Anomalous intramolecular hydrogen bond in 10-(hydroxyalkyl)isoalloxazines, as revealed by photoelectron spectroscopy, and the implications for optical spectra

On the Enigma of Old Yellow Enzyme's Spectral Properties

“…This red shift is considerably larger than usually observed for the isoalloxazine So + S1 transition upon changing the solvent polarity [25,26]. Also the maximum of the near-ultraviolet transition is slightly red-shifted as compared to free FMN, but this behaviour is usual for the isoalloxazine So-+S, transition [28]. Apart from the appearance of the new absorption band centered at 16750 cm-' (597 nm), the most marked change induced by the binding of hydroxybutylbenzamide to old yellow enzyme is a decrease of the absorbance over the entire flavin spectrum and a blue-shift of the 21 700 cm-' (461 nm) band maximum, returning to the position of free FMN.…”

“…In previous work [28], we showed this transition to contain both nn* and nn* character and it was found that rotation of methyl substituents considerably affects a-n separation in orbitals localized on the isoalloxazine carbonyl groups [27]. It is well known that carbonyl nn* transitions have large intrinsic magnetic transition moments [41], hence for the isoalloxazine So+S2 we encounter the situation of a relatively small p (m* component more or less sign-forbidden [28]) and a relatively large m. This situation is particularly sensitive to external perturbation, which may explain the difference in circular dichroic absorption observed for the two long-wavelength bands in the native enzyme (Fig. IB).…”

“…An attractive alternative explanation would be a complex of FMN with a L-amino acid in the apoprotein, presumably L-tyrosine, which has been indicated to be involved in the binding of FMN to the apoprotein [34,43]. The substantial difference in character between the isoalloxazine S1 and S2 electronically excited states [28] is probably the reason why the dichroism of the So+S2 transition is not fully compensated in the complex of FMN with the apoprotein. In previous work [28], we showed this transition to contain both nn* and nn* character and it was found that rotation of methyl substituents considerably affects a-n separation in orbitals localized on the isoalloxazine carbonyl groups [27].…”

“…IB). For the So-+Sl transition, having a large p [25,26,28], it would be fortuitous that hindered rotation of methyl groups alone should be capable of compensating the effect of chirality of the FMN D-ribose side chain. An attractive alternative explanation would be a complex of FMN with a L-amino acid in the apoprotein, presumably L-tyrosine, which has been indicated to be involved in the binding of FMN to the apoprotein [34,43].…”

“…Electron paramagnetic resonance spectra were measured on a Bruker 200-tt ESR spectrometer equipped with a liquid helium cooling system. CNDO/S-CI calculations on isoalloxazine-phenol complexes were performed using the parameters and isoalloxazine molecular geometry from previous work [27,28]. The phenol molecular structure was taken from the literature [32,33].…”