New modular manganese(i) tricarbonyl complexes as PhotoCORMs: In vitro detection of photoinduced carbon monoxide release using COP-1 as a fluorogenic switch-on probe

Article abstract of DOI:10.1039/c4dt00254g

Five manganese(i) tricarbonyl complexes of the general formulae [Mn(bpea^NCHC6H4R)(CO)₃]PF₆ and [Mn(bpea ^NHCH2C6H4R)(CO)₃]PF₆ based on the tridentate bis(pyrazolyl)ethylamine (bpea) ligand, each containing a pendant 4-substituted phenyl group with R = H, I, and CC-H, were synthesized and fully characterized, including X-ray structure analysis for three compounds. All complexes are stable in the dark in aqueous buffer for an extended period of time. However, CO-release could be triggered by illumination at 365 nm, establishing these compounds as novel photoactivatable CO-releasing molecules (PhotoCORMs). The influence of the imine vs. amine group in the ligands on the electronic structure and the photophysical behavior was investigated with the aid of DFT and TDDFT calculations. Solution IR studies on selected compounds allowed identification of intermediates resulting from the photoreaction. Finally, light-induced CO release from a model compound was demonstrated both in PBS buffer and in vitro in human umbilical vein endothelial cells (HUVECs) using COP-1 as a fluorescent switch-on probe.

Full text of DOI:10.1039/c4dt00254g

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
New modular manganese(I) tricarbonyl complexes
as PhotoCORMs: in vitro detection of
photoinduced carbon monoxide release using
COP-1 as a fluorogenic switch-on probe†
Cite this: Dalton Trans., 2014, 43,
8664
Sandesh Pai,a Maryam Hafftlang,b George Atongo,a Christoph Nagel,a
Johanna Niesel,a Svetlana Botov,c Hans-Günther Schmalz,c Benito Yardb and
Ulrich Schatzschneider*a
Five manganese(I) tricarbonyl complexes of the general formulae [Mn(bpeaNvCHC6H4R)(CO)3]PF6 and
[Mn(bpeaNHCH2C6H4R)(CO)3]PF6 based on the tridentate bis(pyrazolyl)ethylamine (bpea) ligand, each con-
taining a pendant 4-substituted phenyl group with R = H, I, and CuC–H, were synthesized and fully
characterized, including X-ray structure analysis for three compounds. All complexes are stable in the
dark in aqueous buffer for an extended period of time. However, CO-release could be triggered by illumi-
nation at 365 nm, establishing these compounds as novel photoactivatable CO-releasing molecules
(PhotoCORMs). The influence of the imine vs. amine group in the ligands on the electronic structure and
the photophysical behavior was investigated with the aid of DFT and TDDFT calculations. Solution IR
studies on selected compounds allowed identification of intermediates resulting from the photoreaction.
Finally, light-induced CO release from a model compound was demonstrated both in PBS buffer and
in vitro in human umbilical vein endothelial cells (HUVECs) using COP-1 as a fluorescent switch-on probe.
Received 23rd January 2014,
Accepted 1st March 2014
DOI: 10.1039/c4dt00254g
www.rsc.org/dalton
issues associated with the use of an odorless gaseous com-
pound which is highly toxic when overdosed, a major problem
Introduction
Carbon monoxide is now well-established as one of the three of inhalative application is that the final attainable cellular
major small signalling molecules in higher organisms, includ- concentration is determined by the partition ratio of CO in
ing humans.1–4 It is enzymatically generated by haem degra- the lungs, body fluids, and tissues, which are fixed values.
dation by the constitutive (HO-2) and inducible (HO-1) Therefore, over the last decade, carbonyl compounds have
isoforms of haem oxygenase (HO), also leading to the for- been explored as solid, easy-to-handle prodrugs for carbon
mation of ferrous iron and biliverdin, with the latter further monoxide delivery to biological systems.13 Most of these CO
converted to bilirubin.5–8 Its main physiological role is in vaso- carrier systems, collectively referred to as CO-releasing mole-
dilation and response to oxidative stress, but the concen- cules (CORMs),14–17 are based on transition metal carbonyl
tration-dependent switch between cytoprotective and cytotoxic complexes, although
a
few main group carbonyl
activity is also increasingly explored now in the context of compounds18–20 and, more recently, purely organic systems
potential therapeutic applications.9–12 There are some devices have been explored.21,22 In addition to ligand exchange reac-
available which allow for a controlled application of CO gas tions with medium,23 there are also other trigger mechanisms
mixtures to patients, but in addition to the general safety to initiate the CO release. These include enzymatic cleavage of
remote bonds by esterases and phosphatases in enzyme-trig-
gered CORMs (ET-CORMs),24–27 magnetic heating of CORM-
aInstitut für Anorganische Chemie, Julius-Maximilians-Universität Würzburg,
loaded maghemite nanoparticles,28 and probably most promi-
Am Hubland, D-97074 Würzburg, Germany.
nently, light-induced liberation of carbon monoxide from the
E-mail: ulrich.schatzschneider@uni-wuerzburg.de
bV. Medizinische Klinik, Universitätsmedizin Mannheim, Theodor-Kutzer-Ufer 1-3,
metal coordination sphere in photoactivatable CO-releasing
molecules (PhotoCORMs).29–31 Iron,32–35 molybdenum,36 tung-
sten,37 rhenium,38 and manganese39–46 carbonyl complexes
have received particular attention in this context. While most
of these systems are relatively simple, low-molecular weight
compounds, more recently, conjugation of CORMs to
D-68167 Mannheim, Germany
cInstitut für Organische Chemie, Universität zu Köln, Greinstr. 4, D-50939 Köln,
Germany
†Electronic supplementary information (ESI) available. CCDC 976263, 976264
and 976267. For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c4dt00254g
8664 | Dalton Trans., 2014, 43, 8664–8678
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
macromolecular and nanoparticle carrier systems has also group, in which R = H, I, or CuC–H, with the latter two provid-
received more and more attention, in order to achieve ing a handle for further bioconjugation via Sonogashira or
additional control of CORM tissue accumulation and CO bio- CuAAC “click” reactions. Reduction of the imines 9–10 with
logical activity. In addition to soft peptide,36,47,48 dendrimer,49 sodium borohydride gave ligands 12–13 with the more stable
and polymer50 conjugates and CORM-loaded micelles,51 hard amine group. All five ligands were obtained in good yields of
CORM-modified nanomaterials have been reported,52,53 70–80%. The facial manganese(I) tricarbonyl complexes 14–18
including those in which the CO is loaded into a porous (Fig. 1) were obtained in moderate to good yields by reaction
material54 instead of surface decoration of nanoparticles with of ligands 9–13 with manganese pentacarbonyl bromide in
molecular metal-carbonyl groups. Such polyfunctional CO acetone at reflux under a dinitrogen atmosphere with exclusion
delivery systems would significantly benefit from a modular of light (Scheme S2†). Prominent IR vibrational bands of
ligand design, allowing facile introduction of different func- 14–18 are listed in Table 1. Compared to the free imine ligands
tional groups onto the periphery, which can undergo mild, 9–11, the CvN stretching vibrations are shifted to lower wave-
bioorthogonal coupling reactions55–58 with macromolecular numbers upon coordination to the Mn(I) center in 14–16.
and nanoscale carriers. Since the modified tris(pyrazolyl)-
Furthermore, all complexes show two intense bands at
methane (tpm)47,52,53 and 2,2′-bipyridine (bpy)36,49,59 ligands about 1930 and 2040 cm−1, which are assigned to the antisym-
used in our previous studies do not allow such a modular metrical and symmetrical CuO stretching vibrations of the
approach, we have recently started to explore the use of a bis- metal tricarbonyl moiety, respectively, see Fig. 2 for a selected
(pyrazolyl)ethylamine (bpea)-based ligand system, which offers example. As expected due to the large number of intervening
an easy, flexible, and modular introduction of reactive func- bonds, no influence of the para-substituent on the phenyl ring
tional groups for further bioconjugation via condensation with on their position is observed.
para-substituted benzaldehydes, leading to either imine- or
The 1H NMR chemical shifts of ligands 9–13 and their
amine-containing N,N,N-chelators, depending on a subsequent corresponding complexes 14–18 are collected in Table S1.†
reduction step or not. In the present work, we would like to The signal of the methine proton shifts from about 6.8 ppm in
report on the CO-release properties of manganese(I) tricarbo- the imine ligands 9–11 to 7.8 ppm for complexes 14–16, while
nyl complexes of the parent bpea ligand as well as iodo- and a slightly smaller shift from 6.5 ppm in the amine ligands
alkyne-functionalized derivatives thereof, prepared for poten- 12–13 to 7.4 ppm for complexes 17–18 was observed upon
tial use in Sonogashira and CuAAC “click” reactions. Further- coordination to the Mn(I) center. Furthermore, the imine
more, it was the aim of our study to explore the utility of the proton shifts from 8.2 ppm in ligands 9–11 to 9.4 ppm for
recently reported fluorescent COP-1 probe60–62 to monitor the complexes 14–16 (Fig. S1–S3†). Only a broad singlet appears at
photoinduced CO-release from our novel PhotoCORMs in 5.6 ppm due to the amine proton in 17–18 (Fig. S4, S5†). The
living human umbilical vein endothelial cells (HUVECs) to two protons of the methylene group adjacent to both the
confirm that carbon monoxide delivery using these com-
pounds indeed works in a complex biological environment.
