concentration of bovine ventricular myosin and myosin-S1 were determined from the absorption at 280 nm using extinction coeffi- cients (0.1%, w/v, 1 cm) of 0.55 and 0.64 (12). The concentration of bovine ventricular actin was determined using an extinction coeffi- cient at 280 nm of 1.15 (0.1%, w/v, 1 cm). The concentration of myofibrils was estimated from E;% of 7.0 measured after dissolving the myofibrils in sodium dodecyl sulfate (13) and removing a small amount of insoluble material by centrifugation. The concentration of myosin heads in myofibrils was calculated assuming that 40% of the

is 480,000. myofibrillar protein is myosin4 and the molecular weight of myosin

The following nomenclature is used to identify rate and equilib- rium constants. Positive subscripts identify rate and equilibrium constants of association; negative subscripts identify rate equilibrium constants of dissociation. Single subscripts, A = actin, D = ADP, T = ATP, P = Pi, a-ADP = e-aza-ADP, a-ATP = e-aza-ATP, refer to the binding of the respective ligand to myosin-SI, which is denoted as M. The subscript H refers to the hydrolysis of ATP to ADP and Pi. For multiple subscripts, the final letter of the string identifies the ligand associating (or dissociating) to myosin-S1 and all other letters refer to ligands already bound. For example, KAT refers to the binding of ATP to actomyosin-S1.

Experimentally measured rate constants of kinetic steps associ- ated with myosin intermediates that are in rapid equilibrium with actin may represent combinations of several rate constants. For example, the koa. of the first fluorescence transition, which occurs after e-aza-ATP binds to actomyosin-S1

k 'AT AM-To - AM-T1*

k'-m

KOTA 11 K'TA 11 k 'T

M-To kl_T is the weighted average of the rates of the processes that occur for M-T and AM-T.

R. F. Siemankowski, unpublished data.

Etheno-ma Nucleotides-e-Aza-ATP and c-aza-ADP were synthe- sized from the respective 1-N6-etheno-nucleotides (14, 15) and puri- fied as described in the accompanying paper (7). A molar extinction coefficient of 1530 M-' cm" at 354 nm was used to calculate concen- trations of e-aza-nucleotides (16).

Fluorescence Measurements-Fluorescence measurements were made using a Spex Fluorolog spectrofluorimeter equipped with dual monochromators in both excitation and emission optics. Actomyosin- S1 or myofibril suspensions were stirred constantly by a magnetic star within the fluorescence cell. Additions of reagents were made using spring loaded Hamilton syringe, CR700-200; reactions with rate constants less than 1 s-' could be measured by this procedure (6). Observations were made using front face fluorescence; this minimizes secondary effects of light scattering changes upon the observed fluo- rescence enhancement (the inner filter effect) (11). The selectivity for fluorescence uersus light scattering under these conditions is 10'-

Stopped Flow Fluorescence Measurements-Some of the transient kinetic measurements made on actomyosin-S1 were done in a stopped flow fluorometer with a 2-cm path length cell in which fluorescence is measured perpendicular to the excitation path. Excitation wave- lengths, 365 nm for nucleotide fluorescence measurements, were obtained using interference filters (Oriel Corp.) to select appropriate lines from the spectrum of a mercury arc lamp (Osram 100/W2). A 450-nm interference filter with a 40-nm band width was used to reduce the fluorescence emission of the unbound nucleotide.

Measurements of the kinetics of e-aza-nucleotide binding to my- ofibrils were made using a front faced stopped flow fluorometer constructed in this laboratory. A block diagram of the light path of the front face stop-flow cell is shown in Fig. 1. The 365-nm excitation light was brought to the cell with a 3-mm diameter liquid light guide (Oriel Corp). A 475-nm interference filter with a 15-nm band width was used as a dichroic mirror at an angle of 45" to reflect the 365-nm excitation light onto the cell surface. The band pass of the filter at this angle is 440-460 nm. The dichroic mirror limits the emission band to 440-460 nm and passes any residual excitation light in the 440-460-nm region away from the cell. A sharp cut off filter (Corning KV418) was used to provide additional blocking of the excitation light. This optical geometry provides the high selection of fluorescence relative to scattered light of a double monochromator system, -lo6, but is simpler and has better optical efficiency. Linear stepping motors (Airpax L92421-P2) were used to drive the syringes (5-ml Hamilton, "gas tight") of the front faced stopped flow at a maximum

lo7.

