PURIFIED NATIVE SUBUNITS OF BACTERIAL LUCIFERASE ARE ACTIVE IN THE BIOLUMINESCENCE REACTION BUT FAIL TO ASSEMBLE INTO THE ALPHA-BETA STRUCTURE

We have expressed the alpha and beta subunits of bacterial luciferase, encoded by luxA and luxB, from separate plasmids in Escherichia coli and developed an efficient purification scheme that yields many milligrams of protein of greater than 90% homogeneity. Earlier experiments showed that subunits synthesized separately assume conformations that do not assemble into the active luciferase heterodimer without prior denaturation. This observation led to the proposal that formation of the luciferase heterodimer involved interactions between intermediate conformations on the folding pathway of one or both of the subunits [Waddle, J. J., Johnston, T. C., & Baldwin, T. O. (1987) Biochemistry 26, 4917-4921]. Both of the subunits catalyze reduced flavin- and aldehyde-dependent bioluminescence reactions that are similar to that of the heterodimer in terms of reduced flavin binding affinity, aldehyde binding and inhibition, and kinetics of the overall bioluminescence reaction, but at an efficiency of about 5 x 10(-6) that of the heterodimer. Spectrophotometric analyses suggest that the structures of the individual subunits are similar to, but not identical to, the structures of the subunits in the heterodimer. Mixing of the two subunits under nondenaturing conditions did not lead to formation of the high specific activity heterodimer, even after prolonged incubation. Likewise, treatment of a stoichiometric mixture of the individual subunits with 5 M urea followed by 50-fold dilution of the urea did not yield the active heterodimer under the same conditions that yield high levels of active enzyme following denaturation of the native heterodimer [Ziegler, M. M., Goldberg, M. E., Chaffotte, A. F., & Baldwin, T. O. (1993) J. Biol. Chem. 268, 10760-10765]. However, refolding of the alpha and beta subunits together from 5 M urea following unfolding with 5 M guanidine HCl resulted in formation of the high specific activity alphabeta species, suggesting that the native isolated alpha and/or beta species is resistant to unfolding by 5 M urea. The results indicate that formation of the heterodimer in vivo must occur by interaction of transient subunit species that are distinct from the stable forms of the subunits that we have purified from cell extracts.