- Email: [email protected]
Kinetics of [35S]dATPαS interaction with P2Y1 purinoceptor in rat brain membranes
Neuroscience Letters 355 (2004) 9–12 www.elsevier.com/locate/neulet
Kinetics of [35S]dATPaS interaction with P2Y1 purinoceptor in rat brain membranes Aldo Oras, Jaak Ja¨rv* Institute of Organic and Bioorganic Chemistry, University of Tartu, 2 Jakobi Str., 51014 Tartu, Estonia Received 19 August 2003; received in revised form 26 September 2003; accepted 6 October 2003
Abstract Kinetics of [35S]dATPaS (20 deoxyadenosine-50 -[a-35S]-thiotriphosphate) interaction with rat brain membrane fragments was studied at 25 8C and at radioligand concentrations from 2 to 250 nM. At least two different ways of [35S]dATPaS interaction with the membranes were distinguished on the basis of radioligand on-rate. Firstly, the binding sites characterized by ‘fast’ on-rate can be observed. Secondly, the ‘slow’ binding sites were kinetically identified and quantified. As in both cases the bound radioligand could be displaced by excess of ATP, all these binding sites can be defined as ‘specific sites’. In the ‘slow’ binding sites isomerization of the receptor-ligand complex was observed, as is typical for interaction of antagonists with G-protein coupled receptors, and the kinetic parameters for this interaction were similar with the appropriate data for the hP2Y1 receptors expressed in 1321N1 astrocytoma cells Therefore these sites could be assigned to the same receptor subtype in brain membranes while the ‘fast’ binding sites belong to other membrane-bound proteins, also interacting with ATP and its analogues. The kinetic properties of the latter sites were not analysed in detail. q 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Purinoceptor; Rat brain; P2Y1 subtype; Radioligand assay; Kinetic mechanism; Receptor – antagonist complex isomerization
Metabotropic P2Y1 nucleotide receptors are widely distributed in nerve system, including different regions of brain, as mapped by specific anti-P2Y1 antibody [4,5] and shown by distribution of the receptor messenger RNA [17]. In parallel, attempts have been made to label these receptors on brain membranes by radioactive nucleotides and their analogues [8,10,11,13,14]. Selection of these radioligands has been made foremost on the basis of pharmacological data, i.e. proceeding from selectivity of these compounds against the receptor subtype. On the basis of these data 20 desoxyadenosine-50 -O-(1-thiotriphosphate) ([35S]dATPaS) was recognized as a reasonably specific probe for P2Y1 receptor subtype (see discussion in refs. [6,10]). However, in spite of the favourable pharmacological specificity, binding of this radioligand at various other binding sites of non-receptor origin interfered with the radioligand assay of P2Y1 receptor under the equilibrium conditions [10]. Therefore search for suitable P2Y1 radioligands has been continued with special attention on non-nucleotide purinoceptor antagonists, and recently one of these compounds, 2* Corresponding author. Tel.: þ 372-7-375-246; fax: þ372-7-375-247. E-mail address: [email protected] (J. Ja¨rv).
