Using voltage-clamp procedures on Xenopus oocytes, agonist-evoked ionic currents by P2X receptors resulting from the coexpression of P2X(2) and P2X(3) subunits were compared against the agonist responses of homomeric P2X(2) and P2X(3) receptors. With the quantity of P2X(3) mRNA kept constant and quantity of P2X(2) mRNA progressively increased, expressed P2X receptors changed from a P2X(3)-like receptor to a P2X(2)-like receptor. In all cases, however, agonist-evoked responses comprised biphasic (fast and slow) currents-the former showing the properties of P2X(3) receptors and latter consistent with the presence of P2X(2) and P2X(2/3) receptors. Using desensitization procedures, the P2X(3)-like fast current was selectively removed to allow the slow current to be studied in isolation. P2X(2/3) receptors were then characterized by slowly inactivating inward currents that were reproducible within 30 s of washout and whose pharmacological profile [selective agonists, Ap(5)A > alpha,beta-methylene ATP >> beta,gamma-methylene ATP > UTP; antagonists, TNP-ATP >> suramin > or = Reactive blue-2 (RB-2)] contrasted with the profile of P2X(2) receptors (Ap(5)A, alpha,beta-methylene ATP, beta,gamma-methylene ATP, and UTP inactive; antagonists, RB-2 > TNP-ATP > suramin). Thus, our experiments reveal that coexpression of two P2X subunits, which of themselves can generate functional homomeric receptors, results in a complex population of heterogeneous P2X receptors-in this case P2X(2), P2X(3), and P2X(2/3) receptors. Depending on the relative levels of P2X subunit coexpression, the operational profile of the resultant P2X receptors can change from one phenotype to another. This spectrum may explain the variability of agonist responses in small sensory neurons that also express P2X(2) and P2X(3) subunits in different amounts.