Developing iron nitrosyl complexes as NO donor prodrugs

Article abstract of DOI:10.1039/b702461d

A novel class of water-soluble iron nitrosyl complexes has been developed for use as NO donor prodrugs. To elaborate these NO prodrugs various water-soluble ligands were used such as P(CH₂OH)₃, 1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane (PTA), 1,2-bis[bis(hydroxymethyl) phosphino]ethane (HMPE), 1,2-bis[bis(hydroxymethyl)phosphino]benzene (TMBz), cysteamine, cysteamine hydrochloride, l-cysteine ethyl ester hydrochloride (LCEE) and pyrimidine-2-thiol (pyrim). The mononuclear complexes Fe(NO) ₂P(CH₂OH)₃Cl 1, Fe(NO)₂(P(CH ₂OH)₃)₂ 2, Fe(NO)₂(PTA)₂ 3, Fe(NO)₂HMPE 4, Fe(NO)₂TMBz 5, [Fe(NO) ₂pyrimI] 10, [Fe(NO)₃P(CH₂OH)₃][X] (X = PF₆, SbF₆, BF₄) 11-13 and the dinuclear species [Fe(NO)₂S(CH₂)₂NH₃Cl] ₂ 6a, [Fe(NO)₂S(CH₂)₂NH ₃I₂] 6b, [Fe(NO)₂LCEE]₂ 8 and [Fe(NO)₂pyrim]₂ 9 were obtained. Complexes 1, 2, 4, 6a, 6b, 11, 12 and 13 are water-soluble. 1, 2 and 4 were identified as nitroxyl and 6a, 6b, 11, 12 and 13 as nitric oxide donors by applying an EPR NO-trap assay. To determine the amount of nitric oxide which was released from the nitric oxide donors, an additional electrochemical methodology was used. The equilibrium release or the trapping concentration of NO was also studied by a UV-vis method, which allowed the rate constant of NO release to be determined. The Royal Society of Chemistry.

Full text of DOI:10.1039/b702461d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Developing iron nitrosyl complexes as NO donor prodrugs†‡
Sandra A. T. Dillinger, Helmut W. Schmalle, Thomas Fox and Heinz Berke*
Received 15th February 2007, Accepted 11th June 2007
First published as an Advance Article on the web 17th July 2007
DOI: 10.1039/b702461d
A novel class of water-soluble iron nitrosyl complexes has been developed for use as NO donor
prodrugs. To elaborate these NO prodrugs various water-soluble ligands were used such as P(CH2OH)3,
1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane (PTA), 1,2-bis[bis(hydroxymethyl)phosphino]ethane
(HMPE), 1,2-bis[bis(hydroxymethyl)phosphino]benzene (TMBz), cysteamine, cysteamine
hydrochloride, L-cysteine ethyl ester hydrochloride (LCEE) and pyrimidine-2-thiol (pyrim). The
mononuclear complexes Fe(NO)2P(CH2OH)3Cl 1, Fe(NO)2(P(CH2OH)3)2 2, Fe(NO)2(PTA)2 3,
Fe(NO)2HMPE 4, Fe(NO)2TMBz 5, [Fe(NO)2pyrimI] 10, [Fe(NO)3P(CH2OH)3][X] (X = PF6, SbF6,
BF4) 11–13 and the dinuclear species [Fe(NO)2S(CH2)2NH3Cl]2 6a, [Fe(NO)2S(CH2)2NH3I2] 6b,
[Fe(NO)2LCEE]2 8 and [Fe(NO)2pyrim]2 9 were obtained. Complexes 1, 2, 4, 6a, 6b, 11, 12 and 13 are
water-soluble. 1, 2 and 4 were identified as nitroxyl and 6a, 6b, 11, 12 and 13 as nitric oxide donors by
applying an EPR NO-trap assay. To determine the amount of nitric oxide which was released from the
nitric oxide donors, an additional electrochemical methodology was used. The equilibrium release or
the trapping concentration of NO was also studied by a UV-vis method, which allowed the rate
constant of NO release to be determined.
Twenty years ago, NO was discovered to be involved in several
physiological and pathophysiological processes in mammalians.1,2
A lack of NO production has been related to diseases such
as hypertension or arteriosclerosis. In such cases organic and
metal organic NO prodrugs, which under physiological conditions
release nitric oxide, have been widely used in clinical treatments.
The most important of which are the triglyceryl3 and the sodium
nitroprusside (SNP) compounds.4 Meanwhile major problems
such as nitrate tolerance for triglyceryl and cyanide poisoning for
SNP have been revealed for some patients.5 In this paper we present
the synthesis of new nitroxyl and nitric oxide donors and their
analysis by biophysical methods such as electron paramagnetic
resonance2,6 (EPR-NO trap method) and electrochemistry. These
methods not only allowed us to detect and identify7 the nature
of the released NO, but also to quantify8 the amount of NO
generated under pseudo-physiological conditions. In such a way,
kinetic studies using pseudo-first order conditions have also been
carried out to determine the rate constant of NO release.
elemental analysis could not be obtained. The 31P NMR spectrum
shows one singlet at around d = 59.5 ppm corresponding to a metal
bound phosphine.11 In the IR spectrum of 1 the NO bands are
found at 1720 and 1678 cm−1. They are shifted to lower frequencies
as compared to the dinuclear starting material where the bands are
found at 1810 and 1725 cm−1. Conductivity measurements showed
that 1 behaves in water as a 1 : 1 electrolyte. Here the chloro ligand
seems to ionise off, as demonstrated in Fig. 1.
Results and discussion
The paramagnetic d7 Fe(NO)2P(CH2OH)3Cl 1 is a 17e− complex,9
10
which can be prepared by the reaction of [Fe(NO)2Cl]2 with
11
two equivalents of P(CH2OH)3 at r.t. Compound 1 is a rose
powder with high solubility and stability in water and alcohols.
It crystallizes with varying amounts of solvent so that correct
Fig. 1 Conductivity of a dilution series of Fe(NO)2P(CH2OH)3Cl in water
at 25 ◦C. Conductivity Km = 195.56 S cm2 mol−1
.
The diamagnetic d6 complex Fe(NO)2[P(CH2OH)3]2 2 can be
obtained from the reaction between Fe(CO)2(NO)2 and two
equivalents of P(CH2OH)3. Complex 2 is a red solid and possesses
good solubility and stability in polar solvents such as water,
alcohols and THF. The 31P NMR spectrum displays one singlet
at d = 52.1 ppm corresponding to the two chemically equivalent
phosphorus nuclei. The 13C NMR spectrum shows one multiplet
12
Anorganisch-chemisches Institut, Universita¨t Zu¨rich, Winterthuerstrasse
190, 8057, Zu¨rich, Switzerland. E-mail: hberke@aci.unizh.ch.; Fax: +41
44 635 68 02; Tel: +41 44 635 46 81
† CCDC reference numbers 610298–610305. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/b702461d
‡ Electronic supplementary information (ESI) available: Complete struc-
tural plot of compound 6b and the calibration curve of the ISO NO
electrode. See DOI: 10.1039/b702461d
13
3562 | Dalton Trans., 2007, 3562–3571
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
at 59.3 ppm that corresponds to an AAꢀXXꢀ spin system and
is assigned to the CH2 groups. Furthermore the two nitrosyl
bands shift from 1810 and 1756 cm−1 for Fe(CO)2(NO)2 to 1711
and 1668 cm−1 for 2. Such lower values of m(NO) bands are in
accordance with data for related complexes bearing two phosphine
ligands.14 2 was recrystallized from water to obtain crystals suitable
for an X-ray diffraction study15 (Fig. 2).
crystallographic twofold axis. The THF solvate molecule lies on
a center of inversion (1.5 − x, 1.5 − y, −z) and was refined with
nine distance restraints.
