Nonheme iron-thiolate complexes as structural models of sulfoxide synthase active sites

Article abstract of DOI:10.1039/d0dt03403g

Two mononuclear iron(ii)-thiolate complexes have been prepared that represent structural models of the nonheme iron enzymes EgtB and OvoA, which catalyze the O2-dependent formation of carbon-sulfur bonds in the biosynthesis of thiohistidine compounds. The series of Fe(ii) complexes reported here feature tripodal N4 chelates (LA and LB) that contain both pyridyl and imidazolyl donors (LA = (1H-imidazol-4-yl)-N,N-bis((pyridin-2-yl)methyl)methanamine; LB = N,N-bis((1-methylimidazol-2-yl)methyl)-2-pyridylmethylamine). Further coordination with monodentate aromatic or aliphatic thiolate ligands yielded the five-coordinate, high-spin Fe(ii) complexes [FeII(LA)(SMes)]BPh4 (1) and [FeII(LB)(SCy)]BPh4 (2), where SMes = 2,4,6-trimethylthiophenolate and SCy = cyclohexanethiolate. X-ray crystal structures revealed that 1 and 2 possess trigonal bipyramidal geometries formed by the N4S ligand set. In each case, the thiolate ligand is positioned cis to an imidazole donor, replicating the arrangement of Cys- and His-based substrates in the active site of EgtB. The geometric and electronic structures of 1 and 2 were analyzed with UV-vis absorption and M?ssbauer spectroscopies in tandem with density functional theory (DFT) calculations. Exposure of 1 and 2 to nitric oxide (NO) yielded six-coordinate FeNO adducts that were characterized with infrared and electron paramagnetic resonance (EPR) spectroscopies, confirming that these complexes are capable of binding diatomic molecules. Reaction of 1 and 2 with O2 causes oxidation of the thiolate ligands to disulfide products. The implications of these results for the development of functional models of EgtB and OvoA are discussed. This journal is

Full text of DOI:10.1039/d0dt03403g

Dalton
Transactions
View Article Online
View Journal
PAPER
Nonheme iron–thiolate complexes as structural
models of sulfoxide synthase active sites†
Cite this: DOI: 10.1039/d0dt03403g
a
a
b
a
Danushka M. Ekanayake, Anne A. Fischer, Maya E. Elwood, Alexandra M. Guzek,
a
b
a
Sergey V. Lindeman, Codrina V. Popescu * and Adam T. Fiedler
*
Two mononuclear iron(II)–thiolate complexes have been prepared that represent structural models of the
nonheme iron enzymes EgtB and OvoA, which catalyze the O -dependent formation of carbon–sulfur
bonds in the biosynthesis of thiohistidine compounds. The series of Fe(II) complexes reported here feature
2
A
B
A
tripodal N
4
chelates (L and L ) that contain both pyridyl and imidazolyl donors (L = (1H-imidazol-4-yl)-N,
B
N-bis((pyridin-2-yl)methyl)methanamine; L = N,N-bis((1-methylimidazol-2-yl)methyl)-2-pyridylmethyl-
amine). Further coordination with monodentate aromatic or aliphatic thiolate ligands yielded the five-coor-
II
A
II
B
dinate, high-spin Fe(II) complexes [Fe (L )(SMes)]BPh
methylthiophenolate and SCy = cyclohexanethiolate. X-ray crystal structures revealed that 1 and 2 possess
trigonal bipyramidal geometries formed by the N S ligand set. In each case, the thiolate ligand is positioned
4 4
(1) and [Fe (L )(SCy)]BPh (2), where SMes = 2,4,6-tri-
4
cis to an imidazole donor, replicating the arrangement of Cys- and His-based substrates in the active site of
EgtB. The geometric and electronic structures of 1 and 2 were analyzed with UV-vis absorption and
Mössbauer spectroscopies in tandem with density functional theory (DFT) calculations. Exposure of 1 and 2
to nitric oxide (NO) yielded six-coordinate FeNO adducts that were characterized with infrared and electron
paramagnetic resonance (EPR) spectroscopies, confirming that these complexes are capable of binding dia-
Received 1st October 2020,
Accepted 9th November 2020
DOI: 10.1039/d0dt03403g
2
tomic molecules. Reaction of 1 and 2 with O causes oxidation of the thiolate ligands to disulfide products.
rsc.li/dalton
The implications of these results for the development of functional models of EgtB and OvoA are discussed.
Two sulfoxide synthases have been reported to date, EgtB and
OvoA, which catalyze the key steps in the production of
Introduction
Several mononuclear nonheme iron (MNHI) enzymes involved ergothioneine (Egt) and ovothiol A, respectively, from Cys and
1
6–21
in O
substrates. The best-studied example is cysteine dioxygenase in formation of a S–C bond at the 2-position of the imidazole
CDO), which initiates Cys catabolism by catalyzing its oxidation ring, whereas the OvoA reaction occurs at the 4-position of the
2
activation employ L-cysteine (Cys) or its derivatives as His building blocks (Scheme 1).
The EgtB reaction results
1–3
(
4–6
to L-cysteine sulfinic acid (Scheme 1). Other members of the ring. The precise functions of Egt and ovothiol A within organ-
thiol dioxygenase family include cysteamine dioxygenase isms are not fully understood, but several reports have empha-
7
,8
9–11
15,22–25
(
ADO) and 3-mercaptopropionate dioxygenase (MDO).
In sized their antioxidant properties.
addition to S-dioxygenation, certain bacterial MNHI enzymes
Crystallographic studies have revealed close similarities
catalyze the O -dependent formation of S–C bonds in the bio- between the active-site structures of EgtB and CDO. In both
2
12–15
syntheses of thiohistidine compounds.
These enzymes are cases, the mononuclear iron(II) center is facially ligated by
referred to as sulfoxide (SO) synthases because the reaction three His residues (the 3His triad), resulting in three cis-labile
2
6–29
combines S–C bond formation with monooxygenation of the sites for simultaneous binding of substrate(s) and O2.
The
S-atom, thereby utilizing all four oxidizing equivalents of O
2
.
structure of Mn-substituted EgtB suggests that the γ-glutamyl
cysteine (γGC) and N-α-trimethyl histidine (TMH) substrates
coordinate directly to iron (Fig. 1), and do not occupy outer-
2
7
a
sphere binding sites as originally supposed. Seebeck and co-
workers have explored the importance of a second-sphere tyro-
sine residue (Tyr377) in the catalytic function of EgtB from
Mycobacterium thermoresistibile. Substitution of Tyr377 by Phe
converts EgtB into a thiol dioxygenase that favors dioxygena-
Department of Chemistry, Marquette University, Milwaukee, Wisconsin 53201, USA.
E-mail: adam.fiedler@marquette.edu
b
Department of Chemistry, The College of Arts and Sciences, University of
St. Thomas, St. Paul, Minnesota 55105, USA. E-mail: ewba2202@stthomas.edu;
Fax: (+1) 414-288-7066
†
Electronic supplementary information (ESI) available. CCDC 2034830–2034833.
For ESI and crystallographic data in CIF or other electronic format see DOI:
0.1039/d0dt03403g
3
0
tion of γGC instead of S–C bond formation. Interestingly, the
1
dioxygenase activity of Y377F EgtB still requires TMH, indicat-
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Scheme 1
Fig. 1 Metal coordination environment in the active site of Mn-substi-
27
tuted EgtB, as revealed by X-ray crystallography (PDB: 4X8D). The Mn
II) center is coordinated by the 3His triad, N,N-dimethyl histidine (DMH),
γGC, and solvent-derived H O. Dioxygen is presumed to bind to the site
occupied by H O.
Scheme 2
(
2
2
direct the S–C bond formation towards the 2- and 4-positions,
respectively, of the imidazole ring (Scheme 1).
ing that the same quaternary EgtB : Fe : γGC : TMH complex is
The development of synthetic SO synthase models has the
formed in both pathways. Mutation of the analogous Tyr potential to elucidate key aspects of both enzymatic structure
residue in OvoA (Tyr417) has also been shown to alter the and mechanism. While numerous CDO models have been
3
1
36–44
product ratio in favor of cysteine sulfinic acid.
reported by us and others,
to the best of our knowledge
The catalytic cycles of SO synthases remain poorly iron–thiolate complexes designed to mimic the structure of
1
5
defined. An ordered mechanism is presumed in which the substrate-bound SO synthases are currently lacking in the lit-
His- and Cys-based substrates coordinate to the Fe(II) center erature. In this manuscript, we report the generation of two
prior to O
transfer from an outer-sphere Tyr residue facilitates formation the EgtB and OvoA active sites prior to O
of an iron(III)-hydroperoxo species, which carries out sulfoxida- in Scheme 2, these complexes feature tetradentate N4 ligands
2
activation. In one proposed mechanism, H-atom high-spin Fe(II) complexes that serve as structural models of
2
binding. As shown
3
2
A
B
tion of the substrate after formation of the S–C bond. This (L and L ) consisting of one or two pendant imidazole
mechanism is supported by QM/MM calculations recently pub- donors. The point of attachment for the imidazole donor is
3
3
A
B
lished by de Visser and coworkers. In contrast, the compu- different in the L and L ligands in an effort to control the
3
4
35
tational studies of Liao and Liu favor an alternative mecha- regioselectivity of the C–S bond formation reaction.
nism in which sulfoxidation occurs prior to S–C bond for- Coordination of a monodentate aryl or alkyl thiolate ligand
mation. Unfortunately, to date, it has not been possible to yields the desired five-coordinate N4S environment in com-
observe intermediates derived from reaction of the enzymes plexes 1 and 2, as confirmed by X-ray crystallography. In these
with O
2
, which would allow one to discriminate between these models, the facial amine and pyridyl/imidazolyl N-donors
mechanistic alternatives. It is also unclear how EgtB and OvoA account for the 3His triad, while a pendant imidazole occupies
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
a position adjacent to the thiolate (Scheme 2). This arrange- ware WMOSS4 (Ion Prisecaru, http://www.wmoss.org) and
ment provides a vacant coordinate site cis to both the imid- SpinCount (Dr Michael Hendrich, Carnegie Mellon University).
azole and thiolate ligands, similar to SO synthase active sites.
