A low-spin alkylperoxo-iron(III) complex with weak Fe-O and O-O bonds: Implications for the mechanism of superoxide reductase

Article abstract of DOI:10.1021/ja064525o

The synthesis of a mononuclear, five-coordinate ferrous complex [([15]aneN₄)Fe^II(SPh)](BF₄) (1) is reported. This complex is a new model of the reduced active site of the enzyme superoxide reductase (SOR), which is comprised of a [(N_His)₄(S_cys)Fe^II] center. Complex 1 reacts with alkylhydroperoxides (tBuOOH, cumenylOOH) at low temperature to give a metastable, dark red intermediate (2a: R = tBu; 2b: R = cumenyl) that has been characterized by UV-vis, EPR, and resonance Raman spectroscopy. The UV-vis spectrum (-80 °C) reveals a 526 nm absorbance (ε = 2150 M^-1 cm^-1) for 2a and a 527 nm absorbance (ε = 1650 M^-1 cm^-1) for 2b, indicative of alkylperoxo-to-iron(III) LMCT transitions, and the EPR data (77 K) show that both intermediates are low-spin iron(III) complexes (g = 2.20 and 1.97). Definitive identification of the Fe(III)-OOR species comes from RR spectra, which give ν(Fe-O) = 612 (2a) and 615 (2b) cm^-1, and ν(O-O) = 803 (2a) and 795 (2b) cm^-1. The assignments for 2a were confirmed by ¹⁸O substitution (tBu¹⁸O¹⁸OH), resulting in a 28 cm^-1 downshift for ν(Fe-¹⁸O), and a 46 cm^-1 downshift for ν(¹⁸O-¹⁸O). These data show that 2a and 2b are low-spin Fe^III-OOR species with weak Fe-O bonds and suggest that a low-spin intermediate may occur in SOR, as opposed to previous proposals invoking high-spin intermediates. Copyright

Full text of DOI:10.1021/ja064525o

Published on Web 10/14/2006
A Low-Spin Alkylperoxo-Iron(III) Complex with Weak Fe-O and O-O Bonds:
Implications for the Mechanism of Superoxide Reductase
Divya Krishnamurthy,† Gary D. Kasper,† Frances Namuswe,† William D. Kerber,†
Amy A. Narducci Sarjeant,† Pierre Moe¨nne-Loccoz,*,‡ and David P. Goldberg*,†
Department of Chemistry, Johns Hopkins UniVersity, Baltimore, Maryland 21218, and Department of
EnVironmental and Biomolecular Systems, OGI School of Science and Engineering, Oregon Health & Science
UniVersity, BeaVerton, Oregon, 97006
Received June 26, 2006; E-mail: dpg@jhu.edu; ploccoz@ebs.ogi.edu
Until recently, the removal of superoxide anion (O2-) from the
biological milieu was believed to take place almost exclusively
through the action of the superoxide dismutases. The discovery of
superoxide reductase1 (SOR), a mononuclear iron enzyme that
-
converts O2 to H2O2 via a one-electron reduction pathway, has
altered this view. The active site of SOR is unique, being comprised
of a [(NHis)4(Scys)FeII] center (reduced active form, SORred). The
four neutral N donors are positioned in the equatorial plane of a
square pyramid, while the thiolate donor occupies an axial position
-
trans to the putative O2 binding site. Thus, the iron coordination
Figure 1. (a) ORTEP diagram of the cation of 1 showing 50% probability
ellipsoids with some of the hydrogen atoms omitted for clarity. Selected
sphere in SOR resembles that of the heme iron in cytochrome P450.
Spectroscopic evidence suggests that an FeIII-OO(H) intermediate
bond lengths (Å): Fe(1)-N(1) 2.164(3), Fe(1)-N(2) 2.273(3), Fe(1)-N(3)
forms during SOR turnover,1 but unlike in P450, where this species
2.138(3), Fe(1)-N(4) 2.206(3), Fe(1)-S(1) 2.3316(11). (b) Packing diagram
highlighting the intermolecular NH-S hydrogen bonds.
undergoes O-O bond cleavage to form a high-valent FedO species,
the hydroperoxo group is released as H2O2 after protonation. It has
been suggested that the spin state of the FeIII-OO(H) intermediate
Scheme 1
1
5
(S ) /2 vs /2) may play an important role in determining the
outcome of this reaction,1b,2 but the mechanism of action of SOR
and, in particular, the factors that account for the dramatically
different outcomes of the chemistry of SOR versus P450 remain
to be determined.
