Fluorescent coumarin-labeled nucleotides to measure ADP release from actomyosin.

Several coumarin-labeled nucleotides have been synthesized, based on 2'(3')-O-(2-aminoethyl)carbamoyl-ATP (edaATP). The fluorescent coumarins coupled with the free amino group are 7-diethylaminocoumarin-3-carboxylic acid (to give deac-edaATP), coumarin 343 (but-edaATP) and 7-ethylamino-8-bromocoumarin-3-carboxylic acid (mbc-edaATP). The carbamoyl linkage of these nucleotide analogs undergoes interconversion between 2'- and 3'-hydroxyl attachment very slowly, so that the 2'- and 3'-isomers were separated and stored with minimal equilibration. 3'-Deac-edaADP had fluorescence excitation and emission maxima at 430 nm and 477 nm, with a fluorescence quantum yield of 0.012. The equivalent data for 3'-but-edaADP are 445 nm, 494 nm, and 0.51, respectively, and for 3'-mbc-edaADP, 405 nm, 464 nm, and 0.62. The interaction with skeletal myosin subfragment 1 was measured in the absence and presence of actin. In each case the fluorescence was decreased when bound to subfragment 1, 3-fold for 3'-deac-edaADP, 7-fold for 3'-but-edaADP, and 11-fold for 3'-mbc-edaADP. Steady-state ATPase measurements and the kinetics of binding and release of nucleotides were similar to those reported for the natural nucleotide. Large fluorescence changes could be observed for the release of these analogs from actomyosin subfragment 1, enabling a direct measurement of the kinetics of this process. In the case of 3'-deac-edaADP a rate constant of 474 s(-1) was measured (at pH 7.0, 20 degrees C, and low ionic strength).

Defining the roles of individual residues in the single-stranded DNA binding site of PcrA helicase.

Crystal structures and biochemical analyses of PcrA helicase provide evidence for a model for processive DNA unwinding that involves coupling of single-stranded DNA (ssDNA) tracking to a duplex destabilization activity. The DNA tracking model invokes ATP-dependent flipping of bases between several pockets on the enzyme formed by conserved aromatic amino acid residues. We have used site-directed mutagenesis to confirm the requirement of all of these residues for helicase activity. We also demonstrate that the duplex unwinding defects correlate with an inability of certain mutant proteins to translocate effectively on ssDNA. Moreover, the results define an essential triad of residues within the ssDNA binding site that comprise the ATP-driven DNA motor itself.

A fluorescent sensor of the phosphorylation state of nucleoside diphosphate kinase and its use to monitor nucleoside diphosphate concentrations in real time.

A sensor for purine nucleoside diphosphates in solution based on nucleoside diphosphate kinase (NDPK) has been developed. A single cysteine was introduced into the protein and labeled with the environmentally sensitive fluorophore, N-[2-(iodoacetamido)ethyl]-7-diethylaminocoumarin-3-carboxamide. The resultant molecule shows a 4-fold fluorescence increase when phosphorylated on His117; this phosphorylation is on the normal reaction pathway of the enzyme. The emission maximum of the phosphoenzyme is at 475 nm, with maximum excitation at 430 nm. The fluorescent phosphorylated NDPK is used to measure the amount of ADP and the unphosphorylated to measure ATP. The labeled protein is phosphorylated to > 90%, and the resultant molecule is stable on ice or can be stored at -80 degrees C. The fluorescence responds to the fraction of protein phosphorylated and so to the equilibrium between ADP plus NDPK approximately P and ATP plus NDPK. In effect, the sensor measures the ADP/ATP concentration ratio. The enzyme has a broad specificity for the purine of the nucleotides, so the sensor also can measure GDP/GTP ratios. The fluorescence and kinetic properties of the labeled protein are described. The binding rate constants of nucleotides are approximately 10(5) M(-1) s(-1), and the fluorescence change is at >200 s(-1) when the ADP concentration is >1 mM. Results are presented with two well-defined systems, namely, the kinetics of ADP release from myosin subfragment 1 and GDP release from the small G protein, human rho. The results obtained with this novel sensor agree with those from alternate methods and demonstrate the applicability for following micromolar changes in nucleoside diphosphate in real time.

Uncoupling DNA translocation and helicase activity in PcrA: direct evidence for an active mechanism.

