Iron-porphyrin NO complexes with covalently attached N-donor ligands: Formation of a stable six-coordinate species in solution

Article abstract of DOI:10.1021/ja904368n

A series of substituted tetraphenylporphyrin type macrocycles (TMP or To-F₂PP) with covalently attached N-donor ligands (pyridine or imidazole linker) have been synthesized. Linkers with varying chain lengths and designs have been applied to systematically investigate the effect of chain length and rigidity on the binding affinity of the linker to the corresponding Fe(II)-NO heme complexes. The binding of the linker is monitored in solution using a variety of spectroscopic methods including UV-vis absorption, EPR, and IR spectroscopy. Both the N-O stretching frequency and the imidazole ¹⁴N hyperfine coupling constants show a good correlation with the Fe-(N-donor) bond strength in these systems. The complexes with covalently attached pyridyl and alkyl imidazole ligands only exhibit weak interactions of the linker with iron(II). However, the stable six-coordinate complex [Fe(To-F₂PP-BzIM)(NO)] (4) is obtained when a rigid benzyl linker is applied. This complex exhibits typical properties of six-coordinate ferrous heme-nitrosyls in which an N-donor ligand is bound trans to NO, including the Soret band at 427 nm and the typical nine line ¹⁴N hyperfine splitting in the EPR spectrum. A crystal structure has been obtained for the corresponding zinc complex. Here, we report the first systematic study on the requirements for the formation of stable six-coordinate ferrous heme nitrosyl complexes in solution at room temperature in the absence of excess axial N-donor ligand.

Full text of DOI:10.1021/ja904368n

Published on Web 11/06/2009
Iron-Porphyrin NO Complexes with Covalently Attached
N-Donor Ligands: Formation of a Stable Six-Coordinate
Species in Solution
Timothy C. Berto, V. K. K. Praneeth, Lauren E. Goodrich, and Nicolai Lehnert*
Department of Chemistry, UniVersity of Michigan, Ann Arbor, Michigan 48109
Received May 29, 2009; E-mail: lehnertn@umich.edu
Abstract: A series of substituted tetraphenylporphyrin type macrocycles (TMP or To-F2PP) with covalently
attached N-donor ligands (pyridine or imidazole linker) have been synthesized. Linkers with varying chain
lengths and designs have been applied to systematically investigate the effect of chain length and rigidity
on the binding affinity of the linker to the corresponding Fe(II)-NO heme complexes. The binding of the
linker is monitored in solution using a variety of spectroscopic methods including UV-vis absorption, EPR,
and IR spectroscopy. Both the N-O stretching frequency and the imidazole 14N hyperfine coupling constants
show a good correlation with the Fe-(N-donor) bond strength in these systems. The complexes with
covalently attached pyridyl and alkyl imidazole ligands only exhibit weak interactions of the linker with
iron(II). However, the stable six-coordinate complex [Fe(To-F2PP-BzIM)(NO)] (4) is obtained when a rigid
benzyl linker is applied. This complex exhibits typical properties of six-coordinate ferrous heme-nitrosyls in
which an N-donor ligand is bound trans to NO, including the Soret band at 427 nm and the typical nine line
14N hyperfine splitting in the EPR spectrum. A crystal structure has been obtained for the corresponding
zinc complex. Here, we report the first systematic study on the requirements for the formation of stable
six-coordinate ferrous heme nitrosyl complexes in solution at room temperature in the absence of excess
axial N-donor ligand.
Introduction
(IM) type ligands that are covalently tethered to the porphyrin
core in order to model this proximal histidine.4
Heme proteins are involved in many important biological
processes, including electron transfer, catalysis, and signaling.
Many of these functions involve the interaction of ferrous hemes
with diatomics, in particular O2, NO, and CO.1 Because of the
exceptional significance of heme diatomic interactions, much
research has been devoted to the synthesis of model complexes
for the corresponding heme proteins. In many cases, a proximal
histidine is present as the trans ligand to the diatomic bound at
the distal site. Prominent examples of this structural motif are
found in the O2 transport and storage proteins hemoglobin (Hb)
and myoglobin (Mb), the NO sensor soluble guanylate cyclase
(sGC), and peroxidases.2 A similar active site with a copper
center in close proximity is also found in cytochrome c oxidase
(CcO). Here, the heme-copper center binds and reduces dioxy-
gen, and thus facilitates respiration within living organisms.
Finally, the active site of bacterial nitric oxide reductase
(NorBC) contains a heme with axial histidine coordination,
which has been proposed to influence the binding and reactivity
of NO bound at the distal site.3 Because of this, many synthetic
model complexes, for example those of Mb, Hb, CcO, and
NorBC, have been equipped with pyridine (Py) or imidazole
Model systems of biological hemes that include the proximal
histidine usually consist of a modified synthetic porphyrin with
a covalently tethered N-donor ligand.4 Original work in this
field, carried out by Traylor and co-workers, led to the
development of the so-called “cyclophane” porphyrins5 in which
alkyl imidazole or thiolate linked chains were anchored at the
ꢀ-pyrrole positions of protoheme.6,4d,7 These models served as
mimics for the active sites of globins and peroxidases. In
separate studies, tethers at the ꢀ-pyrrole positions of porphyrin
ligands have also been used to covalently connect two separate
hemes via peptide linkages.8 Work by Dolphin and co-workers
focused on the investigation of the electronic structure of such
dimeric porphyrin species as a function of the length of the
(4) See refs in: (a) Walker, F. A.; Simonis, U. Iron Porphyrin Chemistry.
In: Encyclopedia of Inorganic Chemistry, 2nd ed.; King, R. B., Ed.;
John Wiley & Sons, Ltd.: Chichester, 2005; Vol. IV, p 2390. (b)
Collman, J. P.; Boulatov, R. ; Sunderland, C. J.; Fu, L. Chem. ReV.
2004, 104, 561. (c) Kim, E.; Chufa´n, E. E.; Kamaraj, K.; Karlin, K. D.
Chem. ReV. 2004, 104, 1077. (d) Traylor, T. G. Acc. Chem. Res. 1981,
14, 102.
(5) Traylor, T. G.; Diekmann, H.; Chang, C. K. J. Am. Chem. Soc. 1971,
93, 4068–4070.
(1) The Smallest Biomolecules: Diatomics and their Interactions with
Heme Proteins; Ghosh, A., Ed.; Elsevier: Amsterdam, 2008.
(2) (a) Garbers, D. L.; Lowe, D. G. J. Biol. Chem. 1996, 269, 30741–
30744. (b) Waldman, S. A.; Murad, F. Pharmacol. ReV. 1987, 39,
163–196. (c) Riggs, A. F. Curr. Opin. Struct. Biol. 1991, 1, 915. (d)
Hersleth; et al. Inorg. Chim. Acta 2008, 361, 831–843.
(3) Moe¨nne-Loccoz, P.; de Vries, S. J. Am. Chem. Soc. 1998, 120, 5147–
5152.
(6) (a) Traylor, T. G.; Chang, C. K.; Geibel, J.; Berzinis, A.; Mincey, T.;
Cannon, J. J. Am. Chem. Soc. 1979, 101, 6716–6731. (b) Chang, C. K.;
Traylor, T. G. Proc. Natl. Acad. Sci. U.S.A. 1973, 70, 2847–2650. (c)
Chang, C. K.; Traylor, T. G. J. Am. Chem. Soc. 1973, 95, 5810–5811.
(7) Traylor, T. G. Pure Appl. Chem. 1991, 63, 265–274.
(8) Schwarz, F. P.; Gouternam, M.; Muljiani, Z.; Dolphin, D. H. Bioinorg.
Chem. 1972, 2, 1–32.
9
17116 J. AM. CHEM. SOC. 2009, 131, 17116–17126
10.1021/ja904368n CCC: $40.75  2009 American Chemical Society

Formation of a Stable Six-Coordinate Species
A R T I C L E S
(CH2)n tether (n ) 0-8).9 Later, Momenteau and co-workers
were able to improve the solubility of imidazole-tethered
protohemes through the use of propionic acid side chains.6b,10
While a considerable amount of insight has been gained from
these ꢀ-pyrrole and propionic acid tethered model systems, they
have been shown to exhibit undesirable intermolecular binding
through the tethered donor ligands.10 As an alternative, the use
of tetraphenylporphyrin derivatives offers a convenient strategy
for attaching such tethers by substituting the ortho positions of
the meso-phenyl groups available in these systems. These
models offer the distinct benefit that, due to the perpendicular
orientation of the phenyl rings with respect to the porphyrin
plane, ortho-phenyl substituents are conveniently directed
toward the axial positions of the heme center and thus,
intermolecular interactions are avoided. Collman and co-workers
have shown the utility of such systems through their use of
multitethered “picket fence porphyrins” which employ either
ether or amide linkages.11-13 Walker and co-workers have also
utilized such porphyrin model systems to elucidate the properties
of cytochromes c, b, and a3 with respect to axial N-donor ligand
geometry.14,4a,15 Recently, model systems from the Collman,
Karlin, and Naruta groups have continued to exploit the
effectiveness of meso-phenyl tethered porphyrin systems in their
models for the active site of cytochrome c oxidase.16-19
Our particular interest is focused on bacterial nitric oxide
reductase (NorBC) which is an enzyme found in soil dwelling
bacteria that is responsible for the conversion of nitric oxide
(NO) to nitrous oxide (N2O) via a two-electron reduction:
seen in the heme active sites of Hb and Mb. The dimetallic
heme/nonheme motif is catalytically active in the diferrous
form.20 Detailed investigations into the properties and reactivity
of the heme component of the active site of NorBC could, in
principle, be based on the Mb and Hb model complexes
described above. However, in the case of NorBC, the interaction
of NO, rather than O2 or CO, with ferrous heme model
complexes needs to be studied to arrive at a detailed structural
and mechanistic understanding of this enzyme. This has
important consequences for the design of model systems as the
generation of six-coordinate (6C) ferrous heme nitrosyls con-
stitutes a significant challenge.
