Effects of Organic Solvents, Methylamines, and Urea on the Affinity for Pi of the Ca2+-ATPase of Sarcoplasmic Reticulum*

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 263, No. 1, Issue of January 5, pp Printed in U. S.A. Effects of Organic Solvents, Methylamines, and Urea on the Affinity for Pi of the Ca2+-ATPase of Sarcoplasmic Reticulum* (Received for publication, June 23, 1987, and in revised form, August 31, 1987) Leopoldo de Meis* From the Znstituto de Ciencios Biomedicas, Departamento de Bioquimica, Uniuersidade Federal do Rio de Janeiro, RJ 21910, Brazil Giuseppe Inesi From the Department of Biological Chemistry, School of Medicine, University of Maryland, Baltimore, Maryland The Ca2+-ATPase of sarcoplasmic reticulum can be phosphorylated by Pi, forming an acylphosphate residue at the catalytic site of the enzyme. In a previous report (de Meis, L., Alves, E., and Martins, 0. B. (1980) Biochemistry 19, ), it was shown that organic solvent such as dimethyl sulfoxide and glycerol cause a decrease in the apparent K, for Pi. In this report it is shown that a similar effect is obtained with the methylamines glycine betaine and trimethylamine N-oxide. The apparent K, value for Pi in totally aqueous medium and in the presence of either 6.4 M glycerol, 1.4 M dimethyl sulfoxide, 0.4 M trimethylamine N-oxide, or 1 M glycine betaine were found to be respectively 2.85,0.52,0.52,0.81, and 0.93 mm at ph 6.2 and >10.0, 1.08, 2.53, 3.05, and 2.05 mm at ph 7.5. In contrast to the effect of methylamines, urea caused an increase in the apparent K, for Pi. When mixed in the appropriate concentration ratio, the effect of either organic solvent or methylamines is cancelled by urea. The Ca2+-ATPase of sarcoplasmic reticulum vesicles can catalyze both the hydrolysis and the synthesis of ATP (1-3). During the hydrolysis of ATP, Ca2+ is accumulated by the vesicles and in the reverse process, coupled with the synthesis of ATP, Ca2+ is released by the vesicles at a fast rate. The synthesis of ATP is initiated by phosphorylation of the ATPase by Pi, forming an acylphosphate residue at the catalytic site of the enzyme (1, 4). This reaction occurs without the need for the energy derived from the Ca2+ gradient (2, 3, 5). The affinity of the enzyme for Pi varies greatly with the ph of the medium (2, 5-7). The concentration of Pi needed for half-maximal enzyme phosphorylation is in the range of 2-3 mm at ph 6.0 and increases progressively to values above 10 mm as the ph of the medium is raised to the physiological values (2, 5-7). The ph dependence of the phosphorylation reaction can be artificially abolished in vitro by adding organic solvents such as dimethyl sulfoxide or glycerol to the assay medium (8-10). These solvents promote a decrease in the enzyme s apparent K,,, for Pi, an effect which is * This investigation was supported by United States Public Health Service Grant HL27867, Organization of American States, Financiadora de Estudos e Projetos, and Conselho de Desenvolvimento Cientifica e Technologico, Brasil. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $Recipient of a fellowship from the John Simon Guggenheim Memorial Foundation. 157 more pronounced at alkaline than at acidic ph values. Thus, in the presence of organic solvents, a given Pi concentration leads to the same phosphoenzyme level at ph 6.0 as at ph 7.5. Quaternary methylamines such as trimethylamine-n-oxide and glycine betaine are phylogenetically widespread compounds. Among vertebrates, they are found in the tissues of species ranging from the urea-rich marine cartilaginous fishes (sharks and rays) to mammals (11-14). In this report it is shown that these two methylamines, like organic solvents, can abolish the effect of ph on the Pi phosphorylation reaction of the Ca*+-ATPase and that theffect of both methylamines and of organic solvents are antagonized by urea. MATERIALS A METHODS Sarcoplasmic reticulum vesicles were prepared from rabbit skeletal muscle by the method of Eletr and Inesi (15). In a few experiments, vesicles loaded with calcium phosphate were prepared as previously described (6). Purification of radioactive Pi and determination of phosphoenzyme