Insight into the preferential N-bindingversusO-binding of nitrosoarenes to ferrous and ferric heme centers

Article abstract of DOI:10.1039/d0dt03604h

Nitrosoarenes (ArNOs) are toxic metabolic intermediates that bind to heme proteins to inhibit their functions. Although much of their biological functions involve coordination to the Fe centers of hemes, the factors that determine N-binding or O-binding of these ArNOs have not been determined. We utilize X-ray crystallography and density functional theory (DFT) analyses of new representative ferrous and ferric ArNO compounds to provide the first theoretical insight into preferential N-bindingversusO-binding of ArNOs to hemes. Our X-ray structural results favored N-binding of ArNO to ferrous heme centers, and O-binding to ferric hemes. Results of the DFT calculations rationalize this preferential binding on the basis of the energies of associated spin-states, and reveal that the dominant stabilization forces in the observed ferrous N-coordination and ferric O-coordination are dπ-pπ* and dσ-pπ*, respectively. Our results provide, for the first time, an explanation whyin situoxidation of the ferrous-ArNO compound to its ferric state results in the observed subsequent dissociation of the ligand.

Full text of DOI:10.1039/d0dt03604h

Volume 50
Number 10
1
4 March 2021
Pages 3381-3736
Dalton
Transactions
An international journal of inorganic chemistry
rsc.li/dalton
ISSN 1477-9226
PAPER
George Richter-Addo, Yong Zhang et al.
Insight into the preferential N-binding versus O-binding of
nitrosoarenes to ferrous and ferric heme centers
Anniversary
Volume

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Insight into the preferential N-binding versus
O-binding of nitrosoarenes to ferrous and ferric
heme centers†
Cite this: Dalton Trans., 2021, 50,
3487
a
b
a
b
Erwin G. Abucayon,
Jia-Min Chu,
Megan Ayala, Rahul L. Khade,
b
a
Yong Zhang * and George B. Richter-Addo
*
Nitrosoarenes (ArNOs) are toxic metabolic intermediates that bind to heme proteins to inhibit their func-
tions. Although much of their biological functions involve coordination to the Fe centers of hemes, the
factors that determine N-binding or O-binding of these ArNOs have not been determined. We utilize
X-ray crystallography and density functional theory (DFT) analyses of new representative ferrous and ferric
ArNO compounds to provide the first theoretical insight into preferential N-binding versus O-binding of
ArNOs to hemes. Our X-ray structural results favored N-binding of ArNO to ferrous heme centers, and
O-binding to ferric hemes. Results of the DFT calculations rationalize this preferential binding on the basis
of the energies of associated spin-states, and reveal that the dominant stabilization forces in the observed
ferrous N-coordination and ferric O-coordination are dπ–pπ* and dσ–pπ*, respectively. Our results
provide, for the first time, an explanation why in situ oxidation of the ferrous-ArNO compound to its ferric
state results in the observed subsequent dissociation of the ligand.
Received 17th October 2020,
Accepted 10th February 2021
DOI: 10.1039/d0dt03604h
rsc.li/dalton
drugs to their nitroso derivatives has been known for
Introduction
1
1–13
decades.
Confirmation of RNO as an inhibitory ligand
Nitrosoarenes and -alkanes (R–NvO; R = aryl, alkyl) are gener- that can bind to heme Fe centers was first reported by Mansuy
1
4,15
ated in vivo and in vitro from the reduction of nitroorganics in 1977 for a synthetic heme model system,
with the heme
(
RNO ) or from the oxidation of amine-containing (RNH ) model-RNO product displaying a similar UV-vis spectrum to
2
2
drugs. This class of compounds frequently displays biological that of the valence isoelectronic oxyferrous heme. Surprisingly,
1
–5
activity as a result of interactions with metalloproteins. The only a few heme protein-RNO derivatives of ferrous Hb, legHb,
coordination chemistry of RNO compounds is fairly well and myoglobin (Mb) have been characterized by X-ray
6
–8
16–19
established.
For monomeric RNO ligands, their coordi- crystallography,
and the data to date reveal an N-binding
nation to monometallic systems may occur through the mode of the RNO ligands to the ferrous heme centers.
N-atom, the O-atom, or through both atoms in a side-on N,O-
binding fashion (Fig. 1).
A particularly interesting class of RNO compounds are
those that contain para-amino functionalities, namely the
Interactions of RNO compounds with heme proteins are p-nitrosodialkylanilines (Fig. 2). Both NODMA and NODEA are
7
,8
20
particularly relevant to their bioinorganic chemistry.
For toxic, mutagenic, and exhibit bactericidal effects. NODMA-
2
1
example, in heme proteins containing exposed cysteine resi- dependent alcohol dehydrogenase enzymes are also known.
dues (e.g., human hemoglobin (Hb)), the RNO species may As with other nitrosoarenes, NODMA is known to interact with
interact directly with heme Fe and/or with the cysteine resi-
9
,10
dues (e.g., βCys93 in Hb) to alter their functions.
Importantly, the inhibition of heme enzymes such as cyto-
chrome P450 after metabolic activation of amine-containing
a
Department of Chemistry and Biochemistry, University of Oklahoma, 101
Stephenson Parkway, Norman, OK 73019, USA. E-mail: grichteraddo@ou.edu
b
Department of Chemistry and Chemical Biology, Stevens Institute of Technology,
Castle Point on Hudson, Hoboken, NJ 07030, USA. E-mail: yong.zhang@stevens.edu
†
Electronic supplementary information (ESI) available: CCDC 2011046–2011048.
For ESI and crystallographic data in CIF or other electronic format see DOI:
0.1039/d0dt03604h
Fig. 1 Binding modes of monomeric RNO ligands to monometallic
centers.
1
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3487–3498 | 3487

View Article Online
Paper
Dalton Transactions
1
3,39–41
42
43,44
cyt P450,
NO synthase, microperoxidase 8,
and
4
5
prostaglandin H synthase, all results in the dissociation of
the respective RNO groups from the ferric centers. In some
cases, ferric intermediates “Fe(III)-RNO” (middle of Fig. 3) with
presumed weak interactions between the ferric centers and
Fig. 2 The nitrosoarenes NODMA and NODEA.
4
2,43,45
RNO ligands were observed,
although the exact nature of
RNO binding to the ferric centers was not established.
We previously reported our preliminary results of nitrosoar-
ene N-binding to the ferrous center of (TPP)Fe(PhNO)
Hb to inhibit its oxygen transport function resulting in
difficulty breathing, and can induce the onset of
2
, and
2
2–24
methemoglobinemia.
Coordination and bioinorganic
+ 27
O-binding to the ferric center of [(TPP)Fe(NODEA)
2
] .
compounds of NODMA and NODEA have been reported, with
both N- and O-binding to the transition metals established by
However, issues with extensive disorder in the crystal structure
of the latter O-bound derivative, and the fact that two different
nitrosoarenes were used for these two derivatives, prevented a
reliable comparison of their structural properties to assess the
effects of N-binding versus O-binding on their relative stabi-
lities. In this paper, we report the investigation of preferential
binding modes of the NODMA and NODEA ligands to ferrous
and ferric porphyrin centers. Importantly, we employ X-ray
crystallography to provide the first direct comparison regard-
ing the geometrical binding preferences as a function of Fe
oxidation state in heme models that are relevant to some bio-
logical systems. In addition, our Density Functional Theory
2
5–30
X-ray crystallography.
A useful historical predictor of the mode of binding of
nitrosoarenes derives from consideration of the Hard–Soft
Acid–Base concept. In a seminal paper by Pearson on the
3
1
3+
3+
topic, metal cations such as Fe and Co are classified as
hard acids” that have favorable interactions with hard bases
“
2
+
+
+
such as O-donors, whereas and Pt , Ag , Cu are considered
soft acids” that favor interaction with soft bases such as
“
2
+
2+
2+
S-donors. Metal cations such as Fe , Co , and Zn are con-
sidered “borderline”. Indeed, this predictor, in many cases,
has helped rationalize binding modes of ligands in several
coordination complexes, especially those of biological rele-
vance. The Hard–Soft Acid–Base concept, despite being useful
in many cases, has not been sufficient in explaining some
experimental observations of preferred binding modes,
especially when it comes to heme model complexes. For
example, the X-ray crystal structure of the model heme
(
DFT) calculations offer the first theoretical support of such a
differential coordination mode change due to the Fe oxidation
state with data from energies and optimized structures. Our
DFT results also revealed previously unknown electronic
insights of charges and molecular orbital features into the pre-
ferred stabilities of the experimentally observed coordination
modes. These results help provide an understanding of the
biological binding motifs of RNO compounds in ferrous and
ferric heme proteins and their model systems.