DNA footprinting and nuclease protection studies of PcrA helicase complexed with a 3'-tailed DNA duplex reveal a contact region that covers a significant region of the substrate both in the presence and absence of a non-hydrolysable analogue of ATP, ADPNP. However, details of the interactions of the enzyme with the duplex region are altered upon binding of nucleotide. By combining this information with that obtained from crystal structures of PcrA complexed with a similar DNA substrate, we have designed mutant proteins that are defective in helicase activity but that leave the ATPase and single-stranded DNA translocation activities intact. These mutants are all located in domains 1B and 2B, which interact with the duplex portion of the DNA substrate. Taken together with the crystal structures, these data support an 'active' mechanism for PcrA that involves two distinct ATP-dependent processes: destabilization of the duplex DNA ahead of the enzyme that is coupled to DNA translocation along the single strand product.

Demonstration of unidirectional single-stranded DNA translocation by PcrA helicase: measurement of step size and translocation speed.

Using a fluorescent sensor for inorganic phosphate, the kinetics of ATP hydrolysis by PcrA helicase were measured in the presence of saturating concentrations of oligonucleotides of various lengths. There is a rapid phase of inorganic phosphate release that is equivalent to several turnovers of the ATPase, followed by slower steady-state ATP hydrolysis. The magnitude of the rapid phase is governed by the length of single-stranded DNA, while the slow phase is independent of its length. A kinetic model is presented in which the rapid phase is associated with translocation along single-stranded DNA, after the PcrA binds randomly along the DNA. There is a linear relationship between the length of single-stranded DNA and both the duration and amplitude of the rapid phase. These data suggest that the translocation activity occurs at 50 bases/s in unidirectional single-base steps, each requiring the hydrolysis of 1 ATP molecule.

A kinetic analysis of the nucleotide-induced allosteric transitions of GroEL.

Single-point mutants of GroEL were constructed with tryptophan replacing a tyrosine residue in order to examine nucleotide-induced structural transitions spectrofluorometrically. The tyrosine residues at positions 203, 360, 476 and 485 were mutated. Of these, the probe at residue 485 gave the clearest fluorescence signals upon nucleotide binding. The probe at 360 reported similar signals. In response to the binding of ATP, the indole fluorescence reports four distinct structural transitions occurring on well-separated timescales, all of which precede hydrolysis of the nucleotide. All four of these rearrangements were analysed, two in detail. The fastest is an order of magnitude more rapid than previously identified rearrangements and is proposed to be a T-to-R transition. The next kinetic phase is a rearrangement to the open state identified by electron cryo-microscopy and this we designate an R to R* transition. Both of these rearrangements can occur when only a single ring of GroEL is loaded with ATP, and the results are consistent with the occupied ring behaving in a concerted, cooperative manner. At higher ATP concentrations both rings can be loaded with the nucleotide and the R to R* transition is accelerated. The resultant GroEL:ATP14 species can then undergo two final rearrangements, RR*-->[RR](+)-->[RR](#). These final slow steps are completely blocked when ADP occupies the second ring, i.e. it does not occur in the GroEL:ATP7:ADP7 or the GroEL:ATP7 species. All equilibrium and kinetic data conform to a minimal model in which the GroEL ring can exist in five distinct states which then give rise to seven types of oligomeric conformer: TT, TR, TR*, RR, RR*, [RR](+) and [RR](#), with concerted transitions between each. The other eight possible conformers are presumably disallowed by constraints imposed by inter-ring contacts. This kinetic behaviour is consistent with the GroEL ring passing through distinct functional states in a binding-encapsulation-folding process, with the T-form having high substrate affinity (binding), the R-form being able to bind GroES but retaining substrate affinity (encapsulation), and the R*-form retaining high GroES affinity but allowing the substrate to dissociate into the enclosed cavity (folding). ADP induces only one detectable rearrangement (designated T to T*) which has no properties in common with those elicited by ATP. However, asymmetric ADP binding prevents ATP occupying both rings and, hence, restricts the system to the T*T, T*R and T*R* complexes.

Phosphate release during microtubule assembly: what stabilizes growing microtubules?