The diamagnetic d6 compound Fe(NO)2HMPE 4 (HMPE =
1,2-bis[bis(hydroxymethyl)phosphino]ethane) was obtained from
the reaction between Fe(CO)2(NO)2 and one equivalent of
HMPE.17,18 4 is a red solid with higher solubility in water than 2
and 3. As for the previous complexes the nitrosyl bands are shifted
in the IR spectrum to lower frequencies (1711, 1664 cm−1). 4 could
be recrystallized from water allowing isolation of crystals suitable
for an X-ray diffraction study, which proved that the bisphosphine
HMPE binds to the metal center in a bidentate fashion (Fig. 4).
Fig. 2 Molecular structure of Fe(NO)2(P(CH2OH)3)2 2 (showing 30%
probability displacement ellipsoids). Hydrogen atoms have been omitted
for clarity.
The diamagnetic d6 Fe(NO)2(PTA)2 compound 3 (PTA = 1,3,5-
triaza-7-phosphatricyclo[3.3.1.1]decane) can be prepared from
Fe(NO)2(CO)2 with two equivalents of PTA.16 3 is a red solid and is
slightly water-soluble, but highly soluble in polar organic solvents.
The 31P NMR spectrum shows one singlet at d = −25.5 ppm
corresponding to two chemically equivalent phosphorus atoms.
In the IR spectrum of 3 we again observe a shift of the NO
bands to lower frequencies (1715, 1668 cm−1) in comparison with
the starting material Fe(CO)2(NO)2. 3 cocrystallized with THF
giving crystals suitable for an X-ray diffraction study17 (Fig. 3).
Compound 3 exhibits twofold symmetry with the Fe atom on a
Fig. 4 Molecular structure of Fe(NO)2HMPE 4 (showing 30% probabil-
ity displacement ellipsoids). Hydrogen atoms have been omitted for clarity.
Another bisphosphine which was explored to function as
a
water-soluble ligand in iron dinitrosyl complexes, was
1,2-bis[bis(hydroxymethyl)phosphino]benzene (TMBz). This bis-
phosphine has similar water-solubility and air stability as
HMPE. TMBz reacts with the dicarbonyl dinitrosyl iron com-
pound to afford the substitution product 5 in a relatively low yield
of 20%. As in compound 4 the TMBz ligand is assumed to bind
to the metal center in a bidentate fashion. In contrast to HMPE, 5
was not water-soluble and crystals suitable for an X-ray diffraction
study could not be isolated.
The ionic d7 dinuclear species [Fe(NO)2·S(CH2)2NH3]2Cl2 6a
and [Fe(NO)2·S(CH2)2NH3]2I2 6b were EPR silent, which pointed
to strong antiferromagnetically coupled iron centers.19 6a can
be obtained in THF by the reaction of Fe(CO)2(NO)2 with one
equivalent of cysteamine hydrochloride.20 6a is an ochre solid and
highly soluble in water. The IR spectrum of 6 exhibits NO bands
at 1794 and 1753 cm−1, whereas a band for a N–H vibration is
found at 1468 cm−1 corresponding in its position to an ammonium
derivative. Recrystallization of 6a gave crystals suitable for an X-
ray diffraction study (see ESI‡). Its structure is similar to that of
compound 6b. The iron–iron contact Fe(1)–Fe(1)#1 amounts to
˚
2.7154(11) A and is characteristic for a ligand supported metal–
metal single bond.19
Species [Fe(NO)2·S(CH2)2NH3]2I2 6b could be prepared in THF
21
by the reaction of the dinuclear complex [Fe(NO)2I]2 with two
equivalents of the cysteamine ligand. 6b is a brown solid and highly
soluble in water. The IR spectrum shows a shift of the NO bands
to lower frequencies (1782 and 1751 cm−1), and the m(NH) band
of the non-coordinated ammonium unit is found at 1464 cm−1. 6b
could be recrystallized from water to obtain crystals suitable for an
Fig. 3 Molecular structure of Fe(NO)2(PTA)2 3 (showing 50% probability
displacement ellipsoids). Hydrogen atoms have been omitted for clarity.
Atoms at the left-hand side with an “a” added to the labels are at equivalent
1
positions (1 − x, y, − z).
2
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3562–3571 | 3563
©

View Article Online
˚
The paramagnetic mononuclear [Fe(NO)2(pyrim)I] complex 10
is obtained by splitting of [Fe(NO)2I]2 with two equivalents of
pyrimidine-2-thiol. 10 is a black solid which is insoluble in water,
but highly soluble in polar organic solvents. The IR spectrum
exhibits two NO bands at 1790 and 1722 cm−1. 10 can be
recrystallized from THF–ether to give suitable crystals for a X-
ray diffraction study23 (Fig. 7). The protonation of atom S1 could
be verified by difference electron density calculations. Protonation
at N3 and N4 was not observed.
X-ray diffraction study (Fig. 5). The Fe(1)–Fe(1)#1 (2.7075(10) A)
bond length is very similar to that of 6a and again corresponds to
an iron–iron single bond.19
Fig. 5 Molecular structure of Fe(NO)2S(CH2)2NH2]I2 6b (showing 50%
probability displacement ellipsoids). Hydrogen atoms have been omitted
for clarity. One water solvate molecule (not shown) completes the structure.
The complete structure of 6b is given in the ESI‡ (Fig. 5).
Fig. 7 Molecular structure of [Fe(NO)2pyrimI] 10 (showing 50% prob-
ability displacement ellipsoids). Hydrogen atoms have been omitted for
clarity.
Complexes containing amino acid ligands could only be ob-
tained with L-cysteine ethyl ester hydrochloride. It was not possible
to substitute CO groups in Fe(NO)2(CO)2 with glycine or cysteine
or induce related reactions with the [Fe(NO)2X]2 complexes (X =
Cl, I). One equivalent of ethyl L-cysteine hydrochloride was thus
reacted with Fe(CO)2(NO)2 in THF at r.t. giving 8 in a yield of
50%. 8 is only slightly water-soluble. Crystals suitable for an X-ray
diffraction study could not be obtained, because 8 decomposed in
organic solvent after a few days.
In 1978 the first trinitrosyl iron compounds containing
trimethylphosphine and triphenylphosphine were synthesized
by M. Herberhold et al.24 These complexes are cationic nitrosyl
metal systems and readily release NO at r.t. We prepared the
−
trinitrosyl complexes 11, 12 and 13 possessing PF6−, BF4 and
−
SbF6 counter ions. These compounds were obtained via the
reaction of dicarbonyldinitrosyl iron and various nitrosyl salts25
(NOPF6, NOBF4 and NOSbF6) in the presence of one equivalent
of the water-soluble P(CH2OH)3 ligand. 11, 12 and 13 are deep
green solids highly soluble in water and polar organic solvents.
In the IR spectrum the m(NO) bands are found at 1925 and
1832 cm−1. They are shifted to higher frequencies in comparison
to the Fe(CO)2(NO)2 complex. 11 could be recrystallized from
CH3NO2–toluene to afford tiny crystals, which were still suitable
for an X-ray diffraction study. Due to the small size and irregular
shape of the crystal an absorption correction could not be carried
out. A split position has been observed for one OH group of the
P(CH2OH)3 ligand (Fig. 8). This is the first crystal structure of
such a trinitrosyl iron complex.