II
A
Synthesis of [Fe (L )(CH CN) ](OTf) (3)
3
2
2
Indeed, we demonstrate here that both SO synthase mimics
are capable of binding small molecules like NO (a surrogate of Equimolar amounts of the L ligand (0.412 g, 1.48 mmol) and
O ), thereby yielding metastable intermediates that have been Fe (OTf) (0.522 g, 1.48 mmol) were added to CH CN (20 mL).
2 2 3
A
II
characterized with UV-vis absorption, electron paramagnetic The resulting reddish-brown solution was stirred for 3 h until all
resonance (EPR), and infrared spectroscopies. These results of the solid material had dissolved. The solution was then filtered
constitute a promising first step towards the development of and concentrated under vacuum. Layering with Et
2
O provided
structural and functional mimics of EgtB and OvoA.
brown crystalline product after one day. Yield = 0.70 g (66%).
−1
Anal. Calcd for C H F FeN O S (M = 715.4 g mol ): C, 36.93;
22
23
6
7
6 2
W
H, 3.24; N, 13.70. Found: C, 36.61; H, 3.18; N, 13.45. UV-vis [λmax
,
nm (ε, M− cm ) in CH
1
−1
CN]: 387 (2700), 560 (130). FTIR (cm ,
−1
Experimental section
3
1
solid): ν = 3236 [ν(N–H)], 1607, 1445, 1233, 1153. H NMR (CDCl ,
3
Materials and methods
4
5
00 MHz): δ = −6.2 (s, 2H, pyr-H), 16.9 (br s, 1H), 34.0 (s, 1H),
3.2 (s, 2H, pyr-H), 53.6 (s, 2H, pyr-H), 78.6 (br s, 2H, –CH –), 82.6
Unless otherwise noted, solvents and reagents were purchased
from commercial sources and used as received. Some solvents
2
(s, 1H), 85.6 (br s, 2H, –CH –), 99.3 (br s, 2H, –CH –), 125 (br s,
2
2
19
2 2 3 2 2
(CH Cl , CH CN, THF, Et O) were dried over CaH or NaH, fol-
2H, pyr-H). F NMR (CD
3
CN, 400 MHz): δ = −76.6 ppm (OTf). μeff
Cl ), 3.9 μ (CD CN).
lowed by distillation. Multiple freeze–pump–thaw cycles were
performed to remove oxygen and other gases, and the solvents
were stored in the glovebox over activated molecular sieves.
Sodium cyclohexanethiolate was prepared by the reaction of
the thiol with NaH in THF and isolated as a white powder. The
Fe(II) complexes were synthesized and handled under an inert
atmosphere using a Vacuum Atmospheres Omni-Lab glovebox
(Evans method) = 4.9 μ
B
(CD
2
2
B
3
II
A
Synthesis of [Fe (L )(SMes)]OTf (1-OTf)
Sodium methoxide (0.016 g, 0.30 mmol) and 2,4,6-trimethyl-
thiophenol (0.044 g, 0.29 mmol) were dissolved in THF and
stirred for 30 minutes, followed by dropwise addition of 3
(
0.212 g, 0.297 mmol) in THF. The resulting orange mixture was
A
equipped with a freezer set to −30 °C. The L chelate was gen-
erated using a previously-reported procedure.
stirred for 90 min, filtered, and the volume reduced under
vacuum. Vapor diffusion of pentane into this solution yielded
4
5
yellow crystals of 1-OTf suitable for X-ray crystallography. Yield =
Physical methods
19
0
.13 g (69%). F NMR (CD CN, 400 MHz): δ = −78.7 ppm (OTf).
3
UV-vis absorption spectra were collected with an Agilent 8453
diode array spectrometer equipped with a Unisoku Scientific
II
A
Synthesis of [Fe (L )(SMes)]BPh (1)
4
Instruments (Osaka, Japan) cryostat for low temperature Complex 1 was prepared by dissolving 2,4,6-trimethylthiophenol
1
measurements. H NMR spectra were measured on a Varian (0.025 g, 0.17 mmol) and sodium methoxide (0.009 g,
4
00 MHz spectrometer featuring variable-temperature capabili- 0.17 mmol) in THF. Dropwise addition of 3 (0.119 g,
ties. This instrument was also used for magnetic susceptibility 0.166 mmol) in THF gave an orange solution that was stirred
measurements carried out using the Evans NMR method. for 1.5 h. The solution was filtered, the solvent was removed
Routine baseline corrections were applied using the SpinWork4 under vacuum, and the orange residue dissolved in MeOH
software program. Elemental analyses were performed at (5 mL). Addition of NaBPh (0.057 g, 0.17 mmol) gave a yellow
4
Midwest Microlab, LLC in Indianapolis, IN. Infrared (IR) spectra precipitate that was allowed to settle to the bottom. The solvent
of solid samples were measured with a Thermo Scientific was decanted and the solid was washed with MeOH and Et O.
2
Nicolet 670 FTIR spectrophotometer. Mass spectra were col- Drying under vacuum provided 1 as a yellow powder that was
lected using a GC-MS instrument consisting of an Agilent used without further purification. Yield = 0.080 g (60%). Anal.
= 805.7 g mol− ): C, 73.05; H, 6.01;
1
5
973 mass spectrometer attached to a 6850 gas chromatography Calcd for C49
H48BFeN S (M
5 W
system. X-band EPR spectra were measured using a Bruker EMX N, 8.69. Found: C, 72.80; H, 5.94; N, 8.78. UV-vis [λmax, nm (ε,
−
1
−1
instrument equipped with an ER4112 SHQ resonator. EPR data
M
2 2
cm ) in CH Cl ]: 356 (1400), 402 (1300), 1010 (30). FTIR
−
1
were collected at liquid He temperatures using a ColdEdge/ (cm , solid): ν = 3312 [ν(N–H)], 1607, 1477, 1262, 1155, 1050,
Bruker RDK-408 Stinger recirculating cryocooler. Mössbauer 1030 cm− . H NMR (CD CN, 400 MHz): δ = −9.8 (s, 2H, pyr-H),
1
1
3
spectra were recorded on a closed-cycle refrigerator spectro- 6.86 (s, 4H, BPh
meter, model CCR4K (SeeCo, Edina, MN) equipped with a 0.07 2H), 24.1 (br s, 6H, –CH
T permanent magnet, maintaining temperatures between 6 and (s, 1H), 56.4 (s, 2H, pyr-H), 65.2 (s, 2H, pyr-H), 68.5 (s, 1H), 72.0
00 K. The samples consisted of either solid powders sus- (br s, 4H, –CH –), 91.3 (br s, 2H, –CH –), 128.2 (br s, 2H, pyr-H).
), 4.7 μ (CD CN).
4 4 4
), 7.02 (s, 8H, BPh ), 7.29 (s, 8H, BPh ), 21.2 (s,
3 3
), 30.7 (s, 3H, –CH ), 33.2 (s, 1H), 51.7
3
2
2
pended in degassed mineral oil or butyronitrile solutions of
μ
eff (Evans method) = 4.8 μ
B
(CD Cl
2 2
B
3
high concentration (∼50 mM). These samples were placed in
B
Synthesis of L
1
.00 mL Delrin cups and frozen in liquid nitrogen. The isomer
shifts are quoted at 6 K with respect to an iron metal standard 1-Methyl-2-imidazolcarboxaldehyde (0.193 g, 1.75 mmol) was
at 298 K. The Mössbauer spectra were analyzed using the soft- dissolved in MeOH (20 mL) in a round bottom flask. To this
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
B 2 2
solution was added N-(1-methyl-1H-imidazol-2-ylmethyl)pyrid- 1H), 78.4 (br s, 1H), 121.7 (br s, 1H). μeff = 5.0 μ (CD Cl ), 4.4
4
6
2
-ylmethylamine (0.356 g, 1.75 mmol) in MeOH (30 mL), fol- μ (CD CN).
B 3
3
lowed by addition of glacial acetic acid (2 mL). NaBH CN was
added to the cooled reaction mixture, which was then stirred
Crystallographic studies
under inert atmosphere for 24 h at room temperature. The Complexes 1-OTf and 2–4 were structurally characterized using
reaction mixture was acidified with conc. HCl and the solvent an Oxford Diffraction SuperNova kappa-diffractometer (Rigaku
removed under rotary evaporation. The resulting residue was Corp.) equipped with both Cu and Mo X-ray sources. Table S1†
dissolved in CH Cl (50 mL) and washed multiple times with a provides details concerning the data collection, analysis, and
2
2
saturated Na
bined, dried over MgSO
2
CO
3
solution. The organic fractions were com- crystallographic parameters for each complex. Structures were
4
7,48
4
, and filtered. Evaporation gave a light solved using the SHELXS program
and refined with the
yellow solid that was used without further purification. Yield = SHELXL program found within the Olex2 crystallographic
1
49
0
(
(
.31 g (65%). H NMR (CDCl
s, 4H), 3.83 (s, 2H), 6.78 (s, 2H), 6.92 (s, 2H), 7.16 (t, 1H), 7.27 obtained from the Cambridge Crystallographic Data Centre
m, 1H), 7.61 (t, 1H), 8.54 (d, 1H) ppm. using the deposition numbers: 2034830 (1-OTf), 2034831 (2),
034832 (3), and 2034833 (4).† In the structure of complex 2,
3
, 300 MHz): δ = 3.35 (s, 6H), 3.71 package.
