Role of Phospholipids in the Calcium-dependent ATPase of the Sarcoplasmic Reticulum ENZYMATIC AND ESR STUDIES WITH PHOSPHOLIPID-REPLACED MEMBRANES*

the The temperature dependence of Ca*+ -activated ATPase of these preparations clearcut differences; with DOL-enzyme there was no appreciable break in the Arrhenius plot in the 3-40” range; DPL-enzyme showed a break at 29”, and del-enzyme and sarcoplasmic reticulum one at 18”. Transition temperatures obtained ESR studies with the use of spin-labeled acid incorporated into the membranes agreed with those derived from ATPase assays. Thermodynamic

Three types of partially purified ATPase enzymes having different phospholipid contents and compositions have been prepared: (a) an enzyme whose phospholipid moiety has been replaced predominantly by dioleoyl lecithin (DOL-enzyme), with about the same phospholipid content as the original sarcoplasmic reticulum, (b) dipalmitoyl lecithin-replaced enzyme whose phospholipid content is 30% of that of DOL-enzyme (DPL-enzyme), and (c) a partially delipidated enzyme with about the same phospholipid content as DPL-enzyme but with the original sarcoplasmic reticulum phospholipid composition (del-enzyme).
The temperature dependence of Ca*+ -activated ATPase activity of these preparations showed clearcut differences; with DOL-enzyme there was no appreciable break in the Arrhenius plot in the 3-40" range; DPL-enzyme showed a break at 29", and del-enzyme and sarcoplasmic reticulum one at 18". Transition temperatures obtained from ESR studies with the use of spin-labeled stearic acid incorporated into the membranes agreed with those derived from ATPase assays. Thermodynamic analysis of the ATP hydrolysis rates shows that DPL-enzyme has considerably larger values of activation enthalpy and activation entropy below the transition temperature (29") than those of the other preparations, while all enzyme preparations show similar free energies of activation. The ESR data show that below their transition temperatures DPL-enzyme, and to a lesser degree del-enzyme, have a strongly restricted motion of their phospholipid molecules as compared with either DOL-enzyme or sarcoplasmic reticulum.
Studies on the formation and decomposition of phosphoenzyme have been carried out with the three types of ATPase preparations.
At O", the rate of inorganic phosphate liberation is 8 times lower in DPL-enzyme than in del-enzyme with little difference in the steady state level of phosphoenzyme. In DOL-enzyme, the level of phosphoenzyme and the rate of inorganic phosphate liberation are 1.8 and 3.5 times higher than the corresponding values obtained with del-enzyme. Addition of ADP to the phosphorylated intermediate of DPL-enzyme induces a fast reversal of the phosphorylation reaction. These results indicate that the physical state of the phospholipid molecules associated with the enzyme affects the decomposition of phosphoenzyme, with little effect on the phosphorylation reaction and its reversal.
Phospholipids are essential components of the Ca*+ -depend-purified ATPase enzyme and phospholipids (2-4) suggest that ent ATPase activity in the sarcoplasmic reticulum. Removal of they are also essential for the Ca*+ accumulation by the SR' the membrane-associated phospholipids by means of phos-vesicles. pholipase digestion leads to loss of the ATPase activity, while In the last few years some insight has been gained into the readdition of phospholipids restores it (1). Recent reports that mechanism by which phospholipids affect the ATPase reaction Ca*+-accumulating vesicles can be reconstituted from the and the coupled Ca*+ transport. ESR studies using spinlabeled analogues of stearic acid and spin probes attached to * This work was supported by grants from the National Institutes of the ATPase protein have shown that ATPase activity and Ca*+ Health (HL-5949 and AM-16922), the National Science Foundation (GB43484X), the American Heart Association and the Muscular transport are related to the fluidity of the membrane lipids, Phospholipids in Sarcoplasmic Reticulum Calcium ATPase 4225 motion of the phospholipid molecules by reducing the temperature inhibits ATPase activity (5). Delipidation also produces a substantial decrease in label mobility and a strong inactivation of ATPase activity; upon readdition of lipids these effects are reversed (6).
