Heme/non-heme diiron(II) complexes and O2, CO, and NO adducts as reduced and substrate-bound models for the active site of bacterial nitric oxide reductase

Article abstract of DOI:10.1021/ja0458773

As a first generation model for the reactive reduced active-site form of bacterial nitric oxide reductase, a heme/non-heme diiron(II) complex [( ⁶L)Fe^II-Fe^II-(CI)]⁺ (2) {where ⁶L = partially fluorinated tetraphenylporphyrin with a tethered tetradentate TMPA chelate; TMPA = tris(2-pyridyl)amine} was generated by reduction of the corresponding μ-oxo diferric compound [(⁶L) Fe^III-O-Fe^III-CI]⁺ (1). Coordination chemistry models for reactions of reduced NOR with O₂, CO, and NO were also developed. With O₂ and CO, adducts are formed, [(⁶L) Fe^III(O₂^-)(thf)...Fe^II-Cl] B(C₆F₅)₄ (2a·O₂) {λ_max 418 (Soret), 536 nm; ν_O-O = 1176 cm ^-1, ν_Fe-O = 574 cm^-1 and [( ⁶L)Fe^II(CO)(thf)Fe^II-Cl]B(C₆F ₅)₄ (2a·CO) {ν_CO 1969 cm ^-1}, respectively. Reaction of purified nitric oxide with 2 leads to the dinitrosyl complex [(⁶L)Fe(NO)Fe(NO)-Cl]B(C₆F ₅)₄ (2a·(NO)₂) with ν_NO absorptions at 1798 cm^-1 (non-heme Fe-NO) and 1689 cm^-1 (heme-NO).

Full text of DOI:10.1021/ja0458773

Published on Web 02/16/2005
Heme/Non-Heme Diiron(II) Complexes and O2, CO, and NO
Adducts as Reduced and Substrate-Bound Models for the
Active Site of Bacterial Nitric Oxide Reductase
Ian M. Wasser,† Hong-wei Huang,‡ Pierre Moe¨nne-Loccoz,‡ and
Kenneth D. Karlin*,†
Contribution from the Department of Chemistry, Johns Hopkins UniVersity, Charles and
34th Streets, Baltimore, Maryland 21218, and Department of EnVironmental & Biomolecular
Systems, OGI School of Science & Engineering, Oregon Health & Science UniVersity,
BeaVerton, Oregon 97006
Received July 9, 2004; E-mail: karlin@jhu.edu
Abstract: As a first generation model for the reactive reduced active-site form of bacterial nitric oxide
6
reductase, a heme/non-heme diiron(II) complex [(6L)FeII‚‚‚FeII-(Cl)]+ (2) {where L ) partially fluorinated
tetraphenylporphyrin with a tethered tetradentate TMPA chelate; TMPA ) tris(2-pyridyl)amine} was
generated by reduction of the corresponding µ-oxo diferric compound [(6L)FeIII-O-FeIII-Cl]+ (1).
Coordination chemistry models for reactions of reduced NOR with O2, CO, and NO were also developed.
With O2 and CO, adducts are formed, [(6L)FeIII(O2-)(thf)‚‚‚FeII-Cl]B(C6F5)4 (2a‚O2) {λmax 418 (Soret), 536
nm; νO-O ) 1176 cm-1, νFe-O ) 574 cm-1 and [(6L)FeII(CO)(thf)FeII-Cl]B(C6F5)4 (2a‚CO) {νCO 1969 cm-1},
respectively. Reaction of purified nitric oxide with 2 leads to the dinitrosyl complex [(6L)Fe(NO)Fe(NO)-
Cl]B(C6F5)4 (2a‚(NO)2) with νNO absorptions at 1798 cm-1 (non-heme Fe-NO) and 1689 cm-1 (heme-
NO).
Introduction
Bacterial denitrification is the anaerobic alternative to aerobic
respiration or photosysnthesis, in which nitrogen oxides are used
as terminal electron acceptors.1,2 The four reductive steps (see
below) in this cycle are catalyzed by metalloenzymes. In these
regards, considerable efforts are being devoted to understanding
the interactions of metalloproteins and/or relevant coordination
compounds with nitrogen oxides.2-7
Figure 1. Schematic representation of the NOR active site showing the
binuclear center where NO(g) reductive coupling occurs, producing nitrous
oxide (N2O).
nitrogen atoms from three histidine ligands with possible ligation
by an additional glutamate residue. The NorB subunit, which
includes the diiron active site, exhibits weak homology to
subunit I of cytochrome c oxidase (CcO), which couples the
reduction of O2 to water with membrane proton translocation;
the six histidine residues that ligate heme a, heme a3, and CuB
in CcO are conserved in NOR.14,15 In essence, the CuB site of
CcO has been replaced by a non-heme iron (FeB) in NOR; CcO
and NOR are genetic cousins.1,2,12,13,16
NO3- f NO2- f NO f N2O f N2
(1)
Nitric oxide reductase (NOR) is a membrane-bound bacterial
enzyme that catalyzes the third step (eq 1), the reduction of
nitric oxide to nitrous oxide, using electrons donated from
soluble cytochrome c.2,3,8,9 NOR isolated from Paracoccus
denitrificans has been shown to contain one high-spin and one
low-spin heme b, one low-spin heme c, and a non-heme iron
(FeB) per enzyme (Figure 1).10-13 The FeB is coordinated by
Our own interest in this area stems from the examination of
heme/copper oxidase model complexes and their interactions
with dioxygen, relevant to the heme a3/CuB binuclear active
site in CcO.17-21 One iron/copper system studied employs a
† Johns Hopkins University.
‡ Oregon Health & Science University.
6
binucleating ligand (e.g., L; Chart 1) {6L is the dianion of
(1) Zumft, W. G. Microbiol. Mol. Biol. ReV. 1997, 61, 533-616.
(2) Wasser, I. M.; de Vries, S.; Moe¨nne-Loccoz, P.; Schro¨eder, I.; Karlin, K.
D. Chem. ReV. 2002, 102, 1201-1234.
(10) Hendriks, J. H. M.; Gohlke, U.; Saraste, M. J. Bioenergetics Biomembranes
1998, 30, 15-24.
(11) Hendriks, J. W. A.; Gohlke, U.; Haltia, T.; Ludovici, C.; Lubben, M.;
Saraste, M. Biochemistry 1998, 37, 13102-13109.
(12) Girsch, P.; de Vries, S. Biochim. Biophys. Acta 1997, 1318, 202-216.
(13) Moe¨nne-Loccoz, P.; de Vries, S. J. Am. Chem. Soc. 1998, 120, 5147.
(14) de Vries, S.; Schro¨eder, I. Biochem. Soc. Trans. 2002, 30, 662-667.
(15) Saraste, M.; Castresana, J. FEBS Lett. 1994, 341, 1-4.
(16) Pinakoulaki, E.; Gemeinhardt, S.; Saraste, M.; Varostis, C. J. Biol. Chem.
2002, 277, 23407-23413.
(3) Averill, B. A. Chem. ReV. 1996, 96, 2951-2964.
(4) Ford, P. C.; Lorkovic, I. Chem. ReV. 2002, 102, 993-1017.
(5) Coppens, P.; Novozhilova, I.; Kovalevsky, A. Chem. ReV. 2002, 102, 861-
884.
(6) Butler, A.; Megson, I. L. Chem. ReV. 2002, 102, 1155-1166.
(7) McCleverty, J. A. Chem. ReV. 2004, 104, 403-418.
(8) Moura, I.; Moura, J. J. G. Curr. Opin. Chem. Biol. 2001, 5, 168-175.
(9) Richardson, D. J.; Watmough, N. J. Curr. Opin. Chem. Biol. 1999, 3, 207-
19.
9
3310
J. AM. CHEM. SOC. 2005, 127, 3310-3320
10.1021/ja0458773 CCC: $30.25 © 2005 American Chemical Society

Models for the Active Site of Bacterial NOR
A R T I C L E S
Chart 1
heme/non-heme diiron complex with dioxygen and nitric
oxide.24 Here, we report the synthesis and characterization of
the complex [(6L)FeII‚‚‚FeII-Cl]+ (2), a reduced heme/non-heme
diiron complex, representing a spectroscopic and/or structural
model for the putative active (reduced) site in NOR. Further-
more, we have examined the reactivity of 2 with dioxygen,
carbon monoxide, and nitric oxide, as it is important to lay the
foundation for understanding the interactions and basic coor-
dination chemistry of these small molecules in a heme/non-
heme diiron(II) environment. Bacterial NOR has been found to
reduce dioxygen in a manner similar to CcO (i.e., having oxidase
activity),26,27 while CO often serves as an O2 or NO surrogate
or probe for spectroscopic and structural studies of biologically
relevant iron enzymes or complexes.28-38 A detailed examination
of the dioxygen reactivity of [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a)
provides proof that this dioxygen adduct is [(6L)FeIII(O2-)‚‚‚
FeII-Cl]B(C6F5)4 (2a‚O2), a superoxo rather than a peroxo
species (Chart 1). We have further examined the formation of
the heme-dioxygen (superoxo) complex in a closely related
“empty tether” complex, the reaction of [(6L)FeII] (3) + O2
producing [(6L)FeIII(O2-)] (3‚O2) (Chart 1). Finally, we present
the first substrate bound (e.g., NO and analogue CO) model
complexes for NOR, room-temperature stable monocarbonyl
(2a‚CO), and dinitrosyl (2a‚(NO)2) adducts. The characterization
includes vibrational studies of the related nitrosyl compounds
formed from reactions of [(6L)FeII] (3) or [(F8)FeII] (4) and
NO(g). We find that complexes 4 and 4‚NO do not catalyze
the disproportionation of NO(g).