The dinuclear [Fe(NO)2pyrim]2 9 (pyrim = pyrimidine-2-thiol)
compound is obtained by mixing Fe(CO)2(NO)2 in THF with one
equivalent of the ligand pyrimidine-2-thiol.22 9 is a black solid
which is insoluble in water, but highly soluble in polar organic
solvents. The IR spectrum exhibits two NO bands at 1794 and
1759 cm−1, which are shifted to lower frequencies in comparison
with Fe(CO)2(NO)2. 9 was recrystallized from THF–ether to
obtain crystals suitable for an X-ray diffraction study (Fig. 6). 9 is
structurally related to 6a,b showing an iron–iron bond distance of
˚
2.7242(10) A. As in 6a,b there is strong antiferromagnetic coupling
of the iron centers.
Detection and identification of NO release: the EPR NO-trap
method
The EPR NO-trap assay has been widely used in the biomedical
area to detect the release of small amounts of NO in vitro. Addi-
tionally in 2000, Y. Xia et al. published a method to identify the
oxidation state of any NO species in water.26 They demonstrated
that a Fe(MGD)2 (N-methyl-D-glucamine dithiocarbamato iron)
can act as a NO trap and allows discrimination of NO from
NO− (nitroxonium) release depending on the redox state of
the Fe(MGD)2 complex used. Fe(II)(MGD)2 reacts selectively
with NO to give the paramagnetic (NO)Fe(II)(MGD)2 complex,
while the Fe(III)(MGD)2 species reacts only with NO− to form
the same paramagnetic complex. The Fe(MGD)2 mononitrosyl
complex gives rise to a characteristic three line EPR spectrum.
Fig. 6 Molecular structure of [Fe(NO)2pyrim]2 9 (showing 50% probabil-
ity displacement ellipsoids). Hydrogen atoms have been omitted for clarity.
Atoms labelled “a” denote inversion symmetry (−x, −y, −z + 1).
3564 | Dalton Trans., 2007, 3562–3571
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table 2 Detection of nitric oxide release from compounds 11, 12, 13, 6a
and 6ba by EPR spectroscopy
b
NO− donors
Fe(II)(MGD)2
Fe(III)(MGD)2
[Fe(NO)3P(CH2OH)3]PF6 11
[Fe(NO)3P(CH2OH)3]BF4 12
[Fe(NO)3P(CH2OH)3]SbF6 13
[Fe(NO)2S(CH2)2NH3)]2Cl2 6a
[Fe(NO)2S(CH2)2NH3]2I2 6b
Three line
Three line
Three line
Three line
Three line
Three line
Three line
Three line
Three line
Three line
a Solvent = water, T = 245–255 K. b Approximately half of the intensity
of Fe(II)(MGD)2.
Here both traps revealed a three line spectrum corresponding to
(NO)Fe(I)(MGD)2.26
Fig. 8 Molecular structure of [Fe(NO)3P(CH2OH3)]PF6 11 showing 10%
probability displacement ellipsoids. Hydrogen atoms have been omitted
for clarity. One (CH2OH) group was found to be disordered.
The difference between each of these pairs of spectra lies in the
intensity of the signal. For the ferric trap the intensity of the signal
is around 40–50% of the ferrous one.
Upon coordination of the nitric oxide (NO) or nitroxyl (NO−)
ligands to the Fe(II)–(MGD)2 or the Fe(III)–(MGD)2 complexes,
it is in both cases the +ON–Fe(I)–(MGD)2 ↔ NO-Fe(II)–(MGD)2
species that are observed by EPR. The mechanism of NO trapping
by the Fe(III)–(MGD)2 complex (reductive nitrosylation) is shown
below.
Table 1 Detection of nitroxyl release from compounds 1, 2 and 4a by
EPR spectroscopy
NO− donors
Fe(II)(MGD)2
Fe(III)(MGD)2
Fe(NO)2P(CH2OH)3Cl 1
Fe(NO)2(P(CH2OH)3)2 2
Fe(NO)2HMPE 4
—
—
—
Three line
Three line
Three line
First step: reduction of the Fe3+–MGD to the Fe2+–MGD
species. Second step: trapping of NO by the Fe2+–MGD molecule.
a Solvent = water, T = 245–255 K.
NO + (MGD)2-Fe3+ → (MGD)2-Fe3+-NO
ꢁ
We systematically carried out two measurements with each of the
water-soluble complexes. The Fe(II)(MGD)2 and Fe(III)(MGD)2
traps were prepared according to literature procedures.27 The EPR
measurements were carried out in frozen water at 245–255 K.
Fe(NO)2P(CH2OH)3Cl 1, Fe(NO)2(P(CH2OH)3)2 2, and
Fe(NO)2HMPE 4 complexes have thus been identified as nitroxyl
donors (Table 1). A three line spectrum was observed only for the
ferric form of the trap, which corresponded to the paramagnetic
(NO)Fe(I)(MGD)2 species. The signal possesses a g factor of
around 2.05, a center field of 3300 G and a hyperfine coupling
constant aN = 12.5 G (Fig. 9).
−
H
O,OH
2
2+
2+
-
(MGD)2-Fe -NO ←−−−−−− (MGD)2-Fe -NO+
2
(MGD)2-Fe2+ + NO → (MGD)2-Fe2+-NO
NO may also directly react with the Fe(II)–(MGD)2 trap.
Whereas the nitroxyl molecule (NO−) reacts with Fe(III)–
(MGD)2 as shown below:
2+
(MGD)2-Fe3+ + NO → (MGD)2-Fe -NO
-
These results are in accord with those obtained by Y. Xia
et al.26,28 One equivalent of NO is necessary to reduce the
Fe(III)(MGD) complexes to the Fe(II)(MGD) whereas the second
equivalent of NO released reacts with the ferrous trap to give the
paramagnetic (NO)Fe(I)(MGD)2 species.
Quantification of nitric oxide (NO) release by applying the
electrochemical method
The electrochemical method for nitric oxide quantification offers
several possibilities that are not available with spectroscopic
approaches. Perhaps the most important is the capability of
microelectrodes to directly sense NO in single cells in situ in close
proximity to the NO source. Electrochemical methods currently
available for NO detection are based on the electrochemical
oxidation of NO on solid electrodes.29 If the current generated
during this oxidation is linearly proportional to the concentration,
the anodic current can be taken as an analytical signal. The
calibration of the ISO-NO electrode is carried out with the
chemical generation of NO and is described elsewhere.30 Once the
Fig. 9 Three-line spectrum of the (NO)–Fe(III)–(MGD)2 species at
245–255 K with 50.10−4 M Fe(III)(MGD)2 and 10−4 M–NO donors.
On the other hand, the [Fe(NO)3P(CH2OH)3][X] compounds
(X = PF6 11, BF4 12, SbF6 13), the [Fe(NO)2SCH2CH2NH3]2Cl2 6a
and the [Fe(NO)2SCH2CH2NH3]2I2 6b species have been identified
as nitric oxide donors (Table 2). Again for each complex, measure-
ments with Fe(III)(MGD)2 and Fe(II)(MGD)2 were carried out.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3562–3571 | 3565
©

View Article Online
Table 4 Amount of NO released in water determined by electro-
sensor has been calibrated direct measurements of the NO released
from the NO-donors is possible at the calibration temperature.
chemistrya
Compound
Amount of NO/mM
Lifetime/s
Results. To carry out electrochemical NO measurements we
used an ISO-NOP 2.0 mm sensor. This electrode is available
from World Precision Instruments, USA. It was applied in the
amperometric mode with a permanent potential of 865 mV.
It was necessary to determine a calibration curve at the
temperature at which the measurements were to be carried out
before any electrochemical NO assay with the NO prodrug was
analysed. The calibration of the ISO-NO electrode takes place
with chemical generation of NO. It utilises the reaction between
a stock solution of a known concentration of sulfuric acid and
potassium iodide (0.01 M) and a stock solution of a NO-donor,
like sodium nitrite (50 × 10−7 M).
