TWO FUNCTIONAL STATES O F S R ATPASE
Res. Commun. 64, 701-707. Fink, A. L., and Bender, M. L. (1969), Biochemistry 8, 5109-5118. Fink, A. L., and Good, N. E. (1974), Biochem. Biophys. Res. Commun. 58, 126- 13 1. Fink, A. L., McGarraugh, G., and Farzami, B. (1976), J. Biol. Chem. (submitted). Fink, A. L., and Wildi, E. (1974), J . Biol. Chem. 249, 6087-6089. Glazer, A. N., and Smith, E. L. (1971), Enzymes 3, 502546. Hamaguchi, K. (1964), J. Biochem. (Tokyo) 56, 441-449. Henry, A. C., and Kirsch, J. F. (1967), Biochemistry 6 , 3536-3544. Herskovits, T. T. (1967), Methods in Enzymol. 11, 748775. Holloway, M. R., and Hardman, M. J. (1973), Eur. J . Biochem. 32, 537-546. Hui Bon Hoa, G., and Douzou, P. (1973), J. Biol. Chem. 248,
4649-4654, Inagami, T., and Sturtevant, J, M. (1960), Biochim. Biophys. Acta 38, 64-79. Kasserra, H. P., and Laidler, K. J. (1970), Can. J. Chem. 48, 1793-1802. Lake, A. W., and Lowe, G. (1966), Biochem. J . 101, 402410. Lowe, G., and Williams, A. (1965), Biochem. J . 96, 199204. Lukton, A., Donahue, H., and Bettleheim, F. A. (1961), Nature (London) 191, 565-567. Mares-Guia, M., and Figueiredo, A. F. S. (1972), Biochemistry 11, 2091-2098. Pohl, F. M. (1968), Eur. J. Biochem. 7 , 146-152. Singer, S. J. (1 962), Adu. Protein Chem. 17, 1-68. Travers, F., and Douzou, P. (1974), Biochimie 56, 509514. Vratsanos, S. (1960), Arch. Biochem. Biophys. 90, 132138.
Two Functional States of Sarcoplasmic Reticulum ATPase? Giuseppe Inesi,* Joel A. Cohen, and Carol R. Coan
ABSTRACT: The “total” ATPase activity of rabbit sarcoplasmic reticulum (SR) vesicles includes a Ca2+-independent component (“basic”) and a Ca2+-dependent component (“extra”). Only the “extra” ATPase is coupled to Ca2+ transport. These activities can be measured under conditions in which the observed rates approximate maximal velocities. The “basic” ATPase is predominant in one of the various SR fractions obtained by prolonged density-gradient centrifugation of SR preparations already purified by repeated differential centrifugations and extractions a t high ionic strength. This fraction (low density, high cholesterol) has a protein composition nearly identical with that of other SR fractions in which the “extra” ATPase is predominant. In these other fractions the ratio of “extra” to “basic” ATPase activities is temperature dependent, being approximately 9.0 at 40 “ C and 0.5 at 4 OC. In all the fractions and at all temperatures studied, similar steady-state levels of phosphorylated SR protein are obtained in the presence of A T P and Ca2+. Furthermore, in
’
Vesicular fragments of sarcoplasmic reticulum (SR) provide an isolated membrane system specifically differentiated for ATP-dependent calcium transport (Ebashi, 1964; Hasselbach, 1964; Weber, 1966). In the absence of Ca2+,rabbit SR vesicles + From the Laboratory of Physiology and Biophysics, University of the Pacific. San Francisco. California 941 15. Received June 26. 1976. SUDported in part by the National Institutes of Health (HL-16607). the Ngtional Science Foundation (PCM 74-01870), and the Muscular Dystrophy Association. I Abbreviations used: SR, sarcoplasmic reticulum; “extra” and “basic” ATPase activity, the Ca2+-dependentand Ca2+-independentcomponents of “total” ATPase activity, respectively; Mops, morpholinopropanesulfonate; EGTA, ethylene glycol bis(@-aminoethyl ether)-N,N’- tetraacetate.
