This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes. Article Cite This: ACS Omega 2019, 4, 10729−10740 http://pubs.acs.org/journal/acsodf Oxidative Cleavage of Cellobiose by Lytic Polysaccharide Monooxygenase (LPMO)-Inspired Copper Complexes Andrea. C. Neira,† Paulina R. Martínez-Alanis,§ Gabriel Aulloń ,§ Marcos Flores-Alamo,‡ Paulino Zeroń ,‡ Anna Company,∥ Juan Chen,⊥ Johann B. Kasper,⊥ Wesley R. Browne,⊥ Ebbe Nordlander,# and Ivan Castillo*,† † Instituto de Química and ‡Facultad de Química, División de Estudios de Posgrado, Universidad Nacional Autónoma de México, Circuito Exterior, CU, 04510 Ciudad de México, México § Departament de Química Inorgànica i Orgànica and Institut de Química Teòrica i Computacional, Universitat de Barcelona, Martí i Franquès 1-11, 08028 Barcelona, Spain ∥ Institut de Química Computacional i Catàlisi (IQCC), Departament de Química, Universitat de Girona, C/ M. Aurèlia Capmany 69, 17003 Girona, Catalonia, Spain ⊥ Molecular Inorganic Chemistry, Stratingh Institute for Chemistry, Faculty of Science and Health, University of Groningen, Nijenborgh 4, 9747AG Groningen, The Netherlands # Chemical Physics, Department of Chemistry, Lund University, Box 124, SE-221 00 Lund, Sweden * S Supporting Information ABSTRACT: The potentially tridentate ligand bis[(1-methyl- 2-benzimidazolyl)ethyl]amine (2BB) was employed to prepare copper complexes [(2BB)CuI]OTf and [(2BB)- CuII(H2O)2](OTf)2 as bioinspired models of lytic poly- saccharide copper-dependent monooxygenase (LPMO) en- zymes. Solid-state characterization of [(2BB)CuI]OTf re- vealed a Cu(I) center with a T-shaped coordination environment and metric parameters in the range of those observed in reduced LPMOs. Solution characterization of [(2BB)Cu I I (H 2 O) 2 ](OTf) 2 indicates that [(2BB)- CuII(H2O)2]2+ is the main species from pH 4 to 7.5; above pH 7.5, the hydroxo-bridged species [{(2BB)CuII(H2O)x}2(μ- OH)2]2+ is also present, on the basis of cyclic voltammetry and mass spectrometry. These observations imply that deprotonation of the central amine of Cu(II)-coordinated 2BB is precluded, and by extension, amine deprotonation in the histidine brace of LPMOs appears unlikely at neutral pH. The complexes [(2BB)CuI]OTf and [(2BB)CuII(H2O)2](OTf)2 act as precursors for the oxidative degradation of cellobiose as a cellulose model substrate. Spectroscopic and reactivity studies indicate that a dicopper(II) side-on peroxide complex generated from [(2BB)CuI]OTf/O2 or [(2BB)CuII(H2O)2](OTf)2/H2O2/NEt3 oxidizes cellobiose both in acetonitrile and aqueous phosphate buffer solutions, as evidenced from product analysis by high-performance liquid chromatography-mass spectrometry. The mixture of [(2BB)CuII(H2O)2](OTf)2/H2O2/NEt3 results in more extensive cellobiose degradation. Likewise, the use of both [(2BB)CuI]OTf and [(2BB)CuII(H2O)2](OTf)2 with KO2 afforded cellobiose oxidation products. In all cases, a common Cu(II) complex formulated as [(2BB)CuII(OH)(H2O)]+ was detected by mass spectrometry as the final form of the complex. ■ INTRODUCTION The recently reported lytic polysaccharide monooxygenases primary amine of the histidine brace,4 as well as the potential involvement of dioxygen5 or hydrogen peroxide6 as oxidants. (LPMOs),1 also known as the auxiliary activity (AA) family of However, details of the mechanism, potential intermediates in copper-dependent enzymes,2 feature a conserved type 2 copper the key C−H activation step, and subsequent polysaccharide binding site with a N3 donor set defined by two imidazoles and hydroxylation remain to be determined. Although cupric− a backbone nitrogen from one of the histidine residues, in a superoxo and oxyl complexes have been suggested as the structural motif described as the “histidine brace” (Chart 1).2a,3 reactive intermediates, a mechanistic insight may be obtained These enzymes oxidatively cleave the polymeric chains of from synthetic models. In this regard, few biomimetic LPMO- recalcitrant polysaccharides to assist hydrolytic enzymes and have potential applications in biomass conversion to renewable Received: March 21, 2019 biofuels. Recent studies have established some of the aspects Accepted: June 6, 2019 that govern substrate binding, such as H-bonding to the Published: June 20, 2019 © 2019 American Chemical Society 10729 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article Chart 1. (a) Schematic Representation of the Active Site of subsequent reactivity studies were hampered by the low LPMO, (b) 2BB, and (c) Mercury Diagram of the Dication solubility of the tetrafluoroborate or hexafluorophosphate [(2BB)CuII(H2O)2]2+15 complexes in dichloromethane.13,14 However, the use of trifluoromethanesulfonate anions (triflate, OTf) resulted in species soluble in acetonitrile amenable for solution character- ization and reactivity studies. Electrospray ionization mass spectrometry (ESI-MS) revealed the presence of a peak at m/z 396 assigned to the monomeric [(2BB)CuI]+ [Figure S1 in the Supporting Information]. Further characterization by 1H NMR and Fourier transform infrared (FTIR) spectroscopy con- firmed the identity of [(2BB)CuI]OTf; its 1H NMR spectrum in CD3CN shows relatively broad aromatic signals for the benzimidazole groups at δ 7.64, 7.24, and 7.06 ppm, the resonance for the N-methyl groups was observed at δ 3.42, whereas the methylene protons gave rise to resonances at δ 3.21 (−CH2−N) and 3.01 ppm (−CH2−C). FTIR spectros- copy displayed notable bands at 3273 (w) cm−1 assigned to the central amine N−H stretch, 1027 and 1257 (s) assigned to S− O stretching bands of the triflate anion, as well as a pair of strong bands at 1147 and 1227 cm−1 assigned to C−F inspired copper complexes have been reported to date,7 stretching vibrations; a band at 1614 cm−1 was assigned to the despite the numerous reports on biologically inspired copper− benzimidazole-based CN stretching mode. The main oxygen systems.8 differences with the previously reported cupric complex Due to the ubiquitous presence of coordinated histidine [(2BB)CuII(H2O)2](OTf)2 include the frequencies of the residues in metalloenzyme active sites, most bioinspired N−H stretch at 3241 (w) cm−1, the slightly higher intensity of complexes employ amines and nitrogen-containing hetero- the CN stretching mode previously observed at 1617 cm−1, cycles as histidine models. This approach has relied heavily on and the shoulder at 3356 cm−1 that was assigned to O−H amine-, pyrazole-, and pyridine-based ligands. In contrast, stretching modes in the cupric complex. reports on ligands that incorporate the more electronically Solid-State Structure of [(2BB)CuI]OTf. [(2BB)CuI]OTf relevant imidazolyl and/or benzimidazolyl donor substituents was dissolved in a minimum amount of anhydrous CH3CN, remain relatively scarce,9 with the notable exception of the and after slow evaporation, colorless X-ray quality crystals of binucleating scaffolds developed by the groups of Casella.10 [(2BB)CuI]OTf were obtained. The metric parameters around We have recently exploited 2-substituted benzimidazoles to the Cu(I) center in the [(2BB)CuI]+ cation are in the range of assemble chelating ligands for mononuclear CuII/I complexes; those reported for the enzymes, displaying the characteristic T- these ligands offer versatility as good σ-donors and reasonable shape of the reduced form of the active sites of LPMOs, as π-acceptors toward copper ions, as well as varying degrees of shown in Figure 1 and Tables S1−S3. The N1−Cu1−N4 open steric protection.11 Exploration of their reactivity has revealed the formation of free superoxide anions, presumably released by solvent displacement from the putative CuII-superoxide species obtained from the corresponding cuprous complexes and O2.11d These steps have been suggested to occur in LPMOs on the basis of DFT calculations5 and recently reported structural data,6,12 in a rapid inner-sphere reductive activation of O2.5 In this context, we recently reported the use of bis[(1- methyl-2-benzimidazolyl)ethyl]amine (2BB,13,10b Chart 1) as a simple structural model of the ligand environment in mononuclear active sites of LPMO.14,15 Our interest was spurred by the presence of a methylated nitrogen atom in one of the histidinic imidazoles of fungal LPMO2a and the presence of a central secondary amine available for potential H-bonding with substrates. Additionally, the 2BB/CuII system afforded Figure 1. Mercury diagram of [(2BB)CuI]+ at the 50% probability complexes with bonding parameters in the range of those of level. H atoms, except for H3 on the central amine, and the triflate LPMO active sites (Cu−N distances of ∼1.98 Å, N−Cu−N counter ion are omitted for clarity; color code: C, grey; N, blue; Cu, angles of ∼97 and ∼165°). In continuation of these turquoise. preliminary studies, we herein report detailed spectroscopic studies of [(2BB)CuII/I]2+/+ complexes, their interaction with angle of 165° in [(2BB)CuI]+ is identical to the corresponding O2, H2O2, and KO2, as well as their oxidative reactivity toward one in the AA9−AA11 active sites,16 whereas the two closed cellobiose as a cellulose model substrate. ■ angles, N3−Cu1−N4 and N1−Cu1−N3 of 98.01 and 96.82°, respectively, for [(2BB)CuI]+, are equivalent to that of 97° in RESULTS AND DISCUSSION LPMO. A comparison of the bond distances is presented in Synthesis of Cu+ Complex. Previous attempts to isolate Table 1, showing a larger Cu−N distance to the central cuprous 2BB complexes required for O2 binding and nitrogen donor than that to the benzimidazole N atoms. 