6a
1.49 × 10−5 (zone A)
1.37 × 10−5 (zone B)
1.21 × 10−5 (zone C)
0.87 × 10−5 (zone A)
1.62 × 10−5 (zone B)
0.63 × 10−5 (zone C)
200
—
—
200
—
—
6b
a Experiments were carried out at 37◦ C for compounds 6a (2.59 mM) and
6b (1.09 mM).
When compounds 11, 12 and 13 were dissolved in water, typical
oxidation curves were detected (Fig. 10). Single oxidation peaks
observed due to the release of only one NO or all NO molecules
simultaneously from these compounds during their decomposition
in water. The short lifetime, between 50–150 s, of these NO donors
mirror their low thermal stability in water.
The method of calibration is based on the following reaction:
2KNO2 + 2KI + 2H2SO4 → 2NO + I2 + 2K2SO4 + 2H2O
Since a known amount of KNO2 is added to produce a known
amount of NO, the quantity (and so the concentration) of NO
generated can be calculated directly from the stoichiometry, if the
concentration of the reactants are known. The potassium iodide
and sulfuric acid are present in great excess, potassium nitrite is the
limiting reagent generating NO for a period sufficient to calibrate
the ISO-NO sensor. Once the sensor has been calibrated, direct
measurements of NO from the NO-donors can be carried out.
Generally, the oxidation of NO on solid electrodes proceeds
with an “EC mechanism”: electrochemical reaction followed by a
chemical reaction.2 The first electrochemical step is a one-electron
transfer from the NO molecule to the electrode resulting in the
formation of a nitrosyl cation:
-
-e
NO −→ NO+
NO+ is a relatively strong Lewis acid, and in the presence of
OH− it is converted to the nitrite anion (NO2−):
Fig. 10 Decomposition curve of compounds [Fe(NO)3P(CH2OH)3]PF6
11, [Fe(NO)3P(CH2OH)3]BF4 12 and [Fe(NO)3P(CH2OH)3]SbF6 13 in
water at r.t.
NO+ + 2OH− → NO−2
The calibration curve can be found in the ESI‡.
Dissolution of compound 6a and 6b in water at 37 ◦C leads to
rapid decomposition of the latter accompanied by the release of
NO (Fig. 11 and 12). At least three oxidation steps are observed
here for each complex, which have been designated by zones A, B
and C. From Fig. 11 and 12 we can conclude that the lifetime of
the dinuclear species 6a and 6b is around 200 s.
Another main difference between the two classes of complexes
is the mode of NO release. According to the shape of the curves
it can be rationalized that the trinitrosyl complexes are able to
release either only one or all NO molecules at once, whereas the
dinuclear species 6a and 6b are able to release NO in a stepwise
fashion and practically all of the nitric oxide molecules present in
the complex.
Firstly, electrochemical measurements were carried out with
nitric oxide donors that have been identified by the EPR-trap
method i.e. the series of the trinitrosyl species 11, 12, and 13 and
the sulfur containing complexes 6a and 6b.
For the trinitrosyl series the electrochemical quantification of
the NO release had to be carried out at room temperature, because
of high instability of the compounds at elevated temperature,
whereas for the compounds 6a and 6b the measurements were
carried out at 37◦ C. The results are shown in Tables 3 and 4.
Table 3 Amount of NO released in water determined by electro-
chemical methodsa
Compound
Amount of NO/mM
Lifetime/s
We have noticed that the amount of NO released from these
NO prodrugs is less than the expected quantities. It can be
rationalized that for the trinitrosyl series (11, 12 and 13) a very
fast decomposition takes place generating only one peak implying
that the release of all NO ligands occurs simultaneously, and that
only part of the total amount of NO is detected. It is also known
that NO can react at higher concentration with dissolved oxygen
11
12
13
2 × 10−3
7.57 × 10−3
1.53 × 10−2
75
50
150
a Experiments were carried out at r.t. for compound 11 (0.086 mM), 12
(0.17 mM), 13 (0.45 mM).
3566 | Dalton Trans., 2007, 3562–3571
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
the equilibrium of NO release from these NO prodrugs can be
described by the rate constants k1 and k−1 of the pre-equilibrium.
k
1
−−−→
←−−−
{Fe(NO)2} + H2O
{Fe(NO)} + NO
(1)
k
¬1
The pseudo-first-order condition requires that the concentra-
tion of the reactant (here the NO prodrug) is at least 10 times
smaller than that of all the other reactants (here the water).
Plotting ln(A0–At) versus time gave in all cases a straight line.
Kinetics of the NO release of the nitric oxide donors. The high
instability of the trinitrosyl series compounds 11, 12 and 13 in
water allowed kinetic studies31 only at room temperature in a
phosphate buffer. These compounds exhibit bands in the UV-
vis region at around 595 nm. 6a and 6b were also studied under
the same pseudo-physiological conditions. The kinetics of the
decomposition were established by monitoring the disappearance
of the UV-vis band at 762 nm over time.32
Fig. 11 Decomposition curve of [Fe(NO)2SCH2CH2NH3]2I2 6a in water
at 37 ◦C.
Based on a life time of approximately 100 s for 11, 12 and 13, and
300 s for 6a and 6b, the kinetic measurements of these compounds
are not in accord with the results obtained by the electrochemical
pathway. This can presumably be attributed to the fact that the
disappearance of the UV-vis bands did not correspond to a process
releasing NO, but to another, yet unidentified decomposition
reaction. Therefore only the kinetic experiments of the nitroxyl
donors can be taken with confidence in this section.
Kinetics of the NO− release of the nitroxyl donors
Fe(NO)2P(CH2OH)3Cl 1 starts to release NO after an induction
period of 10 min. The decomposition process is completed after
100 min. This process was followed by a decrease in the intensity
of the band of 1 at around 520 nm. As demonstrated earlier in this
paper this indicates the release of nitroxyl. A plot of the intensity
of the absorption band at 520 nm permits the overall kinetics
according to the ln(A0–At) = f (t) law to be established. The release
of nitroxyl from 1 is slow and steady, while 2 decomposes in an
incoherent way. It might be that the kinetics of the decomposition
of 2 are more complex than the pseudo-first law would suggest. As
a consequence, it was not possible to determine the rate constant
k1 of NO release for compound 2. Finally Fe(NO)2HMPE 4 com-
pound was studied under the same conditions. The latter complex
possesses a functionalised bisphosphine with high stability under
pseudo-physiological conditions. 4 is completely stable even during
an induction period of 260 min. After this it starts to decompose
within 20 min (Table 5).
Fig. 12 Decomposition curve of [Fe(NO)2SCH2CH2NH3]2I2 6b in water
at 37 ◦C.
and water to give NOx compounds as decomposition products,
which can not be detected by the electrode.
For 6a and 6b, each step should correspond to the release of
one NO molecule. The small amount of NO detected by the ISO-
NO electrode could be explained by assuming that each free and
detectable NO molecule is leaving the coordination sphere by
replacement with a water molecule to give the hydrated species.
However, if attack of a water molecule onto the still coordinated
nitric oxide molecule occurs, other undetectable species such as
HNO2 are produced.