2 6
complex [(TPP)Co(NODMA) ]SbF (TPP = tetraphenylporphyri-
nato dianion) reveals an experimental N-binding of the ligand
3
+
28
to the hard Co metal center, in preference to the predicted
O-binding mode. Some flexibility was also observed when soft
+
cations such as Cu interact with nitrosoarenes to result in
Results and discussion
complexes displaying either the N-binding or O-binding
2
9,32,33
modes.
Further, an O-binding mode of a nitrosoarene
As this study focuses on the structural and electronic conse-
2
+
34
II
III
was established in a complex of the borderline Zn cation.
quences of nitrosoarene binding to Fe and Fe heme centers,
it is informative to first consider the properties of the free
This situation is complicated further upon consideration of
the fact that nitrosoarenes are themselves redox active that can
4
6,47
48
ligands. The crystal structures of NODMA
and NODEA
8
,35,36
serve as π-acid ligands towards metal centers.
have been reported. Both structures suffer from disorder in
their –CNO fragments, but the overall geometrical data
sufficiently define a substantial contribution of the zwitter-
ionic quinoidal structure shown on the right of Fig. 4.
To date, N-binding of RNO ligands to ferrous heme proteins
and models appears to be the favored binding mode based on
the experimental data. Oxidation of the ferrous heme–RNO
complexes generally results in spectral changes that are
accompanied by the loss of the RNO ligand or its modified
form (Fig. 3). In particular, addition of ferricyanide as an
Consistent with the significant zwitterionic contribution
are (i) the planarity of the ONC
long (L) and short (S) bond-length alteration within the aryl
– fragment, and (iii) the larger (O)NCC angles cis to the
6 4 2
H NR core, (ii) the observed
3
7,38
37
oxidant to solutions of RNO-adducts of ferrous Hb,
Mb,
6 4
–C H
nitroso O than trans to O (by ∼10–15°) attributed to intra-
Fig. 3 Oxidation of ferrous heme-RNO compounds.
Fig. 4 Zwitterion contributions to NODMA and NODEA.
3488 | Dalton Trans., 2021, 50, 3487–3498
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
molecular repulsive interactions involving the nitroso O-atom. complexes, it is not surprising that the unambiguous determi-
This zwitterionic contribution appears to correlate with the nations of N- vs. O-binding modes of NODMA/NODEA to metal
difficulty of assigning the vibrational stretching frequency of centers have been through the use of X-ray crystallography. For
the NO bond (νNO). Unlike most nitrosoalkanes and nitrosoar- example, a proposed O-binding mode of NODMA to a cobalt
5
5,56
enes where νNO’s have been assigned with reasonable center based on IR spectroscopy
confidence,
was revised to an
7
,49
25
the νNO’s of NODMA and NODEA (both free N-binding mode based on X-ray crystallography. Indeed, the
and liganded) have historically been the subject of much con- IR spectra of the complexes prepared in this current work
troversy, as discussed by Gowenlock, Cameron, and (Experimental section, and Fig. S6 and S8–S10 in the ESI†)
5
0–53
15
18
Lüttke.
Contributing to this difficulty in νNO assignment is reveal several N-nitroso and O-nitroso isotope sensitive
the extensive vibrational coupling between νNO and νCC and bands, making it difficult to unambiguously assign the νNO
νCH. This is evidenced by the number of IR bands that shift in vibrations in these compounds. Consequently, obtaining
1
5
18
response to N-nitroso and O-nitroso isotopic substitution crystal structures of both the ferrous and ferric derivatives of
for both NODMA (Fig. 5, top) and NODEA (Fig. 5, bottom); a NODMA and NODEA became an absolute requirement for our
dynamic visual of this vibrational coupling is shown in the study in order to assign the binding modes with confidence.
Fig. S5 movie file in the ESI.† Perhaps the most reliable
The ferrous systems
reported assignment of νNO of NODMA to date is that provided
5
4
6
II
in the Ph.D. dissertation of Knieriem that documents a Reaction of NODEA with in situ-generated ferrous d (OEP)Fe ,
1
5
similar observation of multiple N-isotope sensitive band in a manner similar to that used for the preparation of (TPP)Fe
shifts, and assigns a νNO value of 1363 cm− based on both (PhNO)2, resulted in the formation of the mono-nitrosoarene
1
27
1
5
2
NO and H isotope substitutions.
2 6 4 2
adduct (OEP)Fe(NODEA)(NH C H NEt -p) (eqn (1)).
Given the historical complexity of νNO assignment in
NODMA and NODEA in both the free ligands and their metal
ð1Þ
The complex was isolated in good yield and is air-stable as
a solid for several days. To date, we have been unable to obtain
suitable crystals of the expected bis-ArNO (OEP)Fe(NODEA)2
derivative. The six-coordinate mono-NODEA derivative was
likely obtained due to the serendipitous in situ Zn-reduction of
the NODEA reagent present in excess in the reaction mixture.
We note that the chemical reduction of nitrosoarenes such as
NODMA (e.g., by Zn or Fe, with proton sources) to their amines
5
7
are well-known.
The molecular structure of (OEP)Fe(NODEA)(NH
2 6 4 2
C H NEt -
p) was identified by X-ray crystallography and is shown in
Fig. 6; selected bond lengths and angles for the structures
obtained in this work are collected in Table 1. The axial N/O
atoms in the crystal structure of (OEP)Fe(NODEA)-
(
NH C H NEt -p) exhibit a 90 : 10 positional disorder across
2 6 4 2
the porphyrin plane (Fig. S7†). The Fe–N(por) bond lengths of
.99–2.01 Å in (OEP)Fe(NODEA)(NH NEt -p) are consist-
1
2
C
6
H
4
2
6
58
ent with those expected for ferrous d low-spin hemes. The
axial Fe–N(O) bond length of 1.827(2) Å is shorter than that for
the trans Fe–NH
being close to the 2.028(2)–2.043(3) Å bond lengths observed
in the bis primary amine complexes (TPP)Fe(NH R) (R =
Ar bond in
OEP)Fe(NODEA)(NH C H NEt -p) is likely due to the presence
of the trans π-acceptor ArNO moiety. Consistent with this latter
feature is the slight apical displacement of 0.13 Å of the Fe
atom from the 24-atom porphyrin plane towards the ArNO
2
Ar bond length of 2.100(2) Å, with the latter
2
2
5
9
alkyl). The slight lengthening of this Fe–NH
2
Fig. 5 Truncated FTIR spectra of (A) NODMA and (B) NODEA (as KBr
(
1
5
18
2
6
4
2
pellets) and their N-nitroso (broken line trace) and O-nitroso (dotted
line trace) isotope-substituted derivatives. The major isotope-sensitive
bands in these truncated regions are shown in the respective boxes. See
Fig. S1–S4 in the ESI† for additional characterization data.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3487–3498 | 3489

View Article Online
Paper
Dalton Transactions
associated with the ON–aryl link are similar for N7–C47–C48
(
at 119.0(2)°) and N7–C47–C52 (at 121.2(2)°), with ∼2° differ-
ence being much smaller than the 10–12° observed in the free
ligand. Third, both the ON–C and (aryl)C–NEt bond lengths
2
are longer than those observed in the free ligand that has sig-
nificant quinoidal character. Fourth, the aryl C–C bond
lengths do not show the substantial alternating long-short-
long trend observed in the free ligand (Table 1; cf. Fig. 4). We
note that N-binding of NODEA/NODMA in metal derivatives
does not necessarily result in such deviations from the
3
0
quinoid structure of the free ligand, and a twist angle of only
∼
4° from planarity was observed in an N-bound Co–NODMA
2
5
complex.
We had anticipated that the observed significant deviation
from planarity and quinoidal character of the NODEA ligand
in (OEP)Fe(NODEA)(NH C H NEt -p), in effect making the
2 6 4 2
NODEA more of a “normal” ArNO ligand, would have allowed
us to estimate the νNO in this complex. For example, Zhang
and coworkers have used experimental IR data and detailed
computational methods to establish an inverse correlation of
d(N–O) with νNO in a series of heme–RNO/ArNO complexes.
Using their inverse correlation as a predictive tool, the experi-
Fig. 6 The molecular structure of (OEP)Fe(NODEA)(NH
2
C
6
H
4
NEt
2
-p)
49
2
with thermal ellipsoids drawn at 35%. Only the major axially NO/NH -
disordered (∼90%) component is shown (see Fig. S7†).
mental N–O bond length of 1.281(3) Å in (OEP)Fe(NODEA)
−
1
(
NH
2
C
6
H
4
NEt
2
-p) should correspond to a νNO of ∼1250 cm
.