10730 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article Table 1. Bond Distances and Angles in [(2BB)CuI]OTf and support for deprotonation of a coordinated water molecule Comparison with the Corresponding Parameters in over an amine donor is provided by studies on the effect of Reduced LsAA9A16 metal coordination on the pKa values of biologically relevant ligands, with transition-metal ions increasing the acidity by 1− parameter [(2BB)Cu]OTf parameter Cu(I)LsAA9A 3 log units.20 These observations suggest that the potential Cu1−N4 1.901 Å Cu−Nσ (His1) 1.9 Å involvement of a deprotonated amine during the catalytic cycle Cu1−N1 1.908 Å Cu−Nε (His78) 2.0 Å may be ruled out in our system and plausibly in the active site Cu1−N3 2.173 Å Cu−NH2(His1) 2.2 Å of LPMO as well. In the enzymatic system, the proposed N4−Cu1−N1 164.54° Nσ−Cu−Nε 165° deprotonation of the terminal amine of the histidine brace12 N4−Cu1−N3 98.01° Nσ−Cu−NH2 97° seems unlikely near neutral pH,4−6 particularly considering N1−Cu1−N3 96.82° Nε−Cu−NH2 97° determinations of the pKa values of terminal amino groups in related ATCUN peptide complexes.21 Quantum refinement of Interestingly, the triflate anion interacts with [(2BB)CuI]+ via X-ray and neutron diffraction data also argues against −NH2 hydrogen bonding with the central amine N−H, defined by an deprotonation.22 O2···N3 distance of 3.363(3) Å and an O2···H3−N3 angle of Electrochemical Studies. The species distribution is 149(2)°. supported by electrochemical results obtained at pH 5.0, 7.0, Solution Studies of [(2BB)CuII(H2O)2](OTf)2. Insight into 9.0, and 10.5 of aqueous solutions of [(2BB)CuII(H2O)2]- the nature of the cupric complex in an aqueous solution was (OTf)2. The system behaves quasireversibly, with one redox gained through equilibrium studies to determine its speciation. wave assigned to the CuII/I couple at E1/2 = 226 mV vs Potentiometric titration of 2BB afforded pKa values of 3.68, standard hydrogen electrode (SHE) at pH 5.0−7.0, which is in 4.86, and 7.84 for the protonated forms of the two the range of the redox potentials reported for LPMOs (150− benzimidazolic and amine sites, respectively. The values for 370 mV vs SHE). Differences were observed at pH 9.0−10.5; the benzimidazolic sites are lower (in water at 25 °C) than in the former case, a shoulder at a slightly more negative those of benzimidazole at 5.58,17a 1-methylbenzimidazole, 1- potential indicates the presence of an additional species. This is ethylbenzimidazole at 4.88, and even 1-methyl-5-chlorobenzi- confirmed at pH 10.5, where the shoulder becomes the only midazole at 3.88.17b Subsequently, the stability constant for wave observed (Figure 3). In CH3CN, electrochemical analysis [(2BB)CuII(H2O)2]2+ was determined at log K = 12.44 after of [(2BB)CuI]OTf shows a quasireversible behavior, with E1/2 refinement of the potentiometric data with Hyperquad = 76 mV vs SHE (−272 mV vs ferrocenium/ferrocene, Figure software,18 considering temperature, volumes, and concen- S3). trations of analytes (HNO3, NaOH, potassium biphthalate, 2BB, and its cupric complex, see Figure S2). This log K value is high compared to that of Cu(II) with bidentate 2- hydroxymethylbenzimidazole and 1-methyl-2-hydroxymethyl- benzimidazole at 9.30 and 9.66, respectively.17a The species distribution was determined with MEDUSA software19 (ionic strength 0.1 M NaNO3), where the dicationic species [(2BB)CuII]2+ clearly predominates in the pH range of 4−8; see Figure 2. One or two molecules of water are likely Figure 3. Cyclic voltammograms of 1 mM [(2BB)CuII(H2O)2]- (OTf)2 solutions in phosphate buffer (PB) at pH 5.0, 7.0, and 9.0, and in carbonate buffer at pH 10.5, acquired at 1 V s−1 with a glassy carbon electrode (at pH 10.5, the system is heterogeneous). Figure 2. Speciation diagram for [(2BB)CuII]2+. Insight into the identity of the additional species formed at pH 9.0−10.5 was obtained through ESI-MS analysis of a coordinated to the cupric ion in solution to form [(2BB)- solution of [(2BB)CuII(H2O)2](OTf)2. A signal was detected CuII(H2O)n]2+ (n = 1, 2), as observed in the solid state.15 at m/z 887, which was assigned to the mixed hydroxo/ Interestingly, a monodeprotonated species is predicted to be carbonato-bridged dicopper complex shown in Scheme 1 on present around pH 7.5 and above, which was assigned to a the basis of its mass and isotope distribution. The complex cupric−hydroxo complex [(2BB)CuII(OH)(H2O)n−1]+, likely detected may be formed upon aerobic CO2 insertion into an in the form of a bridged dimer [{(2BB)CuII(H2O)n−1}2(μ- initially formed dihydroxo-bridged dimeric complex [{(2BB)- OH)2]2+, with log K = 15.88. This type of species would be CuII(H2O)x}2(μ-OH)2]2+.23 UV−vis spectroscopic evidence of favored over a complex with a deprotonated central amine, on the formation of the putative [{(2BB)CuII(H2O)x}2(μ- the basis of the additional experimental evidence for a OH)2]2+ was obtained by addition of NEt3 to an acetonitrile hydroxo-bridged dicopper complex (vide infra). Further solution of [(2BB)CuII(H2O)2](OTf)2 at room temperature. 10731 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article Scheme 1. Proposed Hydroxo/Carbonato-Bridged Dicopper Moreover, the optimum pH value for enzymatic activity is Complex Formed from [(2BB)CuII(H2O)2](OTf)2 in Water slightly acidic,4,5 rendering the deprotonation of the terminal at pH 9.0 with Experimental (Left) and Simulated Isotopic amine even less likely. Mass Spectrometry Patterns (Right) Reactivity of [(2BB)CuI]OTf with O2. The reactivity of [(2BB)CuI]OTf with dioxygen was explored initially by UV− vis spectroscopy at low temperature in an acetonitrile solution. Gentle bubbling of O2 through 0.3 mM solutions of [(2BB)CuI]OTf at 243 K resulted in the appearance of a broad band at 360 nm, assigned to ligand-to-metal charge transfer (LMCT) of a copper−oxygen species, accompanied by d−d transition at 680 nm; a similar behavior was observed in acetone at 193 K. The species was stable for at least an hour, although decay was observed when the temperature was raised; see Figure S5. These changes were not observed under similar conditions or at lower temperature in tetrahydrofuran (THF) solutions, and reactivity studies in this solvent were not pursued. In aqueous phosphate buffer (PB) at 273 K, no significant changes were observed by UV−vis spectroscopy upon bubbling O2 through in situ-reduced (with 1.2−1.5 equiv of sodium ascorbate) [(2BB)CuII(H2O)2](OTf)2. In both A new band was observed at 330 nm (ε ∼ 2000 M−1 cm−1), acetonitrile and PB solutions, the products generated at low which can be assigned as a hydroxo to Cu(II) ligand-to-metal temperature were characterized by ESI-MS, revealing the charge-transfer (LMCT) transition; see Figure S4. presence of an intense signal at m/z 396 assigned to These observations confirm that deprotonation of a [(2BB)CuI]+; see Figure S1. coordinated water molecule is more facile than that of the The mixture was characterized by electron paramagnetic amine in our system. This deprotonation occurs at a higher pH resonance (EPR) spectroscopy, revealing that although value (7.5) than that used in the study suggesting terminal [(2BB)CuI]OTf reacts slowly with dioxygen in acetonitrile amine deprotonation (i.e., pH 7.0),12 further emphasizing that at 243 K, a signal characteristic of a cupric complex in an axial the amine is likely to remain protonated at enzymatic pH. coordination environment appears after ca. 10 min [g∥ = 2.250, Scheme 