A considerable body of evidence has accumulated indicating the existence of various states of the ATPase enzyme, including one or more phosphorylated forms, in the course of the hydrolysis of ATP. In SR vesicles delipidated by means of digestion with phospholipases, both the rate and level of phosphoenzyme formation are about the same as in the intact SR, while the steady state P, liberation is considerably inhibited (1, 7). This suggests the delipidation primarily affects the decomposition of the phosphoenzyme (for an opposing view, see Ref. 8).
Recently methods have been developed for the replacement of the endogenous phospholipids of the purified ATPase with various synthetic lecithins (4). It has been shown that replacement with dipalmitoyl lecithin and dimyristoyl lecithin -phosphatidyl cholines that contain saturated hydrocarbon chains-completely inhibits ATPase activity at lower temperatures, while further replacement with dioleoyl lecithin, a phospholipid that contains an unsaturated fatty acid moiety, releases the inhibition of ATPase at low temperatures (9). The present studies have been carried out in an attempt to further clarify the relation of physical properties of the membrane and certain aspects of the ATPase reaction to the structure of the fatty acid moiety of the phospholipids.
From parallel studies of ATPase activity and mobility of spin-labeled stearic acids incorporated into ATPase enzymes containing predominantly either DOL or DPL, it appears that decreased fluidity of the lipid moiety decreases the ATPase activity. Thermodynamic analysis,indicates that low fluidity makes the transition associated with activation energetically unfavorable, which is reflected in higher activation enthalpy, at the same time that it imposes "order" on the enzyme, which is reflected in higher activation entropy. Studies on the formation and decomposition of the phosphorylated intermediate show that phospholipids affect primarily the dephosphorylation process. Thus replacement with DPL results in a strong inhibition of the rate of ATP hydrolysis at temperatures at which the mobility of the DPL molecules is strongly restricted, while there is only a small effect on the phosphorylation reaction. These results have confirmed and extended previous suggestions, originated from the results obtained with phospholipasedigested SR, concerning selective involvement of SR lipids in the decomposition of phosphoenzyme during the ATPase reaction.

Replacement
of Endogenous Phospholipids-Fragmented SR was prepared from rabbit white skeletal muscle as reported previously (IO). For the replacement of phospholipids the original method as described by Warren et al. (4) was modified as follows. Freshly prepared SR (25 to 60 mg of protein) was incubated with slow stirring at 0" for 2 hours with a sonicL 3 ?d mixture of deoxycholate and phospholipid in 3 to 6 ml of a solution containing 1 M KCl, 0.3 M sucrose, 10 mM p-mercaptoethanol, and 50 rnM Tris-HCl, pH 8.0. The weight ratio of protein, deoxycholate and phospholipid used was 1:l:l for replacement by DOL, and 1.5:1:2 for replacement by DPL. One milliliter of the resulting clear solutions was placed on top of a discontinuous sucrose gradient obtained by layering 2 ml of 15% (w/v) sucrose on 2 ml of 50% sucrose; both sucrose solutions contained 1 M KCl, 50 mM Tris-HCl, pH 8.0, and 10 rnM fl-mercaptoethanol.
The tubes were centrifuged for 17 hours at 5" in a Beckman SW 50.1 rotor at 150,000 x g. Both phospholipid-replaced preparations formed turbid bands; DOLenzyme at the interface of the two sucrose layers and DPL-enzyme in the 50% sucrose layer, between the upper and middle third. After dropwise collection from the bottom of the tubes, the phospholipidreplaced preparations were diluted 5-fold with 50 rnM Tris-maleate, pH 7.0, and were centrifuged at 4' for 60 min at 100,000 x g. The resulting pellets were homogenized in a solution of 0.3 M sucrose/20 rnM formation, the reaction was stopped by adding trichloroacetic acid to a final concentration of 6.7%; 0.5-ml fractions were filtered through Millipore filters (HA, 0.45 p pore size) previously washed with 1 ml of 0.1 mM ATP to prevent binding of ATP to the filter. The acid-denatured protein retained on the filter was washed with 5-ml portions of a solution containing 5% trichloroacetic acid and 0.2 mM NaH,PO,.
To determine the amount of inorganic phosphate liberated, centrifuged portions of the reaction mixture were treated with activated charcoal to remove the unreacted [Y-~~P]ATP, followed by filtration through surgical cotton. Portions (0.2 ml) of the filtrate were placed on filter paper strips, which were then dried and counted.