all cases the “basic” (Ca2+-independent) ATPase acquires total Ca2+ dependence upon addition of the nonionic detergent Triton X-100. This detergent also transforms the complex substrate dependence of the SR ATPase into a simple dependence, displaying a single value for the apparent K,. The experimental findings indicate that the ATPase of rabbit SR exists in two distinct functional states (El and E2), only one of which (E2) is coupled to Ca2+ transport. The E1 + E2 equilibrium is temperature-dependent and entropy-driven, indicative of its relation to the physical state of the ATPase protein in its membrane environment. The nonlinearity of Arrhenius plots of Ca2+-dependent(“extra”) ATPase activity and Ca2+ transport is explained in terms of simultaneous contributions from both the free energy of activation of enzyme catalysis and the free energy of conversion of E1 to E2. Thermal equilibrium between the two functional states is drastically altered by factors which affect membrane structure and local viscosity.
display an ATPase activity (“basic” activity) which markedly increases on addition of Ca2+ (“total” activity). The difference between “total” and “basic” activities is known as the “extra” or Ca2+-dependent ATPase. Only this “extra” ATPase is coupled to Ca2+ transport (Hasselbach, 1964). We now find that the ratio of Ca2+-dependent to Ca2+independent activities varies in SR fractions of different densities but identical protein composition. Furthermore, this ratio is highly temperature dependent. In all cases, Ca2+-independent ATPase acquires Ca2+ dependence in the presence of the detergent Triton X-100. These and other experiments described below indicate that the ATPase of rabbit S R resides in thermal equilibrium between two distinct functional states, only one state being coupled to Ca2+ transport. Conversion of BIOCHEMISTRY,
VOL.
15,
NO.
24, 1976
5293
INESI. COHEN. A N D COAN
were filtered through HA 0.45-p Millipore filters for separation of the S R vesicles from the medium. The residual calcium in the medium was then measured in a radioactivity scintillation counter. Control samples were made in which SR, but no ATP, was present. The pH of the reaction mixture was adjusted to 6.8 for each temperature. Steady-state levels of phosphorylated enzyme intermediate (ENP) were measured in reaction mixtures containing 20 mM Mops a t pH 6.8,80 mM KCI, 5 mM MgC12,O.l mM CaCI2, 0.5 mM [ Y - ~ ~ P I A Tand P , 0.3-0.4 mg of S R protein/ml. The reaction was quenched with 0.5 N HCI. Mixing of the reagents and quenching of the mixture were accomplished with a Durrum multimixing apparatus, allowing a reaction time of 70-100 ms. The quenched samples were further diluted with I N perchloric acid and washed repeatedly by centrifugation and resuspension. The final sediment was dissolved in hyamine and used for measurement of radioactivity.
CBP CA
I 2 5 4
Results
I’ltiURE I: EleclrophoreticseparationofSR proteinssolubilizedin sodium dodecylsulfate.Gels I , 2.3.and4carrespand tof~~actiansS,.S2,S,.and S4 obtained by centrifugation of two different S R preparations on a multistep gradient of 29.32.39, and 43% sucrose. The bands correspond 10 ATPase. calcium-binding protein (CBP). and calsequestrin (CA) as indicated
ATPase from one state to the other involves changes in the conformation of the protein andjor its membrane environment. Materials and Methods Vesicular fragments of sarcoplasmic reticulum membrane (SR) were prepared from white muscle of rabbit hind legs by homogenization and differential pelleting as described previously (Eletr and Inesi, 1972). Further purification was then carried out on a discontinuous sucrose gradient, collecting the fraction sedimenting between 26 and 29% sucrose after a 3-h centrifugation at 76 OOOg. In some experiments the S R preparation obtained by differential pelleting was subdivided by a 20-h centrifugation a t 68 OOOg in a multiple-step, sucrose gradient (20. 32, 39, and 43%), as described by Meissner (1975). Protein concentration was determined with the biuret reagent standardized with micro Kjeldahl nitrogen determinations. Solubilization of S R in sodium dodecyl sulfate and gel electrophoresis were carried out as previously described (Inesi and Scales, 1974). ATPase activity was measured in the presence of 20 mM morpholinopropanesulfonate (Mops), EO mM KCI, 2 mM potassium ethylene glycol bis(Saminoethy1 ether)-N.N‘tetraacetate (EGTA), 5 mM potassium oxalate, and 0.18 mg of S R protein/ml. CaCl2 (2 mM) was added when “total” ATPase, as opposed to “basic” ATPase, was measured. The reaction was started by the addition of 5 mM M p A T P . The pH of the reaction mixture was adjusted to 6.8 at each temperature. Samples were taken at different times and quenched in equal volumes of 10% trichloroacetic acid. Inorganic phosphate was measured by the method of Fiske and Subbarow (1925). Calcium transport was measured in the presence of 20 mM Mops, 80 mM KCI, 0.1 mM EGTA, 0.1 mM 45Ca.CaC12,and 22-25 pg of S R protein/ml. The reaction was started by the addition of 2.5 mM M p A T P . At appropriate times samples
5294
B I O C H E M I S T R Y ,V O L . 1 5 . N O .