chloro-N 6-methyl-(N)-methanocarba-20 -deoxyadenosine30 ,50 -bisphosphate ([3H]MRS2279), was proposed as radioligand for P2Y1 receptor assay [16]. This antagonist is highly selective against the P2Y1 subtype and shows no binding on cell membranes lacking the receptor protein [16]. However, this assay has to be carried out in ice-water bath, and even then the off-rate of this ligand from its complex with the receptor was relatively fast (half-life approx 0.9 min). Thus the application of icecold buffer cannot be used to avoid the loss of the specifically bound radioligand during the dilution, filtration and washing steps of the binding assay, and therefore should lead to underestimated receptor content. Differently from [3H]MRS2279, the half-life of the offrate for [35S]dATPaS is above 3 min at room temperature and application of ice-cold buffer can effectively stop the process and make the loss of the receptor-bound radioligand negligible during the filtration assay. Therefore, proceeding from this consideration, and the possibility of kinetic differentiation between the receptor and non-receptor binding sites proposed in ref. [6], kinetics of [35S]dATPaS interaction with brain membranes was studied, keeping in
0304-3940/03/$ - see front matter q 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2003.10.029
10
A. Oras, J. Ja¨rv / Neuroscience Letters 355 (2004) 9–12
mind the implication of this approach for the assay of P2Y1 receptor sites. Brains of 7 days old rats were homogenized in ice-cold 50 mM Tris – HCl buffer (pH 7.4) and centrifuged for 10 min at 1500 £ g and 4 8C. The supernatant was decanted and centrifuged (27,000 £ g, 40 min) and the pellet was rehomogenized in the same buffer (1.5 ml per brain). Samples were stored at 2 80 8C. Dye binding method [7] was used for protein determination. The reaction of radioligand binding was initiated by addition of [35S]dATPaS (Amersham Pharmacia Biotech) into the membrane suspension (final volume 7 ml, 50 mM Tris – HCl buffer, pH 7.4, 25 8C, protein concentration 0.09 – 0.1 mg/ml). At appropriate time intervals 0.5 ml aliquots of the reaction mixture were diluted into 5 ml of ice-cold Tris – HCl buffer (50 mM, pH 7.4) and immediately filtered over pre-soaked glass-fiber filters (GF/B, Whatman). The filters were washed twice with 5 ml of the cold buffer and the filter-bound radioactivity was counted on a Beckman LS 1801 liquid scintillation counter (Ecoscint A, ICN, efficiency 95%). Before counting the filters were kept 12 h in the scintillation cocktail. The non-specific binding of [35S]dATPaS was determined in the presence of 1 mM ATP [9], added to the membrane fragments 3 min before the radioligand. In the off-rate experiment the membranes were pre-incubated with 50 nM [35S]dATPaS during 10 min. The radioligand dissociation was initiated by addition of ATP (final concentration 1 mM) and the time course of the process was followed as described above. The rate constants koff were calculated from the rate equation for monomolecular decay. In the equilibrium binding studies the incubation time was 20 min and the Kd as well as the Bmax values were calculated from a common binding isotherm. The kinetic data were analysed by means of non-linear least squares regression method (GraphPad PRISMe, San Diego, CA, USA). A typical time-course of [35S]dATPaS binding with rat brain membrane fragments is shown in Fig. 1 (total binding – filled symbols, non-specific binding – open symbols). It can be seen that most of the membrane-bound radioligand was displaced by the excess of the non-radioactive ligand and thus followed the criteria of specific binding. At the same time the level of the non-specific binding (Bnonsp) was rather low. This kinetic approach clearly differentiated between the ‘fast’ and ‘slow’ specific binding sites. The radioligand binding at the ‘fast’ sites occurred rapidly and this process was complete before the first sample was applied to the filter (approx 10 s). The radioligand binding in the ‘slow’ sites could be followed during several minutes and was well described by the first-order rate equation: Bt ¼ Bnonsp þ Bfast þ Bslow ½1 2 expð2kobs tÞ;
ð1Þ
where Bt is the amount of the membrane-bound radioligand at time t, Bnonsp þ Bfast is the intercept value extrapolated from the exponential part of the Bt vs t plot to t ¼ 0, Bslow is
Fig. 1. Time course of 50 nM [35S]dATPaS interaction with rat brain membranes (X) and the non-specific binding of this radioligand measured in the presence of 1 mM ATP (W). The exponential part of the kinetic curve corresponds to the radioligand binding with the ‘slow’ binding sites and was analyzed by Eq. (1) in the text.
the amplitude of the exponential function and corresponds to the amount of the ‘slow’ binding sites. The ‘slow’ sites could be assigned to nucleotide receptors, localized on the brain membrane fragments, and this conclusion is based on the following. Firstly, in similar kinetic experiments with the 1321N1 astrocytoma cell membranes, the ‘slow’ sites were observed on the transfected hP2Y1-1321N1 cells, while the wild cells,
Fig. 2. Dependence of the pseudo-first-order rate constant of [35S]dATPaS binding to ‘slow’ binding sites in rat brain membranes (kobs) on the radioligand concentration. The data were processed by Eq. (2) as described in the text. The standard errors of separate determinations are shown as bars.
