Proc. Nati. Acad. Sci.

USA
Vol. 76, No. 6, pp. 2620--2624, Juise 1979
Biochemistry

Inhibition of myosin ATPase by vanadate ion

(enzyme mechanisms/enzyme kinetics/active-site modification/metal toxicity/protein conformation)
CHARLES C. GOODNO
Department of Biophysics and Theoretical Biology, The University of Chicago, Chicago, Illinois 60637
Conmmunicated by Donald F. Steiner, March 12, 1979

ABSTRACT Inhibition of the myosin ATPase by vana- ATPase, to the extent that stoichiometric concentrations of Vi
date ion (V1) has been studied in 90 mM NaCl/5 mM MgCl2/20 produce almost total inhibition. The mechanism of inhibition
mM Tris HCI, pH 8.5, at 250C. Although the onset of inhibition by Vi was examined, and the results provide a simple expla-
during the assay is slow and dependent upon Vi concentration nation for the lack of inhibition observed in earlier studies.
(kap -t 0.3 M-1 s-'), the final level of inhibition approaches
iOO4o, provided the V; concentration is in slight excess over the MATERIALS AND METHODS
concentration of ATPase sites. Inhibition is not reversible by
dialysis or the addition of reducing agents. The source of this Materials. Na2ATP and Na2ADP were products of Sigma
irreversible inhibition consists of the formation of a stable, in- and [a-32PIATP and [3HjADP were products of New England
active complex with the composition MADP-V1 (where M rep- Nuclear. Na3VO4 and V205 were supplied by Fisher. The in-
resents a single myosin active site). The complex has been iso-
lated, and its mechanism of formation from M, ADP, and Vi has dicator dye 4-(2-pyridylazo)-resorcinol (PAR), was a product
been studied. Omission of ATP increases the rate of formation of Aldrich. Other chemicals were of reagent grade.
by about 35-fold (kapp 11 M-1 s-'), yet this rate is still low in Proteins. Myosin and chymotryptic heavy meromyosin
comparison with the rates of simple protein-ligand association (HMM) were prepared according to the procedures of Perry
reactions. This slowness is interpreted in terms of a rate-limiting (9) and Weeds and Taylor (10), respectively. The HMM fraction
isomerization step that follows the association of M, ADP, and that precipitated between 45 and 60% saturated ammonium
V1: M*ADP*V1 _ Mt.ADP.V1 (t indicates the inactive product sulfate was dialyzed free of ammonium sulfate, centrifuged 30
of the isomerization). The properties of Mt.ADP-Vi are com-
pared with those of the ATPase intermediate M**-ADP.Pi, and min at 40,000 X g, and used within 10 days. The concentration
the possible role of V; as an analog of Pi is discussed. of HMM was expressed in terms of the ATPase-site concen-
tration, which was determined spectrophotometrically by using
Evidence is accumulating that suggests that enzymes involved a value of A2801% = 6.47 (11), assuming a molecular weight of
in phosphotransferase or phosphohydrolase reactions are ca- 340,000 and two ATPase sites per molecule.
pable of accepting vanadate ion (Vi) as an analog of inorganic ATPase assays were carried out in buffer A (0.09 M NaCl/5
phosphate (Pi). To date, there is evidence that six enzymes ex- mM MgCl2/20 mM Tris-HCI, pH 8.5) at 250C by the addition
hibit this property: ribonuclease A (1), acid phosphatase (2), of MgATP (final concentration, 1 mM) to HMM at 1-7 AM sites.
alkaline phosphatase (3), glyceraldehyde-3-phosphate dehy- Assay times ranged from 0.1 to 5 hr. The reaction was stopped
drogenase (4), Na+,K+-ATPase (5-7), and the dynein ATPase in aliquots (1 ml) of the assay solution with 1 ml of 10% tri-
(8). Lindquist et al. (1) observed that V1 was a competitive in- chloroacetic acid. The aliquots were then clarified by centrif-
hibitor of ribonuclease and explained this finding in terms of ugation and half of each was analyzed for Pi by the procedure
the formation of a complex between V1 and uridine that re- of Taussky and Shorr (12). V1 concentrations below 10 mM
sembled the intermediate uridine 2',3'-phosphate. Van Etten caused less than 1% interference.