24, 1976
Properties of SR Fractions Obtained by Multiple-Step. Sucrose-Gradient Centrifugation. Rabbit S R preparations obtained by differential pelleting were further separated into four fractions by multiple-step, sucrose-gradient centrifugation. The four fractions, collected a t the interface between 29 and 32% sucrose (SI), and within the 32 (S2). 39 (S3).and 43% (S4) sucrose layers, correspond to 2, 7 I , 16, and 11% of the total protein in the original preparations, respectively. The protein composition of the four SR fractions, as revealed by sodium dodecyl sulfate gel electrophoresis, is shown in Figure I. A very heavy band corresponding to the ATPase protein (MacLennan, 1970; Martonosi and Halpin, 1971; Meissner and Fleischer, 1971) is noticed in all fractions, and the patterns of fractions SI,S2, and S3 are nearly identical. Fraction S4 contains a greater proportion of the protein “calsequestrin”, originally described by MacLennan and Wong (1971 ). It should be pointed out that the S R preparation must be purified by multiple differential centrifugation and extractions in high ionicstrength (Eletr and Inesi, 1972) previous to multistep centrifugation, to avoid the presence of contaminating proteins in fraction SI. Functional characterization of the four fractions was performed in the presence of oxalate and substrate concentrations saturating both “extra” and “basic” ATPase. In these conditions S R vesicles sustain Ca2+ transport and A T P hydrolysis a t steady-state rates close to maximal velocity. The Ca2+ transport and ATPase activities of S R in the four fractions are shown in Table 1. As reported by Meissner (1975), fractions Sz and S3are the most active, and calcium transport is coupled to Caz+-dependent ATPase by factor of approximately two. Fraction Sq. on the other hand, appears to be partially uncoupled. Fraction SI,although accounting for only a minor percentage of the total SR preparation, is the most interesting with regard to Ca2+-independent ATPase. This fraction displays mostly “basic” ATPase. However, its calcium transport activity is still coupled to its low “extra” ATPase by a factor of approximately two. It evidently corresponds to the low-density S R fraction described by Flaherty et al. (1975), having high “basic” ATPase and a high cholesterol content. The Effect of Detergent Solubilization. The Ca2+-independent ATPase of fraction SI becomes Ca2+ dependent in the presence of the detergent Triton X-100. Table I shows that, after addition of Triton X-100, the “total” ATPase activity of fraction SI remains approximately the same; however, Ca2+
TWO FUNCTIONAL STATES OF S R ATPASE ~~
TABLE I:
~
Steady-State Activity (rnmol min-’ (g of protein)-’) of Various SR Fractions.u
Fraction
s4
+ +
SI Triton S; Triton S , a t 4 OC S?a t 37 OC
Ca2+ Transport
“Total” ATPase
“Basic” ATPase
“Extra” ATPase
“Extra”/ “Total”
0.44 f 0.10 1.14 f 0.10 1.30 f 0.25 0.63 f 0.15
0.78 f 0.20 0.80 f 0.25 0.86 f 0.30 0.82 f 0.20 0.88 f 0.10 1.09 f 0.15 0.08 f 0.01 3.89 f 0.45
0.60 f 0.15 0.26 f 0.08 0.24 f 0.06 0.21 f 0.06 0.07 f 0.01 0.06 f 0.01 0.05 f 0.01 0.42 f 0.03
0.18 0.54 0.62 0.61 0.8 1 1.03 0.03 3.47
0.23 0.68 0.72 0.74 0.92 0.94 0.38 0.89
0.08 f 0.02 7.82 f 0.35 ~~~~
~
Fractions SI, S2, S3, and S4 were obtained by multistep sucrose-gradient centrifugation of purified S R preparations. Reactions were carried out at 20 OC, except where specified. The Triton X-100 concentration was 0.1% in the final reaction mixtures. Media and methods for determination of activities a r e described in the text. “Total” and “Basic” refer to A T P hydrolysis catalyzed in the presence and in the absence of C a 2 + ,respectively. “Extra” is the difference between “Total” and “Basic”. 0
