Binding of O2 and CO to metal porphyrin anions in the gas phase

Article abstract of DOI:10.1002/anie.201303200

The binding energies of O₂ and CO to iron(II) and manganese(II) porphyrin anions has been determined in the gas phase. Low-pressure ion-molecule equilibria have been measured in a cryogenically cooled trap of an FT-ICR mass spectrometer, and binding energies of (40.8±1.3) kJ mol^-1 and (66.3±2.6) kJ mol^-1 have been obtained for oxygen and carbon monoxide, respectively, with a heme-analogue Fe^II porphyrin complex. Copyright

Full text of DOI:10.1002/anie.201303200

Angewandte
Chemie
Communications
Model Heme Complexes
T. Karpuschkin, M. M. Kappes,
O. Hampe*
&&&& — &&&&
Binding of O2 and CO to Metal Porphyrin
Anions in the Gas Phase
The binding energies of O2 and CO to
iron(II) and manganese(II) porphyrin
anions has been determined in the gas
phase. Low-pressure ion–molecule equi-
libria have been measured in a cryogeni-
cally cooled trap of an FT-ICR mass
spectrometer, and binding energies of
(40.8ꢀ1.3) kJmolꢁ1 and (66.3ꢀ
2.6) kJmolꢁ1 have been obtained for
oxygen and carbon monoxide, respec-
tively, with a heme-analogue FeII porphy-
rin complex.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
DOI: 10.1002/anie.201303200
Model Heme Complexes
Binding of O2 and CO to Metal Porphyrin Anions in the Gas Phase**
Tatjana Karpuschkin, Manfred M. Kappes, and Oliver Hampe*
Metal porphyrins and their derivatives are the key cofactors
in many heme enzymes, which themselves are present in
virtually all living organisms. In heme proteins (e.g. cytochro-
me P450 or hemoglobin), an iron porphyrin complex serves as
the prosthetic group and acts as a potent catalyst for oxygen
transfer and transport.[1] These enzymes ultimately allow
living cells to use otherwise unreactive molecular oxygen
directly from the environment. In most such biochemical
processes the catalytic cycle starts by binding dioxygen to the
ferrous heme group, that is, after the iron porphyrin complex
has been reduced from its resting ferric(III) state to an active
iron(II) center.[2] It has long been recognized that oxygen
binding in heme enzymes occurs in a highly cooperative
manner[3] and is crucially modulated by a proximal axial
ligand at the FeII center and its propensity for electron
donation or withdrawal. Fourfold-coordinated iron porphyr-
ins (4c-FeP; no axial ligand) are known from spectroscopic
experiments to exhibit a triplet ground state (S = 1).[4] For 5c-
FeP, the spin multiplicity is found to be a quintet (S = 2), with
the exact nature of the axial ligand giving rise to subtle but
important differences on the reactivity towards dioxygen. As
suggested recently,[5] the weakly donating imidazole and
histidine ligands (as in hemoglobin or myoglobin, respec-
tively) lead to reversible Fe-O2 binding, whereas more
strongly donating ligands (such as anionic imidazolate) may
be responsible for stronger oxygen fixation (as in cytochro-
me P450).
ues) for the binding of various neutral reagents such as
pyridine and NO.[8] In another ICR study (over the elevated
temperature range 308–342 K), the association reaction of
several iron(II) porphyrin cations (with different peripheral
substituents) with NO was investigated by Ridge and co-
workers.[9] They derived NO binding energies to be in the
range of 103.8–120.9 kJmolꢁ1, that is, with only small varia-
tions depending on the peripheral ligand of the porphyrin.
Moreover, virtually identical numbers were obtained for
singly and doubly charged iron porphyrin cations.
Here we report the first experimental determinations of
the binding energies of molecular oxygen and carbon mon-
oxide to model metal porphyrin complex ions in vacuo. For
this, the association reactions of deprotonated iron(II)
tetrakis(4-sulfonatophenyl)porphyrin
tetraanions
[FeII-
(tpps)]4ꢁ (Figure 1) and the manganese(II) tetrakis(4-sulfo-
natophenyl)porphyrin species [MnII(tpps)]4ꢁ with O2 and CO
were probed over a temperature range of 120 to 210 K in
a variable-temperature Penning ion trap (Figure 2).[10]
Towards an improved understanding of the elementary
binding processes occurring in model heme complexes, gas-
phase experiments—in particular those involving mass-to-
charge ratio selected ions—offer the advantage that the
oxidation state of the metal center may be controlled,[6] while
environmental effects such as counterions or solvent mole-
cules can be rigorously excluded. Along these lines, Crestoni
and co-workers[7] have extensively studied the gas-phase ion
chemistry of metal porphyrin ions in ion cyclotron resonance
(ICR) cells at room temperature. They found no strong ligand
effect for the low-pressure association equilibria of FeII and
FeIII heme cations in the systems studied, but were able to
derive a thermochemical ladder (i.e. differences in DG val-
Figure 1. Oxygen complex of [FeII(tpps)]4ꢁ. tpps=tetrakis(4-sulfonato-
phenyl)porphyrin.