A. Oras, J. Ja¨rv / Neuroscience Letters 355 (2004) 9–12
11
Table 1 Results of kinetic analysis of [35S]dATPaS interaction with ‘slow’ specific binding sites in rat brain membrane fragments (50 mM Tris–HCl buffer, pH 7.4, 25 8C) compared with similar results for transfected hP2Y1-1321N1 astrocytoma cell membranes from ref. [6] Parameters
Rat brain membrane fragments
hP2Y1-1321N1 cell membranes [6]
Ka ki k2i koff Kd (‘slow’ sites) Bmax (‘slow’ sites) Ki ( ¼ k2i/ki)
99 ^ 29 nM (1.1 ^ 0.1) 1022 s21 (3.4 ^ 0.5) 1023 s21 (3.2 ^ 0.6) 1023 s21 12 ^ 5 nM 20 ^ 2 pmol/mg 0.3
59 ^ 16 nM (9.0 ^ 0.8) 1023 s21 (3.9 ^ 0.7) 1023 s21 (2.8 ^ 0.7) 1023 s21 31 ^ 8 nM 2.5 ^ 0.2 pmol/mg 0.4
which do not express the P2Y 1 receptor, revealed [35S]dATPaS interaction only with the ‘fast’ binding sites of non-receptor origin [6]. Secondly, the receptor origin of the ‘slow’ binding sites in brain membranes is supported by the hyperbolic kobs vs [35S]dATPaS concentration plot, shown in Fig. 2: kobs ¼
ki ½A þ k2i : KA þ ½A
ð2Þ
This plot points to the two-step mechanism of the radioligand binding, as analysed in detail by Strickland et al. [15], and formalized by the following reaction scheme: KA
ki
R þ A Y RA Y ðRAÞ;
ð3Þ
k2i
This two-step ligand binding mechanism was found to be characteristic for interaction of specific antagonists with various G-protein coupled receptors [1,3,12], including
Fig. 3. Binding isotherms for [35S]dATPaS interaction with rat brain membranes. The specific binding (W) was measured as difference between the total and non-specific binding, the amount of the ‘slow’ binding sites (X) was calculated from kinetic curves as the amplitude of the exponential function, presented by Eq. (1) in the text. The insert shows the same binding data in Scatchard transformation.
P2Y1 receptors in the transfected hP2Y1-13212N1 cell line [6]. Therefore it could be suggested that the ‘slow’ binding sites traced by [35S]dATPaS in brain membranes also belong to a G-protein coupled receptor. Thirdly, this suggestion was also supported by comparison of binding curves constructed for the overall specific binding (characterized by sum of Bslow þ Bfast) and for Bslow alone (Fig. 3). It can be seen in Fig. 3 that the ‘slow’ binding sites, quantified by Bslow, were saturated by the radioligand as a homogeneous population, while the overall specific binding of this radioligand in brain membranes revealed clear heterogeneity. Proceeding from data on pharmacological specificity of [35S]dATPaS [10], it can be assumed that the homogeneous binding sites belong to the P2Y1 subtype. Finally, the ‘slow’ [35S]dATPaS binding sites on brain membranes and on hP2Y1-1321N1 cells have very close kinetic properties. This can be seen from comparison of the kinetic parameters listed in Table 1, including the experimental values of the radioligand dissociation rate constant koff. This means that the ‘slow’ [35S]dATPaS binding sites on the transfected astrocytoma cell membranes and in rat brain most probably belong to the same receptor subtype, P2Y1. It is important to mention that not only the off-rate but also the monomolecular equilibrium between RA and (RA), characterized by ratio of the k2i and ki values, is important for quantification if the receptor sites, as only the isomerized complex (RA) can be explicitly determined by the filtration assay (see discussion in ref. [2]). Proceeding from data in Table 1, approx 2/3 of the overall receptor sites can be directly quantified by the Bmax value in brain membranes. Therefore the genuine density of the ‘slow’ binding sites for [35S]dATPaS should be around 30 pmol/mg protein. This is significantly higher than 0.2 pmol/mg protein traced by [3H]MRS2279 in rat brain [16]. Similarly, lower amount of the binding sites was determined by [3H]MRS2279 in the transfected hP2Y1-1321N1 astrocytoma cells (0.4 pmol/mg protein, [16]) if compared with the results for [35S]dATPaS (3 pmol/mg protein [6]). This situation seems to agree with the above mentioned possibility of underestimation of the receptor sites due to relatively fast off-rate of [3H]MRS2279 in cold buffer. Moreover, the uncertainty of the titration
12
A. Oras, J. Ja¨rv / Neuroscience Letters 355 (2004) 9–12
results can be increased by the unknown Ki value for the latter compound, as the state of the appropriate monomolecular equilibrium should be considered in determination of the genuine receptor density proceeding from the Bmax value [2]. These questions cannot be finally addressed until the kinetic mechanism for [3H]MRS2279 interaction with the receptor is studied.