The molecular mechanism underlying microtubule dynamic instability depends on the relationship between the addition of tubulin-GTP to a growing microtubule and its hydrolysis in the microtubule lattice to tubulin-GDP, with release of inorganic phosphate (Pi). Since this relationship remains controversial, we have re-examined the release of Pi upon microtubule assembly using a fluorometric assay for Pi, based on the phosphate-binding protein of Escherichia coli [Brune M., Hunter, J. L., Corrie, J. E. T., and Webb, M. R. (1994) Biochemistry 33, 8262-8271]. Microtubule assembly and Pi release were monitored simultaneously in a standard fluorimeter as an increase in the turbidity and fluorescence, respectively, in tubulin-GTP solutions assembled under conditions supporting dynamic instability. At the steady state of assembly, Pi release is nonlinear with respect to time, proceeding at a rate determined by the following: (a) the intrinsic GTPase activity of the nonpolymerized tubulin-GTP, and (b) the microtubule number concentration, which decreases progressively. Direct observation of the time course of nucleated microtubule assembly indicates that Pi release is closely coupled to microtubule elongation, even during the initial stages of assembly when uncoupling of tubulin-GTP addition and GTP hydrolysis would be most evident. Studies of the inhibition and reversal of the growth phase by cytostatic drugs show no evidence of a burst of Pi release. We conclude that nucleotide hydrolysis can keep pace with tubulin-GTP addition rates of 200 molecules per second per microtubule and that extended caps of tubulin-GTP or tubulin-GDP-Pi are not generated in normal assembly, nor are they required to stabilize growing microtubules or to support the phenomenon of dynamic instability of microtubules at the steady state.

The efficiency of contraction in rabbit skeletal muscle fibres, determined from the rate of release of inorganic phosphate.

1. The relationship between mechanical power output and the rate of ATP hydrolysis was investigated in segments of permeabilized fibres isolated from rabbit psoas muscle. 2. Contractions were elicited at 12 degrees C by photolytic release of ATP from the P3 -1-(2-nitrophenyl) ester of ATP (NPE-caged ATP). Inorganic phosphate (Pi) release was measured by a fluorescence method using a coumarin-labelled phosphate binding protein. Force and sarcomere length were also monitored. 3. ATPase activity was determined from the rate of appearance of Pi during each phase of contraction. The ATPase rate was 10.3 s-1 immediately following release of ATP and 5. 1 s-1 during the isometric phase prior to the applied shortening. It rose hyperbolically with shortening velocity, reaching 18.5 s-1 at a maximal shortening velocity > 1 ML s-1 (muscle lengths s-1). 4. Sarcomeres shortened at 0.09 ML s-1 immediately following the photolytic release of ATP and at 0.04 ML s-1 prior to the period of applied shortening. The high initial ATPase rate may be largely attributed to initial sarcomere shortening. 5. During shortening, maximal power output was 28 W l-1. Assuming the free energy of hydrolysis is 50 kJ mol-1, the efficiency of contraction was calculated from the power output at each shortening velocity. The maximum efficiency was 0.36 at a shortening velocity of 0.27 ML s-1, corresponding to a force level 51 % of that in the isometric state. 6. At the maximal shortening velocity, only 10 % of the myosin heads are attached to the thin filaments at any one time.

The interaction between rac1 and its guanine nucleotide dissociation inhibitor (GDI), monitored by a single fluorescent coumarin attached to GDI.

The interaction of rac with guanine nucleotide dissociation inhibitor protein (rhoGDI) is described, using GDI fluorescently labeled on its single cysteine with N-[2-(1-maleimidyl)ethyl]-7-diethylaminocoumarin-3-carboxamide (MDCC). The labeled GDI shows a 70% decrease in fluorescence emission on binding geranylgeranylated rac1.GDP and has an affinity for rac1 within a factor of 2 of the unlabeled GDI. The labeled GDI was used to determine the kinetic mechanism of the interaction by measuring the association and dissociation in real time. The kinetics are interpreted in terms of a two-step mechanism: binding of rac to GDI and then a conformational change of the complex with an overall dissociation constant of 0.4 nM. The conformational change has a rate constant of 7.3 s-1 (pH 7.5, 30 degrees C), and the reverse has a rate constant of 1.4 x 10(-)3 s-1. To overcome difficulties inherent in using and manipulating lipid-modified rac, we also used a combination of unmodified rac1, expressed in Escherichia coli and produced with C-terminal truncation (thus lacking the cysteine that is the site of lipid attachment), and farnesylated C-terminal peptide. This combination can mimic geranylgeranylated rac1, producing a complex with the coumarin-labeled GDI, and was used to examine the relative importance of different regions of rac1 in interaction with GDI.

Real-time visual detection of Pi released by flagellar dynein ATPase.