Experimental Section
Materials and Methods. Reagents were obtained from commercial
sources. Dichloromethane (CH2Cl2), tetrahydrofuran (THF), and toluene
were purified and dried by passing reagent-grade solvent through a
series of two activated alumina columns. Acetone was dried over and
then distilled from indicator-free calcium sulfate. Deoxygenation of
these solvents was performed by bubbling with argon for 20 min
followed by three freeze/pump/thaw cycles. Deuterated tetrahydrofuran
(THF-d8) was distilled from a sodium/benzophenone ketyl and degassed
by three freeze/pump/thaw cycles. Using the modified procedures based
on the methods of Ford and co-workers,4 nitric oxide (research grade/
BOC gases) was purified by passage through a series of two Ascarite
II columns and finally distilled at 173 K. 15NO (99%, ICON isotopes,
Summit, NJ) was distilled at 173 K prior to use. Carbon monoxide
(Matheson Gas Products) was purified by passage through an R & D
Separations oxygen/moisture trap (model OT3-4). Bulbs of 18O2 (99%,
5-(ortho-O-[(N,N-bis(2-pyridylmethyl)-2-(6-methoxyl)pyridine-
methanamine)phenyl]-10,15,20-tris(2,6-difluorophenyl)porphine}
featuring a porphyrin tethered to a tetradentate chelate (TMPA,
tris(2-pyridylmethyl)amine) able to coordinate copper.17 Given
the similarities between the active sites in CcO and NOR, an
obvious extension would be to incorporate a non-heme iron ion
at the TMPA tether, which we have previously accomplished
in investigations featuring “resting” or oxidized active site NOR
models.22-25
Thus, our previous efforts have included the synthesis of
several µ-oxo heme/non-heme diiron compounds.23,25 There has
also been a cursory examination of the reactivity of one reduced
(17) Ghiladi, R. A.; Ju, T. D.; Lee, D.-H.; Moe¨nne-Loccoz, P.; Kaderli, S.;
Neuhold, Y.-M.; Zuberbu¨ehler, A. D.; Woods, A. S.; Cotter, R. J.; Karlin,
K. D. J. Am. Chem. Soc. 1999, 121, 9885-9886.
(26) Fujiwara, T.; Fukumori, Y. J. Bacteriol. 1996, 178, 1866-1871.
(27) Butland, G.; Spiro, S.; Watmough, N. J.; Richardson, D. J. J. Bacteriol.
2001, 183, 189-199.
(18) Ju, T. D.; Ghiladi, R. A.; Lee, D.-H.; van Strijdonck, G. P. F.; Woods, A.
S.; Cotter, R. J.; Young, V. G., Jr.; Karlin, K. D. Inorg. Chem. 1999, 38,
2244-2245.
(28) Spiro, T. G.; Zgierski, M. Z.; Kozlowski, P. M. Coord. Chem. ReV. 2001,
219-221, 923-936.
(29) Spiro, T. G.; Kozlowski, P. M. Acc. Chem. Res. 2001, 34, 137-144.
(30) Zhang, Y.; Palosky, M. A.; Brown, C. A.; Westre, T. E.; Hedman, B.;
Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1992, 114, 9189-
9191.
(19) Obias, H. V.; van Strijdonck, G. P. F.; Lee, D.-H.; Ralle, M.; Blackburn,
N. J.; Karlin, K. D. J. Am. Chem. Soc. 1998, 120, 9696-9697.
(20) Kim, E.; Helton, M. E.; Wasser, I. M.; Karlin, K. D.; Lu, S.; Huang, H.-
w.; Moenne-Loccoz, P.; Incarvito, C. D.; Rheingold, A. L.; Honecker, M.;
Kaderli, S.; Zuberbuhler, A. D. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
3623-3628.
(31) Brown, C. A.; Palosky, M. A.; Westre, T. E.; Zhang, Y.; Hedman, B.;
Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1995, 117, 715-732.
(32) Randall, C. R.; Zang, Y.; True, A. E.; Que, J. L.; Charnock, J. M.; Garner,
C. D.; Fujishima, Y.; Schofield, C. J.; Baldwin, J. E. Biochemistry 1993,
32, 6664-6673.
(21) Kim, E.; Chufan, E. E.; Kamaraj, K.; Karlin, K. D. Chem. ReV. 2004, 104,
1077-1133.
(22) Moe¨nne-Loccoz, P.; Richter, O.-M. H.; Huang, H.-W.; Wasser, I. M.;
Ghiladi, R. A.; Karlin, K. D.; de Vries, S. J. Am. Chem. Soc. 2000, 122,
9344-9345.
(33) Que, J. L.; Dong, Y. Acc. Chem. Res. 1996, 29, 190-196.
(34) Que, J. L. J. Chem. Soc., Dalton Trans. 1997, 3922-3940.
(35) Hsu, H.-F.; Yoo, S. J.; Mu¨nck, E.; Que, J. L. J. Am. Chem. Soc. 2000,
122, 3789-3790.
(23) Martens, C. F.; Murthy, N. N.; Obias, H. V.; Karlin, K. D. J. Chem. Soc.,
Chem. Commun. 1996, 629-630.
(36) MacMurdo, V. L.; Zheng, H.; Que, J. L. Inorg. Chem. 2000, 39, 2254-
(24) Ju, T. D.; Woods, A. S.; Cotter, R. J.; Moe¨nne-Loccoz, P.; Karlin, K. D.
Inorg. Chim. Acta 2000, 297, 362-372.
2255.
(37) Costas, M.; Chen, K.; Que, J. L. Coord. Chem. ReV. 2000, 200-202, 517-
544.
(25) Wasser, I. M.; Martens, C. F.; Verani, C. N.; Rentschler, E.; Huang, H.-
w.; Moeenne-Loccoz, P.; Zakharov, L. N.; Rheingold, A. L.; Karlin, K. D.
Inorg. Chem. 2004, 43, 651-662.
(38) Kryatov, S. V.; Rybak-Akimova, E. V.; MacMurdo, V. L.; Que, J. L. Inorg.
Chem. 2001, 40, 2220-2228.
9
J. AM. CHEM. SOC. VOL. 127, NO. 10, 2005 3311

A R T I C L E S
Wasser et al.
ICON Isotopes, Summit, NJ) gas were held at 193 K prior to use. The
preparation and handling of air-sensitive materials was carried out with
Schlenk techniques under argon or in an MBraun Labmaster 130
glovebox held under nitrogen gas (O2 < 1 ppm; H2O < 1 ppm).
Elemental analyses were performed by Desert Analytics (Tucson,
AZ) or Quantitative Technologies, Inc. (QTI, Whitehouse, NJ). 1H NMR
spectra were recorded at 300 MHz on a Bruker AMX-300 instrument.
Chemical shifts are reported as δ values relative to an internal standard
(Me4Si) and the residual solvent proton peak. Infrared spectra were
obtained on either a Mattson Galaxy 4030 FT-IR spectrometer or an
in situ ReactIR 1000 FTIR spectrometer (RIR) with SiComp ATR probe
tip (Mettler-Toledo, Millersville, MD). Solid samples were prepared
by dissolving a sample in solution in the glovebox, spotting a CaF2 IR
window, and allowing the solvent to evaporate. Alternatively, IR
measurements were made by evaporating solutions of a sample, under
a stream of argon, onto a SiComp ATR probe window. Air-sensitive
UV-visible samples were prepared in the glovebox in the appropriate
deoxygenated solvent. Low-temperature UV-visible spectra were
recorded with a Hewlett-Packard 8453 diode array spectrometer
equipped with HPChemstation software. Low-temperature UV-visible
spectra were recorded using a modified literature procedure.39,40 Briefly,
a quartz-windowed vacuum dewar was connected (via copper tubing)
to a methanol-filled external recirculating cold bath (Neslab Endocal).
The methanol temperature within the dewar was monitored using a
thermocouple probe (Omega Model 651). The cuvette assembly
consisted of a four-window quartz cuvette (1 cm path) connected, via
a 12 cm glass tube, to a 14/20 female ground glass joint and a glass
stopcock. Room-temperature UV-visible spectra were recorded with
a Varian Cary-50 spectrophotometer. Electron paramagnetic resonance
(EPR) spectra were obtained on a Bruker EMX spectrometer operating
at X-band. Matrix-assisted laser desorption-ionization time-of-flight
mass spectrometry (MALDI-MS) was performed using a Kratos
Kompact 4 spectrometer and the porphyrin complexes were examined
without added matrix. MALDI samples were prepared on stainless steal
plates (20 well) in the glovebox and sealed in a long air-free reaction
flask with a rubber septum. Once removed from the glovebox and
positioned near the spectrometer, MALDI plates were quickly removed
from the flask under a stream of dry argon and inserted into the
instrument to minimize exposure to the ambient atmosphere.
[(6L)FeII‚‚‚FeII-Cl]PF6 (2b). A procedure, identical to the one
employed for the synthesis of 2a, was utilized with the exception that
[(6L)FeIII-O-FeIII-Cl]PF6 (1b) was reduced. The resulting product is
a bright red solid (192 mg, 80%). UV-Vis (THF; nm, ꢀ mol-1 L-1):
310, 23000; 424, 205000; 544, 9900. Anal. Calcd for C67H48ClF12-
Fe2N8O2P (2b‚C4H8O): C, 57.35; H, 3.45; N, 7.99. Found: C, 57.18;
H, 3.43; N, 8.17.
[(6L)FeIII(O2-)‚‚‚FeII-Cl]B(C6F5)4 (2a‚O2). UV-visible scale samples
were prepared using a THF solution (5 × 10-5 M) of 2a in an air-free
cuvette assembly fitted with a rubber septum. Subsquent cooling to
193 K was followed by the gentle bubbling with dioxygen via syring
needle. Samples of 2a‚O2 were prepared for RR spectroscopy as
described below using either 16O2 or 18O2. The warming of samples of
2a‚O2 to room temperature, following preparation with 18O2, forms a
labeled Fe-(18O)-Fe end-product species, as determined by RR
spectroscopy {νas(Fe-18O-Fe) ) 803 cm-1
}
25 or by FTIR spectroscopy
{νas(Fe-18O-Fe) ) 803 cm-1}.