2. Proposed Reaction Manifolds and Putative Copper Complexes Generated from [(2BB)CuI]+/O2 (Top Left) or [(2BB)CuII(S)2]2+/H2O2/NEt3 (Top Right) 10732 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article g⊥ = 2.082, A∥ = 423 MHz (151 G), integration 21% relative to weak donors in the axial positions [(2BB)CuII(OOH)(S)]+ or Cu(II) external standard]. A reasonable explanation would [(2BB)CuII(OH)(S)]+ (S = solvent, see Scheme 2). imply initial formation of an EPR silent cupric−superoxo Reactivity of [(2BB)CuI]OTf and [(2BB)CuII(H2O)2]- complex [(2BB)CuII(O2)]+, likely in equilibrium with a side- (OTf)2 with KO2. The reactivity of the copper complexes on peroxo-bridged dicopper(II) complex [{(2BB)CuII}2(μ- with a one-electron reduced form of dioxygen was explored η2:η2-O2)]2+ (on the basis of UV−vis spectroscopy and H2O2 with KO2; this may be important to establish the oxidant in reactivity section below). H-abstraction from solvent or LPMOs, since H2O2 has been suggested instead of the cellobiose may result in the EPR-active cupric hydroperoxo originally proposed O2.6 Thus, addition of KO2 to solutions of [(2BB)CuII(OOH)(S)]+ and/or the further downstream [(2BB)CuI]OTf was monitored by UV−vis absorption hydroxo complex [(2BB)Cu II (OH)(S)] + (S = H 2 O, spectroscopy to identify potential copper−oxygen species, CH3CN). These species may give rise to the shoulder at 330 both in acetonitrile and PB solution at pH 7.0. In acetonitrile, nm by UV−vis absorption spectroscopy (Figure S5). Cu(I) the main spectral change consisted of a shoulder at 365 nm, oxygenation leading to a Cu(II)-hydroperoxo complex through which intensified slightly after addition of cellobiose. Analysis H-abstraction has been previously observed.24 Finally, both of the reaction mixture by ESI-MS revealed the presence of a Cu(II)-hydroperoxo and dicopper(II)-peroxo may further peak at m/z 431 assigned to [(2BB)CuII(OH)(H2O)]+, along react to generate the final product [(2BB)CuII(OH)(S)]+, in with peaks at m/z 396 and 545 characteristic of [(2BB)CuI]+ equilibrium with the dimeric form [{(2BB)CuII(S)x}2(μ- and [(2BB)CuII(OTf)]+ (Figure S6). When cellobiose was OH)2]2+ (Scheme 2). added, ESI-MS evidenced the presence of oxidative degrada- In PB/glycerol at pH 7.0 and 273 K, [(2BB)CuII(H2O)2]- tion products (vide infra). An analogous behavior was (OTf)2 reduced in situ with 1.2 equiv of sodium ascorbate observed in PB by UV−vis, although the shoulder was produced no changes after bubbling of O2, changes were not observed at 345 nm. Three relevant peaks were detected by observed. However, addition of cellobiose to the oxygenated ESI-MS: the one at m/z 412 is consistent with a PB solution resulted in an axial signal with g∥ = 2.275 and g⊥ = monooxygenation product [(2BB)CuII(O)]+, which could 2.054 and A∥ = 502 MHz (179 G) and A⊥ = 70 MHz (25 G); correspond to ligand hydroxylation, although such a product see Figure 4; the spectrum was simulated as containing one or was not detected after demetallation with excess aqueous ammonia; a cupryl species cannot be firmly supported without further evidence. A peak at m/z 428 was assigned to the putative cupric−superoxo [(2BB)CuII(O2)]+ and the one at m/z 602 was tentatively assigned to [(2BB)CuII(O)(OTf)- (CH3CN)]+ (Figure S7); the aforementioned species at m/z 396 and 545 were also detected. These observations imply that highly reactive copper−oxygen species are formed in the Cu(I)/KO2 system, allowing detection of the relatively stable downstream product [(2BB)CuII(OH)(H2O)]+. EPR spectra recorded from flash frozen samples over the course of the reaction reveal that after addition of KO2 to [(2BB)CuI]OTf in acetonitrile, a signal characteristic of a cupric complex in an axial environment appears, with g∥ = 2.250 and g⊥ = 2.050 and A∥ = 536 MHz (191 G) and A⊥ = 81 Figure 4. EPR spectra of in situ-generated [(2BB)CuI]+ with 1.5 MHz (29 G), Figure 4. The signal becomes more intense upon equiv of sodium ascorbate after bubbling O2 and adding 10 equiv of addition of cellobiose, possibly due to generation of the cellobiose (3 mM in PB, blue trace), and mixture of [(2BB)CuI]OTf, proposed [(2BB)CuII(OH)(H2O)]+ after H-abstraction,27 1.2 equiv of KO2, and 10 equiv of cellobiose (3 mM in CH3CN, red which was detected by ESI-MS. The analogous reaction of trace). [(2BB)CuII(H2O)2](OTf)2 in either CH3CN or PB solutions with KO2 did not show considerable changes in its UV−vis two Cu(II) species, with only the former case resulting in a absorption spectra, even after addition of cellobiose. No good fit. Examples of mononuclear copper−hydroperoxo evidence for copper−oxygen species was obtained by ESI-MS complexes featuring tripodal pyridylamine and pyridylamine− or EPR spectroscopy in the reactions of the Cu(II) complex thioether ligands have been characterized by EPR spectroscopy with KO2. Thus, a cupric−superoxo complex cannot be and assigned to trigonal bipyramidal [g∥ = 2.004, g⊥ = 2.207, detected under these conditions. A∥ = MHz 305 (109 G), A⊥ = 210 MHz (75 G)]25 or square Reactivity of [(2BB)Cu]OTf and [(2BB)CuII(H2O)2](OTf)2 pyramidal geometries [g∥ = 2.24, g⊥ = 2.06, A∥ = 471 MHz with H2O2. The reaction of [(2BB)CuII(H2O)2](OTf)2 with (168 G); g∥ = 2.25, g⊥ = 2.04, A∥ = 504 MHz (180 G)].26 A H2O2 was monitored by UV−vis spectroscopy to determine similar situation has been observed with monomeric cupric− whether different species are formed relative to the reaction of hydroxo complexes: EPR spectroscopic data for tripodal the Cu(I) complex with O2, both in acetonitrile and PB (pH complexes have been observed to give rise to “inverse” signals, 7.0). In acetonitrile, no changes with respect to the original whereas axial signals with some degree of rhombicity have spectrum were observed, even after addition of cellobiose; an been observed in the few cases of tetragonal Cu(II)−OH analogous behavior was observed in PB. High-performance complexes [gx = 2.032, gy = 2.055, gz = 2.185, ACu = 588 MHz liquid chromatography-mass spectrometry (HPLC-MS) anal- (189 G), AN = 40−53 MHz (14−19 G)].27 Resemblance of ysis indicated that cellobiose did not undergo degradation. In the parameters to those of the tetragonal species suggests a contrast, reaction of [(2BB)CuII(H2O)2](OTf)2 with 1:1 square pyramidal coordination environment for the putative mixtures of H2O2/Et3N in acetonitrile resulted in spectral Cu(II)−hydroperoxo7c,26 or hydroxo complexes, likely with changes characterized mainly by an absorbance maximum 10733 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article around 365 nm (see Figures S8 and S9) and a d−d transition at 670 nm. Another copper complex appears to be present, on the basis of the shoulder observed at ca. 330 nm. A mixture of the proposed dimeric [{(2BB)CuII}2(μ-η2:η2-O2)]2+ and monomeric [(2BB)Cu II (OOH)(S)] + and/or [(2BB)- CuII(OH)(S)]+ complexes (cf. Scheme 2) may be responsible for the optical features described, with ample precedent for μ- η2:η2-peroxo dicopper(II) complexes with intense absorption bands around 360 nm.10,28 The few examples reported of monomeric, tetragonal cupric hydroperoxo, and hydroxo complexes feature LMCT bands that are blue-shifted relative to the dimeric peroxo-bridged species,25−27 which is consistent with our observations. The similarity with the UV−vis absorption spectrum of the reaction mixture with [(2BB)CuI]+ and dioxygen (see Figure S10) further suggests the formation Figure 5. UV−vis absorption spectrum of [(2BB)CuII(H2O)2]- of peroxo dicopper(II) species in the reaction of Cu(I) with O2 (OTf)2 (0.3 mM in CH3CN, 3 equiv of NaOAc in 60 μL of CH3CN/ described above. [{(2BB)CuII}2(μ-η2:η2-O2)]2+ may form by MeOH, black trace) after addition of 3 equiv of 1:1 H2O2/Et3N rapid capture of an initially formed cupric−superoxo [(2BB)- (green), and 30 equiv of the same (brown); follow-up of the reaction CuII(O2)]+ by a second equivalent of [(2BB)CuI]+. The and the final spectrum after 100 min (purple trace). reaction of [(2BB)CuI]OTf with H2O2 was also analyzed by UV−vis absorption and EPR techniques, resulting in spectra due to formation of the antiferromagnetically coupled with poorly defined features. [{(2BB)CuII}2(μ-η2:η2-O2)]2+ (Figure S13). The presence of dicopper species in acetonitrile solution was Resonance Raman Spectroscopy. Further character- confirmed by cryospray ionization mass spectrometry (CSI- ization of the proposed dimer [{(2BB)CuII}2(μ-η2:η2-O2)]2+ MS) under conditions that are similar to those employed for and the