Enhanced Rates of C-H Bond Cleavage by a Hydrogen-Bonded Synthetic Heme High-Valent Iron(IV) Oxo Complex

Abstract

Secondary coordination sphere interactions are critical in facilitating the formation, stabilization, and enhanced reactivity of high-valent oxidants required for essential biochemical processes. Herein, we compare the C-H bond oxidizing capabilities of spectroscopically characterized synthetic heme iron(IV) oxo complexes, F₈Cmpd-II (F₈ = tetrakis(2,6-difluorophenyl)porphyrinate), and a 2,6-lutidinium triflate (LutH⁺) Lewis acid adduct involving ferryl O-atom hydrogen-bonding, F₈Cmpd-II(LutH⁺). Second-order rate constants utilizing C-H and C-D substrates were obtained by UV-vis spectroscopic monitoring, while products were characterized and quantified by EPR spectroscopy and gas chromatography (GC). With xanthene, F₈Cmpd-II(LutH⁺) reacts 40 times faster (k₂ = 14.2 M^-1 s^-1; -90 °C) than does F₈Cmpd-II, giving bixanthene plus xanthone and the heme product [F₈Fe^IIIOH₂]⁺. For substrates with greater C-H bond dissociation energies (BDEs) F₈Cmpd-II(LutH⁺) reacts with the second order rate constants k₂(9,10-dihydroanthracene; DHA) = 0.485 M^-1 s^-1 and k₂(fluorene) = 0.102 M^-1 s^-1 (-90 °C); by contrast, F₈Cmpd-II is unreactive toward these substrates. For xanthene vs xanthene-(d₂), large, nonclassical deuterium kinetic isotope effects are roughly estimated for both F₈Cmpd-II and F₈Cmpd-II(LutH⁺). The deuterated H-bonded analog, F₈Cmpd-II(LutD⁺), was also prepared; for the reaction with DHA, an inverse KIE (compared to F₈Cmpd-II(LutH⁺)) was observed. This work originates/inaugurates experimental investigation of the reactivity of authentic H-bonded heme-based Fe^IV═O compounds, critically establishing the importance of oxo H-bonding (or protonation) in heme complexes and enzyme active sites.

Figure 1.

Hydrogen-bonding interactions in (A) cytochrome c peroxidase (CcP) Cmpd-I wherein the iron(IV) oxo group is coupled to a Trp residue radical (PDB 4CVJ) and (B) ascorbate peroxidase (APX) protonated Cmpd-II (PDB 5JPR). It is of note in A the hydrogen-bonding between the oxo ligand with distal residues Trp51 and Arg48 and with interstitial water molecules. In B hydrogen-bonding interactions are seen between Trp51 and the oxo ligand as well as between the hydroxide proton and an interstitial water and Arg38. Also, there are proximal hydrogen-bonding interactions in CcP and APX involving the histidine axial ligands in CcP (His175 with Asp235 and Trp191) and in APX (His163 with Asp208 and Trp179). Crystal structures were obtained from the Protein Data Bank (PDB) and generated in PyMOL.

Figure 2.

UV–vis (A) and EPR (B) spectra monitoring the decay of F₈Cmpd-II(LutH⁺) upon addition of 50 or 20 equiv of xanthene, respectively, at −90 °C in 1:9 MeTHF:toluene and the subsequent formation of [F₈Fe^IIIOH₂]⁺ over time (blue to green).

Figure 3.

Reactivity between various C–H substrates (xanthene, dihydroanthracene (DHA), fluorene, and triphenylmethane) and F₈Cmpd-II or F₈Cmpd-II(LutH⁺). Also see the text for further details.

Figure 4.

Rates of reactivity as determined via UV–vis spectroscopy of 0.1 mM F₈Cmpd-II (red) or F₈Cmpd-II(LutH⁺) (blue) with various concentrations of xanthene (2–5 mM) at −90 °C in 1:9 MeTHF:toluene. Each data point represents three trials with error bars denoting the standard deviation between the trials. F₈Cmpd-II(LutH⁺) reacts 40 times faster than F₈Cmpd-II with xanthene under pseudo-first-order conditions. See also Figures S21 and S23.

Figure 5.

Linear correlation of the logarithmic rate constants and substrate C–H BDEs suggests a Bell–Evans–Polanyi relationship,^, supporting a rate-limiting hydrogen-atom abstraction step.

Figure 6.

Kinetic isotope effect (k_H/k_D) determined upon monitoring the rate of reactivity via UV–vis spectroscopy by F₈Cmpd-II(LutH⁺) toward various concentrations of DHA and DHA-d₄ under pseudo-first-order conditions. Each data point is an average of three trials, and error bars represent the standard deviation. Note: The scale of concentrations (the abscissa range) differs for plots A and B. Error bars in A, for 50 equiv of DHA, are smaller than the plot marker.

Figure 7.

(A) Representative scheme of tunneling (green arrow), resulting in a lower activation energy than needed to reach the transition state. (B) DFT-calculated structure of F₈Cmpd-II-(LutH⁺) showing the steric encumbrance by the lutidine molecule. (C) For HAA by iron(IV) oxo complexes, the substrate can approach at either a 180° or a 120° angle with respect to the Fe=O bond (as previously suggested by Shaik and co-workers); these approaches are consistent with the sp²-hybridized oxo ligand. However, in the present system with the lutidinium present, substrates can only approach at a 120° angle.

Figure 8.

Kinetic isotope effect (k_H/k_D) determined upon monitoring the rate of reactivity via UV–vis spectroscopy of F₈Cmpd-II(LutH⁺) and F₈Cmpd-II(LutD⁺) toward various concentrations of DHA under pseudo-first-order conditions. Each data point is an average of three trials, and error bars represent the standard deviation. F₈Cmpd-II(LutD⁺) reacted 1.5× faster, indicating both HAA and subsequent protonation are part of the rate-determining step.

Scheme 1.

Formation of F₈Cmpd-II and Protic Lewis Acid (LutH⁺-triflate) Adduct F₈Cmpd-II(LutH+) in 1:9 MeTHF:toluene at −90 °C

Scheme 2.

Reactivity of F₈Cmpd-II(LutH⁺) with Xanthene and Fluorene Yields Oxygenated Major Products (with xanthene also affording a coupled minor product), While DHA Is Oxidized to Anthracene^{a a}The final inorganic product in all cases is the ferric aqua complex, [F₈Fe^IIIOH₂]⁺.