Mimicking Elementary Reactions of Manganese Lipoxygenase Using Mn-hydroxo and Mn-alkylperoxo Complexes

Manganese lipoxygenase (MnLOX) is an enzyme that converts polyunsaturated fatty acids to alkyl hydroperoxides. In proposed mechanisms for this enzyme, the transfer of a hydrogen atom from a substrate C-H bond to an active-site MnIII-hydroxo center initiates substrate oxidation. In some proposed mechanisms, the active-site MnIII-hydroxo complex is regenerated by the reaction of a MnIII-alkylperoxo intermediate with water by a ligand substitution reaction. In a recent study, we described a pair of MnIII-hydroxo and MnIII-alkylperoxo complexes supported by the same amide-containing pentadentate ligand (6Medpaq). In this present work, we describe the reaction of the MnIII-hydroxo unit in C-H and O-H bond oxidation processes, thus mimicking one of the elementary reactions of the MnLOX enzyme. An analysis of kinetic data shows that the MnIII-hydroxo complex [MnIII(OH)(6Medpaq)]+ oxidizes TEMPOH (2,2′-6,6′-tetramethylpiperidine-1-ol) faster than the majority of previously reported MnIII-hydroxo complexes. Using a combination of cyclic voltammetry and electronic structure computations, we demonstrate that the weak MnIII-N(pyridine) bonds lead to a higher MnIII/II reduction potential, increasing the driving force for substrate oxidation reactions and accounting for the faster reaction rate. In addition, we demonstrate that the MnIII-alkylperoxo complex [MnIII(OOtBu)(6Medpaq)]+ reacts with water to obtain the corresponding MnIII-hydroxo species, thus mimicking the ligand substitution step proposed for MnLOX.


Introduction
Manganese lipoxygenase (MnLOX) is an enzyme that oxidizes C-H bonds of polyunsaturated fatty acids to generate alkyl hydroperoxide products [1][2][3][4]. The hydroperoxides are metabolized to oxylipins, such as leukotrienes and jasmonates, which act as inflammatory mediators and reproductive or growth regulators in plants [5,6]. MnLOXs are also found in fungi that are pathogenic to crops such as wheat (Gaeumannomyces graminis) [3] and rice (Magnaporthe oryzae) [7,8]. Because of the role of these pathogens in crop disease, MnLOXs have garnered interest as a target for pathogenesis [9]. An X-ray crystal structure of the MnLOX enzyme from the fungus Magnaporthe oryzae provides important information regarding the structure of the enzyme [10]. The active site consists of a mononuclear Mn center coordinated by three histidine ligands, a carboxylate from the C-terminus of the protein, a carbonyl donor from an asparagine group, and a solvent molecule. The crystals contained a Mn II center, and the solvent ligand was presumed to be water. It is commonly assumed that the Mn III oxidation state contains a hydroxo ligand [9]. Kinetic studies of MnLOX enzymes [4,8,9,11] and comparisons to Fedependent LOX enzymes [12,13] have led to the proposed mechanism in Scheme 1. The ratedetermining step for substrate oxidation involves the abstraction of a hydrogen atom from the fatty acid substrate by an active-site Mn III -hydroxo unit, yielding a carbon-centered radical and a Mn II -aqua species. In MnLOX from M. oryzae, the k cat parameter shows a large substrate C-H/C-D kinetic isotope effect of 60-80 [9], consistent with dominant hydrogen-atom tunneling in the rate-determining step [14]. The carbon-based substrate radical can rearrange, after which it is trapped by O 2 to generate an oxygen-centered radical. This radical can abstract a hydrogen-atom from the Mn II -aqua complex to yield product and regenerate the Mn III -hydroxo center (Path A in Scheme 1). Alternatively, the oxygen-centered radical can displace the aqua ligand and oxidize the Mn II center to yield a Mn III -alkylperoxo complex (Path B in Scheme 1). Support for this alternate pathway is provided by an X-ray crystal structure of an Fe-alkylperoxo complex from soybean lipoxygenase [15]. In this path, the Mn III -alkylperoxo complex reacts with water via a ligand substitution reaction to yield the substrate and the Mn III -hydroxo center. The proposed mechanism for MnLOX has inspired attempts to model the ratedetermining hydrogen-atom abstraction step, where a C-H bond transfers a hydrogenatom to the Mn III -hydroxo center. Stack and co-workers were the first to report a Mn IIIhydroxo center capable of oxidizing a C-H bond [16]. Using the neutral, pentadentate PY5 ligand, the Mn III -hydroxo complex [Mn III (PY5)(OH)] 2+ (PY5 = 2,6-bis(bis(2-pyridyl)methoxymethane)pyridine) is able to oxidize several hydrocarbons, including xanthene [16]. The bis-benzylic C-H bonds in xanthene provide a reasonable mimic of the bis-allylic C-H bonds in the native substrate of MnLOX. More recently, our lab reported xanthene oxidation by a pair of Mn III -hydroxo complexes supported by anionic, pentadentate ligands, [Mn III (OH)(dpaq)] + and [Mn III (OH)(dpaq 2Me )] + (dpaq = 2-[bis(pyridin-2ylmethyl)]amino-N-quinolin-8-yl-acetamidate) [17,18]. An analysis of kinetic data for these complexes revealed that the rate of xanthene oxidation by [Mn III (PY5)(OH)] 2+ is roughly 10 and 30 fold faster than that of [Mn III (OH)(dpaq)] + and [Mn III (OH)(dpaq 2Me )] + , respectively [16,[18][19][20].
