- Email: [email protected]
Regulation of Chicken Gizzard Ecto-ATPase Activity by Modulators That Affect Its Oligomerization Status
Archives of Biochemistry and Biophysics Vol. 387, No. 1, March 1, pp. 107–116, 2001 doi:10.1006/abbi.2000.2216, available online at http://www.idealibrary.com on
Regulation of Chicken Gizzard Ecto-ATPase Activity by Modulators That Affect Its Oligomerization Status Charles C. Caldwell, 1 Stephen C. Hornyak, Erik Pendleton, Dawn Campbell, and Aileen F. Knowles 2 Department of Chemistry, San Diego State University, San Diego, California 92182-1030
Received September 13, 2000, and in revised form November 10, 2000; published online January 26, 2001
The major ectonucleoside triphosphate phosphohydrolase in the chicken gizzard smooth muscle membranes is an ecto-ATPase, an integral membrane glycoprotein belonging to the E-ATPase (or E-NTPDase) family. The gizzard ecto-ATPase is distinguished by its unusual kinetic properties, temperature dependence, and response to a variety of modulators. Compounds that promote oligomerization of the enzyme protein, i.e., concanavalin A, chemical cross-linking agent, and eosin iodoacetamide, increase its activity. Compounds that inhibit some ion-motive ATPases, e.g., sulfhydryl reagents, xanthene derivatives, NBD-halides, and suramin, also inhibit the gizzard ecto-ATPase, but not another E-ATPase, the chicken liver ecto-ATP-diphosphohydrolase, which contains the same conserved regions as the ecto-ATPase. Furthermore, inhibition of the gizzard ecto-ATPase by these compounds as well as detergents is not prevented by preincubation of the membranes with the substrate, ATP, indicating that their interaction with the enzyme occurs at a locus other than the catalytic site. On the other hand, the inhibitory effect of these compounds, except suramin, is abolished or reduced if the membranes are preincubated with concanavalin A. It is concluded that these structurally unrelated modulators exert their effect by interfering with the oligomerization of the ectoATPase protein. Our findings suggest that, under physiological conditions, the gizzard smooth muscle ecto-ATPase may exhibit a range of activities determined by membrane events that affect the status of oligomerization of the enzyme. © 2001 Academic Press Key Words: ecto-ATPase; E-NTPDase; gizzard; oligomerization.
1 Present address: Laboratory of Immunology, NIAID, NIH, Bethesda, MD 20892-0001. 2 To whom correspondence and reprint requests should be addressed. Fax: (619)594-4634. E-mail: [email protected].
0003-9861/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
E-ATPases (or E-NTPDases) are membrane-bound cell surface phosphohydrolases that have broad substrate specificity. While they hydrolyze most nucleoside triphosphates (NTP) 3 and nucleoside diphosphates (NDP), their physiological substrates are most likely extracellular ATP, UTP, and ADP, ligands of the purinergic receptors. Extracellular ATP is also a substrate for ecto-protein kinases (1). To date, more than a dozen E-ATPases have been cloned. The deduced primary sequences indicate that they are integral membrane proteins of approximately 54 kDa with two transmembranous segments near the N- and C-termini, and all contain five conserved regions (2). Based on the extent of their sequence homology, three subfamilies can be discerned, the ecto-ATPases, the CD39s, and the ecto-ATP/Dases. Expressed ectoATPases from smooth muscle, brain, and tumor cells exhibit low activity toward NDP (3–5). On the other hand, CD39s and ecto-ATP/Dases, whether purified or expressed from their cDNAs, hydrolyze NDP as well as NTP (6 –13). However, CD39s are the vascular ectoATP/Dases (6, 9, 10, 13) and are distinct from the other characterized ecto-ATP/Dases (7, 8, 11, 12) both in the extent of sequence identity and tissue distribution. In addition to their different abilities to hydrolyze NDP, the ecto-ATPases and the ecto ATP/Dases differ in their responses to a variety of modulators. High concentrations of azide inhibit most ecto-ATP/Dases but not ecto-ATPases (14 –16). The inhibitory mechanism of azide on the purified chicken oviduct ecto-ATP/ 3
Abbreviations used: ATP/Dase, ATP diphosphohydrolase; ConA, concanavalin A; DSS, disuccinimidyl suberate; DMSO, dimethyl sulfoxide; EIAA, eosin iodoacetamide; NBD-Cl, 7-chloro-4-nitrobenz2-oxa-1,3-diazole; NBD-F, 7-fluoro-4-nitrobenz-2-oxa-1,3-diazole; pCMPS, p-chloromercuripheylsulfonate; pHMB, p-hydroxymercuribenzoate; T-tubule, transverse tubule, CMC, critical micelle concentration; NTP, nucleoside triphosphate; NDP, nucleoside diphosphate. 107
108
CALDWELL ET AL.
Dase has been investigated (17). On the other hand, we showed that the activity of the chicken gizzard smooth muscle ecto-ATPase is affected by several compounds that have no effect on the chicken liver ecto-ATP/Dase (16). Therefore, in spite of the presence of the same conserved regions in their sequences, significant differences exist between the ecto-ATPases and the ectoATP/Dases with respect to their interaction with nucleotide substrates and modulators. The different effects of the modulators suggest different regulatory mechanisms of the two enzymes. Further characterization of these differences will be important in future structure–function studies of the E-ATPases. In this report, we describe the effect of an extensive list of modulators on the chicken gizzard ecto-ATPase. Based on these results, a model of the regulation of the gizzard ecto-ATPase activity is presented. MATERIALS AND METHODS Materials. DSS was purchased from Pierce (Rockford, IL). EIAA, NBD-Cl, and NBD-F were purchased from Molecular Probes (Eugene, OR). Xanthene derivatives were purchased from Aldrich (Milwaukee, WI). All other reagents were purchased from Sigma Chemical Company (St. Louis, MO). Preparation of chicken gizzard plasma membranes. Chicken gizzards were obtained within 16 h of slaughter. Preparation of the plasma membranes from the tissue homogenate by differential and sucrose gradient centrifugation was described previously (16). The membrane preparations (1–3 mg protein/ml) were stored in 20 mM Mops, pH 7.3, 30% sucrose at ⫺20°C until use. ATPase activity was stable for several months. Protein concentration of the plasma membranes was determined using the Bio-Rad DC protein assay reagent with bovine serum albumin as standard. ATPase assays. ATPase activity was determined either by the coupled enzyme assay which determined the rate of ADP release (18) or by a colorimetric method that determined the rate of phosphate release (16). Because of the nonlinear nature of the time course, the decrease of absorbance at 340 nm of NADH between 1 and 2 min after the initiation of the reaction by ATP in the coupled enzyme assay was used to calculate the rate of ATP hydrolysis. In the colorimetric assay, the ATPase reaction was terminated after 5 min. Unless otherwise indicated, assay temperature was 28°C. One unit of ATPase activity is the amount of enzyme which releases 1 mol of ADP or phosphate per minute. All assays were carried out with triplicate samples. Each experiment was repeated at least once. Data shown were from representative experiment. Addition of effectors to gizzard plasma membranes. For most experiments in which the selected effectors on enzyme activity were tested, preincubation with the membranes was carried out at a membrane protein concentration of 0.5– 0.9 mg/ml prior to enzyme assays. ConA was dissolved in water (2 mg/ml). Unless otherwise stated, ConA was added to the membranes at the indicated ratios. After incubation for 60 min on ice at pH 7.3, aliquots of membranes (5–10 g) were taken for activity determination. DSS was dissolved in DMSO (100 mg/ml). Aliquots were added to the membranes to bring about the indicated DSS to protein (w/w) ratios. Incubation was carried out at pH 8.0 on ice for 60 min before being quenched with 50 mM Tris–Cl, pH 8.0 (final concentration). EIAA was dissolved in DMSO to make a 2 mM stock solution. Aliquots were added to the membranes to the indicated EIAA con-
centrations. Incubation was carried out at pH 8.0 on ice for 60 min before being quenched by 150 mM DTT. NBD-Cl and NBD-F were dissolved in DMSO to make a 0.1 M stock solution. Aliquots were added to the membranes to attain the indicated concentrations. The mixture was incubated on ice for 60 min at the indicated pH prior to being quenched with 50 mM Tris–Cl, pH 8.0 (final concentration). Xanthene derivatives were dissolved in DMSO, aliquots were added directly to the coupled enzyme assay mixture containing 5–10 g gizzard membranes. When DMSO was used as a solvent for the compounds tested, the final concentration of DMSO in the assay reaction mixtures did not exceed 5% (v/v) and resulted in less than 5% variance in activity. Experimental conditions that differed from those above are described in the figure legends.