The question of axial ligand binding is of direct relevance
for the reactivity of ferrous heme nitrosyls. It has been shown
that the trans ligand modulates the amount of radical character
on the NO, and hence, the chemical behavior of these
complexes. This is evident from spectroscopic studies including
EPR, MCD, vibrational spectroscopy (coupled to normal
coordinate analysis), and DFT calculations on five- and six-
coordinate ferrous heme nitrosyl model systems.22 In five-
coordinate (5C) complexes a strong Fe-NO σ-bond is present
2
between the singly occupied π* orbital of NO and dz of
Fe(II).22a,b Additional backbonding is observed between the dxz
and dyz orbitals of iron and the remaining unoccupied π* orbital
of NO. The strong σ-bond and substantial sharing of the
2
unpaired electron via the dz orbital of iron gives rise to the
strong σ-trans interaction between NO and the proximal
N-donor ligand in the corresponding 6C complexes. This has
two consequences: (a) the binding of axial ligands trans to NO
is weak, and (b) upon coordination of an N-donor ligand trans
to NO, the Fe-NO bond is weakened and the unpaired electron
is pushed back from the iron(II) to the NO ligand resulting in
an electronic structure with Fe(II)-NO(radical) character in the
6C case.22a,b In this way, the N-donor ligand could help to
activate the bound NO for catalysis. This is particularly relevant
for the activation of NO in NorBC since ferrous heme nitrosyls
are intrinsically stable and unreactive.23
To investigate this point further, 6C ferrous heme nitrosyl
model complexes that are stable in solution at room temperature
are needed. This is challenging because the binding constants
of N-donor ligands trans to NO are generally very small (Keq
≈ 1 to 30 M-1) due to the σ-trans effect detailed above.22b,24
This is very different compared to CO and O2 complexes where
such a trans effect is lacking. Correspondingly, a recent report
on Fe(II)-NO complexes of protoheme with covalently linked
IM shows that these complexes are indeed only 5C in solution.25
In fact, only one model complex is known so far where the
covalently tethered N-donor ligand seems to remain bound to
iron(II) after coordination of NO without the formation of 5C
species in solution.26 Systematic investigations to optimize axial
ligand binding properties in tailed ferrous heme nitrosyl model
systems are completely lacking. It is apparent from the literature
2NO + 2e- + 2H+ f N2O + H2O
(1)
This enzyme fulfills a vital role in the process of denitrifi-
cation where nitrate is reduced in a stepwise fashion to
dinitrogen.20 The site of catalytic NO reduction within the
enzyme consists of a dinuclear iron center with both heme and
nonheme type coordination. The nonheme iron site has three
histidine ligands and has been proposed to also contain glutamate
ligation.21 Located 3.5 Å from the nonheme iron is a heme b
site with additional proximal histidine ligation similar to that
(9) Paine, J. B., III; Dolphin, D. Can. J. Chem. 1978, 56, 1710–1712.
(10) Momenteau, M.; Rougee, M.; Loock, B. Eur. J. Biochem. 1976, 71,
63–76.
(11) Collman, J. P.; Gagne, R. R.; Halbert, T. R.; Marchon, J. C.; Redd,
C. A. J. Am. Chem. Soc. 1973, 95, 7868–7870.
(12) Collman, J. P.; Gagne, R. R.; Reed, C. A. J. Am. Chem. Soc. 1974,
96, 2629–2631.
(13) Gerothanassis, I. P.; Momenteau, M.; Barrie, P. J.; Kalodimos, C. G.;
Hawkes, G. E. Inorg. Chem. 1996, 35, 2674–2679.
(14) Safo, M. K.; Gupta, G. P.; Walker, F. A.; Scheidt, W. R. J. Am. Chem.
Soc. 1991, 113, 5497.
(15) Walker, F. A.; Reis, D.; Balke, V. L. J. Am. Chem. Soc. 1984, 106,
6888–6898.
(16) Kopf, M. A.; Karlin, K. D. Inorg. Chem. 1999, 38, 4922–4923.
(17) Kim, E.; Shearer, J.; Lu, S.; Moënne-Loccoz, P.; Helton, M. E.;
Kaderli, S.; Zuberbhler, A. D.; Karlin, K. D. J. Am. Chem. Soc. 2004,
126, 12716–12717.
(22) (a) Praneeth, V. K. K.; Neese, F.; Lehnert, N. Inorg. Chem. 2005, 44,
2570. (b) Praneeth, V. K. K.; Na¨ther, C.; Peters, G.; Lehnert, N. Inorg.
Chem. 2006, 45, 2795. (c) Praneeth, V. K. K.; Haupt, E.; Lehnert, N.
J. Inorg. Biochem. 2005, 99, 940 Erratum: Praneeth, V. K. K.; Haupt,
E.; Lehnert, N. J. Inorg. Biochem. 2005, 99, 1744. (d) Lehnert, N.;
Praneeth, V. K. K.; Paulat, F. J. Comput. Chem. 2006, 27, 1338.
(23) Lim, M. D.; Lorkovic, I. M.; Ford, P. C. J. Inorg. Biochem. 2005, 99,
151.
(18) Liu, J.-G.; Naruta, Y.; Tani, F. Angew. Chem. 2005, 117, 1870–1874.
(19) (a) Collman, J. P.; Ghosh, S.; Dey, A.; Decreau, R. A.; Yang, Y. J. Am.
Chem. Soc. 2009, 131, 5034–5035. (b) Collman, J. P.; Fu, L.;
Herrmann, P. C.; Zhang, X. Science 1997, 275, 949–951. (c) Collman,
J. P.; Devaraj, N. K.; Decreau, R. A.; Yang, Y.; Yan, Y.-L.; Ebina,
W.; Eberspacher, T. A.; Chidsey, C. E. D. Science 2007, 315, 1565–
1568. (d) Collman, J. P.; Dey, A.; Decreau, R. A.; Yang, Y.; Hosseini,
A.; Solomon, E. I.; Eberspacher, T. A. Proc. Natl. Acad. Sci. U.S.A.
2008, 105, 9892–9896.
(24) (a) Bohle, D. S.; Hung, C.-H. J. Am. Chem. Soc. 1995, 117, 9584. (b)
Choi, I.-K.; Ryan, M. D. Inorg. Chim. Acta 1988, 153, 25. (c) Liu,
Y.; DeSilva, C.; Ryan, M. D. Inorg. Chim. Acta 1997, 258, 247.
(25) Cullotti, M.; Santagostini, L.; Monzani, E.; Casella, L. Inorg. Chem.
2007, 46, 8971.
(20) Zumft, W. G. J. Inorg. Biochem. 2005, 99, 194–215.
(21) Butland, G.; Spiro, S.; Watmough, N. J.; Richardson, D. J. J. Bacteriol.
2001, 183, 189–199.
9
J. AM. CHEM. SOC. VOL. 131, NO. 47, 2009 17117

A R T I C L E S
Berto et al.
Table 1. Crystallographic Data for Compound [Zn(To-F2PP-BzIM)]
that this issue has not been given enough consideration, in many
cases, where it has been simply assumed that the tethering of
an N-donor ligand will always lead to coordination to the central
metal ion. In order to advance the general design of 6C
Fe(II)-heme NO model complexes without the undesirable
necessity to add a large excess of the axial ligand (which is
unfavorable for reactivity studies),22b a detailed evaluation of
various tailed porphyrin designs is required. This study repre-
sents the first step toward a much needed systematic investiga-
tion into the binding properties of tethered N-donor ligands in
6C ferrous heme nitrosyls in solution at room temperature.
empirical formula
C56H33Cl2F6N7OZn
Formula weight (g/mol)
T (K)
1070.16
85
j
Space group
a (Å)
b (Å)
Triclinic, P1
10.522
13.206
19.752
103.955
101.557
100.116
2536.1
2
c (Å)
R (deg)
ꢀ (deg)
γ (deg)
V (Å3)
Z
µ (mm-1
λ (Å)
)
0.659
Experimental Section
0.71073
54977
12597
0.0460
1.075
In general, reactions were performed applying inert Schlenk
techniques. Preparation and handling of air sensitive materials was
carried out under an argon atmosphere in an MBraun glovebox
equipped with a circulating purifier (O2, H2O < 0.1 ppm). Infrared
spectra were obtained from KBr disks or in chloroform solution
on a Perkin-Elmer BX spectrometer. Proton magnetic resonance
spectra were recorded on a Varian Inova 400 MHz and a Varian
Mercury 300 MHz instrument. Electronic absorption spectra were
measured using an Analytical Jena Specord 600 instrument.
MALDI-TOF mass spectra were obtained on a Micromass TofSpec-
2E mass spectrometer whereas LCT-ESI mass spectra were obtained
on Micromass LCT Time-of-Flight mass spectrometer. Electron
paramagnetic resonance spectra were recorded on a Bruker X-band
EMX spectrometer equipped with an Oxford Instruments liquid
nitrogen or liquid helium cryostat. EPR spectra were typically
obtained on frozen solutions using ∼20 mW microwave power and
100 kHz field modulation with the amplitude set to 1 G. Sample
concentrations employed were ∼1 mM.
Collected reflns
Unique reflns
Rint
GOF
R1 [I > 2σ(I)]
wR2 (all data)
0.0570
0.1770
H2TMPm-Py (Ligand L1). The aminoporphyrin H2(NH2)TMPP
(0.25 g, 0.33 mmol) was dissolved in 6 mL of DMF. To this, 1
mL of N,N-diethylaniline was added, followed by the addition of
a solution of 3-(3′-pyridyl)propionic acid chloride in 3 mL of DMF
[prepared in situ by reacting 0.2 g of 3-(3′-pyridyl)propionic acid
with 3 mL of SOCl2], and then stirred for 4 h at room temperature.
The excess SOCl2 and the solvent were removed under vacuum.