Using a novel fluorescent probe for Pi, a method for the direct visualization of Pi release from reactivated flagellar dynein ATPase has been developed. The probe undergoes a fluorescence increase when it binds Pi. The technique involves simultaneous imaging of demembranated sperm tails by epi-fluorescence and dark-field microscopy, and the use of the caged ATP technique for axoneme reactivation. To limit diffusion and thus maintain the released Pi within the observed field of view, the assay is carried out within a minute droplet under oil (volume 5-15 pl). The video output of a recursively filtered ICCD camera is used to visualize the fluorescence signal, which is subsequently digitized and automatically analysed on a PC. A major advantage of this technique is that it enables simultaneous analysis of the ATP-utilization rate and the motility of the reactivated axonemes.

Time-resolved measurements of phosphate release by cycling cross-bridges in portal vein smooth muscle.

The rate of release of inorganic phosphate (Pi) from cycling cross-bridges in rabbit portal-anterior mesenteric vein smooth muscle was determined by following the fluorescence of the Pi-reporter, MDCC-PBP (Brune, M., J. L. Hunter, S. A. Howell, S. R. Martin, T. L. Hazlett, J. E. T. Corrie, and M. R. Webb. 1998. Biochemistry. 37:10370-10380). Cross-bridge cycling was initiated by photolytic release of ATP from caged-ATP in Triton-permeabilized smooth muscles in rigor. When the regulatory myosin light chains (MLC20) had been thiophosphorylated, the rate of Pi release was biphasic with an initial rate of 80 microM s-1 and amplitude 108 microM, decreasing to 13.7 microM s-1. These rates correspond to fast and slow turnovers of 1.8 s-1 and 0.3 s-1, assuming 84% thiophosphorylation of 52 microM myosin heads. Activation by Ca2+-dependent phosphorylation subsequent to ATP release resulted in slower Pi release, paralleling the rate of contraction that was also slower than after thiophosphorylation, and was also biphasic: 51 microM s-1 and 13.2 microM s-1. These rates suggest that the activity of myosin light chain kinase and phosphatase ("pseudo-ATPase") contributes <20% of the ATP usage during cross-bridge cycling. The extracellular "ecto-nucleotidase" activity was reduced eightfold by permeabilization, conditions in which the ecto-ADPase was 17% of the ecto-ATPase. Nevertheless, the remaining ecto-ATPase activity reduced the precision of the estimate of cross-bridge ATPase. We conclude that the transition from fast to slow ATPase rates reflects the properties and forces directly acting on cross-bridges, rather than the result of a time-dependent decrease in activation (MLC20 phosphorylation) occurring in intact smooth muscle. The mechanisms of slowing may include the effect of positive strain on cross-bridges, inhibition of the cycling rate by high affinity Mg-ADP binding, and associated state hydrolysis.

Vibrational study of phosphate modes in GDP and GTP and their interaction with magnesium in aqueous solution.

Raman and infrared spectra were examined for guanosine 5'-diphosphate (GDP) and guanosine 5'-triphosphate (GTP) in aqueous solution. The vibrational modes were assigned on the basis of isotopic frequency shifts and relative intensities in the Raman and infrared spectra. The observed frequency shifts on 18O isotope labeling made it possible to identify the bands from each phosphate group (alpha, beta, gamma). Frequency shifts were observed as Mg2+ complexes with GDP and GTP. The results suggested that Mg2+ binds to GDP in a bidentate manner to the alpha, beta P[symbol: see text]O bonds and in a tridentate manner to the alpha, beta and gamma P[symbol: see text]O bonds of Mg.GTP. The results indicate that structure of Mg2+ coordinated to GTP in aqueous solution differs somewhat to that found for Mg.ATP.

Raman difference studies of GDP and GTP binding to c-Harvey ras.