Conclusions
NO release kinetics: determination of the pseudo-first order rate
constant of NO release from the NO donors using UV-vis
spectroscopy
Water-soluble iron polynitrosyl complexes were prepared and
tested for their potential application as NO donors. This became
possible via their modification with water-soluble ligands such
The most important factors of any NO prodrug are the equilibrium
release and trapping amount of NO. The complexes to be studied
were expected to possess bands in the UV-vis absorption region;
it is then possible to determine the overall rate constant of
NO release. The measurements were carried out under pseudo-
physiological conditions (phosphate buffer pH = 7.4 and T =
37 ◦C) and concentrations in the millimolar range. The overall
rate constant k1 of the different NO donors was determined
using pseudo-first-order conditions (eqn (1)). We assumed that
Table 5 Values of rate constant k1 of NO− release for the nitroxyl donors
1 and 4 in phosphate buffer
Compounds
k1/s−1
[Fe(NO)2P(CH2OH)3Cl]
[Fe(NO)2HMPE]
3.71 × 10−4
1.35 × 10−5
a Experiments were performed at 37◦ C.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3562–3571 | 3567
©

View Article Online
as functionalised phosphine, bisphosphine and sulfur containing
molecules. Depending on the nature of the ligand, nitric oxide
or nitroxyl donors were generated. Both compounds may be
applicable for further in vitro and in vivo studies, because it
has recently been suggested that nitric oxide and nitroxyl are
both suitable substrates for the NOS (nitric oxide systems).
The biophysical measurements using EPR and electrochemistry
allowed us to identify and quantify the NO release from these
prodrugs. Kinetic measurements carried out under pseudo-first
order conditions allowed us to determine the pseudo-first order
rate constant k1 of the NO release under pseudo-physiological
conditions. From all these experiments a step-wise mechanism of
NO release is suggested.
toluene was added the tris(hydroxymethyl) phosphine (0.080 g,
0.645 mmol) dissolved in 6 ml of water. The mixture was stirred
for 5 min and filtered off. The toluene phase was removed and
the aqueous phase was evaporated under vacuum to give a red
oil. This oil was washed with THF and recrystallized from ether.
Yield: 70 mg (77%). IR (methanol): m/cm−1 = 1720 (m, NO); 1678
1
(s, NO). H NMR (water-d2): d 4.17 (s, 6H, CH2OH). 13C NMR
(water-d2): d 52.5 (d, 1J = 75.8 Hz, CH2OH). 31P NMR (water-d2):
d 59.5 (s, PCH2OH). Conductivity (H2O): 195.56 S cm2 mol−1.
Dinitrosylbis(tris(hydroxymethyl)phosphine)iron(II) 2. To
a
solution of dicarbonyl dinitrosyliron (0.100 g, 0.582 mmol) in
40 ml of THF was added tris(hydroxymethyl)phosphine (0.130 g,
1.056 mmol). The solution was refluxed for 20 h. The red solution
was filtered off and the solvent was removed under vacuum.
Recrystallization from THF–ether gave 2. Yield: 105 mg (50%).
Found: C, 20.06; H, 4.71; N, 7.32%. C6H18FeN2O8P2 required: C,
19.80; H, 4.94; N, 7.69%. IR (THF): m/cm−1 = 1711 (m, NO); 1668
(s, NO). 1H NMR (THF-d8): d 3.58 (s, 6H, CH2); 4.56 (s, 3 H, OH).
31P NMR (THF-d8): d 52.1 (s, 2P). 13C NMR (THF-d8): d 59.3 (m,
3C, CH2OH).
Experimental
Material and methods
All reactions were carried out in an inert atmosphere using
standard Schlenk-tube techniques or a dry-box (M. Braun 150 B-
G). The solvents were dried using standard methods15 and stored
under nitrogen prior to use. The deuterated solvents were also
dried using standard methods and degased using the “pump–
thaw” method.
Dinitrosylbis(1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane)iron-
(II) 3.
A solution of dicarbonyldinitrosyliron (0.100 g,
NMR spectra were recorded as follows: 1H NMR: Bruker
Avance DRX500, frequency 500.2 MHz and Varian Gemini 300,
0.582 mmol) and 1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane
(0.165 g, 0.105 mmol) in THF (40 ml) was refluxed for 16 h.
The red solution was filtered off and the solvent removed under
vacuum. Recrystallization from CH2Cl2–ether gave 3 as a light
red compound. Yield: 75 mg (30%). Found: C, 32.24; H, 5.85.
1
frequency 300.1 Hz; 13C{ H} NMR: Bruker Avance DRX500,
frequency 125.8 MHz and Varian Gemini 300, frequency 75.46 Hz;
19F NMR: Varian Gemini 300, frequency, 282.3 MHz; 31P NMR:
Varian Gemini 300, frequency 121.5 MHz. Chemical shifts (d)
are given in ppm. If not otherwise mentioned, the spectra were
C12H24FeN2O8P2 required: C, 32.52; H, 5.58. IR (THF): m/cm−1
=
1
1715 (m, NO); 1668 (s, NO). H NMR (THF-d8): d 4.05 (s, 6H,
PCH2N); 4.55 (s, 6H, NCH2N). 13C NMR (THF-d8): d 56.1 (m,
1J = 7 Hz, 3C, CH2P); 73.6 (s, 3C, CH2N). 31P NMR (THF-d8): d
−25.5 (s).
recorded at room temperature. The references for the 1H and 13
C
NMR spectra were the signals of the deuterated solvents (internal
standard). For 19F NMR trifluorotoluene, and for 31P NMR
spectra H3PO4 were used as an external standards. Some NMR
spectra were also recorded using two-dimensional techniques (13C–
1H). IR spectra were recorded on a Bio-Rad FTS-45 Fourier
IR spectrometer and low temperature IR spectra were measured
using a Bio-Rad FTS-3500 Fourier IR spectrometer. UV-vis
spectra were recorded on a Cary 500 scan UV-vis spectrometer.
EPR spectra were recorded on a Bruker EMX-081 spectrometer
between 245–255 K. EPR spectrometer settings were as follows,
incident microwave power, 2 mW, modulation frequency, 100 kHz,
frequency of 9.46 GHz, modulation amplitude, 10 G, and scan
conditions (time constant, 2048 ms, sweep width, 700 G, sweep
time, 167, 77 s); each spectrum was the result of signal of one scan.
Conductivity was recorded with an Amel-160 conductimeter with
Pt-cells. CHN-analyses were carried out using Leco CHNS-932
analyser.
The iron salts (FeSO4·7H2O, FeSO4(NH4)2SO4·6H2O, FeCl3;
FeCl2; FeBr2; Fe) were purchased from Fluka or Aldrich and used
without further purification. The nitrosylating agents [NO][X]
with X = BF4, PF6, SbF6 were purchased from ABCR or Aldrich
and used without further purification. NO gas (quality 2, Pangas,
Suisse), for each experiment, was purified using a cold trap
(pentane, liquid nitrogen, −90 ◦C) to remove all the impurities
(N2O4, N2O3, NO2).
Dinitrosyl(1,2-bis(dihydroxymethylphosphine)ethane)iron(II) 4.
A solution of dicarbonyldinitrosyliron (0.100 g, 0.582 mmol)
and HMPE (1,2-bis(dihydroxymethylphosphino)ethane) 0.125 g,
0.528 mmol) in 40 ml of THF was refluxed for 2 d. Removal of
the solvent and recrystallization in THF–ether gave 4 as a red
crystalline compound. Yield: 230 mg (74%). Found: C, 22.18; H,
4.86; N, 8.26%. C12H16FeN2O6 required: C, 21.85; H, 4.85; N,
1
8.49%. IR (THF): m/cm−1 = 1711 (m, NO); 1664 (s, NO). H
2
NMR (THF-d8): d 1.98 (d, J = 16.5 Hz, 4H, PCH2CH2P); 4.0
(m, 8H, CH2OH); 4.36 (m, 4H, OH). 31P NMR (THF-d8): d =
2
82.6 (s, 2P, PCH2OH), 13C NMR (THF-d8): d 19.4 (dd, 1J = J =
20 Hz, 2C, PCH2CH2P); 60.3 (m, 4C, CH2OH).
Dinitrosyl(1,2-bis(dihydroxymethyl phosphine)benzene)iron(II) 5.