The residue obtained was dissolved in CH2Cl2 and washed several
times with deionized water. The solution was then dried with
Na2SO4 and the solvent was removed using a rotary evaporator.
The desired product was purified using column chromatography
(silica, CH2Cl2/CH3OH ) 99:1). Yield: 0.20 g, 68%.
Crystal structure determination was carried out using a Bruker
SMART APEX CCD-based X-ray diffractometer equipped with a
low temperature device and a fine focus Mo-target X-ray tube
(wavelength ) 0.71073 Å) operated at 1500 W power (50 kV, 30
mA). Measurements were taken at 85 K and the detector was placed
5.055 cm from the crystal. The data was processed with SADABS
and corrected for absorption.27b The structure was solved and
refined with the Bruker SHELXTL (ver. 2008/3) software package
(cf. Table 1).27a,c
Materials. All solvents and reagents were purchased and used
as supplied except as follows. Toluene was distilled from sodium
under argon. Dried and air free THF and n-hexane were obtained
after passing through an MBraun solvent purification system.
1-Methylimidazole was vacuum distilled from KOH and degassed
via five freeze-pump-thaw cycles. Nitric oxide (Airgas, Royal
Oak, MI) was purified by first passing through an ascarite II column
(NaOH on silica gel) and then through a cold trap at -80 °C to
exclude higher nitrogen oxide impurities. The free base porphyrin
ligands H2TMP-mPy and H2To-F2PP-C3IM (ligands L1 and L2,
respectively) were synthesized following reported procedures17,28,29
with some modifications as described in the following. The
precursor porphyrins 5,10,15-Tris-(2,4,6-trimethyl-phenyl)-20-(2-
amino-phenyl)-porphyrin [H2(NH2)TMPP] and 5,10,15-Tris-(2,6-
difluoro-phenyl)-20-(2-amino-phenyl)-porphyrin [H2F6(NH2)TPP]
were prepared using reported procedures.29
UV-vis [nm] in CH2Cl2: 418, 514, 545, 590 and 646.
1H NMR (400 MHz, CDCl3): 8.69-8.64 (m, 9H, ꢀ-pyrrole (8H)
and aminophenyl(1H)); 8.13 (m, 1H, pyridyl); 8.0-7.97 (m, 2H,
pyridyl); 7.8 (m, 1H, aminophenyl); 7.48 (m, 1H, pyridyl); 7.26
(s, 6H, mesitylphenyl); 6.9 (m, 1H, aminophenyl); 6.85 (s, 1H, NH-
CdO); 6.78 (m, 1H, aminophenyl); 2.61 (s, 6H, para-CH3); 2.47
(t, 2H, -CH2-Pyridly); 1.84-1.80 (m, 18H, ortho-CH3); 1.61 (t,
2H, -CH2-CH2-Pyridyl); -2.55 (s, 2H, NH pyrrole). MS (MAL-
DI) for C61H56N6O: Calculated: 889, found: 889.
Further Control Experiment. The column purified ligand L1
was reacted with Zn(OAc)2 · 2H2O to yield [Zn(TMP-mPy)] in
CH2Cl2/CH3OH solution. After shaking with H2O to remove the
unreacted Zn(OAc)2, the solvent was removed under reduced
pressure and UV-vis spectra were recorded. The UV-vis spectrum
of the product in CH2Cl2 shows the Soret band at 430 nm, which
corresponds to a five-coordinate Zinc(II)-porphyrin complex.30 This
further proves the presences of the pyridyl linker in ligand L1.
[Fe(TMP-mPy)(Cl)]. 0.15 g (0.17 mmol) of H2TMP-mPy (L1)
was stirred in 10 mL of dry and air-free THF. Anhydrous FeCl2
(0.21 g, 1.7 mmol) was then added and the resulting reaction
mixture was refluxed at 60 °C under an argon atmosphere for 1 h.
The reaction was then stopped and the solvent was removed under
reduced pressure. The obtained residue was chromatographed on
silica gel using 0.5% methanol in CH2Cl2. The metalated porphyrin
fraction was collected and the solvent was then removed using a
rotary evaporator. The obtained product was dried; yield: 0.11 g,
66%. UV-vis [nm] in CH2Cl2: 376, 418, 509, 576, 655, and 692.
1H NMR (CDCl3, 400 MHz): 80 (m, 8H, ꢀ-pyrrole).
(26) Collman, J. P.; Yang, Y.; Dey, A.; Decreau, R. A.; Ghosh, S.; Ohta,
T.; Solomon, E. I. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 15660–
15665.
(27) (a) Sheldrick, G. M. SHELXTL, V. 2008/3; Bruker Analytical X-ray:
Madison, WI, 2008. (b) Sheldrick, G. M. SADABS, V. 2008/1, Program
for Empirical Absorption Correction of Area Detector Data; University
of Gottingen: Gottingen, Germany, 2008. (c) Saint Plus, V. 7.53a;
Bruker Analytical X-ray: Madison, WI, 2008.
[Fe(TMP-mPy)(NO)] (1). 0.05 g (0.05 mmol) of [Fe(TMP-
mPy)(Cl)] were placed in a 100 mL Schlenk flask and freshly
distilled CHCl3 (8 mL) and CH3OH (0.5 mL) were added. NO gas
was then passed through this solution, and the resulting solution
(28) (a) Collman, J. P.; Zhong, M.; Wang, Z.; Rapta, M. Org. Lett. 1999,
1, 2121–2124.
(29) Collman, J. P.; Brauman, J. I.; Doxee, K. M.; Halbert, T. R.;
Bunnenberg, E.; Linder, R. E.; LaMar, G. N.; Gaudio, J. D.; Lang,
G.; Spartalin, K. J. Am. Chem. Soc. 1980, 102, 4182–4192.
(30) Lin, C.-L.; Fang, M.-Y.; Cheng, S.-H. J. Electroanal. Chem. 2002,
531, 155–162.
9
17118 J. AM. CHEM. SOC. VOL. 131, NO. 47, 2009

Formation of a Stable Six-Coordinate Species
A R T I C L E S
was stirred for one hour. n-Hexane (10 mL) was slowly added to
the reaction mixture, which was then stored in a freezer (-22 °C)
for 3 days. The resulting precipitate was filtered off using a fine
pore Schlenk filter funnel. The yield of the product obtain was very
low, 11 mg (22%), and contained some impurities. FT-IR [cm-1]:
ν(NO) ) 1694.
Yield: 0.010 g, 52%: FT-IR in KBr [cm-1]: ν(NO) ) 1686.
To test whether complex 2 is really a monomer, we performed
dilution experiments where the concentration was lowered to an
intensity of 0.2 absorbance units. No shift of the Soret band was
observed, indicating that the complex does not correspond to a
dimer (oligomer) where two complexes (or more) share linkers.
N-(3-Methoxycarbonyl)benzylimidazole. Imidazole (3.8 g, 55.8
mmol) and methyl 3-bromomethyl-benzoate (3.1 g, 13.5 mmol)
were combined in 4 mL of DMF. The resulting faint yellow solution
was stirred at room temperature for 20 h. Upon completion, the
reaction was diluted with 60 mL of water and extracted 3× with a
total of 120 mL CH2Cl2. The organic layer was collected and
extracted 3× with a total of 150 mL of a 10% HCl solution. The
acidic solution was then alkalized with triethylamine to pH ) 10,
extracted 3× with a total of 210 mL ethyl acetate followed by
drying over Na2SO4. Rotary evaporation of the solvent yielded a
faint-yellow oil in quantitative amounts. 1H NMR (300 MHz,
CDCl3): 7.99 (d, 1H), 7.88 (s, 1H), 7.56 (s, 1H), 7.41 (t, 1H), 7.32
(d, 1H), 7.01 (s, 1H), 6.89 (s, 1H), 5.16 (s, 2H), 3.91 (s, 3H).
r-Imidazolyl-m-toluic Acid Hydrochloride. N-(3-Methoxycar-
bonyl)benzylimidazole (3.0 g, 13.9 mmol) was dissolved in 27 mL
of conc. HCl and brought to reflux while stirring for 2 h. The HCl
was then removed on a vacuum line and the resulting off-white
solid was dried under vacuum overnight. The next day, the product
was washed with hot acetonitrile and filtered to remove a yellow-
colored impurity. The resulting white solid was again dried under
H2To-F2PP-C3IM (Ligand L2). 4-(N-Imidazolyl)butyric acid
hydrobromide29 (0.18 g, 0.4 mmol) was ground to a fine powder
and stirred in 5 mL of dry and air-free CH2Cl2. The solution was
brought to reflux under argon and 35 mg (0.3 mmol) SOCl2 was
then added. The solution was stirred at reflux under an argon
atmosphere. After 30 min, the excess SOCl2 and CH2Cl2 were
removed under vacuum and a solution of H2F6(NH2)TPP (0.1 g,
0.135 mmol) in 5 mL of CH2Cl2 was added and the resulting green
solution was stirred for 1 h under argon at room temperature. The
reaction was then stopped by pouring the solution into 20 mL of
CH2Cl2 and successive washing with 2 × 20 mL of Na2CO3, and
3 × 20 mL of water, and then drying over Na2SO4. The solvent
was removed under reduced pressure and the residue obtained was
column purified on silica gel (CH2Cl2/CH3OH ) 95:5) to give the
desired C3 imidazole linked porphyrin ligand H2To-F2PP-C3IM.
Yield: 0.045 g, 38%.
1H NMR (300 MHz, d-acetone): 9.15-8.80 (m, 8H, ꢀ-pyrrole);
8.55 (d, 1H, aminophenyl); 8.44 (s, 1H, amide); 8.11 (dd, 1H,
aminophenyl); 8.05-7.95 (m, 3H, para fluorophenyl); 7.86 (t, 1H,
aminophenyl); 7.65-7.45 (m, 7H, meta fluorophenyl (6H) and
aminophenyl (1H)); 7.02 (s, 1H, imidazole); 6.44 (s, 1H, imidazole);
6.42 (s, 1H, imidazole); 3.45 (t, 2H, C3H6 tether); 1.58 (t, 2H, C3H6
tether); 0.90 (m, 2H, C3H6 tether); -2.69 (s, 2H, pyrrole NH). See
Figure S15 (Supporting Information).