[(6L)Fe(CO)(thf)FeII-Cl]B(C6F5)4 (2a‚CO). In the glovebox, a THF
solution (12 mL) of [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a) (200 mg, ca. 10
mmol) was charged to a 50 mL air-free reaction flask fitted with a
rubber septum and subjected to three freeze/pump/thaw cycles, finally
leaving it at room temperature under argon. To this static system was
added carbon monoxide (10 mL, 1 atm) via syringe with gentle bubbling
through the solution. The solution was allowed to stir at room
temperature under a CO atomosphere for 2 h and then the solvent was
removed under reduced pressure. Recrystallization of the resulting solid
from CH2Cl2/heptane gave a red microcrystalline product (155 mg,
75%). 1H NMR (THF-d8, 300 MHz): δ 92 (br), 81 (br), 60 (s), 53-50
(m), 46-42 (m), 39-29 (m), 8.9 (pyrrole-H), -5, -10, -19. UV-
Vis (THF, nm): 412, 532. FTIR (film, cm-1) 1969 (νCO), 1642, 1623,
1605, 1583, 1513, 1463, 1413, 1372, 1333, 1271, 1235, 1201. Anal.
Calcd for C93H50BCl3F26Fe2N8O3 (4‚CH2Cl2‚C4H8O): C, 54.48; H, 2.46;
N, 5.47. Found: C, 54.73; H, 2.53; N, 5.16.
[(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2). In the glovebox, a
THF solution (12 mL) of [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a) (200 mg,
ca. 10 mmol) was charged to a 50 mL air-free reaction flask fitted
with a glass stopper. The reaction flask was subjected to three freeze/
pump/thaw cycles and finally left frozen in liquid nitrogen under a
static vacuum. To this was added nitric oxide (50 mL, 1 atm) and the
reaction vessel was subsequently allowed to warm to room temperature
as a closed system. The reaction mixture, now a bright orange-red color,
was allowed to stir at room tempearture for 4 h. The solvent was then
removed under reduced pressure. Recrystallization of the resulting solid
Synthesis. [(6L)FeIII-O-FeIII-Cl]B(C6F5)4 (1a),25 [(6L)FeIII-O-
FeIII-Cl]PF6 (1b),25 [(6L)FeII] (3),18,41 and [(F8)FeII] (4)42,43 were
prepared as described in the literature.
[(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a). Under argon, a solution of [(6L)-
FeIII-O-FeIII-Cl]B(C6F5)4 (1a) (250 mg, 0.132 mmol) in deoxygenated
CH2Cl2 (40 mL) was added to a 1 M solution of sodium dithionite in
deoxygenated water (50 mL). The heterogeneous phases were mixed
with vigorous argon bubbling for 35 min, at which time the organic
layer had turned from a dark red to a bright red color. The solvent was
removed and the crude material recrystallized from a THF and heptane
mixture overnight at room temperature. The resulting microcrystalline
material was filtered and dried under reduced pressure to yield a purple-
1
from CH2Cl2/heptane gave an orange-red powder (140 mg, 70%). H
NMR (THF-d8, 300 MHz): δ 159, 142, 135, 130, 126, 116, 101, 93,
87, 76, 64, 61, 60, 55, 51.2, 49, 47, 43, -22, -29, -50. UV-Vis
(THF, nm): 412, 545. FTIR (film, cm-1) 1798 (νNO), 1689 (νNO), 1642,
1625, 1606, 1583, 1575, 1513, 1462, 1413, 1374, 1344, 1274, 1235,
1204. EPR (THF solution, 198 K): g ≈ 2.0. Anal. Calcd for C95H58-
BCl3F26Fe2N10O3 (3‚CH2Cl2‚C7H16): C, 54.07; H, 2.77; N, 6.64.
Found: C, 53.80; H, 2.36; N, 6.49. MALDI-TOF-MS: m/z 1247
(M + H+ - BArF-)+.
1
red solid (200 mg, 81%). H NMR (THF-d8, 300 MHz): δ 91 (br,
R-H), 82 (br, methylene -CH2-), 59 (s, PY-H, â), 57 (s, PY-H, â),
50-47 (m, pyrrole-H), 46-44 (m, PY-H), 43-35 (m, PY-H), 21
(m, pyrrole-H), 7.9-7.2 (m- and p-phenyl-H), -6.1 (PY-H, γ), -7.5
(PY-H, γ), -19 (br, R-H). UV-Vis (THF; nm, ꢀ mol-1 L-1): 310,
23000; 424, 205000; 544, 9900. FTIR (film, cm-1) 1644, 1625, 1606,
1583, 1513, 1463, 1413, 1374, 1336, 1274, 1235, 1204. MALDI-TOF-
MS: m/z 1188 (M + 2H+ - BArF-)+.
UV-Visible Spectroscopic Test for Reversible O2-Binding. In the
glovebox, solutions of 2a were prepared in THF solvent and transferred
into a low-temperature quartz cuvette with Schlenk-type air-free sidearm
flask. Subsequent cooling of the cuvette assembly to 193 K was
followed by direct bubbling of dioxygen into the solution via needle,
thereby generating a heme-O2 adduct. The removal of excess dioxygen
was accomplished by application of a vacuum at 193 K. The test for
reversible dioxygen binding was carried out by bubbling the solution
with a steady stream of dry argon at 193 K for several minutes.
Dioxygen Uptake Determined by Spectrophotometric Titration.
Using a published procedure,44 the dioxygen uptake for room-
(39) Karlin, K. D.; Cruse, R. W.; Gultneh, Y.; Farooq, A.; Hayes, J. C.; Zubieta,
J. J. Am. Chem. Soc. 1987, 109, 2668-2679.
(40) Karlin, K. D.; Haka, M. S.; Cruse, R. W.; Meyer, G. J.; Farooq, A.; Gultneh,
Y.; Hayes, J. C.; Zubieta, J. J. Am. Chem. Soc. 1988, 110, 1196-1207.
(41) Ghiladi, R. A.; Karlin, K. D. Inorg. Chem. 2002, 41, 2400-2407.
(42) Kopf, M.-A.; Neuhold, Y.-M.; Zuberbuehler, A. D.; Karlin, K. D. Inorg.
Chem. 1999, 38, 3093-3102.
(44) Ghiladi, R. A.; Kretzer, R. M.; Guzei, I.; Rheingold, A. L.; Neuhold, Y.-
M.; Hatwell, K. R.; Zuberbu¨ehler, A. D.; Karlin, K. D. Inorg. Chem. 2001,
40, 5754-5767.
(43) Nanthakumar, A.; Fox, S.; Karlin, K. D. J. Chem. Soc., Chem. Commun.
1995, 499-501.
9
3312 J. AM. CHEM. SOC. VOL. 127, NO. 10, 2005

Models for the Active Site of Bacterial NOR
A R T I C L E S
temperature solutions of [(6L)FeII‚‚‚FeII-Cl]+ (2) was determined using
UV-visible spectroscopy. An air-free cuvette assembly was charged
with a THF solution of 2, and then known quantities of dioxygen (from
a 1.01% O2 in Ar gas mixture) were directly bubbled via syringe into
the solutions. The reaction mixture was allowed to equilibrate for 12
h before a measurement was made. Measurements indicated that roughly
0.60 or 0.49 (avg ∼0.55) equiv of dioxygen were required for the
complete conversion of [(6L)FeII‚‚‚FeII-Cl]+ (2) to [(6L)FeIII-O-FeIII-
Cl]PF6 (1). Further injection(s) of dioxygen, in excess of 0.55 equiv of
dioxygen, resulted in no further spectroscopic changes.
Scheme 1
Dioxygen Evolution during Conversion of 2 to 1. Spectrophoto-
metric measurements, using alkaline pyrogallol solutions, were used
to determine dioxygen evolution from low-temperature heme-O2
adducts according to published procedures.41,44 Anaerobic THF solutions
(12 mL) of [(6L)FeII‚‚‚FeII-Cl]+ (2) (20 mg) were prepared in an air-
free flask and fitted with a rubber septum. The solutions were removed
from the glovebox and then connected to the vacuum/argon manifold.
Solutions of 2 were cooled under an argon atmosphere to 193 K, and
then an excess of dioxygen was bubbled through the solution. To
remove excess unbound O2, 2a‚O2 was subjected to five freeze/pump/
thaw cycles with each thaw cycle involving the warming of the solution
from liquid nitrogen temperature (77 K) to roughly 193 K. Subse-
quently, the low-temperature adduct was warmed to room temperature
and the dioxygen evolution (0.41, 0.49, 0.52 equiv O2) was quantified
(UV-visible spectroscopy) by pyrogallol trapping.
Nitric Oxide Evolution. The addition of excess axial base to
solutions of heme-nitrosyl complexes, resulting in the liberation of
NO(g), has been reported.45 Mixing an excess of 1,5-dicyclohexylimi-
dazole or (dimethylamino)pyridine with THF solutions of [(6L)Fe(NO)-
Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2) resulted in the evolution of NO(g).
Nitric oxide was directly detected in the headspace of the reaction using
gas chromatography. Gas chromatography was performed on a Hewlett-
Packard 5890 Series II instrument equipped with a thermal conductivity
detector (TCD) using a 12 ft Porapak-Q packed column. In a second
qualitative method, an air-free cuvette containing a THF solution of
[(F8)FeII] (4) was connected, via rubber tubing, to a flask containing
the THF mixture of base/2a‚(NO)2. The released NO(g) was “trapped”
by 4, resulting in conversion to the complex [(F8)Fe(NO)] (4‚NO), as
determined by UV-visible spectroscopy.
desire to “model” the active site reactivity of the heme/non-
heme diiron complexes with relevant substrates such as NO,
CO, and O2 prompted the synthesis of fully reduced analogues
of these µ-oxo compounds. Deaerated dichloromethane solutions
of the µ-oxo complexes 1a/1b are readily reduced, in a biphasic
reaction, using aqueous sodium dithionite to yield either [(6L)-
FeII‚‚‚FeII-Cl]B(C6F5)4 (2a) or [(6L)FeII‚‚‚FeII-Cl]PF6 (2b),
respectively (Scheme 1). The reaction is presumably quantita-
tive; we report isolated yields of ∼80% due to limitations in
physically separating the organic compounds (2a/2b) from the
aqueous dithionite solutions.
