Reduction of ferric iron could drive hydrogen tunneling in lipoxygenase catalysis: Implications for enzymatic and chemical mechanisms

Recent kinetic measurements of the oxidation of linoleic acid by soybean lipoxygenase show that the thermal rate constants, k(H) = 280 (+/-12) s(-1) and k(D) = 5.0 (+/-0.1) s(-1), are weakly temperature dependent within the temperature interval 30-50 degrees C. The primary kinetic k(H)/k(D) isotope effect is almost temperature independent, and is one order of magnitude larger than expected for a single rate determining isotopically sensitive step. It has therefore been predicted that hydrogen tunneling predominates in this enzyme-catalyzed reaction (Jonsson, T.; Glickman, M. H.; Sun, S.; Klinman, J. P. J. Am. Chem. Sec. 1996, 118, 10319). Our analysis shows that the lipoxygenase reaction can indeed proceed through a wholly quantum-mechanical pathway. While neither a tunneling correction nor groundstate tunneling of a proton through a one-dimensional potential barrier gives a satisfactory explanation for the isotope effects as measured in the experiment, a dissociation involving a two-step mechanism can explain the experimental observations. The first step is the rate-determining step where a hydrogen species tunnels from linoleic acid to lipoxygenase through a two-dimensional potential barrier. The second step is a relaxation process where an electron is transferred from the metastable intermediate to the Fe3+ cofactor in the enzyme active site. This electron transfer causes hydrogen tunneling to be effectively irreversible without introducing a temperature dependence to the reaction. We should note, however, that the two steps are not separable, and cannot be seen as a sequence of independent processes. A consequence of this model is that the tunneling pathways of hydrogen and deuterium are not identical, influencing the magnitude of the measured k(H)/k(D) isotope effect. This model also serves as an explanation for the role of iron in lipoxygenase, in contrast to the role of iron in many other oxygenases. Iron centers in oxygenases are often necessary for oxygen activation, yet there is no evidence that the iron cofactor in lipoxygenase interacts with molecular oxygen at any stage of the catalytic cycle (Glickman, M. H.; Klinman, J. P. Biochemistry 1996, 35, 12882). In lipoxygenase, it is possible that reduction of the ferric iron cofactor serves as an electron sink that drives hydrogen tunneling. An electron-gated hydrogen transfer mechanism, such as suggested, could also have relevance to an array of non-enzymatic organometallic reactions.