[y-3T]ATP was prepared according to Glynn and Chappell (13) (14) with 2 x 2 ml of chloroform/methanol (2/l) followed by 2 ml of chloroform/methanol (l/2). The combined extracts plus 2 ml of chloroform were shaken for 30 s with 2.0 ml of 0.01 N HCl. After brief centrifugation the aqueous phase was discarded and the organic solvents were evaporated to dryness under a stream of nitrogen.
The residue was dissolved in benzene and stored at -20". The phospholipid content was determined was used for development in the first dimension, and a mixture of chloroform, methanol, acetone, glacial acetic acid, and water (50/10/20/10/5) for development in the second dimension.
The individual phospholipid spots were located by exposing the plates to iodine vapor. After scraping the spots from the plates their phospholipid content was determined. A factor of 22.5 was used to convert milligrams of phosphate to milligrams of phospholipid. Fatty acid compositions of phospholipid replaced preparations and SR were determined by gas-liquid chromatography.
The fatty acids were transesterified to their methyl esters and analyzed in a Varian model 2700 gas liquid chromatographic apparatus. ESR measurements-N-Oxyl-4',4'-dimethyloxazolidine derivatives of stearic acid having the general formula I(m,n): were used as spin probes. They were dissolved in benzene at a concentration of 10 mM and stored at -20". Aqueous solutions of the labels were prepared by evaporating to dryness a small amount of the benzene solution prior to dissolving it in 0.025 ml of methanol to which 0.5 ml of 0.3 M sucrose/20 rnM Tris-maleate, pH 7.0, was added. For incorporation of spin labels into membranes, the spin label solution was mixed with the membrane suspension-the final concentration of methanol was less than 0.5%-by vigorous shaking at room temperature. Final concentration of protein was 5 to 10 mg/ml. The spin label was added in ratios ranging from 0.4 to 1.7 rmol/lOO mg of protein.
These values correspond to 0.6 to 2.6 spin labels per 100 molecules of membrane lipid for SR and DOL-enzyme and 2 to 9 spin labels per 100 molecules of lipid in the delipidated preparations, uiz. del-enzyme and DPLenzyme.
Similar spin label/lipid ratios have been used by other workers (6). There were no noticeable differences in the spectra in the range of spin label ratios used.
Three different methods were used to interpret the recorded ESR spectra. For spectra that indicated strongly hindered motion of the spin label, the splitting between the low and high field extrema (2T,3 was used as an indication of immobilization. With other preparations it was feasible to calculate the order parameter, S, introduced for fatty acid spin labels undergoing anisotropic motion around their long molecular axis, as reported previously (16). For the spin label 1(1,14) spectra were evaluated in terms of an empirical motion parameter for nearly isotropic motion, T (17): where W,, is the line width of the midfield line and h, and h-, are the heights of the mid and high field lines of the spectrum, respectively (see Fig. 7). Greater freedom of motion is associated with smaller values of either 27?,, S or 7. All ESR spectra were recorded using a Varian-4052 spectrometer equipped with a temperature controlling unit. The spin-labeled stearic acid derivatives were purchased from Synva Co.

Phospholipid
Content and Composition after Replacement - Table  I shows the phospholipid content of several types of enzyme preparations and SR. DOL-enzyme has the same amount of phospholipid as SR, whereas the phospholipid content of DPL-enzyme is reduced by 70%. By using the procedure described for phospholipid replacement and varying the deoxycholate to protein ratio without adding exogenous phospholipids, preparations with different phospholipid content can be obtained. The higher the ratio of deoxycholate to protein, the lower the amount of phospholipid remaining associated with the protein. To obtain an enzyme preparation with the same phospholipid content as DPL-enzyme, 0.75 mg of deoxycholate/mg of protein were used. The resulting preparation (del-enzyme, Table I), has the same phospholipid composition as the original SR, which differs significantly from the phospholipid composition of DOL-enzyme and DPL-enzyme (Table  II). In agreement with previous reports (7, 18) five different phospholipid species are present in SR, of which phosphatidyl choline amounts to 67% of the total. On the other hand, both DPL-enzyme and DOL-enzyme contain more than 90% of phosphatidyl choline. The results of the fatty acid analysis show that DPL-enzyme contains mainly saturated fatty acids, of which palmitic acid (16:O) represents 85%. In contrast, the DOL-enzyme system contains mostly unsaturated fatty acids of which oleic acid (18: 1) accounts for about 85%.