As expected, [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a) and [(6L)FeII‚
‚‚FeII-Cl]PF6 (2b) have identical UV-visible signatures in THF
solvent, with very intense Soret transitions at λmax ) 424 nm
(ꢀ ) 205000 M-1 cm-1). Of lesser intensity are the Q-bands at
λmax ) 544 nm (ꢀ ) 9900 M-1 cm-1). These features are
consistent with those of known ferrous heme compounds in
coordinating solvents.41,44 It should be noted that the UV-visible
features are dominated by porphyrin-based π f π* transitions
(ꢀ ≈ 10000-200000 M-1 cm-1) which precludes the observa-
tion of the expected much weaker (TMPA)FeII absorption (λmax
≈ 400 nm, ꢀ ≈ 3000 M-1 cm-1).46
Examination of the 1H NMR spectrum of 2a in THF-d8
reveals an array of paramagnetically shifted proton signals,
Figure 2, ranging downfield from 91 ppm to upfield at -19
Resonance Raman (RR) Spectroscopy. RR samples were prepared
using a 3-5 mM solution of 2a in an appropriate solvent. This solution
was then transferred to a 5 mm NMR tube and fitted with a rubber
septum, placed into a cold bath (193 K) and bubbled with either 16O2
or 18O2 gas (10 mL) using a Hamilton gastight syringe fitted with an
air-free three-way valve. The oxygenated samples were held cold for
10 min, before being frozen in liquid nitrogen and flame-sealed.
RR spectra were obtained using frozen samples thermostated to 90
K with a liquid nitrogen coldfinger. A 413 nm Kr+ laser (Innova 302,
Coherent, Santa Clara, CA) was used for excitation in conjunction with
a Kaiser Optical (Ann Arbor, MI) supernotch filter to attenuate Rayleigh
scattering. The backscattered light was analyzed with a McPherson
(Acton, MA) 2061/207 spectrograph (0.67 m with variable gratings)
equipped with a Princeton Instrument (Roper Scientific, Trenton, NJ)
liquid-nitrogen cooled (LN-1100PB) charge-coupled device (CCD)
detector. Frequencies were calibrated relative to known standards and
are within (1 cm-1. The laser power was kept below 20 mW to
minimize sample decomposition and to prevent undesirable photo-
chemical side reactions.
Figure 2. Proton NMR spectrum of [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a),
recorded at 298 K in THF-d8.
ppm. The spectrum represents of juxtaposition of proton signals
arising from the influence of the two independent high-spin FeII
ion centers, with paramagnetic resonances observed for (pre-
sumably) pyrrole, pyridyl (TMPA), and methylene (TMPA)
protons. The spectrum is further complicated by the inherent
lack of symmetry (as compared with the 6L µ-oxo compounds),25
as the tethered-TMPA portion of the system is no longer
Results and Discussion
Synthesis and Characterization of [(6L)FeII‚‚‚FeII-Cl]+
(2). The complexes [(6L)FeIII-O-FeIII-Cl]B(C6F5)4 (1a) and
[(6L)FeIII-O-FeIII-Cl]PF6 (1b), differing only in counteranion
identity, were prepared according to literature procedures.25
A
(46) Kojima, T.; Leising, R. A.; Yan, S.; Que, L., Jr. J. Am. Chem. Soc. 1993,
115, 11328-35.
(45) Bohle, D. S.; Hung, C.-H. J. Am. Chem. Soc. 1995, 117, 9584-9585.
9
J. AM. CHEM. SOC. VOL. 127, NO. 10, 2005 3313

A R T I C L E S
Wasser et al.
Figure 3. Reaction of the reduced complex [(6L)FeII‚‚‚FeII-Cl]+ (2) with
dioxygen, to form a µ-oxo complex [(6L)FeIII-O-FeIII-Cl] (1), via a low-
spin heme-O2 (FeIII-O2-) intermediate. The stoichiometries of reactions
and incorporation of O-label into 1 have been confirmed, see text.
Figure 4. UV-visible spectrum (THF, 193 K) of [(6L)FeII‚‚‚FeII-Cl]B-
(C6F5)4 (2a) (solid trace; λmax 424, 544 nm) and the spectrum of the resulting
dioxygen adduct [(6L)FeIII(O2-)‚‚‚FeII-Cl]B(C6F5)4 (2a‚O2) (dashed trace;
λmax 414, 536 nm).
“anchored” directly to the porphyrin iron and we expect
considerable room-temperature fluxionality of the TMPA-tether
(relative to the plane of the porphyrin). Reasonable assignments
include the following: (1) Pyrrole proton resonances found as
a set of multiplets between 47 and 50 ppm and a singlet near
21 ppm, consistent with other synthetic five-coordinate high-
spin hemes in THF-d8.18,24,41,44 (2) Pyridyl-H and methylene
-CH2- (TMPA) proton resonances observed at 91, 82, 59, 57,
46-44, 43-35, -6.1, -7.5, and -19 ppm. The assignments
of the TMPA-Fe proton resonances are consistent with the work
of Que and co-workers,47 who have noted that the inclusion of
a methyl substituent in the 6-pyridyl position in (TMPA)FeII
complexes (i.e., as occurs here for one of the pyridyl donor
groups in the ligand 6L) favors iron ions with larger ionic radii
(e.g., high-spin Fe(II)). Such S ) 2 complexes have paramag-
netically shifted proton resonances, typically between -24 and
95 ppm.47
Reactivity of Dioxygen with [(6L)FeII‚‚‚FeII-Cl]+ (2).
Addition of an excess of dioxygen to room-temperature solutions
of [(6L)FeII‚‚‚FeII-(Cl)]B(C6F5)4 (2a) in THF results in the clean
formation of [(6L)FeIII-O-FeIII-Cl]B(C6F5)4 (1a) (Figure 3).
Room-temperature spectrophotometric titrations (see Experi-
mental Section) of 2a with dioxygen in THF reveal that 0.55
equiv of dioxygen are required to form 1 equiv of [(6L)FeIII-
O-FeIII-Cl]B(C6F5)4 (1a). However, an intermediate is ob-
served by direct bubbling of O2 (∼30 s) into a cold (193 K)
THF solution of [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a) {λmax ) 424
(Soret) and 544 nm}; this is formulated as a heme-superoxo
complex [(6L)FeIII(O2-)‚‚‚FeII-Cl)]B(C6F5)4 (2a‚O2), having
UV-visible features at 318, 414 (Soret), and 536 nm (Figures
3 and 4). This finding is similar to that observed for related
complexes [(6L)FeII] (3) and [(F8)FeII] (4) (Chart 1), which also
form heme-superoxo adducts in THF solvent, [(6L)Fe(O2-)]
(3‚O2)41 and [(F8)Fe(O2-)] (4‚O2),20,44 respectively. While O2-
binding to 3 and 4 is reversible, the formation of [(6L)FeIII-
(O2-)‚‚‚FeII-Cl)]B(C6F5)4 (2a‚O2), derived from the heme/non-
heme diiron(II) complex 2a, is not. Direct bubbling of argon
or application of a vacuum does not result in the release of
dioxgyen. These results suggest that either the non-heme iron
stabilizes the heme-dioxygen interaction in [(6L)FeIII(O2-)‚‚‚
FeII-Cl]B(C6F5)4 (2a‚O2) or an alternative decomposition
pathway (i.e., other than simple loss of O2) is readily available.
Warming of 2a‚O2 to room temperature cleanly produces the
µ-oxo complex [(6L)FeIII-O-FeIII-Cl]B(C6F5)4 (1a) (Figure
3) with the evolution of ∼0.5 equiv of dioxygen (see Experi-
mental Section), consistent with the stoichiometric relationships
between 2, 1, and 3, those indicated in Figure 3.
Resonance Raman (RR) Spectroscopy. Low-temperature
(90 K) laser excitation (413 nm) into the Soret band of [(6L)-
FeII‚‚‚FeII-Cl]B(C6F5)4 (2a) in THF yields a RR spectrum
dominated by porphyrin skeletal modes.48 The ν4 and ν2 modes
are observed at 1345 and 1537 cm-1, respectively, indicating
2a is best formulated as a high-spin ferrous complex (see
Supporting Information, Figure S1). Indeed, for typical five-
coordinate high-spin [(P)FeII], the oxidation-state marker ν4 is
observed at 1342 cm-1 and the spin-state marker ν2 is at 1537
cm-1 48,49
These results and assignments are consistent with the
.
1H NMR pyrrole proton resonance assignment (vide supra)
suggesting an S ) 2, high-spin heme complex with an axial
THF ligand for 2a. In contrast, porphyrin skeletal modes in the
high-frequency region for the dioxygen complex [(6L)FeIII(O2-)‚
‚‚FeII-Cl]B(C6F5)4 (2a‚O2) indicate the presence of a low-spin
ferric heme species with marker bands (ν4 and ν2) at 1370 cm-1
and 1572 cm-1, respectively, by comparison with other known
low-spin [(TPP)FeIII] compounds (ν ) 1370, 1568 cm-1).49
Presumably, the formally ferric-heme in 2a‚O2 is coordinated
by axial superoxo (O2-) and THF ligands (Figure 3), giving an
overall six-coordinate low-spin complex, as is well known.41,44
The hexa-coordinated nature of the superoxo complex is
confirmed by the frequency displayed by the ν(Fe-O2) de-
scribed below.
