Cis–trans isomerisation in rhodopsin occurs in picoseconds

Abstract

IT has been believed for some time that the primary event in vision, the photochemical formation of bathorhodopsin, can be attributed to a cis–trans photoisomerisation¹. Recently this model has been questioned. Busch et al. proposed that the less-than-6-ps formation time of bathorhodopsin from rhodopsin does not allow significant isomerisation of the 11-cis chromophore to an all-trans isomer². This apparent difficulty with the cis–trans photoisomerisation model has prompted alternative models including (1) a mechanism involving deprotonation of the Schiff base nitrogen³, (2) proton transfer from the retinal methyl at position five to opsin^4,5 (involving the shifting of double bonds along the polyene chain to form a ‘retro’ type retinal), and (3) a photoinduced electron transfer to retinal from a protein donor group⁶. We have approached this question by performing picosecond absorption kinetic measurements on the formation time of bathorhodopsin from bovine rhodopsin and isorhodopsin. The essence of this experiment is that bathorhodopsin, being the common photo-product of rhodopsin (11-cis retinal) and isorhodopsin (9-cis retinal) must be an isomerised product of at least one of these pigments, but could be a product of both pigments (that is, basically all-trans retinal). Thus formation time measurements of bathorhodopsin from the two primary pigments can settle whether isomerisation can take place on the picosecond time scale.