While there are now several studies mimicking the C-H bond oxidation step of Mn-LOX, there are far fewer examples investigating the potential ligand substitution reaction of the Mn III -alkylperoxo intermediate. Kovacs and co-workers have described a family of Mn III -alkylperoxo complexes supported by anionic, pentadentate N 4 S − and N 4 O − ligands [26][27][28]. These complexes decay thermally by O-O homolysis, which is distinct from the chemistry proposed for the Mn III -alkylperoxo complex of MnLOX (Scheme 1). Our lab reported Mn III -alkylperoxo complexes supported by the dpaq and dpaq 2Me ligands; however, the instability of these complexes, and the requirement of a large excess of t BuOOH to achieve their formation, made studies of reactivity unfeasible [29]. More recently, we reported that a derivative of the dpaq ligand, with 6-Me-pyridyl groups, is able to support Mn III -alkylperoxo complexes that are stable at room temperature and can be formed by adding an equivalent of an alkylhydroperoxide to the corresponding Mn III -hydroxo complex [Mn III (OH)( 6Me dpaq)] + (Scheme 2) [30]. The observation that a set of thermally stable Mn III -hydroxo and Mn III -alkylperoxo complexes could be generated using the same supporting ligand presents a unique opportunity to model multiple steps in the proposed mechanism of MnLOX. In this present study, we describe the reactivity of [Mn III (OH)( 6Me dpaq)] + towards TEMPOH, thereby mimicking the initial substrate oxidation step in MnLOX. To understand the reason behind rate variations among the Mn III -hydroxo complexes, we explore the spectroscopic and thermodynamic properties of [Mn III (OH)( 6Me dpaq)] + using the density functional theory (DFT) computations. We also examine the reaction of the Mn III -alkylperoxo complex [Mn III (OO t Bu)( 6Me dpaq)] + with water. This reaction yields the Mn III -hydroxo complex [Mn III (OH)( 6Me dpaq)] + , thereby mimicking a potential step for substrate release by MnLOX.

Materials and Methods
All chemicals obtained from commercial sources were of ACS grade or higher and were used as obtained, unless otherwise noted. Acetonitrile, diethyl ether, and methanol were dried and degassed using a PureSolv Micro solvent purification system. The H 6Me dpaq ligand and the [Mn II (OH 2 )( 6Me dpaq)](OTf), [Mn III (OH)( 6Me dpaq)](OTf), and [Mn III (OO t Bu) ( 6Me dpaq)](OTf) complexes were synthesized according to a previously reported procedure (OTf − = trifluoromethanesulfonate) [30]. All synthetic experiments were performed under dinitrogen atmosphere in a glovebox unless otherwise noted. Electronic absorption experiments were performed using either a Varian Cary 50 Bio UV-visible spectrophotometer (Agilent Technologies, Santa Clara, CA, USA), equipped with a Unisoku cryostat and stirrer (for low temperature experiments) or a Quantum Northwest temperature controller equipped with a stirrer (for high temperature experiments). Electrospray ionization mass spectrometry (ESI-MS) experiments were performed using an LCT Premier MicroMass electrospray time-of-flight instrument (Waters, Milford, MA, USA).