+21 +I
-1
’
I
”?$
L
3.1
3.3
0 2 t
3.5
IO F I G U R E 2: Semilogarithmic plots of V/T vs. 1 / T for ATP hydrolysis catalyzed by SR vesicles obtained by differential pelleting and further purification by a 3-h centrifugation at 76 OOOg and sedimentation between 26 and 29% sucrose. Reaction mixture described in the text. V is the near-maximal velocity for ATP hydrolysis measured in mmol min-’ (g of SR protein)-’. The form log ( V / T )vs. I / T is used here in adherence to eq 4 in the text. ( 0 )“Total” ATPase velocity; dots are experimental data; solid line is best linear regression fit. (9) “Basic” ATPasevelocity; dots are experimental data; dashed line is calculated curve from eq 7a in the text. (0)“Extra” ATPase velocity; dots are data points calculated from “total” minus “basic” velocities; dashed line is calculated curve from eq 7b in the text.
is now necessary for enzyme activation. Fraction S3 is similarly affected. It should be pointed out that, independent of the presence or the absence of detergent, neither Na+K+ stimulated nor mitochondrial ATPase activity was found in any of the SR fractions. The Effect of Temperature. The presence of “basic” ATPase activity is not an exclusive feature of fraction SI. This can be demonstrated in rabbit SR preparations obtained by differential pelleting and further purified by a 3-h sucrose centrifugation and collection of the vesicles sedimenting between 26 and 29% sucrose. In this preparation, which resembles in all respects the fractions S2 and S3, the ratio between Ca2+dependent and Ca2+-independent ATPase activities is reduced simply by lowering the temperature of the reaction mixture. For example, a t 37 “ C the Ca2+-dependent ATPase accounts
20
30
40
50
T W Fraction of “total” ATPase activity which is “extra”, as a function of temperature. SR and reaction mixture as for Figure 2. V approximates maximal velocity. The velocity ratio also indicates the fraction of total ATPase enzyme residing in the Ca2+-dependent functional state since, from eq 2 and 3 of the text, Vmaxextra/Vmaxtota’ = P z l / [Eo]. F I G U R E 3:
for -89% of the total ATPase activity, while at 4 O C it accounts only for -38% (Table I). The temperature dependences of “total”, Ca2+-independent, and Ca2+-dependent ATPase activities within a 4-45 O C temperature range are shown in Figure 2. While “total” ATPase generates a semilogarithmic plot of near-maximal velocity V/T vs. 1 / T having a constant slope, the Ca2+-independent and Ca2+-dependent activities generate nonlinear plots. It is apparent that, as the temperature is raised, the percentage of total activity corresponding to “extra” ATPase is increased, as shown in Figure 3. We have previously demonstrated that, over this temperature range, only the “extra” ATPase is coupled to calcium transport, with a calcium-toA T P molar ratio of approximately two (Davis et al., 1976). Nonlinear Arrhenius plots for the Ca2+-dependent ATPase of SR have been obtained repeatedly when the experimental temperatures were extended over a sufficiently wide range (Deamer, 1973; Inesi et al., 1973; Lee et al., 1974; Madeira et al., 1974). Moreover, we find that the high Ca*+-independent activity observed for this fraction at low temperatures is converted to BIOCHEMISTRY, VOL.
15,
NO.
24, 1 9 7 6
5295
I Y E S I . C O H E N . .AND C O A N
‘\
o 2
L 3G
[ A T pj
(u)
.