RESULTS
General enzymatic properties of chicken gizzard smooth muscle ecto-ATPase. We previously showed by biochemical and immunblot analyses that the major E-ATPase in chicken gizzard smooth muscle plasma membranes was the 66-kDa ecto-ATPase (16). Similar to the chicken T-tubule ecto-ATPase (19), the gizzard ecto-ATPase displayed non-Michaelis–Menten behavior and ATPase activity decreased at substrate concentration greater than 3 mM. In standard ATPase assays, 1 mM ATP was used. The ecto-ATPase activity increased with temperature from 15°C to approximately 28°C but declined at higher temperatures (Fig. 1). At the physiological temperature of the chicken, ⬃40°C, the ecto-ATPase activity was only 30% of the maximal activity. The decrease of activity in the range of physiological temperatures is not due to irreversible inactivation or denaturation of the enzyme. Gizzard membranes could be maintained at 40°C for 1 h with little loss of activity and subsequently still display the same temperature dependence as shown in Fig. 1. Activators of chicken smooth muscle ecto-ATPase. The gizzard ecto-ATPase activity was increased in the presence of certain lectins (21) and chemical crosslinking agents (16, 22). Cunningham et al. found a sixfold increase in ATPase activity when the gizzard membranes were preincubated at 37°C for 10 min with ConA, lentil lectin, wheat germ agglutinin, and Phaeseolus vulgaris erythroagglutinin at a lectin/membrane protein ratio of 2:1 (21). We examined the effect of ConA over a broader range of temperature. Figure 1 shows that in contrast to the activity obtained in the absence of ConA, which decreased at temperatures greater than 28°C, the ecto-ATPase activity continued to increase up to 45°C in the presence of ConA. Therefore, while stimulation by ConA was usually two- to threefold at 28°C, it was markedly higher at physiological temperatures. ConA also had a marked effect on the kinetic behavior of the ecto-ATPase. It abolished the time-dependent decrease of activity during the assay and saturation kinetics was evident (data not shown). Only one K m value, 52 ⫾ 1 M, was obtained,
REGULATION OF CHICKEN ECTO-ATPase ACTIVITY
FIG. 1. Temperature dependence of chicken gizzard ecto-ATPase activity in the absence and presence of activators. Chicken gizzard membranes (0.5 mg/ml) were incubated with ConA at a ConA/membrane ratio of 2:1 (w/w), with 75 M EIAA, or with DSS at a DSS/ membrane ratio of 5:1 for 60 min at 0°C at pH 7.3. ATPase activity was determined at the indicated temperatures with 5 g membrane protein using the coupled enzyme assay. The specific activity obtained at 15°C, 0.5 mol/min/mg, was taken to be the 100% activity.
which was intermediate between the two K m values, 19 ⫾ 9 M and 200 ⫾ 10 M, obtained in the absence of ConA (20). In contrast to the tetravalent ConA, preincubation of the membranes with the divalent succinyl-ConA increased the ecto-ATPase activity by less than 50% (data not shown). Thus the higher order oligomer formed in the presence of ConA was more active than the lower order oligomer formed in the presence of succinyl-ConA. Figure 1 also shows that the gizzard ecto-ATPase activity was stimulated by two other compounds, DSS and eosin iodoacetamide. DSS, a lysine specific chemical cross-linking agent, also increased the ectoATPase activity and altered the kinetic behavior. A K m of 76 ⫾ 2 M was obtained with DSS-treated membranes. However, the stimulatory effect of DSS at 40°C was approximately fivefold, which was less than that obtained with ConA . Eosin iodoacetamide (EIAA), a fluorescein iodoacetamide, which forms thioether bond with cysteine residues at pH ⬎7.5, activated the gizzard ATPase activity. Similar to ConA and DSS, pretreatment of the membranes with 75 M EIAA altered the kinetic behavior of the ecto-ATPase and a single K m of 96 ⫾ 1 M was obtained.
109
The effect of the activators was not additive. Figure 2A shows that at suboptimal DSS:membrane ratios, the maximal activation obtained upon subsequent addition of ConA was similar. At the optimal DSS:protein ratio of 5:1, the effect of ConA on the DSS-treated membrane was slightly greater than stimulation by ConA alone. Figure 2B shows that maximal stimulation by EIAA was less than the stimulation by ConA alone. Furthermore, incubation of the membranes with both EIAA and ConA did not result in greater stimulation than that due to ConA alone. Inhibitors of chicken smooth muscle ecto-ATPase. Results from this and other laboratories have shown that E-ATPases are not affected by the specific inhibitors of the F-, P-, and V-type ATPases. However, both ecto-ATPases and ecto-ATP/Dases are inhibited by diethyl pyrocarbonate and inhibition can be reversed by hydroxylamine (16, 23–26). In addition, inhibition by diethyl pyrocarbonate is reduced if the enzymes are preincubated with substrates (23, 26), indicating that the enzyme–substrate complex is refractile to DEPC modification. Taken together, these results suggest that the most likely target of modification is the sole histidine residue in one of the five apyrase consensus regions (27) present in all E-ATPases. We found, however, that the following compounds with different modes of action inhibited the chicken gizzard ectoATPase and not the chicken liver ecto-ATP/Dase. More significantly, their inhibitory effects were not reduced by preincubation of the enzyme with ATP, suggesting that their interaction with the gizzard ecto-ATPase occurs at sites other than the catalytic site. Mercurials. Figure 3A shows that pCMPS, which covalently modifies cysteine residues, inhibited the gizzard ecto-ATPase in a pH-dependent manner. Greater inhibition was obtained in solutions of lower pH. At pH 6.4, the ATPase activity was inhibited by 50 – 60% at pCMPS concentration greater than 10 M, whereas negligible inhibition was observed at pH 8.17. pHMB was also effective in similar concentration range as pCMPS at pH 6.4, whereas HgCl 2 was less effective (data not shown). Interestingly, significant ConA activation of ATPase was still observed with membranes pretreated with pCMPS (Fig. 3B). Furthermore, the inhibitory effect of pCMPS was markedly reduced if it was added to membranes preincubated with ConA (Fig. 3B), suggesting that ConA has a protective effect against pCMPS inhibition. Suramin. Suramin, a polysulfonated naphthylamine, has been shown to be an inhibitor of several ion-motive ATPases (28 –30), an antagonist of the P2 receptors (31), and an inhibitor of ecto-ATPases (32– 36). Figure 4A shows that suramin inhibited the gizzard ecto-ATPase also in a pH dependent manner. Like the mercurials, inhibition was greater at lower pH.
110
CALDWELL ET AL.
FIG. 2. Effect of ConA on the ecto-ATPase activities of chicken gizzard membranes treated with DSS or EIAA. (A) DSS was added to the membranes (0.9 mg/ml) at the indicated DSS/protein ratios in 20 mM Mops, pH 8. After incubation at 0°C for 60 min, the reaction was terminated by the addition of Tris–Cl, pH 8, to a final concentration of 50 mM. ConA was then added to the untreated or DSS-treated membranes at a ConA/protein ratio of 2:1. After further incubation for 60 min at 0°C, ATPase activity was determined in the coupled enzyme assay with 5 g membrane protein. (B) Experimental conditions were similar to that described in (A) except EIAA at the indicated concentrations was used in the pretreatment.
Inhibition obtained with 50 M suramin at pH 6.4 was ⬃75% but was 40% at pH 8.17. Inhibition by suramin was similar whether added before or after ConA treatment (Fig. 4B). Therefore, in contrast to pCMPS, ConA pretreatment of the membranes did not prevent the ecto-ATPase from inhibition by suramin. Halo-NBDs. NBD-Cl and NBD-F possess a reactive halogen which can be readily replaced by nucleophiles, e.g., -NH 2, -OH, and -SH, with the formation of 7-substituted 4-nitrobenzofurazan derivatives. NBD-Cl inactivates the mitochondrial ATPase (37), Na,K-pump ATPase (38), Ca-pump ATPase (39), and others. Modification occurs at cysteine or tyrosine residues. However, migration of NBD from tyrosine to lysine residue has been reported (40). Inhibition of the gizzard ecto-ATPase by NBD-Cl and NBD-F was more effective at higher pH (Fig. 5A) and was time-dependent (Fig. 5B). Inhibition by 1 mM NBD-Cl was ⬃10% at pH 6, but was ⬃65% at pH 8.0, which is most likely due to the greater nucleophilicity of the amino acid residues involved in the reaction. Inhibition by halo-NBD was not reversible by the addition of dithiothreitol or glutathione (data not shown), excluding cysteine as the target of modification. Preliminary data indicated the involvement of a tyrosine residue.
In contrast to the ion-motive ATPases affected by NBD-Cl, inhibition of the gizzard ecto-ATPase by NBD-Cl is not protected by the substrate, ATP (23), suggesting that the NBD-Cl reactive site is not at the catalytic site. On the other hand, Fig. 5B shows that ConA also protects the ecto-ATPase from inhibition by NBD-Cl, further indicating its action at a regulatory site. Xanthene derivatives. Tetraiodofluorescein has been shown to interact with the nucleotide binding sites on a number of ATP utilizing enzymes (41) and is an inhibitor of the chicken skeletal T-tubule ectoATPase (42). We examined the effect of several related xanthene derivatives on the gizzard ecto-ATPase. The structure of the parent fluorescein, substituents on the xanthene molecule of the different derivatives, and the Kiapp are shown in Table I. The data inidicate that the relatively hydrophobic parent molecule, fluorescein, was the least effective inhibitor. Inhibition increased as the number of polarizable halogen and the volume of the inhibitors increased. Rose bengal, which has a volume of 461.8 Å 3, had a Kiapp that was 2 orders of magnitude lower than fluorescein which has a volume of 273.3 Å 3. Dixon plot analyses indicated that inhibition by the xanthene derivatives was of a mixed nature. Therefore, inhibition occurs by interaction at both
REGULATION OF CHICKEN ECTO-ATPase ACTIVITY
111
FIG. 3. (A) Inhibition of chicken gizzard ecto-ATPase activity by pCMPS at different pH. Chicken gizzard membranes (20 g) were preincubated in 0.45 ml of a solution containing 12.5 mol buffer (Mops for pH 6.4 and Tris for pH 7.18 and 8.17), 2.5 mol MgCl 2 and the indicated concentrations of pCMPS for 5 min at room temperature. Reactions were started by the addition of 0.5 mol of ATP. After 10 min at 28°C, the reaction was terminated by the addition of 0.1 ml 10% TCA. After centrifugation to remove the protein precipitate, an aliquot of the supernatant was used for phosphate determination by the colorimetric method. (B) Effect of ConA on gizzard ecto-ATPase activity added before or after pCMPS. Preincubation of chicken gizzard membranes (20 g) with the indicated concentrations of pCMPS was carried out similar to that described in (A) at pH 6.4. pCMPS was either added alone or added before or after ConA (100 g). Incubation time with pCMPS and ConA was 5 min each.
