Two-stage nucleotide binding mechanism and its implications to H+ transport inhibition of the uncoupling protein from brown adipose tissue mitochondria.

The uncoupling protein (UCP) from brown adipose tissue mitochondria is the simplest H+ translocator known. H+ transport is regulated by fatty acids as activators and by pruine nucleotides as inhibitors. Nucleotide binding again is strongly influenced by the pH [Klingenberg, M. (1988) Biochemistry 27, 781-791]. Previously, by using fluorescent 2'-O-dansyl (DANS) derivatives of purine nucleotides, a two-stage binding mechanism was unraveled with a slow transition from a loose into a tight conformational state in the isolated UCP [Huang, S.-G., & Klingenberg, M. (1995) Biochemistry 34, 349-360]. Whereas with the unsubstituted nucleotides the transition to the tight state is nearly complete, various DANS and DAN (dimethylaminonaphthoyl) nucleotides bind more to the loose state. Here we investigated the relationships between the two-stage nucleotide binding and the inhibition of the H+ transport activity in reconstituted proteoliposomes. Further, limited tryptic digestion was used as an indicator of conformational change induced by the nucleotide binding in the isolated protein. The inhibition of H+ transport activity in reconstituted UCP proteoliposomes correlated only with the fraction of tight state of nucleotide binding. Unsubstituted nucleotides (ATP, GTP, and ADP) as well as DANSGTP inhibit fully the H+ transport, whereas DANSATP and DANSADP inhibit only to about 50%, and DANSAMP is nearly ineffective. Even for the loose conformational state the nucleotide derivatives exhibit considerable affinity. This allows DANSAMP to replace prebound ATP from UCP and relieve the inhibition of H+ transport by reversing the distribution of UCP from the tight into the loose conformational state. The pH dependence of the fraction of nucleotide binding in the tight state correlates closely with the pH dependence of the degree of H+ transport inhibition. Titration with DANS nucleotides of UCP incorporated into phospholipid vesicles revealed that over 70% of binding sites had an affinity comparable with that for the isolated UCP while the remaining sites displayed substantially lower affinity, due to nonhomogeneity of the reconstituted system. The sensitivity against trypsin digestion is inversely correlated with the fraction of nucleotide binding in the tight state. Whereas unsubstituted nucleotides and DANSGTP protect strongly against trypsinolysis, DANSATP and DANSADP do only partially, and DANSAMP does not at all. The counteracting influences of the DANS substitution are shown with DANSAMP, which has an affinity comparable to that of DANSATP or DANSADP but cannot form the tight inhibited complex. These data show that nucleotide binding only in the tight state is associated with a strong conformational change, which further causes an inhibition of H+ transport. In conclusion, UCP can exist in a loose noninhibited and a tight inhibited conformational state. The equilibrium between these two conformations is shifted to the tight state with unsubstituted nucleotides but remains to variable degrees in the loose state with DANS and DAN derivatives. The DANS group hinders progressively the transition to the tight state as the binding affinity of the underlying nucleotide decreases.