1
vacuum. Yield: 2.0 g, 61%. H NMR (300 MHz, CD3OD): 9.13
(s, 1H), 8.06 (s, 1H), 8.05 (s, 1H), 7.66 (d, 1H), 7.60 (s, 1H), 7.57
(m, 2H), 5.56 (s, 2H).
H2To-F2PP-BzIM (Ligand L4). (Adapted from ref 31). R-Imi-
dazolyl-m-toluic acid hydrochloride (0.1 g, 0.42 mmol) and an
excess thionyl chloride (0.1 mL, freshly distilled under Ar) in
methylene chloride (2 mL) were refluxed under an argon atmo-
sphere. After 1 h the solution became clear. The excess SOCl2 and
CH2Cl2 were removed in vacuo to yield crude R-imidazolyl-m-
toluic acid chloride as an off-white solid. In a separate flask, a
solution of the porphyrin H2F6(NH2)TPP (0.088 g, 0.119 mmol) in
15 mL dry and air-free CH2Cl2 was stirred. To this stirred solution
a suspension of R-imidazolyl-m-toluic acid chloride (from above)
in CH2Cl2 (4 mL) was slowly added in small portions under an
argon atmosphere at room temperature (color changed to green).
The reaction was monitored by TLC (silica, CH2Cl2) until the only
visible porphyrin remained at the baseline. The solution was diluted
with 100 mL CH2Cl2 and shaken with 40 mL saturated NaHCO3,
washed 3× with 50 mL water, and dried with sodium sulfate. Rotary
evaporation of the solvent yielded a purple solid. The product was
purified on silica gel using column chromatography with 5% MeOH
in CH2Cl2 as eluent. Yield: 100 mg, 92%. UV-vis [nm] in CH2Cl2:
416, 509, 542, 586, and 640. 1H NMR (300 MHz, CD2Cl2):
9.0-9.85 (m, 9H, ꢀ-pyrrole (8H) and aminophenyl (1H)); 8.22 (dd,
1H, benzyl); 7.95-7.80 (m, 4H, para fluorophenyl (3H) and
aminopheyl (1H)); 7.65 (m, 1H, aminophenyl); 7.50-7.37 (m, 8H,
meta fluorophenyl (6H) and benzyl (2H)); 6.78 (s, 1H, NH-CdO);
6.47-6.38 (m, 2H, benzyl (1H) and aminophenyl (1H)); 6.33 (m,
1H, imidazolyl); 6.19 (s, 1H, imidazolyl); 6.07 (S, 1H, imidazolyl);
3.88 (s, 2H, -CH2-benzyl); -2.75 (s, 2H, NH pyrrole). See Figure
S16 (Supporting Information). MS (MALDI) for C54H34N7F6O;
Calculated: 922, found: 922.
UV-vis [nm] in CH2Cl2: 415, 509, 543, 587, and 642. MS (LCT-
ESI) for C51H33N7F6O: Calculated: 873.8, found: 874.
Further Control Experiment. The column purified ligand L2
was reacted with Zn(OAc)2 · 2H2O to yield [Zn(To-F2PP-C3IM)]
in a CH2Cl2/CH3OH solution. After shaking with H2O to remove
the unreacted Zn(OAc)2, the solvent was removed under reduced
pressure and UV-vis spectra were recorded. The UV-vis spectrum
of the product in CH2Cl2 shows the Soret band at 427 nm, which
is in accordance with similar values observed for five-coordinate
Zinc(II)-porphyrin complexes30 (substituted tetraphenylporphyrins
with electron withdrawing or donating groups attached to the phenyl
rings). This further proves the presence of the imidazolyl linker in
ligand L2.
MS (LCT-ESI) for ZnC51H31N7F6O: Calculated: 935.1, found:
1
936.1. H NMR spectra of the zinc metalated complex [Zn(To-
F2PP-C3IM)] reveal a shift of the imidazole proton peaks into the
porphyrin aromatic region. This is most likely due to an exposure
of the imidazole unit to the aromatic porphyrin ring current upon
binding to the zinc center.
[Fe(To-F2PP-C3IM)(Cl)]. 0.03 g (0.034 mmol) of H2To-F2PP-
C3IM were stirred in 8 mL of dry and air free THF. Anhydrous
FeCl2 (0.043 g, 0.34 mmol) was then added and the resulting
reaction mixture was refluxed under an argon atmosphere for 1 h.
The reaction was then stopped and the solvent was removed under
reduced pressure. The obtained residue was chromatographed on
silica gel using 3% methanol in CH2Cl2. The metalated porphyrin
fraction was collected, and the solvent was then removed using a
rotary evaporator. Yield: 0.023 g, 70%.
[Fe(To-F2PP-BzIM)(Cl)]. 0.03 g of H2To-F2PP-BzIM were
stirred in 10 mL of dry and air free THF. Anhydrous FeCl2 (0.039
g) was then added, and the resulting reaction mixture was refluxed
under an argon atmosphere for 1 h. The reaction was then stopped
and the solvent was removed under reduced pressure. The obtained
residue was chromatographed on silica gel using 5% methanol in
CH2Cl2. The metalated porphyrin fraction (slowest moving) was
collected, and the solvent was removed using a rotary evaporator.
The product obtained was finally dried. Yield: 27 mg, 82%. UV-vis
UV-vis [nm] in CH2Cl2: 350, 415, 510, 557, and 642. FT-IR in
KBr [cm-1]: νCdO ) 1700.
[Fe(To-F2PP-C3IM)(NO)] (2). To a solution of [Fe(To-F2PP-
C3IM)(Cl)] (0.02 g, 0.02 mmol) in 10 mL of freshly distilled CHCl3
and 0.5 mL of CH3OH was added excess nitric oxide and the
resulting solution was stirred for one hour. n-Hexane (15 mL) was
slowly added to the reaction mixture, which was then stored in a
freezer (-20 °C) for 1 day. The resulting precipitate was filtered
off and the obtained compound was stored inside a glovebox. The
IR spectrum shows the NO stretching band at 1686 cm-1, indicative
of the formation of 2.
(31) Young, R.; Chang, C. K. J. Am. Chem. Soc. 1985, 107, 898–909.
9
J. AM. CHEM. SOC. VOL. 131, NO. 47, 2009 17119

A R T I C L E S
Berto et al.
[nm] in CH2Cl2: 337, 413, 577, and 650. FT-IR in KBr [cm-1]:
νCdO ) 1684. MS (LCT-ESI) calculated: 975 (M-Cl), found: 975.
[Zn(To-F2PP-BzBr)]. The same procedure as for [Zn(To-F2PP-
BzIM)] was used, except that To-F2PP-BzBr was synthesized with
an acid chloride made from SOCl2 and methyl 3-bromomethyl-
benzoate to yield To-F2PP-BzBr where the imidazole is lacking to
prevent binding of the linker to zinc. UV-vis (nm) in CH2Cl2: 416,
510, 543, 585.
[Zn(To-F2PP-BzIM)]. 0.026 g of H2To-F2PP-BzIM were stirred
in 10 mL of dry and air free CH2Cl2. Zn(OAc)2 dihydrate (0.250
g) was then added along with 1 mL MeOH, and the resulting
reaction mixture was stirred at room temperature for 1.5 h. The
color of the solution changed from deep red to magenta. The
reaction was washed 3× with water, and dried with sodium sulfate.
Rotary evaporation of the solvent yielded a magenta colored solid.
Purification on a silica column eluted with CH2Cl2 yielded a pure
magenta solid (23.8 mg, 84%). UV-vis (nm) in CH2Cl2: 427, 559,
598.
it is most important to perform room temperature measurements.
This is because the binding of axial ligands is entropically
favored at low temperature, due to the temperature dependent
component of the Gibbs-Helmholtz equation. Methods of
assessing the strength of the N-donor coordination to the heme
center which require low temperatures are thus not a definitive
means of determining the room temperature behavior of these
systems. They can, however, still provide a means by which to
compare different systems. In this study, we use room temper-
ature UV-vis and IR spectroscopy, coupled to low temperature
EPR spectroscopy, to access the binding properties of our
tethered N-donor ligands. In the case of UV-vis spectroscopy,
the Soret band maximum shows a direct correlation with the
coordination number of the iron center. Five-coordinate (5C)
heme nitrosyls show Soret band positions of about 405 nm. With
IM bound, the Soret maximum shifts to 426 nm and the complex
is 6C. A corresponding, direct correlation is also observed
between the N-O stretching mode ν(N-O) and the coordination
number: 5C complexes exhibit ν(N-O) around 1675-1700
cm-1, whereas in 6C species, ν(N-O) drops to ∼1630 cm-1 in
the presence of IM. Additionally, low temperature EPR spec-
troscopy can be an effective probe of N-donor coordination to
ferrous heme nitrosyls.32 Species which are 5C in solution show
g values of 2.10, 2.06, and 2.01 as well as a three line hyperfine
splitting on g(min). On the other hand, 6C ferrous heme nitrosyls
with axial N-donor coordination exhibit smaller g values around
2.08, 2.00, and 1.97 as well as a distinctive nine line hyperfine
splitting observed on g(mid). Since EPR spectroscopy requires
low (at least liquid nitrogen) temperatures, which, as discussed
above, entropically favors the coordination of the axial N-donor
ligand, it can be expected that N-donor binding at room
temperature will generally be somewhat weaker than determined
from EPR. Finally, Raman spectroscopy could potentially be
applied to ferrous heme nitrosyls to characterize the strength
of the Fe-NO bond via measurement of the ν(Fe-NO) stretch.