bandparr f i Iter

365 nm intarfaronce filter I

\\

"" >400 nm - 365nm FIG. 1. Block diagram of the optical pathways of excitation

and emission of the front face stop-flow fluorimeter. Indicated wavelengths are those used to monitor e-aza-nucleotide binding as described under "Materials and Methods."

15158 Etheno-Aza Nucleotide Binding to Myofibrils and Actomyosin-S1

flow rate of approximately 12 ml/s. The drive was stopped in -4 ms by holding the windings of the stepping motor in one position. The solutions were mixed with an 8-jet Berger ball mixer with a 3-mm orifice (Research Instrumentation). Temperature was regulated by running coolant through the syringe block and cell housing. The dead time of the stopped flow cell can be estimated from the cell volume (dimensions 7 X 15 X 1 mm) and the flow rate of 12 ml/s to be -10 ms; this allows the measurement of rate constants as fast as 20 s-l with less than 10% loss of amplitude. The time course of the fluores- cence enhancement observed for e-aza-ATP binding to bovine cardiac actomyosin-S1 was the same as that observed in a conventional stop flow with a dead time of 3 ms in which fluorescence emission is measured perpendicular to excitation (7).

Analysis of the data were carried out as described in the accom- panying paper (7).

RESULTS

Time Course of t-Aza-ATP Fluorescence during Hydrolysis by Cardiac Actomyosin-S1 and Myofibrils-The time courses of enhancement of nucleotide fluorescence emission observed upon mixing substoichiometric amounts of e-aza-ATP with bovine cardiac actomyosin-S1 and myofibrils are shown in Fig. 2, A and B, respectively. The time course of enhanced nucleotide fluorescence observed upon addition of e-aza-ATP to myofibrils (Ca2+ < M), (Fig. 2B) is virtually identical

I I E-azo-ADP + A-SI

""- - -""""" - """" e-azo-ATP

B

I I e-aza-ADP t MF

""_ - """"""" _""" E-azo-ATP

SECONDS 3

FIG. 2. Time course of the fluorescence emission during single turnover hydrolysis of e-aza-ATP by cardiac acto- myosin-S1 and myofibrils. Actomyosin-S1 (A) or myofibrils (B) containing -20 p~ myosin heads were mixed with 10 p~ e-aza-ATP in the front face stopped flow cell described in the legend to Fig. 1 and under "Materials and Methods." The solid lines drawn through the experimental data are the best fits to the equation I ( t ) = Ile-hl ' + Z z e - k d + C . The fit rate constants are &bl = 1.0 s-l, kobs2 = 0.04 s-', and corresponding amplitude coefficients are 1, = 1.0, Iz = -0.7. The dotted lines are the fluorescence intensity measured upon adding e-aza-ADP to actomyosin-S1 or myofibrils. The dashed lines are the fluorescence intensity expected from noninteracting solutions of (A) actomyosin-S1 and 10 p~ t-aza-ATP or ( B ) myofibrils and 10 p~ e- aza-ATP. The same fluorescence intensities were obtained with ac- tomyosin-S1 ( A ) or myofibrils ( B ) to which 1 mM ATP was added before mixing with e-aza-ATP. Experimental conditions: 100 mM KC1, 10 mM MOPS, 5 mM MgC12, 10 p M A,,A, 0.1 mM EGTA, 0.1 mM dithiothreitol, pH 7,O "C. Excitation was at 365 nm and emission at 440-460 nm.