Results and discussion
Synthesis and characterization
Five new functionalized ligands 9–13 based on the bis(2-pyra-
zolyl)ethylamine (bpea) moiety were synthesized by conden-
sation of amine 5 with para-substituted benzaldehydes 6–8
(Scheme 1 and Scheme S1†). Three ligands 9–11 incorporate
an imine backbone with a pendant 4-substituted phenyl Fig. 1 fac-Manganese(I) tricarbonyl bpea complexes 14–18.
Scheme 1 Synthesis of ligands 9–13.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8664–8678 | 8665

View Article Online
Paper
Dalton Transactions
Table 1 Prominent IR vibrational band positions (in cm−1) of ligands
9–13 and complexes 14–18
Table 2 Crystallographic parameters for 15 and 18
Compound
15
18
CvN
CuOantisym
CuOsym
Empirical formula
Formula weight
Dimensions (mm)
Crystal system
Space group
a (Å)
b (Å)
c (Å)
C
18H14F6IMnN5O3P
C21H22BrIMnN5O4
670.19
0.30 × 0.10 × 0.10
Triclinic
ˉ
P1
9.8908(3)
11.9141(4)
12.1870(3)
98.487(2)
110.618(3)
106.055(3)
1242.87(6)
2
675.15
9
1649
1644
1641
—
—
—
—
—
—
—
—
—
0.30 × 0.25 × 0.15
Monoclinic
P2(1)/n
14.7315(7)
11.1201(3)
15.6814(7)
90
116.031(6)
90
2308.27(19)
4
10
11
12
13
14
15
16
17
18
—
—
—
1625
1624
1633
—
1932
1931
1930
1939
1928
2040
2043
2041
2045
2041
α (°)
β (°)
γ (°)
—
V (Å3)
Z
ρcalc (g cm−3
T (K)
)
1.943
113(2)
2.059
1.791
113(2)
3.417
µ (mm−1
)
λ (Å) (Mo Kα)
2Θmax (°)
0.71073
25.00
0.71073
25.00
Reflections measured
Unique refl./[I > 2σ(I)]
Data completeness
Variables
R (I ≥ 2σ(I))
wR (I ≥ 2σ(I))
Largest difference map
peak/hole in e Å−3
Goodness of fit (GOF)
7728
4066/3435
0.999
316
0.0301
0.0643
0.802/−0.463
13 673
4376
0.998
302
0.0200
0.0460
0.593/−0.291
1.026
0.954
Table 3 Selected bond lengths (Å) and angles (°) for complexes 15 and
18
15
18
Fig. 2 ATR IR spectrum of [Mn(bpeaNHCH2C6H5)(CO)3]PF6 17, showing
the symmetrical and antisymmetrical CuO stretching vibrations at 2045
Mn1–C1
Mn1–C2
Mn1–C3
Mn1–N1
Mn1–N3
Mn1–N5
C1–O1
C2–O2
C3–O3
C5–N2
C5–N4
C4–C5
C4–N1
N1–C12
1.812(3)
1.822(4)
1.804(4)
2.091(2)
2.042(3)
2.042(3)
1.144(4)
1.141(4)
1.147(4)
1.451(4)
1.454(4)
1.510(4)
1.479(4)
1.289(4)
1.805(2)
1.801(2)
1.818(2)
2.133(2)
2.036(2)
2.035(2)
1.153(3)
1.151(3)
1.146(3)
1.456(3)
1.453(3)
1.519(3)
1.484(3)
1.501(3)
and 1939 cm−1
.
coordinated amine and the methine give rise to individual
signals at 3.1 and 3.5 ppm, respectively, due to the diastereoto-
pic environment generated by metal coordination of the
amine, and both are further split by coupling to the neighbor-
ing CH and NH protons. In contrast, in the free ligands 12–13,
these give rise to only a single doublet at around 3.7 ppm. In
the imine compounds, this signal is much less sensitive to a
coordination-induced shift and for both ligands 9–11 and
complexes 14–16 appears at about 4.6 ppm.
C1–Mn1–N1
C2–Mn1–N3
C3–Mn1–N5
(Mn1–N1–C4–C5)–(N1–C12–C13)
175.09(13) 174.30(8)
178.48(13) 177.90(9)
176.86(12) 178.03(9)
8.1°
48.7°
93.7°
Molecular structures of the complexes
(N1–C12–C13)–(C13–C14–C15–C16–C17–C18) 9.7°
Single crystals suitable for X-ray structure analysis were
obtained by slow diffusion of n-hexane into acetone solutions
of 15, 17, and 18. Crystallographic parameters are listed in
Table 2 and Table S2.† Selected bond distances and angles of
the complexes are collected in Table 3 and Table S3.† Mole-
cular structures are shown in Fig. 3 and 4 as well as Fig. S6.†
In each case, the three carbonyl ligands are coordinated to the
Mn(I) center in a facial manner. Three nitrogen atoms of the
tridentate bpea ligand, two from the pyrazol groups and one
from the imine/amine, are located opposite to the CO ligands
to give an approximate C2v symmetry of the Mn(CO)3N3 core,
which is however broken by the orientation of the substituents
on the sp3 amine nitrogen atom in 17 and 18. The effect of the
imine vs. amine ligation is also clearly reflected in the
Mn1–N1 bond distance, which is 2.091(2) Å in 15 but extended
by about 0.04 Å to 2.133(2) Å in 18. The other parameters, in
particular the Mn–C, Mn–Npyrazole, and CuO bond distances,
show much less variation between the compounds and gener-
ally do not deviate by more than 0.02 Å from 15 to 18. The
N1–C12 bond length is 1.289(4) Å in imine complex 15 and
1.501(3) Å in amine compound 18. Bond angles around the
metal center are close to 90° and 180°, respectively, as expected
8666 | Dalton Trans., 2014, 43, 8664–8678
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 3 Molecular structure of the cationic unit of 15 with ellipsoids
drawn at the 50% probability level. The hexafluorophosphate counterion
is not shown for clarity.
Fig. 5 UV/Vis absorption spectra of 14–18 in DMSO.
Table 4 Absorption maxima and molar extinction coefficient of com-
plexes 14–18
Compound
λmax [nm]
ελ [M−1 cm−1
]
ε365 nm [M−1 cm−1
]
max
14
15
16
17
18
341
348
347
355
356
3460 60
4410 60
5500 310
2135 80
2275 80
3005 55
3550 40
5495 95
2020 85
2155 65
Fig. 4 Molecular structure of the cationic unit of 18 with ellipsoids
drawn at the 50% probability level. The bromide counterion and a
solvate acetone molecule are not shown for clarity.
nature of the para-substituent on the phenyl ring, which is
thought to be due to a decrease in electronic communication
with the metal center for NHCH2 vs. NvCH. The stability of
the compounds was also studied in aerated DMSO solution.
They did not show any change in absorbance when kept in the
dark over a course of 14 h, as shown for 17 as an example in
Fig. 6. However, illumination at 365 nm using a UV lamp leads
to a pronounced decrease in absorption (Fig. S7†). A plateau
for a near-octahedral coordination environment. The most pro-
minent difference between the imine and amine complexes is,
however, the orientation of the phenyl ring relative to the bis-
(pyrazolyl) core. In imine compound 15, it is essentially
located in the mean plane intersecting the N3–Mn1–N5 and
C2–Mn1–C3 angles, as defined by Mn1, N1, C4, and C5, and
only slightly distorted relative to the CvN group, with a (Mn1–
N1–C4–C5)–(N1–C12–N13) angle of 8.1° and a (N1–C12–C13)–
phenyl angle of 9.7°. In amine complex 18, on the other hand,
the sp3 configuration of N1 leads to significantly expanded di-
hedral angles of 48.7° and 93.7° for the orientation of these
atoms. The unsubstituted parent compound 17 shows a
similar effect, with the two independent molecules in the unit
cell having essentially identical parameters, with the exception
of the relative orientation of the phenyl ring (Table S3†).