An earlier, less detailed study24 showed that an analogue
complex of [(6L)FeII‚‚‚FeII-Cl]+ (2), [(5L)FeII‚‚‚FeII-Cl]+, was
(48) Nakamoto, K. Infrared and Raman Spectra of Inorganic and Coordination
Compounds; Wiley-Interscience: New York, 1997.
(49) Burke, J. M.; Kincaid, J. R.; Peters, S.; Gagne, R. R.; Collman, J. P.; Spiro,
T. G. J. Am. Chem. Soc. 1978, 100, 6083-6088.
(47) Zang, Y.; Kim, J.; Dong, Y.; Wilkinson, E. C.; Appelman, E. H.; Que, L.,
Jr. J. Am. Chem. Soc. 1997, 119, 4197-4205.
9
3314 J. AM. CHEM. SOC. VOL. 127, NO. 10, 2005

Models for the Active Site of Bacterial NOR
A R T I C L E S
Figure 6. Empty tether complex [(6L)FeII] (3) reacts with dioxygen to give
a low-temperature heme-superoxo adduct 3‚O2. Dioxygen may perhaps
bind to either the top or the bottom face of the tethered heme.
Scheme 2
Figure 5. RR spectra of [(6L)FeIII(O2-)‚‚‚FeII-Cl]B(C6F5)4 (2a‚O2
)
-
formed by oxygenation with either 16O2 (solid trace) or 18O2 (dashed trace).
The difference spectrum (16O - 18O) is shown as the bottom trace. The
inset shows the high-frequency region using both 16O2 (solid trace) or 18O2
(dashed trace), with the bottom trace showing the difference spectrum (16O
-
18O). The spectra were recorded at 90 K using 413 nm excitation.
observed to form a low-temperature stable dioxygen adduct,
but whether dioxygen was bound as a peroxo or a superoxo
ligand was not ascertained. Here, however, examination of the
mid-frequency RR regions for [(6L)FeIII(O2-)‚‚‚FeII-Cl]B-
(C6F5)4 (2a‚O2) confirms that dioxygen is bound as a superoxo
(O2-) entity (Figure 5). RR spectra of 2a‚O2, prepared with 16O2
gas, reveals a ν(O-O) at 1176 cm-1, that downshifts by 63
cm-1 to 1113 cm-1 with 18O2 gas. In the low-frequency region,
the ν(Fe-O) is detected at 574 cm-1 and shifts by -25 cm-1
with 18O2 gas. These frequencies unambiguously identify 2a‚
O2 as a hexa-coordinated iron(III)-superoxo end-on com-
plex.48,50,51 Warming the 16O2 and 18O2 superoxo adducts 2a‚
O2 to room temperature re-forms [(6L)FeIII-O-FeIII-Cl]B(C6F5)4
(1a), with concomitant incorporation of the labeled oxygen
species into the Fe-O-Fe bridge, based upon observed
signatures between the complexes [(6L)FeIII(O2-)‚‚‚FeII-Cl]B-
(C6F5)4 (2a‚O2) and [(6L)FeIII(O2-)] (3‚O2) indicates that the
non-heme iron does not influence the O-O or Fe-O stretching
vibrations when dioxygen binds to the heme iron in 2a‚O2. THF
solutions (193 K) of [(F8)FeII(O2-)] (4‚O2), formed by reaction
of [(F8)FeII] (4) + O2 (Chart 1), have similar RR vibrational
properties.52
νas(Fe-16O-Fe) {844 cm-1} and νas(Fe-18O-Fe) {803 cm-1
}
stretching frequencies, by either RR or FTIR spectroscopies.25
Thus, the bridging oxo ligand is derived from dioxygen (Figure
3).
Direct bubbling of dioxygen into low-temperature (198 K)
THF solutions of the “empty tether” complex [(6L)FeII] (3)
(Chart 1) also results in the formation of a low-temperature
stable superoxo intermediate (Figure 6), [(6L)FeIII(O2-)] (3‚O2),
which has been previously characterized by UV-visible and
NMR spectroscopies.41 This dioxygen complex is a low-spin
six-coordinate species with an axial THF ligand. Here, RR
spectroscopic insights are provided (Supporting Information,
Figure S2), obtained using either 16O2 or 18O2, which confirm
the presence of a superoxo complex with a ν(O-O) ) 1176
cm-1 (∆(18O2) -64 cm-1) and a ν(Fe-O) ) 572 cm-1
(∆(18O2) -24 cm-1). The close similarity in vibrational
Proposed Mechanism for µ-Oxo Formation. Several TM-
PA-Fe complexes with methyl substituents in the 6-pyridyl
positions, in the absence of ancillary O-donor ligands, generally
have high redox potentials that favor FeII over the FeIII oxidation
state.47 Thus, given the stoichiometry of reactions observed
(Figure 3) and the clear characteristics of [(6L)FeIII(O2-)‚‚‚FeII-
Cl]B(C6F5)4 (2a‚O2), we infer that the non-heme iron(II) in 2a
does not bind O2 and that the heme-superoxo moiety in 2a‚O2
is the sole dioxygen intermediate observed at 193 K.
Scheme 2 provides a proposed series of reactions which may
account for a dioxygen (formally, superoxide) disproportion-
(50) Proniewicz, L. M.; Paeng, I. R.; Nakamoto, K. J. Am. Chem. Soc. 1991,
113, 3294-3303.
(51) Vogel, K. M.; Kozlowski, P. M.; Zgierski, M. Z.; Spiro, T. G. J. Am. Chem.
Soc. 1999, 121, 9915-9921.
(52) Kim, E.; Helton, M. E.; Wasser, I. M.; Karlin, K. D.; Lu, S.; Huang, H.-
W.; Moe¨nne-Loccoz, P.; Incarvito, C. D.; Rheingold, A. L.; Honecker, M.;
Kaderli, S.; Zuberbuhler, A. D. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
3623-8.
9
J. AM. CHEM. SOC. VOL. 127, NO. 10, 2005 3315

A R T I C L E S
Wasser et al.
ation, starting from 2a‚O2 and forming the µ-oxo complex [(6L)-
FeIII-O-FeIII-Cl]B(C6F5)4 (1a). The disproportionation reac-
tion stoichiometry is (eq 2)
showed a prominent new band (compared to 2a) at ν(CO) )
1969 cm-1 (Supporting Information, Figure S3), consistent with
the presence of a mononuclear heme carbonyl species and
indicating that the CO ligand is not bridging the two iron sites.48
Typical non-heme iron-carbonyl complexes are limited to
binary carbonyl and cyclopentadienyl ligand systems;63 the non-
heme iron in 2a appears to have little affinity for CO. A
comparison of the heme-CO stretching vibrations of the
complex 2a‚CO {ν(CO)Fe 1969 cm-1} to the CO vibration of
the empty-tether complex [(6L)FeII(CO)] (3‚CO) in THF
2FeIII-(O2 )‚‚‚FeII
-
III
-
III
f O +
(2)
2Fe -(O2 )-Fe
2
(2a‚O2)
1a
Superoxo compound 2a‚O2 combines with a second equivalent
of 2a‚O2, forming an intermolecular heme-peroxo-heme
complex, while releasing an equivalent of dioxygen. Subsequent
thermally induced (i.e. warming) cleavage of the O-O bond
would result in a short-lived mixed-valent heme ferryl/non-heme
ferrous complex, which could then combine with the non-heme
iron(II) in an intramolecular fashion to give the diferric µ-oxo
product [(6L)FeIII-O-FeIII-Cl]B(C6F5)4 (1a). Similar mecha-
nisms, involving µ-peroxo diferric heme or non-heme diiron
compounds that thermally decompose to yield µ-oxo diferric
systems, are known.53-58 Attempts to identify a peroxo inter-
mediate here, e.g., [heme Fe-O-O-Fe heme] or [heme Fe-
O-O-Fe non-heme] species, have been unsuccessful.
Synthesis of Monocarbonyl Compound [(6L)FeII(CO)(thf)-
FeII-Cl]B(C6F5)4 (2a‚CO). Solutions of [(6L)FeII‚‚‚FeII-Cl]B-
(C6F5)4 (2a) in THF were bubbled with an excess of carbon
monoxide at room temperature, turning from red to peach in
color, to give a compound formulated as [(6L)FeII(CO)(thf)-
FeII-Cl]B(C6F5)4 (2a‚CO) (Chart 1). The complex could be
isolated as a solid. In THF solutions, 2a‚CO exhibits UV-
visible features at 412 (Soret) and 532 nm. Similar spectral
changes have been observed upon carbonylation of the “empty
tether” complex [(6L)FeII] (3) {λmax ) 424, 543 nm} to form
the corresponding carbonyl complex [(6L)FeII(CO)] (3‚CO)
{λmax ) 413 (Soret), 532 nm}.59 Similarly, the parent complex
[(F8)Fe(CO)] has electronic transitions that support the above
formulation of 2a‚CO.60 Other synthetic heme-CO and natural
heme-CO complexes show UV-vis features of a similar
nature.61,62
{ν(CO)Fe ) 1976 cm-1} or [(F8)Fe(CO)] in THF {ν(CO)Fe
)
1979 cm-1} suggests that 2a‚CO posesses a six-coordinate,
heme-CO moiety, with THF as a likely axial ligand.59,60 Axial
ligation by THF in [(F8)(thf)Fe(CO)] has been established by
an X-ray crystal structure of this six-coordinate low-spin heme
compound.60
A 1H NMR spectrum for [(6L)FeII(CO)(thf)FeII-Cl]B(C6F5)4
(4) (Supporting Information, Figure S4) recorded in THF-d8
reveals a “simplification” of the paramagnetic region relative
to the diferrous complex [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a). The
most prominent change is the loss of signals between 40 and
60 ppm and a appearance of new, broad resonance near 8.9
ppm, corresponding to the pyrrole protons. Similar upfield shifts
1
in the H NMR pyrrole resonances have been observed upon
addition of CO to the complex (6L)FeII in THF-d8 {50-60 ppm
f 9 ppm}.59 Such a change in pyrrole-proton resonances is
indicative of a spin-state change: from a five-coordinate high-
spin, ferrous heme to a six-coordinate, low-spin diamagnetic
ferrous heme-carbonyl. As already described, 2a‚CO incor-
porates an axial THF ligand derived from the bulk solvent.