associated complex formed upon addition of NaOAc UV−vis absorption spectroscopy. As soon as 3 equiv of H2O2/ was attempted by resonance Raman spectroscopy. Since these Et3N were added to [(2BB)CuII(H2O)2](OTf)2 in acetoni- species give rise to the bands observed by UV−vis absorption trile, the sample was injected at 243 K, identifying spectroscopy at 365 nm and 414 and 440 nm, respectively, [(2BB)2Cu2II(μ-CO3)(OTf)]+, [(2BB)2Cu2II(μ-O2)]2+, and measurements were carried out with mixtures of [(2BB)- [(2BB)CuII(OH)(H2O)]+. Spectral comparison with the CuII(H2O)2](OTf)2 and 1:1 H2O2/Et3N, with and without injected sample acquired at 293 K (once the band at 365 added sodium acetate, and excitation at 355 and 457 nm. nm decayed) shows that similar species are formed, but the Raman spectra recorded with excitation at 355 nm was bimetallic species are favored at higher temperature, on the employed at 243 K to favor detection of [{(2BB)CuII}2(μ- basis of their intensities (Table S4 and Figure S11). η2:η2-O2)]2+, whereas excitation at 457 nm was employed at Analysis of the reaction mixture by EPR spectroscopy reveals 293 K to favor detection of the unidentified species that gives that after reaction of [(2BB)CuII(H2O)2](OTf)2 with 1:1 absorption maxima at 414 and 440 nm (Figures S14 and S15). H2O2/Et3N in acetonitrile, the signal characteristic of a cupric Although strong signals are present, no isotope-sensitive bands complex in an axial coordination environment becomes weak are clearly discernible (H216O2 vs H218O2), as the spectra after ca. 10 min, likely due to formation of the appear to be dominated by ligand-based vibrations. In the case antiferromagnetically coupled [{(2BB)CuII}2(μ-η2:η2-O2)]2+; of [{(2BB)CuII}2(μ-η2:η2-O2)]2+, the main absorption band in this species is stable for at least 2 h at 243 K, on the basis of the UV−vis spectrum is associated to a 2BB-based transition UV−vis and EPR measurements. (see the theoretical section and Figures S21 and S22). Addition of the potentially chelating acetate anion (OAc−, 3 Preparative Scale Oxidations of Cellobiose with equiv in methanol/acetonitrile) to 0.3 mM [(2BB)- [(2BB)Cu]OTf and [(2BB)Cu(H2O)2](OTf)2. Oxidations of CuII(H2O)2](OTf)2 in acetonitrile was tested to interrogate cellobiose using 0.3 mM acetonitrile solutions of [(2BB)- the nature of the dimeric species formed upon addition of 30 CuII(H2O)2](OTf)2, 10 equiv of H2O2/Et3N, and 10 equiv of equiv of H2O2/Et3N.29,30 The reaction was monitored by UV− cellobiose at 243 K were undertaken (cf. Figure S16 and Table vis absorption spectroscopy at 243 K, showing the generation S6). Similarly, a comparative reaction was performed by of the band at 365 nm, attributed to [{(2BB)CuII}2(μ-η2:η2- generating the Cu(I) complex in situ from the reduction of O2)]2+, along with new bands at 414 and 440 nm; see Figure 5. [(2BB)CuII(H2O)2](OTf)2 with 10 equiv of sodium ascorbate The band at 365 nm decays over a period of 75 min at 243 and subsequent bubbling of O2, followed by addition of 10 K, with concomitant increase in the intensity of the bands at equiv of cellobiose (Figure S17 and Table S7). As a common 414 and 440 nm; this transformation occurs almost product of these reactions, sodium gluconate was detected by instantaneously by increasing the temperature to 293 K. HPLC-MS (Scheme 3 and Table 2), demonstrating that the Similar results were observed upon addition of O2, KO2, or [(2BB)CuII/I]2+/+ system has biomimetic activity for oxidative H2O2 to cuprous and cupric complexes in the presence of cellobiose degradation in the presence of the appropriate OAc−. CSI-MS analysis of the final products at 243 K revealed oxidizing agent. It should be noted that the reaction that the presence of [(2BB) 2 Cu2 II (μ-η 2-OAc)2 (OTf)]+ and employed in situ-generated [(2BB)CuI]+ in PB with O2 as [{(2BB)CuII}2(μ-η2:η2-O2)]2+ (Table S5), with their inten- oxidant is slightly more selective toward gluconate formation, sities increasing upon warming up the mixture to 293 K; see since it was detected in higher yield by HPLC-MS. In PB with Figure S12. EPR spectroscopy reveals that once H2O2/Et3N is H2O2/Et3N as oxidant, the doubly oxidized glucose in Scheme added, the signal characteristic of the cupric complex in an 3 is the main product (Figure S18). The H2O2 reaction shows axial coordination environment becomes weak after ca. 10 min a greater degree of cellobiose degradation, but more products 10734 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article Scheme 3. Cellobiose Degradation by [(2BB)CuII/I]2+/+ and with those observed for the structurally characterized Several Oxidants, with Main Products Detected by HPLC- complexes. MS Initial Complexes. [(2BB)CuI]+ and [(2BB)CuII]2+ were considered with either a molecule of water or CH3CN coordinated, since those solvents were employed for physical measurements and reactivity tests. For both solvents, the coordination of one molecule is energetically favorable, with values between 3 and 30 kcal mol−1 (Table 3). The Gibbs free energies for the solvent-coordinated cuprous complexes notwithstanding, calculated for [(2BB)CuI(H2O)]+ and [(2BB)CuI(NCCH3)]+ at −3 and −5 kcal mol−1, are rather low compared with the binding constants calculated for their cupric counterparts. This is consistent with the lack of experimental evidence for coordinated solvent molecules in the solid state or in solution by ESI-MS, which can be are observed (i.e., H2O2 is a more active but less selective attributed to the high lability of the Cu(I) centers. The oxidant, Figures S16−S18, Tables S6−S8). Reactions were also geometries of [(2BB)CuI]+ and the related solvent-coordi- performed using in situ-reduced cupric complex (with 1.2 and nated [(2BB)CuI(H2O)]+ were optimized, and their main 2.4 equiv ascorbate) and KO2 as an oxidizing agent, with geometrical parameters are shown in Table S11. Initial similar degradation products detected, along with unreacted optimization was probed with [(2BB)CuI]+ and subsequently cellobiose (Table 2). Control experiments with [(2BB)- CuII(H2O)2](OTf)2, CuSO4, H2O2/NEt3, and CuSO4/H2O2/ tested with one H2O ligand, revealing that the nitrogen atoms NEt3 show no oxidative degradation of cellobiose, although a are maintained in a planar configuration. In the resulting previously observed dehydration product was detected at m/z [(2BB)CuI(H2O)]+, the cuprous center should be defined as 334 in the first case. Care must be exercised in the four-coordinate but the calculated distance of the nitrogen interpretation of chromatograms, since all samples containing atom from the central amine to the Cu+ ion is considerably H2O2 have a leaching product from the HPLC column (Figure longer than in the absence of a bound H2O (2.407 vs 2.940 Å). S19). Redox Properties. The redox potentials for [(2BB)- Glucose oxidation was tested in PB with 0.3 mM CuII/I]2+/+, [(2BB)Cu(CH3CN)II/I]2+/+, and [(2BB)Cu- [(2BB)CuII(H2O)2](OTf)2 and 10 equiv of H2O2/NEt3 (H2O)II/I]2+/+ couples were estimated by simulating the (Figure S20 and Table S9), resulting in mono-oxidation as oxidation process in an acetonitrile solution and referenced the major product relative to double oxidation (33 vs 14%). In to the ferrocenium/ferrocene couple.31 The calculated addition to unreacted glucose, sodium gluconate and glucose potential for trigonal [(2BB)CuII/I]2+/+ was 219 mV, which aldehyde were detected in small quantities (Table 2). The only differs significantly from that calculated with one coordinated precedent of related inorganic systems capable of cleaving molecule of acetonitrile (−135 mV) or water (−305 mV). glucose derivatives features 4-nitrophenolate as the leaving Since the latter values with solvent coordinated to the copper group.7b,c centers are reasonably close to the experimentally determined Computational Studies. The thermodynamic feasibility data in acetonitrile solution (E1/2 = −272 mV), solvent of the copper−oxygen species that may be formed in the coordination seems to be plausible. In all cases, redox pairs presence of the oxidants tested (O2, KO2, and H2O2) was involve a metal-centered process, i.e., the CuII/I couple. evaluated by density functional studies. Reactions of [(2BB)- Dimeric Species. During our attempts to detect copper− CuI]+, [(2BB)CuII]2+, [(2BB)CuI(S)]+, and [(2BB)CuII(S)]2+ oxygen species that may be responsible for the oxidation of (S = H2O, CH3CN) with dioxygen, potassium superoxide, or cellobiose, the main candidates are the dicopper(II) species hydrogen