Temperature Dependence of Caz+-ATPase Activity-A sharp break occurs at 29" in the Arrhenius plot of DPL-enzyme indicating a steep drop in Cal'ATPase activity below that temperature (Fig. 1). The plots for SR, in agreement with previous results (5, 19), and of del-enzyme show a less steep bend at 18". In the case of DOL-enzyme there is no break in the Arrhenius plot in the temperature range studied (3-40'); this absence of a transition temperature might be attributed to the fact that the phase transition for pure DOL occurs below 0" (20). The activation energies and the transition temperatures of the Ca*+ATPase activity for these different preparations are summarized in Table III. Below the transition temperature the  Transition temperatures, T1, and activation energies, ES, were obtained from Arrhenius plots of the Ca2+ATPase activity in the temperature range of 3"-40". For SR, del-enzyme and DPL-enzyme, the first value of each pair is the value obtained above TI, the second the value obtained below Tt. To calculate A@ and A@ above and below Tt, temperature values of 30" and 15", respectively, were used. In the case of DOL-enzyme, the average value is given. activation energy for DPL-enzyme, 43.5 kcal/mol, is considerably higher than those of the other preparations, which fall in the range of 30 kcal/mol below their transition temperatures. Above the transition temperatures, however, all activation energies are similar and in the range of 20 to 25 kcal/mol. Although del-enzyme has the same transition temperature as SR, the ATPase rate of the former has a higher activation energy, suggesting that the considerably reduced lipid content of del-enzyme might have some effect on the enzyme reaction rates. As shown in Table III, the differences in activation energy are compensated by differences in entropy of activation, resulting in values of free energy of activation roughly independent of variations in temperature, phospholipid content and composition (16 to 18 kcal/mol).

Membrane
Fluidity as a Function of Temperature-The ESR spectra of different preparations labeled with I(5,lOl at various temperatures fall into two categories (Fig. 2). Spinlabeled SR ( Fig. 2A) and DOL-enzyme (Fig. 2B) have spectra that reflect a label which changes from a fairly immobilized state at low temperatures (e.g. 3.5') to an almost isotropic motion around 25". In contrast, the spectra obtained with labeled DPL-enzyme and del-enzyme indicate a strongly immobilized label in the whole temperature range studied. Semilogarithmic plots of 2Th (Fig. M) or of (l-S)/S (Fig. 3B) uersus l/T show breaks at 17" in the case of SR labeled with I(5,lO) but no discernible break in the case of DOL-enzyme. The same type of plots for 1(5,10)-labeled del-enzyme and DPL-enzyme show breaks at 18" and 29". respectively (Fig. 4). Also the slopes are considerably steeper in the case of SR and DOL-enzyme than for DPL-enzyme and del-enzyme. Fig. 5 shows the ESR spectra of DPL-enzyme and delenzyme obtained with 1(1,141, which has the nitroxide label at a more distal position from the polar group. Increasing the temperature in this case produces a much larger increase in label mobility than observed with I(5,lOl (cf. Fig. 2). Because of the increased mobility of the label, the parameter T (see "Experimental Procedure") was more appropriate for the analysis of the spectra. A semilogarithmic plot of r uersus l/T (Fig. 6) also shows breaks at 29" and 18" for DPL-enzyme and del-enzyme, respectively. In contrast to 1(5,10), with which both preparations showed similar values of 2T,', at all temperatures, labeling with I( 1,14) results in significantly higher values of r for DPL-enzyme. This would indicate that the segments of the alkyl chains near the methyl ends are in a more fluid environment in del-enzyme than in DPL-enzyme. The differences become smaller at higher temperatures.