et al. (2) found that acid phosphatase was competitively in- Vanadium Analysis. Stock solutions of V1 were prepared
hibited by Vi with a Ki value at least 100-fold lower than that from either Na3VO4 (adjusted to pH 10 with 6 M HCl) or V205
for Pi. Evidence that Vi binds to the same site as Pi was provided (adjusted to pH 10 with 10 M NaOH) and then boiled to destroy
by Lopez et al. (3), who found that Vi and Pi inhibit alkaline yellow polymeric species such as V100286- (13). Standard so-
phosphatase in mutually exclusive fashion. lutions were prepared by volumetric dilution. In order to
In studies on glyceraldehyde-3-phosphate dehydrogenase, minimize the pH-dependent polymerization of Vi, all studies
DeMaster and Mitchell (4) made the remarkable observation were carried out under the alkaline conditions used for the
that Vi appears to function as an alternative substrate for the ATPase assays (buffer A). UV-visible spectra of Vi standard
enzyme in place of Pi. Recently, Josephson and Cantley (5) and solutions were obtained by using a Cary 14 spectrophotometer,
Cantley et al. (6, 7) have shown that Vi strongly inhibits the and the extinction coefficient was determined: Xrnax = 265 nm,
Na+,K+-ATPase, and Gibbons et al. (8) have shown that it in- C265 = 2925 M-1 cm-1. The Vi concentration was determined
hibits the dynein ATPase as well. A plausible explanation for spectrophotometrically wherever possible.
all these results is provided by the suggestions of Lindquist et Where this was unfeasible (e.g., in the presence of protein),
al. (1) and Van Etten et al. (2) that the tetrahedral vanadate ion vanadium was determined by a modification of the colorimetric
is structurally analogous to Pi. procedure of Pribil (14), using the metallochromic dye PAR.
Although preliminary investigations by two groups con- To a 1-ml sample in buffer A was added 100 Al of 1 M imidazole
cluded that V1 did not inhibit the myosin ATPase (5, 8), the mass (pH 6.0) and subsequently 100 Al of 2 mM PAR. After 30 min
of experimental results with other enzymes suggested that in- of color development, the absorbance was read at 550 nm.
hibition of myosin was likely to occur. In this investigation, a Vanadium determinations were sometimes made in the pres-
detailed examination was made of the effect of V1 on myosin. ence of 1% sodium dodecyl sulfate (NaDodSO4), for which 63
The results show that V1 is an effective inhibitor of the myosin gl of 20% NaDodSO4 was added to the usual assay. Although
The publication costs of this article were defrayed in part by page Abbreviations: HNIM, heavy meromyosin; M, single active site of the
charge payment. This article must therefore be hereby marked "ad- myosin ATPase; PAR, 4-(2-pyridylazo)-resorcinol; NaDodSO4, sodium
vertisement" in accordance with 18 U. S. C. §1734 solely to indicate dodecyl sulfate; Vi, vanadate ion (unspecified degree of protona-
this fact. tion)-i.e., V043-, HV0427 or H2VO4 -
2620
 Biochemistry: Goodno Proc. Natl. Acad. Sci. USA 76 (1979) 2621

it was necessary to correct for small changes in absorbance

produced by HMM and NaDodSO4, the caibition e rqptvs
linear in the range of 1-20,M Vs. Because- -dfte senrXsif-bf
the analysis, the vanadium content of glass posed an interfer-
ence, which was circumvented by the use of plastic vessels.
M-ADP-Vj complexes (M denotes a single active site of my-
osin ATPase) were isolated at room temperature from incu-
bation mixtures of HMM + ADP + V1 by two chromatographic
procedures: Gel filtration (0.5-0.8 ml sample) was carried out
by using a 1.5 X 22 cm column of Sephadex G-25, which was
eluted at a flow rate of 0.5 ml/min. Ion exchange chromatog-
raphy (1- to 5-ml sample) was carried out by using 0.5 X 2 cm
columns of Dowex-1 X 8 (200-400 mesh), which were eluted
at a flow rate of 0.2 ml/min. Both types of columns were
equilibrated and eluted with buffer A. Controls with HMM +
ADP and HMM + V1 confirmed the ability of these columns
to remove at least 99% of unbound ADP or Vs.