the catalytic and a regulatory site and that the extent of inhibition is dependent on both the bulk and the polarity of the xanthene derivatives. The effectiveness of the xanthene derivatives decreased if the membranes were preincubated with the activators. The K 0.5 of tetraiodofluorescein (erythrosin in Table I), 1.2 ⫾ 0.1 M, was increased to 3.8 ⫾ 0.3 M, 5.1 ⫾ 0.1 M, and 6.8 ⫾ 0.7 M when the membranes were preincubated with ConA, EIAA, and DSS, respectively. Detergents. Many commonly used detergents for isolating membrane proteins have a deleterious effect on a number of E-ATPases with some exceptions, e.g., the chicken oviduct and liver ecto-ATP/Dases (7, 43). Previously, we showed that the gizzard ecto-ATPase is inhibited by Triton X-100 (20, 21), NP-40, and lysophosphatidylcholine (20) at concentrations below the CMC of these detergents. Digitonin is the only detergent which has a stimulatory effect on the gizzard ecto-ATPase (21), and has been used in its solubilization and purification (44). At sub-CMC concentrations of Triton X-100, the gizzard ecto-ATPase appears to be more sensitive to inhibition than the skeletal muscle T-tubule ecto-ATPase (21), however, pretreatment of the membranes with ConA prevented inhibition of both
enzymes by Triton X-100 (data not shown). Furthermore, Fig. 6 shows that ConA can restore the majority of the lectin-stimulated activity in Triton X-100treated membranes. In the absence of ConA, the ectoATPase activity of the membrane treated with Triton X-100 at a ratio of 20:1 was only 25% of the untreated membrane. Incubation of the Triton X-100-treated membranes with ConA restored the activity to approximately 90% of the lectin-stimulated activity. At higher concentrations of Triton X-100, the detergent interfered with the protein–ConA interaction, and restoration was significantly reduced. DISCUSSION
Previous work from this and other laboratories showed that the ATPase activity of avian skeletal muscle T-tubule membranes (19, 21, 45) and gizzard smooth muscle membranes (20, 21) displayed unusual enzymatic properties and may have a complex regulatory mechanism (42). The enzymes responsible for the activity are members of a subfamily of the E-ATPases, the ecto-ATPases. In a recent study, we showed that the activity of the gizzard ecto-ATPase activity is af-
112
CALDWELL ET AL.
FIG. 4. (A) Inhibition of chicken gizzard ecto-ATPase activity by suramin at different pH. Experimental conditions were similar to that described in the legend of Fig. 3A. (B) Effect of ConA on gizzard ecto-ATPase activity added before or after suramin. Experimental conditions were similar to that described in the legend of Fig. 3B.
fected by several modulators, while that of another E-ATPase, the liver ecto-ATP/Dase, is not (16). The previous results suggest that reagents which promote protein oligomerization, i.e., ConA and DSS, increase the gizzard ecto-ATPase activity, whereas compounds which disrupt oligomerization, i.e., detergents at concentrations below their CMC, decrease its activity. In addition to smooth and skeletal muscle ecto-ATPases, brain and some tumor cell ecto-ATPases are similarly affected by ConA and detergents (21, 46 – 48), suggesting a common regulatory mechanism for the ectoATPases. In this study,we have examined the effect of other compounds that activate or inhibit the gizzard ecto-ATPase individually and in conjunction with ConA. These modulators are not only useful in distinguishing ecto-ATPase and ecto ATP/Dase activities, they have also provided further insight into the regulation of the smooth muscle ecto-ATPase. Among the inhibitory modulators, the mercurials and NBD-Cl were previously shown to inhibit ion-motive ATPases by modifying amino acid residues near the catalytic site. When tested on the gizzard ectoATPase and liver ecto-ATP/Dase, these compounds inhibit only the former and not the latter. Since all EATPases contain five conserved regions (the apyrase consensus regions) that have been shown to be important for catalytic activity (49, 50), the differential effects of pCMPS and NBD-Cl on the two E-ATPases
suggest that these compounds are not directly affecting the catalytic site of the gizzard ecto-ATPase. Furthermore, unlike diethyl pyrocarbonate, which inhibits all E-ATPases and whose inhibitory effect is abolished by prior addition of substrates (23, 26), ATP does not afford protection to the ecto-ATPase against inhibition by pCMPS and NBD-Cl. These results further support the conclusion that their reaction with the enzyme occurs at a regulatory site. The lack of protective effect by ATP was also observed for the other modulators used in this study, i.e., suramin, the xanthene derivatives, and Triton X-100, all amphiphiles that are capable of interfering with membrane protein interactions. The regulatory effects of the selected activating and inhibitory modulators are now incorporated in a model that is based on the thesis that the activity of the gizzard ecto-ATPase is determined by its oligomerization state. The model in Fig. 7 shows that, at physiological temperatures, the “normal” oligomerization state in the native membranes is composed of a mixed population of monomers and dimers. In the presence of the divalent succinyl-ConA, the enzyme exists mostly as dimers and a small increase in activity is obtained. EIAA activates the ecto-ATPase activity to the same extent as succinyl-ConA, suggesting that it also promotes formation of mostly dimers, probably facilitated by the ability of the eosin moiety to form stacked dimers (53). Conversion of the enzyme to a highly oli-
REGULATION OF CHICKEN ECTO-ATPase ACTIVITY
113
FIG. 5. (A) Inhibition of gizzard ecto-ATPase by NBD-Cl and NBD-F at different pH. Chicken gizzard membranes (15 g) were incubated with or without 1 mM NBD-Cl and NBD-F at the indicated pH for 60 min at 0°C. The buffers used were Mes for pH 6.0 – 6.9 and Hepes for pH 6.9 – 8.6. After the reaction was terminated by 50 mM Tris–Cl, pH 7.3, 5–10 g of the membrane protein were used in the coupled enzyme assay. (B) Effect of NBD-Cl on gizzard ecto-ATPase activity without and with preincubation with ConA. Gizzard membranes were incubated on ice for 60 min without or with ConA (2:1). After adjusting the pH to 8.0, NBD-Cl was added to the membranes at 1 and 10 mM and incubation continued on ice. The reaction was quenched at the indicated time with 50 mM Tris–Cl, pH 7.3, and 5–10 g membrane protein were used for activity determination.
gomeric state by DSS and the tetravalent ConA gives rise to higher activity. Oligomerization promoted by EIAA, DSS, and ConA not only alters the reaction rate, but also the apparent K m for ATP, indicating structural changes induced by these compounds affect the interaction of the enzyme with the substrate. The fact that ecto-ATPase activity of the gizzard membranes is low at physiological temperatures (37– 40°C), but can be markedly increased due to oligomerization at the same temperatures, suggests that the gizzard ecto-ATPase activity is probably not constant in vivo, but can vary as determined by physiological events which affects oligomerization of the enzyme. In the model of Fig. 7, we also indicate that detergents and other covalent and noncovalent inhibitors decrease enzyme activity by promoting the state where the enzyme exists mostly as monomers. Other than the halo-NBDs, the noncovalent reagents are characterized by their amphiphilicity. This is most clearly illustrated by the xanthene derivatives. Derivatives of greater volume and polarity are more effective inhibitors than the parent molecule, fluorescein (Table I). These compounds most likely affect the interaction of the hydrophobic regions of the enzyme. In this regard, it is of interest that Wang et al. showed that deletion of either of the two transmembranous regions of the rat
CD39 ecto-ATP/Dase resulted in a decrease of activity and evidence was presented that the oligomer formed by the native CD39 protein is a tetramer (54). Of the other inhibitory modulators, the effect of pCMPS is of particular interest. This sulfhydryl reagent modifies proteins by reacting with cysteine residues. All E-ATPases have ten conserved cysteine residues, which may be involved in the formation of disulfide bonds that are prevalent in most cell surface proteins. Examination of the sequences of five cloned ectoATPases, including the chicken gizzard ecto-ATPase, reveals an additional conserved cysteine residue close to the extracellular membrane surface near the Nterminus. We propose that this is the most likely target of pCMPS and that its modification interferes with oligomerization, resulting in inhibition. The inhibitory effect of pCMPS was reversed to a significant degree by the subsequent addition of ConA and inhibition by pCMPS was negligible if the membranes were pretreated by ConA (Fig. 3B). Thus, pCMPS modification does not appear to interfere with oligomerization induced by ConA which binds to glycans of a glycoprotein. The involvement of the particular cysteine residue in the action of pCMPS will be investigated in future mutagenesis studies.
114
CALDWELL ET AL. TABLE I
Structure, Volume, and K i app of the Inhibitory Xanthene Derivatives
Xanthene 1. 2. 3. 4. 5. 6. 7. 8.
Fluorescein Dichlorofluorescein Dibromofluorescein Diiodofluorescein Eosin Y Phloxine B Erythrosin B Rose Bengal
R1
R2
R3
Volume (Å 3)
K i app (M)
OH OH OH OH OBr OBr OI OI
OH OCl OBr OI OBr OBr OI OI
OH OH OH OH OH OCl OH OCl
273.3 300.6 311.2 340.0 349.9 404.9 406.7 461.8
124 100 35.9 9.30 6.10 1.50 1.20 0.68
Note. Kiapp were obtained from Dixon plot analyses using data obtained at two ATP concentrations (0.25 and 1 mM) and suitable range of inhibitor concentrations. Enzyme activity was determined using the coupled enzyme assay. Molecular volume determination was carried out on a silicon Graphics workstation using SPARTAN 4.1.