Indeed, the IR spectrum of (OEP)Fe(NODEA)(NH
2
C
6
H
4
NEt -p)
2
Table 1 Selected bond lengths (Å) and angles (°) for the structurally
characterized ferrous and ferric compounds obtained in this work
1
5
−1
reveals an N-nitroso isotope sensitive band at 1230 cm
Fig. S6†). However, we are hesitant to assign this band to an
isolated vibration, as extensive vibrational coupling within
(
1
5
18
NODMA/NODEA results in multiple bands being N- and O-
isotope sensitive as described above (Fig. 5).
The ferric systems
Reactions of the ferric porphyrin precursors (por)FeFSbF
5
(por
Ferrous-OEP
Ferric-OEP
Ferric-TTP
=
OEP, TTP) in CH Cl with ∼1.5 equiv. of the nitrosoarenes
2
2
a
(
ArNO = NODMA and NODEA) result in the generation and
Fe–N(O)
Fe–O(N)
N–O
a
b/b’
c/c’
1.827(2)
—
—
—
1.281(3)
1.463(3)
1.388(3)/1.385(3)
1.379(3)/1.388(3)
1.407(3)/1.405(3)
1.390(3)
122.69(16)
—
1.9680(17)
1.318(2)
1.339(3)
1.431(3)/1.412(3)
1.350(3)/1.355(4)
1.453(3)/1.444(4)
1.333(3)
1.920(4)
1.334(5)
1.313(7)
1.425(8)/1.429(7)
subsequent isolation of the mono-nitrosoarene derivatives
a
a
[(por)Fe(ArNO)]SbF containing the uncoordinated anion. The
6
use of <2 equiv. of the nitrosoarene favors, in our hands, the
1.347(8)/1.339(8) isolation of the mono-nitrosoarene compounds that could be
d/d’
1.452(8)/1.434(8)
1.312(7)
—
113.5(3)
113.6(4)
crystallized into well-resolved structures.
e
a
∠
∠
∠
FeNO
FeON
ONC
—
115.52(13)
114.64(19)
a
110.59(18)
a
Data for the major (∼90%) NO/NH
2
-disordered component.
ð2Þ
ligand. In this structure, the NO group is oriented in a position
that essentially bisects adjacent porphyrin N atoms.
There are several interesting structural features of the the Evans’ method, of 4.8–4.9 BM suggesting admixed-spin
bound NODEA ligand in the crystal structure of (OEP)Fe systems of S = 3/2 and 5/2 in solution.
These five-coordinate [(por)Fe(ArNO)]SbF6 compounds in
CDCl3 solvent displayed magnetic moments, determined by
6
1
(
NODEA)(NH
angle involving the nitroso group of the NODEA ligand is 58.2 [(OEP)Fe(NODEA)]SbF6 and [(TTP)Fe(NODMA)]SbF6 are dis-
4)°, and this large deviation from the planarity substantially played in Fig. 7. The most important feature of these structures
disrupts the overlap of the NO and aryl π systems observed in is the determination of the O-binding mode of the nitrosoar-
2
C
6
H
4
NEt
2
-p). First, the O1–N7–C47–C48 torsion
The crystal structures of the cations of the ferric derivatives
(
6
0
the free nitrosoarene. Second, the (O)NCC bond angles ene ligands to the ferric centers. The structure of the [(OEP)Fe
3490 | Dalton Trans., 2021, 50, 3487–3498
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
III
+
O-bound [(OEP)Fe (NODEA)] . The nitroso N–O bond length
of 1.318(2) Å in the ferric O-bound complex is longer than the
related distance of 1.281(3) Å in the ferrous N-bound deriva-
tive. Consistent with this is also the shorter (O)N–C bond
length of 1.339(3) Å in the ferric complex compared with 1.463
(
3) Å in the ferrous case. Of particular note is the essential pla-
narity of the ONC N-moiety in the ferric O-bound complex
6
H
4
with an O1–N5–C37–C38 torsion angle of −0.6(3)°, and the
larger N5–C37–C38 angle (125.1(2)°; cis to nitroso-O) compared
with the N5–C37–C42 angle (116.0(2)°; trans to nitroso-O). In
addition, the aryl C–C bonds in the O-bonded ferric system
show the alternating long-short-long bond lengths similar to
that observed in the free ligand (e.g., right of Fig. 4).
Similar geometrical parameters are extant in the crystal
+
structure of the ferric [(TTP)Fe(NODMA)] derivative (Table 1).
As with the ferrous-NODEA system, we are unable to determine
a reliable assignment of νNO in these ferric derivatives due to
1
5
extensive vibrational coupling even with N-nitroso isotopic
substitution (Fig. S8 and S9 in the ESI†).
Computational insight into the preferential N- versus
O-binding of the nitrosoarenes
In order to understand the electronic reasons for the preferen-
tial binding modes in the experimentally determined struc-
tures of the ArNO liganded ferrous and ferric hemes, we per-
formed a quantum chemical investigation of model systems
using ωB97XD, a recently developed hybrid Hartree–Fock and
DFT method with dispersion correction. We have found this
Fig. 7 The crystal structures of the cations of (a) [(OEP)Fe(NODEA)] method to yield accurate predictions of various experimental
SbF
6
6
, and (b) [(TTP)Fe(NODMA)]SbF , with thermal ellipsoids drawn at
spectroscopic properties, structural features, and reactivity
35%.
62–68
results of iron porphyrin complexes.
We focused on the
electronic structures of the bis-ArNO and mono-ArNO liganded
systems with no other axial ligands, to exclude possible sec-
NODEA)] cation was ordered except for one of the ethyl C-atoms ondary electronic effects of other trans ligands. Using the
+
(
of the terminal NEt2 group. The Fe–N(por) bond lengths of parent unsubstituted porphine macrocycle, we calculated the
.0284(18)–2.0529(18) Å, the axial Fe–O length of 1.9680(17) Å, optimized geometries for both the N-binding mode for the
2
II
and the apical displacement of the Fe atom by +0.40 Å from the ferrous (Fe –N) system (left panel of Fig. 8) and O-binding
III
24-atom mean porphyrin plane towards the NODEA ligand are mode for the ferric (Fe –O) mono-NODMA system, as well as
II
III
consistent with its admixed-spin state. The axial Fe–O–N moiety the alternate but not observed Fe –O and Fe –N systems
is situated in a position that eclipses a porphyrin N-atom, with a (right panel of Fig. 8).
(
por)N2–Fe–O–N(NODEA) torsion angle of ∼0.2°.
The crystal structure of the [(TTP)Fe(NODMA)] cation is yielded results consistent with experiment. The optimized
The geometry optimizations and energy calculations
+
II
also ordered, with the exception of a methyl group of one of structure of the ferrous Fe –N mode showed that the ground
the porphyrin tolyl substituents. The geometrical data are also state is a singlet (S = 0), with the triplet and quintet states
in the range of those determined for an admixed-spin system, being >10 kcal mol− higher in Gibbs free energy that are
with the NODMA ligand in this case oriented in a manner that accompanied by the dissociation of one or both ligands. This
essentially bisects a pair of adjacent porphyrin N-atoms, with a singlet ground state agrees with the experimental data for
1
2
7
15,60
(
por)N1–Fe–O–N5 torsion angle of ∼35°, and the Fe atom api- ferrous (por)Fe(ArNO)2 and (por)Fe(ArNO)L compounds,
cally displaced by +0.48 Å from the 24-atom mean porphyrin and is also consistent with the fact that six-coordination in
plane towards the axial ligand.
Important differences are evident when comparing the geo- state.
metrical parameters of the nitrosoarene ligands in the
ferrous porphyrins is generally associated with the low-spin
6
9,70
The calculations also showed that the ground state of the
III
O-bound ferric complexes with that in the N-bound ferrous experimentally observed ferric Fe –O mode is an admixed S =
system described earlier. We will focus on the crystal structures 3/2 and S = 5/2 spin state, as these two spin states are very
of the ferrous and ferric OEP/NODEA pair, namely the close in energy; the energy (ΔG) of the high-spin state is only
II
−1
N-bound (OEP)Fe (NODEA)(NH C H NEt -p) versus the 2.22 kcal mol higher than that of the intermediate-spin state
2
6
4
2
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3487–3498 | 3491

View Article Online
Paper
Dalton Transactions
Table 3 Spin densities (in e units) of the ferric porphyrins with favor-
able spin states
Fe
Por
L
a
System
S
ρ
ρ
ρ
αβ
αβ
αβ
III
Fe –O
3/2
2.836
3.729
2.023
3.765
0.044
1.221
0.961
1.225
0.120
0.050
0.004
0.001
5
/2
3/2
/2
III
Fe –N
5
a
L is the axial ligand.
difference in spin density distribution is not unlike those
7
1
observed in related S = 5/2 and S = 3/2 iron porphyrins.