The vibrational spectra of phosphate modes for GDP and GTP bound to the c-Harvey p21(ras) protein have been determined using 18O isotope edited Raman difference spectroscopy. A number of the phosphate stretch frequencies are changed upon GDP/GTP binding to ras, and the results are analyzed by ab initio calculations and through the use of empirical relationships that relate bond orders and bond lengths to vibrational frequencies. Bound GDP is found to be strongly stabilized by its interactions, mostly electrostatic, with the active site Mg2+. Bound GTP also interacts with the active site Mg2+ via its beta-phosphate group, as expected on the basis of crystallographic studies of bound GppNp. The angle between the nonbridging P&bondDot;O bonds of the gamma-phosphate of bound GTP increase by about 1-2 degrees compared to its solution value, thus bringing about a geometry that is closer to planar for these bonds as expected for the putative pentacoordinated transition state geometry of the phosphotransfer reaction. Modeling of the interactions at the nucleotide binding site suggests that the water molecule in-line with the P-O bond is positioned to bring about the change in bond angle. Moreover, a weak fifth bond (about 0.03 vu) appears to be formed between it and the gamma-phosphorus atom of bound GTP with a concomitant weakening of the O-P bond between the GDP leaving group and the gamma-phosphorus atom. Hence, an important role of the active site structure appears to be the strategic positioning of this in-line water. These structural results are consistent with a reaction pathway for GTP hydrolysis in ras of synchronous bond formation between the gamma-phosphorus of GTP and the attacking nucleophile and bond breaking between the gamma-phosphorus and the GDP leaving group.

Mechanism of inorganic phosphate interaction with phosphate binding protein from Escherichia coli.

The mechanism of Pi interaction with phosphate binding protein of Escherichia coli has been investigated using the A197C mutant protein labeled with a coumarin fluorophore (MDCC-PBP), which gives a fluorescence change on binding Pi. A pure preparation of MDCC-PBP was obtained, in which the only significant inhomogeneity is the presence of equal amounts of two diastereoisomers due to the chiral center formed on reaction of the cysteine with the maleimide. These diastereoisomers could not be separated, but Pi binding data suggest that they differ in affinity and fluorescence change. When Pi binds to MDCC-PBP, the fluorescence quantum yield increases 8-fold and the fluorescence intensity at 465 nm increases 13-fold. The kinetics of Pi binding show saturation of the rate at high Pi concentrations, and this together with other information suggests a two-step mechanism with the fluorescence change after binding, concomitant with a conformational change of the protein that closes the cleft containing the Pi binding site. Cleft closure has a rate constant of 317 s-1 (pH 7.0, 5 degrees C), and opening has a rate constant of 4.5 s-1. The fluorescence increase is likely to arise from a change in the hydrophobic environment during this closure as the steady state fluorescence emission (lambdamax and intensity) on Pi binding is mimicked by the addition of ethanol to aqueous solutions of an MDCC-thiol adduct. Fluorescence lifetimes in the absence and presence of Pi were 0.3 and 2.4 ns, respectively, consistent with the change in quantum yield. The rotational correlation time of the coumarin increases only 2-fold from 15 to 26 ns on binding Pi as measured by time-resolved polarization, consistent with the main rotation being determined by the protein even in the open conformation, but with greater local motion. Circular dichroism of the coumarin induced by the protein is weak in the absence of Pi and increases strongly upon saturation by Pi. These data are also consistent with an open to closed conformational model.

Crystal structure of phosphate binding protein labeled with a coumarin fluorophore, a probe for inorganic phosphate.

Crystal structures are presented for the A197C mutant of Escherichia coli phosphate binding protein (PBP) and the same mutant labeled at Cys197 with N-[2-(1-maleimidyl)ethyl]-7-(diethylamino)coumarin-3-carboxamide (MDCC). Both proteins are complexed with inorganic phosphate. The latter molecule, MDCC-PBP, exhibits a large increase in fluorescence on binding inorganic phosphate. The resulting high-fluorescence state of the coumarin and the ability of this coumarin to monitor the conformational changes associated with inorganic phosphate binding are interpreted in terms of the specific interactions of MDCC with the protein. The structure helps to explain why this particular label gives a high-fluorescence state on binding inorganic phosphate, while several other related labels do not, and hence aids our general understanding of environmentally sensitive fluorescence probes on proteins.

ATPase and shortening rates in frog fast skeletal myofibrils by time-resolved measurements of protein-bound and free Pi.