[*] Dr. T. Karpuschkin, Prof. Dr. M. M. Kappes, Priv.-Doz. Dr. O. Hampe
Institut fꢀr Nanotechnologie und Institut fꢀr Physikalische Chemie,
Karlsruher Institut fꢀr Technologie (KIT)
Karlsruhe (Germany)
E-mail: oliver.hampe@kit.edu
Figure 2. Schematic representation of the experimental setup.
Homepage: http://www.int.kit.edu/english/1111.php
[**] This work was supported by the DFG through the collaborative
research centre TR-SFB 88 “Cooperative effects in homo- and
heterometallic complexes (3MET)” and by the BMBF through the
Helmholtz association (POF “STN”).
Trapped [FeII(tpps)]4ꢁ and [MnII(tpps)]4ꢁ were allowed to
react at low temperatures with O2 or CO to give the molecular
association products. Figure 3 exemplifies this type of reac-
tion for the system [FeII(tpps)]4ꢁ/O2 at T= 123 K, without any
side reaction being observable in the corresponding mass
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201303200.
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
The values obtained for the rate coefficients are also
summarized in the Supporting Information.
This then allows the equilibrium constants Kp to be
determined from the ratio of the rate coefficients kad/kdes and
known reactant gas pressure. Figure 5 shows such results as
Figure 3. Negative-ion mass spectra of [FeII(tpps)]4ꢁ a) before and
b) after addition of oxygen for 18 s at a partial pressure of
4ꢁ10ꢁ7 mbar and at a temperature of 123 K.
spectra. Similarly, [FeII(tpps)]4ꢁ already reacts readily with
CO at temperatures around 180–210 K, as does [MnII(tpps)]4ꢁ
with O2.[11]
Kinetic data obtained by evaluating the ion concentration
at constant reactant pressure versus reaction time clearly
show that the reaction comes to an equilibrium for longer
times at sufficiently low temperatures (shown in Figure 4 for
the iron/O2 system) and the reaction can simply be described
by Equation (1).
Figure 5. Van’t Hoff plot of experimentally determined equilibrium
constants Kp for the gas-phase association reactions indicated. Reac-
tion enthalpies were derived from the slope of linear fits to the data
(solid lines).
a function of temperature for the three reaction systems
studied. From such a vanꢀt Hoff plot of the equilibrium
constants (ꢁlnKp versus inverse temperature) and a linear fit
to the data one can directly deduce the reaction enthalpies
and entropies for the reaction of O2 or CO to the metal
porphyrin (Table 1). Note that the negative of the reaction
kad
II
4ꢁ
½FeIIðtppsÞꢂ4ꢁ þ O
½Fe ðtppsÞ ꢃ O ꢂ
ƒ!
ð1Þ
ƒ
2
2
kdes
Table 1: Measured reaction enthalpies DHr and entropies DSr for the
association reaction to form the respective O2/CO complex.
[FeII(tpps)·O2]4ꢁ [FeII(tpps)·CO]4ꢁ [MnII(tpps)·O2]4ꢁ
DHr [kJmolꢁ1 [a]
DSr [JKꢁ1 molꢁ1
]
ꢁ40.8ꢀ1.4
ꢁ134ꢀ10
ꢁ66.3ꢀ2.6
ꢁ153ꢀ13
ꢁ67.4ꢀ2.2
ꢁ168ꢀ11
]
[a] Errors are estimated from possible systematic errors in temperature
(ꢀ5 K) and pressure (factor of 2) as well as from the statistical fit errors.
enthalpy can be equated to the binding energy of the
respective diatomic molecule. Our results show that molec-
ular oxygen is bound to [FeII(tpps)]4ꢁ by 40.8 kJmolꢁ1. The
corresponding carbon monoxide binding energy is consider-
ably larger (66.3 kJmolꢁ1). At 67.4 kJmolꢁ1, dioxygen is
found to bind significantly more strongly to the MnII center
than to the iron porphyrin.