Kinetic Studies of TEMPOH and Xanthene Oxidation by [Mn III (OH)( 6Me dpaq)](OTf)
A 1.25 mM solution of [Mn III (OH)( 6Me dpaq)] + was prepared in 2.0 mL of MeCN in a nitrogen-filled glovebox and transferred to a quartz cuvette that was sealed with a rubber septum. The cuvette was removed from the glovebox and allowed to equilibrate at -35 • C for 10 min on the UV-vis spectrometer. A 100 µL solution of TEMPOH, with concentrations ranging from 10-60 equiv. relative to [Mn III (OH)( 6Me dpaq)] + , was added to the cuvette using a gastight syringe that had been purged with N 2 gas. The addition of TEMPOH led to the disappearance of the 510 nm electronic absorption feature of [Mn III (OH)( 6Me dpaq)] + . The change in absorbance as a function of time was fit to obtain a pseudo-first-order rate constant (k obs ). The reported k obs represent an average from three separate measurements. A linear fit of the plot of k obs vs. the concentration of TEMPOH provided the second-order rate constant (k 2 ).
The reactivity of [Mn III (OH)( 6Me dpaq)] + with xanthene was investigated using a similar approach. In this case, 250 equiv. xanthene (relative to Mn) was prepared anaerobically in 300 µL dichloromethane in a 400 mL vial. A solution of xanthene was added to 2 mL of a 1.25 mL solution of [Mn III (OH)( 6Me dpaq)] + that had equilibrated at 50 • C for 10 min on the spectrometer. The decay of the 510 nm feature of the [Mn III (OH)( 6Me dpaq)] + was monitored by electronic absorption spectroscopy over a period of 1000 min.

Reaction of [Mn III (OO t Bu)( 6Me dpaq)] + with Protic Solvents and Kinetic Investigations with Water
The propensity for [Mn III (OO t Bu)( 6Me dpaq)] + to ligand substitution reactions was first discovered in an attempt to prepare the complex in various protic solvents. In a representative procedure, [Mn III (OO t Bu)( 6Me dpaq)] + was prepared in MeCN and then dried in vacuo to remove the solvent, leaving behind an oily film. The oily film was taken into the glovebox and dissolved in 2,2,2-trifluoroethanol (TFE). The solution was transferred to a quartz cuvette, sealed with a rubber septum, and wrapped with Parafilm. The sample was then removed from the glovebox, inserted in the UV-vis spectrometer, and heated to 50 • C. The initial absorption spectra collected under these conditions revealed an electronic absorption band of [Mn III (OO t Bu)( 6Me dpaq)] + at 650 nm. This band decayed over the course of 60 min, with the concomitant formation of a band at 510 nm ( Figure S1). An ESI-MS analysis of the product solution revealed a prominent ion peak at 564.07 m/z, consistent with [Mn III (OCH 2 CF 3 )( 6Me dpaq)] + (m/z = 564.14; see Figure S2). Thus, while [Mn III (OO t Bu)( 6Me dpaq)] + initially forms through the dissolution of the [Mn III (OO t Bu)( 6Me dpaq)](OTf) salt in TFE, a ligand substitution reaction occurs that replaces the alkylperoxo ligand with a CF 3 CH 2 Oligand. This ligand substitution reaction with TFE was only observed at 50 • C. A similar, albeit far more rapid, ligand substitution reaction occurs via the dissolution of [Mn III (OO t Bu)( 6Me dpaq)](OTf) in MeOH at 25 • C ( Figure S3). Moreover, the addition of 100 µL MeOH to a 1.0 mM solution of [Mn III (OO t Bu)( 6Me dpaq)](OTf) in MeCN leads to its conversion to the corresponding Mn III -methoxy complex ( Figures S3 and S4).
To determine rate constants for ligand substitution reactions of [Mn III (OO t Bu)( 6Me dpaq)] + with water, we prepared an acetonitrile solution of [Mn III (OO t Bu)( 6Me dpaq)] + in a glovebox, transferred the solution to a quartz cuvette sealed with a pierceable rubber septum and wrapped with Parafilm. The cuvette was removed from the glovebox and an aliquot of degassed H 2 O (100-400 equiv. relative to [Mn III (OO t Bu)( 6Me dpaq)] + ) was injected while monitoring the reaction by electronic absorption spectroscopy at 25 • C.