The total vanadium content of the M-ADP-Vj complex was Time, hr
determined by the colorimetric procedure in the presence of
1% NaDodSO4. The V1 sequestered by the complex was de- FIG. 1. Inhibition of HMM ATPase by V1. ATPase assays were
termined as the difference in apparent vanadium content of carried out with 2.5 ,uM HMM (sites) in 90 mM NaCl/5 mM MgCl2/20
the complex before and after denaturation with 1% NaDodSO4. mM Tris-HCl, pH 8.5/1 mM MgATP (buffer A) at 250C. When V1 and
Of the total vanadium in the isolated complex, 85-95% was ADP were added, the final concentrations were 0.5 mM and 1.0 mM,
found to be sequestered. respectively. After a 5-min preincubation of the HMM, the assay was
begun by addition of ATP. Although more types of assays were carried
Incorporation of ADP into the complex was determined by out, the time courses were identical in several cases, leading to only
using suitable dilutions of either [3HIADP or [a-32P]ADP four distinct classes of assay curves. Assays A and B contained no
generated in situ from [a-32P]ATP. [a-32P]ADP was deter- addition and added ADP, respectively. Assays C-E contained Vi
mined by Cerenkov counting in H20, and [3H]ADP was de- alone: C, V1 added to assay; D, V1 added to preincubation mixture; E,
termined by scintillation counting under standard condi- V1 incubated with ATP for 1 hr before addition to assay. Assays F and
tions. G contained both ADP and Vi: F, V1 added to preincubation mixture
and ADP added to assay, G, ADP added to preincubation mixture and
Fluorescence Studies. Fluorescence emission spectra of V1 added to assay. Assay H contained ADP and V;, both of which were
HMM and its complexes were obtained at an active site con- added to the preincubation mixture. Assay I contained ADP and V1
centration of 3 MM in buffer A at 250C on a Hitachi-Perkin which were incubated together for 1 hr prior to addition to the assay.
Elmer MPF-44A fluorescence spectrophotometer. An excitation (Inset) Plot of natural logarithm of the residual ATPase activity (%)
wavelength of 295 nm was used, and the emission maximum versus time (data from assay C). The slope gives a value of k0b1 = 1.4
was at 335 nm. Although elevated concentrations of Vi inter- X 10-4 s-1, and the intercept corresponds to an initial inhibition of
16%.
fered with fluorescence measurements, concentrations below
10 MM caused less than 1% interference.
amined by preincubation of HMM with Vi prior to the assay.
RESULTS Fig. 1 (curve D) shows that this preincubation had no effect on
ATPase inhibition studies were carried out at pH 8.5 to ensure the rate of inhibition (confirmed in later experiments in which
that the predominant form of vanadate was the monomeric a 2.5-hr preincubation was used), suggesting that a direct re-
HV042, rather than the various polymeric species that are action between HMM and Vi did not occur. The possible role
favored at lower pH (13). Fig. 1 (curve A) shows that the of ADP as an obligatory cofactor for inhibition was evaluated
ATPase assay was linear for at least 3 hr, indicating that the by the addition of excess ADP (equivalent to total hydrolysis
enzyme was stable under assay conditions. In the presence of of the ATP) to the assay. If this hypothesis were correct, the
0.5 mM V1 (curve C), a slight initial inhibition was followed by ADP should have produced immediate inhibition of about 80%.
a progressive inhibition of more than 80% over a period of 3 hr. Curves F and G show that the excess ADP produced only a
The first-order plot (Fig. 1 inset) shows that the initial inhibition nominal acceleration of inhibition by Vi, whereas ADP alone
was about 16% and the observed rate constant (k0b1) for the (curve B) had no effect. Thus, the slow inhibition could not be
progressive inhibition was 1.4 X 10-4 S-1, corresponding to an explained by the accumulation of ADP. Neither could this in-
apparent second-order rate constant (kapp) of 0.28 M-1 s-1. The hibition be attributed to a slow reaction between ATP and Vi,
slow onset of inhibition was atypical of a simple reversible in- because 1-hr preincubation of ATP and Vi produced no en-
teraction between enzyme and inhibitor, suggesting that a re- hancement of the inhibition (curve E).