A few model complexes of SOR have recently emerged,
including an example of an FeIII-OOH species (S ) 1/2) that
contains thiolate ligation3 and a high-spin [N4Sthiolate]FeIII-OOtBu
complex.2d Here, we report the synthesis of a novel analogue of
SORred, [([15]aneN4)FeII(SPh)]BF4 (1), and its reactivity with
alkylperoxides. Specifically, we have characterized a low-spin
[N4Sthiolate]FeIII-OOR intermediate which exhibits a bonding pattern
that is dramatically different from those of all previously character-
ized low-spin FeIII-OOR species.
The iron(II) complex 1 is obtained by reaction of [15]aneN4 with
FeII(BF4)2 followed by addition of NaSPh (Scheme 1). The structure
of 1 (Figure 1) reveals that a single phenylthiolate ligand is
coordinated to the iron(II) center, resulting in a five-coordinate
geometry between that of square pyramidal (sp) and trigonal
bipyramidal (tbp) (τ ) 0.49).4 The four neutral nitrogen donors
together with an N(2)-H(2)-S(1′) angle of 158(3)°, provide good
evidence for the presence of an intermolecular N-H-S hydrogen
bond. It is not known at this time if these intermolecular H-bonds
persist in solution, but it is worth noting that, in SOR, the
coordinating SCys appears to be hydrogen bonded to two N-Hpeptide
groups.1a
Reaction of 1 with the alkyl hydroperoxides tBuOOH or
cumeneOOH in CH2Cl2 at low temperature (-78 °C) leads to the
formation of dark red intermediates 2a (tBuOOH) and 2b (cume-
nylOOH) (Scheme 1). These dark red species, not observed at room
temperature, exhibit relatively short lifetimes at -80 °C (2a: kobs
) 5.4 × 10-3 s-1), persisting for ∼10 min at -80 °C. This brief
window of stability was enough to characterize the red intermediate
by low-temperature UV-vis spectroscopy. The red intermediate
for the tBuOOH reaction gives rise to a 526 nm absorbance (ꢀ )
2150 M-1 cm-1 assuming total conversion of 1 to 2a) (Figure 2).
In comparison, the cumenyl derivative exhibits a λmax at 527 nm
(ꢀ ) 1650 M-1 cm-1). Both the position and intensity of these
bands are indicative of alkylperoxo-to-iron(III) LMCT transitions,
suggesting that the red intermediate is an FeIII-OOR complex.2d,7
The EPR spectra of 2a,b reveal characteristic patterns of low-
spin iron(III) complexes, with g ) 2.20 and 1.97 (Figure S1).8
Definitive identification of the chromophoric intermediate as an
FeIII-OOR species is provided by resonance Raman (RR) spectra
obtained with a 514 nm excitation in the transient absorption. The
RR spectrum of 2a exhibits bands at 439, 483, 612, and 803 cm-1
and one thiolate ligand match the donor set found in SORred
,
although the geometry is closer to sp in the enzyme. A similar self-
assembly reaction was used to prepare the SORred model [(Me4-
cyclam)FeII(SPh-p-OMe)]+, which also exhibits a distorted geom-
etry between sp and tbp (τ ) 0.50).5a The Fe-N and Fe-S bond
lengths for 1 are in line with other FeII-N and FeII-S bond
distances for high-spin iron(II).6 The N-H groups are arranged so
that H(2) points toward the axial thiolate ligand, while H(1), H(3),
and H(4) lie on the opposite side of the macrocyclic plane. The
NH(2)-S(1′) and N(2)-S(1′) distances of 2.60(5) and 3.489(4) Å,
† Johns Hopkins University.
‡ Oregon Health & Science University.
9
14222
J. AM. CHEM. SOC. 2006, 128, 14222-14223
10.1021/ja064525o CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
agreement with the presumed mechanism of cytochrome P450 in
which a low-spin FeIII-OO(H) leads to O-O cleavage and
formation of a high-valent FedO intermediate. In SOR, a high-
spin Fe3+-OO(H) intermediate with a weak Fe-O bond could favor
H2O2 release rather than O-O bond cleavage.2
The vibrational signatures of intermediates 2a and 2b show that
these FeIII-OOR species exhibit weak Fe-O bonds, in sharp
contrast with the data from other low-spin FeIII-OOR complexes.