Cyclic Voltammetry
Cyclic voltammograms were recorded using a Basi ® PalmSens EmStat3+ potentiostat (PalmSens BV, Houten, Utrecht, The Netherlands). The working electrode was a glassy carbon electrode with a Pt wire as the counter electrode. A 0.01 M AgCl solution was prepared using a 0.1 M Bu 4 NPF 6 electrolyte solution in CH 3 CN. The 0.01 M AgCl solution was used for the Ag/AgCl quasi-reference electrode. The Fc + /Fc potential was measured as an external reference. Additionally, 2 mM solutions of [Mn III (OH)( 6Me dpaq)](OTf) were prepared from 10 mL of a degassed 0.1 M Bu 4 NPF 6 electrolyte solution in CH 3 CN. These sample solutions were sparged with nitrogen gas with the aid of Teflon tubing for 15 min before measurement. The Teflon tubing was placed above the surface of the solution during measurement to continue flushing the headspace, without disturbing the solution in the electrochemical cell. All measurements were performed at room temperature. Data were referenced to the cathodic peak potential for Fc + /Fc in MeCN.

Electronic Structure Calculations
All DFT calculations were performed using ORCA 4.2.1. [31]. For the geometry optimizations the B3LYP [32,33] functional with the def2-TZVP basis set for Mn, N and O atoms were used, while the def2-SVP basis set was used for C and H atoms [34,35]. Grimme's D3 dispersion correction [36][37][38][39] was also applied with a fine integration grid (Grid6 and GridX6 in ORCA). Analytical frequency calculations were performed using the same level of theory. The zero-point energies, thermal corrections, and entropies were obtained from the analytical frequency calculations. Single point energies were obtained for all structures using the same B3LYP-D3 functional but with the larger def2-TZVPP basis set on all atoms and a finer integration grid (Grid7 and GridX7). In all cases, solvation was accounted for by using the SMD solvation model with default parameters for acetonitrile [40]. The RIJCOSX approximation, together with def2/J auxiliary basis set, was used for all calculations [41,42]. A detailed discussion of the calculation of thermodynamic parameters is included in the Supplementary Materials.

Formation of [Mn III (OH)( 6Me dpaq)] + by Aerobic Oxidation
In a previous study, we reported that the Mn III -hydroxo complex [Mn III (OH)( 6Me dpaq)] + could be generated via the oxidation of the Mn II -aqua complex [Mn II (H 2 O)( 6Me dpaq)](OTf) using 0.5 equiv. iodosobenzene (PhIO) [30]. Because a number of Mn III -hydroxo complexes can be generated by aerobic oxidation of their Mn II analogues [17,18,20,24,[43][44][45] Figure 1). The exposure of a MeCN solution of this complex to O 2 results in the eventual growth of a more intense band near 510 nm that is attributed to [Mn III (OH)( 6Me dpaq)] + [30]. This reaction is very slow, with approximately 75% conversion to the Mn III -hydroxo complex achieved after 48 h (Figure 1). In contrast, the [Mn II (dpaq)](OTf) complex and several derivatives show complete oxidation to Mn III products within 0.5-5 h [17,18,20]. The sluggish nature of the reaction of [Mn II (H 2 O)( 6Me dpaq)] + with O 2 may be caused by a more electron-deficient Mn II center. The 6-Me-pyridyl groups in [Mn II (H 2 O)( 6Me dpaq)] + give rise to Mn-N pyridine bonds that are 0.02-0.04 Å longer than [Mn II (dpaq)](OTf) and its derivatives [17,18]. These longer bonds presumably mitigate electron donation from the pyridyl ligands to the Mn II center, leading to a more electron-poor metal center. Similarly, the [Mn II (dpaq 5NO2 )](OTf) complex, which contains a strongly electron-donating nitro group on the quinolinyl moiety, showed essentially no reactivity with dioxygen [20].

Spectroscopic Properties and Electronic Structure of [Mn III (OH)( 6Me dpaq)] +
The electronic absorption spectrum of [Mn III (OH)( 6Me dpaq)] + in MeCN at 25 • C shows a single broad absorption feature from 800 to 470 nm with λ max at 510 nm (ε = 250 M −1 cm −1 ) ( Figure 2). This spectrum deviates significantly from that observed for [Mn III (OH) (dpaq)] + and its derivatives. The electronic absorption spectra of those complexes showed two absorption maxima with λ max values at 770 and 500 nm ( Figure 2) [17,18,20,46]. To understand the origin of the spectral perturbations for [Mn III (OH)( 6Me dpaq)] + , we predicted the electronic absorption spectrum of this complex using time-dependent density functional theory (TD-DFT) calculations. Although TD-DFT calculations have known drawbacks, this method has performed exceptionally well for mononuclear Mn III complexes [47,48], potentially because ligand-field transitions dominate the electronic absorption spectra of these complexes.