Electronic absorption spectra and photolysis of the complexes
The UV/Vis absorption spectra of complexes 14–18 were
recorded in dimethylsulfoxide and are shown in Fig. 5. For all
five complexes, a broad maximum centered at around 350 nm
with an extinction coefficient of about 2500–5000 M−1 cm−1 is
observed (Table 4), which is assigned to a metal-to-ligand
charge transfer transition (vide infra). The intensity of this
band is somewhat smaller for the amine vs. imine compounds,
Fig. 6 Changes in the absorbance at 355 nm of 17 upon incubation in
dimethylsulfoxide in the dark for the initial 14 h and during subsequent
photolysis of the same solution with an UV lamp at 365 nm. See Fig. S7†
and for the former complexes also show less variation with the for the full spectral traces.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8664–8678 | 8667

View Article Online
Paper
Dalton Transactions
value was reached after a couple of minutes, with the half-life
determined as 1.5 min from proper fitting of the data. When
the same experiment was repeated on 17 with LED illumina-
tion at 410 nm, essentially the same spectral changes are
observed (data not shown), but the half-life is significantly
extended to 52 min, which is due to the much lower extinction
coefficient at this wavelength (see Fig. 5).
Photoinducible CO release and quantum yield measurements
The dark stability and photoactivated release of CO from com-
plexes 14–18 was assessed with the standard myoglobin
assay.63,64 In these experiments, a freshly prepared concen-
trated solution of metal complex in dimethylsulfoxide was
added to a buffered solution (PBS, pH 7.4) of horse skeletal
myoglobin (Mb), which was reduced with excess sodium
dithionite under dinitrogen in the dark. When kept in the
dark for up to 14 h under the conditions described above,
none of the compounds 14–18 induced any spectral changes
in the Q-band region of myoglobin, as shown for 17 in Fig. 7
as an example, thus establishing them as dark-stable CORM
prodrugs. However, when illuminated at 365 nm, the absorp-
tion of Mb at 557 nm decreased and the typical peaks of
MbCO at 540 and 577 nm grew in intensity with increasing
illumination time (Fig. S8†). The amount of MbCO formed at
the plateau level (Fig. 8) was calculated using the published
Fig. 8 Comparison of the amount of MbCO (in µM) formed with
increasing illumination time of a solution of complex 14–18 (10 µM) and
myoglobin (60 µM) in 0.1 M PBS pH 7.4 at 365 nm.
410 nm), the half-life of CO release is significantly extended to
132 min at 410 nm due to the lower absorption at this wave-
length. The reaction quantum yields of complexes 17–18 are
slightly lower than those of the imine compounds 14–16. Still,
the
Ф values are much lower than those reported for
[Mn(CO)3(tpa)]ClO4 and [Mn(CO)3(dpa)]Br66 at similar wave-
lengths and in the order of 10−3, which is due to the internal
shielding by the horse skeletal myoglobin absorption at the
excitation wavelength of 365 nm. Nevertheless, the present
data establish manganese(I) tricarbonyl complexes 14–18 as a
new family of Mn-based photoactivatable CO-releasing mole-
cules (PhotoCORMs) with good dark stability and modular syn-
thesis, facilitating further modification and bioconjugation.
extinction coefficient of MbCO [λ540 nm = 15.4 (mM)−1 −1].65
L
The half-lives of CO release, equivalents of carbon monoxide
liberated from complexes 14–18, and their reaction quantum
yields determined at 365 nm (Ф365 nm) are listed in Table 5.
All five complexes release about two equivalents of CO
under the conditions of the myoglobin assay at 365 nm
illumination. However, the half-lives of the imine complexes
14–16 are slightly shorter than those of the amine compounds
17–18 at 19–22 min vs. 26–29 min. To investigate the effect of
lower energy photoactivation, complex 17 was also studied
DFT and TDDFT calculations
using 410 nm illumination from a LED array. While there is To obtain a more detailed picture of the effect of the imine vs.
essentially no change in the number of CO ligands released at amine group in the ligands on the photophysical and photo-
this excitation wavelength (2.1 eq. at 365 nm vs. 2.4 eq. at chemical properties of the title compounds and to assign the
Fig. 7 Absorbance values at four different wavelengths in the Q-band region of myoglobin (60 µM) upon incubation with complex 17 (10 µM) in the
dark for up to 14 h in 0.1 M PBS at pH 7.4 with sodium dithionite (10 mM) under a dinitrogen atmosphere and (inset) UV/Vis absorption spectra
recorded at different time points during the dark incubation.
8668 | Dalton Trans., 2014, 43, 8664–8678
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 5 CO-release data and reaction quantum yield values at 365 nm for 14–18
Compound Conc. of MbCO [µM] Equiv. of CO released Half-life, t1/2 a [min] Rate constant kCO b [s−1
]
Reaction quantum yield (Ф365 nm
)
c
14
15
16
17
18
22.5 0.3
24.8 0.8
23.2 0.2
20.7 0.3
21.3 0.3
2.3 0.1
2.5 0.1
2.3 0.1
2.1 0.1
2.1 0.1
21.6 0.3
21.3 0.5
18.5 0.2
25.8 0.2
28.7 0.8
0.00642
0.01060
0.01208
0.00783
0.00745
(5.9 0.2) × 10−3
(6.3 0.4) × 10−3
(6.6 0.1) × 10−3
(4.4 0.2) × 10−3
(4.3 0.2) × 10−3
a Under the conditions of the myoglobin assay. b Determined in DMSO solution. c Calculated using a photon flux of the UV lamp of (2.82 0.05)
× 10−8 Einstein s−1
.
CuO vibrations of intermediate species resulting from photo- imine vs. amine group on the electronic structure of the title
induced carbon monoxide release from the metal coordination compounds, TDDFT calculations were carried out on the
sphere, DFT and TDDFT calculations were carried out minimum geometries and the first 45 singlet excited states
on imine [Mn(bpeaNvCHC6H5)(CO)3]+
[Mn(bpeaNHCH2C6H5)(CO)3]+ B, generated from the X-ray struc-
A
and amine obtained this way (Table 7). As expected for a Mn(I) center with
low-spin 3d6 electronic configuration, the HOMO to
a
tures of 15 and 18, respectively, by removal of counterions and HOMO−2 orbitals are essentially of metal d character in A as
solvate molecules as well as replacement of the iodo group by well as B, and well-separated by 0.79 and 0.45 eV, respectively,
hydrogen in the para-position of the phenyl ring. The ORCA from the next lowest occupied orbital, which is of π phenyl
program package with the RI-BP86 (geometry optimization character (Tables S5 and S7†). The LUMO is of mixed phenyl/
and calculation of vibrational frequencies) or RIJCOSX-B3LYP imine character in A, but centered on the two pyrazole groups
functionals (TDDFT) and a TZVP basis set on all atoms was in B. Further unoccupied orbitals are either of π*(pyrazole),
used in both cases, taking into account the DMSO solvent with π*(phenyl), or π*(CO) character. The low-energy states of rele-
the COSMO model. As already observed in DFT studies of vant oscillator strength are all of MLCT character (Table 7).
related manganese(I) tricarbonyl complexes using a similar However, while those in the range of 300–350 nm are from the
model chemistry,49 the deviations between calculated bond manganese to the pyrazoles for both A and B, the imine com-
distances and angles from those obtained by X-ray crystallo- pound A is predicted to show an additional Mn→phenyl/imine
graphy are very small and do not exceed 2.5% as averaged over MLCT at around 470 nm. Presumably, conjugation of the
all parameters (Table S4†).
NvCH group with the phenyl ring lowers the energy of one of
No imaginary frequencies were found among the calculated the π* orbitals of this group by about 2 eV, making it the
vibrational modes, establishing the geometries obtained as LUMO and leading to its significant contribution to the
minimum structures, but due to significant deviation from the lowest-energy state. Only below about 300 nm, the difference
experimental values they had to be scaled by a factor of 1.036, densities show additional LLCT and ILCT transitions (Tables
which leads to an excellent agreement with all deviations S6 and S8†). The manganese→pyrazole MLCT transitions close
below 1% or, in absolute terms, less than 20 cm−1. It should to the excitation wavelength of 365 nm will decrease the metal
be noted that this scaling factor deviates from the ones electron density in the excited state, which results in a weaken-
reported in some literature, but was obtained from a specific ing of the back-donation to the CO π* orbitals and sub-
training set based on metal carbonyl compounds. A compari- sequently facilitates M–CO bond breaking, resulting in
son of calculated and experimental vibrations is presented in liberation of the carbon monoxide from the metal coordi-
Table 6 together with the assignment of the major modes. The nation sphere, as has been demonstrated for other manga-
broad antisymmetrical CuO stretch could not be resolved in nese-based PhotoCORMs before.41,49
the two contributions in the experimental ATR IR spectra. As
observed previously,49 the antisymmetrical CuO stretches are
CO release studied by solution IR spectroscopy
underestimated by a maximum of about 20 cm−1, while the
In order to follow the CO release upon photoactivation, solu-
tions of 14 and 17 were prepared in DMSO. The sample solu-
tion was filled in a liquid IR cell and was illuminated at
365 nm for fixed time intervals after which IR spectra were
symmetrical one is overestimated by about the same value
when scaled as above. To investigate the influence of the
Table 6 Comparison of major experimental and scaled calculated
recorded. The carbonyl vibrational region of the IR spectra is
shown in Fig. 9 and Fig. S9† for compounds 17 and 14,
vibrational frequencies for 14 and 17 vs. A and B. The deviation from the
experimental values is given in parentheses (all in cm−1
)
respectively. The two bands at 2038 and 1939 cm−1, which are
typical for fac-Mn(CO)3 complexes, decrease in intensity upon
illumination while three to four new bands appear in the
region between 1640 and 1900 cm−1. Interestingly, these
signals persist even after prolonged illumination, which is
not the case with other related compounds such as
Mode
14 exp.