action was occurring in addition to the hydrolysis of ATP. Three The nature of the reaction was clarified by the finding that
possible explanations for this type of behavior were considered: preincubation of HMM with both ADP + Vi resulted in an
First, a slow reaction between HMM and Vi might be respon- immediate inhibition of 50%, which approached 100% after
sible for inactivation of the ATPase during the assay. Products 3 hr of assay (curve H). This suggested that ADP was, indeed,
of ATP hydrolysis might also play an obligatory role in inhi- a cofactor for inhibition although its role was somewhat ob-
bition by V1, so that a requisite amount of ATP needed to be scured by the ATP in the assay (see Discussion). Although the
hydrolyzed before appreciable inhibition could occur. Finally, synergistic effect of ADP provided insight into the nature of
a slow reaction between ATP and V1 might lead to formation the inhibitory complex, it left unexplained the fact that inhi-
of a new chemical species (e.g. a vanadium-ATP complex) bition by Vi was still unexpectedly slow, even when excess ADP
which was the true inhibitor, rather than Vi. These possibilities was present (t1/2 ; 5 min). It was therefore necessary to con-
were evaluated by carrying out ATPase assays in various ways sider the possibility that the true inhibitory species might be
(Fig. 1). formed by a slow interaction between ADP and Vi. Preincu-
The possibility of reaction between HMM and Vi was ex- bation of ADP and Vi was carried out for 1 hr prior to incuba-
2622 Biochemistry: Goodno Proc. Natl. Acad. Sci. USA 76 (1979)
tion with HMM. The assay (curve I) showed no acceleration in
the rate of inhibition, indicating that the slowness of inhibition
was not due to a slow reaction between ADP and Vi. The pos-
sibility that a new inhibitory species was formed by an en-
zyme-catalyzed side reaction was also evaluated. The accu-
mulation of such an inhibitory species should have been re-
flected by the ability of previously inhibited HMM to accelerate
the inhibition of fresh HMM. This was tested by incubation of
HMM with ADP + Vi until the ATPase was almost completely
inhibited. Fresh HMM was then assayed in the presence and
absence of the inhibited HMM (Fig. 2). Although the inhibited
HMM actually slowed the inhibition of fresh HMM (curves A
and B), it was found that the inhibited HMM had a slight re-
sidual ATPase activity (curve C). When the appropriate cor-
cO 4-a
C
rection was made, the rates of inhibition in both assays were 0
almost indistinguishable, leading to the conclusion that either 0)
a new inhibitory species was formed but not released from the C.
active site, or it was not formed at all. C
0X
2S-
The simplest explanation of the inhibition results, then, () -

seemed to be that a ternary complex formed slowly between

HMM, ADP, and Vi. Such a slowly formed complex would have
to be relatively stable in order to be inhibitory at all. Thus,
HMM was subjected to prolonged incubation with stoichio-
metric [a-32P]ADP and excess V1, followed (within 2 hr of the
end of the reaction) by gel filtration of the products and analysis
for bound ADP + Vi. Fig. 3A shows that ADP and V1 coeluted 4-
with the HMM in a mole ratio of 0.82 ADP and 0.92 Vi per mol
of ATPase sites. When ADP and Vi were separately incubated
with HMM (Fig. 3 B and C), only small amounts (about 0.15
mol per mol of sites) were associated with the HMM. Moreover, 0 -Xt '\-11
the HMM that was treated with ADP and Vi separately retained 0 10 20 30
100% of the untreated control ATPase activity, whereas the Fraction
HMM treated with both retained only 11% activity, which was FIG. 3. Gel filtration of M-ADP.Vi complex. HMM (52 jIM) in
comparable with the fraction of unmodified ATPase sites. Thus, buffer A was preincubated 10 min at 250 with 52 jM la-32P]ATP to
the inhibition of HMM by V1 was due to the formation of a effect quantitative conversion to [a-32P]ADP. Vi (200 jiM) was then
stable complex with the composition M-ADP-Vi. Prolonged added, and the reaction was allowed to proceed for 2.5 hr at 250C. An
incubation of M, ADP, and Vi in the ratio of 1:1:1 produced aliquot of the reaction mixture (0.75 ml) was applied to a 1.5 X 22 cm
similar results. column of Sephadex G-25 and eluted at 0.5 ml/min with buffer A
The properties of this complex were investigated by several (1.5-ml fractions). Column fractions were analyzed for: 0, protein
(A 2s); x, ADP (scintillation counting); and, A, Vi (colorimetric assay).
