A N₃S_(thioether)-Ligated Cu^II-Superoxo with Enhanced Reactivity

Abstract

Previous efforts to synthesize a cupric superoxide complex possessing a thioether donor have resulted in the formation of an end-on trans-peroxodicopper(II) species, [{(Ligand)Cu^II}₂(μ-1,2-O₂²⁻)]²⁺. Redesign/modification of previous N₃S tetradentate ligands has now allowed for the stabilization of the monomeric, superoxide product possessing a S_(thioether)-ligation, [(^DMAN₃S)Cu^II(O₂^•−)]⁺ (2^S), as characterized by UV-vis and resonance Raman (rR) spectroscopies. This complex mimics the putative Cu^II(O₂^•−) active species of the copper monooxygenase PHM and exhibits enhanced reactivity towards both O-H and C-H substrates in comparison to close analogues [(L)Cu^II(O₂^•−)]⁺, where L contains only nitrogen donor atoms. Cu-S_(thioether) ligation with its weaker donor ability (relative to an N-donor) are demonstrated by comparisons to the chemistry of analogue compounds.

The copper monooxygenases peptidylglycine-α-hydroxylating monooxygenase (PHM) and dopamine-β-monooxygenase (DβM) possess a dicopper active site, but it is “noncoupled”; the two copper ions are about 11 Å apart.¹ Extensive biochemical and biophysical research has shown that one copper ion (designated Cu_H or Cu_A) receives and passes electron reducing equivalents to the His₂Met N₂S_(thioether) ligated Cu_M (or Cu_B) center, where O₂ and substrate binding occur. Recent computational analyses² lead to the hypothesis that an initial O₂-adduct, a Cu_M centered cupric superoxide {Cu^II(O₂^•−)} moiety, forms from oxygenation of the fully reduced (Cu^I…Cu^I) enzyme. This species performs the H-atom abstraction of the peptide prohormone substrate (in PHM) leading to C–H hydroxylation and formation of the product hormone. Other reaction mechanisms suggesting other possible reactive intermediates being responsible for substrate C–H attack have been proposed.³ In support of the importance of the Cu^II(O₂^•−) reaction intermediate’s involvement, a crystal structure obtained by Amzel and co-workers in the presence of a substrate inhibitor, reveals dioxygen bound to Cu_M in an end-on fashion,^1b as depicted in Scheme 1a.

Scheme 1.

Undoubtedly, the thioether ligand plays a critical role in determining the electronic structure and functions of Cu_M site leading to C-H bond activation.² However, the precise role of Met coordination and the actual PHM reaction mechanism has yet to be fully elucidated.

Within the sub-field of copper-dioxygen synthetic bioinorganic chemistry, one long-standing goal has been to produce Cu^II(O₂^•−) species that can be studied in detail, and to this end several cupric-superoxo complexes have been reported with ligands containing either three or four N-atoms.⁴ In an attempt to mimic the active site donors of PHM and DβM, considerable effort has been devoted toward generating thioether containing tridentate or tetradentate ligands to study their Cu^I/O₂ reactivity.⁵ In our own efforts we have been able to generate binuclear peroxodicopper complexes with S_(thioether) ligation,⁶ but there has been no report of a mononuclear Cu^II(O₂^•−) species coordinated by a S_(thioether) donor. Herein, for the first time we present the spectroscopic evidence and reactivity of the new mononuclear cupric superoxo complex, (^DMAN₃S)Cu^II(O₂^•−) (2^S) (Scheme 1b) possessing thioether S-ligation that exhibits enhanced reactivity towards both O-H and C-H substrates.

This new ligand, ^DMAN₃S,⁷ differs from our previously reported N₃S ligands in two important ways. First, it possesses highly electron-rich dimethylamino (DMA)⁸ groups at the para position of the two pyridyl donors, a strategy from previously stabilized a mononuclear Cu^II(O₂^•−) complex containing N₄ tripodal tetradentate ligand.^4c,4h,9 Second, the thioether moiety is capped with a bulkier o-methyl benzyl substituent to slow dimerization relative to previous designs.^6d Treatment of ^DMAN₃S with [Cu^I(CH₃CN)₄]B(C₆F₅)₄ in THF, followed by pentane addition leads to the isolation of bright yellow powders with formula [(^DMAN₃S)Cu^I]B(C₆F₅)₄ (1).⁷ Single crystals could be grown from 2-methyltetrahydrofuran (MeTHF)/pentane at RT under Ar. As shown in Figure 1a, the cuprous complex is a four-coordinate monomer ligated by two pyridyl, one tertiary amine, and one thioether atom in a distorted pyramidal geometry providing a very open site for potential dioxygen binding. The Cu-S bond length in 1 (2.2327(5) Å) falls in the usual range (2.2 – 2.44 Å) for Cu^I–S_(thioether) complexes.⁶ Many previously studied N₃S_(thioether)-Cu^I complexes form dimers in the solid state,^6d and the incorporation of the larger o-methylbenzyl group may disfavor their formation.