The inhibitory effect of the nucleotide mimetics, NBD-Cl and NBD-F, on the gizzard ecto-ATPase differs from that on the ion-motive ATPases, such as the mitochondrial ATPase. The failure of ATP in protecting the gizzard ecto-ATPase from inhibition by NBD-Cl or NBD-F (23) suggests that they react with amino acid residues distant from the active site but are involved in oligomerization of the ectoATPase molecule. This conclusion is supported by the observation that their inhibitory effect is abolished by ConA. While preliminary results suggest their reaction with a tyrosine residue, future analysis using purified and expressed protein should reveal the precise target of their modification. In contrast to pCMPS, NBD-Cl, and the xanthene derivatives, the effect of suramin is independent of ConA. Inhibition by suramin is the same whether added before of after ConA, suggesting that ConApromoted oligomerization of the gizzard ecto-ATPase does not prevent interaction of the enzyme with suramin. Suramin affects many ATP-utilizing enzymes and is a potent antagonist of the purinergic receptors (31). It has been suggested that suramin may compete with ATP for its binding site on the receptor. However, the fact that suramin does not inhibit ecto-ATP/Dases (36, 43), argues against a direct effect on the catalytic site of ecto-ATPase. At
FIG. 6. Restoration of ecto-ATPase activity in Triton X-100-treated gizzard membranes by ConA. Gizzard membranes were incubated with Triton X-100 at the indicated protein/Triton X-100 ratios for 60 min at 0°C. Final Triton X-100 concentrations ranged from 27 to 81 M. ConA was then added at a ConA/protein of 2:1 to the Triton X-100-treated membranes and incubated for another 60 min after which aliquots were taken for activity determination.
REGULATION OF CHICKEN ECTO-ATPase ACTIVITY
FIG. 7.
115
Scheme of proposed regulation of chicken gizzard ecto-ATPase activity by activators and inhibitors.
present, the precise mechanism of action of suramin is unknown (55). It will be necessary to conduct further studies to determine if it inhibits the ectoATPase by interfering with its oligomerization or by an unrelated mechanism. In summary, we have shown that the activity of the gizzard smooth muscle ecto-ATPase, in contrast to the
liver ecto-ATP/Dase can be easily altered. The effect of most of the modulators in this study can be explained by their influence on the oligomerization state of the enzyme molecule. Since many of the ecto-ATPases occur in excitable tissues, the facile regulation of this enzyme is probably of physiological significance which remains to be determined.
116
CALDWELL ET AL.
ACKNOWLEDGMENTS We gratefully acknowledge the contribution of Dr. A. S. Dahms to this study. This work was supported in part by the California Metabolic Research Foundation (A.S.D. and A.F.K.) and the Rees-Stealy Research Foundation (A.S.D.). C. C. Caldwell is a recipient of the Arne N. Wick predoctoral research fellowship.
REFERENCES 1. Redegeld, F. A., Caldwell, C. C., and Sitkovsky, M. V. (1999) Trends Pharmacol. Sci. 20, 453– 459. 2. Zimmermann, H. (1999) Trends Pharmacol. Sci. 20, 231–235. 3. Kirley, T. L., Gerber, L. K., and Smith, T. M. (1999) IUBMB Life 48, 67–72. 4. Kegel, B., Braun, N., Heine, P., Maliszewski, C. R., and Zimmermann, H. (1997) Neuropharmacology 26, 1189 –1200. 5. Mateo, J., Harden, T. K., and Boyer, J. L. (1999) Br. J. Pharmacol. 128, 396 – 402. 6. Christoforidis, S., Papamarcaki, T., Galaris, D., Kellner, R., and Tsolas, O. (1995) Eur. J. Biochem. 234, 66 –74. 7. Strobel, R. S., Nagy, A., Knowles, A. F., Buegel, J., and Rosenberg, M. D. (1996) J. Biol. Chem. 271, 16232–16331. 8. Lewis-Carl, S., and Kirley, T. L. (1997) J. Biol. Chem. 272, 23645–23652. 9. Wang, T. F., and Guidotti, G. (1996) J. Biol. Chem. 271, 9898 –9901. 10. Kaczmarek, E., Koziak, K., Se´vigny, J., Siegel, J. B., Anrather, J., Beaudoin, A., Bach, F. H., and Robson, S. C. (1996) J. Biol. Chem. 271, 33116 –33122. 11. Smith, T. M., and Kirley, T. L. (1998) Biochim. Biophys. Acta 1386, 65–78. 12. Se´vigny, J., Robson, S., Waelkens, E., Csizmadia, E., Smith, R. N., and Lemmens, R. (2000) J. Biol. Chem. 275, 5640 –5647. 13. Lemmens, R., Vanduffel, L., Kittel, A., Beaudoin, A. R., Benrezzak, O., and Se´vigny, J. (2000) Eur. J. Biochem. 267, 4106 – 4114. 14. Knowles, A. F., Isler, R. E., and Reece, J. F. (1983) Biochim. Biophys. Acta 731, 88 –96. 15. Beaudoin, A. R., Se´vigny, J., and Picher, M. (1996) in Biomembranes, (A. G. Lee, Ed.), Vol. 5, pp. 369 – 401, JAI Press. 16. Caldwell, C. C., Davis, M. D., and Knowles, A. F. (1999) Arch. Biochem. Biophys. 362, 46 –58. 17. Knowles, A. F., and Nagy, A. K. (1999) Eur. J. Biochem. 262, 349 –357. 18. Kang, J-J., Cunningham, H. B., Jachec, C., Priest, A., Dahms, A. S., and Sabbadini, R. A. (1991) Arch. Biochem. Biophys. 290, 214 –223. 19. Moulton, M. P., Sabbadini, R. A., Norton, K. C., and Dhams, A. S. (1986) J. Biol. Chem. 261, 12244 –12251. 20. Caldwell, C., Norman, V., Urbina, A., Jarvis, A., Quinonez, C., Stemm, M., and Dahms, A. S. (1997) Biochem. Biophys. Res. Commun. 238, 728 –732. 21. Cunningham, H. B., Yazaki, P. J., Domingo, R., C., Oades, K. V., Bohlen, H., Sabbadini, R. A., and Dahms, A. S. (1993) Arch. Biochem. Biophys. 303, 32– 43. 22. Stout, J. G., and Kirley, T. L. (1996) Biochemistry 35, 8289 – 8298. 23. Hornyak, S. C. (1999) M.S. thesis, San Diego State University. 24. Kirley, T. L. (1988) J. Biol. Chem. 263, 12682–12689. 25. Saborido, A., Moro, G., and Megias, A. (1991) J. Biol. Chem. 266, 23490 –23498. 26. Dzhandzhugazyan, K. N., and Plesner, L. (2000) Biochim. Biophys. Acta 1466, 267–277.
27. Handa, M., and Guidotti, G. (1996) Biochem. Biophys. Res. Commun. 218, 916 –923. 28. Fortes, P. A. G., Ellory, J. C., and Lew, V. L. (1973) Biochim. Biophys. Acta 318, 262–272. 29. Calcaterra, N. B., Vicario, L. R., and Roveri, O. A. (1988) Biochem. Pharmacol. 37, 252–257. 30. Moriyama, Y., and Nelson, N. (1988) FEBS Lett. 234, 383–386. 31. Hoyle, C. H. V., Knight, G. E., and Burnstock, G. (1990) Br. J. Pharmacol. 99, 617– 621. 32. Crack, B. E., Beukers, M. W., McKechnie, K. C. W., IJzerman, A. P., and Leff, P. (1994) Br. J. Pharmacol. 113, 1432–1438. 33. Stout, J. G., and Kirley, T. L. (1995) Biochem. Mol. Biol. Int. 36, 927–934. 34. Chen, B. C., Lee, C-M., and Lin, W-W. (1996) Br. J. Pharmacol. 119, 1628 –1634. 35. Delgado, J., Moro, G., Saborido, A., and Megias, A. (1997) Biochem. J. 327, 899 –907. 36. Heine, P., Braun, N., Heilbronn, A., and Zimmermann, H. (1999) Eur. J. Biochem. 262, 102–107. 37. Ferguson, S. J., Lloyd, W. J., Lyons, M. H., and Radda, G. K. (1975) Eur. J. Biochem. 54, 117–126. 38. Cantley, L. C., Gelles, J., and Josephson, L. (1978) Biochemistry 18, 418 – 425. 39. Bailin, G. (1994) Biochim. Biophys. Acta 1208, 197–203. 40. Ferguson, S. J., Lloyd, W. J., and Radda, G. K. (1975) Eur. J. Biochem. 54, 127–133. 41. Neslund, G. G., Miara, J. G., Kang, J.-J., and Dahms, A. S. (1984) Curr. Top. Cell Regul. 24, 444 – 469. 42. Sabbadini, R. A., and Dahms, A. S. (1989) J. Bioenerg. Biomembr. 21, 163–213. 43. Knowles, A. F., Strobel, R. S., Nagy, A. K., and Wu-Weis, M. In preparation. 44. Stout, J. G., and Kirley, T. L. (1994) J. Biochem. Biophys. Methods 29, 61–75. 45. Megı´as, A., and Saborido, A. (1990) Arch. Biochem. Biophys. 278, 113–119. 46. Nagy, A. K., Knowles, A. F., and Treiman, L. J. (2000) in EctoATPases and Related Ectonucleotidases (L. Vanduffel and R. Lemmens, Eds.), pp. 53– 60, Shaker, Maastricht, The Netherlands. 47. Knowles, A. F., and Kaplan, N. O. (1981) Biochem. Biophys. Res. Commun. 99, 1443–1448. 48. Shi, X-J., and Knowles, A. F. (1994) Arch. Biochem. Biophys. 315, 177–184. 49. Smith, T. M., and Kirley, T. L. (1999) Biochemistry 38, 321– 328. 50. Schulte am Esch, II, J., Se´vigny, J., Kaczmarek, E., Siegel, J. B., Imai, M., Koziak, K., Beaudoin, A. R., and Robson, S. C. (1999) Biochemistry 38, 2248 –2258. 51. Beeler, T. J., Gable, K., and Keffer, J. (1983) Biochim. Biophys. Acta 734, 221–234. 52. Stout, J. G., and Kirley, T. L. (1996) Biochemistry 35, 8289 – 8298. 53. Ajtai, K., Ilich, P., Ringler, A., Sedarous, S., Toft, D., and Burghardt, T. (1992) Biochemistry 31, 12431–12440. 54. Wang, T-F., Ou, Y., and Guidotti, G. (1998) J. Biol. Chem. 273, 24814 –24821. 55. Voogd, T. E., Vansterkenburg, E. L. M., Wilting, J., and Janssen, L. H. (1993) Pharmacol. Rev. 45, 177–203.