(A) Elution profile of complete reaction mixture containing HMM,
ADP, and V1. (B) Elution profile of control containing only HMM and
ADP. (C) Elution profile of control containing only HMM and V1. V1
0.3- eluted in fractions 55-90. See Discussion for a possible explanation
of early V1 elution in A.

methods: Additional procedures were evaluated for isolation

0.2 of the complex, and chromatography on Dowex-1 was found
E to be an effective alternative to gel filtration. It was observed
that the apparent Vi content of the isolated complex, as mea-
sured by the calorimetric procedure, was dependent upon the
structural integrity of the enzyme. Measurements in the pres-
ence and absence of 1% NaDodSO4 revealed that 0.85-0.97 mol
C of Vi was released per mol of sites when the complex was de-
natured. This indicated that the binding of Vi to native HMM
was sufficiently strong to sequester it from reaction with the
0 ~~~20
Time, min
40 60 metal indicator dye. The specificity of this binding was sug-
gested by the fact that HMM failed to sequester V1 when ADP
FIG. 2. Evidence against synthesis of a new inhibitor. The inhi- was omitted from the reaction mixture. Additional evidence
bition of HMM (6.9 !LM sites) by 1 mM ADP and 1 mM Vi was studied of this specificity was found in the steady-state fluorescence
in the absence of previously inhibited HMM (curve A) and in its
presence (curve B). Assay conditions were otherwise identical to those spectrum of the complex, which was found to be indistin-
described in the legend to Fig. 1. Inhibited HMM was prepared by guishable from that of HMM alone. When ATP was added to
preincubation of HMM (6.9 ,gM sites) with 1 mM ADP and 1 mM Vi the complex, no fluorescence enhancement occurred, although
for 10 min at 250C. The residual ATPase activity of an aliquot of the controls of HMM + ATP, HMM + ATP + Vi, and HMM +
inhibited HMM (curve C) was determined under the same conditions ADP + ATP all showed 16-17% enhancement, comparable
used for curve A. To another aliquot, fresh, concentrated HMM was with that reported by Werber et al. (15). Thus, the reduction
added to 6.9 jM, and another assay was carried out (curve B). When
the residual activity (curve C) was subtracted from curve B, progress of fluorescence enhancement was additional evidence of the
curves A and B became practically identical. formation of the M-ADP-Vi complex. The composition of the
 Biochemistry: Goodno Proc. Natl. Acad. Sci. USA 76 (1979) 2623

Table 1. Composition of stable M-ADP.Vi complex DISCUSSION

Method ADP/M_ * Vi/M Vaaa+e ion is an effective inhibitor of the myosin ATPase.
Gel filtration 0.83 0.91 Difring the assay, when-ATP is present, however, the onset of
Dowex-1 chromatography 0.92 1.08 inhibition is slow (t1/2 t 1.5 hr with 0.5 mM Vi). Although
Sequestered V 0.95 pretreatment of HMM with V1 has no effect, pretreatment with
Vi + ADP causes much more rapid inhibition than that ob-
The complex was prepared and isolated by gel filtration as de- served during the assay. These observations are explained by
scribed in the legend to Fig. 3 ([a-32PJADP was used). For chroma- the fact that myosin forms a stable, inactive, complex with the
tography on Dowex-1, the complex was prepared in the same manner,
except that [3H]ADP was used, and chromatography was carried out composition M-ADP-Vi. In the presence of ATP, formation of
on a 0.5 X 2 cm column of Dowex-1 after the addition of a 30-fold M-ADP-Vj is slow, whereas in the presence of ADP formation
excess of unlabeled ATP. Sequestered vanadium was determined from is about 35-fold faster. The slowness of formation of the in-
the difference in vanadium content of native and NaDodSO4-dena- hibitory complex under assay conditions is probably the reason
tured samples of the complex, as measured with the metallochromic that the preliminary studies of Josephson and Cantley (5) and
dye PAR. Gibbons et al. (8) failed to detect inhibition of myosin by Vi.