Additional information in CIF format can be
2
II
B
Synthesis of [Fe (L )(CH CN)(OTf)](OTf) (4)
B
3
the L ligand is rotationally disordered with ∼22% interchange
B
Equimolar amounts of the L ligand (0.178 g, 0.600 mmol) between one imidazole donor and the pyridyl ring. The cyclo-
II
and Fe (OTf)
2
(0.214 g, 0.605 mmol) were combined in CH
3
CN hexyl moiety of the thiolate ligand is disordered in a similar
(20 mL) and stirred for 3 h. The resulting dark brown solution 22 : 78 ratio as a result.
was filtered and the volume reduced under vacuum. Layering
with Et O yielded a brown crystalline product. Yield = 0.35 g
78%). Anal. Calcd for C20 FeN (M
mol ): C, 34.74; H, 3.35; N, 14.18. Found: C, 34.98; H, 3.34; N, tion of NaNO (or Na NO ) with ascorbic acid and CuCl cata-
Reactivity with O and NO
2
2
(
H
23
F
6
7
O
6
S
2
W
= 691.4 g Nitric oxide (NO) gas was generated in situ by the aqueous reac-
−
1
15
2
2
2
−
1
−1
50
1
3
4.25. UV-vis [λmax, nm (ε, M cm ) in CH CN]: 340 (400). lyst under an Ar atmosphere. The resulting NO gas was trans-
−
1
1
FTIR (cm , solid): ν = 1608, 1502, 1233, 1153, 1022. H NMR ferred via cannula into the reaction vial containing the iron
(
CD CN, 400 MHz): δ = −13.8 (s, 1H, pyr-H), 14.5 (s, 6H, complex. The reaction of complex 1 with O was performed at
3
2
–NCH
3
), 40.6 (br s, 2H), 50.6 (s, 1H, pyr-H), 56.7 (s, 1H, pyr-H), room temperature in CH
2
Cl
2
. After stirring for one hour, the
6
3.4 (s, 2H), 78.8 (br s, 2H, –CH
2
–), 90.7 (br s, 2H, –CH –), 93.8 solution was filtered through a plug of silica to remove ionic
2
1
9
(s, 2H, –CH –), 112.8 (br s, 1H, pyr-H). F NMR (CD CN, species, including iron-containing by-products of the reaction.
2
3
4
00 MHz): δ = −77.9 ppm (OTf). μeff (Evans method) = 4.9 μ
Cl ), 4.2 μ (CD CN).
B
The solvent was evaporated under vacuum and the resulting
residue was taken up in CDCl . Yields were calculated using
naphthalene as an internal standard (Fig. S7†). Yield of bis
(CD
2
2
B
3
3
II
B
Synthesis of [Fe (L )(SCy)]BPh (2)
1
4
(
2,4,6-trimethyl-phenyl)disulfide (MesS–SMes) = 79%; H NMR
A solution of 4 (0.144 g, 0.21 mmol) in MeOH (10 mL) was (400 MHz, CDCl ): δ = 2.21 (12H, o-CH ), 2.26 (6H, p-CH ), 6.85
added slowly to a methanolic solution of sodium cyclohexa- (4H, m-H-Ar) ppm. GC-MS data (m/z): {M} Calc. for C H S
22 2
3
3
3
+
1
8
+
nethiolate (0.029 g, 0.21 mmol). The reddish-brown solution 302.50, found 302; {M–SMes} Calc. for C
was stirred for 2 h, followed by addition of NaBPh (0.071 g, 151. The reaction of 2 with O was carried out in CH
.21 mmol) to give a yellow precipitate. After allowing the pre- the products were isolated by addition of acidified H O (pH =
9
H
11S 151.25, found
4
2
3
CN and
0
2
cipitate to settle to the bottom of the flask, the solution 2–3), followed by stirring for 2 hours. The aqueous layer was
removed by pipette and the remaining solvent was removed extracted with CH Cl and methyl tert-butyl ether. The organic
under vacuum. The material was washed with MeOH and fractions were combined and filtered through a plug of silica
Et O, dried under vacuum, and used without further purifi- gel. The solvent was removed under vacuum, and the resulting
2
cation. Yield = 120 mg (73%). Crystals suitable for X-ray crystal- residue was taken up in CD Cl . H NMR data of the reaction
2
2
2
1
2
lography were obtained by vapor diffusion of Et O into a con- products were analyzed by comparison to published spectra
2
centrated 1,2-dichloroethane (DCE) solution. Anal. Calcd for and/or spectra measured with commercially-available material
−
1
C
1
46
H
51BFeN
0.68. Found: C, 67.41; H, 6.73; N, 10.51. The discrepancy in an internal standard. Two thiolate-derived products were
the carbon value is due to small amounts of NaOTf that were observed for the reaction of 2 with O . Dicyclohexyldisulfide
6 W
S (M = 786.7 g mol ): C, 70.23; H, 6.53; N, (see Fig. S8†). Yields were calculated using p-benzoquinone as
2
1
9
−1
1
detected using F NMR spectroscopy. UV-vis [λmax, nm (ε, M
(CyS–SCy): yield = 35% (average of three trials). H NMR
−
1
cm ) in CH Cl ]: 328 (2530), 369 (sh), 425 (sh), 1030 (50). (400 MHz, CD Cl ): δ = 1.2–1.4 (10H), 1.61 (2H), 1.76 (4H), 2.02
2
2
2
2
FTIR (cm− , solid): ν = 3044, 2918, 1595, 1484, 1430, 1265, (4H), 2.68 (2H, S–CH) ppm. GC-MS data (m/z): {M} Calc. for
1
+
1
+
1
1
2
152, 957, 704. H NMR (CD
2
Cl
2
, 400 MHz): δ = −7.3 (s, 1H),
12 22 2 6 12 2
C H S 230.43, found 230; {M–Cy + H} Calc. for C H S
.75 (s, 1H), 3.28 (s, 1H), 3.46 (s, 1H), 3.77 (s, 1H), 4.21 (br s, 148.29, found 148. Dicyclohexyl thiosulfonate (CyS–S(O) Cy):
2
1
H), 4.72 (s, 2H), 6.11 (s, 2H), 6.69 (s, 4H, BPh
), 7.22 (s, 8H, BPh ), 14.8 (s, 6H, –NCH ), 15.5 (br s, 1H), CD
1.8 (s, 2H), 53.8 (br s, 3H), 57.2 (s, 2H), 60.1 (s, 2H), 64.5 (s, (1H, O S–CH). GC-MS data (m/z): {M} calc. for C H O S
12 22 2 2
4
), 6.80 (s, 8H, yield = 14% (average of three trials). H NMR (400 MHz,
BPh
4
4
4
3
2
Cl ): δ = 1.2–2.2 (18H), 2.30 (2H), 3.08 (1H, S–CH), 3.44
2
+
2
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
+
2
2 6
62.43, found 262; {M–CySO } calc. for C H11S 115.22, found
1
15.
Computational methods
Density functional theory (DFT) calculations were carried out
using the ORCA 4.0 software package developed by Dr F. Neese
5
1,52
(MPI-CEC).
Unless otherwise noted, all calculations uti-
lized Becke’s three-parameter hybrid functional for exchange
5
3,54
and the Lee–Yang–Parr correlation functional (B3LYP).
Geometry optimizations employed the Karlsruhe valence
double-ζ basis set with polarization functions (def2-SVP),
whereas single-point calculations of spectroscopic features
used the corresponding valence triple-ζ basis set combined
Fig. 2 Thermal ellipsoid plots (50% probability level) derived from X-ray
crystal structures of 3 (left) and 4 (right). Hydrogen atoms and non-
coordinating counteranions have been omitted for clarity. Selected
interatomic distances (Å) for 3: Fe–N1 = 2.010(3), Fe–N2 = 1.948(3), Fe–
5
5
with polarization functions (def2-TZVP). The resolution of
5
6
identity and chain of sphere (RIJCOSX) approximations were N3 = 1.958(4), Fe–N5 = 1.960(4), Fe–N6 = 1.930(3), Fe–N7 = 1.953(3).
applied in conjunction with the appropriate auxiliary basis Selected interatomic distances (Å) for 4: Fe–N1 = 2.314(2), Fe–N2 =
5
7,58
2.167(2), Fe–N4 = 2.116(2), Fe–N6 = 2.124(2), Fe–N7 = 2.117(2), Fe–O1 =
.197(2).
sets.