The weaker Fe-O bonds in 2 may be due to a trans effect from
the sulfur ligand, provided it is coordinated trans and not cis to the
peroxide.9 The sulfur ligand is believed to remain coordinated to
the iron in 2 since preliminary experiments with substituted
arylthiolate ligands show shifted transient absorption maxima (data
not shown). Interestingly, recent DFT calculations have determined
that a low-spin FeIII-OOH intermediate should be energetically
favored for SOR.2a Our results support these predictions and the
possibility of a low-spin intermediate with a weak Fe-O bond in
Figure 2. UV-vis spectra of 1 (black, dashed line) and 2a (blue, solid
line) in CH2Cl2 at -80 °C, and RR spectra of 16O-2a (A) and 18O-2a
(B). The 16O-18O difference spectrum (C, green) is overlapped with a
simulated trace composed of Gaussian peaks (C, red).
Table 1. Spin States and Vibrational Data for FeIII-OOR Species
-
the course of O2 reduction by SOR.
spin
νFe
νO
-
-
(cm-
O
O
1
Acknowledgment. We thank the NIH for financial support
(D.P.G., GM62309; P.M.L., GM074785). W.D.K. is grateful for a
Dreyfus Environmental Postdoctoral Fellowship. We thank K. D.
Karlin for the use of his low-temperature UV-vis instrumentation.
1
complex
statea
)
(cm-
)
ref
[FeIII([15]aneN4)(SPh)(OOtBu)]+
[FeIII([15]aneN4)(SPh)(OOCm)]+
[FeIII(TPA)(OHx)(OOtBu)]x+
[FeIII(L8py2)(SAr)(OOtBu)]+
LS
LS
LS
HS
612
615
696
623
803
795
796
b
b
7f
2d
830, 874
Supporting Information Available: Experimental details, EPR
spectra, RR data, and X-ray structure files for 1 (CIF). This material is
available free of charge via the Internet at http://pubs.acs.org.
a LS ) low-spin (S ) /2); HS ) high-spin (S ) /2). b This work.
1
5
(Figures 2A and S2-A). Intermediate 2b displays similar frequencies
at 430, 490, 615, and 795 cm-1 (Figure S2-B). On the basis of
earlier studies,7 signals below 500 cm-1 can be assigned to (C-
C-C) and (C-C-O) deformation modes of the alkylperoxo ligand.
The bands at 803 and 795 cm-1 are within the expected range of
O-O stretching vibrations in metal-alkylperoxo complexes, and
they compare well with other low-spin FeIII-OOR species (Table
1). In contrast, O-O stretching modes from high-spin FeIII-OOR
species are higher in energy, as seen in Table 1. RR bands at 612
cm-1 for 2a and 615 cm-1 for 2b are consistent with Fe-O
stretching vibrations, but they are dramatically lower in energy than
those observed for low-spin FeIII-OOR adducts.2d,7
To confirm these assignments, intermediate 2a was prepared with
tBu18O18OH. The RR spectrum of 18O-labeled 2a is shown in Figure
2B (middle), along with the 18O-16O difference spectrum. The ν-
(Fe-18O) is observed at 584 cm-1 which represents a 28 cm-1 18O-
downshift. The ν(18O-18O) is downshifted 46 cm-1 and splits as a
Fermi doublet centered at 757 cm-1 (Figure 2B). These 18O-shifts
are in complete agreement with those expected for isolated Fe-O
and O-O diatomic oscillators. In contrast, vibrations observed
below 500 cm-1 show only marginal 18O-shifts, consistent with
their assignment to alkyl C-C-C and C-C-O deformation modes
(Figure S3). The perfect match between observed 18O-shifts and
predicted values for isolated diatomics supports an analysis of the
observed ν(Fe-O) and ν(O-O) frequencies as group vibrations
and permits correlations between vibrational frequencies and bond
strengths.
References
(1) (a) Kurtz, D. M., Jr. Acc. Chem. Res. 2004, 37, 902-908. (b) Mathe´, C.;
Nivie`re, V.; Houe´e-Levin, C.; Mattioli, T. A. Biophys. Chem. 2006, 119,
38-48. (c) Clay, M. D.; Yang, T. C.; Jenney, F. E., Jr.; Kung, I. Y.;
Cosper, C. A.; Krishnan, R.; Kurtz, D. M., Jr.; Adams, M. W. W.;
Hoffman, B. M.; Johnson, M. K. Biochemistry 2006, 45, 427-438. (d)
Yeh, A. P.; Hu, Y.; Jenney, F. E., Jr.; Adams, M. W. W.; Rees, D. C.
