The synthesis and preliminary pharmacological evaluation of 8,9-dihydroxy-1,2,3,11b-tetrahydrochromeno[4,3,2,-de]isoquinoline (5, now named dinoxyline) is described. This molecule was designed as a potential bioisostere that would conserve the essential elements of our beta-phenyldopamine D(1) pharmacophore (i.e., position and orientation of the nitrogen, hydroxyls, and phenyl rings). Previously, we have rigidified these elements using alkyl bridges, as exemplified in the dopamine D(1) full agonist molecules dihydrexidine (1) and dinapsoline (2). This approach has been modified and we now show that it is possible to tether these elements using an ether linkage. Preliminary pharmacology has revealed that 5 is a potent full D(1) agonist (K(0.5) <10 nM; EC(50)=30 nM), but also has high affinity for brain D(2)-like and cloned D(2) and D(3) receptors. Interestingly, whereas 1 and 2 and their analogues have only moderate affinity for the human D(4) receptor, 5 also has high affinity for this isoform. Moreover, although N-alkylation of 1 and 2 increases D(2) affinity, the N-allyl (15) and N-n-propyl (17) derivatives of 5 had decreased D(2) affinity. Therefore, 5 may be engaging different amino acid residues than do 1 and 2 when they bind to the D(2) receptor. This is the first example of a ligand with high affinity at all dopamine receptors, yet with functional characteristics similar to dopamine. These rigid ligands also will be useful tools to determine specific residues of the receptor transmembrane domains that are critical for agonist ligand selectivity for the D(4) receptor.