RESEARCH REPORT Phosphate-Substituted ATP Analogs Are Antagonists at Human P2Y 1 Purinoceptors 1 Katrin Sak,* Gerda Raidaru,* Tania E. Webb,† and Jaak Ja ¨ rv* ,2 *Institute of Chemical Physics, Tartu University, 2 Jakobi Street, 51014 Tartu, Estonia; and Cell Signaling Laboratory, Department of Biological Sciences, De Montfort University, Leicester, The Hawthorn Building, The Gateway, Leicester LE1 9BH, United Kingdom Received May 17, 2000 Key Words: ATP phosphate-substituted analog; an- tagonist; P2Y 1 receptor; inositol phosphate assay. Among the G-protein-coupled nucleotide receptors the P2Y 1 subtype can be distinguished by its specificity for ade- nine nucleotides (1). ADP is a potent agonist at this receptor and its activity tolerates several chemical modifications at the diphosphate group as well as at the adenosine moiety (2– 4). In contrast, ATP has been found to be an antagonist at this subtype (3, 5), although its inhibitory effect may be (and commonly is) overshadowed by the agonistic activity of ADP, rapidly formed through ATP degradation under assay condi- tions (3, 5, 6). To avoid this accompanying effect, it is neces- sary to test more chemically stable ATP derivatives, obtained through substitution of the phosphorus-bridging oxygen at- oms by methylene or imido groups in the triphosphate moi- ety. In this study, the activity of these ATP derivatives at the transfected human P2Y 1 receptors was studied by measuring their effect on the breakdown of inositol phospholipids in hP2Y 1 -1321N1 cells. ATP analogs with a methylene group between the first and the second phosphorus atoms (methyleneATP or Ap[CH 2 ]pp) and the second and the third phosphorus atoms (methyleneATP or App[CH 2 ]pp) were obtained from Sigma. The purity of these ligands was above 99% as deter- mined by HPLC analysis (7) and no appropriate diphos- phates were detected in the samples used. It was found that these ATP analogs were stable under the assay conditions used and were unable to initiate the P2Y 1 receptor response in hP2Y 1 -1321N1 cells at ligand concentrations of up to 1 mM. At the same time both these derivatives inhibited the receptor response, elicited by 2MeSADP (Figs. 1 and 2). The Schild plot for the inhibitory effect for App[CH 2 ]p had a slope of 1.0 0.1 and a K i of 66 13 M, while that for Ap[CH 2 ]pp had a slope 1.1 0.2 and a K i of 191 60 M. Thus, both of these derivatives were competitive antagonists at the P2Y 1 receptors. However, the positioning of the methylene group had a clear influence on the potency of these ATP derivatives. The response, elicited by 2MeSADP in hP2Y 1 -1321N1 cells, was also inhibited by imidoATP (App[NH]p), the only com- mercially available imido-derivative of ATP (Fig. 3). Some activation of the inositol phosphate formation was observed in the presence of App[NH]p at micromolar concentrations. However, as the extent of this effect was clearly increased if longer assay incubation times were used, this agonist re- sponse was related to the degradation products of App[NH]p and is analogous to the agonistic activity observed in the presence of ATP and discussed in detail in our previous report (5). Although the chemical stability of App[NH]p is known to be greater than that of ATP (8, 9), a significant amount (up to 10%) of the appropriate diphosphate App[NH 2 ] was found even in the commercial sample (Sigma, A2647) of this ATP analog. Therefore, the antagonistic effect of App[NH]p was quantified after subtraction of the small ago- nist response observed and a K i of 3.8 0.9 M was calcu- lated from the Schild plot (Fig. 3). The latter K i value was similar to that obtained for ATP (5.0 1.2 M), measured under the same assay conditions (5). This is not surprising as the spatial structures of these ligands are rather similar, as the angles between the P–O–P and P–N–P bonds remain close, 130 and 127°, respectively (8). On the other hand, introduction of the methylene group into the triphosphate moiety yielded the P–C–P angle of 117° (8), obviously leading to a more significant disturbance of the ligand conformation. 1 This work was supported by EU INCO-Copernicus Grant IC 15-CT96-0919. T.E.W. thanks the Wellcome Trust for support. The authors are grateful to Prof. S. P. Kunapuli (Temple University Medical School, Philadelphia) for the cell line hP2Y 1 -1321N1 and to Mrs. K. Samuel (National Institute of Chemical Physics and Bio- physics, Tallinn) for cell cultivation. 2 To whom correspondence should be addressed. Fax: +372-7-375- 247. E-mail: jj@chem.ut.ee. 0003-9861/00 $35.00 171 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved. Archives of Biochemistry and Biophysics Vol. 381, No. 1, September 1, pp. 171–172, 2000 doi:10.1006/abbi.2000.1975, available online at http://www.idealibrary.com on