Dependence of ATPase activity on ATP concentration at 25 and 10 “C. S R and reaction mixture as for Figure 2. with the addition of 3 ,,,M p ~ o s p ~ o e n o ~ p y r u vand a t e 100 pg/ml ofpyruvate kinase for ATP regeneration. MgCI2, 2 mM, in all samples, These curves were obtained ATPase by direct o~orthophosphateliberation: g‘total~3 activity; (0) “basic” ATPase activity. F I G U R E 4:
Ca2+-dependent ATPase by the addition of Triton X- 100, similarly to the case of fraction SI at 20 “C. Thus, the detergent effect is not unique to S Ibut bestows Ca2+ dependence to the enzyme activity in all cases. A T P Dependence. The dependence of “basic” and “total” ATPase un A T P substrate concentration, a t two different temperatures, is shown in Figure 4. As previously reported (Weber et al., 1966; Inesi et al., 1967), the “basic” ATPase requires high A T P for activation. Lineweaver-Burk plots of the data shown in Figure 4 yield apparent K , values for the “basic” ATPase of 2 X IOv4 M a t 25 “C and 3 X lop4 M a t I O “C. The substrate dependence of the “total” ATPase is complex and, while an apparent saturation is obtained at approximately 0.1 m M ATP, further activation occurs at higher A T P concentrations (Figure 4). Accordingly, Lineweaver-Burk plots of these data do not yield simple straight lines. The activation produced by high A T P concentrations has been reported previously (Yamamoto and Tonomura, 1967; Inesi et al., 1967; de Meis and de Mello, 1973). The experiments shown in Figure 4 indicate that the substrate dependence of S R ATPase is not strongly affected by temperature and, thus, the A T P concentrations used in our experiments are sufficient to saturate both “basic” and “total” ATPase over our temperature range. It is also demonstrated that the (relatively) increased Ca*+-independent activity observed a t low temperatures has a substrate dependence very similar to that of the “basic” ATPase a t 25 “C. The effect of Triton X-100 on the A T P dependence of the S R ATPase is shown in Figure 5. The detergent transforms the complex substrate dependence of the “total” SR ATPase (Figure 4) into a simple dependence (Figure 5) displaying a single value of 1.2 X M for the apparent K,. Figures 4 and 5 are consistent with the conversions of Ca2+ dependence described above and indicate that the functional behavior of the S R ATPase is intimately related to physical factors of membrane structure. Phosphorylated Enzyme Intermediate. It is known that the catalytic mechanism of SR ATPase in the presence of Ca2+ includes a phosphorylated enzyme intermediate (Yamamoto and Tonomura, 1967; Makinose, 1969). Formation of the intermediate is very rapid, and steady-state levels of the intermediate are reached within a few milliseconds (Froehlich and Taylor, 1975). In our experiments we found these levels to be 3.1 f 0.8 pmollg of protein, with no significant difference
5296
BIOCHEMISTRY. VOL.
1 5 , NO. 24, 1916
J
TRITON
I
i
z .21
J
2 50 c
i
/
J.---5
-4
;OG [ATP]
-3 (M)
N G L R E 5: Effect of Triton x-100 on ATPdependenceof ATPaseactivity. SR and reaction mixture as for Figure 4, with the addition of 0.1% (v/v) Triton X- 100: ( 0 )“total” ATPase activity; (0) “basic” ATPase activi-
tY
among the various S R fractions, and independent of the reaction temperature. Discussion Purified S R vesicles retain a Ca2+-independent ATPase activity in addition to the Ca2+-dependent activity which is coupled to Ca2+ transport. The Ca*+-independent ATPase is prevalent in a minor (“light”) fraction of SR vesicles displaying a protein electrophoretic pattern identical with that of purified SR. All other fractions catalyze comparable rates of Ca2+independent and Ca*+-dependent activities a t low temperatures (5- 10 “C) while, as the temperature is raised, the ATPase activity gradually acquires Ca2+ dependence, becoming mostly Ca2+ dependent a t 37 “C. On the other hand, the steady-state levels of phosphorylated enzyme intermediate are similar for all fractions of purified SR, at all temperatures. Furthermore, in the presence of the detergent Triton X-100, the Ca*+-independent ATPase becomes Ca2+ dependent for all fractions irrespective of the temperature. This detergent also transforms the complex substrate dependence of SR ATPase to a simple substrate dependence displaying a single K , value. The experimental findings reported here indicate that the