However, photodecomposition is a serious problem in this case
when these complexes are exposed to the laser radiation required
for Raman measurements.
2. Application of Pyridine versus Imidazole as a Proximal
Ligand. In this study, “tailed” porphyrins with covalently
attached imidazole (IM) or pyridine (Py) ligands have been
investigated with the specific aim of determining the require-
ments for the generation of truly 6C ferrous heme nitrosyl model
complexes in solution at room temperature. These tailed
porphyrins are preferred over the use of excess free IM, because
they offer (a) control over the molar stoichiometry, (b) defined
structures of the complexes, and (c) prevention of side reactions
due to the presence of free axial ligand.33 The first consideration
involved in designing a 6C ferrous heme nitrosyl complex is
the choice of the axial N-donor ligand. We started this
investigation with a model complex based on H2TMP (TMP )
tetramesitylporphyrin) with a covalently attached Py (ligand L1
in Figure 1). Reaction of the ferric precursor with NO gas in
the presence of a small amount of methanol generates the
corresponding ferrous heme nitrosyl complex. However, the
compound obtained, [Fe(TMP-mPy)(NO)] (1), forms only a 5C
complex in solution, that is, the Py does not bind to the
Further Control for IM Binding. The UV-vis spectrum of
[Zn(To-F2PP-BzIM)] in CH2Cl2 shows the Soret band at 427 nm,
which is in accordance with similar values observed for five-
coordinate Zinc(II)-porphyrin complexes.30 In comparison, the
analogous complex [Zn(To-F2PP-BzBr)] which lacks the IM ligand
shows the Soret band at 416 nm. See Results and Discussion 4.
Crystallization of [Zn(To-F2PP-BzIM)]. [Zn(To-F2PP-BzIM)]
(18 mg) was dissolved with two 5S in 2 mL of 10% chlorobenzene
in CH2Cl2 in a Schlenk tube which was cooled to 4 °C. This tube
was then connected via glass joints to a Schlenk flask with n-hexane
stirred at 35 °C. After four days in the dark, enough n-hexane had
diffused over to obtain crystals suitable for crystallographic analysis.
[Fe(To-F2PP-BzIM)(NO)] (4). To a solution of the iron(III)-
porphyrin, [Fe(To-F2PP-BzIM)(Cl)] (0.02 g, 0.02 mmol) in 10 mL
of freshly distilled CHCl3 and CH3OH (0.5 mL), excess nitric oxide
was added, and the resulting solution was stirred for one hour under
an NO atmosphere. n-Hexane (15 mL) was slowly added to the
reaction mixture, which was then stored in a freezer (-20 °C) for
1 day. The resulting precipitate was filtered off using a fine pore
Schlenk filter funnel, and the obtained compound was stored inside
a glovebox. The IR spectrum shows the NO stretching band at 1644
cm-1, indicative of the formation of 4. Yield: 14 mg, 70%.
H2To-F2PP-C4IM (L3), [Fe(To-F2PP-C4IM)(Cl)], and
[Fe(To-F2PP-C4IM)(NO)] (3). The preparation of H2To-F2PP-C4IM
was performed by the same method as described for H2To-F2PP-
BzIM, but using 5-bromopentanoic acid in place of R-bromotoluic
1
acid to build the alkyl chain. H NMR (300 MHz, d-acetone):
9.10-8.80 (m, 8H, ꢀ-pyrrole); 8.54 (d, 1H, aminophenyl); 8.41 (s,
1H, amide); 8.09 (dd, 1H, aminophenyl); 8.05-7.90 (m, 3H, para
fluorophenyl); 7.86 (t, 1H, aminophenyl); 7.65-7.45 (m, 7H, meta
fluorophenyl (6H) and aminophenyl (1H)); 7.04 (s, 1H, imidazole);
6.52 (s, 1H, imidazole); 6.44 (s, 1H, imidazole); 3.30 (t, 2H, C4H8
tether); 1.44 (t, 2H, C4H8 tether); 1.08 (m, 2H, C4H8 tether); 0.90
(m, 2H, C4H8 tether); -2.72 (s, 2H, pyrrole NH). See Figure S15
(Supporting Information). UV-vis [nm] in CH2Cl2: 414, 510, 541,
586, and 641. MS (LCT-ESI) for C52H35N7F6O: Calculated: 887.9,
found: 888.
Iron insertion into H2To-F2PP-C4IM and the synthesis of [Fe(To-
F2PP-C4IM)(NO)] (3) also followed the same procedures as
described for 4.
[Fe(To-F2PP-C4IM)(Cl)]. UV-vis [nm] in CH2Cl2: 342, 414,
503, 574, and 650.
FT-IR in KBr [cm-1]: νCdO ) 1719.
Results and Discussion
1. Spectroscopic Distinction between 5C and 6C Ferrous
Heme Nitrosyls. To investigate the formation of a six-coordinate
(6C) ferrous heme nitrosyl in solution according to the equation:
(32) Lehnert, N. EPR and Low-Temperature Magnetic Circular Dichroism
Spectroscopy of Ferrous Heme Nitrosyls. In The Smallest Biomol-
ecules: Diatomics and their Interactions with Heme Proteins; Ghosh,
A., Ed.; Elsevier: Amsterdam, 2008.
(33) Wasser, I. M.; Huang, H.; Moënne-Loccoz, P.; Karlin, K. D. J. Am.
Chem. Soc., 2005, 127, 3310–3320.
FeII(NO) + IM a FeII(NO)(IM) (IM ) imidazole)
(2)
9
17120 J. AM. CHEM. SOC. VOL. 131, NO. 47, 2009

Formation of a Stable Six-Coordinate Species
A R T I C L E S
Figure 1. Drawings of the iron(II)-porphyrin NO complexes with covalently
attached N-donor ligands employed in this study.
Figure 2. Electronic absorption spectrum of [Fe(To-F2PP-C3IM)(NO)] (2,
red) in comparison to 5C [Fe(To-F2PP)(NO)] (black), and the 6C complexes
[Fe(To-F2PP)(MI)(NO)] (blue, MI ) free 1-methylimidazole) and [Fe(To-
F2PP)(Py)(NO)] (purple, Py ) free pyridine). Spectra were recorded in
CH2Cl2 or toluene solution at room temperature.
Fe(II)-NO center. This is evident from the low-temperature
(LT) EPR data of 1 in Figure S1 (Supporting Information),
which show a characteristic 5C spectrum with g values of 2.10,
2.04, and 2.01 as well as a three line 14N hyperfine pattern on
the smallest g value, g(min).32 The room temperature FT-IR
spectrum of 1 exhibits ν(NO) at 1694 cm-1 (Figure S3,
Supporting Information), in agreement with values of other 5C
complexes such as [Fe(TPP)NO] and [Fe(TMP)NO] at 1697
cm-1 and 1676 cm-1, respectively.34 Hence, the stereotype that
simple attachment of an N-donor ligand to the porphyrin core
should promote its coordination to the iron center is misleading.
These results, however, agree with the very weak binding
affinity of Py to Fe(II)-NO previously observed for different
types of tetraphenylporphyrin ligands.22b,35 In these studies it
was found that binding constants of free pyridine are only in
the range of 3 M-1 to 7 M-1 depending on the nature of the
porphyrin ligand employed. During our axial ligation titration
experiments reported in ref 22b, we found that, in general, the
binding constants for 1-methylimidazole (MI) are one order of
magnitude larger (typically between 20-40 M-1) than those
seen for Py. Therefore, Py is not a suitable axial ligand for 6C
ferrous heme nitrosyls and IM-type ligands are preferred due
to their larger binding constants. However, even in the case of
IM-type N-donors, the equilibrium between the 5C Fe(II)-NO
complex and the 6C (IM)N-Fe(II)-NO adduct shown in eq 2
disfavors the formation of the 6C species. This is because the
presence of 1 equiv of an N-donor ligand with a typical binding
constant (Keq) < 50 M-1 will not lead to the formation of
significant amounts of the 6C species in solution due to the
equilibrium strongly favoring the reactant side.36 In this case,
the strong σ-trans effect of NO prevents binding of the axial
ligand.22b,37 The formation of 6C Fe(II)-NO complexes in the
absence of excess N-donor ligands thus presents a significant
challenge.
F2PP) and free imidazole. In this case, a dramatic increase of
the IM binding constant to 2055 M-1 is observed.22b On the
other hand, the application of the porphyrin H2To-F2PP hardly
affects the properties of the Fe(II)-NO subunit as indicated by
the ν(N-O) stretching mode, which is only 6 cm-1 lower in
[Fe(To-F2PP)(MI)(NO)] compared to [Fe(TPP)(MI)(NO)].22b
The binding constant of free MI to these fluorinated porphyrins
therefore shows a nearly 300-fold increase over the pyridine
derivative of the same system. In conclusion, the first require-
ments for the design of a truely 6C ferrous heme nitrosyl in
solution are to (a) use IM as the N-donor ligand and (b) utilize
ortho difluorophenyl substituted TPP or other slightly electron
poor porphyrins. This provides a good basis in order to
overcome the strong σ trans effect of NO and obtain a 6C
complex in solution at room temperature in the presence of only
1 equiv of the N-donor ligand. Based on these results, we
decided to use IM ligands tethered to the fluorinated porphyrin
To-F2PP for our further studies. In the next step, the linker arm
which covalently links the N-donor ligand to the porphyrin core
needs to be optimized.
3. Imidazole Ligation with Alkyl Chain Tethers. In light of
the above-mentioned results, we synthesized ligand L2 in Figure
1 employing a C3 alkyl chain-linked IM attached to a fluoro-
substituted tetraphenylporphyrin. The UV-vis spectrum of the
obtained ferrous heme nitrosyl, [Fe(To-F2PP-C3IM)(NO)] (2),
exhibits the Soret band maximum at 417 nm as shown in Figure
2 (red curve). Interestingly, this is intermediate between the
Soret position of the 5C complexes [Fe(TPP*)(NO)] (TPP* )
tetraphenylporphyrin type ligand) at ∼405 nm, and the IM
coordinated (6C) species [Fe(TPP*)(IM)(NO)] at ∼425 nm (cf.