were carried out as previously described (6, 8, 16). Composition of the assay media is described in the figure legends. In previous reports (8,17) it was observed that organic solvents promote a small but significant change in the pk values of the different buffers used. In all experiments, the ph of the assay medium was measured in the test tube after all the reagents had been added and, when necessary, it was adjusted to the desired value. The incubation time at 35 C was 2 min. It was previously established that equilibrium levels of phosphoenzyme are obtained after 10 s of incubation (2, 8). Formation of the phosphoenzyme by Pi requires removal of Ca2+ from the medium with EGTA and no phosphoenzyme formation is observed when Ca2+ is added to the medium, either in the absence (2,3, 5) or in the presence of methylamines (Fig. 1). The pk, of Pi in totally aqueous medium and in the presence of 1 M of either glycine betaine or urea was determined by titration with standardized solutions of NaOH or HC1 (8, 17). RESULTS Effect of Methylamines-Phosphorylation of the Ca2+- ATPase by Pi occurs more readily at acid than at alkaline ph (2,5-7). In Fig. 1 the equilibrium level of phosphoenzyme was measured in the presence of 1 mm Pi. At ph 6.2 (Fig. hi), methylamines moderately increased the phosphoenzyme level. At ph 7.6 (Fig. lb), a considerably greater increase in the phosphoenzyme level was observed following the addition of either 0.4 M TMAO or 2 M glycine betaine to the medium. The experiment of Fig. 1 was performed using empty vesicles, The abbreviations used are: EGTA, [ethylenebis(oxyethylenenitri1o)ltetraacetic acid TMAO, trimethylamine N-oxide; MES, 2(Nmorpho1ino)ethanesulfonic acid EPPS, N-(2-hydroxyethyl)piperazine-N -3-propanesulfonic acid; MOPS, 3-(N-morpholino)propanesulfonic acid.

2 158 Methylamines and Sarcoplasmic Reticulum 0 'ml-+t=l 0 OJO~".p-r, I I I 2 4 Trimethylamine, M FIG. 1. Effect of methylamines on the equilibrium level of phosphoenzyme. In A the assay medium composition was 50 mm MES-Tris buffer (ph 6.2), 1 mm 32Pi, 5 mm MgC12, and either 10 mm EGTA (0, 0) or 1 mm CaClz (0, B). In B the assay medium composition was the same except that the buffer used was EPPS (ph 7.6). The concentrations of TMAO (0) and glycine betaine (0) are given in the figure. The reaction was started by the addition of empty vesicles to a concentration of 0.35 mg of protein/ml and arrested after 2 min at 37 "C by the addition of 2.5 volumes of an ice-cold 0.25 M perchloric acid solution containing 4 mm nonradioactive Pi. There was no phosphoenzyme formation in the presence of Ca2+. Bars show standard error of four experiments. ATPase TABLE I Apparent affinity of the Ca2'-ATPase for Pi Other additions and experimental conditions were as described in Figs. 2, 4, and 5. The values represent the averages f standard errors of three experiments. Apparent K, for Pi ph 6.2 ph 7.5 mm None C 0.30 >10.00 Glycerol, 6.4 M C f 0.12 Dimethyl sulfoxide 1.4 M 0.52 f C M 0.10 f f M TMAO, 0.4 M 0.81 f f M 0.93 f f M 0.80 f 0.05 Urea, 0.8 M 6.80 f 0.30 Dimethyl sulfoxide 1.4 M plus.2.90 f 0.20 >10.00 urea 1.8 M Glycine betaine 1 M plus urea 3.10 f 0.15 > M 0 I Pi,mM '/Pi.(mM") FIG. 2. Effect of methylamines on the Pi concentration dependence of the phosphorylation reaction. The assay medium composition was 50 mm EPPS buffer (ph 7.5), 5 mm MgCl2, 10 mm EGTA, 0.35 mg of empty vesicle protein/ml, the 32Pi concentrations shown in the figure and either no addition (0), 0.4 M TMAO (A), or 1 M glycine betaine (0). Incubation and quenching were as described in Fig. 1. Double-reciprocal plots of the data obtained with TMAO and glycine betaim are shown in B i.e. in the absence of a transmembrane Ca2+ gradient. At ph 7.5 essentially the same degree of activation by glycine betaine as that obtained with empty vesicles was observed when vesicles previously loaded with calcium phosphate (Ca2+ gradient) were used (data not shown). Both glycine betaine and TMAO led to a decrease in the apparent K,,, of the enzyme for Pi. This effect was more pronounced at alkaline than at acidic ph values (Fig. 2 and Table I). As previously described (8), a similar effect