The mechanism of flavin photoreduction by the amino acids, EDTA, and nitrilotriacetate is shown to be due to light-induced charge separation, which is irreversible in the dark. The irreversibility originates from the decarboxylation of the amino acid radical. This fast process changes the redox properties of the radical and makes a further donation of an electron equivalent possible. In the case of EDTA the electron acceptor of the second electron is flavin, which was left unexcited by the flash or is formed by dismutation from the flavosemiquinone, generated in the primary one-electron transfer process. In contrast to this, a mechanism for the flavin photoreduction by nitrilotriacetate is proposed, in which the decarboxylated nitrilotriacetate radical adds to the flavosemiquinone to yield an alkylated flavohydroquinone. The latter decays to free reduced or oxidized flavin, depending on the position of addition at the flavin chromophore. The difference in reaction mechanism between the nitrite anion, EDTA, and nitrilotriacetate is discussed in terms of differences in molecular structure.

Abstract— With respect to literature data, which seemed to be contradictory, we examined the fate of ethylenediaminetetraacetate (EDTA) and nitrilotriacetate (NTA) in flavin‐sensitized photooxidation reactions by quantitation of all possible products, i.e. CO2 by acidimetric titration, CH2O by colometry and glyoxylate enzymatically; N‐methyl groups were quantified by 1H‐NMR. Nitrilotriacetate yields 100% oxidative photodecarboxylation with CH2O formation, while EDTA behaves quite differently, yielding quantitative amounts of CO2 and glyoxylate, but no CH2O. The electron balance in this reaction is found to be maintained by co‐formation of one equivalent >N‐CH3. From these data it is concluded that the photooxidation of the structural element >NCH2C02 containing a tertiary nitrogen atom, occurs by electron transfer, followed by decarboxylation. The neutral radical that is formed is reductant and will give off an electron equivalent to yield the easily hydrolysed structural element >N+= CH2. With EDTA, containing the fragment –O2CCH2 >NCH2CH2N< CH2CO2–, the electron transfer towards flavin is strongly favoured by stabilisation of a cisoid mesomeric radical cation (02cch2 >+ NCH2CH2NCH2CO2 o2cch2 >NCH2CH2N+CH2CO2), which, after loss of CO2 undergoes internal 1,6‐hydrogen shift to yield the more stable radical: CH2>CH2CH2N CH2CO2CH2lt NCH2CH2NHCO2. The latter radical will easily be oxidized to the iminium ion >N*= CH‐C02‐. which in turn hydrolyses to glyoxylate. Chelation of EDTA by metal cations prevents the 1,6‐hydrogen shift, and leads to the product pattern obtained in the NTA photoreaction.

Abstract

Evidence is given for the fact that the excited flavin triplet (3F10*x) exhibits competitive le-and 2e-transfer chemistry, depending on the nature of the photosubstrate. As an 'external' photoreductant, the 2e-donor borohydride has been investigated. Borohydride is found to compete effectively with the 'internal' 1 e-donors, namely excess starting flavin in the ground state (Flax) and, as primary product, (alky1)dihydroflavin (RFl,,dH). It will be shown that some of the flavin radicals, observed by earlier authors in the reaction of 3Fl& with CH or C -COO-substrates or in the autophotolytic side-chain cleavage of riboflavin, are due to the dye-dye reaction (I) :(1) H i In contrast flavin photoreduction by borohydride or hydrocarbon substrates need not involve radicals, but may in fact be a hydride or 'carbanion-plus-proton' addition towards the highly polar and considerably basic (pK = 4.4) acceptor triplet (cf. reaction 11) RCOO-The products are much more photoreactive than the starting substrates, which leads to the secondary photocomproportionation (111) : 3Fl,*, + R-Fl,,dH + HFl + RFl, (R = alkyl or H). (111) This latter reaction is the second source of radicals in the system. This (cf. 11) 'photohydrogenation' of flavin is mechanistically related to the biological reduction of flavin by CH substrates.The photochemistry of flavins is in fact older than the corresponding ground-state chemistry and even biochemistry, since the side-chain photolysis of vitamin Bz was the first flavin reaction to be discovered [l]. The very complex course of this reaction, yielding lumiflavin at alkaline pH and lumichrome at neutrality, is still not completely resolved [2] (and references therein). This is largely due to a common mistake of insisting that such reactions must involve radical Dedicated to Professor Hellmut Bredereck, Universitat Stuttgart, on his 75th birthday.Abbreviations. Fl,,, flavoquinone, flavin; 3Fl,*,, excited flavin triplet; HFI Fl-, acidic and basic flavosemiquinone; Fl', oxidized flavin radical; HZFlred, flavohydroquinone, 1,5-dihydroflavin.

No abstract