DFT geometry optimization for [Mn III (OH)( 6Me dpaq)] + and [Mn III (OH)(dpaq)] + reproduced the trends in metric parameters obtained from X-ray diffraction (XRD) experiments for the complexes ( Table 1) With this ground configuration, each spin-allowed ligand-field transition can be reasonably approximated by a one-electron excitation from a singly occupied Mn III d-based MO to the unoccupied d z 2 -based MO. Consequently, the ligand-field electronic transition energies can be directly related to the Mn III d-orbital splitting pattern. We will therefore briefly discuss the compositions and energies of the d-based MOs of [Mn III (OH)( 6Me dpaq)] + and [Mn III (OH)(dpaq)] + , as differences in these orbitals can account for all perturbations in the electronic absorption spectra of these complexes.  In each complex, the highest-energy d z 2 MO is strongly destabilized by σ-antibonding interactions with both the hydroxo and carboxamido donors (Figure 3). The d z 2 MO in each complex receives ligand contributions from the 2p x,y orbitals of the equatorial N donor atoms, 2p z orbital of the carboxamido nitrogen atom and 2p z orbital of the oxygen atom of the hydroxo ligand (Table S1). The DFT calculations predict nearly identical energies (both roughly −2.2 eV; see Figure 3) for the d z 2 MOs of [Mn III (OH)( 6Me dpaq)] + and [Mn III (OH)(dpaq)] + . Thus, any differences in the spectroscopic or chemical properties of these complexes are not found to be related to differences in Mn III -hydroxo or Mn III -amide bonding interactions. This DFT-based prediction is in accordance with the nearly identical Mn III −O hydroxo (Mn−O1) and Mn III −N amide (Mn−N2) bond lengths observed in both the experimental and DFT-computed structures of these complexes ( Table 1) Table 1). These longer, and therefore weaker Mn III −N pyridine bonds lead to a stabilization of the d x 2 −y 2 MO in [Mn III (OH)( 6Me dpaq)] + by 0.3 eV relative to that of [Mn III (OH)(dpaq)] + (Figure 3). This stabilization creates a larger gap between the d z 2 and d x 2 −y 2 MOs of [Mn III (OH)( 6Me dpaq)] + . The d xz and d yz MOs of each complex have weak π-antibonding interactions with the hydroxo ligand. The nearly identical Mn-hydroxo distances for the [Mn III (OH)( 6Me dpaq)] + and [Mn III (OH)(dpaq)] + give rise to similar Mn-hydroxo π-interactions, causing the d xz and d yz MOs of these complexes to lie at similar energies. The d xy MO of each complex is non-bonding and, for this reason, the energy of this MO shows little variation between these complexes. The TD-DFT-computed electronic absorption spectra of both [Mn III (OH)( 6Me dpaq)] + and [Mn III (OH)(dpaq)] + are in excellent agreement with their experimental counterparts ( Figure 4). Starting with [Mn III (OH)(dpaq)] + , the calculated spectrum shows two bands at 770 and 500 nm, which nearly perfectly reproduce the experimental spectrum (Figure 4, left). An analysis of the electron-density difference maps (EDDMs), illustrating the factors contributing to these bands, shows that each band derives from Mn III ligand-field transitions. The lower-energy band at 770 nm is a result of a one-electron d x 2 −y 2 → d z 2 transition. The band at 500 nm results from two ligand-field transitions-a d yz → d z 2 transition at 505 nm and a d xz → d z 2 transition near 486 nm. Given these assignments, the position of the lower-energy band reflects the difference between Mn-ligand σ-interactions with the axial and equatorial ligands, while the energy of the higher-energy band elucidates the differences between Mn-hydroxo σ-and π-interactions. The TD-DFT-computed electronic absorption spectrum of [Mn III (OH)( 6Me dpaq)] + also reproduces the experimental spectrum well (Figure 4). The computed spectrum shows a weak feature near 630 nm that corresponds with a distinct rise of the absorption intensity in the experimental spectrum. The calculated spectrum also shows a greater intensity at wavelengths less than 550 nm, in good agreement with experiment ( Figure 4, right). Importantly, the TD-DFT-computed spectrum of [Mn III (OH)( 6Me dpaq)] + reproduces all the major differences and similarities relative to [Mn III (OH)(dpaq)] + ; i.e., the lowest-energy band has a pronounced blue-shift, while the higher-energy bands are relatively unperturbed. The lowest-energy band of [Mn III (OH)( 6Me dpaq)] + at 630 nm indicates a one-electron d x  (Figure 3 and Table 1). The higher-energy band in the TD-DFT-computed spectrum of [Mn III (OH)( 6Me dpaq)] + contains the d yz → d z 2 and d xz → d z 2 transitions. The wavelengths of these transitions are only slightly blue-shifted relative to the corresponding transitions in [Mn III (OH)(dpaq)] + (505 vs. 499 nm, and 486 vs. 469 nm, respectively; see Figure 4), which is in agreement with the minor perturbations in the energies of these MOs (Figure 3). Overall, the TD-DFT computations reveal that the spectral perturbations between [Mn III (OH)(dpaq)] + and [Mn III (OH)( 6Me dpaq)] + can be understood on the basis of the elongated Mn III −N pyridine bonds in the latter complex.