Figure 4. Kinetics for the association of oxygen to trapped [FeII(tpps)]4ꢁ
(O2 partial pressure 4ꢁ10ꢁ7 mbar) as a function of temperature. Solid
lines are fits to the experimental data using the underlying association
and dissociation rate equations. Equilibrium constants Kp were
obtained by parameterizing the stationary/equilibrium signals
observed at long times.
It is interesting to compare our results with literature
values from quantum-chemical computations for model iron
and manganese porphyrins as well as to relevant studies on
heme proteins in the condensed phase (Table 2). It has been
pointed out in a number of theoretical studies that the relative
binding energies of O2, CO, and NO to ferrous heme[12]
increase in the order Eb(O2) < Eb(CO) < Eb(NO).[13] How-
ever, a more quantitative theoretical description still appears
challenging. In a recent study Radon and Pierloot[14] con-
The reversibility of the reaction could be clearly demon-
strated in control experiments: upon turning off the O2 or CO
supply after the reaction, the product complex loses the
diatomic molecule to return the reactant metal porphyrin ion
(see Figure S1/S2 in the Supporting Information). Also
included in Figure 4 (as solid lines) are fits of appropriate
rate equations using a simple two-step mechanism comprising
(radiative) association and (activated) dissociation steps.[10b]
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
Table 2: Binding energies of O2 and CO to several model complexes
from computations of FeII and MnII porphyrins and from solution-phase
studies on heme proteins.
reaction steps involved in the corresponding enzymatic
reactions.[25] In particular, it should help to disentangle the
intrinsic properties of the metal–organic center from environ-
mental influences often referred to in the literature as the
“protein” effect.
Model system
Eb [kJmolꢁ1
]
Reference
Fe porphyrin + O2
[Fe(tpps)]4ꢁ
FeP
40.8ꢀ1.3
this work
37.7
DFT, Rovira et al.(1997)[13]
DFT, Estrin et al. (2008)[16]
DFT, Zhang et al. (2007)[17]
DFT, Witko et al. (2006)[18]
DFT, Witko et al. (2007)[19]
DFT, Sun et al. (2009)[20]
DFT/Radon et al. (2008)[14]
CASPT2, Radon et al. (2008)[14]
Gaud et al. (1974)[21]
FeP
54.0
Experimental Section
Experiments were performed with a 7T-FT-ICR mass spectrometer[26]
(APEX II, Bruker Daltonics) equipped with an electrospray ion
source (ESI, Analytica of Branford) and a home-built ion funnel with
jet disrupter.[27] Solutions of MnIII meso-tetra(4-sulfonatophenyl)por-
phin chloride and FeIII meso-tetra(4-sulfonatophenyl)porphin chlo-
ride (cat. nos MnT1239 and FeT1239 from Frontier Scientific, Logan,
UT, USA), were prepared in pure water at a concentration of
ca. 10ꢁ5 molLꢁ1. Some experiments also made use of a home-built
nanospray source. Negative ions leaving the capillary were efficiently
collected in the ion funnel, prestored in a hexapole ion trap, and
pulsed into the ICR cell. Trapped ions were then excited/detected by
standard ICR techniques. To perform temperature-dependent kinetic
experiments of ion–molecule reactions, the Infinity ICR cell was
recently modified[10a] to allow the cell to be heated or cooled within
a temperature range of T= 90 to 420 K.[28] To obtain kinetic data we
followed a mass-spectrometric protocol similar to the one recently
described.[10a] Briefly, the reactant ions were isolated in the ICR cell
and thermalized with helium or argon at a pressure of 10ꢁ5 mbar for
5 s. After a pump delay, the reactant gases (O2 or CO) were dosed into
the chamber of the ion trap to give a constant partial pressure of
typically around 4 ꢁ 10ꢁ7 mbar by means of a pulsed solenoid valve.
The purity of the gas, as monitored online with a residual gas analyzer,
was better than 99.5%.