Importantly, the calculated geometries, especially those
involving the key bond lengths and angles involving the co-
ordinated ONC H NR -p ligands (Table 4), match very well
6 4 2
with the experimentally determined structures, with a mean
percentage error of 3%.
The energies (ΔG) of the alternate, but not experimentally
observed, binding modes for the ArNO ligands (right panel of
Fig. 8, and Table 2) were also probed computationally using
the favorable and experimentally observed spin states. For the
II
ferrous system, the alternate Fe –O binding mode is higher in
II
energy than the experimentally observed Fe –N binding mode
by 8.22 kcal mol− . For the ferric systems, the alternate Fe –N
1
III
binding mode is higher in energy than the experimentally
III
observed Fe –O binding mode by an average of ∼4–6 kcal
−
1
mol . These computational results clearly reveal that the
experimentally observed ArNO coordination modes originate
from the different relative stabilities of the N- and O-binding
forms of the molecular ferrous and ferric systems, and are not
artifacts resulting from crystal packing effects.
To help understand the origin of the observed different
stabilities associated with the ferrous and ferric porphyrins,
we first look into their geometries. As seen from Fig. 8, regard-
less of iron oxidation state, the N- and O-coordinated ligands
are tilted with respect to the porphyrin planes. However, there
is an interesting trend when comparing the experimental
structures of the ferrous N-coordinated and ferric
O-coordinated forms: the Fe–ON bond length is longer than
the Fe–NO, and the O-coordination is associated with a longer
N–O bond length, a shorter (O)N–C bond length, larger phenyl
C–C bond length variations (the difference between average of
II
III
Fig. 8 Optimized structures of the Fe –N and Fe –O (left panel) and
alternate coordination geometries (right panel) of NODMA-coordinated
ferrous and ferric porphines (atom colors: N-blue, O-red, C-cyan,
H-grey, Fe-black).
(
Table 2). This agrees with the experimental magnetic moment
data in solution determined by the Evans method
Experimental section). In contrast, the low-spin state (S = 1/2;
not shown) is significantly higher in energy than the high-spin
state by 6.52 kcal mol . The spin density data in these ferric
Fe –O systems (Table 3) show that for the S = 3/2 spin state, the
Fe center holds most of the spin density (2.836 e) with only
.044 e located on the porphine macrocycle. In contrast, for the
S = 5/2 spin state, the spin density is more generally distributed
between the Fe center (3.729 e) and the porphine (1.221 e). This
(
−1
III
0
(
(
N)C–C and the “internal” C–C bonds), as well as shorter C–N
Et ) bond lengths, compared to the N-coordination. These
2
structural changes point to a resonance structure difference as
shown in Fig. 4, i.e., the O-coordination prefers the right
zwitterionic resonance structure, while the N-coordination
favors the left neutral resonance structure, consistent with the
X-ray structural data shown in Table 1.
Table 2 Relative energy results (in kcal mol−1) of iron porphyrins with
favorable spin states (S)
Mode
S
ΔE
ΔEZPE
ΔH
ΔG
This feature of zwitterionic contribution is further sup-
ported by the charge analysis results as shown in Table 5.
II
Fe –N
0
0
3/2
0.00
11.65
0.00
5.77
4.30
6.44
0.00
10.59
0.00
3.31
2.23
5.29
0.00
11.08
0.00
3.82
2.03
5.38
0.00
II
Fe –O
8.22 Compared to the almost neutral NO moiety (−0.074 e) in the
III
Fe –O
0.00
2.22
4.26
II
N-coordinated ferrous porphyrin Fe –N, the NO fragment is
5
/2
5/2
/2
III
significantly more anionic with a −0.359 e charge in the
Fe –N
III
3
5.99 O-coordinated ferric porphyrin Fe –O, averaged for the S = 3/2
3492 | Dalton Trans., 2021, 50, 3487–3498
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
Table 4 Selected geometric parameters of iron porphyrins with favorable spin states (Units: Å and ° for bond length and bond angles respectively)
a
b
c
c
c
Mode
S
Fe–Np
Fe–N/O
N–O
a
b/d
c
e
∠FeNO/FeON
∠ONC
II
d
Fe –N
Expt.e
1.999
2.020
2.017
2.060
1.988
2.105
2.007
2.088
1.827
2.032
1.984
1.921
2.073
2.087
2.380
2.213
1.281
1.210
1.225
1.334
1.256
1.255
1.220
1.223
1.463
1.421
1.394
1.312
1.348
1.354
1.390
1.384
1.396
1.405
1.411
1.435
1.427
1.425
1.416
1.417
1.384
1.381
1.378
1.343
1.365
1.366
1.373
1.372
1.390
1.373
1.367
1.312
1.344
1.346
1.356
1.353
122.7
119.0
119.7
113.4
114.3
112.7
111.3
109.9
110.6
114.1
114.6
113.6
116.1
116.1
117.2
118.0
Calc.
Calc. f
Expt.
Calc.
0
0
II
Fe –O
III
Fe –O
3/2
5
/2
3/2
/2
III
Fe –N
Calc.
5
a
b
c
d
Averaged for four Fe–Npor bonds. Axial coordination to N or O. average of aryl bond lengths as defined in Table 1. (OEP)Fe(NODEA)
e
f
III
+
2 6 4 2
(NH C H NEt -p) (major component). for the bis-NODMA compound. [(TTP)Fe (NODMA)] .
Table 5 Atomic charges (in e units) of iron porphyrins with favorable
spin states
System
S
Q
Fe
Q
por
Q
NO
II
a
Fe –N
0
0
3/2
0.231
0.399
1.079
1.111
0.776
1.089
−0.737
−0.824
−0.355
−0.299
0.064
−0.074
−0.131
−0.334
−0.384
−0.277
−0.300
II
Fe –O
III
Fe –O
5
/2
3/2
/2
III
Fe –N
5
−0.269
a
Averaged for the same two ligands.
and 5/2 admixed states. In general, because O is more electrone-
gative than N, the O- and N-coordination may prefer the zwitter-
ionic and neutral forms, respectively. It is interesting to note
that, even for the ferrous systems, the change from
II
II
N-coordination (Fe –N) to O-coordination (Fe –O) also results in
a similar structural variation pattern of the nitrosoarene toward
zwitterionic character (although quantitatively smaller): a longer
N–O bond length, a shorter (O)N–C bond length, larger phenyl
C–C bond length variations, and a shorter C–N(Me ) bond
2
length, as seen from Table 4, and a more anionic NO (Table 5).
The same structural and charge variation pattern toward zwitter-
ionic character also occurs in moving from N-coordination
III
III
(
Fe –N) to O-coordination (Fe –O) for the ferric porphyrins.
III
Due to the ferric center in Fe –O having a much higher posi-
tive charge than in the ferrous porphyrin Fe –N (by ∼0.9 e,
II
Table 5), it can be better stabilized by the more zwitterionic
ligand form and thus favor the anionic O-coordination mode,
while the more neutral ferrous center will favor the more neutral
N-coordination mode. This is supported by the calculated
energy trend that shows that N-coordination is more thermo-
dynamically favored than O-coordination for ferrous porphyrins, Fig. 9 The MOs involving Fe and NO interactions for (a) the ferrous
II
III
Fe –N complex (porphine)Fe(NODMA)
[
2
, (b) the ferric Fe –O complex
while the opposite trend is observed for ferric porphyrins.
We then probed the electronic nature of the preferred N-
and O-coordination modes for the ferrous and ferric porphyr-
ins. The relevant MOs that involve significant interactions
+
III
(porphine)Fe(NODMA)] with S = 3/2, and (c) the latter ferric Fe –O
complex with S = 5/2. Contour values = ±0.02 au. The MO numbering is
for the α spin.