to that observed for actomyosin-Sl as is shown in Fig. 2A. The emission intensity rapidly increases to a maximum and then slowly decays to a final steady value. The nucleotide fluorescence remains enhanced by approximately 30% of max- imum even after hydrolysis is complete. The fluorescence intensity is the same as that obtained by mixing t-aza-ADP with actomyosin-S1 or myofibrils (shown by the dotted lines in Fig. 2, A and B). If the actomyosin-S1 or myofibrils are mixed with 1 mM ATP prior to mixing with 6-aza-ATP, the nucleotide fluorescence is the same as that measured for the same concentration of E-aza-nucleotide in the absence of actomyosin-S1 or myofibrils (shown by the dashed lines in Fig. 2, A and B). Greater than stoichiometric amounts of e - aza-ATP produce a steady state enhancement of fluorescence with a duration proportional to the amount of nucleotide added (data not shown). The kcat for steady state hydrolysis of t-aza-ATP, determined from the duration of the fluores- cence enhancement or from the rate of the fluorescence decay in a single turnover (Fig. 2 A ) is 0.04 s-l at 0 "C. These results are consistent with the maximum enhancement of the nucleo- tide fluorescence on addition to actomyosin-S1 and myofibrils being due to a mixture of the steady state actomyosin-eaza- nucleotide intermediates, (A)-M-a-ATP and/or (A)-"a- ADP-P, with the intermediate level of enhancement corre- sponding to (A)-M-a-ADP.

Spectrum of the Steady State Intermediate of Myofibrillar e-Aza-ATP Hydrolysis-The fluorescence emission spectrum of the steady state complex formed during the hydrolysis of e-aza-ATP by myofibrils can be obtained from the maximum fluorescence at each wavelength observed upon addition of substoichiometric amounts of t-aza-ATP as was shown for 440-460 nm emission in Fig. 2B. Such a point by point spectrum for bovine cardiac myofibrils is shown in Fig. 3. During steady state hydrolysis by stirred suspensions of car- diac myofibrils (Ca2+ < M), the fluorescence emission of t-aza-ATP at the maximum is increased -2-3-fold and the maximum is shifted from 485 to 460 nm. The fluorescence emission spectrum of the myofibril-t-aza-ATP complex is very

W 0 z W 0 v) W

0 K 3 -I LL

4

7

t fi \ v \

00 b 440 480 520 V 560 600

WAVELENGTH (nm)

FIG. 3. Fluorescence emission spectra of t-aza-ATP during hydrolysis by cardiac myofibrils. The time course of fluorescence enhancement at each wavelength was monitored on addition of sub- stoichiometric amounts of t-aza-ATP to cardiac myofibrils at 0 "C. Experimental conditions were similar to those used in Fig. 2 except that 20 ~1 of 0.9 mM t-aza-ATP was added to a stirred suspension of 2.0 ml of 10 mg/ml cardiac myofibrils (16 p M myosin heads) in a Spex fluorolog spectrofluorometer in which the emission wavelength could be varied as indicated. The maximum emission at each wave- length minus the value obtained with a blank prior to the addition of t-aza-ATP is plotted to give the point by point spectrum (dashed line). The emission spectrum of 8 WM e-aza-ATP in the absence of myofibrils is shown for comparison (solid line). The fluorescence enhancement a t 440 nm was measured in triplicate and the mean f S.E. is shown.

Etheno-Aza Nucleotide Binding to Myofibrils and Actomyosin-S1 15159

similar to that obtained with myosin-S1 (7). The increase in fluorescence is greatest in the 410-460-nm region of the emission spectrum. In this region the maximum increase in fluorescence is 5-%fold for t-aza-ATP binding and approxi- mately 2-%fold for 6-aza-ADP binding. The large fluorescence enhancement provides a high selectivity for bound compared to free nucleotide; a 25% fluorescence enhancement at 450 nm is observable with a 20-fold molar excess of t-aza-ATP to myosin binding sites.