Acknowledgements This work was supported by Estonian Science Foundation Grant 5214 and Estonian Ministry of Education and Science Grant 2592.
References [1] J. Ja¨rv, B. Hedlund, T. Bartfai, Isomerization of the muscarinic receptor-antagonist complex, J. Biol. Chem. 254 (1979) 5595–5598. [2] J. Ja¨rv, Neurotoxic agents interacting with the muscarinic acetylcholine receptor, in: H. Herken, F. Hucho (Eds.), Selective Neurotoxicity. Handbook of Exp. Pharmacol., 102, Springer-Verlag, Berlin, Heidelberg, New York, London, Paris, Tokyo, Hong Kong, Barcelona, Budapest, 1992, pp. 659– 680. [3] M. Lepiku, A. Rinken, J. Ja¨rv, K. Fuxe, Kinetic evidence for isomerization of the dopamine receptor – raclopride complex, Neurochem. Int. 28 (1996) 591–595. [4] D. Moore, J. Chambers, H. Waldvogel, R. Faull, P. Emson, Regional and cellular distribution of the P2Y1 purinergic receptor in the human brain: striking neuronal localization, J. Comp. Neurol. 421 (2000) 374–384. [5] M. Moran-Jimenz, C. Matute, Immunohistochemical localization of the P2Y1purinergic receptor in neurons and glial cells of the central nervous system, Brain Res. Mol. Brain Res. 78 (2000) 50–58.
[6] A. Oras, K. Kilk, S. Kunapuli, E.A. Barnard, J. Ja¨rv, Kinetic analysis of [35S]dATPalphaS interaction with P2y(1) nucleotide receptor, Neurochem. Int. 40 (2002) 381–386. [7] G.L. Peterson, Determination of total protein, Methods Enzymol. 91 (1983) 95– 119. [8] G. Reiser, W. Laubinger, R. Schaefer, Nucleotide radiolabels as tools for studying P2Y receptors in membranes from brain and lung tissue, Prog. Brain Res. 120 (1999) 45 –58. [9] K. Sak, J. Ja¨rv, Adenosine triphosphate is full antagonist at human P2Y(1) purinoceptors, Neurosci. Lett. 284 (2000) 179 –181. [10] J.B. Schachter, T.K. Harden, An examination of deoxyadenosine 50 (alpha-thio)triphosphate, a ligand to define P2Y receptors and its selectivity as a low potency partial agonist of the P2Y1 receptor, Br. J. Pharmacol. 121 (1997) 338–344. [11] R. Schaefer, G. Reiser, Characterization of [35S]-ATPalphaS and [3H]-alpha, beta-MeATP binding sites in rat brain cortical synaptosomes: regulation of ligand binding by divalent cations, Br. J. Pharmacol. 121 (1997) 913–922. [12] R. Schliebs, V. Bigl, Kinetics of the interaction of dihydroalprenolol with beta-adrenergic receptor in rat cerebral cortex, Gen. Physiol. Biophys. 3 (1984) 31–46. [13] J. Simon, T.E. Webb, E.A. Barnard, Characterization of a P2Y purinoceptor in the brain, Pharmacol. Toxicol. 76 (1995) 302–307. [14] J. Simon, T.E. Webb, E.A. Barnard, Distribution of [35S]dATPalphaS binding sites in the adult rat neuraxis, Neuropharmacology 36 (1997) 1243–1251. [15] S. Strickland, G. Palmer, V. Massey, Determination of dissociation constants and specific rate constants of enzyme-substrate (or proteinligand) interaction from rapid reaction kinetic data, J. Biol. Chem. 250 (1975) 4048–4052. [16] G.L. Waldo, J. Corbitt, J.L. Boyer, G. Ravi, H.S. Kim, X.-D. Ji, J. Lacy, K.A. Jakobson, T.K. Harden, Quantitation of the P2Y1 receptor with high affinity radiolabeled antagonist, Mol. Pharmacol. 62 (2002) 1249–1257. [17] T.E. Webb, J. Simon, E.A. Barnard, Regional distribution of [35S]20 deoxy 50 -O-(1-thio)ATP binding sites and the P2Y1 messenger RNA within the chick brain, Neuroscience 84 (1998) 825 –837.