Effect
of Phospholipid Replacement on Formation and Decomposition of Phosphoenzyme-Upon addition of ATP at 0", the phosphorylated intermediate of del-enzyme reaches its maximum level in 8 s and it remains constant during the steady state of inorganic phosphate liberation (Fig. 7A). A similar feature has been observed with SR (cf. Ref. 21). The formation of phosphoenzyme with DPL-enzyme occurs in two steps; a rapid rise of phosphoenzyme in the first few seconds, followed by a gradual increase to reach in 120 s comparable values to the steady state levels obtained with del-enzyme. In contrast to this similarity between both systems with respect to phosphoenzyme levels, the steady state rate of inorganic phosphate liberation is much smaller in DPL-enzyme than in del-enzyme, with average rates of 0.30 and 2.40 nmol of P,/mg protein/min, respectively (Table IV). The phosphoenzyme level and the rate of inorganic phosphate liberation in DOLenzyme are 1.8 times (cf. Fig. 7, A and B) and 3.5 times (Table  IV)  Spin label: I(5.10). The order parameter S was calculated from the spectra as described elsewhere (16). In order to further characterize the kinetics of DPL-enzyme, phosphate (22). Furthermore, during the course of the ATPase we have studied in this system the reversa1 of the phosphoryla-'reaction either with SR (21) or with the purified CA*+ATPase tion reaction. It has been demonstrated in the intact SR enzyme (231, addition of ADP results in the resynthesis of ATP vesicles that ATP can be synthesized from ADP and inorganic at the expense of phosphoenzyme, with no appreciable changes in the amount of inorganic phosphate liberated. In the experiment shown in Fig. 8,0.2 mM ADP was added to DPL-enzyme at 4 s or 90 s after starting the reaction with 5 ELM ATP. In either case the addition of ADP induced a rapid decay of phosphoenzyme to the level obtained when the reaction was started with 5 pM ATP in the presence of 0.2 mM ADP. There were no appreciable changes in phosphate levels on addition of ADP.
These results indicate that the mechanism responsible for the strong inhibition of inorganic phosphate liberation in DPLenzyme does not interfere with the reversal of the phosphorylation reaction. It also appears that the phosphorylated intermediate of DPL-enzyme is readily accessible to ADP for the reverse reaction in the early and late phases of reaction.

DISCUSSION
The current view of the mechanism of the ATPase reaction coupled with Cal+ transport in sarcoplasmic reticulum is summarized in the following set of equations (24,25):  ). On the other hand, SR preparations delipidated with various phospholipases (1, 71, while having a strongly inhibited ATPase activity, are phosphorylated by ATP to nearly the same extent as the intact membrane. These results have been interpreted in terms of a selective involvement of phospholipids in the steps leading to phosphoenzyme decomposition. A recent kinetic study on delipidated SR (30) lends support to this view.
The discrepancy in the effect of phospholipids on the stability of the phosphoenzyme may be ascribable, at least partially, to several problems inherent in earlier techniques of delipidation.
For instance, it is rather difficult to control the extent of delipidation by lipase digestion, since a considerable fraction of the total phospholipids (uiz. 'about 20% of the original content, Ref. 7) remains undigested and attempts to remove all phospholipids often result in an irreversible inactivation of the enzyme. Furthermore, digestion products, whose nature depends on the species of lipase used, would induce various modifications of the properties of the membrane (1, 7,  8). The recent method of Warren et al. (4). which permits replacement of the endogenous phospholipids present in SR by various exogenously added phospholipids, has various advantages over enzymatic delipidation.
Since the ATPase protein is equilibrated with phospholipid throughout the procedure, no irreversible inactivation occurs. Also, by using a large excess of phospholipid it is possible to achieve almost complete replacement with the species chosen. A modified version of the replacement technique, which has the added advantage of allowing phospholipid replacement and partial purification of the ATPase enzyme at the same time, has been used in the present study with the aim of investigating the effect of phospholipid replacement by well defined phospholipid molecules-DPL and DOL-on the partial kinetic steps of the ATPase reaction. Replacement by DOL yields an enzyme preparation in which steady state levels of phosphoenzyme (at 0') are similar to those obtained with our conventional preparations of purified ATPase (27). but the rates of P, liberation are decreased by 25%. In contrast, replacement by DPL produces at 0" a very large inhibition of P, liberation, with little effect on phosphoenzyme formation. This. inhibition is due, to a large extent, to the replacement of the original phospholipids by DPL and not to the accompanying delipidation produced during replacement, since the phosphate liberation rates in DPL-enzyme are about 8 times lower in Sarcoplasmic Reticulum Calcium ATPase than the corresponding values obtained with del-enzyme having the same reduced phospholipid content as DPLenzyme. This striking inhibition of phosphate liberation observed in DPL-enzyme suggests that the effect of the incorporated DPL is exerted primarily on the reaction steps that lead to phosphoenzyme decomposition (Equations 7 to 91. The fact that addition of ADP to DPL-enzyme at 0" induces a decrease in the amount of phosphoenzyme without increase in inorganic phosphate liberation, suggests that the reaction has reversed from phosphoenzyme in either the PE(Ca), form (Equation 6) or the PE*(Ca)n form to the E(Caln form (Equations 6 and 7). The rate of the reverse reaction with DPL-enzyme is indistinguishable from those obtained with the purified ATPase (23) or SR (21). Thus the fact that DPL substitution has little effect on either the phosphorylation reaction or its reversal further supports the view that the strong inhibition of ATPase activity observed at low temperatures is due to selective inhibition of phosphoenzyme decomposition.