Kinetics of the Fluorescence Enhancement Observed upon Mixing t-Aza-ATP Binding with Cardiac Actomyosin-S1 and Myofibrils-The rapid increase in 6-aza-nucleotide fluores- cence observed upon mixing eaza-ATP with cardiac acto- myosin-S1 or myofibrils at [Ca2+] < M in a front face stopped flow is shown in Fig. 4. At concentrations of t-aza- ATP < 20 the data for both actomyosin-S1 and myofibrils can be approximately fitted to single exponential indicated by the solid line through the data (Fig. 4, A and B). At concentrations of 6-aza-ATP > 20 pM the fluorescence en- hancement is biphasic for both actomyosin-S1 and myofibrils (Fig. 4, C and D). A double exponential equation is required to fit the data well, indicating that two processes with different time courses are occurring. In the example shown in Fig. 4C, where the final concentrations of actomyosin-S1 and t-aza- ATP were 10 and 85 p ~ , respectively, the fast reaction con- tributed -47% of the total fluorescence amplitude. When myofibrils were used, the observed rate constants were very similar to those obtained with actomyosin-S1 for the same concentration of nucleotide and myosin heads; in this case (Fig. 40) the fast reaction contributed -39% of the total fluorescence amplitude. The dependence of the fit rate con- stants upon t-aza-ATP concentration is shown in Fig. 5. For

I

both actomyosin-S1 and myofibrils, the rate of the more rapid phase of fluorescence enhancement is proportional to nucleo- tide concentration with a second order rate constant of 1.7 X lo5 M-' s?. The first order rate constant of the slower com- ponent of the reaction shows a much smaller dependence upon nucleotide concentration with a rate of 2.7 s-l at the maximum concentration of nucleotide used, 85 pM. The line drawn through the slow phase of the data is for an apparent binding constant of 25 p~ and a maximum rate of 4.0 s-l. Data obtained with actomyosin-S1 under identical conditions from fluorescence measured at 90" in a conventional stopped flow were also biphasic and could be fit by a similar set of rate constants. The rate constants of t-aza-ATP binding to cardiac myosin-S1 (7) , shown by the (.-. -) line in Fig. 5, are 2-5 times slower than the rate constants of t-aza-ATP binding to cardiac actomyosin-S1 and myofibrils at concentrations of t-aza-ATP < 100 p ~ . However, the maximum rate for myosin- S1 was also found to plateau at 4-5 s" at 0 "C at higher nucleotide concentrations (>300 p ~ ) and is probably due to the same step of the mechanism as for actomyosin-S1.

No changes in light intensity were detected in experiments in which ATP was mixed with actomyosin-S1 or myofibrils using the same excitation and emission filters that were used to measure t-aza-ATP fluorescence enhancement. Such ex- periments indicate that rejection of scattered light is suffi- ciently good that light scattering changes could not contribute to the biphasic fluorescence transients that we observed. Thus, the biphasic fluorescence enhancement is due to the kinetic mechanism rather than optical artifact.

The biphasic fluorescence enhancement observed upon Z-

aza-ATP binding to bovine cardiac actomyosin-S1 and my- ofibrils is most easily explained by two sequential transitions

I J

0 SECONDS 5.0 0 SECONDS 2 .o FIG. 4. Stopped flow fluorescence of e-ma-ATP binding to cardiac actomyosin-S1 and myofibrils.

Equal volumes of (A, C) actomyosin-S1 or (B, D ) myofibrils containing approximately (A, B ) 10 PM myosin heads or (C, D ) 20 ,,LM myosin heads were mixed with either (A, B ) 21 p~ or (C, D ) 170 PM t-aza-ATP in the front face stopped flow cell described under "Materials and Methods." The solid line drawn through the data are best fits to either (A, B ) I ( t ) = Zoe-k.b.lt + C or (C, D ) Z ( t ) = Z1e-&' + I Z e - b ' + C. The fit rate constants are: A, k b s l = 1.7 s-'; B, kobZ = 1.3 s-'; C, k&sl = 14.8 s-l (0.47) and kobz = 2.7 s-l (0.53); and D, kbl = 16.9 s" (0.39) and kaba2 = 2.7 s-l (0.61). The values in parentheses for the biphasic fits are the fractional amplitude coefficients, Zl/(Zl + I z ) and Zz/(Zl + I,) corresponding to each rate constant. Experimental conditions were otherwise identical to Fig. 2. Excitation was at 365 nm and emission was at 440-460 nm.