Regulation of Chicken Gizzard Ecto-ATPase Activity by Modulators That Affect Its Oligomerization Status Charles C. Caldwell, 1 Stephen C. Hornyak, Erik Pendleton, Dawn Campbell, and Aileen F. Knowles 2 Department of Chemistry, San Diego State University, San Diego, California 92182-1030
Received September 13, 2000, and in revised form November 10, 2000; published online January 26, 2001
The major ectonucleoside triphosphate phosphohydrolase in the chicken gizzard smooth muscle membranes is an ecto-ATPase, an integral membrane glycoprotein belonging to the E-ATPase (or E-NTPDase) family. The gizzard ecto-ATPase is distinguished by its unusual kinetic properties, temperature dependence, and response to a variety of modulators. Compounds that promote oligomerization of the enzyme protein, i.e., concanavalin A, chemical cross-linking agent, and eosin iodoacetamide, increase its activity. Compounds that inhibit some ion-motive ATPases, e.g., sulfhydryl reagents, xanthene derivatives, NBD-halides, and suramin, also inhibit the gizzard ecto-ATPase, but not another E-ATPase, the chicken liver ecto-ATP-diphosphohydrolase, which contains the same conserved regions as the ecto-ATPase. Furthermore, inhibition of the gizzard ecto-ATPase by these compounds as well as detergents is not prevented by preincubation of the membranes with the substrate, ATP, indicating that their interaction with the enzyme occurs at a locus other than the catalytic site. On the other hand, the inhibitory effect of these compounds, except suramin, is abolished or reduced if the membranes are preincubated with concanavalin A. It is concluded that these structurally unrelated modulators exert their effect by interfering with the oligomerization of the ectoATPase protein. Our findings suggest that, under physiological conditions, the gizzard smooth muscle ecto-ATPase may exhibit a range of activities determined by membrane events that affect the status of oligomerization of the enzyme. © 2001 Academic Press Key Words: ecto-ATPase; E-NTPDase; gizzard; oligomerization.
1 Present address: Laboratory of Immunology, NIAID, NIH, Bethesda, MD 20892-0001. 2 To whom correspondence and reprint requests should be addressed. Fax: (619)594-4634. E-mail: [email protected].
0003-9861/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
E-ATPases (or E-NTPDases) are membrane-bound cell surface phosphohydrolases that have broad substrate specificity. While they hydrolyze most nucleoside triphosphates (NTP) 3 and nucleoside diphosphates (NDP), their physiological substrates are most likely extracellular ATP, UTP, and ADP, ligands of the purinergic receptors. Extracellular ATP is also a substrate for ecto-protein kinases (1). To date, more than a dozen E-ATPases have been cloned. The deduced primary sequences indicate that they are integral membrane proteins of approximately 54 kDa with two transmembranous segments near the N- and C-termini, and all contain five conserved regions (2). Based on the extent of their sequence homology, three subfamilies can be discerned, the ecto-ATPases, the CD39s, and the ecto-ATP/Dases. Expressed ectoATPases from smooth muscle, brain, and tumor cells exhibit low activity toward NDP (3–5). On the other hand, CD39s and ecto-ATP/Dases, whether purified or expressed from their cDNAs, hydrolyze NDP as well as NTP (6 –13). However, CD39s are the vascular ectoATP/Dases (6, 9, 10, 13) and are distinct from the other characterized ecto-ATP/Dases (7, 8, 11, 12) both in the extent of sequence identity and tissue distribution. In addition to their different abilities to hydrolyze NDP, the ecto-ATPases and the ecto ATP/Dases differ in their responses to a variety of modulators. High concentrations of azide inhibit most ecto-ATP/Dases but not ecto-ATPases (14 –16). The inhibitory mechanism of azide on the purified chicken oviduct ecto-ATP/ 3
Abbreviations used: ATP/Dase, ATP diphosphohydrolase; ConA, concanavalin A; DSS, disuccinimidyl suberate; DMSO, dimethyl sulfoxide; EIAA, eosin iodoacetamide; NBD-Cl, 7-chloro-4-nitrobenz2-oxa-1,3-diazole; NBD-F, 7-fluoro-4-nitrobenz-2-oxa-1,3-diazole; pCMPS, p-chloromercuripheylsulfonate; pHMB, p-hydroxymercuribenzoate; T-tubule, transverse tubule, CMC, critical micelle concentration; NTP, nucleoside triphosphate; NDP, nucleoside diphosphate. 107
108
CALDWELL ET AL.
Dase has been investigated (17). On the other hand, we showed that the activity of the chicken gizzard smooth muscle ecto-ATPase is affected by several compounds that have no effect on the chicken liver ecto-ATP/Dase (16). Therefore, in spite of the presence of the same conserved regions in their sequences, significant differences exist between the ecto-ATPases and the ectoATP/Dases with respect to their interaction with nucleotide substrates and modulators. The different effects of the modulators suggest different regulatory mechanisms of the two enzymes. Further characterization of these differences will be important in future structure–function studies of the E-ATPases. In this report, we describe the effect of an extensive list of modulators on the chicken gizzard ecto-ATPase. Based on these results, a model of the regulation of the gizzard ecto-ATPase activity is presented. MATERIALS AND METHODS Materials. DSS was purchased from Pierce (Rockford, IL). EIAA, NBD-Cl, and NBD-F were purchased from Molecular Probes (Eugene, OR). Xanthene derivatives were purchased from Aldrich (Milwaukee, WI). All other reagents were purchased from Sigma Chemical Company (St. Louis, MO). Preparation of chicken gizzard plasma membranes. Chicken gizzards were obtained within 16 h of slaughter. Preparation of the plasma membranes from the tissue homogenate by differential and sucrose gradient centrifugation was described previously (16). The membrane preparations (1–3 mg protein/ml) were stored in 20 mM Mops, pH 7.3, 30% sucrose at ⫺20°C until use. ATPase activity was stable for several months. Protein concentration of the plasma membranes was determined using the Bio-Rad DC protein assay reagent with bovine serum albumin as standard. ATPase assays. ATPase activity was determined either by the coupled enzyme assay which determined the rate of ADP release (18) or by a colorimetric method that determined the rate of phosphate release (16). Because of the nonlinear nature of the time course, the decrease of absorbance at 340 nm of NADH between 1 and 2 min after the initiation of the reaction by ATP in the coupled enzyme assay was used to calculate the rate of ATP hydrolysis. In the colorimetric assay, the ATPase reaction was terminated after 5 min. Unless otherwise indicated, assay temperature was 28°C. One unit of ATPase activity is the amount of enzyme which releases 1 mol of ADP or phosphate per minute. All assays were carried out with triplicate samples. Each experiment was repeated at least once. Data shown were from representative experiment. Addition of effectors to gizzard plasma membranes. For most experiments in which the selected effectors on enzyme activity were tested, preincubation with the membranes was carried out at a membrane protein concentration of 0.5– 0.9 mg/ml prior to enzyme assays. ConA was dissolved in water (2 mg/ml). Unless otherwise stated, ConA was added to the membranes at the indicated ratios. After incubation for 60 min on ice at pH 7.3, aliquots of membranes (5–10 g) were taken for activity determination. DSS was dissolved in DMSO (100 mg/ml). Aliquots were added to the membranes to bring about the indicated DSS to protein (w/w) ratios. Incubation was carried out at pH 8.0 on ice for 60 min before being quenched with 50 mM Tris–Cl, pH 8.0 (final concentration). EIAA was dissolved in DMSO to make a 2 mM stock solution. Aliquots were added to the membranes to the indicated EIAA con-
centrations. Incubation was carried out at pH 8.0 on ice for 60 min before being quenched by 150 mM DTT. NBD-Cl and NBD-F were dissolved in DMSO to make a 0.1 M stock solution. Aliquots were added to the membranes to attain the indicated concentrations. The mixture was incubated on ice for 60 min at the indicated pH prior to being quenched with 50 mM Tris–Cl, pH 8.0 (final concentration). Xanthene derivatives were dissolved in DMSO, aliquots were added directly to the coupled enzyme assay mixture containing 5–10 g gizzard membranes. When DMSO was used as a solvent for the compounds tested, the final concentration of DMSO in the assay reaction mixtures did not exceed 5% (v/v) and resulted in less than 5% variance in activity. Experimental conditions that differed from those above are described in the figure legends.