In contrast, del-enzyme, whose phospholipid composition is the same as that of the native SR or the purified ATPase, but which has only 30% of the phospholipid content, shows only a slight inhibition of phosphoenzyme formation and only a S-fold inhibition in the steady state rate of phosphate liberation. It is interesting to note in this context that much larger inhibitions of phosphate liberation have been reported with enzymatically delipidated SR (1, 7, 301. This may be attributable to the differences in phospholipid content between the two types of preparation since, as opposed to the 30% of the original phospholipids remaining in del-enzyme, only 20% remain after enzymatic delipidation. As has been suggested elsewhere (9), there might be a critical amount of phospholipids associated with the enzyme below which considerable inhibition of enzymatic activity would take place.
While this work was in progress, Warren et al. (9) described the effect of lipid replacement by DPL, DML, and DOL on the ATPase activity of purified enzyme preparations derived from SR. Their results show that in the case of the saturated lecithins, uiz. DPL and DML, complete inhibition of ATPase activity takes place below 28" and 24', respectively, and that no phosphorylation of the enzyme takes place below these temperatures. These results are at variance with our findings, since with DPL-enzyme we could detect ATPase activity even at 0". Furthermore, the phosphorylation reaction at this temperature was virtually unaffected by DPL-replacement.
At present we cannot explain the reasons for these discrepancies.
As shown in this paper, the transitional temperatures derived from enzymatic assays are identical with those derived from the ESR spectra (cf. Table V). A similar result has already been reported for native SR (5). Our studies utilizing partially purified ATPase containing well defined phospholipid species show that the phospholipid composition seems to be a crucial factor controlling the transition temperature of both ESR spectra, which presumably reflect membrane fluidity and ATPase activity. Thus SR and del-enzyme, which differ considerably in their total phospholipid content but show similar phospholipid composition, exhibit the same transition temperature (cf . Tables I and V). On the other hand, replacement by either DOL or DPL, whose transition temperatures are -20 and 42' (20) respectively, leads to large changes in the transition temperature of the corresponding enzyme systems. In DOL-enzyme there is no discernible transition temperature in the 3-40" range, suggesting that DOL substitution shifts the transition temperature to a value below 3", while in the DPL-replaced system the transition temperature is increased from 18' to 29'. It should be pointed out that, while in our experiments the break in ATPase plots occurs at the same temperature as in ESR-mobility plots, comparison of preparations that have been subjected to different treatments (DOL enzyme uersus SR) fails to reveal a correlation between absolute values of ATPase activity and spin label mobility.
The existence of a transition temperature, reflected in some parameters of membrane fluidity, is generally attributed to a phase transition from the crystalline to the liquid-crystalline state of the phospholipid moiety (5,31,32). However, the fact that in DPL-enzyme the transition temperature differs appreciably from the one obtained with pure DPL may indicate that lipid-protein interactions are also involved in the regulation of membrane fluidity and ATPase activity. Assuming that lipidprotein interactions modify the transition temperature of the lipid phase, this result could still be attributed to a lipid phase transition.