1,2-bis(dihydroxymethylphosphino)benzene (TMBz) (0.350 g,
1.63 mmol) was added to a solution of dicarbonyldinitrosyliron
(0.280 g, 1.63 mmol) in 40 ml of THF. The mixture was refluxed
during 40 h and then filtered off. Removal of the solvent under
vacuum gave 5 as a brown powder. Yield: 246 mg (20%). Found:
C, 32.17; H, 4.27; N, 7.15%. C10H16FeN2OP2 required: C, 31.78;
H, 4.23; N, 7.41%. IR (THF): m/cm−1 = 1715 (m, NO); 1666 (s,
NO). 1H NMR (THF-d8): d 4.20 (m, 12H, PCH2OH); 7.5 (m, 4H,
ArH). 31P NMR (THF-d8): d 76.0 (s, 2P). 13C NMR (THF-d8): d
61.5 (m, 4 C, PCH2OH); 131.0 (s, 2C, ArC3,4); 133.9 (t, 2J = 13 Hz,
2C, ArC2,5); 145.0 (t, 1J = 38 Hz, 2C, ArC1,6).
Dinitrosyl(tris(hydroxymethyl)phosphine)chloroiron(I) 1. To a
solution of the dimer [Fe(NO)2Cl]2 (0.100 g, 0.330 mmol) in 32 ml
3568 | Dalton Trans., 2007, 3562–3571
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table 6 Crystal data for complexes 2, 3 and 4†
2
3
4
Formula
C6H18FeN2O8P2
364.01
Red, block
C16H32FeN8O3P2
502.29
Red, plate
0.26 × 0.17 × 0.10
Monoclinic
C2/c
19.4660(9)
6.5409(4)
19.0131(9)
90
115.104(5)
90
C6H16FeN2O6P2
330
Red, block
0.33 × 0.33 × 0.32
Triclinic
¯
P1
7.6992(10)
8.5741(11)
11.2515(15)
95.919(15)
106.994(15)
109.827(14)
651.11(18)
2
1.683
183
0.71073
2.87–30.39
13 195
Molecular weight/g mol−1
Crystal description
Crystal size/mm
Crystal system
0.28 × 0.25 × 0.18
Triclinic
¯
P1
Space group
˚
a/A
6.8833(7)
9.4457(9)
11.4636(10)
89.368(11)
72.801(11)
77.584(11)
694.25(13)
2
1.741
153
0.71073
2.97–30.27
11 490
˚
b/A
˚
c/A
a/◦
b/◦
c /◦
˚
V/A
Z
2192.2(2)
4
1.522
Dc/g cm−3
T/K
153
˚
k/A
0.71073
3.32–30.28
12 492
3253
1056
0.0458
0.1215
2151
h range/◦
Total no. of data
No. of unique data
F(000)
3765
376
0.0496
0.141
3544
340
0.0268
0.0716
Rab
ac
wR2
Observed dataa
3064
3022
Max., min. transmission
Data/restraints/parameters
Goodness-of-fit on F2
0.793, 0.704
3765/0/178
1.091
2.13, −1.30
0.918, 0.805
3253/9/131
0.912
0.892, −0.728
0.720, 0.601
3544/0/164
1.078
0.558,−0.694
−3
˚
Max., min. peaks/e A
a Observation criteria I > 2r(I). b R = RꢂFo| − |Foꢂ/RꢂFo|. c wR2 = {R[w(Fo − Fc2)2]/R[w(Fo2)2]}
.
2
1/2
Table 7 Crystal data for complexes 6b, 9, 10 and 11†
6b
9
10
11
Formula
C4H14Fe2I2N6O5S2
659.88
Red, plate
C8H8Fe2N8O4S2
465.06
Red opaque, block
0.31 × 0.24 × 0.19
Triclinic
¯
P1
6.4219(13)
7.626(2)
8.345(2)
75.61(3)
80.83(3)
85.15(3)
390.37(17)
1
1.940
183
0.71073
3.28–30.33
6823
C8H14FeIN4O3S
429.05
Red opaque, block
0.26 × 0.14 × 0.10
Monoclinic
P21/n
7.0648(4)
11.4932(7)
18.8432(10)
90
91.237(7)
90
1529.66(15)
4
1.863
153
0.71073
C3H9F6FeN3O6 P2
414.92
Green, tiny plate
(Irregular shape)
Orthorhombic
Pna21
24.693(3)
6.6317(10)
8.2556(16)
90
90
90
1351.9(4)
4
2.039
Molecular weight/g mol−1
Description
Crystal size/mm
Crystal system
Space group
0.32 × 0.20 × 0.09
Triclinic
¯
P1
˚
a/A
6.3097(5)
6.8752(5)
24.5366(17)
90.403(8)
93.515(8)
116.319(8)
951.58(14)
2
2.303
153
0.71073
2.50–28.02
10 041
˚
b/A
˚
c/A
a/◦
b/◦
c /◦
3
˚
V/A
Z
Dc/g cm−3
T/K
183
˚
k/A
0.71073
3.18–30.52
6204
2046
824
0.0949
0.2201
1023
h range/◦
2.80–30.31
17 573
4499
836
0.0479
0.1252
2709
Total no. of data
No. of unique data
F(000)
4195
628
0.0283
0.0767
2102
228
0.0319
0.1066
Rab
ac
wR2
Observed dataa
3491
1752
Max., min. transmission
Data/restraints/parameters
Goodness-of-fit on F2
0.661, 0.297
4195/0/194
1.009
1.5, −1.582
0.685, 0.554
2102/0/109
1.090
0.550, −1.091
0.744, 0.495
4499/0/166
1.005
2.080, −1.752
—, —
2046/23/199
1.234
0.961, −0.990
−3
˚
Max., min. peaks/e A
2
1/2
a Observation criteria I > 2r(I). b R = RꢂF0| − |Fcꢂ/RꢂF0|. c wR2 = {R[w(Fo − Fc2)2]/R[w(F02)2]}
.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3562–3571 | 3569
©

View Article Online
Di-l-cysteaminehydrochloridetetranitrosyldiiron(I) 6a. To
a
The solution changed from red to green. Then a solution of
trishydroxymethylphosphine (0.036 g, 0.290 mmol) in 3 ml of
nitromethane was added dropwise. To precipitate 11, a large
portion (20 ml) of cold toluene was added. The green product
was filtered off and dried under vacuum. Due to the instability of
11 at room temperature a correct elemental analysis could not be
obtained. Yield: 85 mg (63%). IR (nitromethane, −10 ◦C, CaF2):
m/cm−1 = 1927 (w, NO); 1833 (vs, NO). 1H NMR (nitromethane-
d3): d 4.73 (s, 6H, CH2OH). 13C NMR (nitromethane-d3): d 55.0
solution of dicarbonyldinitrosyliron (0.100 g, 0.582 mmol) in
30 ml of THF was added cysteamine hydrochloride (0.066 g,
0.582 mmol). The mixture was refluxed overnight and the pre-
cipitate was filtered off and dried under vacuum to give 6 as an
ochre precipitate. Yield: 11 mg (40%). Found: C, 13.14; H, 3.02; N,
17.05%. C4H14Cl2Fe2N6O4S2, S(CH2)2NH3Cl required: C, 12.66;
H, 3.68; N, 17.21%. IR (Nujol): m/cm−1 = 1754 (m, NO); 1753
1
(s, NO). H NMR (water-d2): d 1.77 (s, 2H, CH2S); 3.4 (m, 2H,
CH2NH3Cl). 13C NMR (water-d2): d 22.8 (s, 2H, SCH2); 65.7 (s,
(d, J = 50 Hz, CH2OH). 19F NMR (nitromethane-d3): d −72.0
1
2H, CH2NH3Cl).
