Vol. 180, No. 3, 1991 November 14, 1991
ATP
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 1200-1206
diphosphohydrolase is r e s p o n s i b l e f o r e c t o - A T P a s e a n d e c t o - A D P a s e a c t i v i t i e s in b o v i n e a o r t a smooth muscle cells endothelial and Kiyohito Yagi, Masashi Shinbo, Minori Hashizume, Leonard S. Shimba, Sachiko Kurimura and Yoshiharu Miura* Faculty of Pharmaceutical Sciences, Osaka University, 1-6 Yamadaoka, Suita, Osaka 565 Japan
Received September 30, 1991
SUMMARY: An ATP diphosphohydrolase (EC 3.6.1.5) is an enzyme hydrolyzing pyrophosphate bonds in nucleoside di- and triphosphates with broad substrate specificity in the presence of divalent cations. The ATPase and ADPase activities in the enzyme purified to homogeneity from bovine aortic vessel wall were insensitive to oligomycin, ouabain, and various protease treatments, and sensitive to azide and Ap5A. Bovine aorta endothelial and smooth muscle cells were cultured separately to characterize the ectonucleotidase activities. The activities were dependent on the addition of divalent cations and had broad substrate specificity. The ecto-ATPase and -ADPase activities were insensitive to oligomycin, ouabain, and protease treatments, and sensitive to azide and Ap5A. No enzyme degrading only ADP was found in the aortic vessel wall. Moreover, antiserum raised against purified ATP diphosphohydrolase inhibited the ecto-ATPase and -ADPase activities. These results indicated that ecto-ATPase and ecto-ADPase are not separate enzymes but are expressed by one enzyme, ATP diphosphohydrolase. © 199~ AcademioP..... znc.
Extracellular adenine nucleotides have many biological activities including PGI2 secretion (1, 2) and induction of mitogenic response (3, 4) in vascular endothelial cells, relaxation or constriction of smooth muscle and Ca mobilization in astrocytes (5). mediated by P2y-purinoceptor [for review see (6)].
These responses are
The receptor responds to ATP and ADP but
not to AMP or adenosine. The extracellular ATP and ADP are dephosphorylated to adenosine by ectonucleotidases in many cells, tissues, and organs that have the receptor (7, 8).
The enzymes
may be important in the clearance of effector molecules from the receptor, but the enzymes responsible for the ectonucleotidase activities of the cells have not been identified. Recently, we solubilized ATP diphosphohydrolase (EC 3.6.1.5) from the vessel wall of bovine aorta and purified it to homogeneity (9). The enzyme hydrolyzes the pyrophosphate bond of ATP and ADP to form AMP in the presence of divalent cations. The enzyme has been reported in insects (10-12), plants (13-16), and mammalian tissues (17, 18).
We first tried to purify the
enzyme responsible for ecto-ADPase activity, which is presumed to contribute to the antithrombogenicity of the vascular wall.
However, we could not find an ecto-ADPase degrading
To whom correspondence should be addressed. 0006-291X/91 $1.50 Copyright © 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
1200
Vol. 1 80, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
only ADP throughout the purification steps.
The enzyme degrading ADP also dephosphorylated
ATP and other nucleoside di- and tri-phosphates. It is quite probable that the enzyme participates in the reaction from ATP to AMP at the surface of vascular endothelial cells. Since the vessel wall contains smooth muscle cells as well as endothelial cells, we cultured bovine aortic endothelial and smooth muscle cells separately and characterized the ectonucleotidase activities to compare with the enzyme that we purified from bovine aortic vessel wall. MATERIALS AND METHODS Materials Nucleotides, ouabain, azide, and Ap5A were obtained from Wako Chemicals. Oligomycin was obtained from Nacarai tesque. All other materials were of reagent grade. Bovine aortas wore purchased from a local slaughtorhouse. Cells Bovine endothelial cells and smooth muscle cells were gifts from Dr. T. Matsuda (National Cardiovascular Center Research Institute, Suita, Osaka, Japan). Cells were suspended in Dulbecco's modified Eagle's medium (Nissui) containing 15% fetal calf serum (Row Laboratories Inc.) and then plated for experiments in 12-well culture dishes (Sumilon, Sumitumo Bakelite Co., Ltd.) at a density of 1 x 105 cells/cm 2, and incubated at 37 °C under air containing 5% CO2. Measurement of ectonucleotidase activities A dish was washed with the reaction mixture consisting of 50 mM Trls-HCI (pH 7.4), 1 mM CaCI2, 1 mM MgCI2, 95 mM NaCI, 5 mM KCI, 5 mM glucose, 0.05% BSA, and 1 mM Na2HPO4. Two ml of the reaction mixture was added and the plate was shaken reciprocally at a speed of 110 rpm for 1 hour at 37 °C. Then nucleotide was added at a final concentration of 0.5 mM and the plate was shaken at 110 rpm. At regular intervals 100 pl of the solution was taken and mixed with 20 ~ul of solution consisting of 50 mM EDTA and 1 mM NaCI to stop the reaction. The amount of nucleotide in the plate without cells was also measured as a control. The ectonucleotidase activity was calculated from the decrease in nucleotide or increase in orthophosphate concentration for 30 rain. The nucleotide was analyzed by reversed phase HPLC or the amount of orthophosphate released was measured by the method of Fiske and Subbarow (19). Reversed phase HPLC The HPLC was done using a Shimadzu LC-6A system with a Shimedzu Chromatopac data analyzer. The column was C18 reversed phase Shimpack CLC-ODS (i.d. 6.0 mm, 15 cm long, from Shimadzu). The column was equilibrated with the elutlon buffer, NH4H2PO4 (pH 6.0). Ten pl of sample was injected into the column. The buffer ran isocratically at a rate of 1 ml/min for 30 min. Nucleotides were detected at 254 nm using a Shimadzu SPD-6A UV spectrophotometric detector. The retention times of ATP, ADP, and AMP were 8.2, 9.8, and 17.7 min, respectively. RESULTS Extracellular nucleotidase activity Bovine aorta endothelial cells wore cultured as a monolayer and the ecto-ATPase activity was measured as described in Materials and Methods.
Figure 1 shows the time course of adenine
nucleotides hydrolysis in cultured aorta endothelial and smooth muscle cells.
In both types of cells,
the concentration of ATP and ADP decreased rapidly and linearly with time until 60 min of incubation. rapidly.
No transient accumulation of ADP was seen and the concentration of AMP increased
Although cultured cells did not have ecto-pyrophosphatase activity, which liberates
orthophosphate from inorganic pyrophosphate, the accumulation of orthophosphate was observed in the medium.
Therefore, externally added ATP was degraded to ADP and then to AMP by ecto-
ATPase and ecto-ADPase activities, respectively.
The activities were dependent on divalent
cations, Ca2+ or Mg2+. The addition of EDTA completely inhibited the activities.
The survival of
cells used in these experiments were confirmed by trypan blue exclusion test and lactate dehydrogenase activity was not detected in the reaction buffer after the measurement of
1201
Vol. 180, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
600
A
400
600,
600
B
= l J ~
C
400
400
v O~ "4= O (D
"O
0
t'-
'~ c-
200
200
50
600
O cO t= ¢D O ¢O
D
400
L)
0
100
50
60O
E
400
200
100
~ ~ L ~
200 I
0
50
0
. . . . i . . . . i. ,, 50 100
600
,oot/ 200 1
100
50
Time
100
F
!
0~ 0
~
50 100
(rain)
Fi,qure 1. Hydrolysis of extracellular adenine nucleotides by cultured aortic endothelial and smooth muscle cells. At the times indicated, the amounts of ATP ( [ ] ) , ADP ( • ) , and AMP ( • ) were measured as described in Materials and Methods. Values are means of two experiments. Results of endothelial cells are shown in A, B, and C. Results of smooth muscle cells are shown in D, E, and F. Substrates : A and D; ATP, B and E; ADP, C and F; AMP.
nucleotidase activity.
In case of endothelial cells, the amount of AM P gradually decreased and in
turn adenosine accumulated upon longer incubation, but the AMPase activity was very weak as compared with the ATPase and ADPase activities.
On the contrary, cultured smooth muscle cells
had high ecto-AMPase activity which was comparable to the ecto-ATPase and -ADPase activities. That is the reason why the accumulation rate of AMP was lower than the degradation rate of ATP and ADP. Substrate s_Decificitv ATP diphosphohydrolase purified from vessel walls of bovine aorta had a broad substrate specificity to nucleoside di- and triphosphate besides ATP and ADP(9).
Table 1 shows the
substrate specificity of ectonucleotidase activity in cultured endothelial and smooth muscle ceils. Cultured cells also degraded various nucleoside di- and triphosphates at almost the same rate as ATP degradation. Sensitivitv to various ATPase inhibitors and oroteases Effects of various inhibitors on ectonucleotidase activities were examined in cultured endothelial and smooth muscle cells. Table 2 shows the results. The addition of ouabain and oligomycin, inhibitors of Na +, K + -ATPase and H+ -ATPase, respectively, did not inhibit ectoADPase and -ATPase activities. On the contrary, both activities were inhibited by sodium azide and Ap5A which are known as inhibitors of mitochondrial ATPase and adenylate kinase, respectively. When the activities were measured by the release of orthophosphate, similar sensitivities to
1202
V o l . 180, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Table 1.
Ectonucleotidase activities in cultured bovine aortic endothelialand smooth muscle cells Relativeactivity (%) Endothelial cells Smooth muscle cells
Sub~r~e ATP ADP GTP GDP TTP CDP IDP
100 133 79 60 110 61 125
100 123 107 NT NT 59 117
Rate of ATP hydrolysis was taken as 100% in each kind of cell. Values are means of two experiments. NT; not tested.
inhibitors were observed in both types of cells.
ATP diphosphohydrolase was purified to
homogeneity from bovine aorta microsomes solubilized within Triton X-100. The activity had almost the same sensitivities to these inhibitors, as shown in Table 3.
Oligomycin and ouabain did not
affect the ATPase and ADPase activities of purified ATP diphosphohydrolase. On the contrary, the activities were inhibited by the addition of ApSA and sodium azide.
Next we tried to examine the
sensitivity of ATPase and ADPase activities to various proteases in cultured cells and purified ATP diphosphohydrolase.
Purified enzyme was incubated with protease for 60 min at 37°C and then
activity was measured. chymotrypsin, and papain.
The ATPase and ADPase activities were insensitive to trypsin, aChange of the mobility in SDS/PAGE was examined before and after
the protease treatments (data not shown). Trypsin did not change the molecular weight of the enzyme.
It is confirmed that the same concentration of trypsin could degrade the same amount of
albumin as that of the enzyme tested. A slight change of mobility was observed when the enzyme was treated by ~-chymotrypsin and papain.
Cultured
endothelial cells were incubated with
protease for 60 min and then activities were measured.
Detachment of cultured cells was not
observed during the treatment with proteases.
These activities were also insensitive to all
proteases. Trypsin treatment did not affect ecto-ATPase and -ADPase activities in cultured smooth muscle cells.
Table 2. Effects of inhibitors on ecto-ATPase and -ADPase activities in cultured bovine aortic endothelial and smooth muscle cells Inhibition (%) Inhibitors
EndothelialceRs ecto-ATPase - A D P a s e
None Ap5A (0.25 mM) Oligomycin (100 ~Jg/ml) Ouabain (1.0 mM) Sodium azide (10 mM)
0(0) 40 (37) 0 (0) 0(0) 51 (60)
0(0) 40 (40) 0 (0) 0(0) 54 (40)
Smooth muscle cells ecto-ATPase-ADPase 0(0) 56 (45) 0 (0) 0(0) 43 (38)
0(0) 60 (58) 8 (0) 0(0) 50 (50)
Values in parentheses indicate percentage of inhibition in the assay measuring orthophosphate released in the medium. Substrate concentration was 500 IJM. Values are means of two experiments.
1203
Vol. 180, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Table 3. Effects of inhibitors on ATPase and ADPase activity in ATP diphosphohydrolase purified from bovine aortic vessel wall
Inhibition (%) ATP diphosphohydrolase ATPase ADPase
Inhibitors
None Ap5A Oligomycin Ouabain Sodium azide
0 69 0 0 54
(0.25 mM) (100 pg/ml) (1.0 mM) (10 mM)
0 63 0 0 42
Values are means of tWO experiments.
Inhibition by antibody raised against ATP diphos~)hohydrolase Antiserum was raised in rabbit against the purified bovine aorta ATP diphosphohydrolase. Figure 2 shows that the sensitivity of ecto-ATPase activity in cultured endothelial cells to the antiserum. The addition of 1% (v/v) of we-immune serum gave a slight increase in the ecto-ATPase activity.
On the contrary, the ecto-ATPase activity was 45 % inhibited by the addition of 1% (v/v) of
the antiserum.
The antiserum also inhibited 53 % and 42 % inhibited ecto-ADPase and -GTPase
activities, respectively. DISCUSSION ATP diphosphohydrolase is widely distributed in mammalian plasma membranes (17, 18). We purified the enzyme from bovine aortic vessel wall to homogeneity for the first time from an animal source (9).
The enzyme activity detected in various tissues is ouabain-, oligomycin-, and
Ap5A-insensitive, and azide-sensitive.
On the contrary, our homogeneous preparation was
sensitive to Ap5A, which is a potent inhibitor of adenylate kinase.
When the activity of ATP
diphosphohydrolase is measured using ADP as a substrate by detecting liberated orthophosphate,
1000
800 _c r~ ~9 O ¢EL ¢3 c: (2,
600 400 RO0
0 0 ~-
0
•
i 20
i 40
i 60
i 80
100
Time (hr) Figure 2.
Inhibition of ecto-ATPese activity in endothelial cells by anti-serum raised
against bovine aorta ATP diphosphohydrolese.
The increase in orthophosphate
concentration was measured for 90 min in the absence of serum ( • anti-serum ( • ) or pre-immune serum ( • ). values are means _+S.D. of three experiments.
1204
) and presence of the
Vol. 180, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
the orthophosphate might be released by the cooperative action of adenylate kinase and ATPase. But we concluded that bovine aortic enzyme is Ap5A-sensitive because 1) we used a homogeneous preparation of the enzyme, 2) addition of an ATP scavenging system consisting of glucose and hexokinase did not affect the activity, and 3) the ATPase activity of the enzyme was also inhibited to the same extent as ADPase activity by Ap5A. Eoto-ATPase and ADPase activities in cultured endothelial and smooth muscle cells had the same sensitivity to the inhibitors as in case of ATP diphosphohydrolase, that is, those activities were insensitive to oligomycin and ouabain, and sensitive to azide and Ap5A. Lin et al. purified (Ca2+ -Mg2+)-ATPase, which has a similar substrate specificity to ATP diphosphohydrolase, from rat liver plasma membrane (20, 21). The purified enzyme was insensitive to protease treatment (22).
Moreover, ecto-ATPase activity in primarily cultured hepatocytes was
also insensitive to proteases. Similarly, ecto-ATPase and -ADPase activities in cultured endothelial cells as well as ATP diphosphohydrolase purified from bovine aorta were resistant to various protease treatments.
Insensitivity of ectonucleotidases in vascular endothelial cells to protease is
meaningful because those proteins should always be exposed to various proteases in blood. Ecto-adeninenucleotidase activities are reported in vascular endothelial cells (7) and smooth muscle cells (8).
We confirmed that cultured endothelial and smooth muscle cells had
ectonucleotidase activity with broad substrate specificity.
Cultured cells degraded various
nucleoside di- and triphosphate at comparable rates. The broad substrate specificity agrees with that of ATP diphosphohydrolase purified from vessel wall of bovine aorta. Cultured cells and the purified enzyme showed similar behavior to inhibitors, sensitivity to proteases, and substrate specificity.
Moreover, antiserum against purified enzyme inhibited ectonucleotidase activities in
cultured cells. Therefore, we concluded
that ATP diphosphohydrolase is an enzyme responsible
for ecto-ATPase and -ADPase activities in aortic endothelial and smooth muscle cells. The physiological function of ATP diphosphohydrolase is not known.
One possible
function is to scavenge ATP and ADP in the vicinity of the P2y-purinoceptor, because cells and organs confirmed to have P2y-purinoceptor have high ectonucleotidase activities. makes the effects of ATP and ADP transient.
The activity
The ADPase activity in the enzyme purified from rat
liver was about 30% of the ATPase activity (23). On the contrary, the enzyme purified from bovine aorta had high reactivity to ADP. The enzyme might contribute to antlthrombogenicity of the vessel wall by controlling platelet aggregation.
There is another possibility is that the ATP
diphosphohydrolase itself could be the P2¥-purinoceptor.
Similarly to the enzyme, the P2¥-
purinergic effect has broad nucleotide specificity.
The receptor is sensitive to nucleoside tri- and
diphosphates other than ATP and ADP (24-26).
Moreover, it is recently reported that hepatocyte
plasma membrane ecto-ATPase, which was purified as ( Ca2+ -Mg2+)-ATPase, is a substrate for tyrosine kinase activity of the insulin receptor (27, 28).
Therefore, ATP diphosphohydrolase
purified from aortic vessel wall might not merely be a ATP- and ADP-scavenger but may be important in signal transduction of the P2-purinergic effect.
ACKNOWLEDGMENTS We would like to thank Dr. Takehisa Matsuda and Prof. Tadanori Mayumi for their provision of cultured bovine endothelial and smooth muscle cells. We also thank Dr. Seno for preparing antiserum against ATP diphosphohydrolase.
1205
Vol. 180, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
REFERENCES
1 Pearson, J. D., Slakey, L. L. and Gordon, J. L. (1983) Biochem. J. 214, 273-276. 2 Boeynaems, J. M. and Galand, N. (1983) Biochem. Biophys. Res. Commun. 112, 290-296. 3 Van Coevorden, A., Roger, P. P. and Boeynaems, J. M. (1989) Thromb. Haemostasis 62, 190. 4 Boutherin-Falson, O., Reuse, S., Dumont, J. E. and Boeynaems, J. M. (1990) Biochem. Biophys. Res. Commun. 172, 306-312. 5 Pearce, B., Murphy, S., Jeremy, J., Morrow, C. and Dandona, P. (1989) J. Neurochern. 52, 971-977. 6 Boeynaems, J. M. and Pearson, J. D. (1990) Trends Pharmacol. Sd. 11, 34-37. 7 Gordon, E. L, Pearson, J. D. and Slakey, L. L. (1986) J. Biol. Chem. 261, 15496-15507. 8 Pearson, J. D., Coade, S. B. and Cusack, N. J. (1985) Biochern. J. 230, 503-507. 9 Yagi, K., Arai, Y., Kato, N., Hirota, K. and Miura, Y. (1989) Eur. J. Biochem. 180, 509-513. 10 Mant, M. J. and Parker, K. R. (1981) Br.J. Haematol. 48, 601-608. 11 Ribeiro, J. M. C., Sarkis, J. J. F., Rossingnol, P. A. and Spielman, A. (1984) Comp. Biochem. Physiol. 79B, 81-86. 12 Sarkis, J. J. F., Guimaraes, J. A. and Ril0eiro, J. M. C. (1986) Biochem. J. 233, 885-891. 13 Mancilla, M., Kettlun, A. M., Valenzuera, M. A. and Traverso-Cori, A. (1984) Phytochemistry 23, 1397-1400. 14 Vara, F. and Serrano, R. (1981) Biochem. J. 197, 637-643. 15 Traverso-Cori, A., Chaimovich, H. and Cori, O. (1965) Arch. Biochem. Biophys. 109, 173-184. 16 Tognoli, L. and Matte, E. (1981) Biochim. Biophys. Acta 642, 1-14. 17 LeBel, D., Poirier, G. G., Phaneuf, S, St.-Jean, P., Laliberte, J. F. and Beaudoin, A. R. (1980) J. Biol. Chem. 255, 1227-1233. 18 Knowles, A. F., Isler, R. E. and Reece, J, F. (1983) Biochim. Biophys. Acta 731, 88-96. 19 Fiske, C. H. and Subbarow, Y. (1925) J. Biol. Chem. 66, 375-400. 20 Lin, S. -H. and Fain (1984) J. Biol. Chem. 259, 3016-3020. 21 Lin, S. -H. (1985) J. Biol. Chem. 260, 7850-7856. 22 Lin, S. -H. and Russell, W. E. (1988) J. Biol. Chem. 263, 12253-8 23 Lin, S.-H. (1985) J. Biol. Chem. 260, 10976-10980. 24 Okajima, F., Tokumitsu, Y., Kondo, Y. and Ui, M. (1987) J. Biol. Chem. 262, 13483-13498. 25 Forsberg, E. J., Feuerstein, G., Shohami, E. and Pollard, H. B. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 5630-5634. 26 Buxton, D. B., Robertson, S. M. and Olson, M. S. (1986) Biochem. J. 237, 773-780. 27 Lin, S. -H. and Guidotti, G. (1989) J. Biol. Chem. 264, 14408-14414. 28 Margolis, R. N., Schell, M. J., Taylor, S. I. and Hubbard, A. L. (1990) Biochem. Biophys. Res. Commun. 166, 562-566.
1206
ATP
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 1200-1206
diphosphohydrolase is r e s p o n s i b l e f o r e c t o - A T P a s e a n d e c t o - A D P a s e a c t i v i t i e s in b o v i n e a o r t a smooth muscle cells endothelial and Kiyohito Yagi, Masashi Shinbo, Minori Hashizume, Leonard S. Shimba, Sachiko Kurimura and Yoshiharu Miura* Faculty of Pharmaceutical Sciences, Osaka University, 1-6 Yamadaoka, Suita, Osaka 565 Japan
Received September 30, 1991
SUMMARY: An ATP diphosphohydrolase (EC 3.6.1.5) is an enzyme hydrolyzing pyrophosphate bonds in nucleoside di- and triphosphates with broad substrate specificity in the presence of divalent cations. The ATPase and ADPase activities in the enzyme purified to homogeneity from bovine aortic vessel wall were insensitive to oligomycin, ouabain, and various protease treatments, and sensitive to azide and Ap5A. Bovine aorta endothelial and smooth muscle cells were cultured separately to characterize the ectonucleotidase activities. The activities were dependent on the addition of divalent cations and had broad substrate specificity. The ecto-ATPase and -ADPase activities were insensitive to oligomycin, ouabain, and protease treatments, and sensitive to azide and Ap5A. No enzyme degrading only ADP was found in the aortic vessel wall. Moreover, antiserum raised against purified ATP diphosphohydrolase inhibited the ecto-ATPase and -ADPase activities. These results indicated that ecto-ATPase and ecto-ADPase are not separate enzymes but are expressed by one enzyme, ATP diphosphohydrolase. © 199~ AcademioP..... znc.
Extracellular adenine nucleotides have many biological activities including PGI2 secretion (1, 2) and induction of mitogenic response (3, 4) in vascular endothelial cells, relaxation or constriction of smooth muscle and Ca mobilization in astrocytes (5). mediated by P2y-purinoceptor [for review see (6)].
These responses are
The receptor responds to ATP and ADP but
not to AMP or adenosine. The extracellular ATP and ADP are dephosphorylated to adenosine by ectonucleotidases in many cells, tissues, and organs that have the receptor (7, 8).
The enzymes
may be important in the clearance of effector molecules from the receptor, but the enzymes responsible for the ectonucleotidase activities of the cells have not been identified. Recently, we solubilized ATP diphosphohydrolase (EC 3.6.1.5) from the vessel wall of bovine aorta and purified it to homogeneity (9). The enzyme hydrolyzes the pyrophosphate bond of ATP and ADP to form AMP in the presence of divalent cations. The enzyme has been reported in insects (10-12), plants (13-16), and mammalian tissues (17, 18).
We first tried to purify the
enzyme responsible for ecto-ADPase activity, which is presumed to contribute to the antithrombogenicity of the vascular wall.
However, we could not find an ecto-ADPase degrading
To whom correspondence should be addressed. 0006-291X/91 $1.50 Copyright © 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
1200
Vol. 1 80, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
only ADP throughout the purification steps.
The enzyme degrading ADP also dephosphorylated
ATP and other nucleoside di- and tri-phosphates. It is quite probable that the enzyme participates in the reaction from ATP to AMP at the surface of vascular endothelial cells. Since the vessel wall contains smooth muscle cells as well as endothelial cells, we cultured bovine aortic endothelial and smooth muscle cells separately and characterized the ectonucleotidase activities to compare with the enzyme that we purified from bovine aortic vessel wall. MATERIALS AND METHODS Materials Nucleotides, ouabain, azide, and Ap5A were obtained from Wako Chemicals. Oligomycin was obtained from Nacarai tesque. All other materials were of reagent grade. Bovine aortas wore purchased from a local slaughtorhouse. Cells Bovine endothelial cells and smooth muscle cells were gifts from Dr. T. Matsuda (National Cardiovascular Center Research Institute, Suita, Osaka, Japan). Cells were suspended in Dulbecco's modified Eagle's medium (Nissui) containing 15% fetal calf serum (Row Laboratories Inc.) and then plated for experiments in 12-well culture dishes (Sumilon, Sumitumo Bakelite Co., Ltd.) at a density of 1 x 105 cells/cm 2, and incubated at 37 °C under air containing 5% CO2. Measurement of ectonucleotidase activities A dish was washed with the reaction mixture consisting of 50 mM Trls-HCI (pH 7.4), 1 mM CaCI2, 1 mM MgCI2, 95 mM NaCI, 5 mM KCI, 5 mM glucose, 0.05% BSA, and 1 mM Na2HPO4. Two ml of the reaction mixture was added and the plate was shaken reciprocally at a speed of 110 rpm for 1 hour at 37 °C. Then nucleotide was added at a final concentration of 0.5 mM and the plate was shaken at 110 rpm. At regular intervals 100 pl of the solution was taken and mixed with 20 ~ul of solution consisting of 50 mM EDTA and 1 mM NaCI to stop the reaction. The amount of nucleotide in the plate without cells was also measured as a control. The ectonucleotidase activity was calculated from the decrease in nucleotide or increase in orthophosphate concentration for 30 rain. The nucleotide was analyzed by reversed phase HPLC or the amount of orthophosphate released was measured by the method of Fiske and Subbarow (19). Reversed phase HPLC The HPLC was done using a Shimadzu LC-6A system with a Shimedzu Chromatopac data analyzer. The column was C18 reversed phase Shimpack CLC-ODS (i.d. 6.0 mm, 15 cm long, from Shimadzu). The column was equilibrated with the elutlon buffer, NH4H2PO4 (pH 6.0). Ten pl of sample was injected into the column. The buffer ran isocratically at a rate of 1 ml/min for 30 min. Nucleotides were detected at 254 nm using a Shimadzu SPD-6A UV spectrophotometric detector. The retention times of ATP, ADP, and AMP were 8.2, 9.8, and 17.7 min, respectively. RESULTS Extracellular nucleotidase activity Bovine aorta endothelial cells wore cultured as a monolayer and the ecto-ATPase activity was measured as described in Materials and Methods.
Figure 1 shows the time course of adenine
nucleotides hydrolysis in cultured aorta endothelial and smooth muscle cells.
In both types of cells,
the concentration of ATP and ADP decreased rapidly and linearly with time until 60 min of incubation. rapidly.
No transient accumulation of ADP was seen and the concentration of AMP increased
Although cultured cells did not have ecto-pyrophosphatase activity, which liberates
orthophosphate from inorganic pyrophosphate, the accumulation of orthophosphate was observed in the medium.
Therefore, externally added ATP was degraded to ADP and then to AMP by ecto-
ATPase and ecto-ADPase activities, respectively.
The activities were dependent on divalent
cations, Ca2+ or Mg2+. The addition of EDTA completely inhibited the activities.
The survival of
cells used in these experiments were confirmed by trypan blue exclusion test and lactate dehydrogenase activity was not detected in the reaction buffer after the measurement of
1201
Vol. 180, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
600
A
400
600,
600
B
= l J ~
C
400
400
v O~ "4= O (D
"O
0
t'-
'~ c-
200
200
50
600
O cO t= ¢D O ¢O
D
400
L)
0
100
50
60O
E
400
200
100
~ ~ L ~
200 I
0
50
0
. . . . i . . . . i. ,, 50 100
600
,oot/ 200 1
100
50
Time
100
F
!
0~ 0
~
50 100
(rain)
Fi,qure 1. Hydrolysis of extracellular adenine nucleotides by cultured aortic endothelial and smooth muscle cells. At the times indicated, the amounts of ATP ( [ ] ) , ADP ( • ) , and AMP ( • ) were measured as described in Materials and Methods. Values are means of two experiments. Results of endothelial cells are shown in A, B, and C. Results of smooth muscle cells are shown in D, E, and F. Substrates : A and D; ATP, B and E; ADP, C and F; AMP.
nucleotidase activity.
In case of endothelial cells, the amount of AM P gradually decreased and in
turn adenosine accumulated upon longer incubation, but the AMPase activity was very weak as compared with the ATPase and ADPase activities.
On the contrary, cultured smooth muscle cells
had high ecto-AMPase activity which was comparable to the ecto-ATPase and -ADPase activities. That is the reason why the accumulation rate of AMP was lower than the degradation rate of ATP and ADP. Substrate s_Decificitv ATP diphosphohydrolase purified from vessel walls of bovine aorta had a broad substrate specificity to nucleoside di- and triphosphate besides ATP and ADP(9).
Table 1 shows the
substrate specificity of ectonucleotidase activity in cultured endothelial and smooth muscle ceils. Cultured cells also degraded various nucleoside di- and triphosphates at almost the same rate as ATP degradation. Sensitivitv to various ATPase inhibitors and oroteases Effects of various inhibitors on ectonucleotidase activities were examined in cultured endothelial and smooth muscle cells. Table 2 shows the results. The addition of ouabain and oligomycin, inhibitors of Na +, K + -ATPase and H+ -ATPase, respectively, did not inhibit ectoADPase and -ATPase activities. On the contrary, both activities were inhibited by sodium azide and Ap5A which are known as inhibitors of mitochondrial ATPase and adenylate kinase, respectively. When the activities were measured by the release of orthophosphate, similar sensitivities to
1202
V o l . 180, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Table 1.
Ectonucleotidase activities in cultured bovine aortic endothelialand smooth muscle cells Relativeactivity (%) Endothelial cells Smooth muscle cells
Sub~r~e ATP ADP GTP GDP TTP CDP IDP
100 133 79 60 110 61 125
100 123 107 NT NT 59 117
Rate of ATP hydrolysis was taken as 100% in each kind of cell. Values are means of two experiments. NT; not tested.
inhibitors were observed in both types of cells.
ATP diphosphohydrolase was purified to
homogeneity from bovine aorta microsomes solubilized within Triton X-100. The activity had almost the same sensitivities to these inhibitors, as shown in Table 3.
Oligomycin and ouabain did not
affect the ATPase and ADPase activities of purified ATP diphosphohydrolase. On the contrary, the activities were inhibited by the addition of ApSA and sodium azide.
Next we tried to examine the
sensitivity of ATPase and ADPase activities to various proteases in cultured cells and purified ATP diphosphohydrolase.
Purified enzyme was incubated with protease for 60 min at 37°C and then
activity was measured. chymotrypsin, and papain.
The ATPase and ADPase activities were insensitive to trypsin, aChange of the mobility in SDS/PAGE was examined before and after
the protease treatments (data not shown). Trypsin did not change the molecular weight of the enzyme.
It is confirmed that the same concentration of trypsin could degrade the same amount of
albumin as that of the enzyme tested. A slight change of mobility was observed when the enzyme was treated by ~-chymotrypsin and papain.
Cultured
endothelial cells were incubated with
protease for 60 min and then activities were measured.
Detachment of cultured cells was not
observed during the treatment with proteases.
These activities were also insensitive to all
proteases. Trypsin treatment did not affect ecto-ATPase and -ADPase activities in cultured smooth muscle cells.
Table 2. Effects of inhibitors on ecto-ATPase and -ADPase activities in cultured bovine aortic endothelial and smooth muscle cells Inhibition (%) Inhibitors
EndothelialceRs ecto-ATPase - A D P a s e
None Ap5A (0.25 mM) Oligomycin (100 ~Jg/ml) Ouabain (1.0 mM) Sodium azide (10 mM)
0(0) 40 (37) 0 (0) 0(0) 51 (60)
0(0) 40 (40) 0 (0) 0(0) 54 (40)
Smooth muscle cells ecto-ATPase-ADPase 0(0) 56 (45) 0 (0) 0(0) 43 (38)
0(0) 60 (58) 8 (0) 0(0) 50 (50)
Values in parentheses indicate percentage of inhibition in the assay measuring orthophosphate released in the medium. Substrate concentration was 500 IJM. Values are means of two experiments.
1203
Vol. 180, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Table 3. Effects of inhibitors on ATPase and ADPase activity in ATP diphosphohydrolase purified from bovine aortic vessel wall
Inhibition (%) ATP diphosphohydrolase ATPase ADPase
Inhibitors
None Ap5A Oligomycin Ouabain Sodium azide
0 69 0 0 54
(0.25 mM) (100 pg/ml) (1.0 mM) (10 mM)
0 63 0 0 42
Values are means of tWO experiments.
Inhibition by antibody raised against ATP diphos~)hohydrolase Antiserum was raised in rabbit against the purified bovine aorta ATP diphosphohydrolase. Figure 2 shows that the sensitivity of ecto-ATPase activity in cultured endothelial cells to the antiserum. The addition of 1% (v/v) of we-immune serum gave a slight increase in the ecto-ATPase activity.
On the contrary, the ecto-ATPase activity was 45 % inhibited by the addition of 1% (v/v) of
the antiserum.
The antiserum also inhibited 53 % and 42 % inhibited ecto-ADPase and -GTPase
activities, respectively. DISCUSSION ATP diphosphohydrolase is widely distributed in mammalian plasma membranes (17, 18). We purified the enzyme from bovine aortic vessel wall to homogeneity for the first time from an animal source (9).
The enzyme activity detected in various tissues is ouabain-, oligomycin-, and
Ap5A-insensitive, and azide-sensitive.
On the contrary, our homogeneous preparation was
sensitive to Ap5A, which is a potent inhibitor of adenylate kinase.
When the activity of ATP
diphosphohydrolase is measured using ADP as a substrate by detecting liberated orthophosphate,
1000
800 _c r~ ~9 O ¢EL ¢3 c: (2,
600 400 RO0
0 0 ~-
0
•
i 20
i 40
i 60
i 80
100
Time (hr) Figure 2.
Inhibition of ecto-ATPese activity in endothelial cells by anti-serum raised
against bovine aorta ATP diphosphohydrolese.
The increase in orthophosphate
concentration was measured for 90 min in the absence of serum ( • anti-serum ( • ) or pre-immune serum ( • ). values are means _+S.D. of three experiments.
1204
) and presence of the
Vol. 180, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
the orthophosphate might be released by the cooperative action of adenylate kinase and ATPase. But we concluded that bovine aortic enzyme is Ap5A-sensitive because 1) we used a homogeneous preparation of the enzyme, 2) addition of an ATP scavenging system consisting of glucose and hexokinase did not affect the activity, and 3) the ATPase activity of the enzyme was also inhibited to the same extent as ADPase activity by Ap5A. Eoto-ATPase and ADPase activities in cultured endothelial and smooth muscle cells had the same sensitivity to the inhibitors as in case of ATP diphosphohydrolase, that is, those activities were insensitive to oligomycin and ouabain, and sensitive to azide and Ap5A. Lin et al. purified (Ca2+ -Mg2+)-ATPase, which has a similar substrate specificity to ATP diphosphohydrolase, from rat liver plasma membrane (20, 21). The purified enzyme was insensitive to protease treatment (22).
Moreover, ecto-ATPase activity in primarily cultured hepatocytes was
also insensitive to proteases. Similarly, ecto-ATPase and -ADPase activities in cultured endothelial cells as well as ATP diphosphohydrolase purified from bovine aorta were resistant to various protease treatments.
Insensitivity of ectonucleotidases in vascular endothelial cells to protease is
meaningful because those proteins should always be exposed to various proteases in blood. Ecto-adeninenucleotidase activities are reported in vascular endothelial cells (7) and smooth muscle cells (8).
We confirmed that cultured endothelial and smooth muscle cells had
ectonucleotidase activity with broad substrate specificity.
Cultured cells degraded various
nucleoside di- and triphosphate at comparable rates. The broad substrate specificity agrees with that of ATP diphosphohydrolase purified from vessel wall of bovine aorta. Cultured cells and the purified enzyme showed similar behavior to inhibitors, sensitivity to proteases, and substrate specificity.
Moreover, antiserum against purified enzyme inhibited ectonucleotidase activities in
cultured cells. Therefore, we concluded
that ATP diphosphohydrolase is an enzyme responsible
for ecto-ATPase and -ADPase activities in aortic endothelial and smooth muscle cells. The physiological function of ATP diphosphohydrolase is not known.
One possible
function is to scavenge ATP and ADP in the vicinity of the P2y-purinoceptor, because cells and organs confirmed to have P2y-purinoceptor have high ectonucleotidase activities. makes the effects of ATP and ADP transient.
The activity
The ADPase activity in the enzyme purified from rat
liver was about 30% of the ATPase activity (23). On the contrary, the enzyme purified from bovine aorta had high reactivity to ADP. The enzyme might contribute to antlthrombogenicity of the vessel wall by controlling platelet aggregation.
There is another possibility is that the ATP
diphosphohydrolase itself could be the P2¥-purinoceptor.
Similarly to the enzyme, the P2¥-
purinergic effect has broad nucleotide specificity.
The receptor is sensitive to nucleoside tri- and
diphosphates other than ATP and ADP (24-26).
Moreover, it is recently reported that hepatocyte
plasma membrane ecto-ATPase, which was purified as ( Ca2+ -Mg2+)-ATPase, is a substrate for tyrosine kinase activity of the insulin receptor (27, 28).
Therefore, ATP diphosphohydrolase
purified from aortic vessel wall might not merely be a ATP- and ADP-scavenger but may be important in signal transduction of the P2-purinergic effect.
ACKNOWLEDGMENTS We would like to thank Dr. Takehisa Matsuda and Prof. Tadanori Mayumi for their provision of cultured bovine endothelial and smooth muscle cells. We also thank Dr. Seno for preparing antiserum against ATP diphosphohydrolase.
1205
Vol. 180, No. 3, 1991
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
REFERENCES
1 Pearson, J. D., Slakey, L. L. and Gordon, J. L. (1983) Biochem. J. 214, 273-276. 2 Boeynaems, J. M. and Galand, N. (1983) Biochem. Biophys. Res. Commun. 112, 290-296. 3 Van Coevorden, A., Roger, P. P. and Boeynaems, J. M. (1989) Thromb. Haemostasis 62, 190. 4 Boutherin-Falson, O., Reuse, S., Dumont, J. E. and Boeynaems, J. M. (1990) Biochem. Biophys. Res. Commun. 172, 306-312. 5 Pearce, B., Murphy, S., Jeremy, J., Morrow, C. and Dandona, P. (1989) J. Neurochern. 52, 971-977. 6 Boeynaems, J. M. and Pearson, J. D. (1990) Trends Pharmacol. Sd. 11, 34-37. 7 Gordon, E. L, Pearson, J. D. and Slakey, L. L. (1986) J. Biol. Chem. 261, 15496-15507. 8 Pearson, J. D., Coade, S. B. and Cusack, N. J. (1985) Biochern. J. 230, 503-507. 9 Yagi, K., Arai, Y., Kato, N., Hirota, K. and Miura, Y. (1989) Eur. J. Biochem. 180, 509-513. 10 Mant, M. J. and Parker, K. R. (1981) Br.J. Haematol. 48, 601-608. 11 Ribeiro, J. M. C., Sarkis, J. J. F., Rossingnol, P. A. and Spielman, A. (1984) Comp. Biochem. Physiol. 79B, 81-86. 12 Sarkis, J. J. F., Guimaraes, J. A. and Ril0eiro, J. M. C. (1986) Biochem. J. 233, 885-891. 13 Mancilla, M., Kettlun, A. M., Valenzuera, M. A. and Traverso-Cori, A. (1984) Phytochemistry 23, 1397-1400. 14 Vara, F. and Serrano, R. (1981) Biochem. J. 197, 637-643. 15 Traverso-Cori, A., Chaimovich, H. and Cori, O. (1965) Arch. Biochem. Biophys. 109, 173-184. 16 Tognoli, L. and Matte, E. (1981) Biochim. Biophys. Acta 642, 1-14. 17 LeBel, D., Poirier, G. G., Phaneuf, S, St.-Jean, P., Laliberte, J. F. and Beaudoin, A. R. (1980) J. Biol. Chem. 255, 1227-1233. 18 Knowles, A. F., Isler, R. E. and Reece, J, F. (1983) Biochim. Biophys. Acta 731, 88-96. 19 Fiske, C. H. and Subbarow, Y. (1925) J. Biol. Chem. 66, 375-400. 20 Lin, S. -H. and Fain (1984) J. Biol. Chem. 259, 3016-3020. 21 Lin, S. -H. (1985) J. Biol. Chem. 260, 7850-7856. 22 Lin, S. -H. and Russell, W. E. (1988) J. Biol. Chem. 263, 12253-8 23 Lin, S.-H. (1985) J. Biol. Chem. 260, 10976-10980. 24 Okajima, F., Tokumitsu, Y., Kondo, Y. and Ui, M. (1987) J. Biol. Chem. 262, 13483-13498. 25 Forsberg, E. J., Feuerstein, G., Shohami, E. and Pollard, H. B. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 5630-5634. 26 Buxton, D. B., Robertson, S. M. and Olson, M. S. (1986) Biochem. J. 237, 773-780. 27 Lin, S. -H. and Guidotti, G. (1989) J. Biol. Chem. 264, 14408-14414. 28 Margolis, R. N., Schell, M. J., Taylor, S. I. and Hubbard, A. L. (1990) Biochem. Biophys. Res. Commun. 166, 562-566.
1206