However, we are unable to observe the 1H NMR signals derived
from the coordinated THF ligand, due to the expected rapid
exchange with the bulk NMR solvent. The non-heme iron site,
which is not found to bind CO, remains a high-spin ferrous
ion, with paramagnetically shifted signals similar to those
observed for [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a). Thus, this
tethered TMPA FeB site is not affected by coordination of carbon
monoxide at the adjacent heme.
Synthesis of Dinitrosyl Complex [(6L)Fe(NO)Fe(NO)-
Cl]B(C6F5)4 (2a‚(NO)2). Distillation of an excess of NO(g) onto
a frozen THF solution of [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a) and
subsequent warming to room temperature resulted in a color
change from red to red-orange. The product complex [(6L)Fe-
(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2) was isolated after re-
crystallization from CH2Cl2/heptane. Tetrahydrofuran solutions
of 2a‚(NO)2 have a UV-visible spectrum dominated by
porphyrin-based electronic transitions at 412 (Soret) and 545
nm. Similarly, toluene solutions of [(6L)Fe(NO)Fe(NO)-Cl]B-
(C6F5)4 (2a‚(NO)2) have transitions at 408 (Soret) and 542 nm.
Farmer and co-workers64 have reported an electronic absorption
spectrum for [(TPP)Fe-NO] in toluene as λmax 407 (Soret), 537
nm, while Ford and co-workers65 described a similar spectrum
for the same compound in toluene (λmax 406 (Soret), 539 nm).
These observations indicate that the heme portion of [(6L)Fe-
(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2) consists of a five coor-
dinate heme-nitrosyl, similar to (TPP)Fe-NO and related
complexes. The intense heme-based electronic transitions
An IR spectrum of 2a‚CO, obtained by evaporating a
concentrated THF solution of 2a‚CO onto an IR window,
(53) Balch, A. L. Inorg. Chim. Acta 1992, 198-200, 297-307.
(54) Balch, A. L.; Chan, Y.-W.; Cheng, R.-J.; La Mar, G. N.; Latos-Grazynski,
L.; Renner, M. W. J. Am. Chem. Soc. 1984, 106, 7779-7785.
(55) Chiou, Y.-M.; Que, J. L. J. Am. Chem. Soc. 1995, 117, 3999-4013.
(56) Lee, D.; Pierce, B.; Krebs, C.; Hendrich, M. P.; Huynh, B. H.; Lippard, S.
J. J. Am. Chem. Soc. 2002, 124, 3993-4007.
(57) Feig, A. L.; Lippard, S. J. Chem. ReV. 1994, 94, 759-805.
(58) DuBois, J.; Mizoguchi, T. J.; Lippard, S. J. Coord. Chem. ReV. 2000, 200-
202, 443-485.
(59) Kretzer, R. M.; Ghiladi, R. A.; Lebeau, E. L.; Liang, H.-C.; Karlin, K. D.
Inorg. Chem. 2003, 42, 3016-3025.
(60) Thompson, D. W.; Kretzer, R. M.; Lebeau, E. L.; Scaltrito, D. V.; Ghiladi,
R. A.; Lam, K.-C.; Rheingold, A. L.; Karlin, K. D.; Meyer, G. J. Inorg.
Chem. 2003, 42, 5211-5218.
(61) Tsuchida, E.; Komatasu, T.; Arai, K.; Hishide, H. J. Chem. Soc., Dalton
Trans. 1993, 2465-2469.
(63) Cotton, F. A.; Wilkinson, G.; Bochmann, M.; Murillo, C. AdVanced
Inorganic Chemistry, 6th ed.; John Wiley & Sons: New York, 1998.
(64) Lin, R.; Farmer, P. J. J. Am. Chem. Soc. 2001, 123, 1143-1150.
(65) Lorkovic, I.; Ford, P. C. Inorg. Chem. 2000, 39, 632-633.
(62) Collman, J. P.; Brauman, J. I.; Doxsee, K. M.; Halbert, T. R.; Bunnenberg,
E.; Linder, R. E.; La Mar, G. N.; Del Gaudio, J.; Lang, G.; Spartalian, K.
J. Am. Chem. Soc. 1980, 102, 4182-4192.
9
3316 J. AM. CHEM. SOC. VOL. 127, NO. 10, 2005

Models for the Active Site of Bacterial NOR
A R T I C L E S
preclude the observation of absorption signals which could be
assigned to the (TMPA)Fe(NO)(Cl) portion of the system, which
are expected near λmax 350 nm, but with small extinction
coeffecient (ꢀ ≈ 1000 M-1 cm-1).66
Freshly prepared THF solutions of [(6L)Fe(NO)Fe(NO)-
Cl]B(C6F5)4 (2a‚(NO)2) were used to prepare air-free samples
for MALDI mass spectrometry (see Experimental Section). The
spectrum (Supporting Information, Figure S5) is dominated by
a parent ion peak at m/z 1247 amu corresponding to [(6L)Fe-
- +
(NO)Fe(NO)-Cl]+ or a (M + H+ - B(C6F5)4
) cation. A
fragment ion peak corresponding to the loss of a single nitrosyl
ligand is not observed; rather, both nitrosyl ligands are lost to
give rise to a fragment with m/z 1186 amu which is tentatively
assigned as a (M - 2NO - B(C6F5)4-)+ cation. The appearance
of the fragment ion peak at m/z 1167 amu suggests the presence
Figure 7. FTIR spectrum for [(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2).
Non-heme Fe-NO; νNO ) 1798 cm-1. Heme nitrosyl; νNO ) 1689 cm-1
.
Table 1. Iron-Nitrosyl Compounds for Comparison to
[(6L)Fe(NO)Fe(NO)-Cl]+ (2‚(NO)2)
of an ion related to [(6L)FeIII-O-FeIII-Cl]B(C6F5)4 (1a),
specifically an (M + H+ - Cl - B(C6F5)4
)
cation. The
- +
1
complex
ν
NO, cm-
sample type
ref
observation of the µ-oxo species suggests that either; (a) the
sample is partially oxidized during transfer from the reaction
tube to the mass spectrometer; or (b) (we speculate) laser
excitation is of sufficient energy to cause the photorelease and
coupling of the nitrosyl ligands, resulting in the formation of
[(6L)FeIII-O-FeIII-Cl]+ (1) within the mass spectrometer.
[(6L)Fe(NO)2Fe(Cl)]+ (2‚(NO)2)
[(6L)Fe(NO)2Fe(Cl)]+ (2‚(NO)2)
[(TMPA)Fe(NO)(BF)]+
[(TMPzA)Fe(NO)(Cl)]+
[(Me3TACN)Fe(NO)(N3)2]
[(F20-TPP)Fe(NO)]
[(F8-TPP)Fe(NO)]
[(F8-TPP)Fe(NO)]
[(TMP)Fe(NO)]
[(TPP)Fe(NO)]
1798a
1689b
1794
1796
1690
1705
1656
1687
1676
1677c
1697d
film
film
this work
this work
66
66
71
45
this work
45
45
64
64
KBr pellet
KBr pellet
KBr pellet
KBr pellet
film
KBr pellet
KBr pellet
film
The proton NMR spectrum for [(6L)Fe(NO)Fe(NO)-Cl]B-
(C6F5)4 (2a‚(NO)2) (Supporting Information, Figure S6) reveals
an array of paramagnetically shifted signals from 159 to -50
ppm. Such a wide range of signals might be expected due to
the presence of two paramagnetic Fe-nitrosyl species. Similar
up- and downfield shifted paramagnetic NMR signals have been
reported by Que and co-workers for [(TMPA)Fe(NO)] com-
plexes.66 Likely proton assignments, by analogy to the work of
Que and co-workers, include the following: (1) methylene
-CH2- resonances at 159, 142, and 126 ppm; (2) R-H pyridyl
proton resonances at 101, 116, and 93 ppm; (3) â,â′-H pyridyl
proton resonances at 87, 76, 64, 61, 60, and 55 ppm. The
remaining proton signals (-22, -29, -50 ppm), that are found
upfield from TMS, presumably arise from the γ-H pyridyl
protons. Iron nitrosyl complexes of the formulation {FeNO}7
(see also discussion below) are paramagnetic species with a spin
delocalized electron over both the Fe ion and the NO ligand.67,68
Therefore, it should also be noted that the pyrrole and other
proton resonance from heme-nitrosyl species are often broad
and unobserved;69 hence, no definitive assignment for the pyrrole
proton resonance of [(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚
(NO)2) is given here.
[(TPP)Fe(NO)]
film
a Tentatively assigned as a non-heme Fe-NO (see text for further
discussion). b Tentatively assigned as a heme Fe-NO (see text for further
discussion). c Deposited from wet solvent. d After drying film under stream
of N2 for 1 h. Abbreviations used: TPP ) tetraphenylporphyrinate dianion;
TMP ) tetramesitylporphyrinate dianion; F20-TPP ) aryl perfluorinated
tetraphenylporphyrinate dianion; Me3TACN ) N,N′,N′′-trimethyl-1,4,7-
triazacyclononane; TMPzA ) tris((3,5-dimethyl-1-pyrazolyl)methyl)amine.
spectroscopy. FTIR spectra obtained from solid film samples
of [(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2) revealed two
new stretching vibrations, relative to the reduced [(6L)FeII‚‚‚
FeII-Cl]B(C6F5)4 (2a), at 1798 and 1689 cm-1 (Figure 7). The
νNO signal intensities are not equivalent, suggesting unequal
populations of the heme and non-heme iron-nitrosyl sites. This
is likely due to some exposure of the solid FTIR samples to
the ambient atmosphere during the spectral measurement, likely
leading to partial product decomposition. A complete bleach
of the non-heme iron-nitrosyl νNO was observed within an hour
while the heme-nitrosyl νNO decayed at a somewhat slower
rate. These bands were shifted to 1756 and 1652 cm-1 when
15NO(g) was used in sample preparation. These bands are
assigned to heme-NO and non-heme Fe-NO on the basis of
values reported in the literature for mononuclear heme-NO and
mononuclear non-heme Fe-NO complexes (Table 1). In addi-
tion to this criteria, it is known69 that dinitrosyl heme complexes
(e.g., [(P)Fe(NO)2]) do not form at room temperature and
complexes of the type [(TMPA)Fe(NO)2] would require a
second vacant coordination site at the non-heme iron.70 These
alternative structural formulations can thus be discarded.