Biochemistry 2000, 39, 2499-2508.
(2) (a) Silaghi-Dumitrescu, R.; Silaghi-Dumitrescu, I.; Coulter, E. D.; Kurtz,
D. M., Jr. Inorg. Chem. 2003, 42, 446-456. (b) Clay, M. D.; Cosper, C.
A.; Jenney, F. E., Jr.; Adams, M. W. W.; Johnson, M. K. Proc. Natl.
Acad. Sci. U.S.A. 2003, 100, 3796-3801. (c) Kovacs, J. A. Chem. ReV.
2004, 104, 825-848. (d) Bukowski, M. R.; Halfen, H. L.; van den Berg,
T. A.; Halfen, J. A.; Que, L., Jr. Angew. Chem., Int. Ed. 2005, 44, 584-
587.
(3) Theisen, R. M.; Kovacs, J. A. Inorg. Chem. 2005, 44, 1169-1171.
(4) Addison, A. W.; Rao, T. N.; Reedjik, J.; van Rijn, J.; Verschoor, G. C. J.
Chem. Soc., Dalton Trans. 1984, 1349-1456.
(5) (a) Fiedler, A. T.; Halfen, H. L.; Halfen, J. A.; Brunold, T. C. J. Am.
Chem. Soc. 2005, 127, 1675-1689. (b) For examples of related zinc(II)
complexes, see: Notni, J.; Gorls, H.; Anders, E. Eur. J. Inorg. Chem.
2006, 1444-1455.
(6) Krishnamurthy, D.; Sarjeant, A.; Goldberg, D. P.; Caneschi, A.; Totti,
F.; Zakharov, L. N.; Rheingold, A. L. Chem.sEur. J. 2005, 11, 7328-
7341 and references therein.
(7) (a) Menage, S.; Wilkinson, E. C.; Que, L., Jr.; Fontecave, M. Angew.
Chem., Int. Ed. 1995, 34, 203-205. (b) Zang, Y.; Kim, J.; Dong, Y. H.;
Wilkinson, E. C.; Appelman, E. H.; Que, L., Jr. J. Am. Chem. Soc. 1997,
119, 4197-4205. (c) Wada, A.; Ogo, S.; Watanabe, Y.; Mukai, M.;
Kitagawa, T.; Jitsukawa, K.; Masuda, H.; Einaga, H. Inorg. Chem. 1999,
38, 3592-3593. (d) Girerd, J. J.; Banse, F.; Simaan, A. J. Struct. Bonding
2000, 97, 145-177. (e) Lehnert, N.; Ho, R. Y. N.; Que, L., Jr.; Solomon,
E. I. J. Am. Chem. Soc. 2001, 123, 12802-12816. (f) Lehnert, N.; Ho, R.
Y. N.; Que, L., Jr.; Solomon, E. I. J. Am. Chem. Soc. 2001, 123, 8271-
8290. (g) Lehnert, N.; Fujisawa, K.; Solomon, E. I. Inorg. Chem. 2003,
42, 469-481.
(8) Quantitation of the low-spin EPR signal reveals that it accounts for 10-
30% of the starting iron(II) complex. This relatively low value is likely
due to decay of the FeIII-OOR species during the manual mixing time in
the EPR tube. In contrast, the UV-vis spectra for 2 are obtained within
a few seconds after addition of ROOH. Moreover, the ∼5-fold increase
in concentration of reactants in the EPR experiments compared to those
used in the UV-vis will accelerate intermolecular decay pathways. Thus,
it is reasonable to estimate ꢀ values for 2 based on 100% conversion despite
the EPR quantitation measurements.
Many mononuclear non-heme iron enzymes have been proposed
to proceed via FeIII-OOH reaction intermediates. Experimental and
theoretical work on enzymes and models have led to a widely
invoked hypothesis regarding spin state and Fe-O and O-O bond
strengths in FeIII-OOR complexes.2d,7 Low-spin intermediates have
been shown to exhibit high ν(Fe-O) and low ν(O-O), suggesting
strong Fe-O and weak O-O bonds, respectively, while high-spin
species exhibit the opposite pattern (Table 1). This trend is in
(9) Deuteration of 1 at the N-H positions does not change the ν(Fe-O),
ruling out hydrogen bonding as the cause of the low frequency.
JA064525O
9
J. AM. CHEM. SOC. VOL. 128, NO. 44, 2006 14223