Oxidative Reactivity of [Mn III (OH)( 6Me dpaq)](OTf)
To evaluate the effect of the Mn III −N pyridine bond elongations of [Mn III (OH)( 6Me dpaq)] + on chemical reactivity, we explored the reactions of this complex with TEMPOH and xan-thene. The addition of 10 equiv. TEMPOH to a 1.25 mM solution of [Mn III (OH)( 6Me dpaq)] + in MeCN at −35 • C led to the disappearance of the electronic absorption features of the Mn III -hydroxo complex within 100 s, providing a final spectrum identical to that of [Mn II (H 2 O)( 6Me dpaq)] + with approximately 100% yield ( Figure 5). An EPR analysis of the solution, conducted following the reaction, reveals an intense signal from TEMPO radical ( Figure S5). The observed products are consistent with a CPET reaction between the Mn III -hydroxo center and TEMPOH, which would afford TEMPO radical and [Mn II (H 2 O)( 6Me dpaq)] + . Kinetic experiments were performed at −35 • C using 10-60 equiv. TEMPOH to obtain the second-order rate constant (k 2 ). A plot of k obs vs. the concentration of TEMPOH is linear, and a fit to these data provided the second-order rate constant (k 2 ) of 3.4(2) M −1 s −1 at −35 • C ( Figure 6).    [17,20]. Thus, Me substituents on the pyridyl or quinolinyl groups have similar effects on reactivity. We also investigated the reactivity of [Mn III (OH)( 6Me dpaq)] + with xanthene by treating a 1.25 mM solution of this Mn III -hydroxo complex in MeCN with 250 equiv. xanthene at 50 • C. In this case, the absorbance band of [Mn III (OH)( 6Me dpaq)] + at 510 nm decayed by only 10% over the course of 1000 min, which is within the range of the self-decay rate for [Mn III (OH)( 6Me dpaq)] + at this temperature. The apparent lack of reactivity is somewhat unexpected, as [Mn III (OH)(dpaq)] + reacts with xanthene under similar conditions, albeit at a slow rate (0.0008 s −1 ). It is possible that the decreased reactivity for [Mn III (OH)( 6Me dpaq)] + might be due to the steric bulk of the 6-methyl-pyridyl substituents, which could hinder the access of xanthene to the Mn III -hydroxo unit. Notably, [Mn III (OH)(dpaq 2Me )] + reacts with xanthene three times slower than [Mn III (OH)(dpaq)] + (0.00025 s −1 vs. 0.0008 s −1 , respectively). 17 DFT computations demonstrated that the steric bulk of the 2-Me-quinoline moiety in [Mn III (OH)(dpaq 2Me )] + caused xanthene to orientate differently in the transition state when compared to [Mn III (OH)(dpaq)] + . This reorientation introduced a destabilizing effect that led to a higher transition state of [Mn III (OH)(dpaq 2Me )] + than that of [Mn III (OH)(dpaq)] + by around 3 kcal/mol. A similar situation, caused by the 6-Mepyridyl groups of [Mn III (OH)( 6Me dpaq)] + , could hamper the reaction of this complex with xanthene.