15160

15.f

- v) P

10.1

5.1

Etheno-Aza Nucleotide Binding to Myofibrils and Actomyosin-S1

0 -

0- c

..-. *.. . . . . . .(lJ - - - - !""*"" ._._. ._._.___.-.-.".

I ___.___._.-.-. ---.- 1

50

e-azo-ATP (pM) FIG. 5. The dependence upon [e-aza-ATP] of the observed

rate constant (kobs) of e-aza-ATP binding to cardiac myosin- S 1, actomyosin-81, and myofibrils. Experimental conditions were identical to Fig. 2 except that e-aza-ATP concentration was varied as indicated. Open symbols are for data obtained with acto- myosin-S1; solid symbols are for myofibrils. Single exponential fits of data are indicated by triangles and the fast and slow components of double exponential fits are indicated, respectively, by circles and squares. The solid l ine drawn through the data of the fast phase of the reaction is for a second order rate constant of 1.7 X IO5 M-' s-'. The dashed l ine drawn through the slow phase of the reaction extrap- olates to a maximum rate of 4 s-'. The fractional amplitude coeffi- cient, Ifh/Zmt, plotted in the upper panel of the figure is the ratio of the amplitude coefficient of the fast component to the total amplitude. The upper limit of 10% shown at 21 PM e-aza-ATP is the smallest amount of a fast component that could be detected. Single exponential fits of data for c-aza-ATP binding to bovine cardiac myosin-Sl over this nucleotide concentration range, obtained as described in the accompanying paper (7), are indicated by a dotted line, .-. -. . The solid line segments correspond to the nucleotide binding step and first fluorescence transition (AM + a-ATP * AM-a-ATP c) AM-a-ATP*) and the dashed line segments to the transition (AM-a-ATP* * AM- a-ATP**). The observed rate constants cannot be assigned to indi- vidual transitions at concentrations shown by dotted lines, . . . . . . between states that have different levels of fluorescence emis- sion as indicated by the single and double asterisks in Equa- tion 2. The lines in Fig. 5 are theoretical curves for the dependence of the observed rate constants of the slow and fast components of the fluorescence enhancement upon nu- cleotide concentration calculated from the mechanism shown in Equation 2. The upper panel of Fig. 5 shows the expected dependence of the amplitude of the fast component upon nucleotide concentration.

1.7 x lo5 ~ - 1 s-1 4 s-1 ** AM + AM-a-ATP s rAM-a-ATP1 s r AM-a-ATP 1

The slower fluorescence change may correspond either to the transition to a second (A)M-a-ATP intermediate or to the hydrolytic step producing (A)M-a-ADP-P** (17, 18).

0

B

\ SECONDS

0.5

FIG. 6. Kinetics of the displacement of c-ma-ADP by ATP from cardiac actomyosin-S1 and myofibrils. Equal volumes of either (A) cardiac actomyosin-S1 or (€3) myofibrils containing -20 PM myosin heads and 50 PM E-aza-ADP were mixed with 2 mM ATP. The solid lines through the experimental data are fitted curves cor- responding to rate constants of (A) 20 and ( B ) 18 s-'. Fluorescence emission of 6-aza-ADP bound to actomyosin-S1 was measured at 90" to excitation in a 2-cm path length cell; emission of e-aza-ADP bound to myofibrils was measured using the front face cell described in the legend to Fig. 1 and under "Materials and Methods." Experimental conditions were otherwise identical to those described in the legend to Fig. 2.