Figure 2). To test whether the 417 nm Soret band of 2
corresponds to weak binding of the axial ligand, we recorded
the UV-vis spectrum of the 6C Py complex [Fe(To-
F2PP)(Py)(NO)] for comparison, because in this case, it is known
that the Fe(II)-Py interaction is very weak.22b,37 As shown in
Figure 2 (purple curve), the spectra of 2 and [Fe(To-
F2PP)(Py)(NO)] show good agreement. Importantly, the lack
of a shoulder between 400 and 407 nm and around 470 nm for
2 confirms that no (or very little) 5C species is present in
solution. In addition to UV-vis data, FT-IR spectroscopy can
also be used to probe the coordination number of ferrous heme
nitrosyls through the N-O stretching mode ν(N-O), as
Importantly, our previous studies have also shown that a
further improvement of the axial ligand binding constant can
be achieved by using the combination of a weakly electron-
withdrawing tetra(ortho-difluorophenyl)porphyrin ligand (H2To-
(34) Scheidt, W. R.; Ellison, M. K. Acc. Chem. Res. 1999, 32, 350–359.
(35) Choi, I.-K.; Ryan, M. D. Inorg. Chim. Acta 1988, 153, 25–30.
(36) (a) Hoshino, M.; Ozawa, K.; Seki, H.; Ford, P. C. J. Am. Chem. Soc.
1993, 115, 9568–9575. (b) Traylor, T. G.; Sharma, V. S. Biochemistry
1992, 31, 2847–2849.
(37) Wyllie, G. R. A.; Schulz, C. E.; Scheidt, W. R. Inorg. Chem. 2003,
42, 5722–5734.
9
J. AM. CHEM. SOC. VOL. 131, NO. 47, 2009 17121

A R T I C L E S
Berto et al.
indicative that 2 is a 6C complex in agreement with the UV-vis
result. However, the lack of clear nine-line hyperfine splittings
of this signal indicates very weak binding of IM to iron as is
also reflected by the large value of ν(N-O). This is further
supported by the small hyperfine coupling constant of the IM
nitrogen estimated at around 2.0 MHz for 2 compared to 16-19
MHz for N-donor ligands in known 6C complexes.38 EPR
spectra are ideally suited for assessing the Fe-IM bond strength
as the covalency of the Fe-IM bond directly correlates with
the amount of spin density transferred from the Fe(II)-NO unit
to the N-donor atom of IM.22b The spin density present on the
IM nitrogen atom then correlates with the contact shift, and
hence, the magnitude of the 14N(IM) hyperfine coupling
constant. In this sense, weak bonding results in minimal transfer
of spin density between Fe and IM, resulting in almost no
contact shift and a small hyperfine coupling constant, likely
dominated by the dipolar contribution. A stronger Fe-IM bond
will manifest itself in an increase in contact shift, leading to a
larger hyperfine coupling constant, and thus a cleaner resolution
of hyperfine lines on g(mid).
The EPR spectrum of 2 in Figure 3 also shows a fourth g
value which is typically observed for 6C ferrous heme nitrosyl
complexes in both proteins and model systems.32 This signal is
usually referred to as g? and is clearly identified in the case of
both 2 and 3. Interestingly, this additional signal seems to be
absent in 5C heme nitrosyls. The origin of g? has been a matter
of extended discussion in the literature.39-41 It has been
proposed that this signal arises from a second conformation of
the complex where the relative orientation of NO, with respect
to the IM plane, has shifted.32 This is consistent with the
observation that crystal structures of 6C model complexes
usually show disorder of the NO ligand, giving rise to two major
conformations.22b,37,42 To rule out binding of the solvent DMSO
in the case of 2 and 3, the EPR spectrum of the 5C complex
[Fe(TPP)(NO)] was recorded in a DMSO/toluene mixture (1:
1) (cf. Figure S8, Supporting Information). These data resemble
more closely the typical 5C spectrum of this compound in pure
toluene. Additionally, the UV-vis spectrum of [Fe(TPP)(NO)]
in DMSO shows the Soret band at 409 nm, indicative of a 5C
complex. Therefore, binding of DMSO to complex 2 can be
ruled out in the EPR experiments. Based on all available
spectroscopic results, it can be concluded that complex 2 is 6C
in solution at room temperature, but that the interaction of the
C3 imidazole arm with the Fe(II)-NO center is weak.
Figure 3. EPR spectrum of [Fe(To-F2PP-C3IM)(NO)] (2) in DMSO at
liquid nitrogen temperature. A simulation of these data is shown in Figure
S6 (Supporting Information). The additional signal g? is typically observed
for 6C ferrous heme nitrosyls in both proteins and model complexes. See
text for a detailed explanation and relevant references.
mentioned above. Interestingly, as elaborated below, the mag-
nitude of the downshift of ν(N-O) is a direct function of the
Fe-N(IM) bond strength, mediated by the trans interaction
between the N-donor ligand and NO. The FT-IR spectrum of
2, taken at room temperature in a KBr disk, shows a broad
intense band at 1686 cm-1 (Figure S4, Supporting Information),
which is assigned to the ν(N-O) stretch. This energy of ν(N-O)
is not far off from values of known 5C complexes such as
[Fe(TPP)NO] or [Fe(TMP)NO] (vide supra) and thus indicates
a very weak covalent interaction between IM and the heme
center.22b
More insight is available from EPR spectroscopy. The EPR
spectrum of 2, shown in Figure 3, clearly resemble the spectra
of other 6C complexes.32 Hyperfine splittings observed within
these spectra can reveal a great deal of information about the
coordination environment of the central iron in heme nitrosyls
and has traditionally been used to distinguish between 5C and
6C complexes.32 This information manifests itself not only in
the number of hyperfine lines observed, but also in the particular
g value which exhibits the well-resolved hyperfine splittings.
In the case of 5C ferrous heme nitrosyl complexes such as
[Fe(TPP)(NO)], three g values are observed at 2.10, 2.06, and
2.01. The smallest of these, g(min), shows clear three line
hyperfine splittings due to the interaction of the nuclear spin (I
) 1) of 14N(O) with the unpaired electron of the Fe(II)-NO
unit. The fact that the well-resolved hyperfine splittings are seen
on g(min) indicates that the principal axis of this g-value is
aligned closest to the Fe-N(O) axis.22a,32 For 6C complexes
with NO and IM in the axial positions, the g shifts are smaller
with g values of about 2.07, 2.00, and 1.97.32 In addition, the
three hyperfine lines are further split, due to the presence of
14N of IM, to give a nine line hyperfine pattern, which is now
observed on g(mid) because of a rotation of the g tensor.22a,32
The spectrum of 2 at 77 K, shown in Figure 3, exhibits strong
14N hyperfine lines of NO on g(mid), and the g values are
obtained at 2.09, 2.00, and 1.98. In the lq. helium spectrum
additional, small, unresolved hyperfine splittings due to the axial
IM ligand seem to be present on g(mid). The obtained g values
and the presence of hyperfine splittings on g(mid) are generally
There are two possible reasons for the weak binding of the
C3 imidazole linker in ligand L2: (a) the alkyl chain of the ‘C3’
linker is too short to allow for a good interaction of IM with
iron(II), or (b) the dynamic motion of the phenyl rings leads to
a constant alkyl chain motion in solution, which prevents
effective binding of the imidazole ligand to iron(II) (see Results
and Discussion 4). In order to determine the influence of the
alkyl chain length on the binding properties of the linked IM,
we prepared a corresponding complex where the length of the
(38) (a) Kon, H.; Kataoka, N. Biochemistry 1969, 8, 4757–4762. (b)
Wayland, B. B.; Olson, L. W. J. Am. Chem. Soc. 1974, 96, 6037–
6041.
(39) Morse, R. H.; Chan, S. I. J. Biol. Chem. 1980, 255, 7876–7882.
(40) Hu¨ttermann, J.; Burgand, C.; Kappl, R. J. Chem. Soc. Faraday Trans.
1994, 90, 3077–3087.
(41) Tyryshkin, A. M.; Dikanov, S. A.; Reijerse, E. J.; Burgard, C.;
Huttermann, J. J. Am. Chem. Soc. 1999, 121, 3396–3406.
(42) (a) Silvernail, N. J.; Pavlik, J. W.; Noll, B. C.; Schulz, C. E.; Scheidt,
W. R. Inorg. Chem., 2008, 47, 912–920. (b) Silvernail, N. J.;
Barabanschikov, A.; Sage, J. T.; Noll, B. C.; Scheidt, W. R. J. Am.
Chem. Soc. 2009, 131, 2131–2140.
9
17122 J. AM. CHEM. SOC. VOL. 131, NO. 47, 2009

Formation of a Stable Six-Coordinate Species
A R T I C L E S
alkyl chain is increased by one CH2 unit (ligand L3 in Figure
1). Importantly, the UV-vis and EPR spectra of this complex,
[Fe(To-F2PP-C4IM)(NO)] (3) (cf. Figures S9 and S10, Sup-
porting Information), again show weak binding of the IM linker
to the iron(II) center with a Soret maximum at 415 nm. In
particular, the absorption spectrum in Figure S9 (Supporting
Information) is indicative of the presence of a small amount of
5C complex in solution at room temperature, as evidenced by
a very pronounced shoulder at 405 nm. This counterintuitive
result shows that an increase of the alkyl chain length from
“C3” to “C4” actually has a negative effect, i.e. the increase in
length or flexibility of the linker slightly decreases the binding
constant of the tethered IM ligand to iron(II). This is most likely
due to a loss of entropy of internal rotation as discussed before
for tailed Zn-porphyrins.43 Here, entropy loss was observed for
each additional CH2 unit added to the alkyl tethers within these
systems. EPR spectra of complex 3 are in agreement with this
conclusion and show g values at 2.07, 1.99, and 1.97, which
are almost identical to those for 2, indicating that an increase
in tether length does not increase the Fe-IM covalency, and
hence, bond strength (see Figures S6 and S11 in the Supporting
Information). No 5C species is observed in the EPR spectra of
3, but this is likely due to a slight increase in the IM binding
constant at low temperature compared to the room temperature
UV-vis data (see Results and Discussion 1).
