Violet bioluminescence and fast kinetics from W92F obelin: structure-based proposals for the bioluminescence triggering and the identification of the emitting species

Obelin from the hydroid Obelia longissima and aequorin are members of a subfamily of Ca(2+)-regulated photoproteins that is a part of the larger EF-hand calcium binding protein family. On the addition of Ca(2+), obelin generates a blue bioluminescence emission (lambda(max) = 485 nm) as the result of the oxidative decarboxylation of the bound substrate, coelenterazine. The W92F obelin mutant is noteworthy because of the unusually high speed with which it responds to sudden changes of [Ca(2+)] and because it emits violet light rather than blue due to a prominent band with lambda(max) = 405 nm. Increase of pH in the range from 5.5 to 8.5 and using D(2)O both diminish the contribution of the 405 nm band, indicating that excited state proton transfer is involved. Fluorescence model studies have suggested the origin of the 485 nm emission as the excited state of an anion of coelenteramide, the bioluminescence reaction product, and 405 nm from the excited neutral state. Assuming that the dimensions of the substrate binding cavity do not change during the excited state formation, a His22 residue within hydrogen bonding distance to the 6-(p-hydroxy)-phenyl group of the excited coelenteramide is a likely candidate for accepting the phenol proton to produce an ion-pair excited state, in support of recent suggestions for the bioluminescence emitting state. The proton transfer could be impeded by removal of the Trp92 H-bond, resulting in strong enhancement of a 405 nm band giving the violet color of bioluminescence. Comparative analysis of 3D structures of the wild-type (WT) and W92F obelins reveals that there are structural displacements of certain key Ca(2+)-ligating residues in the loops of the two C-terminal EF hands as well as clear differences in hydrogen bond networks in W92F. For instance, the hydrogen bond between the side-chain oxygen atom of Asp169 and the main-chain nitrogen of Arg112 binds together the incoming alpha-helix of loop III and the exiting alpha-helix of loop IV in WT, providing probably concerted changes in these EF hands on calcium binding. But this linkage is not found in W92F obelin. These differences apparently do not change the overall affinity to calcium of W92F obelin but may account for the kinetic differences between the WT and mutant obelins. From analysis of the hydrogen bond network in the coelenterazine binding cavity, it is proposed that the trigger for bioluminescence reaction in these Ca(2+)-regulated photoproteins may be a shift of the hydrogen bond donor-acceptor separations around the coelenterazine-2-hydroperoxy substrate, initiated by small spatial adjustment of the exiting alpha-helix of loop IV.