Thermodynamic Driving Force for TEMPOH Oxidation Using Experimental and Computational Methods
CPET reaction rates of Mn III -hydroxo complexes show a strong correlation to the thermodynamic driving force [20]. For these reactions, the driving force is the difference between the BDFE of the Mn II O(H)-H bond that is formed, and the TEMPO-H bond that is broken (Equation (1)). In this equation, C G,sol is a constant for a given solvent (C G,sol, MeCN = 54.9 kcal/mol). In our previous investigation of [Mn III (OH)(dpaq 5R )] + complexes, we observed that both the reduction potential and pK a changed as a function of the R substituent, although the potential changed more dramatically and was, therefore, largely responsible for the net change in O−H BDFE [20].
We performed cyclic voltammetry (CV) experiments of [Mn III (OH)( 6Me dpaq)] + to understand the thermodynamic effects on the CPET reactivity of the complex . When scanning to negative potentials, we observed a reduction wave (E p,c ) at −0.63 V Fc + /Fc ( Figure S6) that we attributed to a reduction of the Mn III -hydroxo complex. This potential is between those of [Mn III (OH)(dpaq 2Me )] + and [Mn III (OH)(dpaq 5Cl )] + (−0.62 V and −0.66 V vs. Fc + /Fc, respectively) [20]. The E p,c value for [Mn III (OH)( 6Me dpaq)] + and those previously measured for [Mn III (OH)(dpaq)] + and its derivatives are presented in Table 2.
The rate of TEMPOH oxidation by [Mn III (OH)( 6Me dpaq)] + can be plotted against the O−H BDFE of the [Mn II (OH 2 )( 6Me dpaq)] + product ( Figure 7). As shown in this plot, the [Mn III (OH)( 6Me dpaq)] + complex follows the linear correlation previously observed for Mn III -hydroxo complexes supported by the dpaq ligand and its derivatives quite well [20]. The individual contributions of the pK a and E 1/2 to the Mn II -aqua O−H BDFEs for these complexes are presented in Figure S7, and the correlations of these parameters with the reaction rates of the Mn III -hydroxo complexes with TEMPOH are shown in Figure 8. Among this series, the more significant change to BDFE originates from the E 1/2 factors, with only a slight effect enacted by the pK a ( Table 2 and Figure 8). The pK a change across the series is 1.7 pK a units (2.33 kcal/mol), whereas the calculated E 1/2 changes by 0.22 V (5.07 kcal/mol). Thus, the E 1/2 contributes around twice as much as the pK a to the net BDFE across the series.  From Figure 8, we observe a strong linear correlation between the E 1/2 of the Mn III/II couple and the ln(k 2 ) for TEMPOH oxidation, with more positive potentials translating to faster reaction rates. In contrast, the plot of ln(k 2 ) for TEMPOH oxidation versus the Mn IIaqua pK a value shows an inverse correlation, where the more basic complexes experience slower reaction rates. Thus, the pK a and E 1/2 counterbalance each other to affect the rate of reaction of the O−H BDFE.

Ligand Substitution Reactions of [Mn III (OO t Bu)( 6Me dpaq)] + with Water
While the reaction of TEMPOH with [Mn III (OH)( 6Me dpaq)] + mimics one of the elementary reactions proposed in the catalytic cycle of MnLOX, one of the proposed mechanisms for this enzyme postulates a ligand substitution reaction, where a Mn III -alkylperoxo complex reacts with water to give the alkyl hydroperoxo product and a Mn III -hydroxo complex [50]. Since both [Mn III (OH)( 6Me dpaq)] + and its Mn III -alkylperoxo analogue [Mn III (OO t Bu)( 6Me dpaq)] + are well characterized and stable at room temperature, we took advantage of these complexes to explore whether we could mimic this proposed ligand substitution reaction. The addition of 100 equiv. H 2 O to [Mn III (OO t Bu)( 6Me dpaq)] + in an anerobic solution of MeCN resulted in a loss of intensity of the electronic absorption band of the Mn III -alkylperoxo complex at 650 nm and growth in absorbance at~500 nm ( Figure 9). This conversion demonstrates isosbestic behavior (with an isosbestic point at 574 nm), suggesting a simple conversion that does not involve accumulating intermediates. The final spectrum is identical to that of [Mn III (OH)( 6Me dpaq)] + , consistent with a ligand substitution reaction. The decay of the optical signal of [Mn III (OO t Bu)( 6Me dpaq)] + in the presence of H 2 O (650 nm) follows pseudo-first order behavior (Figure 9, inset), allowing us to determine a pseudo-first order rate constant (k obs ). Experiments using different concentrations of H 2 O showed a linear increase in k obs with increasing water concentrations. An analysis of these data yield a second order rate constant for the ligand substitution reaction of 1.13(8) × 10 −3 M −1 s −1 at 25 • C ( Figure 10). A mass spectral analysis of the solution following the reaction of [Mn III (OO t Bu)( 6Me dpaq)] + with H 2 O revealed a peak at m/z = 482.14 ( Figure S8), further confirming the formation of [Mn III (OH)( 6Me dpaq)] + (calculated m/z = 482.14). At longer time periods, we observed the precipitation of a brown solid, which may provide evidence of a degree of demetallation of the [Mn III (OH)( 6Me dpaq)] + due to the presence of water.

