The melatonin-binding protein in chicken brain membranes was characterized using both [I-125]-2-iodomelatonin and [H-3]-melatonin as radioligands, Saturation studies conducted at 25 degrees C revealed a single class of binding site with dissociation constants of 24 +/- 4.8 pM (n = 7) and 125 +/- 21 pM (n = 6) for the iodinated and tritiated ligands, respectively. Calculation of the affinity constant using data from kinetic experiments gave values of 2.2 +/- 0.4 pM and 135 +/- 15 pM for the iodinated and tritiated ligands, respectively. Competition studies showed that the rank order of inhibition of binding by melatonin analogues was similar for both radioligands (2-iodomelatonin > melatonin > 2,3-dihydromelatonin > N-acetyl-5-methoxykynurenamine > N-acetylserotonin > 5-methoxytryptamine). The calculation of Ki, which depends upon the affinity constant, was 22 +/- 4.9 pM and 129 +/- 21 pM for 2-iodomelatonin and melatonin, respectively, when the affinity constant derived from the [I-125]-2-iodomelatonin saturation experiments was used, but 4.9 +/- 1.5 pM and 33 +/- 5.5 pM when the kinetically derived constant was used. When [H-3]-melatonin was used, the Ki for melatonin was 72 +/- 8 pM and 20 +/- 4.6 pM for 2-iodomelatonin and melatonin. Binding of [I-125]-2-iodomelatonin to the membranes was partially reversible at 25 degrees C in contrast to the complete reversibility of [H-3]-melatonin. Examination of the effects of temperature on binding indicated that at 37 degrees C both association and dissociation of both ligands were accelerated. Closer examination showed that at 37 degrees C there was a loss of approximately 40% of the [I-125]-2-iodomelatonin binding sites and little influence upon the affinity of binding with time. By contrast, when [H-3]-melatonin was used, the affinity decreased fourfold, with only a slight change in the number of sites. If membranes were incubated at 37 degrees C and then switched 25 degrees C, binding increased, emphasizing the fact that the binding sites were not destroyed. Whereas there appears to be little doubt that 2-iodomelatonin is a biologically active melatonin agonist, the binding of the radioactive form of this agonist to the putative melatonin receptor binding site is quite different from that of the endogenous ligand. This may have serious consequences in studies where receptor content is determined following physiological or pharmacological interventions.