RESULTS
General enzymatic properties of chicken gizzard smooth muscle ecto-ATPase. We previously showed by biochemical and immunblot analyses that the major E-ATPase in chicken gizzard smooth muscle plasma membranes was the 66-kDa ecto-ATPase (16). Similar to the chicken T-tubule ecto-ATPase (19), the gizzard ecto-ATPase displayed non-Michaelis–Menten behavior and ATPase activity decreased at substrate concentration greater than 3 mM. In standard ATPase assays, 1 mM ATP was used. The ecto-ATPase activity increased with temperature from 15°C to approximately 28°C but declined at higher temperatures (Fig. 1). At the physiological temperature of the chicken, ⬃40°C, the ecto-ATPase activity was only 30% of the maximal activity. The decrease of activity in the range of physiological temperatures is not due to irreversible inactivation or denaturation of the enzyme. Gizzard membranes could be maintained at 40°C for 1 h with little loss of activity and subsequently still display the same temperature dependence as shown in Fig. 1. Activators of chicken smooth muscle ecto-ATPase. The gizzard ecto-ATPase activity was increased in the presence of certain lectins (21) and chemical crosslinking agents (16, 22). Cunningham et al. found a sixfold increase in ATPase activity when the gizzard membranes were preincubated at 37°C for 10 min with ConA, lentil lectin, wheat germ agglutinin, and Phaeseolus vulgaris erythroagglutinin at a lectin/membrane protein ratio of 2:1 (21). We examined the effect of ConA over a broader range of temperature. Figure 1 shows that in contrast to the activity obtained in the absence of ConA, which decreased at temperatures greater than 28°C, the ecto-ATPase activity continued to increase up to 45°C in the presence of ConA. Therefore, while stimulation by ConA was usually two- to threefold at 28°C, it was markedly higher at physiological temperatures. ConA also had a marked effect on the kinetic behavior of the ecto-ATPase. It abolished the time-dependent decrease of activity during the assay and saturation kinetics was evident (data not shown). Only one K m value, 52 ⫾ 1 M, was obtained,
REGULATION OF CHICKEN ECTO-ATPase ACTIVITY
FIG. 1. Temperature dependence of chicken gizzard ecto-ATPase activity in the absence and presence of activators. Chicken gizzard membranes (0.5 mg/ml) were incubated with ConA at a ConA/membrane ratio of 2:1 (w/w), with 75 M EIAA, or with DSS at a DSS/ membrane ratio of 5:1 for 60 min at 0°C at pH 7.3. ATPase activity was determined at the indicated temperatures with 5 g membrane protein using the coupled enzyme assay. The specific activity obtained at 15°C, 0.5 mol/min/mg, was taken to be the 100% activity.
which was intermediate between the two K m values, 19 ⫾ 9 M and 200 ⫾ 10 M, obtained in the absence of ConA (20). In contrast to the tetravalent ConA, preincubation of the membranes with the divalent succinyl-ConA increased the ecto-ATPase activity by less than 50% (data not shown). Thus the higher order oligomer formed in the presence of ConA was more active than the lower order oligomer formed in the presence of succinyl-ConA. Figure 1 also shows that the gizzard ecto-ATPase activity was stimulated by two other compounds, DSS and eosin iodoacetamide. DSS, a lysine specific chemical cross-linking agent, also increased the ectoATPase activity and altered the kinetic behavior. A K m of 76 ⫾ 2 M was obtained with DSS-treated membranes. However, the stimulatory effect of DSS at 40°C was approximately fivefold, which was less than that obtained with ConA . Eosin iodoacetamide (EIAA), a fluorescein iodoacetamide, which forms thioether bond with cysteine residues at pH ⬎7.5, activated the gizzard ATPase activity. Similar to ConA and DSS, pretreatment of the membranes with 75 M EIAA altered the kinetic behavior of the ecto-ATPase and a single K m of 96 ⫾ 1 M was obtained.
109
The effect of the activators was not additive. Figure 2A shows that at suboptimal DSS:membrane ratios, the maximal activation obtained upon subsequent addition of ConA was similar. At the optimal DSS:protein ratio of 5:1, the effect of ConA on the DSS-treated membrane was slightly greater than stimulation by ConA alone. Figure 2B shows that maximal stimulation by EIAA was less than the stimulation by ConA alone. Furthermore, incubation of the membranes with both EIAA and ConA did not result in greater stimulation than that due to ConA alone. Inhibitors of chicken smooth muscle ecto-ATPase. Results from this and other laboratories have shown that E-ATPases are not affected by the specific inhibitors of the F-, P-, and V-type ATPases. However, both ecto-ATPases and ecto-ATP/Dases are inhibited by diethyl pyrocarbonate and inhibition can be reversed by hydroxylamine (16, 23–26). In addition, inhibition by diethyl pyrocarbonate is reduced if the enzymes are preincubated with substrates (23, 26), indicating that the enzyme–substrate complex is refractile to DEPC modification. Taken together, these results suggest that the most likely target of modification is the sole histidine residue in one of the five apyrase consensus regions (27) present in all E-ATPases. We found, however, that the following compounds with different modes of action inhibited the chicken gizzard ectoATPase and not the chicken liver ecto-ATP/Dase. More significantly, their inhibitory effects were not reduced by preincubation of the enzyme with ATP, suggesting that their interaction with the gizzard ecto-ATPase occurs at sites other than the catalytic site. Mercurials. Figure 3A shows that pCMPS, which covalently modifies cysteine residues, inhibited the gizzard ecto-ATPase in a pH-dependent manner. Greater inhibition was obtained in solutions of lower pH. At pH 6.4, the ATPase activity was inhibited by 50 – 60% at pCMPS concentration greater than 10 M, whereas negligible inhibition was observed at pH 8.17. pHMB was also effective in similar concentration range as pCMPS at pH 6.4, whereas HgCl 2 was less effective (data not shown). Interestingly, significant ConA activation of ATPase was still observed with membranes pretreated with pCMPS (Fig. 3B). Furthermore, the inhibitory effect of pCMPS was markedly reduced if it was added to membranes preincubated with ConA (Fig. 3B), suggesting that ConA has a protective effect against pCMPS inhibition. Suramin. Suramin, a polysulfonated naphthylamine, has been shown to be an inhibitor of several ion-motive ATPases (28 –30), an antagonist of the P2 receptors (31), and an inhibitor of ecto-ATPases (32– 36). Figure 4A shows that suramin inhibited the gizzard ecto-ATPase also in a pH dependent manner. Like the mercurials, inhibition was greater at lower pH.
110
CALDWELL ET AL.
FIG. 2. Effect of ConA on the ecto-ATPase activities of chicken gizzard membranes treated with DSS or EIAA. (A) DSS was added to the membranes (0.9 mg/ml) at the indicated DSS/protein ratios in 20 mM Mops, pH 8. After incubation at 0°C for 60 min, the reaction was terminated by the addition of Tris–Cl, pH 8, to a final concentration of 50 mM. ConA was then added to the untreated or DSS-treated membranes at a ConA/protein ratio of 2:1. After further incubation for 60 min at 0°C, ATPase activity was determined in the coupled enzyme assay with 5 g membrane protein. (B) Experimental conditions were similar to that described in (A) except EIAA at the indicated concentrations was used in the pretreatment.
Inhibition obtained with 50 M suramin at pH 6.4 was ⬃75% but was 40% at pH 8.17. Inhibition by suramin was similar whether added before or after ConA treatment (Fig. 4B). Therefore, in contrast to pCMPS, ConA pretreatment of the membranes did not prevent the ecto-ATPase from inhibition by suramin. Halo-NBDs. NBD-Cl and NBD-F possess a reactive halogen which can be readily replaced by nucleophiles, e.g., -NH 2, -OH, and -SH, with the formation of 7-substituted 4-nitrobenzofurazan derivatives. NBD-Cl inactivates the mitochondrial ATPase (37), Na,K-pump ATPase (38), Ca-pump ATPase (39), and others. Modification occurs at cysteine or tyrosine residues. However, migration of NBD from tyrosine to lysine residue has been reported (40). Inhibition of the gizzard ecto-ATPase by NBD-Cl and NBD-F was more effective at higher pH (Fig. 5A) and was time-dependent (Fig. 5B). Inhibition by 1 mM NBD-Cl was ⬃10% at pH 6, but was ⬃65% at pH 8.0, which is most likely due to the greater nucleophilicity of the amino acid residues involved in the reaction. Inhibition by halo-NBD was not reversible by the addition of dithiothreitol or glutathione (data not shown), excluding cysteine as the target of modification. Preliminary data indicated the involvement of a tyrosine residue.
In contrast to the ion-motive ATPases affected by NBD-Cl, inhibition of the gizzard ecto-ATPase by NBD-Cl is not protected by the substrate, ATP (23), suggesting that the NBD-Cl reactive site is not at the catalytic site. On the other hand, Fig. 5B shows that ConA also protects the ecto-ATPase from inhibition by NBD-Cl, further indicating its action at a regulatory site. Xanthene derivatives. Tetraiodofluorescein has been shown to interact with the nucleotide binding sites on a number of ATP utilizing enzymes (41) and is an inhibitor of the chicken skeletal T-tubule ectoATPase (42). We examined the effect of several related xanthene derivatives on the gizzard ecto-ATPase. The structure of the parent fluorescein, substituents on the xanthene molecule of the different derivatives, and the Kiapp are shown in Table I. The data inidicate that the relatively hydrophobic parent molecule, fluorescein, was the least effective inhibitor. Inhibition increased as the number of polarizable halogen and the volume of the inhibitors increased. Rose bengal, which has a volume of 461.8 Å 3, had a Kiapp that was 2 orders of magnitude lower than fluorescein which has a volume of 273.3 Å 3. Dixon plot analyses indicated that inhibition by the xanthene derivatives was of a mixed nature. Therefore, inhibition occurs by interaction at both
REGULATION OF CHICKEN ECTO-ATPase ACTIVITY
111
FIG. 3. (A) Inhibition of chicken gizzard ecto-ATPase activity by pCMPS at different pH. Chicken gizzard membranes (20 g) were preincubated in 0.45 ml of a solution containing 12.5 mol buffer (Mops for pH 6.4 and Tris for pH 7.18 and 8.17), 2.5 mol MgCl 2 and the indicated concentrations of pCMPS for 5 min at room temperature. Reactions were started by the addition of 0.5 mol of ATP. After 10 min at 28°C, the reaction was terminated by the addition of 0.1 ml 10% TCA. After centrifugation to remove the protein precipitate, an aliquot of the supernatant was used for phosphate determination by the colorimetric method. (B) Effect of ConA on gizzard ecto-ATPase activity added before or after pCMPS. Preincubation of chicken gizzard membranes (20 g) with the indicated concentrations of pCMPS was carried out similar to that described in (A) at pH 6.4. pCMPS was either added alone or added before or after ConA (100 g). Incubation time with pCMPS and ConA was 5 min each.