peroxide were evaluated as they appear to generate [{(2BB)CuII}2(μ-η2:η2-O2)]2+ and derivatives having acetate the dimeric [{(2BB)CuII}2(μ-η2:η2-O2)]2+ and potentially co-ligands. Since several geometries for Cu2O2 cores have been [{(2BB)CuIII}2(μ-O)2]2+. The optimized structures were described in the literature,10,28−30 different conformations of validated by comparison of the bonding parameters obtained the ligands were considered in the molecular optimization of Table 2. Product Distribution in Cellobiose Degradation Reactions (% Yield by HPLC Integration Relative to Cellobiose) assignment cellobiose degradation glucose degradation O2 + Cu(I) + 10 equiv sodium ascorbate H2O2/Et3N + Cu(II) H2O2/Et3N + Cu(II) H2O2/Et3N + Cu(II) conditions (H2O) (CH3CN) (PB) (PB) % [Na + cellobiose]+ 65 8 % [Na + cellobiose-2OH]+ 28 70 % [Na + glucose]+ 13 % [2Na + gluconate]+ 5 3 2 1 X% [doubly oxidized glucose + H]+ 18 14 % [oxidized glucose + cellobiose + H]+ 15 33 % [Na + doubly oxidized glucose + 37 2H2O]+ % [Na + glucose aldehyde]+ 10 5 10735 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article Table 3. Free Energies of Formation of [(2BB)CuI(S)]+ and [(2BB)CuII(S)]2+ with S = H2O or CH3CN reactions ΔG (gas phase) ΔG (CH3CN) [(2BB)Cu]+ + CH3CN → [(2BB)Cu(CH3CN)]+ −11.2 −4.8 [(2BB)Cu]2+ + CH3CN → [(2BB)Cu(CH3CN)]2+ −30.9 −16.8 [(2BB)Cu]+ + H2O → [(2BB)Cu(H2O)]+ −8.9 −3.1 [(2BB)Cu]2+ + H2O → [(2BB)Cu(H2O)]2+ −26.6 −19.0 dicopper species (Figure S21). Their relative energies in activations remains a major challenge in copper bioinorganic acetonitrile were calculated (Table S12) and, as expected, the chemistry. ■ bimetallic complexes were found to be more stable than the mononuclear T-shape Cu(I) complex by more than 20 kcal EXPERIMENTAL SECTION mol−1 in CH3CN. Nevertheless, the presence of acetate All synthetic procedures were performed under a dry modifies the dioxygen coordination around the metal centers, dinitrogen atmosphere in a glovebox or by conventional predicting the additional species [{(2BB)CuII}2(μ-η1:η1- Schlenk techniques. THF and diethyl ether were obtained by O2)(μ-AcO)]+. oxygen- and water-free distilling from sodium benzophenone UV−vis spectra were calculated for complexes with and under a N2 atmosphere; acetonitrile was distilled from CaH2. without acetate as co-ligand (Figure S22). Apparently, Reagents were purchased from commercial suppliers and used coordination of a bridging acetate does not alter the main as received; 2BB10b,13 and [(2BB)CuII(H2O)2](OTf)214,15 electronic transitions in the bimetallic complexes. For were synthesized according to literature procedures. 1H and [{(2BB)CuII}2(μ-η2:η2-O2)]2+, an intense band around 330 13 C NMR spectra were recorded on a JEOL Eclipse 300 or a nm with a 2BB to dioxygen charge-transfer character was calculated, together with a weaker one at 420 nm having an Bruker Avance DRX spectrometer at 300 and 75 MHz or 500 MLCT character; the first band is consistent with the and 125 MHz, respectively, using the residual protiated solvent experimentally observed absorption at 365 nm. For optimized signal or TMS as internal references (TMS δ = 0.00, CHCl3 δ [{(2BB)CuII}2(μ-η1:η1-O2)(μ-AcO)]+ in Figure S21 (species = 7.26 ppm). Electrospray ionization mass spectrometry (ESI- D), a medium intensity band is predicted at 420 nm, although MS) experiments were performed with a JEOL JMS-AX505HA the one predicted at around 650 nm is not observed spectrometer. Positive-ion FAB+ mass spectra were acquired experimentally (Figures S22 and S23); assignment of using a JEOL JMS-SX-102A mass spectrometer operated at an [{(2BB)CuII}2(μ-η1:η1-O2)(μ-AcO)]+ is only tentative with- accelerating voltage of 10 kV from a nitrobenzyl alcohol matrix out experimental verification. by using xenon atoms at 6 keV. Cryospray ionization mass ■ spectra were recorded on a Bruker MicrOTOF-Q II instru- ment at Serveis Tècnics of the University of Girona. Samples CONCLUSIONS were introduced into the mass spectrometer ion source by [(2BB)CuI]+ acts as a structural and electronic model of the direct infusion using a syringe pump and were externally reduced site of LPMO enzymes. Speciation studies at different calibrated using sodium formate. A cryospray attachment was pH values, together with electrochemical and mass spectrom- used. Temperature of the nebulizing and drying gases was set etry measurements at pH 9.0−10.5, allow us to establish that at −30 °C. The instrument was operated in both positive- and the pKa of water molecules bound to 2BB-coordinated cupric negative-ion modes. centers is lower than that of the ligand-based amine moiety; X-ray Crystallography. A single crystal of [(2BB)CuI]- this observation is relevant to the enzymatic system, where the OTf mounted on a glass fiber was studied with an Oxford proposed deprotonation of the amine of the histidine brace Diffraction Gemini “A” diffractometer and a CCD area appears to be precluded. detector (λMoKα = 0.71073 Å, monochromator: graphite) Oxidative cleavage of cellobiose, which was employed as a source equipped with a sealed tube X-ray source at 130 K. Unit model polysaccharide substrate, could be effected in the cell constants were determined with a set of 15/3 narrow presence of [(2BB)CuI]+ or [(2BB)CuII(H2O)2]2+ under frame/run (1° in ω) scans. A data set consisted of 235 frames appropriate conditions. Clear identification of a copper− of intensity data collected and a crystal-to-detector distance of oxygen species responsible for the C−H activation that leads to 55.00 mm. The double pass method of scanning was used to the oxidative cleavage of cellobiose was not possible via low- exclude any noise. The collected frames were integrated by temperature UV−vis absorption, MS, Raman, or EPR using an orientation matrix determined from the narrow frame spectroscopic methods in CH3CN or PB. However, the scans. CrysAlisPro and CrysAlis RED software packages were detection of potential reactive intermediates by ESI-MS, as used for data collection and data integration.33 Analysis of the well as [(2BB)CuII(OH)(H2O)]+ as the common end integrated data did not reveal any decay. Final cell constants product, provides some clues regarding the mechanistic were determined by a global refinement of 12 733 reflections possibilities outlined in Scheme 2. (θ < 29.51°). Collected data were corrected for absorbance by Moreover, when a more reduced source of the oxidizing using analytical numeric absorption correction using a agent is provided in our system (i.e., H2O2 vs O2), more multifaceted crystal model based on expressions upon the degradation of cellobiose was observed by HPLC-MS on a Laue symmetry using equivalent reflections.34 Structure preparative scale, at a cost of lower selectivity. This is solution and refinement were carried out with SHELXS- consistent with the tentative mechanisms that have been 201435a and SHELXL-2014;35b WinGX v2014.1 software36 proposed for LPMOs, which imply Cu(I)/H2O2 as the most was used to prepare material for publication. Full-matrix least- competent combination32 and H2O2 as the natural oxidant for squares refinement was carried out by minimizing (Fo2 − Fc2)2. polysaccharide degradation.6 Nonetheless, identification of the All nonhydrogen atoms were refined anisotropically. The H reactive copper−oxygen species involved in the key C−H atom of the amine group (H−N) was located in the difference 10736 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article map and refined isotropically with Uiso(H) = 1.2 for H−N. H in a special cell at 25 °C, stirred under a nitrogen atmosphere, atoms attached to C atoms were placed in geometrically controlling the ionic strength at 0.1 M with NaNO3 under idealized positions and refined as riding on their parent atoms, homogeneous conditions. Potentiometric data were obtained with C−H = 0.95−0.99 Å and with Uiso(H) = 1.2Ueq(C) for in triplicate for acid, base, internal standard, and analytes using aromatic and methylene groups and Uiso(H) = 1.5Ueq(C) for a 702SM Titrino automatic titrator and TIAMO1.3 software; methyl groups. A summary of crystallographic data is presented for each triplicate, the standard deviation was determined and in Tables S1−S3, Supporting Information. Crystallographic the final values were averaged. Hyperquad18 was employed to data have been deposited at the Cambridge Crystallographic simulate the data obtained, so that the constants adjusted the Data Centre CCDC 1821707. model with χ2 lower than 12.6 with 95% confidence and the