Proton Transfer at Helium Temperatures during Dioxygen Activation by Heme Monooxygenases Roman Davydov, † Sergey Chemerisov, § David E. Werst, § Tijana Rajh,* ,§ Toshitaka Matsui, ‡ Masao Ikeda-Saito,* ,‡ and Brian M. Hoffman* ,† Department of Chemistry, Northwestern UniVersity, EVanston, Illinois 60208, Chemistry DiVision, Argonne National Laboratory, Argonne, Illinois 60439, Institute for Multidisciplinary Research for AdVanced Materials, Tohoku UniVersity, Sendai 980-8577, Japan Received September 3, 2004; E-mail: bmh@northwestern.edu In the hydroxylation of substrate (RH) by heme monooxygenases 1 such as cytochromes P450, 2 heme oxygenase (HO), 3 and nitric oxide synthase (NOS), 4,5 the committed portion of the catalytic cycle involves the one-electron reduction of the enzyme’s dioxygen-bound ferroheme (O 2 Fe(Por)); with the addition of two protons this leads to the hydroxylation of substrate. 1 The two protons are delivered by an elaborate distal-pocket proton-delivery network connected by H-bonds to the oxy-ferroheme. 6,7 The physiological reduction and addition of the first proton may well involve proton-coupled electron transfer, 8,9 but radiolytic cryoreduction in general forms a trapped peroxo-ferriheme state ([FeO 2 ] 7 per ; 1), 10,11 thereby decoupling the two processes, and allowing us to monitor at all temperatures both the transfer of the “first” proton to generate the hydroperoxo-ferriheme ([FeO 2 H]; 7 2), eq 1, and the subsequent activation of this species by the second proton. 12 In the first measurement of enzymatic proton transfer at liquid helium temperatures, we examine protonation of 1 in HO in H 2 O and D 2 O solvents at ca. 4 K and above, and compare these finding with analogous measurements for oxy-P450cam and for oxy-Mb. Cryoreduction of oxy-HO frozen in both H 2 O and D 2 O glycerol/ buffer medium at 77 K has been shown to afford a hydroperoxo- ferriheme EPR signal with g-tensor components g a ) [2.37, 2.180, 1.917] (Figure 1, inset). 13 Thus, proton/deuteron 1 is delivered (eq 1) without the need for thermal activation above this temperature. When oxy-HO frozen in H 2 O buffer and situated in the EPR cavity is reduced by an electron beam at ∼4.2 K, 14 a strong EPR signal from 2 (Figure 1) shows that the proton/deuteron has been delivered to the one-electron reduced oxy-heme center even at this temperature. Surprisingly, cryogenic proton transfer is not quenched when the ∼4.2 K experiment is repeated with oxy-HO exchanged into D 2 O buffer, Figure 1. As shown in Figure 1, the signal remains unchanged upon in situ annealing to ∼77 K, and the signal taken at this temperature within ∼ 20 min of irradiation matches that seen upon 77 K irradiation. Disruption of the distal network through mutation of a critical component in HO(D140X), X ) A, F, does quench helium-temperature proton transfer; as reported, eq 1 only occurs in the mutants at temperatures above ∼170-180 K. 13 The prompt delivery of “proton 1” at ∼4-7 K is not seen in P450cam, even though it too has a distal-pocket proton-delivery network. 15 As reported, when the camphor complex of oxy- P450cam is cryoreduced at 4-7 K, the major product is 1; as the sample temperature is raised in situ, substantial proton delivery to generate 2 occurs by ∼55 K and above, 17 a process which is slowed in D 2 O buffer glass. 18,19 As with HO, perturbation of the proton- delivering network in P450cam by mutation D252N disrupts the ready proton transfer (eq 1), which occurs only at temperatures above ∼170 K in the mutant. 17 The behavior of the HO-1(D140X) and P450cam(D251N) mutants in fact is similar to that of the O 2 -carrying proteins, Hb and Mb. Cryoreduction of oxy-Mb and oxy-Hb at 77 K affords 1, and it is stable at this temperature for years; for completeness, we reduced oxy-Mb in glycerol/buffer at ∼4.2 K and confirmed that there is no proton transfer at this temperature or upon annealing to ∼77 K. The oxy-Mb (and oxy-Hb) intermediates 1 do not convert to 2 at temperatures less than 170 K; 20 by 200 K, the reaction, eq 1, is too fast to measure by progressive annealing with either H 2 O or D 2 O solvents, τ , 1 min. We determined the solvent kinetic isotope effect (solV-KIE) for eq 1 in oxy-Mb at 180 K, through measurements in H 2 O and D 2 O glycerol/buffer, 12 Figure 2. At this temperature the decay of 1 is roughly biphasic, as has been seen † Northwestern University. § Argonne National Laboratory. ‡ Tohoku University. Figure 1. g1-region X-band EPR spectra of oxy-HO cryoreduced in situ in EPR cavity at ∼4.2 K; spectra collected at ∼7 K. Rise with increasing field is due to intensity from the m ) 1 /2 H-atom line. Small features at ∼3000 G and above are from minority (<5%) oxy-HO substates. In these spectra, differential H/D broadening is not apparent. Conditions: microwave frequency, 9.502 GHz; modulation amplitude, 7.5 G. (Inset) Full 35 GHz spectrum of HO intermediate 2 (2 K). Published on Web 11/16/2004 15960 9 J. AM. CHEM. SOC. 2004, 126, 15960-15961 10.1021/ja044646t CCC: $27.50 © 2004 American Chemical Society