Conclusions
In this study, we examined the spectroscopic properties and chemical reactivity of the Mn III -hydroxo complex [Mn III (OH)( 6Me dpaq)] + . Using TD-DFT computations, we were able to explain differences in the electronic absorption spectra of [Mn III (OH)( 6Me dpaq)] + and [Mn III (OH)(dpaq)] + in terms of the stabilization of the Mn d x 2 -y 2 orbital in the former complex. Orbital stabilization occurs due to longer Mn III −N pyridine bonds, caused by the 6-Me-pyridyl groups in [Mn III (OH)( 6Me dpaq)] + . Importantly, DFT computations predict no difference in the energy of the redox-active d z 2 MO between the [Mn III (OH)( 6Me dpaq)] + and [Mn III (OH)(dpaq)] + complexes. Thus, any differences in redox reactivity between these complexes must stem from factors beyond the energy of the d z 2 MO. The oxidative capability of the [Mn III (OH)( 6Me dpaq)] + complex was assessed by exploring the rate of TEMPOH oxidation. The reactivity of [Mn III (OH)( 6Me dpaq)] + with TEMPOH is 3-fold faster than that of [Mn III (OH)(dpaq)] + . This rate enhancement results from the greater Mn III/II reduction potential of [Mn III (OH)( 6Me dpaq)] + than [Mn III (OH)(dpaq)] + . The TEMPOH oxidation rate for [Mn III (OH)( 6Me dpaq)] + is in accordance with a previously established linear free energy relationship for Mn III -hydroxo complexes [20]. The related [Mn III (OO t Bu)( 6Me dpaq)] + complex reacts with water to form the Mn III -hydroxo complex [Mn III (OH)( 6Me dpaq)] + , providing a mimic of an elementary step proposed in some mechanisms for MnLOX.

Supplementary Materials:
The following are available online, Figure S1: Electronic absorption spectra showing the reaction of [Mn III (OO t Bu)( 6Me dpaq)] + in TFE at 50 • C to form the [Mn III (OCH 2 CF 3 ) ( 6Me dpaq)] + complex (blue tace). Figure S2 Figure S4: ESI-MS data for [Mn III (OO t Bu)( 6Me dpaq)](OTf) in MeOH. Figure S5: EPR spectrum of the solution following reaction of [Mn III (OH)( 6Me dpaq)] + and TEMPOH. Figure S6: Cyclic voltammetry trace of [Mn III (OH)( 6Me dpaq)] + . Figure S7: Thermodynamic contributions to the O−H BDFE of Mn II -aqua complexes from the Mn III -OH/Mn II -OH reduction potentials and Mn II -aqua pK a values. Figure S8: ESI-MS data for [Mn III (OO t Bu)( 6Me dpaq)] + with H 2 O. Table S1: TD-DFT calculated energies, percent contributions from dominant one-electron excitations, and oscillator strengths for the major electronic transitions in [Mn III (OH)(dpaq)] + and [Mn III (OH)( 6Me dpaq)] + .  Table S3: DFT-Calculated ∆G values (in kcal/mol) for the addition of a pro-ton or electron, used in calculating the E red and pK a values. Table S4: DFT-calculated thermodynamic parameters used to determine Mn II -OH 2 BDFEs for the [Mn II (OH 2 )( 6Me dpaq)] + complex. Tables  S5-S9 Funding: This work was supported by the U.S. National Science Foundation (CHE-1900384 to T.A.J). The U.S. NSF is also acknowledged for funds to purchase the EPR spectrometer (CHE-0946883).