between Fe orbitals and the axial ArNO ligands in both the
II
ferrous and ferric systems are shown in Fig. 9. For ferrous Fe – while the σ-type bonding between the Fe dz2 orbital and the
N (Fig. 9a), the π-type bonding between the Fe d
π
orbitals ligand is in the LUMO region (LUMO+9). This suggests domi-
(including both dxz and dyz) and ligand pπ* orbitals can be nant π-bonding interactions that stabilize the ferrous
clearly seen in the HOMO region (HOMO−4 and HOMO−5), N-coordination mode.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3487–3498 | 3493

View Article Online
Paper
Dalton Transactions
In contrast, there are three significant differences that are Solvents were dried using a Pure Solv 400-5-MD (Innovative
III
evident for the MOs in the ferric Fe –O systems for both the S Technology) solvent purification system, or distilled from
=
3/2 (Fig. 9b) and S = 5/2 (Fig. 9c) spin states. First, the σ-type appropriate drying agents under nitrogen. The free base por-
interaction between the Fe d 2 orbital and the ArNO in-plane phyrin OEPH (octaethylporphyrin) was purchased from
i.e., ligand plane) π* orbitals is now located in the HOMO Frontier Scientific, and TTPH2 (tetratolylporphyrin) was syn-
z
2
(
7
2
region (HOMO−3). Second, unlike the two Fe d
interactions that are present in the ferrous complex, there is FeX (por = OEP, TTP; X = Cl, SbF
only one major Fe d and ArNO π interaction in the case of pared according to published procedures. Silver hexafluoroan-
ferric Fe –O, which involves the out-of-plane (i.e., perpendicu- timonate
lar to the ligand plane) π orbital with antibonding (HOMO−22 (p-Et NC
in Fig. 9b) and bonding (HOMO−26 in Fig. 9c) interactions (p-Me NC H NO; NODMA, 97%), and Dowex 50WX2 were pur-
π
and ArNO π* thesized by the Adler method. The metalloporphyrins (por)
7
3–75
76
6
)
and Zn/Hg were pre-
π
III
(AgSbF
6
,
99%),
N,N-diethyl-4-nitrosoaniline
2
6 4
H NO; NODEA, 97%), N,N-dimethyl-4-nitrosoaniline
2
6
4
1
8
with the Fe dyz orbital in the S = 3/2 and S = 5/2 spin states, chased from Sigma-Aldrich and were used as received. O-
1
5
respectively. Third, the σ-type bonding is more important than labeled water was purchased from Icon Isotopes. Na NO
2
and
the π-type of bonding for ferric Fe –O system, due to the fact chloroform-d (CDCl3, 99.96%D) was purchased from
that the σ-type bonding is near the surface of HOMO region, Cambridge Isotopes; CDCl was deaerated by three freeze–
III
3
while the π-type interaction is located in inner MOs as indi- pump–thaw cycles and stored over molecular sieves. IR spectra
1
cated by the relative MO numbering in Fig. 9b and c. This kind were collected on a Bruker Tensor 27 FTIR spectrometer. H
III
of bonding helps stabilize the Fe –O σ interaction between NMR spectroscopy was performed using a 400 MHz Varian
the ferric center and the anionic O-coordination from the NMR spectrometer. UHPLC-MS measurements were performed
zwitterionic resonance contribution, as also noted above from on a Waters (Milford, MA) Acquity chromatography system
the calculations of the structures and charges.
coupled with a Waters G2-Si Ion Mobility Q-TOF mass spectro-
meter equipped with an electrospray ionization source oper-
ated in positive ion mode.
1
5
15
15
15
N-labeled p-Me
2 6 4
NC H NO ( NODMA). The N-labeled
Conclusion
derivative was prepared in a similar manner to that used for
7
7
We have reported the preparation and crystal structural charac- the preparation of the unlabeled analogue, but with slight
terization of ferrous and ferric Fe–ArNO heme model com- modifications. To a cold (ice-bath) stirred solution of dimethyl-
pounds, and demonstrate that N-binding of the para-amino aniline (0.51 g, 4.21 mmol) in conc. HCl (∼2 mL) was added a
1
5
substituted ArNO ligand is favored for ferrous heme, and solution of Na NO (0.32 g, 4.57 mmol; in ∼1 mL of H O).
2
2
O-binding is favored for ferric heme. Examination of the geo- The solution was stirred for 1 h while cold (<8 °C), during
metrical features reveals that the quinoidal/zwitterionic charac- which time the color turned yellow-orange with formation of a
ter of the para-substituted ArNO ligand is prominent in the dark yellow precipitate. The precipitate was collected by
O-bound ferric system. Our results from DFT calculations on vacuum filtration, washed with HCl : H
2
O (1 : 1 v/v, 3 × 10 mL)
the N-binding and O-binding modes as a function of Fe oxi- followed by ethanol (3 × 10 mL), and subsequently dried under
1
5
dation and spin state are consistent with the experimentally vacuum to give p-Me NC H NO·HCl in ∼60% crude yield.
2
6
4
observed preferential N- and O-binding modes in the ferrous This salt was neutralized by addition of enough water to form
and ferric systems, respectively. a paste of the salt to which aq. NaOH (3 M) was added until
Overall, these results provide the first theoretical comparisons the solution turned basic (as judged using pH paper) and the
of structural features, charges, and molecular orbital interactions color changed to a bright green. The neutralized product was
due to Fe–N/O coordination in ArNO porphyrin complexes, and then extracted using benzene (3 × 10 mL), the extract then con-
reveal that the dominant stabilization forces in the observed centrated by slow evaporation (at ∼80 °C), and the resulting
1
5
ferrous N-coordination and ferric O-coordination are d
π 2 6 4
–pπ* and solution was cooled to yield crystals of the p-Me NC H NO
1
5
d
σ
–pπ*, respectively. These results support the experimentally ( NODMA) product which were isolated by filtration and air-
1
5
observed N-coordination of RNO compounds to ferrous heme dried overnight (78% isolated yield). IR (KBr; major N-
−1 1
proteins and the subsequent dissociation of such ligands upon isotope sensitive bands): 1388, 1360, 1332, and 1299 cm . H
13,37–44
in situ oxidation to the ferric state,
due to the instability of NMR (δ ppm, CDCl
3
, 500 MHz): 7.90 (v br, 2H, aryl-H), 6.69
N-coordination to ferric centers as revealed here.
(br, 2H, aryl-H), 3.18 (s, 6H, –N(CH ) ) (Fig. S1 in the ESI†).
3
2
ESI-TOF MS: m/z 152.0833 (calcd 152.0836) (Fig. S2 (middle) in
the ESI†).
1
5
15
15
N-labeled p-Et NC H NO ( NODEA). The diethyl ana-
2
5
6
4
15
Experimental section
1
logue p-Et
2
NC
6
H
4
NO ( NODEA) was prepared similarly, but
General
using Na CO
2
3
as the neutralization agent (60% isolated yield).
1
5
The reactions were performed anaerobically under an atmo- IR (KBr; major N-isotope sensitive bands): 1362, 1344, and
sphere of nitrogen unless otherwise noted. Air-sensitive 1327 cm− . H NMR (δ ppm, CDCl
1
1
, 500 MHz): 8.70 (v br, 2H,
3
samples and reagents were handled inside a glove box and all aryl-H), 6.67 (br, 2H, aryl-H), 3.51 (q, JCH 7 Hz, 4H, –N
reactions were performed using standard Schlenk glassware. (CH CH ) ), 1.28 (t, J 7 Hz, 6H, –N(CH CH ) ) (Fig. S3 in the
2
3 2
CH
2
3 2
3494 | Dalton Trans., 2021, 50, 3487–3498
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
ESI†). ESI-TOF MS: m/z 180.1158 (calcd 180.1143) (Fig. S4 in vacuo. The solid was redissolved in CH
2
Cl
2
(∼1.5 mL) and
(
bottom) in the ESI†).
carefully layered with n-hexane (∼3 mL) in a vial inside the
O- glove box. Slow evaporation of this solution to dryness inside a
1
8
18
18
18
O-labeled p-Me
2
NC
6
H
4
N O ( O-NODMA). The
labeled nitrite used for this reaction was prepared following a glove box afforded block-shaped crystals that were isolated by
78
literature procedure but with modifications due to our handpicking and identified as [(OEP)Fe(NODEA)]SbF (∼63%
6
inability to obtain anhydrous HCl during the COVID-19 lab isolated yield) by X-ray crystallography. An IR spectrum (KBr)
1
5
shutdown restrictions. To a solution of vacuum dried NaNO
(
2
of the crystals revealed several N-nitroso isotope sensitive
0.38 g, 5.51 mmol) in cold O-labeled water (H2 O, ∼1.5 mL, bands as shown in Fig. S8 in the ESI.† A spin-only magnetic
ice bath) was added Dowex 50WX2 (∼0.50 g) and allowed to moment of 4.91 BM was determined for the crystalline
1
8
18
warm to room temperature and kept at this temperature for complex in CDCl
3
(25 °C) by the Evan’s NMR method, which
7
9,80
∼
24 h. In a separate vial, a stirred solution of dimethylaniline suggests an admixed-spin system of S = 3/2 and 5/2.