Figure 1.

Displacement ellipsoid plots of the cations; (a) [(^DMAN₃S)Cu^I]⁺ (1) and (b) [(^DMAN₃S)Cu^II(H₂O)(ClO₄)]⁺ (3). Hydrogen atoms were removed for clarity.

Furthermore, the copper(II) complex [(^DMAN₃S)Cu^II (H₂O)(ClO₄)](ClO₄) (3) (Figure 1b) was generated and structurally characterized.⁷ A slightly distorted octahedral coordination is observed. The thioether donor group is found in an axial position with Cu-S bond distance of 2.6974(6) Å. Electron paramagnetic resonance (EPR) measurement on 3 reveals a standard axial spectrum of copper(II) complex.⁷

Bubbling O₂ into a MeTHF solution of 1 at −135 °C led to the immediate formation of a new species (λ_max = 425 nm) that rapidly (~50 s) converted to the thermodynamically more stable end-on trans-peroxodicopper(II) species [{(^DMAN₃S)Cu^II}₂ (μ-1,2-O₂²⁻)]²⁺ (2^P) (Figure 2, top) exhibiting strong UV-vis absorptions at λ_max = 534 (6500 M⁻¹cm⁻¹) and 604 (6600 M⁻¹cm⁻¹) nm (Figure 2, bottom) consistent with the feature of previously synthesized S_(thioether)-ligated (μ-1,2-peroxo)Cu^II₂.⁶ The formulation of 2^P is confirmed by rR spectroscopy (ν_O–O = 821 cm⁻¹, ν_Cu–O = 547 cm⁻¹).⁷ Although the initially formed species has a short lifetime, the characteristic λ_max value of 425 nm and the transformation to a more stable (μ-1,2-peroxo)Cu^II₂, strongly suggest it is the S-ligated cupric superoxo species [(^DMAN₃S)Cu^II(O₂^•−)]⁺ (2^S).

Figure 2.

Low-temperature UV-vis absorption spectra of the reaction of 1 with O₂ at −135 °C in MeTHF (0.2 mM). The super-oxo product 2^S is observed immediately upon O₂ addition (red) and converted to the peroxo 2^P (t = 50 s, blue).

Interestingly, when polar or potentially hydrogen-bonding solvents such as acetone, ethanol, methanol, or trifluoroethanol (TFE) are mixed with MeTHF, [(^DMAN₃S)Cu^II–OO^•−]⁺ (2^S) is much more persistent than is found in MeTHF only (Figure 3a; λ_max = 418, 605, 743 nm). There appears to be a shift of the equilibrium constant to the mononuclear superoxide over the binuclear trans-peroxo complex. This result could be due to the formation of a hydrogen bond between an O-atom of the superoxide ligand and the TFE solvent, as observed in a ‘classical’ cobalt(III)-superoxo complex with TFE.¹⁰

Figure 3.

(a) Low-temperature (−135 °C) UV-vis absorption spectrum of 2^S (containing ~11% 2^P, λ_max = 526 nm, ε = 6500)¹² as recorded ~40 s after bubbling O₂ into a MeTHF:TFE (4:1) solution of 1 (0.098 mM), and (b) rR spectra of frozen 2^S (0.62 mM, λ_ex = 413 nm, 77 K) in MeTHF:TFE (4:1).⁷

The rR spectra of 2^S (λ_ex = 413 nm) reveal two dioxygen isotope sensitive vibrations (Figure 3b). An O–O stretch is observed at 1117 cm⁻¹ which shifts to 1056 cm⁻¹ upon ¹⁸O₂ substitution (Δ¹⁸O₂ = −61 cm⁻¹) and a Cu–O stretch at 460 cm⁻¹ (Δ¹⁸O₂ = −20 cm⁻¹). The energy and isotope shifts of these vibrations confirm the assignment of 2^S as an end-on superoxide species. In fact, these parameters closely match those for previously described cupric-superoxo complexes with tripodal tetradentate N₄ ligands (^DMAtmpa and ^DMMtmpa), (Table 1) both having ν_O–O = 1121 cm ^−1.4c,4h The species 2^S is assigned as having an S = 1 (triplet) ground state by EPR spectroscopy.^7,11

Table 1.

Comparison of LCu^II/LCu^I Redox Potentials and Reactivity of Derived [(L)Cu^II(O₂^•−)]⁺ Complexes.

	DMAtmpa4c	DMMtmpa4h	DMAN3S
p-OMe-DTBP a	No b	No b	Yes
AcrH2 a	No b	No b	Yes
E1/2, mV c	−700	−570	−470

^aReaction comparisons of [(L)Cu^II(O₂^•−)]⁺ complexes (MeTHF:TFE (4:1) at −135 °C) toward substrates.