Nevertheless, the possibility that lateral phase separations or cluster formations in the lipids are instrumental in producing the transition temperature cannot be excluded. Recently,Lee et al. (33) have shown that a complex of purified ATPase and DOL exhibits a transition temperature at 29" in terms of its ATPase activity. They attribute this result to the presence of quasicrystalline clusters in the DOL moiety, since they find the same transition temperature in the partition of the spin probe TEMPO incorporated into DOL bilayers. It is difficult to reconcile these results with our finding that in DOL-enzyme, both by ESR and ATPase measurements, no transition temperature is discernible in the 3-40" range. Also worth noting in regard to DOL-enzyme is the fact that while below 17' ESR studies show similar spin label mobilities for this system and native SR, above this temperature a significantly steeper increase in label mobility was found for SR. Although the DOL molecules in DOL-enzyme are most likely to be already in the liquid-crystalline state at 17", the lipids of SR include several types of unsaturated fatty acid molecules with multiple double bonds (81, which above the average transition temperature would produce a much more fluid membrane than DOL, which has only one double bond in each hydrocarbon chain. The ESR spectra of DPL-enzyme indicate the strongest spin label immobilization below the transition temperature. This finding suggests that the lipid phase in DPL-enzyme is in a highly ordered state in this temperature range, as might be expected for a system whose predominant lipid moiety contains two saturated hydrocarbon chains. However, on the basis of the greater label immobilization in del-enzyme than in intact SR, some of the immobilization in DPL-enzyme might be attributable to its reduced phospholipid content. The decrease in label mobility produced by a reduction in total phospholipid content, as observed in del-enzyme with respect to SR, may be due to a large proportion of total lipids being in close contact with protein molecules, the increased lipidprotein interactions resulting in a concomitant decrease in the mobility of the lipids. It has been suggested for other membrane systems that the lipid regions in the immediate vicinity of the protein molecule possess a more ordered configuration than the bulk lipid in the membrane (9,34). case, it remains to be resolved whether the effect of high Ca*+ or alkaline pH involves the same conformational stabilization that we postulate for the phospholipid effect; further experiments are needed to clarify this problem.
The thermodynamic parameters of activation are those of the rate-limiting step(s) of the hydrolysis of ATP; in terms of the current reaction scheme this step(s) follows the formation of the phosphoenzyme. The fact that both AH$ and ASS are positive indicates that the activation step is energetically unfavorable but entropically favorable. That is, the decrease in order compensates for the increase in enthalpy; mechanistically one could visualize this as an increase in randomness of the protein structure coupled with stretching of, or hindered rotation about, some bonds. The increase in AHi and ASS below the transition temperature, indicating a change in the nature of the rate limiting step(s), suggests that the increased order of lipid, reflected in the reduced mobility of the spin label, imposes an increased order on the preactivation state, thereby producing an increased AS& while also somehow energetically impeding the transition associated with activation. The difference between DPL-enzyme and SR can be similarly explained: the increased "order" in the lipid moiety of DPL-enzyme, resulting in an increase in both AH+ and AS$, produces not only greater loss on activation of rigidity or order (e.g. hydration) in the enzyme, but also increases the energy barrier for activation. The same reasoning would explain the larger AH$and ASS obtained with del-enzyme as compared to SR. In this case the reduction in lipid content results in an increased order of the lipid moiety, as discussed above, which in turn would affect the protein in the same way as lipid replacement by DPL, although the magnitude of the effects are much larger in the case of DPL replacement.
In the light of the above discussion, the large inhibition of phosphoenzyme decomposition in DPL-enzyme below its transition temperature might be attributed to stabilization of the phosphoenzyme form by the highly ordered array of the DPL molecules around the enzyme. Alkaline pH (25,35), or high Ca2+ concentrations (27) also exert a stabilizing effect on the phosphoenzyme, with the consequent inhibition of inorganic phosphate liberation. In the case of high Cal+ concentration, it has been suggested that the inhibition of phosphoenzyme decomposition results in accumulation of the PE*(Ca)n form, since all the previous steps are not affected. In the case of the stabilization of phosphoenzyme produced by DPL-replacement, it is not possible to decide which one of the two postulated forms of phosphoenzyme is the predominant species. The inhibition of the reaction might be exerted either at the level of the conformational change that accompanies CaZ+ translocation, in which the enzyme changes from PE(Ca)n to PE*(Ca)n (Equation 7), resulting in net accumulation of PE(Ca)n, or at the level of the decomposition of the PE*(Ca)n form (Equation 8), resulting in its net accumulation.
In either