1
8
(0.62 g, 5.12 mmol) in cold H
2
O (∼1.5 mL, ice bath) was
6 2 2
[(TTP)Fe(NODMA)]SbF . To a CH Cl (10 mL) solution of
5
mixed with Dowex 50WX2 (∼0.75 g), and to this mixture was (TTP)FeFSbF (31.2 mg, 0.033 mmol) was added NODMA
slowly added the mixture of NaNO /Dowex 50WX2, and stirred (7.5 mg, 0.050 mmol, ∼1.5 equiv.). The mixture was stirred for
2
for an additional ∼30 min. The color of the combined reaction 2 h during which time the color changed from orange-red to
solution turned deep green during this period. The solution red. The solution was concentrated under reduced pressure,
was decanted from the Dowex resin, and the product was and the product was precipitated with n-hexane (15 mL). The
extracted using benzene (2 × 10 mL). The benzene extract was supernate was decanted and the solid was washed with
2 3
dried with anhydrous K CO , and the product isolated by evap- n-hexane (3 × 10 mL) and subsequently dried in vacuo. The solid
oration of the benzene at ∼80 °C and air-dried overnight. The was redissolved in CH Cl (∼ 2 mL), and the solution carefully
2
2
remaining mixture, after the benzene extraction step, was neu- layered with n-hexane (5 mL). Placing the mixture at −25 °C for
tralized with NaOH (3 M) to recover more product. The total ∼1 d resulted in the formation of crystals which were isolated
yield of the product, in our hands, was low (∼5%). IR (KBr; by decanting the supernate and drying the crystals using a flow
1
8
major O-isotope sensitive bands): 1387, 1364, 1330, and of nitrogen gas. The crystals were identified as [(TTP)Fe
1
1
294 cm− . ESI-TOF MS: m/z 153.0917 (calcd 153.0908); the (NODMA)]SbF
6
(82% yield) using X-ray crystallography. An IR
1
8
15
ratio of O-labeled : unlabeled NODMA was 2 : 1, indicating spectrum (KBr) of the crystals revealed several N-nitroso
66% isotope incorporation (Fig. S2 (bottom) in the ESI†). isotope sensitive bands as shown in Fig. S9 in the ESI.† A spin-
OEP)Fe(NODEA)(NH NEt -p). To a THF (10 mL) solu- only magnetic moment of 4.78 BM was determined for the crys-
tion of (OEP)FeCl (25.0 mg, 0.040 mmol) was added excess Zn/ talline complex in CDCl (25 °C) by the Evan’s NMR method,
∼
(
2
C
6
H
4
2
3
7
9,80
Hg (48.2 mg, 0.74 mmol in Zn) and the mixture stirred for 1 h, which suggests an admixed-spin system of S = 3/2 and 5/2.
during which time the pale purple solution changed to a
bright red-purple. The supernatant solution was transferred by
X-ray crystallography
cannula into a separate Schlenk tube. To this air-sensitive solu- Single-crystal X-ray diffraction data were collected using a D8
tion was added NODEA (18.0 mg, 0.10 mmol, 2.5 equiv.), and QUEST diffractometer with a Bruker Photon II CPAD area
8
1
the reaction mixture was stirred for an additional 1 h. The THF detector and an Incoatec 1 μs microfocus Mo Kα radiation
was removed in vacuo, and the residue was washed with anhy- source (λ = 0.71073 Å), or with a Bruker APEX ccd area
8
2,83
drous n-hexane (3 × 10 mL). The resulting solid was redis- detector.
Diffraction data were collected from the samples
solved in CH Cl (∼1.5 mL) and transferred to a separate vial, at 100(2) K. The structures were solved by direct methods and
2
2
and the solution carefully layered with n-hexane (∼3 mL). Slow using the SHELXTL system and refined by full-matrix least-
2
84
evaporation of the solvent mixture to dryness inside a glove squares methods on F . Details of crystal data and structural
box resulted in a formation of thin plates that were isolated by refinement parameters are collected in Table S1 in the ESI.†
handpicking and identified by X-ray crystallography as (OEP)Fe CCDC 2011046–2011048 contain the supplementary crystallo-
(
NODEA)(NH
2
NC
6
H
4
NEt
2
-p) in ∼80% yield based on Fe. As graphic data.
(OEP)Fe(NODEA)(NH C H NEt -p).
with the compounds below, X-ray structural determinations
A
red plate-shaped
2
6
4
2
from several crystals from the batch revealed the formation of crystal of dimensions 0.040 × 0.156 × 0.173 mm was selected
only one crystalline product. An IR spectrum (KBr) of the crys- for structural analysis. Cell parameters were determined from
1
5
tals revealed several N-nitroso isotope sensitive bands as a non-linear least squares fit of 9890 peaks in the range 2.19 <
shown in Fig. S6 in the ESI.†
(OEP)Fe(NODEA)]SbF . To a CH
θ < 26.93°. A total of 72590 data were measured in the range
[
6
2
Cl
2
(10 mL) solution of 2.191 < θ < 27.366° using ϕ and ω oscillation frames. The data
8
5
(
OEP)FeFSbF5 (13.8 mg, 0.017 mmol) was added NODEA were corrected for absorption by the empirical method
1.9 mg, 0.023 mmol, ∼1.4 equiv.). The mixture was stirred for giving minimum and maximum transmission factors of 0.942
h during which time the color of the solution slowly changed and 0.986. The data were merged to form a set of 11 216 inde-
from light purple to red. The solution was concentrated to pendent data with R(int) = 0.0828 and a coverage of 99.9%.
about half volume and the product was precipitated using The monoclinic space group P2 /n was determined by systema-
(
2
1
n-hexane (∼15 mL). The supernate was decanted and the solid tic absences and statistical tests and verified by subsequent
was washed with n-hexane (3 × 10 mL) and subsequently dried refinement. Non-hydrogen atoms were refined with anisotropic
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3487–3498 | 3495

View Article Online
Paper
Dalton Transactions
displacement parameters. Hydrogen atom (bonded to carbon) of 99.9%. The monoclinic space group P2
positions were initially determined by geometry and refined by by systematic absences and statistical tests and verified by sub-
a riding model. The axial NO/NH ligands in the structure sequent refinement. Non-hydrogen atoms were refined with
1
/c was determined
2
were disordered. The occupancies of atoms N7, O1 and N5 anisotropic displacement parameters. Hydrogen atom (bonded
refined to 0.901(4) and 0.099(4) for the unprimed and primed to carbon) positions were initially determined by geometry and
atoms, respectively. Restraints on the positional and displace- refined by a riding model. Hydrogen atom displacement para-
ment parameters of the disordered atoms were required. meters were set to 1.2 times (1.5 for methyl) the displacement
Hydrogen atom displacement parameters were set to 1.2 times parameters of the bonded atoms. The anion and one of the
(1.5 for methyl) the displacement parameters of the bonded methyl groups (of TTP) were disordered. The occupancies of
atoms. A total of 611 parameters were refined against 51 the C48 atom refined to 0.50(6) and 0.50(6) for the unprimed
2
restraints and 11 216 data to give wR(F ) = 0.1064 and S = 1.027 and primed atoms, respectively. The occupancies of the anion
2
2
2
for weights of w = 1/[σ (F ) + (0.0330 P) + 3.5800 P], where P = refined to 0.872(5) and 0.128(5) for the unprimed and primed
2
2
[
F + 2F ]/3. The final R(F) was 0.0468 for the 7946 observed, [F atoms, respectively. Restraints on the positional and displace-
o c
>
4σ(F)], data. The largest shift/s.u. was 0.001 in the final ment parameters were required. A CH Cl solvent molecule
2 2
refinement cycle.
was severely disordered and its effects on the intensity data
purple block-shaped were removed using the Squeeze algorithm. A total of 708
8
6
[(OEP)Fe(NODEA)]SbF
6
·CH
2
Cl
2
.
A
crystal of dimensions 0.080 × 0.240 × 0.420 mm was selected parameters were refined against 1157 restraints and 10 138
2
for structural analysis. Cell parameters were determined from data to give wR(F ) = 0.1637 and S = 1.053 for weights of w = 1/
2
2
2
2
2
a non-linear least squares fit of 9975 peaks in the range 2.26 < [σ (F ) + (0.0520 P) + 20.7200 P], where P = [F
o c
+ 2F ]/3. The
θ < 29.22°. A total of 55 564 data were measured in the range final R(F) was 0.0636 for the 7048 observed, [F > 4σ(F)], data.
1
.393 < θ < 29.784° using ϕ and ω oscillation frames. The data The largest shift/s.u. was 0.000 in the final refinement cycle.
8
5
were corrected for absorption by the empirical method
giving minimum and maximum transmission factors of 0.3728 Computational methodology
and 0.4324. The data were merged to form a set of 13 621 inde-
pendent data with R(int) = 0.0400 and a coverage of 100.0%.