^bReactions do occur, but only above −100 °C,

^cE_1/2 (vs Fc⁺/Fc) [(L)Cu^II(solvent)](ClO₄)₂ complexes as determined by cyclic voltammetry.⁷

In MeTHF:TFE (4:1 v:v), an approximately 4:1 (mol/mol) equilibrium mixture of (2^S)/(2^P)¹² forms within ~150 s following oxygenation of [(^DMAN₃S)Cu^I]⁺ (1) at −135 °C. This mixture shows minimal decomposition over an hour and thus, we could carry out reactivity studies of 2^S with the O–H and C–H substrates; 2,6-di-tert-butyl-4-methoxyphenol (p-OMe-DTBP) and N-methyl-9,10-dihydroacridine (AcrH₂). Pseudo first-order kinetic behavior was observed upon addition of p-OMe-DTBP (with t_1/2 = 3 min) or AcrH₂ (t_1/2 = 2 min) as monitored by the disappearance of the 418 nm UV-vis band of 2^S. Following workup at RT the products were analyzed as 2,6-di-tert-butyl-1,4-benzoquinone and 10-methyl-9-acridone, respectively.⁷ Independent observations demonstrate that the complex 2^P is unreactive toward these substrates. Warming up the 2^S solution does not lead to sulfur oxygenation which occurs by bis(μ-oxo)Cu^III₂ species.^6d

To assess the effect of the thioether donor in 2^S, we compared the reactivity of (L)Cu^II(O₂^•−) complexes (L = ^DMAtmpa or ^DMMtmpa) toward the same substrates.^4c,4h Notably, both cupric superoxide complexes showed no reactivity under identical conditions (Table 1), although they do react above −100 °C. Thus, 2^S with S_(thioether)-ligation is more reactive than the closely related N₄ superoxide compounds. The difference in reactivity is rationalized by the different donor strengths of the corresponding ligands. The substitution of a S_(thioether) for a N_(pyridine) donor decreases the ligand field strength, consistent with the Cu^II/I redox potentials for the separately isolated Cu^II complexes in which thioether coordination raises the reduction potential by 230 mV (Table 1).⁷ As a result, we hypothesize that 2^S is more electrophilic and hence better able to accept an H-atom from either an O–H or C–H substrate, as compared to the N₄ complexes.

The influence of the Cu–S interaction on the oxygenated products was further probed by comparing Cu/O₂ species with different donor atoms binding Cu. Three structurally analogous ligands were synthesized replacing the sulfur site with carbon, oxygen, and nitrogen (Scheme 2).⁷ The isolated Cu^I complex with tridentate ^DMAN₃ ligand exhibits a distinctive 382 nm absorption band⁷ upon O₂ addition at −135 °C in MeTHF. This absorption spectrum is consistent with a bis(μ-oxo)Cu^III₂ complex, which is known for many other products from low-temperature oxygenation of Cu^I-N₃ precursors.^6d,14 Under identical reaction conditions, oxygenation of the Cu^I(^DMAN₃O), possessing three N-donors and an ether O-atom, leads to the initial formation of a (μ-1,2-peroxo)Cu^II₂ which rapidly (~ 1 min) converts to a bis(μ-oxo)Cu^III₂ species with very similar UV-vis spectrum as for the case with ^DMAN₃.^6d,15 Thus, as seen before,^6d,16 the oxygen atom of the ether arm in ^DMAN₃O has an extremely weak to non-existent interaction with the copper ion in the oxygenated product. We find that for Cu^I(^DMAN₄) in MeTHF (Scheme 2c), both Cu^II–superoxo and (μ-1,2-peroxo)Cu^II₂ complexes are generated⁷ as in the case of ^DMAN₃S. This strongly suggests that the S_(thioether)-atom donor of ^DMAN₃S ligand is coordinated in both Cu^II–superoxo/μ-1,2-peroxo complexes, 2^S and 2^P. If it were not, a bis(μ-oxo)Cu^III₂ complex would prominently form.¹⁷

Scheme 2.

Oxygenated products of ligand-Cu^I complexes.¹³

In summary, [(^DMAN₃S)Cu^II(O₂^•−)]⁺ (2^S) is the first reported example of a cupric superoxo species supported by a S_(thioether) donor. This advance allowed us to determine that a superoxide species from an N₃S donor was more reactive towards O–H and C–H bonds than the corresponding N₄ complex. These results indicate that one role of the Met ligand in PHM and DβM is to activate the superoxide species for reaction by increasing their electrophilicity. The synthesis of this species will also allow us to further consider the role of thioether ligation on critical downstream O₂-reduced (and protonated) derivatives, such as Cu^II-hydroperoxo and to perform detailed kinetic analysis of the preliminary reactivity study presented here.