FeP/Fe(tpp)
FeP/[FePPIX]2ꢁ
[Fe(heme)(his)]2ꢁ
FeP
FeP
FeP
human HbA
myoglobin
Hb
56.0/75.3
55.2/60.2
89.9
32.8
ꢁ5.4–67.4
44.8
55.2ꢀ1.7
75.7ꢀ1.7
43.1ꢀ4.6
68.2ꢀ4.6
Keyes et al. (1971)[22]
Johnson et al. (1992)[23]
Collman et al. (1978)[24]
Fe{Piv3(5ClImP)Por}
Fe porphyrin + CO
[Fe(tpps)]4ꢁ
FeP
FeP
FeP
FeP
66.3ꢀ2.6
this work
108.8
DFT, Rovira et al. (1997)[13]
DFT, Estrin et al. (2008)[16]
DFT, Radon et al. (2008)[14]
CASPT2, Radon et al. (2008)[14]
DFT, Witko et al. (2007)[19]
Gaud et al. (1974)[21]
139.3
ꢁ4.2–110.9
66.9
[Fe(heme)(his)]2ꢁ
human HbA
myoglobin
156.9
74.1ꢀ1.7
89.5ꢀ1.3
Keyes et al. (1971)[22]
Mn porphyrin + O2
[Mn(tpps)]4ꢁ
MnP
67.4ꢀ2.2
this work
48.9
DFT, Witko et al. (2006)[18]
Received: April 16, 2013
Published online: && &&, &&&&
cluded that density functional theory tends to underestimate
the binding energy—except when using the BP86 functional.
For example, the binding energy of O2 to 4c-FeP was found to
be anywhere between ꢁ5.4 kJmolꢁ1 (B3LYP), that is, not
bound at all, and 67.4 kJmolꢁ1 for BP86. However, in the
same study it was found that a description using methods
based on wave functions, including perturbation theory
(CASPT2), reproduces the binding energetics as found in
solution-phase studies of wild-type myoglobin (a 5c-FeP
complex within its protein environment) reasonably well.[15]
Our gas-phase values for the binding energies of O2 and
CO to iron(II) porphyrin ions agree well with the CASPT2
description (Table 2).
Another theoretical study has considered the influence of
charge on the binding propensity of iron(II) porphyrin.[18] The
oxygen binding energy of the dianionic iron protoporphyr-
in IX (FePPIX), which carries two deprotonated propionic
acids at the porphyrin ring perimeter, was found to increase
by about 10% over the neutral iron porphyrin (FeP). In the
seminal computations of Rovira et al.[13] it was also concluded
that peripheral ligands only have a minor influence on the
chemical behavior of the iron center.
Keywords: biophysical chemistry · ferrous heme complexes ·
mass spectrometry · gas-phase ion chemistry · oxygen binding
.
[1] S. Shaik, S. Cohen, Y. Wang, H. Chen, D. Kumar, W. Thiel,
Chem. Rev. 2010, 110, 949 – 1017.
[2] I. G. Denisov, T. M. Makris, S. G. Sligar, I. Schlichting, Chem.
Rev. 2005, 105, 2253 – 2277.
[3] M. F. Perutz, Nature 1970, 228, 726 – 734.
[4] a) J. P. Collman, J. L. Hoard, N. Kim, G. Lang, C. A. Reed, J.
Am. Chem. Soc. 1975, 97, 2676 – 2681; b) H. Goff, G. N. Lamar,
C. A. Reed, J. Am. Chem. Soc. 1977, 99, 3641 – 3646; c) G. Lang,
K. Spartalian, C. A. Reed, J. P. Collman, J. Chem. Phys. 1978, 69,
5424 – 5427.
[5] C. Hu, C. D. Sulok, F. Paulat, N. Lehnert, A. I. Twigg, M. P.
Hendrich, C. E. Schulz, W. R. Scheidt, J. Am. Chem. Soc. 2010,
132, 3737 – 3750.
[6] D. K. Bçhme, H. Schwarz, Angew. Chem. 2005, 117, 2388 – 2406;
Angew. Chem. Int. Ed. 2005, 44, 2336 – 2354.
[7] a) B. Chiavarino, R. Cipollini, M. E. Crestoni, S. Fornarini, F.
Lanucara, A. Lapi, J. Am. Chem. Soc. 2008, 130, 3208 – 3217;
b) M. E. Crestoni, S. Fornarini, F. Lanucara, Chem. Eur. J. 2009,
15, 7863 – 7866; c) M. E. Crestoni, S. Fornarini, F. Lanucara, J. J.
Warren, J. M. Mayer, J. Am. Chem. Soc. 2010, 132, 4336 – 4343.
[8] a) B. Chiavarino, M. E. Crestoni, S. Fornarini, C. Rovira, Inorg.