Kinetics of [35S]dATPaS interaction with P2Y1 purinoceptor in rat brain membranes Aldo Oras, Jaak Ja¨rv* Institute of Organic and Bioorganic Chemistry, University of Tartu, 2 Jakobi Str., 51014 Tartu, Estonia Received 19 August 2003; received in revised form 26 September 2003; accepted 6 October 2003
Abstract Kinetics of [35S]dATPaS (20 deoxyadenosine-50 -[a-35S]-thiotriphosphate) interaction with rat brain membrane fragments was studied at 25 8C and at radioligand concentrations from 2 to 250 nM. At least two different ways of [35S]dATPaS interaction with the membranes were distinguished on the basis of radioligand on-rate. Firstly, the binding sites characterized by ‘fast’ on-rate can be observed. Secondly, the ‘slow’ binding sites were kinetically identified and quantified. As in both cases the bound radioligand could be displaced by excess of ATP, all these binding sites can be defined as ‘specific sites’. In the ‘slow’ binding sites isomerization of the receptor-ligand complex was observed, as is typical for interaction of antagonists with G-protein coupled receptors, and the kinetic parameters for this interaction were similar with the appropriate data for the hP2Y1 receptors expressed in 1321N1 astrocytoma cells Therefore these sites could be assigned to the same receptor subtype in brain membranes while the ‘fast’ binding sites belong to other membrane-bound proteins, also interacting with ATP and its analogues. The kinetic properties of the latter sites were not analysed in detail. q 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Purinoceptor; Rat brain; P2Y1 subtype; Radioligand assay; Kinetic mechanism; Receptor – antagonist complex isomerization
Metabotropic P2Y1 nucleotide receptors are widely distributed in nerve system, including different regions of brain, as mapped by specific anti-P2Y1 antibody [4,5] and shown by distribution of the receptor messenger RNA [17]. In parallel, attempts have been made to label these receptors on brain membranes by radioactive nucleotides and their analogues [8,10,11,13,14]. Selection of these radioligands has been made foremost on the basis of pharmacological data, i.e. proceeding from selectivity of these compounds against the receptor subtype. On the basis of these data 20 desoxyadenosine-50 -O-(1-thiotriphosphate) ([35S]dATPaS) was recognized as a reasonably specific probe for P2Y1 receptor subtype (see discussion in refs. [6,10]). However, in spite of the favourable pharmacological specificity, binding of this radioligand at various other binding sites of non-receptor origin interfered with the radioligand assay of P2Y1 receptor under the equilibrium conditions [10]. Therefore search for suitable P2Y1 radioligands has been continued with special attention on non-nucleotide purinoceptor antagonists, and recently one of these compounds, 2* Corresponding author. Tel.: þ 372-7-375-246; fax: þ372-7-375-247. E-mail address: [email protected] (J. Ja¨rv).
chloro-N 6-methyl-(N)-methanocarba-20 -deoxyadenosine30 ,50 -bisphosphate ([3H]MRS2279), was proposed as radioligand for P2Y1 receptor assay [16]. This antagonist is highly selective against the P2Y1 subtype and shows no binding on cell membranes lacking the receptor protein [16]. However, this assay has to be carried out in ice-water bath, and even then the off-rate of this ligand from its complex with the receptor was relatively fast (half-life approx 0.9 min). Thus the application of icecold buffer cannot be used to avoid the loss of the specifically bound radioligand during the dilution, filtration and washing steps of the binding assay, and therefore should lead to underestimated receptor content. Differently from [3H]MRS2279, the half-life of the offrate for [35S]dATPaS is above 3 min at room temperature and application of ice-cold buffer can effectively stop the process and make the loss of the receptor-bound radioligand negligible during the filtration assay. Therefore, proceeding from this consideration, and the possibility of kinetic differentiation between the receptor and non-receptor binding sites proposed in ref. [6], kinetics of [35S]dATPaS interaction with brain membranes was studied, keeping in