the catalytic and a regulatory site and that the extent of inhibition is dependent on both the bulk and the polarity of the xanthene derivatives. The effectiveness of the xanthene derivatives decreased if the membranes were preincubated with the activators. The K 0.5 of tetraiodofluorescein (erythrosin in Table I), 1.2 ⫾ 0.1 M, was increased to 3.8 ⫾ 0.3 M, 5.1 ⫾ 0.1 M, and 6.8 ⫾ 0.7 M when the membranes were preincubated with ConA, EIAA, and DSS, respectively. Detergents. Many commonly used detergents for isolating membrane proteins have a deleterious effect on a number of E-ATPases with some exceptions, e.g., the chicken oviduct and liver ecto-ATP/Dases (7, 43). Previously, we showed that the gizzard ecto-ATPase is inhibited by Triton X-100 (20, 21), NP-40, and lysophosphatidylcholine (20) at concentrations below the CMC of these detergents. Digitonin is the only detergent which has a stimulatory effect on the gizzard ecto-ATPase (21), and has been used in its solubilization and purification (44). At sub-CMC concentrations of Triton X-100, the gizzard ecto-ATPase appears to be more sensitive to inhibition than the skeletal muscle T-tubule ecto-ATPase (21), however, pretreatment of the membranes with ConA prevented inhibition of both
enzymes by Triton X-100 (data not shown). Furthermore, Fig. 6 shows that ConA can restore the majority of the lectin-stimulated activity in Triton X-100treated membranes. In the absence of ConA, the ectoATPase activity of the membrane treated with Triton X-100 at a ratio of 20:1 was only 25% of the untreated membrane. Incubation of the Triton X-100-treated membranes with ConA restored the activity to approximately 90% of the lectin-stimulated activity. At higher concentrations of Triton X-100, the detergent interfered with the protein–ConA interaction, and restoration was significantly reduced. DISCUSSION
Previous work from this and other laboratories showed that the ATPase activity of avian skeletal muscle T-tubule membranes (19, 21, 45) and gizzard smooth muscle membranes (20, 21) displayed unusual enzymatic properties and may have a complex regulatory mechanism (42). The enzymes responsible for the activity are members of a subfamily of the E-ATPases, the ecto-ATPases. In a recent study, we showed that the activity of the gizzard ecto-ATPase activity is af-
112
CALDWELL ET AL.
FIG. 4. (A) Inhibition of chicken gizzard ecto-ATPase activity by suramin at different pH. Experimental conditions were similar to that described in the legend of Fig. 3A. (B) Effect of ConA on gizzard ecto-ATPase activity added before or after suramin. Experimental conditions were similar to that described in the legend of Fig. 3B.
fected by several modulators, while that of another E-ATPase, the liver ecto-ATP/Dase, is not (16). The previous results suggest that reagents which promote protein oligomerization, i.e., ConA and DSS, increase the gizzard ecto-ATPase activity, whereas compounds which disrupt oligomerization, i.e., detergents at concentrations below their CMC, decrease its activity. In addition to smooth and skeletal muscle ecto-ATPases, brain and some tumor cell ecto-ATPases are similarly affected by ConA and detergents (21, 46 – 48), suggesting a common regulatory mechanism for the ectoATPases. In this study,we have examined the effect of other compounds that activate or inhibit the gizzard ecto-ATPase individually and in conjunction with ConA. These modulators are not only useful in distinguishing ecto-ATPase and ecto ATP/Dase activities, they have also provided further insight into the regulation of the smooth muscle ecto-ATPase. Among the inhibitory modulators, the mercurials and NBD-Cl were previously shown to inhibit ion-motive ATPases by modifying amino acid residues near the catalytic site. When tested on the gizzard ectoATPase and liver ecto-ATP/Dase, these compounds inhibit only the former and not the latter. Since all EATPases contain five conserved regions (the apyrase consensus regions) that have been shown to be important for catalytic activity (49, 50), the differential effects of pCMPS and NBD-Cl on the two E-ATPases
suggest that these compounds are not directly affecting the catalytic site of the gizzard ecto-ATPase. Furthermore, unlike diethyl pyrocarbonate, which inhibits all E-ATPases and whose inhibitory effect is abolished by prior addition of substrates (23, 26), ATP does not afford protection to the ecto-ATPase against inhibition by pCMPS and NBD-Cl. These results further support the conclusion that their reaction with the enzyme occurs at a regulatory site. The lack of protective effect by ATP was also observed for the other modulators used in this study, i.e., suramin, the xanthene derivatives, and Triton X-100, all amphiphiles that are capable of interfering with membrane protein interactions. The regulatory effects of the selected activating and inhibitory modulators are now incorporated in a model that is based on the thesis that the activity of the gizzard ecto-ATPase is determined by its oligomerization state. The model in Fig. 7 shows that, at physiological temperatures, the “normal” oligomerization state in the native membranes is composed of a mixed population of monomers and dimers. In the presence of the divalent succinyl-ConA, the enzyme exists mostly as dimers and a small increase in activity is obtained. EIAA activates the ecto-ATPase activity to the same extent as succinyl-ConA, suggesting that it also promotes formation of mostly dimers, probably facilitated by the ability of the eosin moiety to form stacked dimers (53). Conversion of the enzyme to a highly oli-
REGULATION OF CHICKEN ECTO-ATPase ACTIVITY
113
FIG. 5. (A) Inhibition of gizzard ecto-ATPase by NBD-Cl and NBD-F at different pH. Chicken gizzard membranes (15 g) were incubated with or without 1 mM NBD-Cl and NBD-F at the indicated pH for 60 min at 0°C. The buffers used were Mes for pH 6.0 – 6.9 and Hepes for pH 6.9 – 8.6. After the reaction was terminated by 50 mM Tris–Cl, pH 7.3, 5–10 g of the membrane protein were used in the coupled enzyme assay. (B) Effect of NBD-Cl on gizzard ecto-ATPase activity without and with preincubation with ConA. Gizzard membranes were incubated on ice for 60 min without or with ConA (2:1). After adjusting the pH to 8.0, NBD-Cl was added to the membranes at 1 and 10 mM and incubation continued on ice. The reaction was quenched at the indicated time with 50 mM Tris–Cl, pH 7.3, and 5–10 g membrane protein were used for activity determination.
gomeric state by DSS and the tetravalent ConA gives rise to higher activity. Oligomerization promoted by EIAA, DSS, and ConA not only alters the reaction rate, but also the apparent K m for ATP, indicating structural changes induced by these compounds affect the interaction of the enzyme with the substrate. The fact that ecto-ATPase activity of the gizzard membranes is low at physiological temperatures (37– 40°C), but can be markedly increased due to oligomerization at the same temperatures, suggests that the gizzard ecto-ATPase activity is probably not constant in vivo, but can vary as determined by physiological events which affects oligomerization of the enzyme. In the model of Fig. 7, we also indicate that detergents and other covalent and noncovalent inhibitors decrease enzyme activity by promoting the state where the enzyme exists mostly as monomers. Other than the halo-NBDs, the noncovalent reagents are characterized by their amphiphilicity. This is most clearly illustrated by the xanthene derivatives. Derivatives of greater volume and polarity are more effective inhibitors than the parent molecule, fluorescein (Table I). These compounds most likely affect the interaction of the hydrophobic regions of the enzyme. In this regard, it is of interest that Wang et al. showed that deletion of either of the two transmembranous regions of the rat
CD39 ecto-ATP/Dase resulted in a decrease of activity and evidence was presented that the oligomer formed by the native CD39 protein is a tetramer (54). Of the other inhibitory modulators, the effect of pCMPS is of particular interest. This sulfhydryl reagent modifies proteins by reacting with cysteine residues. All E-ATPases have ten conserved cysteine residues, which may be involved in the formation of disulfide bonds that are prevalent in most cell surface proteins. Examination of the sequences of five cloned ectoATPases, including the chicken gizzard ecto-ATPase, reveals an additional conserved cysteine residue close to the extracellular membrane surface near the Nterminus. We propose that this is the most likely target of pCMPS and that its modification interferes with oligomerization, resulting in inhibition. The inhibitory effect of pCMPS was reversed to a significant degree by the subsequent addition of ConA and inhibition by pCMPS was negligible if the membranes were pretreated by ConA (Fig. 3B). Thus, pCMPS modification does not appear to interfere with oligomerization induced by ConA which binds to glycans of a glycoprotein. The involvement of the particular cysteine residue in the action of pCMPS will be investigated in future mutagenesis studies.
114
CALDWELL ET AL. TABLE I
Structure, Volume, and K i app of the Inhibitory Xanthene Derivatives
Xanthene 1. 2. 3. 4. 5. 6. 7. 8.
Fluorescein Dichlorofluorescein Dibromofluorescein Diiodofluorescein Eosin Y Phloxine B Erythrosin B Rose Bengal
R1
R2
R3
Volume (Å 3)
K i app (M)
OH OH OH OH OBr OBr OI OI
OH OCl OBr OI OBr OBr OI OI
OH OH OH OH OH OCl OH OCl
273.3 300.6 311.2 340.0 349.9 404.9 406.7 461.8
124 100 35.9 9.30 6.10 1.50 1.20 0.68
Note. Kiapp were obtained from Dixon plot analyses using data obtained at two ATP concentrations (0.25 and 1 mM) and suitable range of inhibitor concentrations. Enzyme activity was determined using the coupled enzyme assay. Molecular volume determination was carried out on a silicon Graphics workstation using SPARTAN 4.1.
The inhibitory effect of the nucleotide mimetics, NBD-Cl and NBD-F, on the gizzard ecto-ATPase differs from that on the ion-motive ATPases, such as the mitochondrial ATPase. The failure of ATP in protecting the gizzard ecto-ATPase from inhibition by NBD-Cl or NBD-F (23) suggests that they react with amino acid residues distant from the active site but are involved in oligomerization of the ectoATPase molecule. This conclusion is supported by the observation that their inhibitory effect is abolished by ConA. While preliminary results suggest their reaction with a tyrosine residue, future analysis using purified and expressed protein should reveal the precise target of their modification. In contrast to pCMPS, NBD-Cl, and the xanthene derivatives, the effect of suramin is independent of ConA. Inhibition by suramin is the same whether added before of after ConA, suggesting that ConApromoted oligomerization of the gizzard ecto-ATPase does not prevent interaction of the enzyme with suramin. Suramin affects many ATP-utilizing enzymes and is a potent antagonist of the purinergic receptors (31). It has been suggested that suramin may compete with ATP for its binding site on the receptor. However, the fact that suramin does not inhibit ecto-ATP/Dases (36, 43), argues against a direct effect on the catalytic site of ecto-ATPase. At
FIG. 6. Restoration of ecto-ATPase activity in Triton X-100-treated gizzard membranes by ConA. Gizzard membranes were incubated with Triton X-100 at the indicated protein/Triton X-100 ratios for 60 min at 0°C. Final Triton X-100 concentrations ranged from 27 to 81 M. ConA was then added at a ConA/protein of 2:1 to the Triton X-100-treated membranes and incubated for another 60 min after which aliquots were taken for activity determination.