is observed when one of the organic solvents glycerol or dimethyl sulfoxide is added to the medium (Table I). The effects of methylamines and organic solvents were additive (Fig. 3). This could be observed clearly in the presence of low concentrations of organic solvent. The maximal number of sites available to be phosphorylated by Pi in sarcoplasmic reticulum vesicles is in the range of pmols/g of protein. In the presence of a large concentration of dimethyl sulfoxide, all the catalytic sites of the enzyme are phosphorylated by Pi, and the effect of adding methylamine can no longer be detected. The double reciprocal plot of Fig Me, SO,M FIG. 3. Effect of dimethyl sulfoxide on the equilibrium level of phosphoenzyme in the absence and in the presence of methylamines. The assay medium composition was 50 mm EPPS buffer (ph 7.5), 1 mm 32Pi, 5 mm MgCl,, 10 mm EGTA and either no addition (O), 0.4 M TMAO (A), or 1 M glycine betaine (B). In the figure, MezSO refers to dimethyl sulfoxide. Other conditions were as in Fig. 1 2B shows that the maximal level of phosphorylation was not affected by methylamines. Partition Coefficient-Previous evidence (8, 9, 16-23) indi- cates that the catalytic of site the ca''-atpase is hydrophobic when the enzyme is in the conformation that is phosphorylated by Pi. These results have been interpreted to mean that different organic solvents facilitate the partition of Pi from the assay medium into the catalytic site of the enzyme by decreasing the difference of hydrophobicity between these two compartments. This would promote an increase in the enzyme affinity for Pi. Supporting this view, it was found that the partition of Pi between a water phase (assay medium) and an organic phase containing benzene and isobutyl alcohol in- creased when cosolvents such as dimethyl sulfoxide and glycerol were included in the water phase (8). In agreement with previous reports (8), an increase in the partition coefficient for Pi was observed when dimethyl sulfoxide was included in the water phase, but this could only be measured when the concentration of dimethyl sulfoxide was raised above 1.4 M (Table 11). In totally aqueous medium and in the presence of either 0.4 M TMAO, 1.4 M dimethyl sulfoxide or 1 M glycine betaine, the partition coefficient was too small to be measured with the method used. The data of Fig. 3 show that the effect of 0.4 M TMAO or of 1 M glycine betaine on the phosphoryl-

3 Methylamines and Sarcoplasmic Reticulum ATPase TABLE I1 Effect of urea and dimethyl sulfoxide on the partition coefficient of Pi The assay media contained 50 mm EPPS buffer, 5 mm EGTA, 5 mm MgCl,, 1 mm 32P; (2 X 10' cpmlpmol Pi) and the additions shown in the table. To 0.5 ml of the assay media was added 1 ml of a mixture containing 35% benzene and 65% isobutyl alcohol (v/v) previously saturated with water. The tube was vigorously stirred for 5 min. After phase separation an aliquot of the benzene-isobutyl alcohol layer was counted in a liquid scintillation counter. The partition coefficient was calculated by dividing the concentration of Pi in the organic phase by its concentration in the aqueous phase (8). The values are the average f standard error of four experiments., nondetectable. None TMAO, 0.4 M Glycine betaine, 2 M Dimethyl sulfoxide 1.4 M 2.8 M 4.2 M Dimethyl sulfoxide 2.8 M plus 1 M NH,C1 Dimethyl sulfoxide 2.8 M plus 1 M urea Dimethyl sulfoxide 2.8 M plus 3 M urea Organic-aqueous phase partition coefficient 5.04 f 1.34 X lo-' 1.41 f 0.37 X lo-' 4.72 f 1.27 X lo-' 7.60 f 3.23 X 5 3 ation of the Ca2'-ATPase by Pi was comparable to that observed with the use of about 1 M dimethyl sulfoxide. There- L 2 fore, it might be possible that the effect of methylamines and 2 of low concentrations of dimethyl sulfoxide is promoted by = - I small changes in the partition coefficient for Pi not measurable by the method used. 0 Kundu and Das (24) measured the transfer energies of Pi and other anions from water to dimethyl sulfoxide-water and UREA,M $i,(rnm-') from water to urea-water mixtures. These measurements re- FIG. 5. Effect of urea on the equilibrium level of phosphoenvealed that dimethyl sulfoxide and urea have