Figure 4. Molecular structure of [Zn(To-F2PP-BzIM)] showing IM bound
to Zn(II) where the Zn ion is displaced form the porphyrin plane by 0.5 Å.
Two CH2Cl2 solvent molecules are present per unit cell and have been
omitted, along with all hydrogen atoms, for clarity. Selected bond lengths
are reported in Table 2. A complete list of all distances and angles is
provided in the Supporting Information.
Table 2. Selected Crystallographic Features of
[Zn(To-F2PP-BzIM)]a
Zn1-N1 (porphyrin)
Zn1-N2 (porphyrin)
Zn1-N3 (porphyrin)
Zn1-N4 (porphyrin)
Zn1-N7 (axial IM)
Zn1 (displacement)
2.077
2.095
2.064
2.073
2.079
0.5
4. Application of the More Rigid Benzyl Linker. Our sys-
tematic studies on IM binding to ferrous heme nitrosyls
presented above therefore show that pyridyl donors and C3-IM
or C4-IM tethers (L1-L3) are poor designs for 6C ferrous
heme nitrosyls in solution. Because of the poor binding seen
for both 2 and 3, it can be inferred that the length of the tether
is not as important as its other structural properties.
a All values are given in Å. A complete list of distances and angles is
given in the Supporting Information.
On the basis of this result, we decided to explore whether a
more rigid (less floppy) linker with a larger mass could change
the dynamics of the linker motion and facilitate a 6C complex.
For this purpose, a benzyl group was incorporated into the linker,
leading to ligand H2To-F2PP-BzIM (L4) in Figure 1. From the
literature, work of Collman and co-workers has presented
evidence that a similar linker could indeed allow for the
formation of a 6C species in solution, although the room
temperature characterization of the corresponding complex was
incomplete.26 Ligand L4 was prepared and crystallographically
characterized as the corresponding Zn complex. The crystal
structure of this compound is shown in Figure 4 and important
structural parameters are listed in Tables 2 and S1-S5 (Sup-
porting Information). The obtained bond distances and angles
reveal very little strain within the tether. The only sizable
perturbation from expected values is seen in the CNC angle of
129.1° for the amide linkage. This value is several degrees larger
than expected for an idealized system but only slightly larger
than amide CNC angles in other corresponding zinc porphyrin
compounds. The zinc complex of R,R,R,R-Tetra-(2-[3,4-
dimethoxyphenyl]-acetamidophenyl)porphyrin, for example,
exhibits an amide CNC angle of 128°.44 The zinc within the
porphyrin center of the L4 complex is clearly displaced toward
the IM tether by about 0.5 Å from the porphyrin plane. The
Zn-N(imidazole) bond distance is 2.079 Å. The average
Zn-N(porphyrin) bond distance is 2.077 Å, which is slightly
longer than the average bond length of 2.050 Å observed in
four-coordinate [Zn(TPP)].45 On the other hand, this value is
in agreement with the 5C complexes [Zn(TPP)(3-APy)] (3-APy
) 3-aminopyridine) and [Zn(OPP)(3-APy)] (OPP2- ) octaphe-
nylporphyrin) where average Zn-N(porphyrin) bond lengths
of 2.080 Å and 2.075 Å have been determined, respectively.46
The structure is well ordered with the single exception of the
phenyl ring located opposite to the tether position. Two possible
orientations are seen in the crystal structure where the phenyl
ring rotates slightly off the perpendicular geometry with respect
to the porphyrin plane, in both possible directions. The porphyrin
ligand itself shows very little distortion and appears only to be
slightly domed toward the IM ligand. The UV-vis adsorption
data obtained for the complex [Zn(To-F2PP-BzIM)] show the
Soret band at 427 nm. This is in agreement with know 5C
Zn(II)-porphyrin complexes.30 The analogous complex [Zn(To-
F2PP-BzBr)], where IM has been replaced by a noncoordinating
bromine, shows typical UV-vis absorption features of four-
coordinate (4C) Zn(II)-porphyrin complexes where the Soret
band is observed at 416 nm (cf. Figure 5). Due to these
characteristic absorption features, zinc-metalated porphyrin
species can be a useful tool to determine the coordination
environment of tethered porphyrin systems, and in particular,
to probe for the presence and binding properties of an attached
linker.
Next, the ferrous heme nitrosyl model complex 4 (shown in
Scheme 1) was synthesized. Importantly, the iron(II)-NO
(45) Terazono, Y.; Patrick, B. O.; Dolphin, D. H. Inorg. Chem. 2002, 41,
6703–6710.
(43) Walker, F. A.; Benson, M. J. Am. Chem. Soc. 1980, 102, 5530–5538.
(44) Cormode, D. P.; Drew, M. G. B.; Jagessar, R.; Beer, P. D. Dalton
Trans. 2008, 47, 6732–6741.
(46) Kojima, T.; Nakanishi, T.; Honda, T.; Harada, R.; Shiro, M.; Fukuzumi,
S. Eur. J. Inorg. Chem., 2009, 727–734.
9
J. AM. CHEM. SOC. VOL. 131, NO. 47, 2009 17123

A R T I C L E S
Berto et al.
Figure 7. EPR spectrum of [Fe(To-F2PP-BzIM)(NO)] (4) in frozen DMSO
at 77 K (14N hyperfine for g(mid) [MHz]: A(NO) ) 62, A(IM) ) 19). A
simulation of these data is shown in Figure S12 (Supporting Information).
The additional signal g? is typically observed for 6C ferrous heme nitrosyls
in both proteins and model complexes. See text for a detailed explanation
and relevant references.
Figure 5. UV-vis spectra of the zinc complexes [Zn(To-F2PP-BzIM)]
(black) and [Zn(To-F2PP-BzBr)] (red). Typical Soret and Q-band features
in 5C zinc tetraphenylporphyrin complexes are seen at 427 and 559 nm,
respectively, as found for the IM-bound complex. Upon replacement of
IM with a noncoordinating bromine, the Soret and Q features shift to
positions typically observed for 4C zinc complexes at 416 and 543 nm,
respectively. This confirms the ability of the IM in ligands L4 to coordinate
to the central metal ion.
is indicative of 5C species in solution, confirms the absence of
any 5C complex for the benzyl-linked compound. To further
investigate the strength of the Fe-IM interaction in 4, the EPR
spectrum of this compound was recorded as shown in Figure
7. Both the observed g values of 2.08, 2.01, and 1.98, as well
as the clean nine line hyperfine pattern on g(mid) with an IM-
nitrogen hyperfine coupling constant of 19 MHz resemble those
of well-known 6C ferrous heme nitrosyls.32 These results
confirm that 4 forms a stable 6C complex in solution where the
interaction of the benzyl IM arm with the Fe(II)-NO center is
strong. Neither the UV-vis data in Figure 6 nor the EPR
spectrum (see fit in Figure S12, Supporting Information) indicate
the presence of any 5C species in solution. To further quantify
the strength of the Fe-IM interaction, solution FT-IR spectra
were recorded at room temperature, since the strength of the
N-O bond, represented by the N-O stretching frequency, is
very sensitive to the strength of the Fe-IM interaction.22 In 5C
complexes of the type [Fe(porphyrin)(NO)], the N-O stretch
ν(N-O) is observed at 1675-1700 cm-1 (1697 cm-1 for
[Fe(TPP)(NO)]). Addition of a variety of free Py derivatives to
[Fe(TPP)(NO)] results in weak coordination of these Py ligands
and thus a moderate shift in the ν(N-O) frequency. In the case
of [Fe(TPP)(4-NMe2Py)(NO)] (4-NMe2Py ) 4-(dimethylami-
no)pyridine), the ν(N-O) frequency shifts down to 1653 cm-1
(recorded in the solid state).37 On the other hand, upon
coordination of free MI, a large shift in the ν(N-O) stretching
frequency to 1630 cm-1 (for [Fe(TPP)(MI)(NO)]) is observed.
This indicates that binding of an N-donor ligand and donation
Scheme 1. Drawing of the Structure of the 6C Complex
[Fe(To-F2PP-BzIM)(NO)] (4)
complex exhibits the Soret band at 427 nm as shown in Figure
6 (red curve), which is indicative of the formation of a stable
6C adduct in solution. The spectral features compare well with
those of the 6C complex [Fe(To-F2PP)(MI)(NO)] (MI ) free
1-methylimidazole), obtained in the presence of an excess of
MI. In particular, the lack of a shoulder around 470 nm, which
2
into the dz orbital of iron weakens the Fe-NO σ-bond, and,
due to the resulting reduced donation from the singly-occupied
π* orbital of NO, also weakens the N-O bond. Such an
interpretation is in agreement with the experimentally observed
direct correlation of the Fe-NO and N-O bond strengths and
vibrational frequencies where binding of the axial N-donor
ligand in fact weakens both of these bonds.22a,b,37,47,48 In this
Figure 6. Electronic absorption spectra of [Fe(To-F2PP-BzIM)(NO)] (4,
red), 5C [Fe(To-F2PP)(NO)] (black), and 6C [Fe(To-F2PP)(MI)(NO)] (MI
) free 1-methylimidazole, blue).
(47) Paulat, F; Berto, T. C.; George, S. D.; Goodrich, L. E.; Praneeth,
V. K. K.; Sulok, C. D.; Lehnert, N. Inorg. Chem. 2008, 47, 11449–
11451.
9
17124 J. AM. CHEM. SOC. VOL. 131, NO. 47, 2009

Formation of a Stable Six-Coordinate Species
A R T I C L E S
the N-O stretching frequency also seems to be the most
sensitive probe for the strength of the Fe-IM interaction.