Computational Details. Unrestricted calculations were species distribution was determined with MEDUSA.19 carried out using the Gaussian 09 package.37 The hybrid Synthesis of [(2BB)CuI]OTf. In an inert atmosphere density functional method known as B3LYP was applied.38 glovebox, 2BB (100 mg, 0.30 mmol) and [CuI(NCCH3)4]OTf Effective core potentials were used to represent the innermost (113 mg, 0.30 mmol) were dissolved in 10 mL of anhydrous electrons of the transition-metal atoms and the valence double- CH3CN; after 3 h, volatiles were evaporated under reduced ζ quality basis set LANL2DZ associated with the pseudopo- pressure and the solid obtained was washed with 10 mL of tentials.39 The basis set used for the main group elements was anhydrous diethyl ether to afford colorless microcrystalline 6-31G*.40 Solvent effects were taken into account through [(2BB)CuI]OTf (39 mg): Yield: 24%; mp 150−152 °C; 1H PCM calculations (acetonitrile, ε = 36.64 and water, ε = NMR (300 MHz, CD3CN): δ 7.64 (s, 5H, Ar), δ 7.24 (s, 1H, 78.3553)41 by using the geometries optimized for the gas Ar), δ 7.06 (p, J = 7.2 Hz, 5H, Ar), δ 4.97 (dd, J = 13.4, 3.3 Hz, phase (single-point calculations). Gibbs free energies in 1H, NH), δ 3.42 (m, 7H, N-methyl), δ 3.21 (m, 1H, CH2N−), solution were calculated from a thermodynamic cycle of the δ 3.01 (dt, J = 40.4, 20.2 Hz, 4H, CH2C−); ESI-MS m/z: chemical process by adding solvent effects. Redox potentials [(2BB)CuI]+ = 396, [(2BB)CuI(OTf)]+ = 545; UV−vis were estimated from these solvent calculations by using as (CH3CN): 264 (1744), 270 (1577), 282 (7044); IR (KBr): reference a value of 5.08 V for the absolute standard hydrogen 3273, 2957, 2906, 2852, 1614, 1503, 1481, 1452, 1410, 1327, electrode (SHE) in acetonitrile.42 The geometrical parameters 1257, 1227, 1147, 1067, 1027, 974, 936, 850, 825, 750, 656, were analyzed using the SHAPE program.43 633, 570, 515, 455, 434; elemental analysis calcd (%) for Vertical excitation energies were obtained from TD-DFT, as C21H22CuF3N5O3S: C, 46.28; H, 4.07; N, 12.85; found: C, implemented in Gaussian 09 in the presence of solvent 46.71; H, 4.15; N, 12.75. (acetonitrile). Calculations were performed at the same Reactivity Studies. The reactivity studies that were carried computational level as singlet states for the dimeric species, out with the 2BB-copper complexes were monitored by UV− using the broken-symmetry approach. Closed-shell singlet and vis, ESI-MS, and EPR: triplet states were also calculated, with small differences. For UV−vis, a 0.3 mM concentration solution of the Electrochemistry. Cyclic voltammetry measurements were complex under study was prepared, 2.5 mL was transferred to a made under N2 in anhydrous CH3CN, carbonate buffer 1 cm optical path cell (Schlenk type in the case of cuprous solution at pH 10.5, or phosphate buffer solutions at pH = 5.0, complex, which was sealed with a rubber septum). The cell was 7.0, and 9.0, with a CH Instruments potentiostat−galvanostat transferred to a precooled cryostat and chilled to 243 or 273 K, equipped with a glassy carbon working electrode and a depending on the solvent used (CH3CN, PB pH 7.0), with 10 platinum wire auxiliary electrode. Potentials were recorded min allowed for equilibration prior to reaction. Oxygenation of versus a pseudoreference electrode of AgBr(s)/Ag(wire) the cuprous complex was achieved by gently bubbling dioxygen immersed in 0.1 M NBu4Br acetonitrile or distilled water through the solution using a long needle for 50 s; addition of solutions. All voltammograms were started from the current KO2, H2O2 35%, or H2O2/Et3N dissolved in CH3CN, H2O, or null potential (Ei = 0) and were scanned in both directions, a MeOH/CH3CN mixture (2:3) was performed with 1 equiv positive and negative, and obtained at scan rates of 0.10 and of the oxidizing agent in 2 μL of solvent. The stability of the 0.20 V s−1. In agreement with IUPAC convention, the formed species was monitored every 5 min over a period of an voltammogram of the ferrocenium-ferrocene (Fc+/Fc) sys- hour by measuring the absorbance. Reversibility of each tem31 was obtained to establish the values of half wave generated system was also determined by bubbling Ar through potentials (E1/2) from the expression E1/2 = (Ea + Ec)/2. To the solution for 50 s using a long needle. Once the copper− obtain the normalized current for each complex, the measured oxygen species was formed, 10 equiv of cellobiose as a current was divided by the exact molar concentration of the substrate was added while stirring in 20 μL of water or a 2:3 electroactive species. H2O/CH3CN mixture; the samples were monitored every 5 Solution Studies. For potentiometric titrations, the min over a period of 40 min by measuring the absorbance. The electrode was calibrated in water, using standard buffers at same samples were injected in ESI-MS spectrometers under 25 °C. Commercial 0.1 M NaOH standard solutions were appropriate conditions. employed, after determination of carbonates present using the Characterization by X-band electron spin resonance (EPR) Gran Method, ensuring that their concentration was below 5%. spectroscopy was undertaken in frozen acetonitrile or PB pH The solution of HNO3 at 8 mM concentration was prepared 7.0/glycerol. Two hundred microliters of a 3 mM stock directly from concentrated acid. Potassium biphthalate was solution of the cuprous or cupric complex in CH3CN or PB/ used as an internal standard (oven vacuum-drying 24 h prior to glycerol were placed in quartz EPR tubes ([(2BB)CuI]OTf use). was prepared in a dinitrogen-filled glovebox) and were sealed The analytes 2BB and of [(2BB)CuII(H2O)2](OTf)2 were with rubber septa. Samples were frozen in liquid nitrogen acidified in a 1:8 ratio with respect to HNO3 to ensure before acquisition at 77 K for CH3CN and PB/glycerol. The complete protonation, maintaining in each case a concen- tubes were warmed in ice (273 for PB solution) or ice/acetone tration of 1.21 × 10−3 mol L−1. The solutions were maintained (243 K for CH3CN solution) baths until the samples became 10737 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article fluid. Oxygenation of cuprous complexes was achieved by Author Contributions bubbling O2 through the solutions using a long needle for 50 s. The manuscript was written through contributions of all The oxidizing agents used were as follows: KO2 in 1:1 authors. All authors have given approval to the final version of methanol/acetonitrile or H2O; H2O2, or H2O2/Et3N, main- the manuscript. taining 1 equiv of the oxidizing agent in 2 μL of solution. After Notes addition, spectra were acquired again at 77 or 133 K, respectively. After formation of copper−oxygen species, 10 The authors declare no competing financial interest. equiv of cellobiose dissolved in 20 μL of water or a 2:3 MeOH/CH3CN mixture was added and the spectra were acquired again at 77 K. In situ generation of the Cu(I) complex ■ ACKNOWLEDGMENTS The authors thank Carmen Márquez and Lucero Rı ́os for mass from [(2BB)CuII(H2O)2](OTf)2 was achieved by adding 1.2 spectrometry, Marı ́a de la Paz Orta for combustion analysis, equiv of sodium ascorbate in 20 μL of distilled water or a 2:3 Rocı ́o Patiño for IR, and Virginia Gómez-Vidales for EPR MeOH/CH3CN mixture. determinations; G.A. and P.R.M.A. acknowledge IQTC-UB for Raman spectra at 355 and 405 nm were recorded at −30 °C computational resources; AC MINECO Spain CTQ2016- (QNW temperature control cell holder) and room temper- 77989-P; A.C.N., E.N., and I.C., the Swedish Research Council ature. Raman spectra recorded at 355 (10 mW Cobalt lasers) (2014-0429); J.C., the Chinese Scholarship Council; J.B.K. and and 457 (50 mW Cobalt lasers) nm used a home-built system W.R.B., the Netherlands Ministry of Education, Culture and in which the laser was focused on the sample in a 180° Science (Gravity Program 024.001.035); A.C.N. and I.C., backscattering arrangement and Raman scattering was DGAPA-PAPIIT (IN210214 and IN203317), CONACyT collected, collimated, and subsequently refocused via a pair (Beca 293842), and CONACyT-ANUIES (ECOS 291247) of 2.5 cm diameter planoconvex lens (f = 15 and 10 cm, for financial support. respectively). The collected light was filtered by an appropriate long pass edge filter (Semrock) and dispersed by a Shamrock500i spectrograph (slit width 80 μm, Andor Technology) with a 2400 L/mm grating blazed at 300 nm ■ REFERENCES (1) (a) Phillips, C. M.; Beeson, W. T., IV; Cate, J. H.; Marletta, M. and a Shamrock300i spectrograph (slit width 80 μm, Andor A. Cellobiose Dehydrogenase and a Copper-Dependent Polysacchar- Technology) with a 1200 L/mm grating blazed at 500 nm, ide Monooxygenase Potentiate Cellulose Degradation by Neurospora respectively. Data were recorded using Andor Solis (Andor crassa. ACS Chem. Biol. 2011, 6, 1399. (b) Beeson, W. T.; Phillips, C. M.; Cate, J. H. D.; Marletta, M. A. Oxidative Cleavage of Cellulose by Technology) with spectral calibration performed using the Fungal Copper-Dependent Polysaccharide Monooxygenases. J. Am. Raman spectrum of acetonitrile/toluene 50:50 (v/v).44 Chem. Soc. 2012, 134, 890. Samples were held in quartz 1 cm path length cuvettes. (2) (a) Quinlan, R. J.; Sweeney, M. D.; Lo Leggio, L.; Otten, H.; Spectra analysis and processing was performed using the Poulsen, J.