Among the 18 ionotropic glutamate receptor subunits identified in the mammalian central nervous system, five (delta1, delta2, GluR7, chi2 and NR3A, formerly called NMDAR-L or chi1) reportedly fail to form functional ion channels in heterologous expression systems. Four of these subunits, delta1, delta2, chi2 and NR3A, have not even been shown to bind glutamatergic ligands, relegating them to the status of 'orphan' receptors. We used a domain transplantation approach to investigate potential functional properties of the putative ion channel domains of four of these subunits. By exchanging ion pore domains between functional glutamate receptors (GluR1, GluR6 and NMDAR1) with known pore properties we first tested the feasibility of the domain swapping method. We demonstrate that ion channel domains can be transplanted between all three functional subfamilies of ionotropic glutamate receptors. Furthermore, exchange of ion pore domains allows identification of those channel properties determined exclusively by the ion pore. We then show that transplanting the pore domain of GluR7 into either GluR1 or GluR6 generates perfectly functional ligand-gated ion channels that allow characterization of electrophysiological and pharmacological properties of the GluR7 pore domain. In contrast, delta1, delta2 and NR3A do not produce functional receptors when their pore domains are transplanted into either the AMPA receptor, GluR1, the kainate receptor, GluR6, or the NMDA receptor, NMDAR1. We speculate that the orphan receptors delta1 and delta2, and the NMDA receptor-like subunit NR3A may serve some modulatory function, rather than contributing to the formation of ion channels.