A calc.
17 exp.
B calc.
ν(CHvN)
1625
1932
1932
2040
1605 (−20)
1929 (−3)
1930 (−2)
2059 (+19)
—
—
νantisym1(CuO)
νantisym2(CuO)
νsym(CuO)
1939
1939
2045
1922 (−17)
1928 (−11)
2066 (+21)
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8664–8678 | 8669

View Article Online
Paper
Dalton Transactions
Table 7 Energies (in nm), oscillator strength ( fosc), main orbital contributions, and the type of transition involved in the most important singlet exci-
tations for imine A and amine B calculated with TDDFT
A
Statea
λ (nm)
fosc
Main transitionsb
Type of transition
2
4
5
469.3
350.0
329.2
0.2602
0.0168
0.0528
102→103 (95%)
102→104 (92%)
101→104 (71%)
102→105 (22%)
102→105 (54%)
101→105 (83%)
97→103 (55%)
98→103 (43%)
102→107 (75%)
101→108 (84%)
97→103 (38%)
98→103 (41%)
MLCT Mn→phenyl + imine
MLCT Mn→pyrazole
MLCT Mn→pyrazole
8c
296.7
299.4
304.4
0.1258
0.0941
0.1377
MLCT Mn→pyrazole(+phenyl)
MLCT Mn→pyrazole
LLCT phenyl→pyrazole
9c
11c
13
15
18
264.7
260.6
258.5
0.0850
0.0362
0.4138
MLCT Mn→CO
MLCT Mn→CO
ILCT phenyl→phenyl
B
Statea
λ (nm)
fosc
Main transitionsb
Type of transition
1
2
4
5
7
8
343.0
337.2
304.8
295.4
280.6
276.2
0.0110
0.0781
0.1017
0.0660
0.0127
0.0446
102→104 (84%)
103→104 (82%)
103→105 (75%)
102→105 (70%)
101→105 (62%)
101→105 (21%)
103→106 (48%)
102→108 (70%)
100→104 (57%)
MLCT Mn→pyrazole
MLCT Mn→pyrazole
MLCT Mn→pyrazole
MLCT Mn→pyrazole
MLCT Mn→pyrazole
MLCT Mn→pyrazole + phenyl
10
13
261.0
253.5
0.0545
0.0190
MLCT Mn→CO
MLCT/LLCT phenyl→pyrazole
a Only strong transitions with an oscillator strength >0.01 in the 250–600 nm range are reported. b Only contributions >20% are listed. c The
reversed order is due to the states experiencing a different degree of solvent shift.
monoxide observed under the conditions of the myoglobin
assay (vide supra).
To assist in the assignment of the species resulting from
the photolysis, DFT calculations were carried out to obtain the
vibrational signature of potential products and intermediates
(Table 8). The general validity of this approach was previously
demonstrated by comparison of optimized geometries to those
of the X-ray structural data of 14 and 17. Two isomers were
examined for each CO release step, depending on whether the
CO liberated originates from a position trans to a pyrazole or
trans to the amine/imine group (Fig. 10). In line with previous
Table 8 Scaled calculated vibrational frequencies for potential pro-
ducts C to F resulting from the prolonged photolysis of imine 14 and
Fig. 9 Spectral changes in the solution IR of 17 (9 mM solution in
DMSO) upon illumination (0–1320 s) at 365 nm. The red trace was
recorded on a freshly prepared solution before the start of the illumina-
tion whereas the blue trace shows the final spectrum recorded after
about 22 min of illumination.
amine 17 at 365 nm with differences to experimental values in round
brackets (all in cm−1
)
Mode
C
C′
Da
D′
ν(CHvN)
νantisym(CuO)
νsym(CuO)
1560 (−89)
1862 (+2)
1935 (+39)
1554 (−95)
1863 (−2)
1940 (+44)
—
—
—
1629 (−20)
—
1848 (+24)b
[Mn(CO)3(tpm)]+, [Mn(CO)3(tpa-κ3N)]+, and [Mn(CO)3(tpp)]+.67
The two signals growing at 1898 and 1862 cm−1 for 17 are
indicative of a cis-Mn(CO)2 moiety remaining after the release
of the first equivalent of carbon monoxide, while the
1648 cm−1 band appears at the same position as found in
imine ligands 9–11 for the CvN stretching vibration. This
result contrasts with the release of two equivalents of carbon
Mode
E
E′
F
F′
νantisym(CuO)
νsym(CuO)
1846 (−16)
1935 (+37)
1835 (−27)
1919 (+21)
—
—
1825b,c
1826b,c
a Could not be converged in the low-spin state. b Monocarbonyl
species, only one CuO vibrational mode. c Not observed, probably
covered by remaining signals of the starting material.
8670 | Dalton Trans., 2014, 43, 8664–8678
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
originates from a modified protein and thus is expected to
have much less general applicability. Since both systems have
so far only been evaluated using CO gas and CORM-3 as the
carbon monoxide source, we were particularly interested to
validate whether a response to photoinduced carbon monoxide
release from our PhotoCORMs could also be detected. Initially,
the COP-1 response was examined in PBS buffer in the pres-
ence and absence of light. A dose-dependent increase in fluo-
rescence signal was observed when increasing amounts of
PhotoCORM 17 (0–100 µM) were added to COP-1 (10 µM in
PBS) followed by photoactivation at 365 nm for 30 min while
in the non-illuminated sample, the fluorescence intensity of
the mixture remained at the basal level (Fig. 11).
Importantly, even at a 1 : 10 molar ratio of COP-1 to Photo-
CORM 17, which translates to a 1 : 20 molar ratio of COP-1 to
labile CO, if one assumes that only two out of the three carbo-
nyl ligands are released from the metal coordination sphere,
the signal was still not fully saturated and a non-linear
response of COP-1 fluorescence with increasing CO concen-
tration was observed.
To check whether the light-triggered release of carbon mon-
oxide from 17 could also be detected by COP-1 in living cells,
human umbilical vein endothelial cells (HUVECs) were incu-
bated in microtiter plates with COP-1 (10 µM) in PBS followed
by addition of PhotoCORM 17 (100 µM) in the dark. The cell
cultures were then exposed to 365 nm light for 30 min, while a
control experiment was performed under exactly the same con-
ditions, but without illumination. Further controls included
phosphate buffer saline (PBS) only as well as COP-1 and
sodium hexafluorophosphate in the same buffer, both in the
presence and absence of UV light. At the end of the experi-
ment, the supernatant and the cell-containing phase were
separated and individually assessed. The molecular probe
Fig. 10 Structure of the cationic unit of iCORM intermediates C–F opti-
mized with BP86 after subsequent removal of carbonyl ligands and
replacement with one or two solvent water molecules. Color coding:
carbon (grey), hydrogen (white), nitrogen (blue), oxygen (red), manga-
nese (magenta).
studies, the octahedral coordination sphere of the manganese
center was retained and departed CO was replaced by water
molecules as models for solvent. As evident from Table 8,
using this methodology, it is not possible to conclusively
decide between these two cases. However, the first intermedi-
ate is clearly identified as a cis-manganese dicarbonyl, and at
least in the case of the imine complex 14, also some amount of
monocarbonyl seems to form after extended photolysis (Fig. S9†).
In vitro detection of CO release using the COP-1 fluorogenic
switch-on probe
Very recently, the groups of Chang and He reported on fluo-
rescence-based molecular probes for the detection of carbon
monoxide in complex biological systems.60–62 We were particu-
larly interested in the application of COP-1, since it is based
on a small-molecule system composed of a BODIPY switch-on
fluorescent dye coupled to a cyclometalated dimeric palladium
unit, which is cleaved upon reaction with carbon monoxide.