0304-3940/03/$ - see front matter q 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2003.10.029
10
A. Oras, J. Ja¨rv / Neuroscience Letters 355 (2004) 9–12
mind the implication of this approach for the assay of P2Y1 receptor sites. Brains of 7 days old rats were homogenized in ice-cold 50 mM Tris – HCl buffer (pH 7.4) and centrifuged for 10 min at 1500 £ g and 4 8C. The supernatant was decanted and centrifuged (27,000 £ g, 40 min) and the pellet was rehomogenized in the same buffer (1.5 ml per brain). Samples were stored at 2 80 8C. Dye binding method [7] was used for protein determination. The reaction of radioligand binding was initiated by addition of [35S]dATPaS (Amersham Pharmacia Biotech) into the membrane suspension (final volume 7 ml, 50 mM Tris – HCl buffer, pH 7.4, 25 8C, protein concentration 0.09 – 0.1 mg/ml). At appropriate time intervals 0.5 ml aliquots of the reaction mixture were diluted into 5 ml of ice-cold Tris – HCl buffer (50 mM, pH 7.4) and immediately filtered over pre-soaked glass-fiber filters (GF/B, Whatman). The filters were washed twice with 5 ml of the cold buffer and the filter-bound radioactivity was counted on a Beckman LS 1801 liquid scintillation counter (Ecoscint A, ICN, efficiency 95%). Before counting the filters were kept 12 h in the scintillation cocktail. The non-specific binding of [35S]dATPaS was determined in the presence of 1 mM ATP [9], added to the membrane fragments 3 min before the radioligand. In the off-rate experiment the membranes were pre-incubated with 50 nM [35S]dATPaS during 10 min. The radioligand dissociation was initiated by addition of ATP (final concentration 1 mM) and the time course of the process was followed as described above. The rate constants koff were calculated from the rate equation for monomolecular decay. In the equilibrium binding studies the incubation time was 20 min and the Kd as well as the Bmax values were calculated from a common binding isotherm. The kinetic data were analysed by means of non-linear least squares regression method (GraphPad PRISMe, San Diego, CA, USA). A typical time-course of [35S]dATPaS binding with rat brain membrane fragments is shown in Fig. 1 (total binding – filled symbols, non-specific binding – open symbols). It can be seen that most of the membrane-bound radioligand was displaced by the excess of the non-radioactive ligand and thus followed the criteria of specific binding. At the same time the level of the non-specific binding (Bnonsp) was rather low. This kinetic approach clearly differentiated between the ‘fast’ and ‘slow’ specific binding sites. The radioligand binding at the ‘fast’ sites occurred rapidly and this process was complete before the first sample was applied to the filter (approx 10 s). The radioligand binding in the ‘slow’ sites could be followed during several minutes and was well described by the first-order rate equation: Bt ¼ Bnonsp þ Bfast þ Bslow ½1 2 expð2kobs tÞ;
ð1Þ
where Bt is the amount of the membrane-bound radioligand at time t, Bnonsp þ Bfast is the intercept value extrapolated from the exponential part of the Bt vs t plot to t ¼ 0, Bslow is
Fig. 1. Time course of 50 nM [35S]dATPaS interaction with rat brain membranes (X) and the non-specific binding of this radioligand measured in the presence of 1 mM ATP (W). The exponential part of the kinetic curve corresponds to the radioligand binding with the ‘slow’ binding sites and was analyzed by Eq. (1) in the text.
the amplitude of the exponential function and corresponds to the amount of the ‘slow’ binding sites. The ‘slow’ sites could be assigned to nucleotide receptors, localized on the brain membrane fragments, and this conclusion is based on the following. Firstly, in similar kinetic experiments with the 1321N1 astrocytoma cell membranes, the ‘slow’ sites were observed on the transfected hP2Y1-1321N1 cells, while the wild cells,
Fig. 2. Dependence of the pseudo-first-order rate constant of [35S]dATPaS binding to ‘slow’ binding sites in rat brain membranes (kobs) on the radioligand concentration. The data were processed by Eq. (2) as described in the text. The standard errors of separate determinations are shown as bars.