opposing effects zyme in absence and in presence of dimethyl sulfoxide. In A the assay media composition was 50 mm MES buffer (ph 6.2), 1 mm on the interactions between anions and water. Anions are 32Pi, 5 mm MgCl,, 10 mm EGTA, and either no addition (O), 1.4 M destabilized in dimethyl sulfoxide-water mixtures and stabi- (O), or 2.8 M dimethyl sulfoxide (A). In B the assay medium compolized in urea-water mixtures, as if the interactions between sition was 50 MOPS buffer (ph 6.1), 5 mm MgCl,, 5 mm EGTA, the water and anions were to decrease after the addition of concentrations of 32Pi shown in the figure and either no addition (O), dimethyl sulfoxide and to increase when urea is included in 1.4 M dimethyl sulfoxide (O), 1.4 M dimethyl sulfoxide plus 1 M urea (A), or 1.4 M dimethyl sulfoxide plus 1.8 M urea the system. Table I1 (A). Incubation and shows that urea antagonized the effect quenching were as described in Fig. 1 of dimethyl sulfoxide on the partition coefficient for Pi. This seems to be a specific effect of urea because it was not TABLE I11 duplicated by NH4C1. From these data we inferred that urea Combined effects of urea and methylamines on the phosphorylation should increase the apparent K,,, for Pi and antagonize the level effect of both dimethyl sulfoxide and methyl amines on the The assay medium composition was 50 mm MES-Tris buffer (ph Ca2'-ATPase. 6.2), 5 mm MgC12, 1 mm 32P;, 10 mm EGTA, and the additions shown Effect of Urea on the Phosphorylation Reaction-In contrast in the table. Incubation and quenching were as described in Fig. 1. to methylamines and organic solvents, urea increased the Pi PhosDhoenzvme concentration required for half-maximal phosphorylation of am1 E-P/g protein the ATPase by Pi (Fig. 4 and Table I). The effects of both None 1.31 f 0.01 dimethyl sulfoxide and methylamines on the Ca2'-ATPase were antagonized by urea (Fig. 5, Tables I and 111). Whereas organic solvents and methylamines decreased the apparent K, for Pi (Table I and Fig. 2), urea increased it (Fig. 4), and in mixtures of appropriate concentration ratios the effect of one reagent canceled the effect of the other (Tables I and I11 and Fig. 5). The effect of 1 M urea was antagonized by approximately 1 M of either methylamine (Table 111) or dimethyl sulfoxide (Fig. 5). The pk2 of Pi-Beil et al. (7) proposed that only primary phosphate (HzPO;) is able to phosphorylate the ATPase. Thus, the changes in apparent K,,, for Pi observed after the addition of methylamines and urea to the assay medium could be related solely to an effect of these compounds on the pk, I E- P p I 2 Pi, rnm '/Pi.(rnM") FIG. 4. Effect of urea on the apparent K, for Pi. The assay medium composition was 50 mm MES buffer (ph 6.2), 10 mm EGTA, 5 mm MgCl,, the concentrations of 32Pi shown in the figure and either no addition (O), 0.4 M urea (W), or 0.8 M urea (A). Incubation and quenching were as described in Fig. 1. Double reciprocal plots of the data obtained in A are shown in B. 4 1 M urea 1 M urea M TMAO 0.44 f * M urea M glycine betaine 0.90 f M urea + 1 M glycine betaine 1.27 f 0.01 of Pi, i.e. to different concentrations of the ionic species HzPOa and HPOT. This possibility can be excluded because the pk, of Pi was practically the same in totally aqueous medium and in the presence of 1 M of either urea or glycine betaine. The values measured were 7.10, 7.10 and 7.25, respectively (data not shown). DISCUSSION Apparent K,,, for Pi-The Ca2'-ATPase undergoes a conformational change during the catalytic cycle (for reviews see 2,

4 160 Methylamines and Sarcoplasmic Reticulum ATPase 3). In one of its conformations (E) the enzyme has a high to water made hypertonic (300 mosmol/liter) with NaCl. affinity for Ca2+ and is phosphorylated by ATP but not by Pi. Yancey and Somero (11, 14) reported that the activity of In the second conformation (*E), the enzyme has a low arginosuccinate lyase from pig kidney or from liver of the affinity for Ca2+ and is phosphorylated by Pi but not by ATP. The experiments presented in this report were performed in thornback ray is noncompetitively inhibited by 0.4 M urea, and that this effect is abolished when 0.2 M TMAO is added