One problem that remains is whether 4 shows intramolecular
binding of the tether as desired, or intermolecular coordination
of the tether to a different complex, potentially forming dimers
or large aggregates in solution. Such intermolecular coordination
has been previously suggested by Momenteau for heme protein
analogues in which tethers were attached to the ꢀ-pyrrole
positions of the porphyrin.10 In this case, the linkers cor-
responded to floppy alkyl chains. Additionally, tethers at the
ꢀ-pyrrole positions should be more susceptible to intermolecular
coordination due to their orientation relative to the porphyrin
plane. A tether employed at the ortho-phenyl position of a
tetraphenylporphyrin derivative is oriented toward the heme
center and thus is less likely to show intermolecular coordination
due to the close proximity of the tethered N-donor ligand to
the Fe center of the same complex. Increasing the rigidity of
the tether will further hinder this undesirable coordination. The
crystal structure of the zinc complex of L4 presented above
provides further evidence that intermolecular coordination is
most likely not occurring in complex 4. To further address this
issue, UV-vis dilution experiments were performed. The
concentration of [Fe(To-F2PP-BzIM)(NO)] was systematically
decreased until the spectra merged with the baseline, and the
Soret position was determined for each concentration. As
described above, the Soret band position is diagnostic for the
coordination mode of the heme nitrosyl. In these experiments
the Soret position remained constant (see Figure S14 in
Supporting Information). This result strongly indicates that the
tethered IM interacts with the iron center of the same complex.
If intermolecular binding was occurring, lower concentrations
should favor dissociation of the intermolecular complex and
thus, a Soret shift would be observed upon concentration
decrease.10
Figure 8. Solution IR spectrum of [Fe(To-F2PP-BzIM)(NO)] (4) showing
the ν(N-O) stretching frequency at 1644 cm-1 (see also Figure S13,
Supporting Information). This value is slightly higher than that observed
for [Fe(To-F2PP)(MI)(NO)] with free MI (1624 cm-1).
Table 3. Properties of 5C and 6C Ferrous Heme Nitrosyl Model
Complexesa
Soret
[nm]
ν(N-O)
b
c
complex
EPR g values
[cm-1
]
[Fe(TPP)(NO)]d
405
425
403
425
406
417
415
427
415
1697
1630
not det.
1624
2.102/2.064/2.010(*)
2.079/2.004(*)/1.972
not det.
[Fe(TPP)(MI)(NO)]d
[Fe(To-F2PP)(NO)]d
[Fe(To-F2PP)(MI)(NO)]d
not det.
e
[Fe(TMP-mPy)(NO)]
1694
2.099/2.040/2.012(*)
2.087/2.002(*)/1.982
2.073/1.991(*)/1.971
2.077/2.009(*)/1.978
2.072/2.002(*)/1.976
e
[Fe(To-F2PP-C3IM)(NO)]
[Fe(To-F2PP-C4IM)(NO)]e
[Fe(To-F2PP-BzIM)(NO)]
∼1686
not det.
1644(s)
1635(?)
e
f
[Fe(TpivPP-IM)(NO)]
Our results presented above are also in agreement with recent
findings of Collman and co-workers where a similarly rigid
benzyl linker was used for the preparation of CcO and NorBC
models, although the coordination number of the heme nitrosyls
was not explicitly addressed.19d,26,49 Nevertheless, room tem-
perature UV-vis data shown in these reports provide evidence
that the similar benzyl linker, used in conjunction with a picket
fence porphyrin, also generates a 6C ferrous heme nitrosyl
complex in solution. Not suprisingly, low temperature data are
also in agreement with this finding. While further room
temperature characterization, in particular IR spectroscopy, of
these compounds is lacking, the benzyl-linked complex does
appear to fulfill all the necessary requirements to generate 6C
ferrous heme nitrosyls in solution as defined in this work.
Besides the presence of the benzyl-IM tether, the amide-
substituted TPP derivative employed in this work provides the
necessary, weakly electron-poor porphyrin ligand needed to
enhance IM binding to the Fe-NO unit. Only one other example
of a potentially 6C iron-porphyrin NO complex with a tethered
axial IM ligand is known.50 In this case, the alkyl chain linker
is attached to the ꢀ-pyrrole carbon of a picket fence porphyrin.
However, whereas the EPR spectrum of this complex clearly
shows binding of the IM-linker to iron(II) at low temperature,
the room temperature UV-vis data exhibit the Soret band at
415 nm, which is more in agreement with the weak binding
observed for 2 and 3. A shoulder at 479 nm also indicates the
a References are included. b Measured at room temperature in a KBr
disk. Solution data are indicated by (s). c Measured at lq. nitrogen
temperature. The asterisk indicates the g value that shows well resolved
hyperfine lines in the spectrum. d References 22a and b. e This work.
f Reference 50. The conditions for the IR measurements are not
provided.
way, binding of an axial N-donor ligand lowers the N-O
stretching frequency, giving rise to the observed inverse
correlation between ν(N-O) and the Fe-N(N-donor) bond
strength. In the case of complexes 1-4 presented here, the
energy of ν(N-O) can also be seen to vary significantly based
on the tether employed and the specific N-donor ligand used.
The IR results in Table 3 show the range of ν(N-O) energies
observed based on ligand choice. Importantly, the ν(N-O)
stretching frequency of 4 is observed at 1644 cm-1 in solution
at room temperature as shown in Figure 8, which further
confirms that 4 corresponds to a 6C species even in solution.
Interestingly, this frequency is somewhat higher in energy
compared to [Fe(To-F2PP)(MI)(NO)] (1624 cm-1) and
[Fe(TPP)(MI)(NO)] (1630 cm-1) with bound free MI. There-
fore, the IR studies show that the covalently attached benzyl-
IM linker still cannot bind as strongly to the Fe(II)-NO unit
as free 1-methylimidazole (MI), but that 4 forms a stable 6C
complex in solution at room temperature without the require-
ment for the presence of excess axial ligand. In this regard,
(49) Collman, J. P.; Dey, A.; Yang, Y.; Decreau, R. A.; Ohta, T.; Solomon,
E. I. J. Am. Chem. Soc. 2008, 130, 16498–16499.
(48) Lehnert, N.; Galinato, M. G. I.; Paulat, F.; Richter-Addo, G.; Sturhahn,
W.; Xu, N.; Zhao, J. Submitted for publication.
(50) Komatsu, T.; Matsukawa, Y.; Tsuchida, E. Chem. Lett. 2000, 1060.
9
J. AM. CHEM. SOC. VOL. 131, NO. 47, 2009 17125

A R T I C L E S
Berto et al.
presence of a distinct amount of 5C species at RT in solution.
Finally, dilution experiments need to be performed in this case,
since ꢀ-pyrrole tailed hemes tend to form intermolecular
aggregates as described above.10
studies. Complex 4 is easy to prepare at relatively high total
yields (2.5%) for a sophisticated porphyrin.
Interestingly, the strength of the Fe-(N-donor) bond in the
benzyl-linked complex 4 is still slightly weaker than that
observed for [Fe(To-F2PP)(MI)(NO)] with free MI, as evidenced
by the higher ν(N-O) stretching frequency. Further studies
should therefore be directed at forming even stronger Fe-(N-
donor) bonds in 6C complexes as well as investigating other
tethers which could potentially facilitate the formation of truly
6C ferrous heme nitrosyls in solution. This knowledge will allow
for the improved synthesis of NorBC model complexes which
can more effectively mimic the structure and function of the
active site of this enzyme.
Conclusions
As shown in this paper, the design of a truly 6C ferrous heme
nitrosyl complex in solution at room temperature depends on
several factors. Application of a strongly binding ligand like
IM, combined with a bulky benzyl linker, is crucial for the
formation of these 6C complexes. In addition, a relatively
electron poor porphyrin ligand seems to facilitate IM binding,
and therefore, needs to be incorporated. This can be ac-
complished by the addition of either fluoro substituents or amide
groups to the meso phenyl rings of a tetraphenylporphyrin type
ligand. These 6C model systems require a very specific design
as compared to those employed for O2 and CO binding studies
because these diatomics do not show the strong σ-trans effect
that is observed for NO. In this study it is demonstrated that a
truly 6C ferrous heme nitrosyl can be generated in solution at
room temperature in the presence of only one equivalent of the
N-donor ligand, if all requirements specified above are fulfilled.
The need for excess IM for the generation of 6C species, which
leads to undesired side reactions including denitrosylation,22b,24a,51
is now obsolete. Compound 4 is therefore ideally suited for
reactivity studies on 6C Fe(II)-NO complexes which are
currently in progress. In particular, the facile synthesis of 4 will
be advantageous for application of this complex in NorBC model
Acknowledgment. This work was supported by the Deutsche
Forschungsgemeinschaft (DFG; grant LE 1393/1-2), the Dow
Corning Corporation, and a grant from the National Science
Foundation (CHE 0846235). We acknowledge Dr. Jeff W. Kampf
(University of Michigan) for his X-ray crystallographic analysis
of [Zn(To-F2PP-BzIM)].
Supporting Information Available: FT-IR spectra of the heme
nitrosyl complexes 1, 2, and 4 are presented in comparison to
the corresponding ferric precursor complexes, UV-vis absorp-
tion spectrum of 3, variable concentration UV-vis spectrum
of 4, EPR spectra along with simulations for complexes 1-4,
complete Tables of distances and angles for [Zn(To-F2PP-
BzIM)], and 1H-NMR spectra of L2, L3, L4, and MI for
comparison, complete ref 2d. This material is available free of
charge via the Internet at http://pubs.acs.org.
(51) Lançon, D.; Kadish, K. M. J. Am. Chem. Soc. 1983, 105, 5610–5617.
JA904368N
9
17126 J. AM. CHEM. SOC. VOL. 131, NO. 47, 2009