Solvent effects in geometry optimizations were
2
accounted for using the conductor-like polarizable continuum
5
9
3
model (C-PCM). Calculations of CH CN-bound complexes
used acetonitrile as the solvent, while optimizations of the Fe/
NO adducts used CH Cl . Time-dependent DFT (TD-DFT) cal-
2
+
(
CH
3
CN)
2
]
cation (the two triflate counteranions are non-
2
2
coordinating). The Fe–N bond distances fall between 1.93 and
2.01 Å, typical of low-spin Fe(II) complexes with related ligand
sets. In contrast, the [Fe (L )(CH CN)(OTf)] cation of 4 fea-
3
tures a coordinated triflate ligand trans to one of the imidazole
donors (Fig. 2). The increase in average Fe–N distance to
2.17 Å in 4 is indicative of a high-spin Fe(II) center. A notable
feature of this structure is the unusually long bond distance of
.314 Å between Fe1 and the central amino donor (N1). For
comparison, Fe–Namine bond distances in high-spin Fe(II) com-
culations employed the cam-B3LYP range-separated hybrid
6
0
functional, which is known to provide excellent agreement
between experimental and computed absorption spectra for
6
7
II
B
+
3
6,61
CDO and iron–thiolate complexes.
and intensities were computed for 40 excited states via the
Tamm-Dancoff approximation.
Absorption energies
6
2,63
Isosurface plots of mole-
cular orbitals and electron density difference maps were pre-
2
pared using the ChemCraft program. For calculations of
6
4
Mössbauer parameters, the “core properties” with extended
6
5
plexes supported by substituted tris(2-pyridylmethyl)amine
polarization [CP(PPP)] basis set and a high resolution grid
with an integration accuracy of 7.0 was used for the Fe atom;
the def2-TZVP basis set was used for all other atoms. Isomer
67–69,70
(
TPA) ligands generally fall between 2.15 and 2.22 Å.
Reaction of 3 with 2,4,6-trimethylthiophenol in the pres-
ence of base in THF yielded the desired iron(II)–thiolate
complex [Fe (L )(SMes)]OTf (1-OTf). Counteranion metathesis
4 4
with NaBPh in MeOH provided [Fe (L )(SMes)]BPh (1) as a
yellow precipitate. Similarly, treatment of 4 with sodium cyclo-
hexanethiolate (NaSCy) in MeOH, followed by addition of
4 4
NaBPh , generated [Fe (L )(SCy)]BPh (2) as a yellow powder.
Q
shifts (δ) and quadrupole splittings (ΔE ) were derived from
II
A
4
1
linear correlation functions reported by Gordon et al. For
II
A
isomer shifts: δ = α(ρ(0) − C) + β, where ρ(0) is the computed
−
3
electron density at the Fe nucleus (in units of a.u. ), α =
3
−1
−1
−3
−
0.440 mm a.u. s , β = 2.104 mm s , and C = 11 813 a.u. .
II
B
For quadrupole splittings: ΔEQ = η(ΔEQ,DFT) − B , where
0
X-ray quality crystals of 1-OTf and 2 were obtained via methods
described in the Experimental section, but attempts to grow
suitable crystals of 1 were not successful.
ΔEQ,DFT is the DFT-computed quadrupole splitting, η = 0.840,
−
1
and B = −0.00193 mm s .
0
The X-ray structures of 1-OTf and 2 are shown in Fig. 3 and
key metric parameters are summarized in Table 1. Both com-
plexes display distorted trigonal–bipyramidal geometries (τ-
Results and discussion
Synthesis and X-ray crystal structures
7
1
values near 1.0) in which the thiolate (S1) and amine (N1)
donors occupy the axial positions, as evident by the S1–Fe–N1
angles near 180°. The pyridyl and imidazole donors are
arranged in the equatorial plane with N–Fe–N angles ranging
from 108 to 122°. The Fe–S bond distances of 2.36 and 2.30 Å
for 1-OTf and 2, respectively, are typical of high-spin iron(II)–
A
The synthesis of L was previously reported by Karlin and co-
4
5
B
2
workers in their studies of Cu(I)/O chemistry. The L sup-
porting ligand, which features two pendant imidazole donors,
was prepared by modifying the procedure described in Bowers
6
6
II
A
et al. Reaction of Fe (OTf)
2
with an equimolar amount of L
3
6–41
B
thiolate complexes,
and the Fe–S1–C1 bond angles are
or L in CH
3
CN yielded the precursor complexes 3 and 4,
sharply bent for both complexes (103° for 1-OTf, 112° for 2).
Comparison of 2 and 4 reveals that coordination of the cyclo-
hexanethiolate ligand further elongates the Fe–N1 bond by
0.05 Å, and the structure of 1-OTf also features a lengthy Fe–
respectively, in good yields. Brown crystals of 3 and 4 suitable
for X-ray diffraction experiments were generated by layering
CH
3
CN solutions with diethyl ether. As shown in Fig. 2, the
II
A
X-ray structure of
3
contains a six-coordinate [Fe (L )
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 3 Thermal ellipsoid plots (50% probability level) derived from X-ray crystal structures of 1-OTf (left) and 2 (right). Hydrogen atoms and non-
coordinating counteranions have been omitted for clarity.
a
Table 1 Selected bond distances (Å) and angles (°) for 1-OTf, 2, and substrate-bound EgtB as determined by X-ray crystallography
b
1-OTf
2
EgtB
Fe–S1
2.3616(7)
2.312(2)
2.133(2)
2.150(2)
2.098(2)
Fe–S1
2.3048(7)
2.389(2)
2.134(2)
2.101(2)
2.094(2)
Fe–N (His51)
Fe–N (His134)
Fe–N (His138)
Fe–N (TMH)
2.13/2.12
2.06/2.14
2.11/2.11
2.10/2.12
2.09/2.12
2.21/2.19
Fe–N1 (amino)
Fe–N2 (pyr)
Fe–N3 (pyr)
Fe–N5 (imid)
Fe–N1 (amino)
Fe–N2 (pyr)
Fe–N4 (imid)
Fe–N6 (imid)
Fe–O (H
Fe–O (H
2
O)
O)
2
N1–Fe–S1
N2–Fe–S1
N3–Fe–S1
N5–Fe–S1
N1–Fe–N2
N1–Fe–N3
N1–Fe–N5
N2–Fe–N3
N2–Fe–N5
N3–Fe–N5
177.33(6)
102.88(6)
103.51(6)
107.21(7)
75.83(8)
75.24(8)
75.45(8)
117.40(8)
108.10(8)
116.26(8)
1.00
N1–Fe–S1
N2–Fe–S1
N4–Fe–S1
N6–Fe–S1
N1–Fe–N2
N1–Fe–N4
N1–Fe–N6
N2–Fe–N4
N2–Fe–N6
N4–Fe–N6
177.39(5)
105.08(6)
108.46(5)
105.51(6)
72.58(8)
73.97(7)
74.32(7)
122.40(8)
106.14(8)
108.09(7)
0.92
Mn–N (His51)
Mn–N (His134)
Mn–N (His138)
Mn–S (γGC)
2.31/2.15
2.12/2.13
2.25/2.10
2.54/2.58
2.25/2.21
2.39/2.30
Mn–N (DMH)
2
Mn–O (H O)
c
c
τ-Value
τ-Value
a
b
The enzymatic results were obtained from ref. 27. Bond distances are provided both active sites in the EgtB structures. The set of Fe–N/O bond
distances were measured for the ternary complex consisting of EgtB, Fe(II), and TMH (PDB: 4X8E). The Mn–N/O/S bond distances are derived
c
from the structure of the quaternary complex containing EgtB, Mn(II), N,N-dimethyl histidine (DMH), and γGC (PDB: 4X8D). See ref. 71 for a
definition of the τ-value.
N1 bond distance of 2.312 Å. In both cases, the S1–Fe–Npyr/imid (1-OTf, 2, and 4). In contrast, the Mn–SγGC bond distance of
angles lie closer to 109.5° than 90° (Table 1). Thus, based on ∼2.55 Å in the quaternary complex is much longer than the
the single-crystal X-ray data, the geometries of 1-OTf and 2 are Fe–S distances of ∼2.3 Å in the synthetic models, although the
best described as intermediate between five-coordinate (trigo- uncertainty in the former value is sizable due to the inherently
nal bipyramidal) and four-coordinate (tetrahedral).
lower resolution (1.98 Å) of the enzymatic data. Regardless,
The metal–ligand bond distances of substrate-bound EgtB from the EgtB structure, it is clear that the γGC substrate par-
are also provided in Table 1 for the sake of comparison. These ticipates in numerous noncovalent interactions with second-
values are derived from two X-ray structures reported by sphere residues that may contribute to elongation of the
2
7
Seebeck and coworkers. The first structure is a ternary metal–thiolate bond. An important parameter with respect to
complex consisting of EgtB, Fe(II), and TMH (PDB: 4X8E), C–S bond formation is the distance between the thiolate
while the second is an inactive quaternary complex containing S-atom and C2 of the His-derived substrate, which measures
EgtB, Mn(II), N,N-dimethyl histidine (DMH) and γGC (PDB: 3.8 Å in the Mn(II)-EgtB structure. The corresponding value is
4
X8D). The divalent metal ions in the EgtB active site are coor- only slightly longer (4.02 Å) in the X-ray structure of 1-OTf,
dinatively saturated due to the binding of one or two H suggesting that this complex has the proper geometry for SO
molecules, which resembles the binding of CH CN and OTf synthase reactivity.