the absence of ATP and Ca2+. In this condition the two to the assay medium together with urea. Ca2+ transport enzyme forms coexist in equilibrium and the phosphorylation ATPases with properties similar to that found in the sarcoby Pi is described as: plasmic reticulum have been described in brain, red cells, (1) (2) (3) E + *E + *E.Pi + *E- P + HZ0 Pi liver, platelets, and kidney (29-36). Using 14N NMR, Balaban and Knepper (13) estimated that the intracellular concentration of methylamine compounds in rat and rabbit kidney is At ph , the equilibrium constant of reaction 1 (*E/E) near 200 mmol/kg of intracellular water. At these concentrais about one (2,25), and it decreases as the ph of the medium tions their effects on the phosphorylation of the Ca2 -ATPase is raised. Thus, at ph 7.4 practically all the enzyme is in the by Pi are not maximal, but in the physiological ph range, form E (2). The equilibrium of reaction 3 strongly favors the they are not trivial. In Fig. lb, for example, the phosphoenformation of the phosphoenzyme (8, 26). Therefore, a small zyme formation doubled in the presence of 0.1 M TMAO or change in the Keq of reaction 2 can lead to a significant change 0.2 M glycine betaine. Therefore, besides protecting the cell in the apparent K, for Pi (8). Experiments described in from the toxic effect of urea, perhaps methylamines may also previous reports (8, 9, 16-23) indicate that the catalytic site be involved in regulating the affinity for Pi of the Ca2+of the enzyme form *E is hydrophobic. Dupont and Pougeois ATPase. (10) suggested that the binding of Pi to the Ca2+-ATPase Effect of Organic Solvent on Different Enzymes-In addition promotes a conformational change of the protein which would to their effects on the Ca2 -ATPase of sarcoplasmic reticulum, lead to the release of a large number of water molecules from organic solvents have been found to promote a decrease in the active site. In this view, the active site of the enzyme the apparent K,,, for Pi of different enzymes involved in processes of energy transduction. These include the soluble would only become hydrophobic after the binding of Pi. This proposal is not supported by the findings obtained recently in F,-ATPase of mitochondria (37-41), yeast inorganic pyrophosphatase (42), and the H -ATPase of yeast plasma memdifferent laboratories. Solvent accessibility studied by fluoresbrane (43). There are large structural differences between the cence quenching (21, 22) and hydrophobic labeling of the molecules of these different enzymes. The Ca2+-ATPase has Ca2 -ATPase with trifluoromethyl-iodophenyl-diazirine (23) only one type of subunit, while the Fl-ATPase is an oligomer indicate that the catalytic site of the enzyme form *E had the containing several different subunits. The Ca2 -ATPase, the hydrophobic character before the addition of Pi and that the (Na + K)-ATPase and the H+-ATPase of yeast are membinding of Pi did not promote a detectable change of polarity brane-bound enzymes, whereas the F, of mitochondria and or water accessibility in the catalytic site. The finding that the yeast pyrophosphatase are water-soluble enzymes without organic solvents increase the partition coefficient of Pi from bound lipids. These differences suggest that the decrease in an aqueous phase into an organic phase (Ref. 8 and Table 11) the apparent K,,, for Pi is not related to a specific interaction indicates that dimethyl sulfoxide changes the properties of of the organic solvent with the enzyme, but rather to a change the solution (assay medium). We infer that this change is of water activity promoted by the cosolvent. The water molresponsible for changes of the Keg of reaction 2 and a decreased ecules that organize around a protein in solution have propapparent K,,, for Pi. The partition coefficient data obtained erties that are different from those of medium bulk water, with methylamines were not sufficient to distinguish whether for example a lower vapor pressure, a lower water mobility the decrease in the apparent K, for Pi was promoted by a and a greatly reduced freezing point (44,45). Similar changes change in the Keq of reaction 1, 2, or 3. The finding