Fig. 11 Dose-dependent fluorescence response of 10 µM COP-1 in
PBS to photoinduced release of CO from increasing amounts of 17 (data
points shown at 0, 3, 6, 12.5, 25, 50, and 100 µM). Fluorescence intensity
was measured at an emission wavelength, λem = 510 nm with λex
=
475 nm. The data are expressed as mean fluorescence intensity SD
The COSer system from the group of He, on the other hand, with experiments performed in quadruplicate.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8664–8678 | 8671

View Article Online
Paper
Dalton Transactions
remaining in the supernatant only shows a very weak fluore- (10 µM) and compound 17 (100 µM) in PBS when kept in the
scence both in the dark control and upon illumination at dark (Fig. 12a, 4th column). Only the combination of added
365 nm for 30 min (Fig. 12a, 2nd column). The same is also metal carbonyl complex 17 and illumination led to a signifi-
true for complex 17 alone (Fig. 12a, 3rd column). Furthermore, cant, about 15-fold increase in fluorescence intensity (Fig. 12a,
sodium hexafluorophosphate as the source of the counter ion 4th column), while under all other conditions, the emission
and a mix of COP-1 and NaPF6 were also examined for back- remained at essentially baseline level. A similar trend was
ground signal but found to be non-emissive (Fig. 12a, 5th and observed with the cell-based fraction, but the increase for the illu-
6th columns). The same is also true for a mixture of COP-1 minated vs. non-illuminated sample was only about 5-fold higher
compared to the controls, and thus significantly smaller than
observed in the supernatant. This could be due to impaired cellu-
lar uptake of either the COP-1 probe or the metal complex, or
both, or some kind of deactivation or quenching process. Thus,
although COP-1 is very important as one of the first in vitro probes
for carbon monoxide and was also demonstrated here to allow
detection of light-induced CO release from PhotoCORMs, clearly
more improved and sensitive derivatives thereof are urgently
needed as tools for the study of small-molecule signalling.
Conclusions
In the current report, a series of five manganese(I) tricarbonyl
complexes based on the bis(2-pyrazolyl)ethylamine (bpea)
moiety with a pendant para-substituted phenyl group in which
R = H, I, or CuC–H has been synthesized. The iodo and
alkyne substituents will provide a flexible handle for bioconju-
gation via Sonogashira or CuAAC “click” reactions, for
example, to further improve the bioavailability and targeting
properties of these compounds, and the easy synthetic access
to the bpea group will also allow facile preparation of
additional derivatives with a tuned “drug sphere”. All com-
pounds are stable in DMSO solution and under the conditions
of the myoglobin assay in the dark for a period of up to 14 h,
thus establishing them as CORM prodrugs. However, they rapidly
release about two equivalents of CO per mole of PhotoCORM
upon illumination at 365 nm with half-lives in the range of
20–30 min, depending somewhat on the phenyl substituent
and the imine vs. amine linker. Light-induced CO release
could also be observed at 410 nm, although with a signifi-
cantly longer half-life. MLCT transitions from the manganese
to the pyrazole groups lead to a weakening of the metal–CO
bond and the photolability of the carbonyl ligands, as inferred
from DFT and TDDFT calculations. To demonstrate that this
CO release also takes place under physiological conditions, the
small-molecule fluorogenic switch-on carbon monoxide probe
COP-1 was used to detect CO release from PhotoCORMs for
the first time. While no response was observed in the presence
Fig. 12 Fluorescence intensity (λex = 475 nm, λem = 510 nm) observed
upon incubation of human umbilical vein endothelial cells (HUVECs)
with 17 in (a) either the supernatant (top) and (b) the cell-containing
of up to 10-fold excess of CORM over the probe when kept in
the dark, illumination led to a rapid dose-dependent increase
of the COP-1 fluorescence, which however did not show a
fraction (bottom); black bars: dark control; grey bars: illuminated for
30 min at 365 nm, see the Experimental section for details; addition of linear relationship with the CORM concentration and did not
(from left to right) PBS buffer only; COP-1 (10 µM) in PBS; 17 (100 µM) in
PBS; COP-1 (10 µM) and 17 (100 µM) in PBS; NaPF6 (100 µM) in PBS;
COP-1 (10 µM) and NaPF6 (100 µM) in PBS; the data are expressed as
mean fluorescence intensity SD with experiments performed in quad-
fully saturate at even a 1 : 20 ratio of COP-1 to labile carbonyl
ligands. When incubated with living HUVEC cells, COP-1
showed a significant response to light-induced CO release
from one of the metal carbonyl complexes compared to dark
ruplicate. The fluorescence intensity is shown in arbitrary units at the
same scale for both panels.
controls as well as other additives. The effect was, however,
8672 | Dalton Trans., 2014, 43, 8664–8678
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
much more pronounced in the supernatant than in the cell- under reduced pressure and the resulting solid was dried
containing fraction (about 15-fold vs. 4-fold increase), indicat- under vacuum.
ing that the cellular uptake of either the CORM, the probe, or
9. Pale yellow solid, yield 73% (2.00 g, 0.01 mol). IR (ATR,
1
both will require further improvement although the system in ν/cm−1): 3108 (w), 2852, 1649 (s, imine, CvN), 1514, 1434. H
general holds great promise as an in vitro reporter of carbon NMR (200 MHz, CDCl3, ppm): δ 8.20 (s, 1H, CHvN), 7.67 (d,
monoxide delivery from artificial carrier systems.
2H, 3J = 2.4 Hz, H3,3′-pz), 7.56 (d, 2H, 3J = 1.5 Hz, H5,5′-pz), 7.39
3
(m, 5H, C6H5), 6.80 (t, 1H, J = 7.00 Hz, CH(pz)2), 6.25 (t, 2H,
3
H4,4′-pz), 4.66 (d, 2H, J = 6.8 Hz, CH2). 13C-NMR (50.33 MHz,
Experimental
General procedures
CDCl3, ppm) δ 63.0 (CH2), 75.2 (C(pz)2), 106.6 (C4-pz), 128.4
(C3,5–C6H5), 128.7 (C2,6–C6H5), 129.2 (C4–C6H5), 131.3 (C5-pz),
135.8 (C1–C6H5), 140.5 (C3-Pz), 164.9 (CvN). MS (FAB+): m/z
266 [M + H]+. Elemental analysis (%): calc. for C15H15N5:
C 67.90, H 5.69, N 26.39; found: C 67.75, H 5.88, N 26.74.
10. White solid, yield 88% (1.17 g, 2.99 mmol). IR (ATR,
ν/cm−1): 3100 (w), 1644 (s, imine, CvN), 1586, 1510, 1436,
Reactions were carried out in oven-dried Schlenk glassware
under an atmosphere of pure dinitrogen when necessary. Sol-
vents were dried over molecular sieves and degassed prior to
use. All reagents were purchased either from Sigma-Aldrich or
Alfa Aesar and used without further purification. Manganese
pentacarbonyl bromide was obtained from Strem Chemicals.
2,2′-Bis(pyrazol-1-yl)ethylamine (bpea) 5 was prepared accord-
ing to the literature as described in the ESI.†68,69 NMR spectra
were recorded on Bruker DPX 200 and DRX 400 spectrometers
(1H: 200.13 and 400.13 MHz, respectively; 13C: 50.33 and
100.62 MHz) at ambient temperature. Chemical shifts δ in
ppm indicate a downfield shift relative to tetramethylsilane
(TMS) and were referenced relative to the signal of the
solvent.70 Coupling constants J are given in Hz. Individual
peaks are marked as singlet (s), doublet (d), doublet-of-
doublet (dd), triplet (t), or multiplet (m). Mass spectra were
measured on VG Autospec (FAB) and Bruker Esquire 6000 (ESI)
instruments. 3-Nitrobenzylalcohol was used as the matrix in
the FAB measurements. Data were recorded in positive ion
mode. Infrared spectra were recorded for pure solid samples
using a Bruker Tensor 27 IR spectrometer equipped with a
Pike MIRacle Micro ATR accessory. The solution IR measure-
ments were carried out with a Jasco FT/IR-4100 spectrometer
using an IR flow-cell holder equipped with calcium fluoride
windows (d = 4 mm) and a teflon spacer (d = 0.5 mm). The
elemental analysis (C, H, N) of the compounds were performed
with a VarioEL from Elementar Analysensysteme GmbH. The
myoglobin assay and all photophysical studies were carried
out in a quartz cuvette (d = 1 cm) with an Agilent 8453 UV/Vis
diode array spectrophotometer using either a UV/Vis hand
lamp (365 nm, UVIlite LF-206LS, 6 W, UVItec Ltd, Cambridge,
UK) or a custom-built LED light source (407–412 nm wave-
length range, 5 mm round type UV-LEDs, model YDG-504VC,
Kingbright Elec. Co., Taipei, Taiwan, http://www.kingbright.
com, part no. 181000-05).
1
1387. H NMR (200 MHz, CDCl3, ppm): δ 8.11 (s, 1H, CHvN),
3
3
7.71 (d, 2H, J = 8.4 Hz, H3,5–C6H4), 7.65 (d, 2H, J = 2.3 Hz,
3
3
H3,3′-pz), 7.55 (d, 2H, J = 1.5 Hz, H5,5′-pz), 7.34 (d, 2H, J = 8.4
3
Hz, H2,6–C6H4), 6.77 (t, 1H, J = 6.8 Hz, CH(pz)2), 6.25 (t, 2H,
H4,4′-pz), 4.64 (d, 2H, J = 6.8 Hz, CH2). 13C-NMR (100.62 MHz,
3
CDCl3, ppm) δ 63.0 (CH2), 75.1 (C(pz)2), 98.1 (C4–C6H4), 106.6
(C4-pz), 129.2 (C5-pz), 129.8 (C2,6–C6H4), 135.2 (C1–C6H4), 138.0
(C3,5–C6H4), 140.5 (C3-pz), 163.9 (CvN). MS (FAB+): m/z 392
[M + H]+, 414 [M + Na]+. Elemental analysis (%): calc. for
C15H14N5I: C 46.05, H 3.60, N 17.90; found: C 46.14, H 3.69,
N 17.82.