2
O
3
moieties to the Fe(II) centers of 3 and 4. The Fe/Mn–N bond
distances involving the 3His triad and TMH/DMH substrates
range between 2.1 and 2.3 Å, similar to the Fe–N distances
Spectroscopic features of complexes 1–4
1
H NMR spectra of 1–4 display an abundance of paramagneti-
observed in X-ray structures of our high-spin Fe(II) complexes cally-shifted peaks between −20 and 130 ppm (Fig. S1–S4†).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
The effective magnetic moments measured using the Evans full width at half maximum (fwhm) of 100 Hz at room temp-
method lie between 4.8 and 5.0 μ in CD Cl , typical of high- erature, which is much broader than the peaks of 1-OTf and 4
B
2
2
spin Fe(II) complexes (S = 2). Complexes 1–4 remain paramag- (fwhm ∼20 Hz). For reference, the PhCF
netic in CD CN with effective magnetic moments ranging OTf both exhibit fwhm values of ∼5 Hz under identical con-
between 3.9 μ (for 3) and 4.7 μ (for 1), indicating that the Fe ditions. Collectively, these results suggest that a rapid equili-
3 4
standard and [NBu ]
3
B
B
(
II) centers are largely high-spin even in coordinating solvents. brium between bound and unbound triflate exists for 3 in
Thus, the low-spin geometry of 3 observed by XRD is likely due solution, whereas the triflate counteranions of 1-OTf and 4 are
1
9
to crystal packing effects. F NMR spectra of 1-OTf, 3, and 4 in “free” to a much greater extent.
CD CN feature a peak close to −78 ppm, typical of “free” The spin states and coordination geometries of 1–4 were
uncoordinated) triflate anions. The broadness of the triflate further examined using Mössbauer (MB) spectroscopy. The MB
peak (relative to a PhCF standard) can be used to assess the spectrum of complex 3 in a powder suspension (Fig. 4, top) con-
3
(
3
7
2
degree of counteranion binding to the Fe(II) center. As sum- sists of three overlapping quadrupole doublets that arise from
marized in Table S2,† the triflate peak of complex 3 exhibits a distinct iron species. The relative contribution of each sub-
species was determined by spectral analysis and least-square
fitting, and the results are summarized in Table 2.
Approximately 75% of the sample is attributed to two species
−1
with similar isomer shifts (δ) of 1.18 and 1.12 mm s – values
that are characteristic of high-spin, six-coordinate Fe(II) centers
73
with N/O donors. These two species are distinguished by their
different quadrupole splittings (ΔE ). A third component (20%
Q
Q
of the mixture, blue line in Fig. 4) displays smaller δ- and ΔE -
values of 0.47 and 0.35 mm s− , respectively, that are typical of
1
7
3
low-spin, six-coordinate Fe(II) centers. This minor species
A
2+
likely corresponds to the [Fe(L )(CH
3 2
CN) ]
complex observed
by X-ray crystallography (vide supra). The three species observed
for 3 are attributed to different combinations of CH CN and
3
3 3
OTf ligands in the coordination sphere (i.e., 2 CH CN, CH CN/
OTf, and 2 OTf). By comparison, the powder MB spectrum of 4
is less complex: nearly 90% of the signal arises from a six-coor-
−1
dinate, high-spin species with an isomer shift of 1.12 mm s
0 mT. The samples were prepared by suspension of powders in consistent with the crystallographic structure.
,
Fig. 4 MB spectra of complexes 3 (A) and 4 (B) measured at 7.0 K and
7
degassed mineral oil. The hatch marks are the experimental data, and
the colored lines are spectral fits generated with the parameters in
Table 2. For clarity, in spectrum A an unknown species with δ = 0.75 and
MB samples of 1 and 2 were prepared as both powder sus-
pensions and concentrated solutions in butyronitrile, and the
resulting spectra are shown in Fig. 5. The MB features of 1 are
nearly identical in the solid state and solution; both spectra
−1
ΔE
Q
= 1.0 mm s accounting for 5% absorption was subtracted from
the raw data.
Table 2 Experimental and DFT-Computed Mössbauer Parameters for Complexes 1–4
a
Experimental MB parameters
DFT-computed MB parameters
δ (mm s−
1
)
ΔE
Q
(mm s−1
)
FWHM (mm s−1
)
Model
C.N., spin
b
δ (mm s−1
)
Q
ΔE )
(mm s−1
Complex, state
1
, solid
Sp1 (73%)
Sp2 (17%)
0.94(3)
0.45(3)
3.52(3)
0.27(3)
0.29
0.33
1 (XRD)
[1(CH CN)]
3
5, S = 2
6, S = 2
0.88
1.08
2.97
3.00
+
+
1
2
, solution
, solid
Sp1 (90%)
Sp1 (96%)
0.93(3)
0.90(3)
3.64(3)
3.12(3)
0.31
0.31
2 (XRD)
[2(CH CN)]
5, S = 2
6, S = 2
0.85
1.07
2.65
2.78
2, solution
Sp1 (65%)
Sp2 (19%)
0.89(3)
1.00(3)
2.83(3)
3.70(3)
0.38
0.40
3
3, solid
Sp1 (55%)
Sp2 (20%)
Sp3 (20%)
1.18(3)
1.12(3)
0.47(3)
2.40(3)
3.00(3)
0.35(3)
0.35
0.30
0.40
3 (XRD)
6, S = 0
0.43
0.37
4, solid
Sp1 (92%)
1.12(5)
2.48(5)
0.30
4 (XRD)
6, S = 2
1.11
2.82
a
The relative contribution of each sub-species (Sp) was determined by least-square fitting of the experimental data. Contributions from very
b
minor species (<15%) are not included in the table; hence, the percentages do not add up to 100%. See the ESI† for further details. C.N. =
coordination number.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 5 MB spectra of complexes 1 and 2 measured at 7.0 K and 70 mT.
The samples consisted of either powders in degassed mineral oil (A and
C) or concentrated butyronitrile solutions (B and D). The hatch marks
are the experimental data, and the red lines are spectral fits generated
with the parameters in Table 2. Contributions from minor impurities
have been removed for the sake of clarity (see ESI† for additional
details).
Fig. 6 UV-vis absorption spectra of
temperature.
2 2
1 and 2 in CH Cl at room
mately 500 cm−1 on average. The lack of major differences
between spectra collected in CH Cl and CH CN is further evi-
2
2
3
dence of that these complexes possess five-coordinate geome-
tries in both solvents.
Q
consist of a dominant quadrupole doublet with δ- and ΔE -
values near 0.94 and 3.6 mm s , respectively (Table 2). In the
−
1
Computational studies of complexes 1–4
case of 2, the parameters of the solid-state sample (δ = 0.90
−
1
Q
and ΔE = 3.1 mm s ) are very similar to those of the major The geometric and electronic structures of complexes 1–4 were
species in the solution sample. The latter also contains a examined by DFT calculations that employed the B3LYP hybrid
minor species (20% of the mixture) with a larger isomer shift functional. Calculations of the precursor complexes (3 and 4)
of 1.00 mm s− . Previous studies have demonstrated that δ- utilized the crystallographic coordinates without further
1
−
1
values around 0.9 mm s are typical for five-coordinate, high- optimization. Two sets of computational models were gener-
4
1
spin Fe(II) complexes, including those with thiolate ligands.
ated for the iron(II)–thiolate complexes: (i) unoptimized, five-
Thus, the powder MB data collected for 1 and 2 are fully con- coordinate structures derived from XRD studies of 1 and 2,
sistent with the crystallographic structures. In addition, the and (ii) geometry-optimized, six-coordinate structures with
lack of significant changes in MB parameters upon solvation bound CH CN to account for inner-sphere solvation (i.e., [1
3
+
+
in butyronitrile indicates that the structures remain largely (CH CN)] and [2(CH CN)] ). Both high-spin and low-spin
five-coordinate even in coordinating solvents. However, the models of [1(CH CN)] and [2(CH CN)] were generated. The
3
3
+
+
3
3
larger isomer shift measured for the minor component of 2 acetonitrile ligand is presumed to adopt a position cis to the
suggests that a small fraction of this complex may bind butyro- thiolate and trans to an imidazole donor. Comparison of the
nitrile in solution to give a six-coordinate species.
DFT and XRD structures suggests that solvent binding causes
The iron(II)–thiolate complexes 1 and 2 are light yellow in a modest elongation of the Fe–S/N bonds by an average 0.07 Å
solution due to a series of intense absorption features in the (1) and 0.11 Å (2) for the high-spin models (Table S3†).
near-UV region (λ < 500 nm). These bands are distinct from
those observed in spectra of the precursor complexes. As the results are summarized in Table 2. These calculations
shown in Fig. 6, the absorption spectrum of 1 in CH Cl exhi- employed the B3LYP functional and the calibration parameters
Mössbauer parameters were calculated for each model and
2
2
bits two peaks at λmax = 356 and 402 nm (ε ∼ 1350 M− cm ). recently reported by Goldberg and coworkers for nonheme
1
−1
4
1
The spectral features of 2 consist of an intense peak at λ
3
4
=
iron–thiolate complexes. The five-coordinate models of 1 and
max
−1
−1
28 nm (ε = 2530 M cm ) and two shoulders at 369 and 2 yield MB parameters that agree nicely with both the solid-
25 nm. These absorption bands in the near-UV region are and solution-state experimental data, whereas the computed δ-
assigned as S → Fe(II) charge transfer (CT) transitions based values of ∼1.08 mm s− for the high-spin, six-coordinate
1
4
1,74
on literature precedents
and the computational results pre- models are too high. Thus, the DFT results provide further cor-
sented in the next section. Both complexes also display a weak roboration that 1 and 2 exist primarily as five-coordinate
−
1
−1
band near 1000 nm (ε = 30–50 M cm ) that is attributed to species in CH CN and butyronitrile solutions. Interestingly,
3
+
an Fe(II) d–d transition. Spectra of 1 and 2 measured in CH
exhibit a nearly identical set of S → Fe(II) CT features
Fig. S5†), although the peaks are blue-shifted by approxi- minor component of the solution-state spectrum of 2, which
3
CN the MB parameters computed for [2(CH CN)] (δ = 1.07 mm
3
−
1
−1
s
; ΔEQ = 2.78 mm s ) are similar to those observed for the
(
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
supports the hypothesis that this feature arises from a six-coor-
dinate species.
As shown in Fig. 8, absorption spectra of 1 and 2 generated
by time-dependent DFT (TD-DFT) calculations are in excellent
Q
For the precursor complexes, the δ- and ΔE -values com- agreement with the experimental spectra. The TD-DFT results
puted for the low-spin (S = 0) model of 3 closely match those confirm that the dominant features in the near-UV region
exhibited by the minor component in the experimental spec- arise from S → Fe(II) CT transitions. In the case of 1, the elec-
trum (Fig. 4 and Table 2), confirming that this doublet arises tron density difference map of the lowest-energy CT band
from a low-spin species. The isomer shift of 1.11 mm s− cal- (labeled a in Fig. 8) indicates that this transition has S
1
→ Fe
σ
culated for the six-coordinate model of 4 is in excellent agree- (3d) character, while the more intense transition at higher
−
1
π
ment with the experimental value of 1.12 mm s . Based on energy (b) has S → Fe(3d) character. These two excitations
these results, we can confidently assign the major components correspond to the experimental peaks at 402 and 356 nm. The
in the experimental spectrum of 3 to high-spin, six-coordinate computed spectrum of 2 features two strong S → Fe(3d) CT
π
species with structures similar to 4.
transitions (a and b) that match the shoulders at 369 and
DFT calculations also provided insights into the electronic 425 nm in the experimental spectrum. Many of the computed
structures of 1 and 2, particularly Fe/S bonding interactions. transitions at shorter wavelengths (like c at 312 nm) possess
Fig. 7 presents energy level diagrams derived from DFT calcu- both S → Fe(3d) and Fe(3d) → pyr(π*) CT character, which
σ
lations of the five-coordinate XRD structures, highlighting the accounts for the sizable intensity of absorption features near
Fe(II)- and thiolate-based molecular orbitals (MOs) most rele- 300 nm.
vant to the spectroscopic features of 1 and 2. For both com-
plexes, the highest-occupied MO (HOMO) in the spin-down (β) Reactivity of 1 and 2 with dioxygen and nitric oxide
manifold is an Fe(3d)-based orbital that bisects the S–Fe–Nimid
2 3
Exposure of 1 and 2 to O at room temperature in CH CN
causes a color change from yellow to dark brown. Absorption
angle. The four remaining spin-down Fe(3d)-based MOs are
unoccupied due to the high-spin nature of these Fe(II) com-
plexes. The two highest-energy thiolate-based MOs ligands
possess largely S(3p) character and are distinguished by their
spectra of the O reaction revealed that the S → Fe(II) CT tran-
2
sitions of 1 and 2 are replaced by broad and ill-defined bands
that tail into the visible region (Fig. S6†), suggesting loss of the
iron–thiolate bonds. In the case of 2, a purple chromophore
with a band at λmax = 520 nm is observed upon initial exposure
to O , but this species decays rapidly (t = 47 s) to give a sha-
π σ
orientation either perpendicular (S ) or parallel (S ) to the Fe–S
bond. The S -based MO of complex 1 exhibits substantial delo-
σ
calization over the mesityl ring, typical of aryl thiolates. The
average percentage of Fe character in the S(3p)-based MOs of 1
and 2 (22% and 11%, respectively) suggests that the Fe–S bond
of the former complex is more covalent. This difference in Fe/S
covalency is attributed to the smaller energy gap between the
S(3p) and Fe(3d) MOs in 1 compared to 2, which also accounts
for the relative energies of S → Fe(II) CT bands in the experi-
mental UV-vis spectra (vide supra).
2
1/2
peless spectrum. No intermediates are observed in the reaction
of 1 with O . Analysis of the oxygenated solutions with GC-MS
after work-up) revealed that the thiolate ligands of 1 and 2 are
2
(
primarily converted into the corresponding disulfides, MesS–
SMes and CyS–SCy, respectively. Formation of these disulfide
1
products was further confirmed by H NMR analysis of the O -
2
treated samples (see Fig. S7 and S8†). Disulfides are com-
Fig. 7 Energy-level diagrams and isosurface plots of spin-down (β) molecular orbitals derived from spin-unrestricted DFT calculations (B3LYP; def-
TZVP) of complexes 1 (left) and 2 (right).
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
drance that prevents O
2
from binding directly to the Fe(II)
center. Such a possibility is suggested by the spectroscopic
studies presented above, which found that CH CN and butyro-
3
nitrile molecules are incapable of coordinating to 1 and 2 in
solution. To probe this hypothesis further, we examined the
reaction of 1 and 2 with nitric oxide (NO) – an analog of O
2
that has been employed in numerous studies of biological and
synthetic nonheme iron sites. The reaction of complexes 1 and
2
2 2
with NO at low temperature (−70 °C) in CH Cl generates
new chromophores (labeled 1-NO and 2-NO) with the UV-
visible absorption features shown in Fig. 9. The spectrum of
NO-treated 2 exhibits a band at 430 nm and less intense fea-
tures at 550 and 630 nm; a similar set of absorption bands is
observed for NO-treated 1. The same species are generated
upon exposure to NO at room temperature, but the absorption
bands are less intense and decay over time. The features of 1-
7
NO and 2-NO are characteristic of six-coordinate {FeNO}
81
species (Enemark–Feltham notation ) with a high-spin (S = 3/
3
6,82–85
2) ground state.
Consistent with this conclusion, ana-
Fig. 8 TD-DFT computed absorption spectra (solid black lines) of com-
plexes 1 (top) and 2 (bottom) compared to experimental spectra (dashed
red lines) measured in CH
indicate the energies and intensities of computed transitions; intense S
Fe(II) CT transitions are marked with letters (a, b, and c). The insets
2 2
Cl at room temperature. The black sticks
→
display electron density difference maps of the lowest-energy S → Fe(II)
CT transition for each complex. The yellow and green areas represent
the loss and gain, respectively, of electron density during the excitation.
monly-observed products of O
2
reactions involving iron–thio-
late complexes, as demonstrated in previous studies of CDO
4
1,75,76
models.
product derived from the aryl thiolate ligand in the reaction of
with O . In the case of 2, a small amount of the thiosulfonate
Dimesityldisulfide was the only observed
1
2
1
ester, CyS–S(O)
2
Cy, was also detected by GC-MS and H NMR
Fig. S8†). The thiosulfonate product could arise from either
dioxygenation of CyS–SCy or reaction of a sulfinic acid inter-
mediate (CySO H; not observed) with cyclohexylthiol(ate). Fig. 9 UV-vis absorption spectra of the reaction of complexes 1 (top)
and 2 (bottom) with NO gas at −70 °C in CH Cl ([1] = 0.95 mM; [2] =
7
7
(
2
There is precedent for both reaction pathways in the
2
2
7
8–80
0.98 mM). The wavelengths of prominent bands in the spectra of 1-NO
literature.
reaction mixtures of either complex.
Sulfinic acid products were not observed in the
and 2-NO (red lines) are indicated. Insets: IR spectra measured after
14
15
treatment of 1 or 2 with NO (red lines) or NO (blue lines) at room
temperature in CH Cl . IR spectra of 1 and 2 in CH Cl (black lines) are
We surmised that the paucity of S-oxygenation products in
2
2
2
2
the reaction of 1 and 2 with O might be due to steric hin- shown for comparison.
2
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
lysis of NO-treated samples with X-band EPR spectroscopy probed with computational methods, which accurately repro-
revealed axial S = 3/2 signals (g ∼ 4.0 and 2.0) typical of duced the experimental spectroscopic features. Both com-
7
86
7
{
FeNO} species (Fig. S9†).
plexes react with NO to generate six-coordinate {FeNO}
The formation of Fe/NO adducts was further confirmed by species, as confirmed by EPR and IR spectroscopies. The struc-
IR spectroscopy. The IR spectrum of 1-NO exhibits an intense tures of these Fe/NO adducts resemble those proposed for
feature at 1727 cm− that is not present in the precursor spec- iron-superoxo intermediates in the catalytic cycles of EgtB and
1
−
1
15,32–35
trum (Fig. 9, inset). This peak downshifts by 33 cm when OvoA.
1
5
NO gas is employed, indicating that it arises from the N–O
Even though 1 and 2 are faithful structural models of the
stretching mode, ν(NO). Similarly, the IR spectrum of 2-NO SO synthase active sites, the major thiolate-derived product
−
1
14 15
2
displays a ν(NO)-based peak at 1702 cm ( N/ N isotope upon reaction with O at room temperature is the corres-
shift of 32 cm− ). The difference in ν(NO) frequencies indicates ponding disulfide. Neither sulfinic acids nor products arising
1
that the aliphatic SCy ligand of 2-NO is more electron-donating from S–C bond formation were observed, indicating that 1 and
8
7
than the aryl SMes ligand of 1-NO. The ν(NO) frequencies of 2 do not replicate the reactivity of either thiol dioxygenases or
-NO and 2-NO are comparable to those previously reported SO synthases. It is likely that second-sphere residues in the
1
7
88–90
for similar thiolate-ligated nonheme {FeNO} species.
Based on the linear correlation presented by Chavez et al.,
active-site pockets of EgtB and OvoA steer the iron-bound sub-
strates towards thiohistidine products and prevent the kind of
9
1
the NO stretching frequencies of 1-NO and 2-NO are indicative deleterious side-reactions observed for our solvent-exposed
of bent nitrosyl ligands with Fe–N–O bond angles around 150°. mimics. Regardless, this first generation of synthetic SO
Computational models of 1-NO and 2-NO generated via DFT synthase models offers a valuable foundation for continuing
geometry optimizations feature Fe–N–O angles of 161° and biomimetic studies aimed at a more complete understanding
1
54°, respectively, further supporting the correlation between of SO synthase catalysis. Future work in our laboratories will
ν(NO) frequency and Fe–N–O angle. seek to promote enzyme-like O reactivity though rational
2
−
1
Interestingly, the decay of the 1702 cm peak of 2-NO at modifications of ligand structure, coordination geometry, and/
room temperature in accompanied by an increase in the inten- or reaction conditions. We will also probe the identity of inter-
−
1
sity of two isotopically-active features at 1772 and 1745 cm
mediates, such as the transient purple species observed in the
1
4
15
(
Fig. S10†). These two peaks, which exhibit N/ N isotope reaction of 2 with O , which promises to shed light on the O
shifts of 35 cm , likely correspond to the ν(NO) modes of a activation mechanisms of nonheme iron enzymes with thiolate
dinitrosyl iron complex (DNIC). Indeed, we occasionally substrates.
observed the S = 1/2 signal of {Fe(NO)2} species in EPR
2
2
−
1
9
2
9
spectra of 1-NO and 2-NO samples. DNICs are common decay
products of thiolate–ligated iron-nitrosyl species because di-
Conflicts of interest
sulfide formation provides the necessary reducing equivalents
7
9
36,93
for the 2{FeNO} → {FeNO
2
}
reaction.
Thus, our results There are no conflicts to declare.
indicate that diatomic molecules like NO (and, by extension,
O ) are capable of binding to the Fe(II) centers of 1 and 2, but
2
the resulting adducts are unstable at elevated temperatures.
Acknowledgements
The authors are grateful for financial support from the
National Institutes of Health (GM126522 to A. T. F.).
Improvements to the X-band EPR instrument at Marquette
Conclusions
In this manuscript, we have reported the syntheses, X-ray struc- University were funded through a grant from the National
tural characterization, spectroscopic features, and O /NO reac- Science Foundation (CHE-1532168). The Mössbauer experi-
2
tivities of two nonheme iron–thiolate complexes that serve as ments were supported by start-up funds from The College of
structural mimics of SO synthases. These synthetic complexes Arts and Sciences (to C. V. P.) and the Undergraduate Research
emulate the substrate-bound active sites of EgtB and OvoA in Opportunities Program (M. E. E.) at The University of
three crucial respects: both 1 and 2 feature (i) a high-spin, five- St. Thomas. C. V. P thanks Dr. Michael Hendrich (Carnegie
coordinate Fe(II) center, (ii) a monodentate thiolate ligand cis Mellon University) for the program Spin Count.
to an imidazole moiety, and (iii) three neutral N-donors in a
facial geometry similar to the enzymatic 3His triad. The
desired coordination geometry was achieved by incorporating
the imidazole donor(s) within the tetradentate L and L
References
A
B
scaffolds, followed by formation of Fe(II) precursors (3 and 4)
1 M. H. Stipanuk, C. R. Simmons, P. Andrew Karplus and
J. E. Dominy, Jr., Amino Acids, 2011, 41, 91–102.
2 M. H. Stipanuk, Annu. Rev. Nutr., 2004, 24, 539–577.
3 S. C. Peck and W. A. van der Donk, J. Biol. Inorg. Chem.,
2017, 22, 381–394.
that readily bind a single thiolate ligand. UV-vis absorption,
1
H NMR, and Mössbauer spectroscopic studies revealed the 1
2 2
and 2 retain their five-coordinate geometries in CH Cl and
nitrile solutions. The nature of the iron–thiolate bond was
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
4
5
M. H. Stipanuk, M. Londono, L. L. Hirschberger, C. Hickey, 30 K. V. Goncharenko and F. P. Seebeck, Chem. Commun.,
D. J. Thiel and L. Wang, Amino Acids, 2004, 26, 99–106. 2016, 52, 1945–1948.
S. C. Chai, A. A. Jerkins, J. J. Banik, I. Shalev, J. L. Pinkham, 31 L. Chen, N. Naowarojna, H. Song, S. Wang, J. Wang,
P. C. Uden and M. J. Maroney, J. Biol. Chem., 2005, 280,
865–9869.
Z. Deng, C. Zhao and P. Liu, J. Am. Chem. Soc., 2018, 140,
4604–4612.
9
6
7
M. H. Stipanuk, I. Ueki, J. E. Dominy, Jr., C. R. Simmons 32 A. R. Stampfli, K. V. Goncharenko, M. Meury, B. N. Dubey,
and L. L. Hirschberger, Amino Acids, 2009, 37, 55–63.
T. Schirmer and F. P. Seebeck, J. Am. Chem. Soc., 2019, 141,
5275–5285.
J. E. Dominy, Jr., C. R. Simmons, L. L. Hirschberger,
J. Hwang, R. M. Coloso and M. H. Stipanuk, J. Biol. Chem., 33 A. S. Faponle, F. P. Seebeck and S. P. de Visser, J. Am. Chem.
007, 282, 25189–25198. Soc., 2017, 139, 9259–9270.
N. Masson, T. P. Keeley, B. Giuntoli, M. D. White, 34 W.-J. Wei, P. E. M. Siegbahn and R.-Z. Liao, Inorg. Chem.,
2
8
M. L. Puerta, P. Perata, R. J. Hopkinson, E. Flashman,
2017, 56, 3589–3599.
F. Licausi and P. J. Ratcliffe, Science, 2019, 365, 65–69.
35 G. Tian, H. Su and Y. Liu, ACS Catal., 2018, 8, 5875–5889.
9
N. Bruland, J. H. Wuebbeler and A. Steinbuechel, J. Biol. 36 A. A. Fischer, N. Stracey, S. V. Lindeman, T. C. Brunold and
Chem., 2009, 284, 660–672. A. T. Fiedler, Inorg. Chem., 2016, 55, 11839–11853.
0 B. S. Pierce, B. P. Subedi, S. Sardar and J. K. Crowell, 37 A. C. McQuilken and D. P. Goldberg, Dalton Trans., 2012,
Biochemistry, 2015, 54, 7477–7490. 41, 10883–10899.
1 E. P. Tchesnokov, M. Fellner, E. Siakkou, T. Kleffmann, 38 M. Sallmann, I. Siewert, L. Fohlmeister, C. Limberg and
1
1
L. W. Martin, S. Aloi, I. L. Lamont, S. M. Wilbanks and
G. N. L. Jameson, J. Biol. Chem., 2015, 290, 24424–24437.
2 Y. Ishikawa, S. E. Israel and D. B. Melville, J. Biol. Chem.,
C. Knispel, Angew. Chem., Int. Ed., 2012, 51, 2234–2237.
39 L. Mueller, S. Hoof, M. Keck, C. Herwig and C. Limberg,
Chem. – Eur. J., 2020, 26, 11851–11861.
40 M. N. Blakely, M. A. Dedushko, P. C. Y. Poon, G. Villar-
Acevedo and J. A. Kovacs, J. Am. Chem. Soc., 2019, 141,
1867–1870.
1
1
1
1
1
974, 249, 4420–4427.
3 R. N. Vogt, H. S. C. Spies and D. J. Steenkamp,
Eur. J. Biochem., 2001, 268, 5229–5241.
4 N. Naowarojna, R. Cheng, L. Chen, M. Quill, M. Xu, 41 J. B. Gordon, J. P. McGale, J. R. Prendergast, Z. Shirani-
C. Zhao and P. Liu, Biochemistry, 2018, 57, 3309–3325.
5 A. R. Stampfli, W. Blankenfeldt and F. P. Seebeck, Curr.
Opin. Struct. Biol., 2020, 65, 1–8.
Sarmazeh, M. A. Siegler, G. N. L. Jameson and
D. P. Goldberg, J. Am. Chem. Soc., 2018, 140, 14807–14822.
42 D. Kumar, G. N. Sastry, D. P. Goldberg and S. P. de Visser,
J. Phys. Chem. A, 2012, 116, 582–591.
1
1
6 F. P. Seebeck, J. Am. Chem. Soc., 2010, 132, 6632–6633.
7 A. Braunshausen and F. P. Seebeck, J. Am. Chem. Soc., 2011, 43 K. Anandababu, R. Ramasubramanian, H. Wadepohl,
1
33, 1757–1759.
8 G. T. M. Mashabela and F. P. Seebeck, Chem. Commun.,
013, 49, 7714–7716.
9 H. Song, M. Leninger, N. Lee and P. Liu, Org. Lett., 2013,
5, 4854–4857.
P. Comba, N. Johnee Britto, M. Jaccob and
R. Mayilmurugan, Chem. – Eur. J., 2019, 25, 9540–9547.
44 S. P. de Visser, M. G. Quesne, B. Martin, P. Comba and
U. Ryde, Chem. Commun., 2014, 50, 262–282.
45 Y. Lee, G.-Y. Park, H. R. Lucas, P. L. Vajda, K. Kamaraj,
M. A. Vance, A. E. Milligan, J. S. Woertink, M. A. Siegler,
A. A. Narducci Sarjeant, L. N. Zakharov, A. L. Rheingold,
E. I. Solomon and K. D. Karlin, Inorg. Chem., 2009, 48,
11297–11309.
1
1
2
2
2
1
0 H. Song, A. S. Her, F. Raso, Z. Zhen, Y. Huo and P. Liu, Org.
Lett., 2014, 16, 2122–2125.
1 N. Naowarojna, P. Huang, Y. Cai, H. Song, L. Wu, R. Cheng,
Y. Li, S. Wang, H. Lyu, L. Zhang, J. Zhou and P. Liu, Org.
Lett., 2018, 20, 5427–5430.
46 J. Chaignon, S.-E. Stiriba, F. Lloret, C. Yuste, G. Pilet,
L. Bonneviot, B. Albela and I. Castro, Dalton Trans., 2014,
43, 9704–9713.
2
2
2 I. K. Cheah and B. Halliwell, Biochim. Biophys. Acta, Mol.
Basis Dis., 2012, 1822, 784–793.
3 M. R. Ariyanayagam and A. H. Fairlamb, Mol. Biochem. 47 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found.
Parasitol., 2001, 115, 189–198.
Crystallogr., 2008, 64, 112–122.
2
2
4 C. Jacob, Nat. Prod. Rep., 2006, 23, 851–863.
5 I. Castellano and F. P. Seebeck, Nat. Prod. Rep., 2018, 35,
48 G. M. Sheldrick, SHELXL-97, Program for crystal structure refine-
ment, University of Göttingen, Göttingen, Germany, 1997.
49 O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard
and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339–341.
1
241–1250.
2
6 J. G. McCoy, L. J. Bailey, E. Bitto, C. A. Bingman, D. J. Aceti,
B. G. Fox and G. N. Phillips, Jr., Proc. Natl. Acad. 50 A. T. Smith, T. Majtan, K. M. Freeman, Y. Su, J. P. Kraus
Sci. U. S. A., 2006, 103, 3084–3089. and J. N. Burstyn, Inorg. Chem., 2011, 50, 4417–4427.
7 K. V. Goncharenko, A. Vit, W. Blankenfeldt and 51 F. Neese, ORCA - An Ab Initio, DFT and Semiempirical
2
2
2
F. P. Seebeck, Angew. Chem., Int. Ed., 2015, 54, 2821–2824.
8 N. Naowarojna, S. Irani, W. Hu, R. Cheng, L. Zhang, X. Li,
J. Chen, Y. J. Zhang and P. Liu, ACS Catal., 2019, 9, 6955–6961.
Electronic Structure Package, version 4.0, Max Planck
Institute for Chemical Energy Conversion, Muelheim
(Germany), 2017.
9 S. P. de Visser and G. D. Straganz, J. Phys. Chem. A, 2009, 52 F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2012, 2,
13, 1835–1846. 73–78.
1
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
5
5
3 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
4 C. T. Lee, W. T. Yang and R. G. Parr, Phys. Rev. B: Condens.
Matter Mater. Phys., 1988, 37, 785–789.
74 A. T. Fiedler, H. L. Halfen, J. A. Halfen and T. C. Brunold,
J. Am. Chem. Soc., 2005, 127, 1675–1689.
75 Y. M. Badiei, M. A. Siegler and D. P. Goldberg, J. Am. Chem.
Soc., 2011, 133, 1274–1277.
5
5 F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005,
7
, 3297–3305.
76 A. A. Fischer, S. V. Lindeman and A. T. Fiedler, Chem.
Commun., 2018, 54, 11344–11347.
77 T. X. T. Luu, T.-T. T. Nguyen, T. N. Le, J. Spanget-Larsen
and F. Duus, J. Sulfur Chem., 2015, 36, 340–350.
78 T. Keshari, R. Kapoorr and L. D. S. Yadav, Synlett, 2016, 27,
1878–1882.
5
6 F. Neese, F. Wennmohs, A. Hansen and U. Becker, Chem.
Phys., 2009, 356, 98–109.
7 F. Weigend, Phys. Chem. Chem. Phys., 2006, 8, 1057–1065.
8 F. Weigend, J. Comput. Chem., 2008, 29, 167–175.
9 V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995–
5
5
5
2
001.
79 P. Natarajan, Tetrahedron Lett., 2015, 56, 4131–4134.
6
6
6
6
6
6
0 T. Yanai, D. P. Tew and N. C. Handy, Chem. Phys. Lett., 80 R. J. Reddy, J. J. Kumar and A. H. Kumari, Eur. J. Org.
004, 393, 51–57. Chem., 2019, 3771–3775.
1 E. J. Blaesi, B. G. Fox and T. C. Brunold, Biochemistry, 2015, 81 J. H. Enemark and R. D. Feltham, Coord. Chem. Rev., 1974,
4, 2874–2884. 13, 339–406.
2 S. Hirata and M. Head-Gordon, Chem. Phys. Lett., 1999, 82 Y. M. Chiou and L. Que, Inorg. Chem., 1995, 34, 3270–3278.
2
5
3
14, 291–299.
3 S. Hirata and M. Head-Gordon, Chem. Phys. Lett., 1999,
02, 375–382.
4 M. Romelt, S. F. Ye and F. Neese, Inorg. Chem., 2009, 48,
84–785.
83 T. A. Jackson, E. Yikilmaz, A. F. Miller and T. C. Brunold,
J. Am. Chem. Soc., 2003, 125, 8348–8363.
3
84 H. Park, M. M. Bittner, J. S. Baus, S. V. Lindeman and
A. T. Fiedler, Inorg. Chem., 2012, 51, 10279–10289.
85 E. C. Wasinger, M. I. Davis, M. Y. M. Pau, A. M. Orville,
J. M. Zaleski, B. Hedman, J. D. Lipscomb, K. O. Hodgson
and E. I. Solomon, Inorg. Chem., 2003, 42, 365–376.
86 T. C. Berto, A. L. Speelman, S. Zheng and N. Lehnert,
Coord. Chem. Rev., 2013, 257, 244–259.
7
5 The “core properties” basis set is derived from the
TurboMole DZ basis set developed by Ahlrichs and
coworkers.
6
6 E. V. Bowers, G. S. Murphy, S. N. Till, C. J. VandenBussche,
M. M. Yaroschak, R. D. Pike, R. J. Butcher and 87 T. C. Berto, M. B. Hoffman, Y. Murata, K. B. Landenberger,
D. C. Bebout, Dalton Trans., 2014, 43, 16475–16485.
7 K. Chen and L. Que, Jr., J. Am. Chem. Soc., 2001, 123, 6327–
E. E. Alp, J. Y. Zhao and N. Lehnert, J. Am. Chem. Soc.,
2011, 133, 16714–16717.
6
6
6
6
337.
8 S. J. Lange, H. Miyake and L. Que, Jr., J. Am. Chem. Soc.,
999, 121, 6330–6331.
88 G. Villar-Acevedo, E. Nam, S. Fitch, J. Benedict,
J. Freudenthal, W. Kaminsky and J. A. Kovacs, J. Am. Chem.
Soc., 2011, 133, 1419–1427.
1
9 Y. Zang, J. Kim, Y. Dong, E. C. Wilkinson, E. H. Appelman 89 A. C. McQuilken, H. Matsumura, M. Durr, A. M. Confer,
and L. Que, Jr., J. Am. Chem. Soc., 1997, 119, 4197–
205.
0 Based on an analysis of structures in the Cambridge
Crystallographic Data Centre.
1 A. W. Addison, T. N. Rao, J. Reedijk, J. Vanrijn and
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–
J. P. Sheckelton, M. A. Siegler, T. M. McQueen, I. Ivanovic-
Burmazovic, P. Moenne-Loccoz and D. P. Goldberg, J. Am.
Chem. Soc., 2016, 138, 3107–3117.
4
7
7
90 M. D. Clay, C. A. Cosper, F. E. Jenney, Jr., M. W. W. Adams
and M. K. Johnson, Proc. Natl. Acad. Sci. U. S. A., 2003, 100,
3796–3801.
1
356.
91 J. Li, A. Banerjee, P. L. Pawlak, W. W. Brennessel and
F. A. Chavez, Inorg. Chem., 2014, 53, 5414–5416.
7
7
2 A. N. Biswas, M. Puri, K. K. Meier, W. N. Oloo, G. T. Rohde,
E. L. Bominaar, E. Munck and L. Que, J. Am. Chem. Soc., 92 H. Lewandowska, M. Kalinowska, K. Brzoska, K. Wojciuk,
2
015, 137, 2428–2431.
G. Wojciuk and M. Kruszewski, Dalton Trans., 2011, 40,
8273–8289.
3 E. Munck, in Physical Methods in Bioinorganic Chemistry:
Spectroscopy and Magnetism, ed. L. Que, University Science 93 M.-L. Tsai, C.-C. Tsou and W.-F. Liaw, Acc. Chem. Res., 2015,
Books, Sausalito, CA, 2000, pp. 287–319. 48, 1184–1193.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.