Displacement of e-aza-ADP Bound to Cardiac Actomyosin- S1 and Myofibrils by ATP-We have previously shown that the kinetic mechanism of 6-aza-ADP dissociation from bovine cardiac actomyosin-S1 is quantitatively accounted for by Equation 3 in which e-aza-ATP is a competitive inhibitor of the ATP binding sites of actomyosin-S1 (7).

k-m KAT k m

AM-a-ADP e AM AM-ATP e A + "ATP (3) k-TA

At high concentrations of ATP such that KATk-~A[ATP] >> ~ A D [ ~ - A D P ] + k m , the observed rate constant is equal to k-m, the rate constant of 6-aza-ADP dissociation from acto- myosin-S1. Measurements of the dissociation of €-aza-ADP from bovine cardiac actomyosin-SI and myofibrils were made by stopped flow measurements of the decrease in enhance- ment of e-aza-ADP fluorescence observed upon mixing with ATP (Fig. 6). The solid lines through the data are single exponential fits of 20 s-' for actomyosin-S1 (Fig. 6A) and 18 s-' for myofibrils (Fig. 6B). The observed rate constant of 6-

aza-ADP dissociation was unchanged upon increasing the ATP concentration from 0.5 to 1.0 mM and is therefore a good measure of AD.

Etheno-Aza Nucleotide Binding to Myofibrils and Actomyosin-S1 15161

DISCUSSION

Although the kinetic mechanism of hydrolysis of ATP by myosin-Sjl and actomyosin-S1 in solution has been studied in considerable detail, relatively little is known about the hydro- lytic pathway in the lattice of actin and myosin filaments which make up the contractile apparatus of muscle. Acto- myosin-S1 in solution differs in several significant ways from the actin and myosin filaments in muscle. 1) Myosin-S1 is a proteolytic fragment of myosin. Therefore, study of the mech- anism of myosin-S1 or actomyosin-S1 ATP hydrolysis would not reveal any modification of the catalytic properties of the myosin head by the S2 or rod portion of myosin, by the regulatory light chains (which are frequently partially proteo- lyzed or missing in myosin-Sl) or by interaction between the myosin heads. For example, the calcium-sensitive regulation of both scallop and smooth muscle actomyosin does not occur with actomyosin-S1(19,20). 2) Several hundred myosin mol- ecules are organized into each thick filament in muscle. In- termolecular interactions between tails of the myosin mole- cules in the thick filament or between the heads on the surface of the filament could alter the kinetic behavior of the catalytic site or its interaction with actin. 3) The unequal helical repeat of the actin and myosin filaments in vertebrate striated mus- cle requires that some cross-bridges are constrained to ge- ometries that may not be optimal for myosin to bind to actin (4). Recent optical reconstruction of electron micrographs of irfsect flight muscle revealed two classes of rigor cross-bridge structures, which differ in their apparent angle of attachment to actin by nearly 30" (21). The precise details of the distri- bution of cross-bridge binding angles are likely to be different in vertebrate striated muscle and insect flight muscle because of the different axial repeats of the actin and myosin lattice positions; however, it would be surprising if the cross-bridges were to have a uniform geometry of attachment in vertebrate striated muscle.

Detailed analysis of the kinetics of t-aza-ATP binding to myofibrils (7) could not be made because the experiments had to be carried out at high protein concentration and a restricted range of nucleotide concentration to obtain adequate signal to noise ratio. However, at a somewhat less detailed level, the kinetics of t-aza-ATP binding to the myosin cross-bridges in bovine cardiac myofibrils are essentially the same as that to actomyosin-S1. The data obtained in these experiments is summarised in Equation 4.

The differences in the rates of nucleotide binding to and dissociation from cardiac myosin-S1 and actomyosin-SI (7) allow us to distinguish kinetically between free and actin bound cross-bridges in myofibrils. Since the rate constant of t-aza-ADP dissociation from cardiac actomyosin-S1,20 s-', is 400 times greater than that from cardiac myosin-S1,0.05 s-l (7), the similarity between the rates of dissociation of t-aza- ADP from cardiac actomyosin-S1 and myofibrils observed here suggests that most of the nucleotide was bound to at- tached cross-bridges. In addition, we have found that the rate constants of the fast and slow components of the fluorescence enhancement of t-am-ATP binding to cardiac actomyosin-S1 and myofibrils are the same, but are as much as 10 times faster at low nucleotide concentrations ((100 PM) than the observed rate constants of the fluorescence enhancement at corresponding nucleotide concentrations with cardiac myosin-