functional state of SR ATPase is dependent on physical factors related to membrane structure (see also Davis et al., 1976). These findings may be explained by postulating an enzyme population which is thermally equilibrated between a state (E*) coupled to Ca2+ transport and a state (El) which is uncoupled from Ca2+ transport. Catalysis of A T P hydrolysis by E2 proceeds through a mechanism including phosphorylation of the enzyme as an intermediate (Ca2+-dependent) step. On the contrary, phosphorylation and hydrolysis are uncoupled in the El state. In this state the phosphorylated enzyme intermediate is formed in the presence of Ca2+,but its subsequent hydrolysis is inhibited. However, in native SR vesicles at sufficiently high substrate concentrations, ATPase in state E1 is able to catalyze ATP hydrolysis independently of the presence of Ca*+ and enzyme phosphorylation. It is apparent that the activities catalyzed by E2 and E l correspond to “extra” and “basic” ATP hydrolysis, respectively. This scheme is consistent with the facts that (a) hydrolysis by the “basic” enzyme is not Ca2+ dependent, (b) Ca2+ transport is coupled only to the “extra” enzyme hydrolytic activity, (c) phosphorylated enzyme intermediate does not occur in the absence of Ca*+, (d) steady-state levels of phosphorylated enzyme intermediate, in the presence of Ca2+,are nearly identical at all temperatures, (e) ATP binding
TWO FUNCTIONAL STATES OF SR ATPASE
to the ATPase is not Ca2+ dependent (Inesi and Almendares, 1968; Meissner, 1973). According to this scheme, the effects of detergent and increasing temperature can be described as conversions of enzyme from the state El to the state E2. At saturating substrate concentrations, the constant ( K E )expressing the equilibrium between the two enzyme states involves substrate-enzyme complex, rather than free enzyme. The equilibration is assumed to be very rapid and insensitive to the absence or presence (lo-+ M) of Ca2+. It is generally accepted that the hydrolysis steps are rate limiting in the SR ATPase reaction (Froehlich and Taylor, 1975). In this regard, both our detergent data (see Table I) and temperature data (see analysis below) indicate that the hydrolysis rate constants for the “basic” and “extra” channels are similar in magnitude (khb = khe kh). Accordingly, in the absence of Ca2+ the measured velocity of enzyme catalysis with saturating substrate concentrations is:
where [Eo] is the total enzyme concentration = [ATP-El] + [ATP.E2]. In the presence of Ca2+, the measured velocity of enzyme catalysis with saturating substrate concentrations is
+
VmaXtotal= khb[ATP-E~] khe[ADP-P-E2]
kh[Eo]
(2)
The “extra” maximal velocity is defined as the difference between Vmaxtotaland V,,,axbasic (Hasselbach, 1964). This definition implies that the “basic” velocity proceeds a t the same rate both in the absence and in the presence of Ca2+. This “extra” ATPase velocity is indeed a functionally meaningful parameter inasmuch as, in intact vesicles, it is related to the velocity of net Ca2+ transport by a coupling factor of two, a t temperatures varying from 4 to 37 “C (Davis et al., 1976; Table I). Hence from eq 1 and 2
Temperature Dependence of ATPase Activities. The detailed experimental data on the temperature dependence of the ATPase activities may then be analyzed in light of the model proposed above. It is shown in Figure 2 that plots of log (V/T) vs. 1 / T are linear for “total” ATPase, but nonlinear for the “basic” and “extra” activities. The linearity of the “total” activity provides further experimental indication of a similarity of the hydrolysis rate constants for the “basic” and “extra” channels which yielded eq 2 above. The temperature dependence of eq 2 can be described from transition-state theory:2
Thus plots of In ( V m a x t a t a l / Tvs.) 1 / T yield straight lines of slope = -AH*/R and intercept = In ( k ~ [ E o ] / h ) A S * / R , provided that AH*, AS*, and [Eo] are temperature-independent constants, as is apparently the case in Figure 2 . Values for AH* and AS* derived from Figure 2 , using [Eo] = 7
+
3. I
33
35
T” r K ) x l O ’ F I G U R E 6: Semilogarithmic plot of Vextra/Vbdsic vs. 1 / T in accordance with eq 6 in the text. SR and reaction mixture as for Figure 2. Dots are data points calculated from experimental values approximating maximal velocities; solid line is best linear regression fit.