11. Colorless solid, yield 72% (585 mg, 2.02 mmol). IR
(ATR, ν/cm−1): 3254 (w), 3137, 2917, 2105 (CuC), 1641 (s,
imine, CvN), 1556, 1433. 1H NMR (400 MHz, CDCl3, ppm):
δ 8.17 (s, CHvN), 7.66 (d, 2H, 3J = 2.4 Hz, H3,3′-pz), 7.57 (d,
3
3
2H, J = 8.4 Hz, H3,5–C6H4), 7.55 (d, 2H, J = 1.6 Hz, H5,5′-pz),
7.47 (d, 2H, 3J = 8.4 Hz, H2,6–C6H4), 6.78 (t, 1H, 3J = 6.8 Hz,
3
CH(pz)2), 6.25 (t, 2H, H4,4′-pz), 4.67 (d, 2H, J = 6.8 Hz, CH2),
3.16 (s, 1H, CuC–H). 13C-NMR (100.62 MHz, CDCl3, ppm)
δ 63.0 (CH2), 75.2 (CuCH), 79.3 (C(pz)2), 83.3 (CuCH), 106.6
(C4-pz), 125.0 (C4–C6H4), 128.2 (C2,6–C6H4), 129.2 (C5-pz), 132.4
(C3,5–C6H4), 135.9 (C1–C6H4), 140.5 (C3-pz), 164.0 (CvN). MS
(FAB+): m/z 290 [M + H]+, 312 [M + Na]+. Elemental analysis
(%): calc. for C17H15N5: C 70.57, H 5.22, N 24.20; found: C
70.49, H 5.31, N 24.27.
General procedure for the synthesis of bpea amine ligands
12–13. An excess of solid sodium borohydride (65 mg,
1.72 mmol for 12 or 38 mg, 1.02 mmol for 13) was added in
portions to a solution of 9 (230 mg, 0.86 mmol) or 10 (160 mg,
0.41 mmol) in methanol (10 mL) and then stirred for 6 h at
room temperature. Water (5 mL) was added to the reaction
mixture. The solvent was then removed under vacuum and the
residue was redissolved in ethyl acetate (40 mL) and washed
Synthetic procedures
General procedure for the synthesis of bpea imine ligands with water (3 × 50 mL). The organic phase was separated, dried
9–11. A solution of benzaldehyde (1.12 mL, 1.00 mmol for 9), over sodium sulfate (∼5 g), and filtered. The solvent was then
4-iodobenzaldehyde (800 mg, 3.38 mmol for 10), or 4-ethynyl- removed from the filtrate under reduced pressure and the
benzaldehyde (365 mg, 2.83 mmol for 11) in methanol (10 mL) resulting solid was dried in vacuo.
was added dropwise to a stirred solution of 2,2′-bis(pyrazolyl)-
12. Pale yellow solid, yield 78% (178 mg, 0.66 mmol). IR
ethylamine (2.00 g, 1.00 mmol for 9 or 600 mg, 3.38 mmol for (ATR, ν/cm−1): 3341 (m, NH), 3137, 2830, 1599, 1493, 1367.
10 or 500 mg, 2.82 mmol for 11) in methanol (60 mL) and 1H-NMR (400 MHz, CDCl3, ppm): δ 7.58 (d, 2H, 3J = 2.4 Hz,
3
heated to reflux overnight. The solvent was then removed H3,3′-pz), 7.55 (d, 2H, J = 1.7 Hz, H5,5′-pz), 7.28 (m, 5H, C6H5),
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8664–8678 | 8673

View Article Online
Paper
Dalton Transactions
3
6.54 (t, 1H, J = 7 Hz, CH(pz)2), 6.27 (m, 2H, H4,4′-pz), 3.83 (s, 6.75 (t, 2H, H4,4′-pz), 4.63 (d, 2H, CH2). 13C-NMR (100.62 MHz,
2H, CH2–C6H5), 3.70 (d, 2H, 3J = 7.0 Hz, CH2). 13C-NMR acetone-d6, ppm) δ 51.4 (CH2), 58.0 (C(pz)2), 88.7 (C4–C6H4),
(100.62 MHz, CDCl3, ppm) δ 51.2 (CH2), 53.4 (CH2–C6H5), 75.2 98.8 (C4-pz), 120.0, 123.0, 125.2, 127.8, 128.0, 137.0 (C3-pz),
(C(pz)2), 106.8 (C4-pz), 127.3 (C4–C6H5), 128.2 (C3,5–C6H5), 164.9, 171.1 (CvN). MS (ESI+, CH3OH): m/z 529.81 [M − PF6]+.
128.7 (C2,6–C6H5), 129.1 (C5-pz), 139.8 (C1–C6H5), 140.4 (C3-pz). Elemental analysis (%): calc. for C18H14F6IMnN5O3P: C 32.02,
MS (FAB+): m/z 268 [M + H]+, 290 [M + Na]+. Elemental analysis H 2.09, N 10.37; found: C 31.83, H 1.91, N 10.26.
[Mn(CO)3(bpeaNvCHC6H4CuCH)]PF6 (16). Yellow solid, yield:
(%): calc. for C15H17N5: C 67.39, H 6.40, N 26.19; found:
C 67.12, H 6.35, N 26.61.
82% (355 mg, 0.62 mmol). IR (ATR, ν/cm−1): 3277 (m, N–H),
3138 (w), 3021 (w), 2041 (s, CuO), 1930 (s, CuO), 1633 (s,
imine, CvN), 1458, 1365, 1262, 1019. 1H-NMR (400 MHz,
acetone-d6, ppm): δ 9.39 (s, 1H, CHvN), 8.43 (d, 2H, 3J =
2.7 Hz, H5,5′-pz), 8.37 (d, 2H, 3J = 2.2 Hz, H3,3′-pz), 7.73 (m, 5H,
CH(pz)2, C6H4), 6.74 (t, 2H, H4,4′-pz), 4.63 (d, 2H, CH2), 3.88 (s,
1H, CuC–H). 13C-NMR (100.62 MHz, acetone-d6, ppm) δ 62.9
(CH2), 69.6 (CuCH), 81.7 (C(pz)2), 83.3 (CuCH), 109.6 (C4-pz),
126.2 (C4–C6H4), 129.3 (C2,6–C6H4), 133.1 (C3,5–C6H4), 135.7
(C5-pz), 137.0 (C1–C6H4), 147.8 (C3-pz), 181.9 (CvN), 220.3
(CuO). MS (ESI+, CH3OH): m/z 427.84 [M − PF6]+. Elemental
analysis (%): calc. for C20H15F6MnN5O3P: C 41.90, H 2.64,
N 12.22; found: C 41.98, H 2.54, N 12.28.
13. Colorless solid, yield 72% (115 mg, 0.29 mmol). IR
(ATR, ν/cm−1): 3327 (m, NH), 3090, 2928, 1510, 1482, 1387,
1283. 1H-NMR (400 MHz, CDCl3, ppm): δ 7.62 (d, 2H, 3J =
3
8.3 Hz, H3,5–C6H4), 7.57 (d, 2H, J = 2.4 Hz, H3,3′-pz), 7.55 (d,
3
3
2H, J = 1.5 Hz, H5,5′-pz), 7.02 (d, 2H, J = 8.3 Hz, H2,6–C6H4),
3
6.52 (t, 1H, J = 7 Hz, CH(pz)2), 6.27 (t, 2H, H4,4′-pz), 3.76 (s,
2H, CH2–C6H4), 3.66 (d, 2H, 3J = 6.9 Hz, CH2). 13C-NMR
(100.62 MHz, CDCl3, ppm) δ 51.1 (CH2), 52.8 (CH2–C6H4), 75.2
(C(pz)2), 92.6 (C4–C6H4), 106.9 (C4-pz), 129.1 (C3,5–C6H4), 120.1
(C2,6–C6H4), 137.7 (C5-pz), 139.6 (C1–C6H4), 140.5 (C3-pz). MS
(FAB+): m/z 394 [M + H]+, 416 [M + Na]+. Elemental analysis
(%): calc. for C15H16N5I: C 45.81, H 4.10, N 17.80; found:
C 45.66, H 4.23, N 17.69.