Solid film samples of [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a) were
also prepared on the RIR ATR probe tip (see Experimental
Section). Subsequent evacuation and exposure to NO(g) dem-
onstrated the similar in situ formation of [(6L)Fe(NO)Fe(NO)-
Cl]B(C6F5)4 (2a‚(NO)2), based upon the new peaks observed
at 1798 and 1687 cm-1 (Supporting Information, Figure S7).
Complex (2a‚(NO)2) exhibits a room-temperature EPR spec-
trum with an intense resonance at g ≈ 2.0 (see Experimental
Section), which would be characteristic of heme-NO spe-
cies.45,68 An EPR signal derived from the non-heme iron-
nitrosyl site, expected at g ≈ 4,66,70 was not observed.
[(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2): Infrared
Spectroscopy. The confirmation of nitrosyl ligand coordination
at both the heme and the non-heme iron sites required infrared
(66) Chiou, Y.-M.; Que, J, L. Inorg. Chem. 1995, 34, 3270-3278.
(67) Walker, F. A. In Porphyrin Handbook; Kadish, K. M., Smith, K. M.,
Guilard, R., Eds.; Academic Press: New York, 2000; Vol. 5, pp 81-183.
(68) Cheng, L.; Richter-Addo, G. B. In The Porphyrin Handbook; Kadish, K.
M., Smith, K. M., Guilard, R., Eds.; Academic Press: New York, 2000;
Chapter 33.
(69) (a) Lorkovic, I.; Ford, P. C. J. Am. Chem. Soc. 2000, 122, 6516-6517. (b)
Wayland, B. B.; Olson, L. W. J. Am. Chem. Soc. 1974, 96, 8037-8041.
(70) Jo, D.-H.; Chiou, Y.-M.; Que, J. L. Inorg. Chem. 2001, 40, 3181-3190.
9
J. AM. CHEM. SOC. VOL. 127, NO. 10, 2005 3317

A R T I C L E S
Wasser et al.
Figure 9. Reaction of [(F8)FeII] (4) with impure NO(g) results in the in
situ formation of the compounds [(F8)Fe(NO)] (4‚NO) and [(F8)FeIII-
(NO)(NO2-)].
in the literature; a KBr pellet IR spectrum of [(F8)Fe(NO)] (4‚
Figure 8. Reaction of empty tether complex [(6L)FeII] (3) with NO(g) to
form the five-coordinate heme nitrosyl [(6L)FeII(NO)] (3‚NO). The NO
ligand is surmised to be either on the top or bottom face of the heme.
NO) exhibits a νNO ) 1687 cm-1 45
In sharp contrast to these
.
findings, the addition of “unpurified” NO(g) to a thin film of
(F8)FeII (4), which was deposited on a Si-Comp ATR window
and held under static vacuum, resulted in the formation of both
(F8)Fe-NO (ν(NO) ) 1656 cm-1; ∆(15N18O) ) 68 cm-1) and
the nitro-, nitrosyl-complex (F8)Fe(NO)(NO2) (ν(NO) ) 1868
A rough plot of absorbance against time (Supporting Informa-
tion, Figure S7) reveals that the [FeNO]2 species is formed at
the heme and non-heme FeB sites at similar rates with apparent
rapid saturation behavior. Additional clarity comes from the
addition of purified NO(g) to THF solutions of the “empty
tether” complex [(6L)FeII] (3) (Figure 8), which resulted in the
cm-1; ∆(15N18O) ) 79 cm-1 and νs(NO2) ) 1294 cm-1
;
∆(15N18O) ) 55 cm-1) (Supporting Information, Figure S9).
It has been suggested that ferrous heme-nitrosyl complexes
can promote the disproportionation of NO(g).64 Our findings
suggest that [(F8)FeII] (4) does not catalyze the disproportion-
ation of nitric oxide to nitrous oxide and nitrogen dioxide (eq
3, below). This result, the first demonstration that an electron-
deficient (i.e., with fluorinated F8 ligand) heme nitrosyl does
not promote NO(g) disproportionation, is similar to the findings
of Ford and co-workers,65 who suggest that the more electron-
rich (TPP)FeII systems do not disproportionate NO(g). These
researchers suggest that disproportionation products such as
[(TPP)Fe(NO)(NO2)] are formed due to the presence of trace
NO2(g) (e.g., Figure 9).75
clean formation of a heme nitrosyl (νNO ) 1683 cm-1
;
Supporting Information, Figure S8). This finding suggests that
the heme nitrosyl portion of [(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4
(2a‚(NO)2) is unperturbed by the presence of a non-heme iron
NO ligand.
According to Enemark-Feltham notation, the electrons in
an {MNO}n system are counted as if this fragment were
completely isolated and n corresponds to the number of metal
d-electrons plus the electron in the π* orbitals of the NO.72 Thus,
typical (P)FeII-NO compounds are formally considered {MNO}7
species. By contrast, a ferric nitrosyl (P)FeIII-NO complex could
formally be considered an {MNO}6 species. For [(6L)Fe(NO)-
Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2), there are two separate {MNO}n
fragments. The heme-nitrosyl (νNO 1689 cm-1) which is
spectroscopically similar to (TPP)Fe-NO, therefore, suggests
an {FeNO}7 formulation.73,74 Similarly, by comparison to the
work of Que and co-workers in their synthesis of (TMPA)Fe-
NO complexes, the non-heme nitrosyl (νNO 1795 cm-1) in 2a‚
(NO)2 is also an {FeNO}7 system.31,66 Thus, overall the complex
[(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2) is best described,
in terms of Enemark-Feltham notation, as [{FeNO}7]2.
Special caution must be taken to ensure the use of pure NO-
(g) in the preparation of [(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚
(NO)2). Common impurities, which can present difficulties with
the synthesis of desired nitrosyl products, include NO2(g), N2O3-
(g), and N2O(g).4 It can be misleading to assume that NO is
the strongest ligand; in fact, NO2 binds to iron complexes with
greater affinity than does NO.65,75 The addition of purified NO-
(g) (see Experimental Section) to a THF solution of [(F8)FeII]
(4) results in the clean formation of [(F8)Fe(NO)] (4‚NO), with
νNO ) 1684 cm-1 (Nujol). Similar vibrational data is reported
3NO f N2O + NO2
(3)
It should be noted that the addition of unpurified NO(g) to
solid films of [(6L)FeII‚‚‚FeII-Cl]B(C6F5)4 (2a) resulted in the
formation of an intractable mixture of heme-nitrosyl and
heme-nitrite complexes, with possible non-heme iron-nitrosyl
and non-heme iron-nitrite species. Nonetheless, from within
these reaction mixtures it was possible to observe the bands
directly related to the compound [(6L)Fe(NO)Fe(NO)-Cl]B-
(C6F5)4 (2a‚(NO)2) under most circumstances. Thus, we con-
clude that there is no NOx redox chemistry occurring in our
heme/non-heme diiron system under the conditions examined
with purified NO(g). As already pointed out by Ford,4,65 our
results and observations also show that extreme care must be
taken in the purification of NO(g) in order to prevent the
observation of anomalous artifacts and non-NO(g)-derived
chemistry caused by the presence of trace impurities.
Relevance to NOR and Attempts at Functional Modeling.
Purified bacterial NOR has been shown to exist in both an
oxidized (e.g. FeIII-O-FeIII) configuration and can be chemi-
cally reduced to the diferrous form. The latter has been most
often discussed as the “active” enzyme complex capable of nitric
oxide reductase activity,1,2 although some studies suggest that
(71) Pohl, K.; Wieghardt, K.; Nuber, B.; Weiss, J. J. Chem. Soc., Dalton Trans.
1987, 187-192.
(72) Enemark, J. H.; Feltham, R. D. Coord. Chem. ReV. 1974, 13, 339-406.
(73) Richter-Addo, G. B.; Legzdins, P. Metal Nitrosyls; Oxford University
Press: New York, 1992.
(74) Scheidt, W. R.; Ellison, M. K. Acc. Chem. Res. 1999, 32, 350-359.
(75) Kurtikyan, T. S.; Martirosyan, G. G.; Lorkovic, I. M.; Ford, P. C. J. Am.
Chem. Soc. 2002, 124, 10124-10129.
9
3318 J. AM. CHEM. SOC. VOL. 127, NO. 10, 2005

Models for the Active Site of Bacterial NOR
A R T I C L E S
the same reactions, CcO has evolved to efficiently catalyze the
reduction of O2 (250 O2/s), while NOR remains the most
efficient at catalyzing the reductive coupling of NO (200 NO/
s).
Much like bacterial NOR, our heme/non-heme diiron complex
[(6L)FeII‚‚‚FeII-Cl]+ (2) reduces dioxygen and subsequently
forms an oxidized FeIII-O-FeIII complex. NOR has been found
to bind CO at heme b3 (νCO 1970 cm-1).13,82 Similarly, [(6L)-
FeII(CO)(thf)FeII-Cl]B(C6F5)4 (2a‚CO) is a low-spin heme-
carbonyl species (νCO 1969 cm-1, film), with the non-heme iron
being unreactive toward CO, under the conditions used.