kmol/g of protein, are: AH* = 18.75 f 0.5 kcal/mol and AS* = 7 f 2 cal deg-l mol-’. The separate temperature dependences of the “basic” and “extra” ATPase activities can be written in terms of AS*, AH*, and the total El and E2 enzyme concentrations, respectively, via equations analogous to eq 4. In the face of a linear semilogarithmic plot of V,axto‘al/T vs. 1 / T , a simple explanation for curvatures in the plots of log ( VmaxbdS’C/T) and vs. 1 / T (Figure 2 ) is that, unlike [Eo], neilog ( VmaxeXtra/T) ther the total El enzyme concentration nor the total E2 enzyme concentration is constant but, in fact, both vary with tempera t ~ r eInspection .~ of the curvatures in Figure 2 indicates that the total El concentration must be decreasing, and the total E2 concentration increasing, as the temperature is raised. This situation is suggestive of a temperature-induced reversible conversion between the two functional states of the enzyme: total) + E2(total) (5) having an effective equilibrium constant, under saturating conditions, K E = [E2](totai)/[E1](t0t~1). The temperature dependence of this equilibrium constant can be treated analogously to the thermal denaturation of proteins (Joly, 1965), and the “free energy of conversion” AC”l-2 defined by the Boltzmann relation K E = exp (- AGO ]-2/RT). The enthalpy and entropy of conversion are defined by AGO1 -2 = AHO1-2 - TAS” 1-2 and their values obtained from the slope and intercept of a plot of log ( VmdXextra/ VmaxbaFic) VS. 1/ T since, from eq 1 and 3
The values fitted by linear regression, as shown in Figure 6, are AHO1-2 = 12.5 f 0.5 kcal/mol and AS’l-2 = +44 f 2 cal deg-’ mol-’ (linear regression correlation. coefficient r2 = 0.924). The highly positive entropy of conversion is of interest,
~~~~
’
According to transitionstate theory, a rate constant kh IS described by kh = kBT/h exp ( - A G * / R T ) , where k~ is Boltzmann’s constant, h is Planck’s constant, R is the gas constant, T the absolute temperature, and AG* the standard-state free energy of activation. In turn, AG* = AH* - TAS*, where AH* and AS* are the standard-state enthalpy and entropy of activation, respectively.
It should be realized that, for khb= khe kh, the parameters AS* and AH* appearing in the temperature-dependent equations for Vmaxbas’C and VmdXeXtrd are identical with those appearing in eq 4 for Vmaxfotal. Thus, they are temperature independent, as indicated in the text, and cannot account for the curvatures in Figure 2. BIOCHEMISTRY, VOL.
15,
NO.
24, 1976
5297
motion of membrane phospholipids (Davis et al., 1976) has as it indicates that conversion of F1 to E2 is accompanied by been previously demonstrated. change in the conformation of the protein and/or its membrane environment. The detailed nonlinearities of the plots of log ( V,,,dxhd~lC/T) Acknowledgments and log ( Vmdxextra/ T ) vs. 1/ T (Figure 2) are now conveniently The authors acknowledge stimulating discussions with Drs. L. de Meis, A. J. Murphy, R. Sabbadini, and L. Peller. We also investigated by use of eq 1-3 and the above definitions, acknowledge very helpful suggestions by the reviewers of an whence earlier version of this manuscript. References -log [ I
+ exp
-log [ l
+ exp ( + A G o ~ - + 2 / R T ) ](7b)
( - S o l
+ l / R T ) ] (7a)
and