S1 (7). This kinetic evidence that most of the cross-bridges in bovine cardiac myofibrils behave as if they were bound to actin is in agreement with several additional pieces of physical and biochemical evidence. The affinity of ADP for rabbit skeletal and bovine ventricular myofibrils is similar to the affinity of ADP for the corresponding actomyosin-S1; less than 10% of the sites have affinities for ADP that would be expected for myosin-S1 (22). This is consistent with the observation that more than 95% of the proteolytic digestion products from myosin in skeletal myofibrils have the peptide composition that is characteristic of that obtained with acto- myosin-S1 rather than that obtained with myosin-S1 (23). Moreover, in glycerinated muscle fibers more than 90% of the myosin heads are bound to actin as indicated by the esr spectra of spin labeled myosin (24).

The biochemical and physiological properties of c-aza-nu- cleotides make them valuable tools to investigate the kinetics of the cross-bridge cycle in muscle. The data that we have obtained here about the kinetics of t-aza-nucleotide binding and hydrolysis by actomyosin-S1 in solution and in the my- ofibril is relevant to an interpretation of such studies. The spectrum of the steady state intermediate that we observe in cardiac myofibrils is similar to that obtained with myosin-SI and actomyosin-S1. This suggests that the same steady state intermediates, presumably (A)M-a-ATP and (A)M-a-ADP-P, are present in the simple structured system and in solution. Nagano and Yanagida (8) have used t-aza-ATP as a probe to study the states of the cross-bridge nucleotide complex during contraction of glycerinated rabbit psoas muscle fibers; their observations indicate that these same intermediates are pres- ent in skeletal muscle fibers during isometric contraction or shortening. In addition, we have found that the second order rate constants of t-aza-ATP binding to skeletal or cardiac actomyosin-S1 and cardiac myofibrils are approximately 10- fold slower than the corresponding rate constant for ATP binding (7). This can be related to the observation that the concentration of t-aza-ATP required for a muscle fiber to attain maximal isometric tension under relaxing conditions is 10 times that required for ATP (8). Such results can be explained if the rates of 6-aza-ATP binding to the muscle fibers are 10-fold slower than the rates of binding of ATP, so that a 10-fold higher concentration of nucleotide is necessary to obtain the same steady state concentration of force pro- ducing cross-bridge intermediate when t-aza-ATP is substrate for the cycle.

On the basis of measurements of the rate of ADP dissocia- tion from actomyosin-Sl, isolated from a variety of vertebrate muscles, one of us has previously proposed that ADP disso- ciation from cross-bridges in a muscle may be sufficiently slow to limit the maximum rate of shortening in that muscle (11,25). For such a hypothesis to be correct, the rates of ADP dissociation from actomyosin-S1 in solution and from the nucleotide binding site of cross-bridges must be the same. The rate constant of t-aza-ADP dissociation from bovine cardiac actomyosin-S1, 20 s" at 0 "C, is similar to that measured for ADP dissociation, 15 s-l at 0 "C, under identical conditions (7, 11). These studies show that the rate of e-aza- ADP dissociation from bovine cardiac actomyosin-SI and myofibrils is the same, suggesting that the rate of this process and probably also that of ADP dissociation is not affected by

1.7 x io5 M" s-' AM - ** 0.04 s-l 20 s-l - AM-a-ADP AM

(4)

M-a-ATP (M-a-ADP-P)

15162 Etheno-Aza Nucleotide Binding to Myofibrils and Actomyosin-S1

geometric constraints imposed upon the cross-bridges in my- ofibrils.