that the in the properties of water are observed in mixtures of organic effect of methylamines is small at ph 6.0 and pronounced at solvents and water (46, 47). At present we do not know the ph 7.5 may indicate that they change the K, of reaction 1. structure of water in the different compartments of the cell. On the other hand, the finding that urea abolishes the effect For instance, because of the very high protein concentration of dimethyl sulfoxide on the partition coefficient of Pi and found in the mitochondrial matrix (48,49), it is unlikely that that it also abolishes the effect of dimethyl sulfoxide and in physiological conditions the solvent structure in this orgamethylamines on the K, for Pi suggests that like organic nelle is the same as that of the aqueous solutions usually used solvents, the methylamines may promote a change in the Keq for experiments in vitro. The data presented raise the possiof reaction 2. More experimentation is required to untangle bility that methylamines and urea may play a physiological this and other possibilities. role in modulating the affinity for Pi of the Ca2+-ATPase and Physiological Role-Methylamines have been observed in perhaps, of other enzymes involved in processes of energy different types of tissues for many years; however, their transduction. In the cell the different methylamines would physiological role is still unknown (13, 27, 28). Recently it modify the properties of bulk water in a manner similar to has been proposed that methylamines may serve to protect that observed with the use of organic solvents, and urea would certain cells from the toxic effect of urea by counteracting its antagonize this effect. effect on protein structures (11, 14). Concentrations of urea varying from M occur commonly in the mammalian REFERENCES kidney and throughout the tissues of cartilaginous fishes (11, 1. Hasselbach, W. (1978) Biochirn. Biophys. Acta 515, ). In these fishes the high concentration of urea serves to 2. de Meis. L. (1981) in Transport in the Life Sciences (Bittar, E. balance the high osmolarity of the sea water. Wilkie and Wray E., ed) Vol. 2, Wiley, New York (28) measured the concentration of the methylamine glycer- 3. Tanford, C. (1984) Crit. Rev. Biochern. 17, Makinose, M. (1972) FEBS Lett. 25, ylphosphorylcholine in the gastrocnemius muscles of frogs by 5. Masuda, H., and de Meis, L. (1973) Biochemistry 12, P NMR spectroscopy. They observed a simultaneous in- 6. de Meis, L. (1976) J. Bwl. Chern. 251, crease in blood urea and in the concentration of the methyl- 7. Beil, F. U., von Chak, D., and Hasselbach, W. (1977) Eur. J. amine in muscle when frogs kept in tap water were exposed Biochern. 81,

5 Methylamines Sarcoplasmic and Reticulum ATPase de Meis. L., Alves, E., and Martins, 0. B. (1980) Biochemistry J. Bwl. Chem. 245, , Trotta, E., E. and de Meis, L. (1978) J. Bwl. Chem. 253, de Meis, L., and Inesi, G. (1982) J. Biol. Chem. 257, Dupont, Y., and Pougeois, R. (1983) FEBS Lett. 156, Chiesi, M., Zurini, M., and Carafoli, E. (1984) Biochemistry 23, 11. Yancey, P. H., and Somero, G. N. (1980) J. Exp. Zool. 212, Lotersztajn, S., Hanoune, J., and Pecker, F. (1981) J. Biol. Chem. 12. Martins, J. J., and Finkelstein, J. D. (1981) Anal. Bwchem. 111, 256, Chan, K.-M., and Junger, K. D. (1983) J. Bwl. Chem. 258, Balaban, R. S., and Knepper, M. A. (1983) Am. J. Physiol. 245, Dean, W. L. (1984) J. Biol. Chem. 259, Somero, N. G. (1986) News Physiol. Sciences 1, Gmaj, P., Zurini, M., Murer, H., and Carafoli, E. (1983) Eur. J. 15. Eletr, S., and Inesi, G. (1972) Biochim. Biophys. Acta. 282, 174- Biochem. 136, Vieyra, A., Nachbin, L., de Dios-Abad, E., Goldfeld, M., Meyer- 16. de Meis, L. (1984) J. Bwl. Chem. 259, Fernandes, J. R., and de Moraes, L. (1986) J. Biol. Chem. 261, 17. de Meis. L., Behrens, M. I., Petretsky, J. H., and Politi, M. J (1985) Biochemistry 24, Sakamoto, J., and Tonomura, Y. (1983) J. Biochem. (Tokyo) 93, 18. de Meis, L. (1986) in: The Molecular Basis of Mooement Through Membranes (Quinn In, P. J., and Pasternak, C. A., eds) 38. Sakarnoto, J. (1984) J. Biochem. (Tokyo) 96, Biochem. Soc. Symp. 50, Yoshida, M. (1983) Biochem. Biophys. Res. Commun. 114, 907-