The temperature dependence of AGO -2 is explicitly determined by the values of A H o 1 -+2 and ASo 1 -2 obtained above in Figure 6. The log ( VmaXtotai/T) curve has also been established previously (solid line in Figure 2). Equation 7 shows the effect on the “basic” and “extra” enzyme velocities of population changes in [El] and [El] resulting from the thermally induced E1 + E2 conversion. The curves calculated from eq 7 are given by the dashed lines in Figure 2 . It should be stressed that these theoretical curves employ no adjustable parameters, i.e., are completely determined by the previously established values for A H o 1-2 and ASoI - 2 . This fit to the nonlinear experimental data underscores the validity of our proposed model and its analysis. Conclusion We have presented several kinds of data indicating that the SR ATPase exists in either of two states (El and Ez), in a thermal equilibrium. In both states a phosphorylated intermediate is formed in the presence of Ca2+. However, subsequent hydrolysis of the intermediate (“extra” activity) proceeds only in one state (E2) which is coupled to Ca2+ transport. Therefore, nonlinear Arrhenius plots of Ca2+-dependent ATPase and Ca2+ transport are explained with simultaneous contributions from both the free energy of activation of enzyme catalysis and the free energy of conversion of El to E>. It is suggested that, in native SR vesicles, the ATPase in state E , catalyzes A T P hydrolysis (“basic” activity) even in the absence of Ca2+ and independently of enzyme phosphorylation. This suggestion is consistent with the linear Arrhenius plots of “total” ATPase, but it is not necessary to explain the nonlinearity of Arrhenius plots for “extra” ATPase and Ca2+ transport as outlined above (cf. eq 3 and 7b). Thermal equilibration between the two states is drastically altered by factors which affect membrane structure and local viscosity (e.g., detergents). In fact, the phospholipid environment of the enzyme has been shown to affect the decomposition of the phosphorylated intermediate (Hidalgo et al., 1976). In addition, a relation between S R enzyme activity and thermal
5298
BIOCHEMISTRY, VOL.
1 5 , NO. 24, 1976
Davis, D. G., Inesi, G., and Gulik-Krzywicki, T . ( I 976), Biochemistry 15, 1271. Deamer. D. W. (1973), J . Biol. Chem. 248, 5477. de Meis, L., and de Mello, M. C. F. (1973), J . B i d . Chem. 248. 3691. Ebashi, S . (1964), Prog. Theor. Phys. 17, 3 5 . Eletr, S., and Inesi, G . (1972), Biochim. Biophys. Acta 282, 174. Fiske. H.. and Subbarow, Y . (1925), J . Riol. Chem. 66. 375. Flaherty, J. O., Barrett, E. .I., Bradley, D. P., and Headon, D. K. ( 1 975), Biochim. Biophys. Acta 401, 177. Froehlich, J. P., and Taylor, E. W. (1975), J . Biol. Chem. 250, 2013. Hasselbach, W. (1964), Prog. Biophys. Mol. Biol. 14, 167. Hidalgo, C., Ikemoto, N., and Gergely, J. (1976), J . Riol. Chem. 251, 4224. Inesi, G., and Almendares, J. (1968), Arch. Biochem. Biophys. 126. 733. Inesi, G.. Goodman, J.. and Watanabe, S. (1967), J . Biol. Chem. 242, 4637. Inesi. G., Millman, M., and Eletr, S. (1973), J . Mol. B i d . 81, 483. Inesi, G., and Scales, I).(1974). Biochemistry 13, 3298. Joly, M. ( 1 9 6 9 , A Physico-Chemical Approach to the Denaturation of Protein, London, Academic Press, p 192. Lee, A . A . . Birdsall, N. J. M., Metcalfe, J . C., Tson. P. A,, and Warren, G. B. (1974), Biochemistry 13, 3699. MacLennan, D. H. (1970), J . Biol. Chem. 245, 4508. MacLennan, D. H., and Wong, P. ’r. S. (1971), Proc. Natl. Acad. Sci. U . S . A . 68, 1231. Madeira, V. M. C., Antunes-Madeira. M . C., and Carvalho, A . P. (1974), Riochem. Biophys. Res. Commun. 58, 897. Makinose, M. (1969), Eur. J . Biochem. 10, 74. Martonosi, A,, and Halpin, R . A . (1971), Arch. Biochem. Biophjs. 144, 66. Ileissner, (3. ( 1 973). Biochim. Biophys. Acta 298, 906. Meissner, G . ( 1 975), Biochim. Biophys. Acta 389, 5 1. Meissner, G., and Fleischer, S. (1971), Biochim. Biophys. Acta 241, 356. Weber, A. ( 1 966), C w r . Top. Bioenerg. 1, 203. Weber. .A, Herz, R.. and Reiss. I . ( 1966). Biochem. Z . 34.5, 329. Yamamoto, T., and Tonomura, Y. (1967), J . Biochem. (Tookyoi 62, 5 5 8 .