The triclinic space group P ˉ1 was determined by systematic
87
All calculations were performed using Gaussian 16. Full geo-
metry optimizations using the unsubstituted porphine (por)
macrocycle were conducted for all studied chemical systems,
with subsequent frequency calculations to verify the nature of
the corresponding stationary states on their potential energy
surfaces and provide zero-point energy corrected electronic
energies (EZPE’s), enthalpies (H′s), and Gibbs free energies (G′
s) at room temperature in addition to electronic energies (E′s).
absences and statistical tests and verified by subsequent
refinement. Non-hydrogen atoms were refined with anisotropic
displacement parameters. Hydrogen atom (bonded to carbon)
positions were initially determined by geometry and refined by
a riding model. Hydrogen atom displacement parameters were
set to 1.2 times (1.5 for methyl) the displacement parameters
of the bonded atoms. The anion was disordered, with occu-
pancies refining to 0.9687(5) and 0.0313(5) for the unprimed
and primed atoms, respectively; restraints on the positional
and displacement parameters of the disordered atoms were
required. The displacement parameters of the two Sb atoms
were constrained to be equal. The occupancies of the ligand
88
The used method includes the ωB97XD functional with the
basis set LanL2DZ for Fe, 6-311++G(2d,2p) for first shell
atoms (porphyrin N atoms and RNO’s NO moiety), and 6-31G
d) for the rest of the atoms, which was the same for all
systems, all bonding situations, and all spin states studied
here. This functional enabled accurate predictions of various
experimental geometric parameters, spectroscopic properties,
89
(
NEt
2
atoms C45 and C46 were also disordered and refined to
62–68
and reactivities of iron porphyrin complexes
and other
0
.867(5) and 0.133(5) for unprimed and primed atoms, respect-
90
transition metal complexes, and this basis set also well
reproduced many experimental properties of similar NO/HNO
The atomic charges and spin densities
reported here are from the Natural Population Analysis (NPA)
and Mulliken schemes respectively, as implemented in
ively. A total of 654 parameters were refined against 624
2
restraints and 13 621 data to give wR(F ) = 0.1056 and S = 1.005
91–93
heme systems.
2
2
2
for weights of w = 1/[σ (F ) + (0.0560 P) + 1.6400 P], where P =
[
2
2
F + 2F ]/3. The final R(F) was 0.0417 for the 11 034 observed,
o c
[F > 4σ(F)], data. The largest shift/s.u. was 0.002 in the final
87
Gaussian 16.
refinement cycle.
(TTP)Fe(NODMA)]SbF . A purple plate-shaped crystal of
[
6
dimensions 0.024 × 0.132 × 0.233 mm was selected for struc-
tural analysis. Cell parameters were determined from a non-
Conflicts of interest
linear least squares fit of 9853 peaks in the range 2.31 < θ < There are no conflicts to declare.
2
4.73°. A total of 80 385 data were measured in the range 2.310
<
θ < 25.383° using ϕ and ω oscillation frames. The data were
corrected for absorption by the semi-empirical from equiva-
lents method giving minimum and maximum transmission
Acknowledgements
8
5
factors of 0.833 and 0.981. The data were merged to form a set This material is based upon work supported by (while GBR-A
of 10 138 independent data with R(int) = 0.0842 and a coverage was serving at) the U.S. National Science Foundation (NSF;
3496 | Dalton Trans., 2021, 50, 3487–3498
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
CHE-1900181). Any opinion, findings, and conclusions or rec- 20 C. M. Goodall, C. M. Moore and O. B. Stephens,
ommendations expressed in this material are those of the Xenobiotica, 1986, 16, 587–593.
authors and do not necessarily reflect the views of the NSF. We 21 T. Daussmann, A. Aivasidis and C. Wandrey,
are also grateful to the U.S. National Institutes of Health Eur. J. Biochem., 1997, 248, 889–896.
GM085774 to Y.Z.) for funding this work, and to the NSF MRI 22 O. Bodansky, Pharmacol. Rev., 1951, 3, 144–196.
program (CHE-1726630 to GBR-A) for funds to purchase the 23 J. H. Harrison and D. J. Hollow, Mol. Pharmacol., 1987, 32,
X-ray diffractometer. We thank Dr Douglas R. Powell (OU) for 423–431.
the X-ray structural solution of [(OEP)Fe(NODEA)]SbF6 and 24 T. E. Kearney, A. S. Manoguerra and J. V. Dunford, Jr.,
assistance with the re-refinement of the other structures. We West. J. Med., 1984, 140, 282–286.
also thank Jennifer Londono for technical assistance. GBR-A is 25 D. B. Sams and R. J. Doedens, Inorg. Chem., 1979, 18, 153–
especially grateful to Professor Armin de Meijere at the 156.
University of Göttingen for retrieving, with the assistance of 26 S. J. Fox, L. Chen, M. A. Khan and G. B. Richter-Addo,
BK’s spouse, hardcopy of the Ph.D. dissertation of Inorg. Chem., 1997, 36, 6465–6467.
Dr Burkhard Knieriem (ref. 54) and mailing it to Oklahoma 27 L.-S. Wang, L. Chen, M. A. Khan and G. B. Richter-Addo,
(
a
during the COVID-19 pandemic.
Chem. Commun., 1996, 323–324.
2
8 L. Chen, J. B. Fox, Jr., G.-B. Yi, M. A. Khan and G. B. Richter-
Addo, J. Porphyrins Phthalocyanines, 2001, 5, 702–707.
9 R. S. Srivastava, M. A. Khan and K. M. Nicholas, J. Am.
Chem. Soc., 2005, 127, 7278–7279.
2
References
1
2
M. Kiese, Arch. Exp. Pathol. Pharmakol., 1959, 235, 360–364. 30 S. Wirth and I. P. Lorenz, Z. Naturforsch., B: J. Chem. Sci.,
F. Jung, Biochem. Z., 1940, 305, 248–260, (Chem. Abs., 235, 2012, 67, 532–542.
72). 31 R. G. Pearson, J. Am. Chem. Soc., 1963, 85, 3533–3539.
M. A. Belisario, R. Pecce, A. Garofalo, N. Sannolo and 32 R. S. Srivastava, N. R. Tarver and K. M. Nicholas, J. Am.
A. Malorni, Toxicology, 1996, 108, 101–108. Chem. Soc., 2007, 129, 15250–15258.
J. H. Harrison and D. J. Jollow, J. Pharmacol. Exp. Ther., 33 M. S. Askari, B. Girard, M. Murugesu and X. Ottenwaelder,
986, 238, 1045–1054. Chem. Commun., 2011, 47, 8055–8057.
M. Roldan, E. Perez-Reinado, F. Castillo and C. Moreno- 34 S. Hu, D. M. Thompson, P. O. Ikekwere, R. J. Barton,
4
3
4
5
6
7
8
9
1
Vivian, FEMS Microbiol. Rev., 2008, 32, 474–500.
M. Cameron, B. G. Gowenlock and G. Vasapollo, Chem. Soc.
Rev., 1990, 19, 355–379.
J. Lee, L. Chen, A. H. West and G. B. Richter-Addo, Chem.
Rev., 2002, 102, 1019–1065.
K. E. Johnson and B. E. Robertson, Inorg. Chem., 1989, 28,
4552–4554.
35 N. C. Tomson, L. A. Labios, T. Weyhermuller, J. S. Figueroa
and K. Wieghardt, Inorg. Chem., 2011, 50, 5763–5776.
36 L. Chen, M. A. Khan, G. B. Richter-Addo, V. G. Young, Jr.
and D. R. Powell, Inorg. Chem., 1998, 37, 4689–4696.
37 D. Mansuy, J. C. Chottard and G. Chottard, Eur. J. Biochem.,
1977, 76, 617–623.
38 D. Keilin and E. F. Hartree, Nature, 1943, 151, 390–391.
39 J. M. Fukuto, J. F. Brady, J. N. Burstyn, R. B. VanAtta,
J. S. Valentine and A. K. Cho, Biochemistry, 1986, 25, 2714–
2719.
N. Xu and G. B. Richter-Addo, Prog. Inorg. Chem., 2014, 59,
3
81–445.
P. Eyer and M. Ascherl, Biol. Chem. Hoppe-Seyler, 1987, 368,
85–294.
2
1
1
1
0 P. Eyer and E. Lierheimer, Xenobiotica, 1980, 10, 517–526.
1 M. R. Franklin, Chem.-Biol. Interact., 1976, 14, 337–346.
2 C. Bensoussan, M. Delaforge and D. Mansuy, Biochem.
Pharmacol., 1995, 49, 591–602.
40 D. Mansuy, P. Beaune, T. Cresteil, C. Bacot, J.-C. Chottard
and P. Gans, Eur. J. Biochem., 1978, 86, 573–579.
41 J. Werringloer and R. W. Estabrook, Arch. Biochem. Biophys.,
1975, 167, 270–286.
1
1
1
1
1
1
1
3 D. Mansuy, P. Gans, J.-C. Chottard and J.-F. Bartoli,
Eur. J. Biochem., 1977, 76, 607–615.
4 D. Mansuy, P. Battioni, J. C. Chottard and M. Lange, J. Am.
Chem. Soc., 1977, 99, 6441–6443.