Chem. 2008, 47, 7792 – 7801; b) F. Angelelli, B. Chiavarino, M.
Crestoni, S. Fornarini, J. Am. Soc. Mass Spectrom. 2005, 16, 589 –
598.
In summary, the experimental work described herein gives
benchmark thermochemical information for one of the most
important biochemical reactions. Such data on the binding
energy of dioxygen and carbon monoxide to iron and
manganese porphyrins may prove helpful towards a quantita-
tive quantum-chemical description of the spin-forbidden
[9] O. N. Chen, S. Groh, A. Liechty, D. P. Ridge, J. Am. Chem. Soc.
1999, 121, 11910 – 11911.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
[10] a) O. Hampe, T. Karpuschkin, M. Vonderach, P. Weis, Y. M. Yu,
L. B. Gan, W. Klopper, M. M. Kappes, Phys. Chem. Chem. Phys.
2011, 13, 9818 – 9823; b) M. Neumaier, F. Weigend, O. Hampe,
M. M. Kappes, J. Chem. Phys. 2005, 122, 104702.
[19] D. Rutkowska-Zbik, M. Witko, G. Stochel, J. Comput. Chem.
2007, 28, 825 – 831.
[20] Y. Sun, X. Hu, H. Li, A. F. Jalbout, J. Phys. Chem. C 2009, 113,
14316 – 14323.
[11] Note that under virtually identical experimental conditions the
porphyrin species with a central metal atom in its oxidation state
+ III and the same number (= 4) of deprotonation sites, that is,
a molecular trianion, does not react to any appreciable extent.
This strongly indicates that the chemical behavior of the metal
porphyrin systems with respect to O2 or CO binding is indeed
governed by the metal center and to a much lesser extent by the
peripheral charges.
[21] H. T. Gaud, B. G. Barisas, S. J. Gill, Biochem. Biophys. Res.
Commun. 1974, 59, 1389 – 1394.
[22] M. H. Keyes, M. Falley, R. Lumry, J. Am. Chem. Soc. 1971, 93,
2035.
[23] C. R. Johnson, D. W. Ownby, S. J. Gill, K. S. Peters, Biochemistry
1992, 31, 10074 – 10082.
[24] J. P. Collman, J. I. Brauman, K. M. Doxsee, T. R. Halbert, K. S.
Suslick, Proc. Natl. Acad. Sci. USA 1978, 75, 564 – 568.
[25] S. Shaik, M. Filatov, D. Schrçder, H. Schwarz, Chem. Eur. J. 1998,
4, 193 – 199.
[12] L. M. Blomberg, M. R. A. Blomberg, P. E. M. Siegbahn, J. Inorg.
Biochem. 2005, 99, 949 – 958.
[13] C. Rovira, K. Kunc, J. Hutter, P. Ballone, M. Parrinello, J. Phys.
Chem. A 1997, 101, 8914 – 8925.
[26] B. Concina, M. Neumaier, O. Hampe, M. M. Kappes, J. Chem.
Phys. 2008, 128, 134306.
´
[14] M. Radon, K. Pierloot, J. Phys. Chem. A 2008, 112, 11824 –
[27] K. Q. Tang, A. V. Tolmachev, E. Nikolaev, R. Zhang, M. E.
Belov, H. R. Udseth, R. D. Smith, Anal. Chem. 2002, 74, 5431 –
5437.
[28] a) X. H. Guo, M. Duursma, A. Al-Khalili, L. A. McDonnell,
R. M. A. Heeren, Int. J. Mass Spectrom. 2004, 231, 37 – 45;
b) R. L. Wong, K. Paech, E. R. Williams, Int. J. Mass Spectrom.
2004, 232, 59 – 66; c) O. P. Balaj, C. B. Berg, S. J. Reitmeier, V. E.
Bondybey, M. K. Beyer, Int. J. Mass Spectrom. 2009, 279, 5 – 9.
11832.
[15] J. S. Olson, G. N. Phillips, J. Biol. Inorg. Chem. 1997, 2, 544 – 552.
[16] D. E. Bikiel, S. E. Bari, F. Doctorovich, D. A. Estrin, J. Inorg.
Biochem. 2008, 102, 70 – 76.
[17] Z. Shi, J. J. Zhang, J. Phys. Chem. C 2007, 111, 7084 – 7090.
[18] D. Rutkowska-Zbik, M. Witko, J. Mol. Catal. A 2006, 258, 376 –
380.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!