[Mn(CO)3(bpeaNHCH2C6H5)]PF6 (17). Yellow solid, yield: 69%
(283 mg, 0.51 mmol). IR (ATR, ν/cm−1): 3307 (m, N–H), 2045
(s, CuO), 1939 (s, CuO), 1604, 1519, 1450, 1315. 1H-NMR
General procedure for the synthesis of manganese(I) tricar-
bonyl-bpea complexes (14–18). To a stirred solution of manga-
nese pentacarbonyl bromide (206 mg, 0.75 mmol) in
anhydrous acetone (45 mL) was added a solution of 9 (200 mg,
0.75 mmol for 14), 10 (293 mg, 0.75 mmol for 15), 11 (217 mg,
0.75 mmol for 16), 12 (200 mg, 0.75 mmol for 17), or 13
(295 mg, 0.75 mmol for 18) in anhydrous acetone (5 mL) and
the mixture was heated to reflux for 5 h under a dinitrogen
atmosphere with exclusion of light. The solvent was then
removed under vacuum. The yellow residue was washed with
diethylether, redissolved in methanol (10 mL) and an aqueous
solution of potassium hexafluorophosphate (1.2 equiv.) was
added. The precipitated product was filtered off, washed with
water and diethylether, and dried under vacuum.
3
(400 MHz, acetone-d6, ppm): δ 8.51 (d, 1H, J = 2.3 Hz, H3′-pz),
8.39 (d, 1H, 3J = 2.3 Hz, H3-pz), 8.33 (d, 1H, 3J = 2.7 Hz, H5′-pz),
3
8.27 (d, 1H, J = 2.7 Hz, H5-pz), 7.41 (m, 6H, CH(pz)2, C6H5),
3
3
6.74 (t, 1H, J = 2.5 Hz, H4-pz), 6.66 (t, 1H, J = 2.5 Hz, H4′-pz),
3
5.59 (s, br, 1H, NH), 4.69 (dd, 1H, J = 13.4, 4.1 Hz, Hb′–CH2–
C6H5), 4.20 (dd, 1H, 3J = 13.4, 10.1 Hz, Hb–CH2–C6H5), 3.52
3
3
(ddd, 1H, J = 13.9, 7.8, 4.0 Hz, Ha′–CH2), 3.12 (ddd, 1H, J =
13.9, 8.9, 1.4 Hz, Ha–CH2). 13C-NMR (100.62 MHz, acetone-d6,
ppm) δ 50.3 (CH2–N), 62.3 (CH2–C6H5), 68.7 (C(pz)2), 108.3
(C4-pz), 108.4 (C4′-pz), 128.5, 128.7, 129.6, 134.3 (C5-pz), 134.7
(C5′-pz), 135.8 (C1–C6H5), 147.0 (C3-pz), 147.1 (C3′-pz). MS
(ESI+, CH3OH): m/z 405.92 [M − PF6]+. Elemental analysis (%):
calc. for C18H17F6MnN5O3P: C 39.22, H 3.11, N 12.70; found:
C 39.21, H 3.27, N 12.76.
[Mn(CO)3(bpeaNvCHC6H5)]PF6 (14). Yellow solid, yield 64%
(263 mg, 0.48 mmol). IR (ATR, ν/cm−1): 3135 (w), 2962 (w),
2040 (s, CuO), 1932 (s, CuO), 1625 (s, imine, CvN), 1414,
[Mn(CO)3(bpeaNHCH2C6H4I)]PF6 (18). Yellow solid, yield: 43%
(218 mg, 0.32 mmol). IR (ATR, ν/cm−1): 3275 (m, N–H), 3139
(w), 2962 (w), 2041 (s, CuO), 1928 (s, CuO), 1520, 1443, 1307,
1
1261, 1102, 1065. H-NMR (400 MHz, acetone-d6, ppm): δ 9.41
3
(s, 1H, CHvN), 8.45 (d, 2H, J = 2.7 Hz, H5,5′-pz), 8.38 (d, 2H,
1
3J = 2.3 Hz, H3,3′-pz), 7.81 (t, 1H, 3J = 2.5 Hz, CH(pz)2), 7.66 (m,
1261, 1148, 1065. H-NMR (400 MHz, acetone-d6, ppm): δ 8.51
(d, 1H, 3J = 2.2 Hz, H3′-pz), 8.40 (d, 1H, 3J = 2.2 Hz, H3-pz), 8.32
(d, 1H, 3J = 2.3 Hz, H5′-pz), 8.27 (d, 1H, 3J = 2.4 Hz, H5-pz), 7.76
(d, 2H, 3J = 8.4 Hz, H2,6–C6H4), 7.43 (d, 1H, 3J = 3.1 Hz, CH
3
3
5H, C6H5), 6.75 (t, 2H, J = 2.5 Hz, H4,4′-pz), 4.67 (dd, 2H, J =
2.6 Hz, 2.1 Hz, CH2). 13C-NMR (100.62 MHz, acetone-d6, ppm)
δ 63.0 (CH2), 69.7 (C(pz)2), 109.6 (C4-pz), 129.0 (C4–C6H5),
129.7 (C3,5–C6H5), 132.3 (C2,6–C6H5), 135.8 (C5-pz), 136.9 (C1–
C6H5), 146.8 (C3-pz), 181.6 (CvN). MS (ESI+, MeOH): m/z
3
3
(pz)2), 7.30 (d, 2H, J = 8.3 Hz, H3,5–C6H4), 6.74 (t, 1H, J = 2.5
3
Hz, H4′-pz), 6.67 (t, 1H, J = 2.5 Hz, H4-pz), 5.60 (s, 1H, NH),
4.68 (dd, 1H, J = 13.5, J = 4.0 Hz, Hb′–CH2–C6H4), 4.21 (dd,
1H, J = 13.5, J = 10.2 Hz, Hb–CH2–C6H4), 3.55 (ddd, 1H, J =
3
2
403.82 [M
−
PF6]+. Elemental analysis (%): calc. for
3
3
3
C18H15F6MnN5O3P·H2O: C 38.11, H 3.02, N 12.35; found:
3
2
3
3
13.7, J = 7.8, J = 3.8 Hz, Ha′–CH2), 3.12 (dd, 1H, J = 13.7, J =
8.9 Hz, Ha–CH2). 13C-NMR (100.62 MHz, acetone-d6, ppm)
δ 51.2 (CH2), 62.4 (CH2–C6H4), 69.6 (C(pz)2), 94.8 (C4–C6H4),
109.3 (C4-pz), 132.7 (C3–C6H4), 135.2 (C5-pz), 135.7 (C5′-pz),
136.6 (C1–C6H4), 138.7 (C2,6–C6H4), 148.0 (C3-pz), 148.1
(C3′-pz). MS (ESI+, CH3OH): m/z 531.77 [M − PF6]+. Elemental
analysis (%): calc. for C18H17F6MnN5O3P: C 31.93, H 2.38,
N 10.34; found: C 32.01, H 2.31, N 10.46.
C 38.36, H 2.83, N 11.99.
[Mn(CO)3(bpeaNvCHC6H4I)]PF6 (15). Yellow solid, yield 43%
(216 mg, 0.32 mmol). IR (ATR, ν/cm−1): 3135 (w), 3021 (w),
2043 (s, CuO), 1931 (s, CuO), 1624 (s, imine, CvN), 1412,
1295, 1103. 1H-NMR (400 MHz, acetone-d6, ppm): δ 9.34 (s,
3
3
1H, CHvN), 8.43 (d, 2H, J = 2.7 Hz, H5,5′-pz), 8.38 (d, 2H, J =
3
2.2 Hz, H3,3′-pz), 8.03 (d, 2H, J = 4.7 Hz, H3,5–C6H4), 7.76 (t,
3
3
1H, J = 2.7 Hz, CH(pz)2), 7.52 (d, 2H, J = 8.1 Hz, H2,6–C6H4),
8674 | Dalton Trans., 2014, 43, 8664–8678
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Myoglobin assay
was then measured at 510 nm (ε = 11 100 M−1 cm−1). A510 was
kept within the range of 0.4–1.0. The photon flux of the
In a quartz cuvette, a solution of horse skeletal muscle myoglo-
bin in 0.1 M phosphate buffer (PBS, pH 7.4) was degassed by
bubbling with dinitrogen and reduced by addition of sodium
dithionite (100 mM, 100 μL) in PBS buffer (0.1 M, pH 7.4) to a
total volume of 614 µL. To this solution, 10 μL of 14–18 in
dimethylsulfoxide was added followed by PBS to give a total
volume of 1000 µL with final concentrations of 10 µM of
CORM, 10 mM of sodium dithionite, 60 μM of myoglobin with
A557 < 1. The solutions were freshly prepared for both the dark
stability measurements and the photoactivation experiments.