Shortening and ATPase rates were measured in Ca2+-activated myofibrils from frog fast muscles in unloaded conditions at 4 degrees C. ATPase rates were determined using the phosphate-binding protein method (free phosphate) and quench flow (total phosphate). Shortening rates at near zero load (V0) were estimated by quenching reaction mixtures 50 ms to 10 s old at pH 3.5 and measuring sarcomere lengths under the optical microscope. As with the rabbit psoas myofibrils (C. Lionne, F. Travers, and T. Barman, 1996, Biophys. J. 70:887-895), the ATPase progress curves had three phases: a transient Pi burst, a fast linear phase (kF), and a deceleration to a slow phase (kS). Evidence is given that kF is the ATPase rate of shortening myofibrils. V0 is in good agreement with mechanical measurements in myofibrils and fibers. Under the same conditions and at saturation in ATP, V0 and kF are 2.4 microm half-sarcomere(-1) s(-1) and 4.6 s(-1), and their Km values are 33 and 200 microM, respectively. These parameters are higher than found with rabbit psoas myofibrils. The myofibrillar kF is higher than the fiber ATPase rates obtained previously in frog fast muscles but considerably lower than obtained in skinned fibers by the phosphate-binding protein method (Z. H. He, R. K. Chillingworth, M. Brune, J. E. T. Corrie, D. R. Trentham, M. R. Webb, and M. R. Ferenczi, 1997, J. Physiol. 50:125-148). We show that, with frog as with rabbit myofibrillar ATPase, phosphate release is the rate-limiting step.

Purine nucleoside phosphorylase: its use in a spectroscopic assay for inorganic phosphate and for removing inorganic phosphate with the aid of phosphodeoxyribomutase.

The kinetics of the phosphorolysis of 7-methylated guanosine analogues catalyzed by purine nucleoside phosphorylase has been analyzed to understand the use of this system as a "Pi mop" to remove Pi from solutions and as a spectroscopic assay for Pi at micromolar concentrations. An expression system was developed for the phosphorylase from Escherichia coli: this protein (subunit molecular mass 26 kDa) and one from a commercial source (29 kDa) were used in this study. Rates of >50 s-1 were obtained for the phosphorolysis at 30 degrees C, so that when the phosphorylase is coupled to the phosphatase being studied, rates of Pi release from the phosphatase can be measured close to this rate. The kinetic mechanism appears to obey the Michaelis-Menten model in the steady state with the bond cleavage rate limiting. Slow hydrolysis of ribose-1-phosphate to Pi catalyzed by the phosphorylase limits the efficiency of the Pi mop. To overcome this, phosphodeoxyribomutase was used to catalyze the conversion of ribose-1-phosphate to ribose-5-phosphate, enabling the Pi mop to remove large amounts of Pi quantitatively. Acyclovir diphosphate provides a simple method to switch off the Pi mop as it is a tight inhibitor (Kd 12 nM) of purine nucleoside phosphorylase.

Kinetics of nucleoside triphosphate cleavage and phosphate release steps by associated rabbit skeletal actomyosin, measured using a novel fluorescent probe for phosphate.

We have measured the kinetics of inorganic phosphate (Pi) release during a single turnover of actomyosin nucleoside triphosphate (NTP) hydrolysis using a double-mixing stopped-flow spectrofluorometer, at very low ionic strength to increase the affinity of myosin-ATP and myosin-ADP-Pi to actin. Myosin subfragment 1 and a series of nucleoside triphosphates were mixed and incubated for approximately 1-10 s to allow NTP to bind to myosin and generate a steady state mixture of myosin-NTP and myosin-NDP-Pi. The steady state intermediates were then mixed with actin. The kinetics of Pi release were measured using a fluorescent probe for Pi, based on a phosphate binding protein [Brune et al. (1994) Biochemistry 33, 8262-8271]. These data are correlated with quenched-flow data, where the extent of the rapid burst of hydrolysis during the first turnover of ATP hydrolysis was followed by chemical quenching of the reaction mix at various times after rapidly mixing ATP and myosin subfragment 1. From the double-mixing actomyosin measurements, the kinetics of Pi release are biphasic. The fast phase corresponds to Pi release from the associated actomyosin-ADP-Pi complex. The slow phase measures the rate of the cleavage step on associated actomyosin. At saturating actin, there is a correlation between the amplitude of the fast phase and the size of the Pi burst observed by quenched flow in the absence of actin: the size of this phase corresponds to the amount of myosin-ADP-Pi formed during the first mix. For ATP at 20 degrees C the rate of the Pi release step is 75 (+/-5) s-1, 25-fold larger than the cleavage step, which is the rate-limiting step of actomyosin ATP hydrolysis at saturating actin. The rate constant of Pi release varies only slightly with nucleoside structure. The rate constant of the slow phase of the Pi release (measuring cleavage) is highly dependent upon the structure of the NTP substrate.