-C. N.; Johansen, K. S.; Krogh, K. B. R. M.; Jorgensen, C. I.; program Spectragryph 1.2.11.45 Tovborg, M.; Anthonsen, A.; Tryfona, T.; Walter, C. P.; Dupree, P.; Preparative Scale Cellobiose and Glucose Degrada- Xu, F.; Davies, G. J.; Walton, P. H. Insights into the oxidative tion. Studies were carried out with 3 mL of 2BB−Cu degradation of cellulose by a copper metalloenzyme that exploits complexes and 10 equiv of oxidizing agent or 1 atmosphere of biomass components. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 15079. O2 at 273 K (PB) or 243 K (CH3CN) and 10 equiv of (b) Hemsworth, G. R.; Taylor, E. J.; Kim, R. Q.; Gregory, R. C.; cellobiose or glucose in 1 mL of distilled water. The mixtures Lewis, S. J.; Turkenburg, J. P.; Parkin, A.; Davies, G. J.; Walton, P. H. were stirred for 2 h prior to characterization by HPLC-MS The Copper Active Site of CBM33 Polysaccharide Oxygenases. J. Am. Chem. Soc. 2013, 135, 6069. (c) Hemsworth, G. R.; Henrissat, B.; with an Agilent 1200 infinity Q-ToF spectrometer equipped Davies, G. J.; Walton, P. H. Discovery and characterization of a new with a Poroshell 120 column. ■ family of lytic polysaccharide mono-oxygenases. Nat. Chem. Biol. 2014, 10, 122. ASSOCIATED CONTENT (3) (a) Lieberman, R. L.; Rosenzweig, A. C. Crystal structure of a * Supporting Information S membrane-bound metalloenzyme that catalyses the biological The Supporting Information is available free of charge on the oxidation of methane. Nature 2005, 434, 177. (b) Balasubramanian, R.; Smith, S. M.; Rawat, S.; Yatsunyk, L. A.; Stemmler, T. L.; ACS Publications website at DOI: 10.1021/acsome- Rosenzweig, A. C. Oxidation of methane by a biological dicopper ga.9b00785. centre. Nature 2010, 465, 115. Materials, spectroscopic data, and methods; electrospray (4) Frandsen, K. E. H.; Simmons, T. J.; Dupree, P.; Poulsen, J.-C. mass spectrum; summary of crystallographic data; bond N.; Hemsworth, G. R.; Ciano, L.; Johnston, E. M.; Tovborg, M.; lengths; cyclic voltammograms; UV−vis spectrum; Johansen, K. S.; von Freiesleben, P.; Marmuse, L.; Fort, S.; Cottaz, S.; general procedure for sample preparations in reactivity Driguez, H.; Henrissat, B.; Lenfant, N.; Tuna, F.; Baldansuren, A.; Davies, G. J.; Lo Leggio, L.; Walton, P. H. The molecular basis of studies (PDF) polysaccharide cleavage by lytic polysaccharide monooxygenases. Nat. ■ Chem. Biol. 2016, 12, 298. AUTHOR INFORMATION (5) Kjaergaard, C. H.; Qayyum, M. F.; Wong, S. D.; Xu, F.; Hemsworth, G. R.; Walton, D. J.; Young, N. A.; Davies, G. J.; Walton, Corresponding Author P. H.; Johansen, K. S.; Hodgson, K. O.; Hedman, B.; Solomon, E. I. *E-mail: joseivan@unam.mx. Spectroscopic and computational insight into the activation of O2 by ORCID the mononuclear Cu center in polysaccharide monooxygenases. Proc. Gabriel Aullón: 0000-0001-7519-4522 Nat. Acad. Sci. U.S.A. 2014, 111, 8797. (6) Bissaro, B.; Rohr, A. K.; Muller, G.; Chylenski, P.; Skaugen, M.; Anna Company: 0000-0003-4845-4418 Forsberg, Z.; Horn, S. J.; Vaaje-Kolstad, G.; Eijsink, V. G. H. Wesley R. Browne: 0000-0001-5063-6961 Oxidative cleavage of polysaccharides by monocopper enzymes Ivan Castillo: 0000-0002-4876-4339 depends on H2O2. Nat. Chem. Biol. 2017, 13, 1123. 10738 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article (7) (a) Neisen, B. D.; Gagnon, N. L.; Dhar, D.; Spaeth, A. D.; superoxo triplet complexes as competent agents for H-abstraction Tolman, W. B. Formally Copper(III)−Alkylperoxo Complexes as relevant to the active site of PHM. Chem. Commun. 2015, 51, 11134. Models of Possible Intermediates in Monooxygenase Enzymes. J. Am. (12) Bacik, J.-P.; Mekasha, S.; Forsberg, Z.; Kovalevsky, A. Y.; Vaaje- Chem. Soc. 2017, 139, 10220. (b) Concia, A. L.; Beccia, M. R.; Orio, Kolstad, G.; Eijsink, V. G. H.; Nix, J. C.; Coates, L.; Cuneo, M. J.; M.; Ferre, F. T.; Scarpellini, M.; Biaso, F.; Guigliarelli, B.; Réglier, M.; Unkefer, C. J.; Chen, J. C. Neutron and Atomic Resolution X-ray Simaan, A. J. Copper Complexes as Bioinspired Models for Lytic Structures of a Lytic Polysaccharide Monooxygenase Reveal Copper- Polysaccharide Monooxygenases. Inorg. Chem. 2017, 56, 1023. Mediated Dioxygen Binding and Evidence for N-Terminal Deproto- (c) Muthuramalingam, S.; Maheshwaran, D.; Velusamy, M.; nation. Biochemistry 2017, 56, 2529. Mayilmurugan, R. Regioselective oxidative carbon-oxygen bond (13) Sorrell, T. N.; Vankai, V. A.; Garrity, M. L. Synthesis and cleavage catalysed by copper(II) complexes: A relevant model study reactivity of dinuclear copper complexes having a m-xylyl spacer for lytic polysaccharides monooxygenases activity. J. Catal. 2019, 372, between coordination units. Inorg. Chem. 1991, 30, 207. 352. (14) Casella, L.; Carugo, O.; Gullotti, M.; Doldi, S.; Frassoni, M. (8) (a) Itoh, S. Developing Mononuclear Copper−Active-Oxygen Synthesis, Structure, and Reactivity of model complexes of copper Complexes Relevant to Reactive Intermediates of Biological nitrite reductase. Inorg. Chem. 1996, 35, 1101. Oxidation Reactions. Acc. Chem. Res. 2015, 48, 2066. (b) Quist, D. (15) Castillo, I.; Neira, A. C.; Nordlander, E.; Zeglio, E. A.; Díaz, D. E.; Liu, J. J.; Karlin, K. D. Activation of dioxygen by Bis(benzimidazolyl)amine copper complexes with a synthetic copper metalloproteins and insights from model complexes. J. Biol. ‘histidine brace’ structural motif relevant to polysaccharide mono- Inorg. Chem. 2017, 22, 253. oxygenases. Inorg. Chim. Acta. 2014, 422, 152. (9) (a) Thompson, L. K.; Ramaswamy, B. S.; Dawe, R. D. Nickel(II) (16) Frandsen, K. E. H.; Poulsen, J.-C. N.; Tandrup, T.; Lo Leggio, and copper(II) complexes of the ‘tripod’ ligand tris(2- L. Unliganded and substrate bound structures of the cellooligosac- benzimidazylmethyl)amine. Can. J. Chem. 1978, 56, 1311. (b) Addi- charide active lytic polysaccharide monooxygenase LsAA9A at low son, A. W.; Hendriks, H. M. J.; Reedijk, J.; Thompson, L. K. Copper pH. Carbohydrate Res. 2017, 448, 187. complexes of the “tripod” ligand tris(2-benzimidazolylmethyl)amine: (17) (a) Lane, T. J.; Quinlan, R. Metal binding of the five- and six-coordinate copper(II) derivatives and some copper(I) benzimidazoles. J. Am. Chem. Soc. 1960, 82, 2994. (b) Davies, M. derivatives. Inorg. Chem. 1981, 20, 103. (c) Balamurugan, R.; T.; Mamalis, P.; Petrow, V.; Sturgeon, B. J. The preparation and Palaniandavar, M.; Gopalan, R. S. Trigonal planar copper(I) complex: spectrographic properties of some benziminazole-cobalt coordination synthesis, structure, and spectra of a redox pair of novel copper(II/I) compounds. J. Pharm. Pharmacol. 1951, 3, 420. complexes of tridentate bis(benzimidazol-2′-yl) ligand framework as (18) Gans, P.; Sabatini, A.; Vacca, A. Investigation of equilibria in models for electron-transfer copper proteins. Inorg. Chem. 2001, 40, solution. Determination of equilibrium constants with the HYPER- 2246. (d) Vaidyanathan, M.; Balamurugan, R.; Sivagnanam, U.; QUAD suite of programs. Talanta 1996, 43, 1739. Palaniandavar, M. Synthesis, structure, spectra and redox of Cu(II) (19) Puigdomenech, I. MEDUSA: Make Equilibrium Diagrams Using complexes of chelating bis(benzimidazole)thioether ligands as Sophisticated Algorithms; KTH, Stockholm, 2009. models for electron transfer blue copper proteins. J. Chem. Soc., Dalton (20) Martin, R. B. Pyrrole hydrogen ionization of imidazole Trans. 2001, 3498. (e) Seo, M. S.; Kim, N. H.; Cho, K.