5 D. Mansuy, P. Battioni, J.-C. Chottard, C. Riche and
A. Chiaroni, J. Am. Chem. Soc., 1983, 105, 455–463.
42 A. Renodon, J.-L. Boucher, C. Wu, R. Gachhui, M.-A. Sari,
D. Mansuy and D. Stuehr, Biochemistry, 1998, 37, 6367–
6374.
6 J. Yi, G. Ye, L. M. Thomas and G. B. Richter-Addo, Chem. 43 R. Ricoux, J.-L. Boucher, D. Mansuy and J.-P. Mahy,
Commun., 2013, 49, 11179–11181. Biochem. Biophys. Res. Commun., 2000, 278, 217–223.
7 D. M. Copeland, A. H. West and G. B. Richter-Addo, 44 R. Ricoux, E. Ludowska, F. Pezzotti and J.-P. Mahy,
Proteins: Struct., Funct., Genet., 2003, 53, 182–192. Eur. J. Biochem., 2004, 271, 1277–1283.
8 S. M. Powell, L. M. Thomas and G. B. Richter-Addo, 45 J. P. Mahy and D. Mansuy, Biochemistry, 1991, 30, 4165–
J. Inorg. Biochem., 2020, 213, 111262. 4172.
9 I. P. Kuranova, A. V. Teplyakov, G. V. Obmolova, 46 C. Rømming and H. J. Talberg, Acta Chem. Scand., 1973, 27,
A. A. Voronova, A. N. Popov, D. M. Kheiker and 2246–2248.
E. G. Harutyunyan, Bioorg. Khim., 1982, 8, 1625–1636 (in 47 K. Lewinski, W. Nitek and P. Milart, Acta Crystallogr., Sect.
Russian). Chem. Abstr., CA1698, 48994. C: Cryst. Struct. Commun., 1993, 49, 188–190.
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3487–3498 | 3497

View Article Online
Paper
Dalton Transactions
4
4
8 H. J. Talberg, Acta Chem. Scand., Ser. A, 1977, 31, 743–751.
9 Y. Ling, C. Mills, R. Weber, L. Yang and Y. Zhang, J. Am.
Chem. Soc., 2010, 132, 1583–1591.
0 M. Cameron, B. G. Gowenlock and G. Vasapollo,
J. Organomet. Chem., 1991, 325, 325–333.
74 E. T. Kintner and J. H. Dawson, Inorg. Chem., 1991, 30,
4892–4897.
75 K. Shelly, T. Bartczak and W. R. Scheidt, Inorg. Chem., 1985,
24, 4325–5330.
5
76 P. D. Caesar, Org. Synth., 1963, Coll. Vol. 4, 695–696.
5
1 M. Cameron, B. G. Gowenlock and G. Vasapollo, 77 G. M. Bennett and E. V. Bell, Org. Synth., 1943, Coll. Vol. 2,
Polyhedron, 1994, 13, 1371–1377.
223–225.
5
5
2 W. Lüttke, Angew. Chem., 1958, 70, 576–577.
3 M. Cameron, B. G. Gowenlock and G. Vasapollo,
J. Organomet. Chem., 1989, 378, 493–496.
78 D. Samuel and I. Wassermann, J. Labelled Compd., 1971, 7,
355–356.
79 C. A. Reed, T. Mashiko, S. P. Bentley, M. E. Kastner,
W. R. Scheidt, K. Spartalian and G. Lang, J. Am. Chem. Soc.,
1979, 101, 2948–2958.
5
5
5
5
5
4 B. Knieriem, Ph.D. Dissertation, University of Göttingen,
1
972.
5 C. J. Popp and R. O. Ragsdale, Inorg. Chem., 1968, 7, 1845– 80 M. M. Maltempo and T. H. Moss, Q. Rev. Biophys., 1976, 9,
848. 181–215.
6 C. J. Popp and R. O. Ragsdale, J. Chem. Soc. (A), 1970, 1822– 81 Bruker AXS Inc, Madison, WI, USA, 2016.
825. 82 APEX2, Bruker-AXS, Madison, WI, USA, 2007.
7 K. Polat, M. L. Aksu and A. T. Pekel, J. Appl. Electrochem., 83 SAINT, Data Reduction: SAINT Software Reference Manual.
000, 30, 733–736, and references therein. Bruker-AXS, Madison, WI., 2007.
8 W. R. Scheidt, in The Porphyrin Handbook, ed. 84 G. M. Sheldrick, Acta Crystallogr., Sect. C: Cryst. Struct.
1
1
2
K. M. Kadish, K. M. Smith and R. Guilard, Academic Press,
New York, 2000, ch. 16, vol. 3.
9 O. Q. Munro, P. S. Madlala, R. A. F. Warby, T. B. Seda and
G. Hearne, Inorg. Chem., 1999, 38, 4724–4736.
Commun., 2015, 71, 3–8.
85 L. Krause, R. Herbst-Irmer, G. M. Sheldrick and D. Stalke,
J. Appl. Crystallogr., 2015, 48, 3–10.
86 A. L. Spek, Acta Crystallogr., Sect. C: Cryst. Struct. Commun.,
2015, 71, 9–18.
5
6
0 N. Godbout, L. K. Sanders, R. Salzmann, R. H. Havlin,
M. Wojdelski and E. Oldfield, J. Am. Chem. Soc., 1999, 121, 87 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
3
829–3844.
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato,
A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts,
B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov,
J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini,
F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson,
D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng,
W. Liang, M. Hada, M. T. Ehara, K. R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
T. Vreven, K. Throssell, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers,
K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi,
J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant,
S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene,
C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin,
K. Morokuma, O. Farkas, J. B. Foresman and D. J. Fox,
Gaussian-16 Rev. C.01, Wallingford, CT, 2016.
6
6
6
6
6
6
6
6
1 D. H. Grant, J. Chem. Educ., 1995, 72, 39–40, and references
therein.
2 R. L. Khade, A. L. Chandgude, R. Fasan and Y. Zhang,
ChemCatChem, 2019, 11, 3101–3108.
3 R. L. Khade, W. Fan, Y. Ling, L. Yang, E. Oldfield and
Y. Zhang, Angew. Chem., Int. Ed., 2014, 53, 7574–7578.
4 R. L. Khade and Y. Zhang, J. Am. Chem. Soc., 2015, 137,
7
560–7563.
5 R. L. Khade and Y. Zhang, Chem. – Eur. J., 2017, 23, 17654–
7658.
1
6 D. A. Vargas, R. L. Khade, Y. Zhang and R. Fasan, Angew.
Chem., Int. Ed., 2019, 58, 10148–10152.
7 Y. Wei, A. Tinoco, V. Steck, R. Fasan and Y. Zhang, J. Am.
Chem. Soc., 2018, 140, 1649–1662.
8 A. Tinoco, Y. Wei, J. P. Bacik, E. J. Moore, N. Ando,
Y. Zhang and R. Fasan, ACS Catal., 2019, 9, 1514–
1
524.
88 J.-D. Chai and M. Head-Gordon, Phys. Chem. Chem. Phys.,
2008, 10, 6615–6620.
6
9 C. A. Reed, T. Mashiko, W. R. Scheidt, K. Spartalian and
G. Lang, J. Am. Chem. Soc., 1980, 102, 2302–2306.
89 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270–283.
7
0 P. G. Debrunner, in Iron Porphyrins, ed. H. B. Gray and 90 Y. Minenkov, Å. Singstad, G. Occhipinti and V. R. Jensen,
A. B. P. Lever, VCH Publishers, New York, 1989, pp.
39–234.
1 Y. Ling and Y. Zhang, J. Am. Chem. Soc., 2009, 131, 6386–
388.
2 A. D. Adler, F. R. Longo, J. D. Finarelli, J. Goldmacher,
J. Assour and L. Korsakoff, J. Org. Chem., 1967, 32, 476.
Dalton Trans., 2012, 41, 5526–5541.
1
91 E. G. Abucayon, R. L. Khade, D. R. Powell, Y. Zhang and
G. B. Richter-Addo, J. Am. Chem. Soc., 2016, 138, 104–107.
92 E. G. Abucayon, R. L. Khade, D. R. Powell, M. J. Shaw,
Y. Zhang and G. B. Richter-Addo, Dalton Trans., 2016, 45,
18259–18266.
7
7
7
6
3 A. D. Adler, F. R. Longo, F. Kampas and J. Kim, J. Inorg. 93 Y. Shi and Y. Zhang, Angew. Chem., Int. Ed., 2018, 57,
Nucl. Chem., 1970, 32, 2443–2445.
16654–16658.
3498 | Dalton Trans., 2021, 50, 3487–3498
This journal is © The Royal Society of Chemistry 2021