One of the principal questions regarding this inhibition is
complex, as determined by several methods, is summarized in whether Vi is the true inhibitor. There is evidence that Vi may
Table 1. complex with the ribose (1) and phosphate (16) moieties of
Preliminary studies of the rate of formation of the M-ADP-Vi nucleotides. Possible, though inconclusive support of this notion
complex were made by determination of the rate at which V1 is found in the gel filtration studies (Fig. 3), which show that
was sequestered (data not shown). Under pseudo-first-order Vi (free of HMM) partially coelutes with ADP ahead of the
conditions for Vi (100 AM Vii, 20 AM ATPase sites) the rate usual Vi elution peak. If a complex of ADP and Vi were essential
constant was a hyperbolic function of ADP concentration, with for inhibition, the concentrations of ADP and Vi would have
a plateau attained near stoichiometric ADP. In the presence of parallel effects on the rate of formation of the enzyme-inhibitor
excess ADP (500MM ADP, 100MM Vi, and 5 MM ATPase sites) complex. When the concentration dependence was examined,
the first-order plot was linear through about 70% modification, however, it was found that the rate approached a plateau at a
giving kobs of 1.1 X 1o-3 S-I (kapp = I1 M-' s-1). Under these stoichiometric ratio of ADP and myosin sites, whereas the
conditions inhibition of ATPase activity occurred with kSbs of variation with Vi did not plateau even at a 10-fold greater
about 1.6 X 10-3 sol, which was within the experimental error stoichiometric level. Moreover, preincubation of the Vi with
for the rate of modification. The rate constant for inhibition ADP had no effect on the rate of ATPase inhibition. These re-
increased linearly with low concentrations of Vi but began to sults indicate that a complex of ADP and Vi is not the inhibitory
level off above 300MM. At 400 MM, the highest V1 concentration species. It is also unlikely that a polymeric form of Vi is the in-
examined, a kobs of 5.7 X 10-3 s-l was obtained. hibitor, because virtually complete inhibition of the ATPase
Dissociation of M-ADP-Vi was examined by fluorescence, and is produced by incorporation of a single vanadium atom per
no dissociation was found after 2.5 hr (no fluorescence en- active site. Therefore, free V1 is the most plausible inhibitor.
hancement with ATP). Dissociation was also examined by ex- The simplest mechanism adequate to explain the incorpo-
tensive dialysis of the complex (Fig. 4). Approximately 20% of ration of ADP and Vi into a stable ternary complex consists of
the Vi was lost with a t 1/2 of about 1 day, and a plateau of about equilibrium binding followed by a slow isomerization:
80% incorporation was approached after 2 days, suggesting that
about 80% was retained for a considerably longer period of K, K2
time. The control retained only 10% stoichiometric Vi. Recovery M-ADP
of the ATPase activity was examined by incubation of the
complex with the reducing agents 2-mercaptoethanol, di- M M-ADP-Vi Mt-ADP-Vi
thiothreitol, and sodium ascorbate for periods ranging up to 25
hr. No reactivation was found, however. M-Vi
K3 K4
11 In this mechanism the formation of M-ADP-Vj is rapid and
reversible, but the isomerization (k5) is essentially irreversible,
so that Mt-ADP-Vi is the stable complex (t indicates the inactive
product of the isomerization). Preliminary evidence of a rapidly
formed complex is found in the small, immediate inhibition that
c
01) Vi produces in the ATPase assay (Fig. 1). Several types of evi-
j1 dence support the hypothesis that the stable complex is formed
by an isomerization: The Vi in the complex is sequestered, and
the fluorescence enhancement produced by ADP is quenched.
Moreover, the rate of sequestration of Vi (kapp = 11 M-' s')
is unusually low in comparison with the rates of simple pro-
tein-ligand association reactions, which have second-order rate
Time, days constants in the vicinity of 107 M-1 s-1 (17, 18). Because the
FIG. 4. Dialysis of M-ADP-Vi complex. HMM (69 ,M) was in- reaction approaches completion even at a stoichiometric ratio
activated by a 10-min incubation at 250C with 0.5 mM ADP and 5 mM of M, ADP, and Vi, it is unlikely that this slowness is due to
V1 in buffer A. The M-ADP-Vi complex was purified on a 0.5 X 5 cm unfavorable equilibria in steps 1-4. It is more plausible that an
column of Dowex-1 and found to have a Vi to site ratio of 0.9. Dialysis intrinsically slow isomerization step occurs after the reversible
of 10 yM M-ADP-Vi was carried out at 40C against 50 vol of buffer A formation of a M-ADP-Vj complex.