REGULATION OF CHICKEN ECTO-ATPase ACTIVITY
FIG. 7.
115
Scheme of proposed regulation of chicken gizzard ecto-ATPase activity by activators and inhibitors.
present, the precise mechanism of action of suramin is unknown (55). It will be necessary to conduct further studies to determine if it inhibits the ectoATPase by interfering with its oligomerization or by an unrelated mechanism. In summary, we have shown that the activity of the gizzard smooth muscle ecto-ATPase, in contrast to the
liver ecto-ATP/Dase can be easily altered. The effect of most of the modulators in this study can be explained by their influence on the oligomerization state of the enzyme molecule. Since many of the ecto-ATPases occur in excitable tissues, the facile regulation of this enzyme is probably of physiological significance which remains to be determined.
116
CALDWELL ET AL.
ACKNOWLEDGMENTS We gratefully acknowledge the contribution of Dr. A. S. Dahms to this study. This work was supported in part by the California Metabolic Research Foundation (A.S.D. and A.F.K.) and the Rees-Stealy Research Foundation (A.S.D.). C. C. Caldwell is a recipient of the Arne N. Wick predoctoral research fellowship.
REFERENCES 1. Redegeld, F. A., Caldwell, C. C., and Sitkovsky, M. V. (1999) Trends Pharmacol. Sci. 20, 453– 459. 2. Zimmermann, H. (1999) Trends Pharmacol. Sci. 20, 231–235. 3. Kirley, T. L., Gerber, L. K., and Smith, T. M. (1999) IUBMB Life 48, 67–72. 4. Kegel, B., Braun, N., Heine, P., Maliszewski, C. R., and Zimmermann, H. (1997) Neuropharmacology 26, 1189 –1200. 5. Mateo, J., Harden, T. K., and Boyer, J. L. (1999) Br. J. Pharmacol. 128, 396 – 402. 6. Christoforidis, S., Papamarcaki, T., Galaris, D., Kellner, R., and Tsolas, O. (1995) Eur. J. Biochem. 234, 66 –74. 7. Strobel, R. S., Nagy, A., Knowles, A. F., Buegel, J., and Rosenberg, M. D. (1996) J. Biol. Chem. 271, 16232–16331. 8. Lewis-Carl, S., and Kirley, T. L. (1997) J. Biol. Chem. 272, 23645–23652. 9. Wang, T. F., and Guidotti, G. (1996) J. Biol. Chem. 271, 9898 –9901. 10. Kaczmarek, E., Koziak, K., Se´vigny, J., Siegel, J. B., Anrather, J., Beaudoin, A., Bach, F. H., and Robson, S. C. (1996) J. Biol. Chem. 271, 33116 –33122. 11. Smith, T. M., and Kirley, T. L. (1998) Biochim. Biophys. Acta 1386, 65–78. 12. Se´vigny, J., Robson, S., Waelkens, E., Csizmadia, E., Smith, R. N., and Lemmens, R. (2000) J. Biol. Chem. 275, 5640 –5647. 13. Lemmens, R., Vanduffel, L., Kittel, A., Beaudoin, A. R., Benrezzak, O., and Se´vigny, J. (2000) Eur. J. Biochem. 267, 4106 – 4114. 14. Knowles, A. F., Isler, R. E., and Reece, J. F. (1983) Biochim. Biophys. Acta 731, 88 –96. 15. Beaudoin, A. R., Se´vigny, J., and Picher, M. (1996) in Biomembranes, (A. G. Lee, Ed.), Vol. 5, pp. 369 – 401, JAI Press. 16. Caldwell, C. C., Davis, M. D., and Knowles, A. F. (1999) Arch. Biochem. Biophys. 362, 46 –58. 17. Knowles, A. F., and Nagy, A. K. (1999) Eur. J. Biochem. 262, 349 –357. 18. Kang, J-J., Cunningham, H. B., Jachec, C., Priest, A., Dahms, A. S., and Sabbadini, R. A. (1991) Arch. Biochem. Biophys. 290, 214 –223. 19. Moulton, M. P., Sabbadini, R. A., Norton, K. C., and Dhams, A. S. (1986) J. Biol. Chem. 261, 12244 –12251. 20. Caldwell, C., Norman, V., Urbina, A., Jarvis, A., Quinonez, C., Stemm, M., and Dahms, A. S. (1997) Biochem. Biophys. Res. Commun. 238, 728 –732. 21. Cunningham, H. B., Yazaki, P. J., Domingo, R., C., Oades, K. V., Bohlen, H., Sabbadini, R. A., and Dahms, A. S. (1993) Arch. Biochem. Biophys. 303, 32– 43. 22. Stout, J. G., and Kirley, T. L. (1996) Biochemistry 35, 8289 – 8298. 23. Hornyak, S. C. (1999) M.S. thesis, San Diego State University. 24. Kirley, T. L. (1988) J. Biol. Chem. 263, 12682–12689. 25. Saborido, A., Moro, G., and Megias, A. (1991) J. Biol. Chem. 266, 23490 –23498. 26. Dzhandzhugazyan, K. N., and Plesner, L. (2000) Biochim. Biophys. Acta 1466, 267–277.
27. Handa, M., and Guidotti, G. (1996) Biochem. Biophys. Res. Commun. 218, 916 –923. 28. Fortes, P. A. G., Ellory, J. C., and Lew, V. L. (1973) Biochim. Biophys. Acta 318, 262–272. 29. Calcaterra, N. B., Vicario, L. R., and Roveri, O. A. (1988) Biochem. Pharmacol. 37, 252–257. 30. Moriyama, Y., and Nelson, N. (1988) FEBS Lett. 234, 383–386. 31. Hoyle, C. H. V., Knight, G. E., and Burnstock, G. (1990) Br. J. Pharmacol. 99, 617– 621. 32. Crack, B. E., Beukers, M. W., McKechnie, K. C. W., IJzerman, A. P., and Leff, P. (1994) Br. J. Pharmacol. 113, 1432–1438. 33. Stout, J. G., and Kirley, T. L. (1995) Biochem. Mol. Biol. Int. 36, 927–934. 34. Chen, B. C., Lee, C-M., and Lin, W-W. (1996) Br. J. Pharmacol. 119, 1628 –1634. 35. Delgado, J., Moro, G., Saborido, A., and Megias, A. (1997) Biochem. J. 327, 899 –907. 36. Heine, P., Braun, N., Heilbronn, A., and Zimmermann, H. (1999) Eur. J. Biochem. 262, 102–107. 37. Ferguson, S. J., Lloyd, W. J., Lyons, M. H., and Radda, G. K. (1975) Eur. J. Biochem. 54, 117–126. 38. Cantley, L. C., Gelles, J., and Josephson, L. (1978) Biochemistry 18, 418 – 425. 39. Bailin, G. (1994) Biochim. Biophys. Acta 1208, 197–203. 40. Ferguson, S. J., Lloyd, W. J., and Radda, G. K. (1975) Eur. J. Biochem. 54, 127–133. 41. Neslund, G. G., Miara, J. G., Kang, J.-J., and Dahms, A. S. (1984) Curr. Top. Cell Regul. 24, 444 – 469. 42. Sabbadini, R. A., and Dahms, A. S. (1989) J. Bioenerg. Biomembr. 21, 163–213. 43. Knowles, A. F., Strobel, R. S., Nagy, A. K., and Wu-Weis, M. In preparation. 44. Stout, J. G., and Kirley, T. L. (1994) J. Biochem. Biophys. Methods 29, 61–75. 45. Megı´as, A., and Saborido, A. (1990) Arch. Biochem. Biophys. 278, 113–119. 46. Nagy, A. K., Knowles, A. F., and Treiman, L. J. (2000) in EctoATPases and Related Ectonucleotidases (L. Vanduffel and R. Lemmens, Eds.), pp. 53– 60, Shaker, Maastricht, The Netherlands. 47. Knowles, A. F., and Kaplan, N. O. (1981) Biochem. Biophys. Res. Commun. 99, 1443–1448. 48. Shi, X-J., and Knowles, A. F. (1994) Arch. Biochem. Biophys. 315, 177–184. 49. Smith, T. M., and Kirley, T. L. (1999) Biochemistry 38, 321– 328. 50. Schulte am Esch, II, J., Se´vigny, J., Kaczmarek, E., Siegel, J. B., Imai, M., Koziak, K., Beaudoin, A. R., and Robson, S. C. (1999) Biochemistry 38, 2248 –2258. 51. Beeler, T. J., Gable, K., and Keffer, J. (1983) Biochim. Biophys. Acta 734, 221–234. 52. Stout, J. G., and Kirley, T. L. (1996) Biochemistry 35, 8289 – 8298. 53. Ajtai, K., Ilich, P., Ringler, A., Sedarous, S., Toft, D., and Burghardt, T. (1992) Biochemistry 31, 12431–12440. 54. Wang, T-F., Ou, Y., and Guidotti, G. (1998) J. Biol. Chem. 273, 24814 –24821. 55. Voogd, T. E., Vansterkenburg, E. L. M., Wilting, J., and Janssen, L. H. (1993) Pharmacol. Rev. 45, 177–203.