The rate constants of most of the steps in the hydrolytic pathway for the analogue 6-aza-ATP are different from those for the physiological substrate, ATP; the exception is that the rate constants of both e-aza-ADP and ADP dissociation from cardiac actomyosin-S1 are similar. Moreover, these processes have the same temperature dependence (7). These observa- tions allow a further indirect test of whether the velocity of unloaded shortening of muscle is limited by the rate at which ADP dissociates from the cross-bridge. This model predicts that if 6-aza-ATP were used as a substrate for a muscle, the velocity of unloaded shortening should be the same as that for ATP. This has been found to be the case for rabbit psoas muscle fibers (8). A proper test would, however, require that shortening velocity measurements and kinetic measurements of the rate of t-aza-ATP dissociation be made on the same type of muscle.

We have shown here that the time courses of the fluores- cence enhancement of eaza-nucleotides associated with t-

aza-ATP binding, e-aza-ADP dissociation, and single turn- overs of t-aza-ATP hydrolysis are essentially the same for bovine cardiac actomyosin-S1 and bovine cardiac myofibrils. These results suggest that the rates of these reactions are not dependent upon the angle of cross-bridge attachment and/or cross-bridge strain since these would be expected to show a distribution of values in myofibrils.

REFERENCES

1. Eisenberg, E., and Greene, L. E. (1980) Annu. Reu. Physwl. 4 2 ,

2. Taylor, E. W. (1979) CRC Crit. Reu. Biochem. 6 , 102-164 3. Sleep, J. A., and Smith, S. J. (1980) Current Top Bioenerg. 11 ,

293-309

239-296

4. Offer, G., Couch, J., O'Brien, E., and Elliot, A. (1981) J. Mol.

5. Sleep, J. A. (1981) Biochemistry 2 0 , 5043-5051 6. White, H. D. (1985) J. Bwl. Chem. 260,982-986 7. Smith, S. J., and White, H D. (1985) J. Bwl. Chem. 260, 15146-

8. Nagano, H., and Yanagida, T. (1984) J. Mol. Biol. 177, 769-785 9. White, H. D., and Taylor, E. W. (1976) Biochemistry 1 5 , 5818-

10. Stein, L. A., Schwarz, R. P., Jr., Chock, P. B., and Eisenberg, E.

11. Siemankowski, R. F., and White, H. D. (1984) J. Biol. Chem.

12. Tada, M., Bailin, G., Barany, K., and Barany, M. (1969) Biochem-

13. Sutoh, K., and Harrington, W. F. (1977) Biochemistry 16,2441-

14. Secrist, J. A., Barrio, J. R., Leonard, N. J., and Weber, G. (1972)

15. Yip, K. F., and Tsou, K. C. (1973) Tetrahedron Lett 33 , 3087-

16. Miyata, H., and Asai, H. (1982) Biochem. Biophys. Res. Commun.

17. Rosenfeld, S. S., and Taylor, E. W. (1984) J. Biol. Chem. 259,

18. Rosenfeld, S. S., and Tavlor. E. W. (1984) J. Biol. Chem. 259,

Biol. 151,663-702

15155

5826

(1979) Biochemistry 18,3895-3909

259,5045-5053

istry 8,4842-4850

2448

Biochemistry 11,3499-3506

3090

105,296-302

11920-11929

11908-11919 '

- . . .

19. Chantler, P. D., and Szent-Gyorgyi, A. G. (1979) J. Mol. Biol. 138,473-492

20. Marston, S. B., and Taylor, E. W. (1978) FEBS Lett. 8 6 , 167-

21. Taylor, K. A., Reedy, M. C., Cordova, L., and Reedy, M. K. (1984)

22. Johnson, R. E., and Adams, P. H. (1984) FEBS Lett. 174 , 11-

23. Lovell, S. E., and Harrington, W. F. (1981) J. Mol. Biol. 149 ,

24. Thomas, D., and Cooke, R. (1980) Biophys. J. 32,891-906 25. Siemankowski, R. F., Wiseman, M. O., and White, H. D. (1985)

170

Nature 310,285-291

14

659-674

Proc. Natl. Acad. Sci. U. S. A. 82,658-662