(changed daily). Aliquots (1 ml) were taken for vanadium analysis.
The control, which consisted of 10 yM HMM + 10 ,M V;, was di- The rate of reaction by this mechanism is given by k5[M.
alyzed in the same way. ADP-Vi], in which [M-ADP-Vi] is determined by the various
2624 Biochemistry: Goodno Proc. Natl. Acad. Sci. USA 76 (1979)
equilibria. Because this is a first-order reaction, the kapp, which stabilized at the myosin active site. Because ADP is required
is an extrapolated quantity, loses its meaning; and only the kobs for Vi incorporation, it is further possible that Vi incorporation
is significant. Under conditions that appear to be near half- involves the formation of a binary transition-state analog at the
saturation for M-ADP-Vi (excess ADP and 0.4 mM Vi), kob, is active site (24, 29). This type of mechanism has been suggested
about 6 X 10-3 s-1. Thus, k5 may be as low as 0.01 s-1, which by Milner-White and Watts (30) to explain the anomalous in-
is reasonable for a protein conformational change (19). Because hibition of creatine kinase by certain pairs of inhibitors.
it is experimentally difficult to monitor the rate of reaction Thanks are due to Drs. E. W. Taylor, F. Kezdy, K. L. Agarwal, and
above 0.3 mM Vi under the conditions used in these studies, this R. L. Van Etten for helpful advice regarding this work. Portions of this
value of k5 is only an order-of-magnitude estimate. work were supported by a postdoctoral fellowship from the Arizona
It is currently uncertain whether the association of ADP and Heart Association and a grant from the National Institutes of Health
V1 with M is truly random. Assuming so, it is possible to draw (HL 20592).
certain conclusions about the relative contributions of the
M-ADP and M-Vi complexes. The dissociation constant for 1. Lindquist, R. N., Lynn, J. L., Jr. & Lienhard, G. E. (1973) J. Am.
M-ADP is known to be in the neighborhood of 1 AiM (20,21) and Chem. Soc. 95,8762-8768.
2. Van Etten, R. L., Waymack, P. P. & Rehkop, D. M. (1974) J. Am.
the plateau in the reaction rate with stoichiometric ADP is Chem. Soc. 96,6782-6785.
consistent with such tight binding. The dependence of the rate 3. Lopez, V., Stevens, T. & Lindquist, R. N. (1976) Arch. Biochem.
on V1 concentration, however, begins to diverge from linearity Biophys. 175,31-38.
only above 0.3 mM, indicating that the dissociation constant 4. DeMaster, E. G. & Mitchell, R. A. (1973) Biochemistry 12,
for V1 is probably in the neighborhood of 0.5 mM. Because ADP 3616-3621.
and V1 compete for free myosin sites, it is apparent that M-ADP 5. Josephson, L. & Cantley, L. C., Jr. (1977) Biochemistry 16,
will be the dominant binary complex so long as ADP is present 4572-4578.
above a stoichiometric level and Vi is below about 0.5 mM. The 6. Cantley, L. C., Jr., Josephson, L., Warner, R., Yanagisawa, M.,
Lechene, C. & Guidotti, G. (1977) J. Biol. Chem. 252, 7421-
rate of reaction, then, is given by: 7423.
7. Cantley, L. C., Jr., Cantley, L. G. & Josephson, L. (1978) J. Biol.
v=k5[M.ADP.Vi] = k5[MKADP][V
K2 +[Vi]
] Chem. 253, 7361-7368.
8. Gibbons, I. R., Cosson, M. P., Evans, J. A., Gibbons, B. H., Houck,
where K2 is the dissociation constant for M-ADP-Vi. This pre- B., Martinson, K. H., Sale, W. S. & Tang, W. J. Y. (1978) Proc.
dicts that the rate will be proportional to the concentration of Natl. Acad. Sci. USA 75,2220-2224.
M-ADP, which is in turn equal to the total active-site concen- 9. Perry, S. V. (1955) Methods Enzymol. 2,582-588.
tration when stoichiometric ADP is present. During the 10. Weeds, A. & Taylor, R. S. (1975) Nature (London) 257, 54-
steady-state of ATP hydrolysis, however, M-ADP drops to about 56.