A. Oras, J. Ja¨rv / Neuroscience Letters 355 (2004) 9–12
11
Table 1 Results of kinetic analysis of [35S]dATPaS interaction with ‘slow’ specific binding sites in rat brain membrane fragments (50 mM Tris–HCl buffer, pH 7.4, 25 8C) compared with similar results for transfected hP2Y1-1321N1 astrocytoma cell membranes from ref. [6] Parameters
Rat brain membrane fragments
hP2Y1-1321N1 cell membranes [6]
Ka ki k2i koff Kd (‘slow’ sites) Bmax (‘slow’ sites) Ki ( ¼ k2i/ki)
99 ^ 29 nM (1.1 ^ 0.1) 1022 s21 (3.4 ^ 0.5) 1023 s21 (3.2 ^ 0.6) 1023 s21 12 ^ 5 nM 20 ^ 2 pmol/mg 0.3
59 ^ 16 nM (9.0 ^ 0.8) 1023 s21 (3.9 ^ 0.7) 1023 s21 (2.8 ^ 0.7) 1023 s21 31 ^ 8 nM 2.5 ^ 0.2 pmol/mg 0.4
which do not express the P2Y 1 receptor, revealed [35S]dATPaS interaction only with the ‘fast’ binding sites of non-receptor origin [6]. Secondly, the receptor origin of the ‘slow’ binding sites in brain membranes is supported by the hyperbolic kobs vs [35S]dATPaS concentration plot, shown in Fig. 2: kobs ¼
ki ½A þ k2i : KA þ ½A
ð2Þ
This plot points to the two-step mechanism of the radioligand binding, as analysed in detail by Strickland et al. [15], and formalized by the following reaction scheme: KA
ki
R þ A Y RA Y ðRAÞ;
ð3Þ
k2i
This two-step ligand binding mechanism was found to be characteristic for interaction of specific antagonists with various G-protein coupled receptors [1,3,12], including
Fig. 3. Binding isotherms for [35S]dATPaS interaction with rat brain membranes. The specific binding (W) was measured as difference between the total and non-specific binding, the amount of the ‘slow’ binding sites (X) was calculated from kinetic curves as the amplitude of the exponential function, presented by Eq. (1) in the text. The insert shows the same binding data in Scatchard transformation.
P2Y1 receptors in the transfected hP2Y1-13212N1 cell line [6]. Therefore it could be suggested that the ‘slow’ binding sites traced by [35S]dATPaS in brain membranes also belong to a G-protein coupled receptor. Thirdly, this suggestion was also supported by comparison of binding curves constructed for the overall specific binding (characterized by sum of Bslow þ Bfast) and for Bslow alone (Fig. 3). It can be seen in Fig. 3 that the ‘slow’ binding sites, quantified by Bslow, were saturated by the radioligand as a homogeneous population, while the overall specific binding of this radioligand in brain membranes revealed clear heterogeneity. Proceeding from data on pharmacological specificity of [35S]dATPaS [10], it can be assumed that the homogeneous binding sites belong to the P2Y1 subtype. Finally, the ‘slow’ [35S]dATPaS binding sites on brain membranes and on hP2Y1-1321N1 cells have very close kinetic properties. This can be seen from comparison of the kinetic parameters listed in Table 1, including the experimental values of the radioligand dissociation rate constant koff. This means that the ‘slow’ [35S]dATPaS binding sites on the transfected astrocytoma cell membranes and in rat brain most probably belong to the same receptor subtype, P2Y1. It is important to mention that not only the off-rate but also the monomolecular equilibrium between RA and (RA), characterized by ratio of the k2i and ki values, is important for quantification if the receptor sites, as only the isomerized complex (RA) can be explicitly determined by the filtration assay (see discussion in ref. [2]). Proceeding from data in Table 1, approx 2/3 of the overall receptor sites can be directly quantified by the Bmax value in brain membranes. Therefore the genuine density of the ‘slow’ binding sites for [35S]dATPaS should be around 30 pmol/mg protein. This is significantly higher than 0.2 pmol/mg protein traced by [3H]MRS2279 in rat brain [16]. Similarly, lower amount of the binding sites was determined by [3H]MRS2279 in the transfected hP2Y1-1321N1 astrocytoma cells (0.4 pmol/mg protein, [16]) if compared with the results for [35S]dATPaS (3 pmol/mg protein [6]). This situation seems to agree with the above mentioned possibility of underestimation of the receptor sites due to relatively fast off-rate of [3H]MRS2279 in cold buffer. Moreover, the uncertainty of the titration
12
A. Oras, J. Ja¨rv / Neuroscience Letters 355 (2004) 9–12
results can be increased by the unknown Ki value for the latter compound, as the state of the appropriate monomolecular equilibrium should be considered in determination of the genuine receptor density proceeding from the Bmax value [2]. These questions cannot be finally addressed until the kinetic mechanism for [3H]MRS2279 interaction with the receptor is studied.