(d, 1J = 714 Hz, 6F, PF6). 31P NMR (nitromethane-d3): d −144,0
(hept., 1J = 715 Hz, PF6); 46,0 (s, P(CH2OH)3).
Di-l-cysteaminehydroiodidetetranitrosyldiiron(I) 6b. A mix-
ture of di-l-iodotetranitrosyldiiron [Fe(NO)2I]2 (0.100 g,
0.205 mmol) and cysteine (0.016 g, 0.205 mmol) in 20 ml of THF
was stirred at room temperature for 18 h. The brown precipitate
was filtered off and washed several times with ether and dried
under vacuum to give 7. Yield: 124 mg (95%). Found: C, 12.63; H,
2.97; N, 12.06%. C4H14I2Fe2N6O4S2, required: C, 13.53; H, 2.81;
N, 11.83%. IR (Nujol): m/cm−1 = 1782 (s, NO); 1715 (s, NO);
Trinitrosyl(tris(hydroxymethyl)phosphine)iron(II) tetrafluorobo-
rate 12. The same procedure as for 11 was used to synthesise the
trinitrosyl(trishydroxymethylphosphine)iron(II) tetrafluoroborate
12, and the hexafluoroantimonate 13, using nitrosoniumtetraflu-
oroborate for 12 and nitrosoniumhexafluoroantimonate for 13
as nitrosylating agents. Due to the instability of 12 at room
temperature a correct elemental analysis could not be obtained.
1464 (m, NH3). 1H NMR (water-d2): d 3.3 (m, 4H, CH2CH2). 13
C
Yield: 74 mg (63%). IR (nitromethane, −10 ◦C, CaF2): m/cm−1
=
NMR (water-d2): d 37.9 (d, 1J = 8 Hz, SCH2); 39.5 (d, 1J = 4 Hz,
1927 (w, NO); 1835 (vs, NO). 1H NMR (nitromethane-d3): d 4.79 (s,
6H, CH2OH). 13C NMR (nitromethane-d3): d 57.0 (sb, CH2OH).
31P NMR (nitromethane-d3): d 47.0 (s, P(CH2OH)3). 19F NMR
(nitromethane-d3): d −144.0 (s, BF4).
CH2NH2).
Di-l-cysteineethanoatetetranitrosyldiiron(I) 8. To a solution of
dicarbonyldinitrosyliron (0.100 g, 0.582 mmol) in 40 ml of THF
the L-cysteine ethyl ester hydrochloride (0.307 g, 0.582 mmol)
was added. The mixture was stirred at room temperature for 3
d. The solution was evaporated under vacuum to give 8 as a deep
green product. Yield: 181 mg (81%). Found: C, 23.00; H, 4.02.
Trinitrosyl(tris(hydroxymethyl)phosphine)iron(II) hexafluoroan-
timonate 13. Due to the instability of 13 at room temperature
a correct elemental analysis could not be obtained. Yield: 34 mg
(21%). IR (nitromethane, −10 ◦C, CaF2): m/cm−1 = 1927 (w, NO);
1835 (s, NO). 1H NMR (nitromethane-d3): d 4.82 (s, 6H, CH2OH).
31P NMR (nitromethane-d3): d 46.0 (s, P(CH2OH)3). 13C NMR
(nitromethane-d3): d 57.5.
C10H20Fe2N6O8S2 required: C, 22.66; H, 4.15. IR (Nujol): m/cm−1
=
1755 (m, NO); 1711 (s, CO(O)Et); 1595 (w, NH2). 1H NMR (THF-
d8): d 0.88 (m, 2H, CH2CH3); 1.35 (m, 3H, CH2CH3); 4.39 (m, 3H,
CH2CH). 13C NMR (THF-d8): d 14.2 (s, S–CH2); 23.1 (s, NH2CH);
26.3 (s, OCH2CH3); 34.9 (s, OCH2CH3); 64.1 (s, CO(OCH2CH3).
Crystal data are given in Tables 6 and 7.
CCDC reference numbers 610298–610305.
Di-l-(pyrimidine-2-thionate)tetranitrosyldiiron(I) 9. To a solu-
tion of dicarbonyldinitrosyliron (0.100 g, 0.582 mmol) in 40 ml
of THF was added pyrimidine-2-thiol (0.065 g, 0.582 mmol). The
solution is refluxed for 16 h. Removal of the solvent under vacuum
gave 9 as a black solid. Yield: 132 mg (50%). Found: C, 21.36; H,
1.40; N, 24.40%. C8H6Fe2N8O4S2 requires: C, 21.07; H, 1.75; N,
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b702461d
References
1 A. R. F. Furchgott and J. V. Zawadzki, Nature, 1980, 288, 373–376;
S. Moncada, R. M. J. Palmer and E. A. Higgs, Pharm. Rev., 1991,
43, 109–141; R. F. Furchgott, Vasodilation: Vasular Smooth Muscle,
Peptides, and Endothelium, ed. P. M. Vanhoutte, Raven Press, New
York, 1988, pp. 401–414; S. Moncada, R. M. J. Palmer and E. A.
Higgs, Pharm. Rev., 1991, 43, 109–141; N. Miki, Y. Kawabe and K.
Kuriyama, Biochem. Biophys. Res. Commun., 1977, 75(4), 851–856.
2 R. F. Furchgott and J. V. Zawadzki, Nature, 1980, 288, 373–376.
3 F. Murad, Angew. Chem., Int. Ed., 1999, 38, 1856–1868; R. F.
Furchgott, Angew. Chem., Int. Ed., 1999, 38, 1870–1880; L.-J. Ignarro,
Angew. Chem., Int. Ed., 1999, 38, 1882–1892.
4 A. R. Butler and C. Glidewell, Chem. Soc. Rev., 1987, 16, 361–380;
E. A. Kowaluk, P. Seth and H.-L. Fung, J. Pharmacol. Exp. Ther.,
1992, 262(3), 916–922.
5 P. G. Wang, M. Xian, X. Tang, X. Wu, Z. Wen, T. Cai and A. J. Janczuk,
Chem. Rev., 2002, 102, 1091–1134.
1
24.57%. IR (THF): m/cm−1 = 1794 (m, NO); 1759 (s, NO). H
NMR (THF-d8): d 7.14 (s, 1H, CHpyr.); 8.55 (s, 2H, 2CNCHpyr.).
13C NMR (THF-d8): d 121.0 (s, Cpyr.HCH); 158.5 (s, 2CpyrN); 174.0
(s, Cpyr.S).
Dinitrosyl(pyrimidine-2-thionate)iodoiron(0) 10. A solution of
[Fe(NO)2I]2 (0.100 g, 0.205 mmol) and pyrimidine-2-thiol (0.023 g,
0.205 mmol) in 40 ml of THF was stirred for 3 h at room
temperature. Recrystallization from THF–ether gave 10 as a black
solid. Yield: 79 mg (67%). Found: C, 18.44; H, 2.10; N, 14.32%.
C8H6IFe2N8O4S2 required: C, 18.39; H, 2.29; N, 14.29%. IR
(THF): m/cm−1 = 1790 (m, NO); 1722 (s, NO). 1H NMR (THF-
d8): d 3.4 (m, 1H, Hpyr.CH); 3.58 (s, 2H, Hpyr.CHN). 13C NMR
(THF-d8): d 15.6 (s, 2C, CHN); 66.3 (s, 2C, CH2; C–S).