11. Young, D. M., Himmelfarb, S. & Harrington, W. F. (1965) J. Biol.
4% of the total site concentration, because kcat is 25-fold greater Chem. 240, 2428-2436.
than the rate of ADP release (22). Under these conditions, the 12. Taussky, H. H. & Shorr, E. (1953) J. Biol. Chem. 202, 675-
rate of irreversible inhibition by V1 drops to about 3%, 685.
suggesting that M-ADP is, in fact, the main binary intermediate. 13. Pope, M. T. & Dale, B. W. (1968) Q. Rev. Chem. Soc. 22,527-
Because the experimental conditions of these studies (Vi < 0.5 548.
mM) select for the M-ADP pathway, additional studies are 14. Pribil, R. (1972) Analytical Applications of EDTA and Related
necessary to determine whether the binding of ADP and V1 is Compounds (Pergamon, Oxford), pp. 304-305.
truly random. 15. Werber, M., Szent-Gyorgyi, A. G. & Fasman, G. (1972) Bio-
It is interesting to compare the properties of Mt-ADP-Vi with chemistry 11, 2872-2883.
16. Ivakin, A. A., Kurbatova, L. D. & Voronova, E. M. (1974) Zh.
those of ATPase intermediate M**-ADP-Pi (23). Direct evi- Neorg. Khim. 19,714-718.
dence has been presented here for a slow isomerization step in 17. Gutfreund, H. (1971) Annu. Rev. Biochem. 40,315-344.
the formation of Mt-ADP-Vi from M, ADP, and V1. A similar 18. Gutfreund, H. (1975) Prog. Biophys. Mol. Biol. 29, 161-195.
mechanism has been inferred by Trentham et al. (22) on the 19. Hammes, G. G. & Schimmel, P. R. (1970) The Enzymes (Aca-
basis of an unusually low calculated rate constant for the for- demic, New York), Vol. 2, pp. 67-114.
mation of M**-ADP-Pi from M, ADP, and Pi. A notable dif- 20. Bagshaw, C. R., Eccleston, J. F., Eckstein, F., Goody, R. S.,
ference between Mt-ADP-Vi and M**-ADP-Pi is that the latter Gutfreund, H. & Trentham, D. R. (1974) Biochem J. 141,
dissociates with a t 1/2 of about 12 sec, whereas the former has 351-364.
a t 1/2 of a day or more. Because the rate of dissociation of 21. Taylor, E. W. (1979) CRC Crit. Rev. Biochem., in press.
22. Trentham, D. R., Eccleston, J. F. & Bagshaw, C. R. (1976) Q. Rev.
M**-ADP-Pi is controlled by the isomerization step M**-ADP-Pi Biophys. 9, 217-281.
- M*-ADP-Pi (22), it is plausible that the dissociation of Mt- 23. Bagshaw, C. R. & Trentham, D. R. (1974) Biochem. J. 141,
ADP-Vi is controlled by the step Mt-ADP-Vi - M-ADP-Vi. 331-349.
Thus, the difference in rates probably derives from the ability 24. Lienhard, G. E., Secemski, I. I., Koehler, K. A. & Lindquist, R.
of Vi to lock the myosin into the Mt-ADP-Vi conformation. This N. (1971) Cold Spring Harbor Symp. Quant. Biol. 36,45-51.
locking might be rationalized in terms of the ability of Vi to 25. Westheimer, F. H. (1968) Acc. Chem. Res. 1, 70-78.
form a coordination complex with nucleophilic residues at the 26. Howarth, 0. W. & Richards, R. E. (1965) J. Chem. Soc., 864-
active site. The tetrahedral vanadate ion has the capacity either 870.
27. Griffith, W. P. & Wickins, T. D. (1966) J. Chem. Soc. A,
to exchange ligands or to accept a fifth ligand (as in crystalline 1087-1090.
metavanadates) to form a trigonal bipyramidal complex (13), 28. Murmann, R. K. (1977) Inorg. Chem. 16,46-51.
which resembles the transition state for phosphoryl transfer (2, 29. Wolfenden, R. (1972) Acc. Chem. Res. 5, 10-18.
24, 25). Although there is no evidence of a five-coordinate V1 30. Milner-White, E. J. & Watts, D. C. (1971) Biochem. J. 122,
species in free solution (13, 26-28), such a complex might be 727-740.