Acknowledgements This work was supported by Estonian Science Foundation Grant 5214 and Estonian Ministry of Education and Science Grant 2592.
References [1] J. Ja¨rv, B. Hedlund, T. Bartfai, Isomerization of the muscarinic receptor-antagonist complex, J. Biol. Chem. 254 (1979) 5595–5598. [2] J. Ja¨rv, Neurotoxic agents interacting with the muscarinic acetylcholine receptor, in: H. Herken, F. Hucho (Eds.), Selective Neurotoxicity. Handbook of Exp. Pharmacol., 102, Springer-Verlag, Berlin, Heidelberg, New York, London, Paris, Tokyo, Hong Kong, Barcelona, Budapest, 1992, pp. 659– 680. [3] M. Lepiku, A. Rinken, J. Ja¨rv, K. Fuxe, Kinetic evidence for isomerization of the dopamine receptor – raclopride complex, Neurochem. Int. 28 (1996) 591–595. [4] D. Moore, J. Chambers, H. Waldvogel, R. Faull, P. Emson, Regional and cellular distribution of the P2Y1 purinergic receptor in the human brain: striking neuronal localization, J. Comp. Neurol. 421 (2000) 374–384. [5] M. Moran-Jimenz, C. Matute, Immunohistochemical localization of the P2Y1purinergic receptor in neurons and glial cells of the central nervous system, Brain Res. Mol. Brain Res. 78 (2000) 50–58.
[6] A. Oras, K. Kilk, S. Kunapuli, E.A. Barnard, J. Ja¨rv, Kinetic analysis of [35S]dATPalphaS interaction with P2y(1) nucleotide receptor, Neurochem. Int. 40 (2002) 381–386. [7] G.L. Peterson, Determination of total protein, Methods Enzymol. 91 (1983) 95– 119. [8] G. Reiser, W. Laubinger, R. Schaefer, Nucleotide radiolabels as tools for studying P2Y receptors in membranes from brain and lung tissue, Prog. Brain Res. 120 (1999) 45 –58. [9] K. Sak, J. Ja¨rv, Adenosine triphosphate is full antagonist at human P2Y(1) purinoceptors, Neurosci. Lett. 284 (2000) 179 –181. [10] J.B. Schachter, T.K. Harden, An examination of deoxyadenosine 50 (alpha-thio)triphosphate, a ligand to define P2Y receptors and its selectivity as a low potency partial agonist of the P2Y1 receptor, Br. J. Pharmacol. 121 (1997) 338–344. [11] R. Schaefer, G. Reiser, Characterization of [35S]-ATPalphaS and [3H]-alpha, beta-MeATP binding sites in rat brain cortical synaptosomes: regulation of ligand binding by divalent cations, Br. J. Pharmacol. 121 (1997) 913–922. [12] R. Schliebs, V. Bigl, Kinetics of the interaction of dihydroalprenolol with beta-adrenergic receptor in rat cerebral cortex, Gen. Physiol. Biophys. 3 (1984) 31–46. [13] J. Simon, T.E. Webb, E.A. Barnard, Characterization of a P2Y purinoceptor in the brain, Pharmacol. Toxicol. 76 (1995) 302–307. [14] J. Simon, T.E. Webb, E.A. Barnard, Distribution of [35S]dATPalphaS binding sites in the adult rat neuraxis, Neuropharmacology 36 (1997) 1243–1251. [15] S. Strickland, G. Palmer, V. Massey, Determination of dissociation constants and specific rate constants of enzyme-substrate (or proteinligand) interaction from rapid reaction kinetic data, J. Biol. Chem. 250 (1975) 4048–4052. [16] G.L. Waldo, J. Corbitt, J.L. Boyer, G. Ravi, H.S. Kim, X.-D. Ji, J. Lacy, K.A. Jakobson, T.K. Harden, Quantitation of the P2Y1 receptor with high affinity radiolabeled antagonist, Mol. Pharmacol. 62 (2002) 1249–1257. [17] T.E. Webb, J. Simon, E.A. Barnard, Regional distribution of [35S]20 deoxy 50 -O-(1-thio)ATP binding sites and the P2Y1 messenger RNA within the chick brain, Neuroscience 84 (1998) 825 –837.