Illumination was carried out under dinitrogen either using a
UV/Vis hand lamp (365 nm, UVIlite LF-206LS, 6 W, UVItec Ltd,
365 nm UV hand lamp was then calculated using Φ365 nm
=
1.21 following the equation:73
ΔAV110ꢀ3V3
Φp ¼
Φλε510V2 t
Cells and cell culture conditions
Human umbilical vein endothelial cells (HUVECs) were
received in collaboration with the Institute of Transfusion
Medicine and Immunology, Medical Faculty Mannheim, Uni-
versity of Heidelberg. Permission for isolation and propagation
of endothelial cells from umbilical cords for research purposes
was granted by the local ethics committee of the clinical
faculty Mannheim, University of Heidelberg with informed
consent in writing. HUVECs were isolated from fresh umbilical
cords as described previously.74 The cells were cultured in
endothelial cell growth medium (EGM) (Promocell, Heidel-
berg, Germany) in T25 flasks (Greiner, Frickenhausen,
Germany) coated with gelatin (1%). Confluent monolayers
were passaged by trypsin/EDTA (Sigma-Aldrich, St. Louis, MO).
Characterization of endothelial cells was performed on the
basis of a positive uptake of acetylated LDL, Factor VIII related
antigen, and PECAM (CD31) expression, and a negative stain-
ing for alpha smooth muscle actin.
Cambridge, UK) or
a
custom-built LED light source
(407–412 nm wavelength range, 5 mm round type UV-LEDs,
model YDG-504VC, Kingbright Elec. Co., Taipei, Taiwan, http://
www.kingbright.com, part no. 181000-05), both positioned
perpendicular to the cuvette at a distance of 3 cm. The illumi-
nation was interrupted in regular intervals to record UV/Vis
spectra on an Agilent 8453 UV/Vis diode array spectro-
photometer. As a dark control, measurements were carried
out by using the automated spectrometer software for a pre-
defined period of time (14 h). All experiments were carried out
in triplicate.
Photolysis experiments monitored by solution IR spectroscopy
An IR flow-cell holder equipped with calcium fluoride In vitro detection of CO release using the COP-1 fluorogenic
windows and a Teflon spacer was filled with a freshly prepared switch-on probe
solution of 14 or 17 in DMSO (9 mM). Illuminations were then
carried out using an UV/Vis hand lamp (365 nm, UVIlite
LF-206LS, 6 W, UVItec Ltd, Cambridge, UK) with the flow-
cell positioned at a distance of 3 cm. The illumination was
(a) In phosphate-buffered saline. On a 96-well microtiter
plate, phosphate-buffered saline (PBS, 100 µL) containing
different final concentrations of compound 17 (3–100 µM) dis-
solved in DMSO was added to each well and then, COP-1 dis-
interrupted in regular intervals and IR spectra recorded on a
solved in PBS was added to a final concentration of 10 µM.
Jasco FT/IR-4100 spectrometer.
The plate was subsequently exposed for 30 min to UV light at
365 nm from a 6 W lamp (UVITEC Cambridge, UK) positioned
parallel to the plate at a distance of 3 cm. Meanwhile, another
Ferrioxalate actinometry
Ferrioxalate actinometry was used to determine the photon plate was prepared under the same experimental conditions
flux of the 365 nm UV hand lamp because of its sensitivity, but kept in the dark for 30 min. The fluorescence was
wide spectral range including the ultraviolet, and the ease of measured immediately afterwards on an Infinite M200 reader
use.71,72 The whole ferrioxalate actinometry procedure includ- (Tecan, Crailsheim, Germany) using the I-control v9 software.
ing the preparation of solutions was carried out under red The fluorescence intensity was assessed at excitation and emis-
safe-light. The concentration of ferrous iron formed was deter- sion wavelengths of 475 and 510 nm, respectively. The exper-
mined spectrophotometrically by complexation with 1,10-phe- iments along with the controls were performed in
nanthroline (phen) to give the colored tris-phenanthroline quadruplicate and the data are expressed as mean fluorescence
complex [Fe(phen)3]2+ at λmax = 510 nm. Thus, in a 1 cm intensity SD.
quartz cell, potassium ferrioxalate (0.006 M) in sulfuric acid
(b) In cell culture. HUVECs were seeded in 96-well plates
(0.05 M, 3 mL) as the chemical actinometer was illuminated (Greiner, Frickenhausen, Germany) at a concentration of 105
with a 365 nm UV hand lamp (6 W, UVITEC, UK) positioned cells per well. After 1 d of seeding, the supernatant was aspi-
3 cm from the cuvette under efficient stirring. The illuminated rated and replaced by a freshly prepared solution of compound
solution (1 mL) was mixed with 0.1% 1,10-phenanthroline in 17 in DMSO diluted with PBS to a final concentration of
water (1 mL) and aqueous sodium acetate buffer (0.5 mL, 1 M, 100 µM and COP-1 dissolved in PBS was added to a final con-
pH 3.5), and then further diluted to 10 mL with water. A refer- centration of 10 µM. As negative controls, pure PBS buffer,
ence sample was prepared in the same way but without the COP-1 (10 µM), 17 (100 µM) and sodium hexafluorophosphate
illumination. Both solutions were placed in the dark (about (100 µM) were also included in the experiments. One plate was
1 h) to allow the complexation to complete. The absorbance subsequently exposed for 30 min to UV light at 365 nm from a
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8664–8678 | 8675

View Article Online
Paper
Dalton Transactions
6 W lamp (LF-206-LS, UVITEC, Cambridge, UK) positioned par- function of organometallic compounds”. The stay of G.A. in
allel to the plate at a distance of 3 cm. Meanwhile, another Germany was supported by the DAAD with an IAESTE fellow-
plate was prepared under the same experimental conditions ship. We would like to thank Manuela Winter and Timo Boller-
but kept in the dark for 30 min. After exposure, the whole mann for help with some of the X-ray structure analysis.
supernatant was transferred to a fresh plate to measure the
fluorescence intensity in the fluid phase. The cell-containing
plates were washed with PBS (2×) and after the last wash,
100 µL of PBS was added to each well. The fluorescence inten-
References
sity associated with the fluid phase and the cells was measured
separately on an Infinite M200 reader (Tecan, Crailsheim,
Germany) using the I-control v9 software. The fluorescence
intensity was assessed at an excitation and emission wave-
length of 475 and 510 nm, respectively. The experiments along
with the controls were performed in quadruplicate and the
data are expressed as mean fluorescence intensity SD.
1 L. Wu and R. Wang, Pharmacol. Rev., 2005, 57, 585–630.
2 H. P. Kim, S. W. Ryter and A. M. K. Choi, Annu. Rev. Phar-
macol. Toxicol., 2006, 46, 411–449.
3 M. Kajimura, R. Fukuda, R. M. Bateman, T. Yamamoto and
M. Suematsu, Antioxid. Redox Signaling, 2010, 13, 157–192.
4 L. Rochette, Y. Cottin, M. Zeller and C. Vergely, Pharmacol.
Ther., 2013, 137, 133–152.
5 S. W. Ryter, J. Alam and A. M. K. Choi, Physiol. Rev., 2006,
86, 583–650.
Density functional theory calculations
6 N. G. Abraham and A. Kappas, Pharmacol. Rev., 2008, 60,
79–127.
7 T. Matsui, M. Iwasaki, R. Sugiyama, M. Unno and M. Ikeda-
Saito, Inorg. Chem., 2010, 49, 3602–3609.
8 T. Matsui, M. Unno and M. Ikeda-Saito, Acc. Chem. Res.,
2010, 43, 240–247.
9 R. Motterlini and L. E. Otterbein, Nat. Rev. Drug Discovery,
2010, 9, 728–743.
10 B. Wegiel, D. W. Hanto and L. E. Otterbein, Trends Mol.
Med., 2013, 19, 3–11.
11 L. S. Nobre, J. D. Seixas, C. C. Romao and L. M. Saraiva,
Antimicrob. Agents Chemother., 2007, 51, 4303–4307.
12 J. L. Wilson, H. E. Jesse, R. K. Poole and K. S. Davidge,
Curr. Pharm. Des., 2012, 13, 760–768.
DFT calculations were carried out on the Linux cluster of the
Leibniz-Rechenzentrum (LRZ) in Munich with ORCA version
2.8,75 using the BP86 functional with the resolution-of-the-
identity (RI) approximation, a def2-TZVP/def2-TZVP/J basis
set,76,77 the tightscf and grid4 options, and the COSMO sol-
vation model with dimethylsulfoxide as the solvent for geome-
try optimizations and subsequent calculation of vibrational
frequencies to characterize the structures obtained as minima
by inspection for the absence of imaginary modes. Reported
vibrational modes are scaled with a factor of 1.036 relative to
the calculated values. The TDDFT calculations employed the
B3LYP functional with the same basis set and settings as
above using the RIJCOSX keyword. The first 45 singlet excited
states were calculated (nroots 45).
13 C. C. Romao, W. A. Blättler, J. D. Seixas and
G. J. L. Bernardes, Chem. Soc. Rev., 2012, 41, 3571–3583.
14 B. E. Mann, in Top. Organomet. Chem., ed. N. Metzler-Nolte
and G. