Full text of DOI:10.1080/15216540050212105

–
IUBMB Life, 50: 99 103, 2000
c
°
Copyright 2000 IUBMB
1521-6543/00 $12.00 + .00
Original Research Article
Dual Effect of Nucleotides on P2Y Receptors
1
2
1
¨
Katrin Sak, Eric A. Barnard, and Jaak Jarv
1Institute of Chemical Physics, Tartu University, 2 Jakobi Street, 51014 Tartu, Estonia
2Department of Pharmacology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QJ, U.K.
minimise uncertainties related to the decomposition of nucleo-
( )
side triphosphates during the assay procedure 6 , only nucleo-
Summary
side diphosphates and their structural analogues were used in
this analysis, given that the corresponding monophosphates and
b
The interaction of ADP, 2MeSADP, and ADP S with the ade-
nine nucleotide receptor P2Y1 in the hP2Y1 -1321N1 cell line and
of UDP with a receptor or receptors recognizing pyrimidine nu-
cleotides in NG108-15 cells was studied over a wide range of ligand
concentrations. Bell-shaped dose–response curves for stimulation
of phosphoinositide hydrolysis were obtained in these cells. This
dual behavior of the agonists studied was characterizedby two dis-
sociation constants, Kagon and Kantag, which quantify the agonistic
and antagonistic activity of these ligands and can be compared
with the conventional EC50 and IC50 values, respectively. The data
revealed a common pattern of agonistic and antagonistic behav-
ior of nucleoside diphosphates and their derivatives at these two
types of P2Y receptors, pointing to some similar properties of their
nucleotide binding sites.
(
)
nucleosides are known to be 1, 2 at best very weak ligands for
any of these receptors.
EXPERIMENTAL PROCEDURES
Chemicals. UDP, ADP, AMP, and adenosine were obtained
from Boehringer Mannheim; UMP, uridine, 2MeSADP, and
ADPbS were from Sigma. The purity of all these compounds
(
)
was analysed before their use by HPLC Gilson on an anion-
(
)
exchange column Mono Q Pharmacia Biotech . Absorbance
was monitored at 258 nm for adenine nucleotides and at 270 nm
for uracil nucleotides. Linear gradients from 0 to 1 M NaCl in
–
IUBMB
Life, 50: 99 103, 2000
Keywords Dual effect; inositol phospholipid hydrolysis; nucleoside
(
)
40 mM phosphate buffer pH 7.0 and from 0.1 to 1.2 M ammo-
nium carbonate in water at the  ow rate 1 ml/min were used. If
necessary, the nucleotides were additionally puriŽ ed by rechro-
matographing with the latter gradient. The purity of nucleotides
diphosphate; P2Y receptor.
INTRODUCTION
in the assay was never <99.5%, and no corresponding triphos-
–
3
The G-protein coupled P2Y receptors belong to the class of
7-transmembrane domain 7-TM proteins, so far, 9 subtypes
of these nucleotide receptors have been found in vertebrates by
DNA cloning 1, 2 . These receptors reveal substantial structural
homology with other 7-TM receptors 3 , which, in turn, sug-
gests the possibility of similar functional aspects. In this report
we have examined the ability of the P2Y receptors to produce
the bell-shaped dose response curves previously observed with
some other G-protein coupled receptors, a Ž nding most thor-
oughly analysed for muscarinic receptor regulated cAMP syn-
[
]
phates were detected in the samples used. myo- 2- H Inositol
(
)
was purchased from Amersham.
Cell Cultures. NG108-15 cells were obtained from the Eu-
(
)
(
)
ropean Collection of Cell Cultures no. 88112302 . hP2Y1-
( )
1321N1 cells were the stable cell line expressing the human
( )
P2Y1 receptor, as described elsewhere 7 . Both cell types were
(
cultured in Dulbecco’s modiŽ ed Eagle’s medium DMEM;
–
)
(
)
Gibco supplemented with 10% v/v fetal calf serum and ty-
–
(
)
losine 8 l g/ml . For NG108-15 cells the medium was further
supplemented with 0.1 mM hypoxanthine, 1 l M aminopterin,
and 16 lM thymidine. The medium for hP2Y1-1321N1 cells
–
+
2
( )
thesis 4 and for the mobilisation of intracellular Ca by mus-
carinic agonists 5 . The dual effect of nucleotides was observed
for both a purine-speciŽ c and a pyrimidine-speciŽ c receptor. To
( )
(
)
was supplemented with 400 l g/ml G-418 sulfate Gibco BRL .
Cells were grown at 37 ±C in a humidiŽ ed atmosphere of 95%
air and 5% CO2 in 75-cm2 culture dishes.
Assay of Inositol Phosphates. Cellswere seeded on 24-well
culture plates at a density of 5 £ 104 cells/well for NG108-
«
Received 11 February 2000; accepted 3 July 2000.
15 and 1 £ 105 cells/well for hP2Y -1321N1 and grown to
¨
Address correspondence to Jaak Jarv. E-mail: jj@chem.ut.ee
«
1
99

100
SAK ET AL.
3
The parameters Kagon and Kantag have the meaning of the ap-
propriate dissociation constants and can be compared with the
concentrations at which one-half of binding is complete or in-
[
]
subcon uence for 3 days. H -Inositol was added to the cells in
200 l l of inositol-free DMEM at 2 l Ci/ml18 h before the assay.
No changes in medium were made subsequent to H inositol
addition. The ligands were added in 50 l l of 50 mM LiCl and
3
[
]
(
)
hibited conventional EC50 and IC50 values , respectively. Ex-
perimental data were expressed in semilogarithmic coordinates
(
)
100 mM phosphate buffer pH 7.2 , and the assay mixture was
incubated 10 min at 37 ±C. A check of the pH of the assay
medium after the assay indicated no systematic change, even at
high ligand concentrations.
(
)
Fig. 1 by means of a nonlinear regression analysis program
PrismTM ver. 2.00 , and the calculatedparameterswere reported
(
)
(
)
with standard errors §SE . The number of experiments used
in calculations corresponded to the number of points shown in
Fig. 1.
The assay was terminated by aspirating the medium and
adding 0.5 ml of ice-cold 5% trichloroacetic acid. The result-
ing supernatant was extracted three times with 0.5 ml of di-
ethyl ether and inositol phosphates were isolated on the Dowex
RESULTS
–
(
AG1- £8 columns Bio-Rad; 100 200 mesh, formate form, bed
)
(
volume 0.8 ml . These columns were washed with water 2 £
Effect of ADP and Its Derivatives on the Human P2Y1
Receptor Stably Expressed in Astrocytoma 1321N1 Cells
ADP, 2MeSADP, and ADPbS induced the synthesis of in-
ositol phosphates in hP2Y1-1321N1 cells, whereas uracil nu-
cleotides were inactive therein. With ADP the time-dependence
of formation of inositol phosphates was analysed. During the
Ž rst 20 min the increase in product concentration was linear
)
(
)
4 ml and 50 mM ammonium formate 8 ml ; inositol phosphates
were eluted with 1 M ammonium formate in 0.1 M formic acid
3
(
)
[
]
2 £ 4 ml . The amount of H inositol phosphates formed was
(
determined by adding scintillation cocktail OptiPhase HiSafe
)
III; Wallac to each eluateand counting the radioactivity present.
Under the experimental conditions used the baseline radioactiv-
ity of the samples remained between 600 and 700 cpm/sample.
The maximal radioactivity measured during the assay time was
2700 2900 cpm/sample for NG108-15 cells and 3200 3400
cpm/sample for hP2Y1-1321N1 cells. When investigating the
time-dependence of accumulation of the inositol phosphates,
incubation times between 3 and 20 min were used.
(
)
Fig. 2A , in agreement with the assumptions that no signiŽ cant
degradation of the ligand occurred during the assay, and the
initial rate of the process was measured. Indeed, separate chro-
–
–
–
matographic analysis revealed that only 5 10% of ADP was
converted into AMP in the assay medium during 10 min.
–
The dose response curves for ADP, 2MeSADP, and ADPbS
–
Data Processing. The bell-shaped dose response curves
were measured at ligand concentrations from 0.1 nM to 10 mM.
At millimolarconcentrations all these compounds caused a clear
were analysed by assuming that the interaction of a ligand A
at the agonistic and antagonistic receptor sites follows the mass
action law and that relative effect can be presented as a combi-
–
downturn of the dose response curves, pointing to inhibition of
–
(
)
the response at high ligand concentration Fig. 1A C . The data
were analysed by using Eq. 1, and the pKagon and pKantag values
( )
nation of two binding isotherms 8 :
(
)
were calculated Table 1 . These constants varied for different
ligands, but the maximum response evoked by all the agonists
was equal. Secondly, the results were independent on the length
[ ]
[ ]
A
+ [ ]
A
A
[ ]
1
Relative effect
=
¡
Kantag
+ [ ]
Kagon
A
Table 1
Effect of nucleoside diphosphates and their analogues on synthesis of inositol phosphates in
hP2Y1-1321N1 and NG108-15 cells at various concentrations of divalent ions
Assay conditions
+
+
2
2
[
]
[
]
Mg
Ca
EDTA
(
)
(
)
(
)
mM
Nucleotide
mM
mM
pKagon
pKantag
hP2Y1-1321N1 cells
ADP
1.80
1.80
—
1.80
1.80
0.813
10.00
—
0.813
0.813
—
—
4.00
—
7.2 § 0.1
5.9 § 0.1
6.8 § 0.1
7.8 § 0.1
6.4 § 0.1
2.1 § 0.1
1.6 § 0.1
2.0 § 0.1
2.9 § 0.1
1.8 § 0.2
2MeSADP
ADPbS
NG108-15 cells
UDP
—
1.80
1.80
—
0.813
10.00
—
—
—
4.00
6.4 § 0.1
6.1 § 0.1
6.4 § 0.1
1.8 § 0.1
2.0 § 0.2
2.0 § 0.3

101
DUAL EFFECT OF NUCLEOTIDES ON P2Y RECEPTORS
(
)
(
(
)
( )
( )
)
on the pyrimidinergic receptor or receptors expressed by NG108-15 cells.
Figure 1. Effect of ADP A, , AMP A, d , 2MeSADP B , and ADPbS C on the hP2Y1 expressed in 1321N1 cells and the
j
(
)
effect of UDP D, and UMP D,
u
±
Figure 2. Time dependence of the accumulation of inositol phosphates. The values are the means §SEM of triplicate determi-
( ) ( ) ( ) ( )
nations in 2 independent experiments. A hP2Y1-1321N1 cells were incubated with 50 lM , 5 l M , 500 nM m , or 50 nM
.
j
( )
d
( )
ADP. B NG108-15 cells were incubated with 100 l M
( )
, 10 l M
u
( )
, 1 l M
r
( ) ( )
, or 100 nM UDP.
4
±

102
of the incubation time, as shown by the linear time dependence DISCUSSION
SAK ET AL.
(
)
of formation of inositol phosphates Fig. 2A .
The receptors, expressed by hP2Y1-1321N1 and NG108-15
cells, revealed clear selectivity against adenine and uracil nu-
cleotides, respectively. This along with the EC50 values listed in
Table 1, is in good agreement with most of the results published
In parallel, the in uence of AMP and adenosine was tested
(
)
withhP2Y1-1321N1 cells.AMPhad onlyaweak effect Fig. 1A ,
whereas adenosine and the uracil nucleotides, including UDP,
clearly had no effect in this system. At 1 mM, neither AMP nor
adenosine produced any antagonist change in the 2MeSADP
curve.
(
)
for these cells receptor subtypes and collected in the P2Y Re-
(
ceptor Ligand Database, available on the Internet http://bioorg.
)
chem.ut.ee/p2y/ .
On the other hand, at high ligand concentrations a cleardown-
–
(
)
turn in the dose response curves was observed Fig. 1 , lead-
ing to a nonconventional bell-shaped plot and pointing to the
dual nature of action of agonists on the purine-selective and
pyrimidine-selective receptors studied.
Effect of UDP on Pyrimidinoceptor(s) Expressed
by NG108-15 Cells
In the NG108-15 cell line, the synthesis of inositol phos-
phates was activated by UDP, whereas the adenine nucleotides
used above had only a minor effect on this process. The NG108-
15 cells express P2Y1, P2Y2, and P2Y6 but not P2Y4, as de-
We excluded the obvious potential artefact of an antagonist
impurity, which becomes signiŽ cant in size only when high
ligand concentrations are applied. First, the purity of agonists
was checked before their use in experiments. Second, there was
no change in the pKagon and pKantag values when EDTA was
added to an amount that would complex polyvalent metal im-
purities in general. Third, the processing of a diphosphate by
ecto-enzymes was shown to be low during the assays, and even
for that low conversion level, the main product from ADP was
AMP, which is itself a weak agonist Fig. 1A on these cells and
therefore irrelevant to this question.
The dual effect of speciŽ c agonists at the G-protein coupled
7-TM receptors and the formation of the bell-shaped dose
response curves can be explained by at least two theoretical
models. The Ž rst model involves interaction of the same ago-
nist with two distinct receptor populations, which have different
afŽ nities for the ligand and produce opposite effects 12 . The
second model assumes that the stimulatory agonistic and in-
hibitory antagonistic effects of the same drug can be related to
the presence of two distinct ligand-binding sites on the receptor
molecule, which are responsible for signal transduction ago-
( )
termined by polymerase chain reaction analysis 6 . Because
( )
(
)
UDP is inactive on P2Y1 9 and P2Y2 10 , we presume that
(
)
the response to UDP 11 occurs through the P2Y6 receptor, al-
though a contribution from some other undetected and unknown
subtype cannot be excluded.
The time dependence of the UDP-initiated formation of inos-
itol phosphates was found to be linear in these cells for at least
(
)
(
)
20 min Fig. 2B , so again the time interval of 10 min was used
–
in all receptor assays. The dose response curve for UDP was
–
monitored with NG108-15 cells at concentrations from 0.1 nM
to 10 mM. Within this concentration interval, the plot revealed
–
(
)
a clear bell-shaped form Fig. 1D , and the values of pKagon and
pKantag were calculated, as listed in Table 1. Under the same
experimental conditions, UMP had only a weak effect on the
(
)
synthesis of inositol phosphates, with EC50
1.0 § 0.6 mM
=
(
)
(
)
Fig. 1D , whereas uridine had no effect at 1 mM concentration.
(
)
At 1 mM, neither UMP nor uridine had an antagonist effect on
the UDP response.
(
)
(
)
( )
nistic site and inhibition antagonistic site , respectively 8 . In
both models, the difference between agonistic and antagonistic
effects of a ligand is determined by differences in speciŽ city
+
+
Effect of Mg2 and Ca2 Ions and EDTA on Synthesis
of Inositol Phosphates
+
+
The concentration of Mg2 and Ca2 ions was varied in the of the appropriate binding sites, and even the parameters calcu-
–
–
assay mixture to analyse the possible role of the nucleotide ion latedfrom the bell-shape dose response curve have quite similar
complexes in formation of the receptor response. For both cell meanings.
+
lines the concentration of Mg2 in the assay medium was in-
( )
According to the two-site receptor model 8 , the dissocia-
creased from 0.8 to 10 mM. In other experiments divalent metal tionconstant Kantag should characterisethe antagonistic receptor
ions were omitted from the assay medium and 4 mM EDTA sites. In the case of the P2Y receptor subtypes, the binding prop-
–
was added. Under these rather extreme conditions, the dose
erties of these sites are still not clearly understood, because only
response curves were still bell-shaped and described by Eq 1, a few and usually nonspeciŽ c antagonists have been described.
–
yielding the dissociation constants pKagon and pKantag close to Therefore, the analysis of the bell-shaped dose response curves
(
)
the values obtained under the routine assay conditions Table 1 . seems to provide a method for characterisation of the speciŽ city
On the other hand, the maximal rate of synthesis of inositol phos- patterns of the antagonistic sites on purinergic and pyrimidiner-
phates in both cell lines was affected under these conditions. For gic receptor subtypes. The afŽ nities there are very weak, but the
+
example, in the presence of 10 mM Mg2 the ADP- and UDP- effect of those sites nevertheless becomes discernible at diphos-
(
)
induced maximal responses were 1.8 and 2.2 times more than phate concentrations >10 lM Fig. 1 .
the maximal effect obtained under the standard assayconditions;
Recent molecular modelling studies on the human P2Y1 re-
in the absence of the divalent metal ions, however, about half of ceptor structure has revealed the possibility of ATP binding to
( )
at least two distinct domains of the receptor molecule 13 . One
the standard response was observed.

103
DUAL EFFECT OF NUCLEOTIDES ON P2Y RECEPTORS
of these domains was proposed to be located outside of the TM REFERENCES
(
)
1. Barnard, E. A., Simon, J., and Webb, T. E. 1997 Nucleotide receptors
segments and was formed by the extracellular loops 2 and 3,
whereas the other proposed binding domain was located within
the TM core. The former binding was calculated to be weaker
and was hypothesised to function to move an ATP molecule into
in the nervous system. An abundant component using diverse transduction
mechanisms. Mol. Neurobiol. 15, 103 129.
2. Ralevic, V., and Burnstock, G. 1998 Receptors for purines and pyrim-
idines. Pharmacol. Rev. 50, 413 492.
3. Barnard, E. A. 1997 Protein structures in receptor classiŽ cation. Ann. N.Y.
Acad. Sci. 812, 14 28.
–
(
)
–
(
)
the functional strong binding site 13 . These domains may well
correspond to the agonistic and antagonistic sites predicted by
the two-site receptor model.
(
)
–
¨
(
)
4. Jarv, J., Toomela, T., and Karelson, E. 1993 Dual effect of carbachol on
–
the muscarinic receptor. Biochem. Mol. Biol. Int. 30, 649 654.
The theoretical analysis of muscarinic receptor interaction
with speciŽ c ligands has also suggested distinct locations of ag-
Ê
+
¨
(
)
5. Jarv, J., Hautala, R., and Akerman, K. E. O. 1995 Dual effect of muscarinic
receptor agonists on Ca2 mobilization in SH-SY5Y neuroblastoma cells.
(
)
(
)
onist carbachol and antagonist N -methylscopolamine mole-
–
Eur. J. Pharmacol. 291, 43 50.
(
)
6. Webb, T. E., and Barnard, E. A. 1999 Molecular biology of P2Y receptors
(
)
cules on the receptor 14 . Given that variations of structure
–
expressed in the nervous system. Prog. Brain Res. 120, 23 31.
7. Ayyanathan, K., Webb, T. E., Sandhu, A. K., Athwal, R. S., Barnard, E. A.,
and Kunapuli, S. P. 1996 Cloning and chromosomal localization of the
human P2Y1 purinoceptor. Biochem. Biophys. Res. Commun. 218, 783
of classical muscarinic agonists and antagonists have revealed
rather different structure activityrelationships for these ligands,
the appropriate binding sites should possess different patterns,
thereby explaining the existence of very potent and highly
–
(
)
–
788.
¨
(
)
8. Jarv, J. 1995 A model of non-exclusive binding of agonist and antagonist
(
)
selective ligands of both types of activity 15 .
–
on G-protein coupled receptors. J. Theor. Biol. 175, 577 582.
9. Simon, J., Webb, T. E., King, B. F., Burnstock, G., and Barnard, E. A. 1995
Characterisation of a recombinant P2Y purinoceptor. Eur. J. Pharmacol.
291, 281 289.
On the other hand, the data listed in Table 1 reveal that vari-
ation in ligand structure resulted in rather similar changes in
Kagon and Kantag for the nucleotides. Thus the putative binding
sites on the P2Y receptors seem to possess rather close speci-
Ž city patterns, presumably dominated by the phosphate groups.
This may be one of the reasons why the discovery of selec-
tive and potent antagonists for these receptors has been elu-
sive for so long. We hope that this information about the puta-
tive antagonistic sites may be useful for development of these
studies.
(
)
–
10. Nicholas, R. A., Watt, W. C., Lazarowski, E. R., Li, Q., and Harden, T. K.
(
)
1996 Uridine nucleotide selectivity of three phospholipase C-activating
P2 receptors: identiŽ cation of a UDP-selective, a UTP-selective, and an
ATP-and UTP-speciŽ c receptor. Mol. Pharmacol. 50, 224 229.
–
¨
(
)
11. Sak, K., Webb, T. E., Samuel, K., Kelve, M., and Jarv, J. 1999 Only
pyrimidinoceptors are functionally expressed inmouse neuroblastoma cell
lines. Mol. Cell Biol. Res. Commun. 1, 203 208.
12. Szabadi, E. 1977 A model of two functionally antagonistic receptor pop-
–
(
)
–
ulations activated by the same agonist. J. Theor. Biol. 69, 101 112.
(
)
13. Moro, S., Hoffmann, C.,and Jacobson, K. A. 1999 Role of the extracellular
loops of G protein-coupled receptors in ligand recognition: a molecular
ACKNOWLEDGEMENT
–
modeling study of the human P2Y1 receptor. Biochemistry 38, 3498 3507.
(
)
This work was supported by the EU Inco-Copernicus Grant
14. Fanelli, F., Menziani, M. C., and Benedetti, P. G. 1995 Molecular dynam-
ics simulations of m3-muscarinic receptor activation and QSAR analysis.
(
)
IC 15-CT96-0919 DG 12-CDPE , and by the Wellcome Trust.
–
Bioorg. Med. Chem. 11, 1465 1477.
¨
(
We thank Kulli Samuel Institute of Chemical Physics and Bio-
¨
(
)
15. Jarv, J. 1992 Neurotoxic agents interacting with the muscarinic acetyl-
)
physics, Tallinn, Estonia for cultivation of the cells and Gerda
choline receptor. In Handbook of Experimental Pharmacology, vol. 102,
Selective Neurotoxicity Herken, H., and Hucho, F., eds , pp. 659 680,
(
)
Raidaru Institute of Chemical Physics, Tartu, Estonia for chro-
matographic analysis of nucleotides.
–
(
)
Springer-Verlag, Berlin.

Copyright of IUBMB Life is the property of Wiley-Blackwell and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.