-B.; So, J. E.; derivatives in metal ion complexes and carbonic anhydrase. Proc. Park, S. K.; Clémancey, M.; Garcia-Serres, R.; Latour, J. -M; Shaik, S.; Nat. Acad. Sci. U.S.A. 1974, 71, 4346. Nam, W. A mononuclear nonheme iron(IV)-oxo complex which is (21) (a) Miyamoto, T.; Kamino, S.; Odani, A.; Hiromura, M.; more reactive than cytochrome P450 model compound I. Chem. Sci. Enomoto, S. Basicity of N-Terminal amine in ATCUN peptide 2011, 2, 1039. regulates stability constant of albumin-like Cu2+ complex. Chem. Lett. (10) (a) Casella, L.; Gullotti, M.; Radaelli, R.; Di Gennaro, P. A 2013, 42, 1099. (b) Kandemir, B.; Kubie, L.; Guo, Y.; Sheldon, B.; tyrosinase model system. Phenol ortho-hydroxylation by a binuclear Bren, K. L. Hydrogen evolution from water under aerobic conditions three-coordinate copper(I) complex and dioxygen. J. Chem. Soc., catalyzed by a cobalt ATCUN Metallopeptide. Inorg. Chem. 2016, 55, Chem. Commun. 1991, 1611. (b) Casella, L.; Monzani, E.; Gullotti, 1355. M.; Gliubich, F.; De Gioia, L. Cytochrome c oxidase models: (22) Caldararu, O.; Oksanen, E.; Ryde, U.; Hedegard, E. D. synthesis and reactivity of iron(III)−copper(II) complexes of Mechanism of hydrogen peroxide formation by lytic polysaccharide deuterohaemin−polybenzimidazole dinucleating ligands. J. Chem. monooxygenase. Chem. Sci. 2019, 10, 576. Soc., Dalton Trans. 1994, 3203. (c) Casella, L.; Monzani, E.; (23) (a) Mukherjee, J.; Balamurugan, V.; Hundal, M. S.; Mukherjee, Gullotti, M.; Cavagnino, D.; Cerina, G.; Santagostini, L.; Ugo, R. R. Fixation of CO2 in air: synthesis and crystal structure of a μ3-CO3- Functional Modeling of Tyrosinase. Mechanism of Phenol ortho- bridged tricopper(II) compound. J. Chem. Sci. 2005, 117, 111. Hydroxylation by Dinuclear Copper Complexes. Inorg. Chem. 1996, (b) Das, B.; Bhadbhade, M.; Thapper, A.; Ling, C. D.; Colbran, S. B. 35, 7516. (d) Garcia-Bosch, I.; Company, A.; Frisch, J. R.; Torrent- A new tri-nuclear Cu-carbonate cluster utilizing CO2 as a C1-building Sucarrat, M.; Cardellach, M.; Gamba, I.; Güell, M.; Casella, L.; Que, block − reactive intermediates, a probable mechanism, and EPR and L., Jr.; Ribas, X.; Luis, J. M.; Costas, M. O2 Activation and Selective magnetic studies. Dalton Trans. 2019, 48, 3576. Phenolate ortho Hydroxylation by an Unsymmetric Dicopper μ- (24) Fujii, T.; Yamaguchi, S.; Funahashi, Y.; Ozawa, T.; Tosha, T.; η1:η1-Peroxido Complex. Angew. Chem., Int. Ed. 2010, 49, 2406. Kitagawa, T.; Masuda, H. Mononuclear copper(II)−hydroperoxo (11) (a) Rodríguez Solano, L. A.; Aguiñiga, I.; López-Ortíz, M.; complex derived from reaction of copper(I) complex with dioxygen as Tiburcio, R.; Luviano, A.; Regla, I.; Santiago-Osorio, E.; Ugalde- a model of DβM and PHM. Chem. Commun. 2006, 4428. Saldívar, V. M.; Toscano, R. A.; Castillo, I. Bis(2- (25) Wada, A.; Harata, M.; Hasegawa, K.; Jitsukawa, K.; Masuda, H.; methylbenzimidazolyl)amine-Derived Copper Complexes and Their Mukai, M.; Kitagawa, T.; Einaga, H. Structural and Spectroscopic Antineoplastic Activity. Eur. J. Inorg. Chem. 2011, 3454. (b) Castillo, Characterization of a Mononuclear Hydroperoxo-Copper(II) Com- I.; Ugalde-Saldívar, V. M.; Rodríguez Solano, L. A.; Sánchez Eguía, B. plex with Tripodal Pyridylamine Ligands. Angew. Chem., Int. Ed. 1998, N.; Zeglio, E.; Nordlander, E. Structural, spectroscopic, and 37, 798. electrochemical properties of tri- and tetradentate N3 and N3S copper (26) (a) Kodera, M.; Kita, T.; Miura, I.; Nakayama, N.; Kawata, T.; complexes with mixed benzimidazole/thioether donors. Dalton Trans. Kano, K.; Hirota, S. Hydroperoxo−Copper(II) Complex Stabilized by 2012, 41, 9394. (c) Martínez-Alanis, P. R.; Sánchez-Eguía, B. N.; N3S-Type Ligand Having a Phenyl Thioether. J. Am. Chem. Soc. 2001, Ugalde-Saldívar, V. M.; Regla, I.; Demare, P.; Aullón, G.; Castillo, I. 123, 7715. (b) Maiti, D.; Lucas, H. R.; Narducci Sarjeant, A. A.; Copper Versus Thioether-Centered Oxidation: Mechanistic Insights Karlin, K. D. Aryl Hydroxylation from a Mononuclear Copper- into the Non-Innocent Redox Behavior of Tripodal Benzimidazoly- Hydroperoxo Species. J. Am. Chem. Soc. 2007, 129, 6998. laminothioether Ligands. Chem. - Eur. J. 2013, 19, 6067. (d) Sánchez- (27) Donoghue, P. J.; Tehranchi, J.; Cramer, C. J.; Sarangi, R.; Eguía, B. N.; Flores-Alamo, M.; Orio, M.; Castillo, I. Side-on cupric− Solomon, E. I.; Tolman, W. B. Rapid C−H Bond Activation by a 10739 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740 ACS Omega Article Monocopper(III)−Hydroxide Complex. J. Am. Chem. Soc. 2011, 133, (41) (a) Tomasi, J.; Persico, M. Molecular Interactions in Solution: 17602. An Overview of Methods Based on Continuous Distributions of the (28) (a) Baldwin, M. J.; Root, D. E.; Pate, J. E.; Fujisawa, K.; Solvent. Chem. Rev. 1994, 94, 2027. (b) Amovilli, C.; Barone, V.; Kitajima, N.; Solomon, E. I. Spectroscopic studies of side-on Cammi, R.; Cances, E.; Cossi, M.; Mennucci, B.; Pomelli, C. S.; peroxide-bridged binuclear copper(II) model complexes of relevance Tomasi, J. Recent advances in the description of solvent effects with to oxyhemocyanin and oxytyrosinase. J. Am. Chem. Soc. 1992, 114, the polarizable continuum model. Adv. Quantum Chem. 1998, 32, 227. 10421. (b) Pidcock, E.; Obias, H. V.; Abe, M.; Liang, H.-C.; Karlin, K. (42) (a) Trasatti, S. The absolute electrode potential: an explanatory D.; Solomon, E. I. Spectroscopic and Theoretical Studies of note (Recommendations 1986). Pure Appl. Chem. 1986, 58, 955. Oxygenated Dicopper(I) Complexes Containing Hydrocarbon- (b) Persson, I. Solvation and complex formation in strongly solvating Linked Bis[2-(2-pyridyl)ethyl]amine Units: Investigation of a solvents. Pure Appl. Chem. 1986, 58, 1153. Butterfly [Cu2(μ-η2:η2)(O2)]2+ Core. J. Am. Chem. Soc. 1999, 121, (43) Llunell, M.; Casanova, D.; Cirera, J.; Alemany, P.; Alvarez, S. 1299. SHAPE, version 2.0; Barcelona, 2010. (29) Ottenwaelder, X.; Rudd, D. J.; Corbett, M. C.; Hodgson, K. O.; (44) McCreery, R. Raman Spectroscopy for Chemical Analysis; John Hedman, B.; Stack, T. D. P. Reversible O−O bond cleavage in Wiley & sons, Inc.: New York, 2000; Chapter 10. copper-dioxygen isomers: impact of anion basicity. J. Am. Chem. Soc. (45) https://www.effemm2.de/spectragryph/. 2006, 128, 9268. (30) Funahashi, Y.; Nishikawa, T.; Wasada-Tsutsui, Y.; Kajita, Y.; Yamaguchi, S.; Arii, H.; Ozawa, T.; Jitsukawa, K.; Tosha, T.; Hirota, S.; Kitagawa, T.; Masuda, H. Formation of a Bridged butterfly-type μ- η2:η2-peroxo dicopper core structure with a carboxylate group. J. Am. Chem. Soc. 2008, 130, 16444. (31) Gagne, R. R.; Koval, C. A.; Lisensky, G. C. Ferrocene as an internal standard for electrochemical measurements. Inorg. Chem. 1980, 19, 2854. (32) Walton, P. H.; Davies, G. J. On the catalytic mechanisms of lytic polysaccharide monooxygenases. Curr. Opin. Chem. Biol. 2016, 31, 195. (33) CrysAlis PRO; Agilent Technologies Ltd: Yarnton, Oxfordshire, England, 2013. (34) Clark, R. C.; Reid, J. S. The analytical calculation of absorption in multifaceted crystals. Acta Crystallogr., Sect. A: Found. Crystallogr. 1995, 51, 887. (35) (a) Sheldrick, G. M. SHELXT - Integrated space-group and crystal-structure determination. Acta Crystallogr., Sect. A: Found. Adv. 2015, 71, 3. (b) Sheldrick, G. M. Crystal structure refinement with SHELXL. Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3−8. (36) Farrugia, L. J. WinGX suite for small-molecule single-crystal crystallography. J. Appl. Crystallogr. 1999, 32, 837. (37) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A.; Bloino, J.; Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J. V.; Izmaylov, A. F.; Sonnenberg, J. L.; Williams-Young, D.; Ding, F.; Lipparini, F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson, T.; Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.; Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Throssell, K.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Millam, J. M.; Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J. Gaussian 09, revision A.02; Gaussian, Inc.: Wallingford CT, 2016. (38) (a) Becke, A. D. Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti correlation- energy formula into a functional of the electron density. Phys. Rev. B 1988, 37, 785. (39) Hay, P. J.; Wadt, W. R. Ab initio effective core potentials for molecular calculations. Potentials for K to Au including the outermost core orbitals. J. Chem. Phys. 1985, 82, 299. (40) (a) Hariharan, P. C.; Pople, J. A. The influence of polarization functions on molecular orbital hydrogenation energies. Theor. Chim. Acta. 1973, 28, 213. (b) Francl, M. M.; Petro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon, M. S.; DeFrees, D. J.; Pople, J. A. Self- consistent molecular orbital methods. XXIII. A polarization-type basis set for second-row elements. J. Chem. Phys. 1982, 77, 3654. 10740 DOI: 10.1021/acsomega.9b00785 ACS Omega 2019, 4, 10729−10740