Recently, however, the Fe/Fe′ active site of NOR obtained from
Bacillus azotoformans was found to bind CO at both the heme
and non-heme iron sites. Subsequent exposure to increasing
concentrations of chloride precluded the formation of a non-
heme iron-carbonyl species.83 This finding suggests that
removal of the chloride ligand from the non-heme iron site, in
synthetic complexes such as 2, could lead to a change in
chemistry and allow the formation of a non-heme iron-
carbonyl. A chloride-free non-heme iron could also possibly
have implications for the NO-reductase activity for model
systems (vide infra).
The overall goal in the design of the complex [(6L)FeII‚‚‚
FeII-Cl]+ (2) was as a functional “model” for the reduction of
nitric oxide to nitrous oxide, as inspired by bacterial NOR.
However, [(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2) is ther-
mally stable and does not seem to react further to give nitrous
oxide upon the addition of proton sources. By analogy to the
putative mechanism for NOR-mediated NO reduction, we also
investigated the effects of adding axial ligands to [(6L)Fe(NO)-
Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2) in an effort to drive N-N
reductive coupling. Unfortunately, the addition of an excess of
1,5-dicyclohexylimidazole or (dimethylamino)pyridine to THF
solutions of [(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2) re-
sulted in the isolation of the reduced complex [(6L)FeII‚‚‚FeII-
(Cl)]+ (2) (eq 4), as confirmed by UV-visible and FTIR
spectroscopies.
Figure 10. Possible mechanism for the binding of 2 equiv of NO by the
NOR enzyme, followed by successive coupling to give nitrous oxide.
a mixed-valent form might also be involved.76,77 Researchers
have also identified both CO- and NO-bound enzyme com-
plexes.13,16,78 Two possible mechanisms for the binding and
reduction of NO at a fully reduced site in NOR are in Figure
10.2,3,13,22 Route A involves the formation of both a heme and
non-heme iron-nitrosyl species formed in close proximity to
each other. This pathway is supported by the findings that
reduced NOR incubated with NO(g) revealed weak EPR signals
near g ) 2 and 4, typical for heme nitrosyl and non-heme iron-
nitrosyl species, repsectively.2,11 As mentioned, [(6L)Fe(NO)-
Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2) does possess an EPR signal
at g ≈ 2. Route B involves the formation of a non-heme iron-
cis-dinitrosyl complex. In either case, there is subsequent
reductive coupling of the adjacent nitrosyl ligands to give nitrous
oxide. Mechanism A does not indicate formation of a µ-oxo
species, although such a complex possibly exists at some point
during catalytic turnover.16,22 While NOR is not involved in
pumping protons across a cell membrane, protons are integral
toward the working of the catalytic cycle. They are needed in
breaking the µ-oxo bridge, thereby forming water and, with the
addition of two electrons, re-forming the reduced diferrous
complex for further NO turnover.
FeII-(NO)2-Fe
2‚(NO)2
II + excess “base” f
II + 2NO(g)
II
Fe ‚‚‚Fe
2
(4)
The evolution of nitric oxide from these systems was observed
using both gas chromatography (NO detected by TCD) and by
trapping with the complex [(F8)FeII] (4) (yielding (F8)FeII-NO
by UV-visible spectroscopy). Ligand-promoted nitric oxide
dissociation from ferrous heme nitrosyl complexes has previ-
ously been reported,45 suggesting that our system is thermo-
dynamically driven to release NO (as free NO and not NO-,
the latter of which could drive N-N coupling) upon the addition
of excess axial ligands.
Additional attempts to drive N-N coupling, or to otherwise
effect NO disproportionation, were also unsuccessful. Further,
[(6L)Fe(NO)Fe(NO)-Cl]B(C6F5)4 (2a‚(NO)2) was stable in the
presence of excess nitric oxide. Attempts to form a dinitrosyl
exclusively at the non-heme iron site, to possibly induce
In addition to catalyzing the reduction of nitric oxide to
nitrous oxide, NOR has an “oxidase” activity and has been found
to catalyze the four electron reduction of dioxygen to water (40
O2/s),26,27 thus the possible relevance of our dioxygen chemistry
(vide supra). Likewise, cytochrome c oxidase (CcO) has been
shown to catalyze the reduction of NO to N2O (0.5 NO/s at
best).79-81 While both CcO and NOR are capable of catalyzing
(76) Gronberg, K. L. C.; Roldan, M. D.; Prior, L.; Butland, G.; Cheesman, M.
R.; Richardson, D. J.; Spiro, S.; Thomson, A. J.; Watmough, N. J.
Biochemistry 1999, 38, 13780-13786.
(77) Groenberg, K. L. C.; Watmough, N. J.; Thomson, A. J.; Richardson, D. J.;
Field, S. J. J. Biol. Chem. 2004, 17, 17120-17125.
(81) Giuffre, A.; Stubauer, G.; Sarti, P.; Brunori, M.; Zumft, W. G.; Buse, G.;
Soulimane, T. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 14718-14723.
(82) Matsuda, Y.; Uchida, T.; Hori, H.; Kitagawa, T.; Arata, H. Biochim.
Biophys. Acta 2004, 1656, 37-45.
(83) Lu, S.; Suharti; de Vries, S.; Moe¨nne-Loccoz, P. J. Am. Chem. Soc. 2004,
126, 15332-15333.
(78) Cheesman, M. R.; Zumft, W. G.; Thomson, A. J. Biochemistry 1998, 37,
3994-4000.
(79) Brudvig, G. W. S., T. H.; Chan, S. I. Biochemistry 1980, 19, 5275-5285.
(80) Brunori, M. Trends Biochem. Sci. 2001, 26, 21-23.
9
J. AM. CHEM. SOC. VOL. 127, NO. 10, 2005 3319

A R T I C L E S
Wasser et al.
coupling, were also unsuccessful. In such reactions, a silver
pseudohalide (e.g., AgOTf, AgSbF6) was added prior to the
introduction of NO to frozen solutions of [(6L)FeII‚‚‚FeII-(Cl)]+
(2), hoping to remove the non-heme chloride ligand in situ,
thereby creating a vacant labile site and enhancing reactivity.
But, these reactions failed to yield the desired nitrous oxide
product. As mentioned, the current mechanism postulated for
the reductive coupling of NO to give N2O by NOR is “catalysis
by approximation”. According to this view, adjacent iron-
nitrosyls have sufficient driving force to couple (with additional
donated electrons) and yield N2O. The fact that we could not
drive the coupling of adjacent stable iron-nitrosyls suggests
that the putative biological mechanism may be overly simplified,
and that other reaction mechanisms need to be contemplated
and tested. Planned second-generation model systems might
benefit by (1) incorporating an intramolecularly tethered axial
base and (2) altering the non-heme iron site ligand donor set,
including that it be chloride free.
complex is the thermal decay by disproportionation (and
evolution of 1/2 O2) to give the µ-oxo complex [(6L)FeIII-O-
FeIII-Cl]+ (1). We suggest that an iron-peroxo-iron species,
of an inter- or intramolecular nature, likely exists as a short-
lived intermediate en route to oxo formation.
Similarly, the CO and NO reactivity of coordination com-
pounds inspired by the reduced NOR active site has been
developed. [(6L)FeII‚‚‚FeII-(Cl)]+ (2) reacts with CO(g) or NO-
(g) to give a stable monocarbonyl species [(6L)Fe(CO)Fe-Cl]+
(2‚CO) or dinitrosyl species [(6L)Fe(NO)Fe(NO)-Cl]+
(2‚(NO)2), respectively. Unfortunately, from the standpoint of
nitric oxide reduction, the intrinsic heme-NO and (TMPA)-
Fe-NO functionalities (i.e., one heme; one non-heme) are both
stable when tethered (i.e., here in 2‚(NO)2) or otherwise exposed
to each other. Further efforts in our laboratory are aimed at
functional modeling of NOR via the design and study of new
ligand systems for heme/non-heme diiron chemistry.
Acknowledgment. This work was supported by National
Institutes of Health Grants GM60353 (to K.D.K.) and GM18865
(to P.M.-L.).
Summary/Conclusions
The heme/non-heme diiron active site of NOR is a (quite)
new prosthetic group active site entity in bioinorganic enzymol-
ogy. The first synthetic model for the reduced diiron center is
presented here, along with biologically relevant chemical
reactivity including interactions with O2, CO and NO gases.
The dioxygen adducts and dioxygen reduction chemistry
observed functionally models the NOR O2 chemistry, as [(6L)-
FeII‚‚‚FeII-(Cl)]+ (2) readily reacts with 1/2 O2 to give the
“resting” or oxidized species [(6L)FeIII-O-FeIII-Cl]+ (1). This
two-electron reduction of dioxygen to an oxo (O2-) product
involves the initial formation of a heme-superoxo species which
was detected at low temperature by both UV-visible and RR
spectroscopies. This represents the first direct detection of a
heme-superoxo intermediate in heme/non-heme diiron chem-
istry en route to the formation of a µ-oxo bridge heme/non-
heme complex. The inevitable fate of the heme-superoxo
Supporting Information Available: High-frequency RR
spectra for the reaction of dioxygen with 2 (Figure S1), high-
frequency RR spectra for the reaction of dioxygen with the
empty tether complex [(6L)FeII] (3) in THF (Figure S2), FTIR
1
spectrum for 2a‚CO (Figure S3), H NMR spectrum for 2a‚
CO (Figure S4), MALDI-TOF-MS for 2‚(NO)2 (Figure S5), 1H
NMR spectrum of 2‚(NO)2 (Figure S6), in situ RIR monitoring
of the reaction of solid films of 2a with NO(g) (Figure S7) and
solid films of [(6L)FeII] (3) with NO(g) (Figure S8), and the
FTIR difference spectra for [(F8)Fe] (4) + unpurified 14NO/
15NO (Figure S9). This material is available free of charge via
the Internet at http://pubs.acs.org.
JA0458773
9
3320 J. AM. CHEM. SOC. VOL. 127, NO. 10, 2005