6 D. J. Stuehr, S. S. Gross, I. Sakuma, R. Levi and C. F. Nathan, J. Exp.
Med., 1989, 169, 1011–1020; Pellat, Y. Henry and J.-C. Drapier, Nitric
oxide from L-arginine: a bioregulatory system, ed. S. Moncada and
E. A. Higgs, Elsevier, Amsterdam, 1990, pp. 281–289.
7 M. E. Murphy and H. Sies, Proc. Natl. Acad. Sci. USA, 1991, 88,
10860–10864.
Trinitrosyl(tris(hydroxymethyl)phosphine)iron(II)
hexafluo-
rophosphate 11. To
a solution of dicarbonyldinitrosyliron
(0.050 g, 0.290 mmol) in 3 ml of nitromethane at −30◦
C
8 K. Shibuki, Neurosci. Res., 1990, 9, 69–76; T. Malinski and Z. Taha,
Nature, 1992, 358, 676–8.
was added NOPF6 (0.050 g, 0.290 mmol) in small portions.
3570 | Dalton Trans., 2007, 3562–3571
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
9 F. L. Atkinson, H. E. Blackwell, N. C. Brown, N. G. Connelly, J. G.
Crossley, A. G. Orpen, A. L. Rieger and P. H. Rieger, J. Chem. Soc.,
Dalton Trans., 1996, 3491–3502.
10 D. Ballivet-Tkatchenko and C. Billard, J. Polym. Sci., Polym. Chem.
Ed., 1981, 19, 1697–1706.
11 J. W. Ellis, K. N. Harrison, P. A. T. Hoye, A. G. Orpen, P. G. Pringle
and M. B. Smith, Inorg. Chem., 1992, 31, 3026–3033.
12 D. W. McBride, S. L. Stafford and G. A. Stone, Inorg. Chem., 1962,
1(2), 386–388.
13 L. Malatesta and A. Araneo, J. Chem. Soc., 1957, 3803–3805; V. G.
Albano, A. Araneo, P. L. Bellon, G. Ciani and M. Manassero, J. Org.
Chem., 1974, 67, 413–422.
14 J. T. Lin, S. Y. Wang, Y. C. Chou, M. L. Gong, Y.-M. Shiow, H.-M.
Gau and Y. S. Wen, J. Org. Chem., 1996, 508, 183–193.
15 J. W. Ellis, K. N. Harrison, P. A. T. Hoye, A. G. Orpen, P. G. Pringle
and M. B. Smith, Inorg. Chem., 1992, 31, 3026–3033; V. G. Albano, A.
Araneo, P. L. Bellon, G. Ciani and M. Manassero, J. Org. Chem., 1974,
67, 413–422.
21 H. Soling and R. W. Asmussen, Acta Chem. Scand., 1957, 11, 1534–
1540; B. Haymore and R. D. Feltham, Inorg. Synth., 1973, XIV,
81–92; T. Moeller and D. H. Wilkins, Inorg. Synth., 1953, IV, 101–
103; S. Y. Tyree, Inorg. Synth., 1953, IV, 104–111; W. Hieber and R.
Marin, Z. Anorg. Allg. Chem., 1939, 240, 241–260; W. Hieber and J. S.
Anderson, Z. Anorg. Allg. Chem., 1933, 211, 132–140.
22 K. Yamanari, K. Okusako and S. Kaizaki, J. Chem. Soc., Dalton Trans.,
1992, 1615–1619; K. Yamanari, K. Okusako, Y. Kushi and S. Kaizaki,
J. Chem. Soc., Dalton Trans., 1992, 1621–1626; D. L. Geffner, M.
Azukizawa and J. M. Hershman, J. Clin. Invest., 1975, 75, 224–229.
23 W. Hieber and K. Kaiser, Z. Anorg. Allg. Chem., 1968, 358, 271–281. 49;
V. M. Kothari and D. H. Busch, Inorg. Chem., 1969, 8(11), 2276–2280;
C. P. Sloan and J. H. Krueger, Inorg. Chem., 1975, 14(7), 1481–1485;
H. C. Freeman, C. J. Moore, W. G. Jackson and A. M. Sargeson, Inorg.
Chem., 1978, 17(12), 3513–3521.
24 F. Seel, Z. Anorg. Allg. Chem., 1942, 249, 309–324.
25 M. T. Mocella, M. S. Okamoto and E. K. Barefield, Synth. React. Inorg.
Met.-Org. Chem., 1974, 4(1), 69–90.
16 D. J. Daigle, T. J. Decuir, J. B. Robertson and D. J. Darens-
bourg, Ligands for Water-Solubilizing Org. Com., Inorg. Synth., ed.
M. Y. Darensbourg, John Wiley & Sons. Inc., New York, 1998, 40–44;
M. Y. Darensbourg and D. Daigle, Inorg. Chem., 1975, 14(5), 1217–
1218.
17 V. G. Albano, A. Araneo, P. L. Bellon, G. Ciani and M. Manassero,
J. Org. Chem., 1974, 67, 413–422.
26 M. N. Hughes and R. Camack, Methods Enzymol., 1999, 301, 279–281;
Y. Xia, A. J. Cardounel, A. F. Vanin and J. L. Zweier, Free Radical Biol.
Med., 2000, 29(8), 793–797.
27 L. A. Shinobu, S. G. Jones and M. M. Jones, Acta Pharmacol. Toxicol.,
1984, 54, 189–194.
28 M. N. Hughes and R. Camack, Methods Enzymol., 1999, 301, 279–
281.
18 V. S. Reddy, D. E. Berning, K V. Katti, C. L. Barnes, W. Y. Volkert and
A. R. Ketring, Inorg. Chem., 1996, 35, 1753–1757.
29 M. Feelisch and J. S. Stamler, Methods in Nitric Oxide Research, John
Wiley & Sons Inc., New York, 1996.
19 (a) R. E. Rundle, J. Phys. Chem., 1957, 61, 45–50; R. V. G. Evens,
Nature, 1948, 161, 530–531; P. S. Braterman, V. A. Wilson and K. K.
Joshi, J. Chem. Soc. A, 1971, 191–195; (b) W. E. Douglas and M. L. H.
Green, J. Chem. Soc., Dalton Trans., 1972, 1796–1800.
20 K. A. Jensen, Z. Anorg. Allg. Chem., 1936, 229, 265–281; D. C. Jicha
and D. H. Busch, Inorg. Chem., 1962, 1(4), 872–877; D. C. Jicha and
D. H. Busch, Inorg. Chem., 1962, 1(4), 878–883; D. H. Busch and D. C.
Jicha, Inorg. Chem., 1962, 1(4), 884–886; T. Yonemura, Z.-P. Bai, K.-I.
Okamoto, T. Ama, H. Kawaguchi, T. Yasui and J. Hidaka, J. Chem.
Soc., Dalton Trans., 1999, 2151–2157.
30 ISO-NO Mark II Instruction Manual, World Precision Instruments,
Inc., Sarasota, FL, USA, 2002.
31 T. Yonemura, S. Nakahira, T. Ama, H. Kawaguchi, T. Yasui, K.-I.
Okamoto and J. Hidaka, Bull. Chem. Soc. Jpn., 1995, 68, 2859–2866;
S.-I. Aizawa, K.-I. Okamoto, H. Einaga and J. Hidaka, Bull. Chem. Soc.
Jpn., 1988, 61, 1601–1606; K.-I. Okamoto, T. Yonemura, T. Konno and
J. Hidaka, Bull. Chem. Soc. Jpn., 1992, 65, 794–798.
32 R. B. Jordan, Reaction Mechanism of Inorg. and Org. Systems,
Oxford University Press, Oxford, 1998; H.-H. Perkampus, UV-Vis-
Spektroscopie und ihre Anwendungen, Springler-Verlag, Berlin, 1986.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3562–3571 | 3571
©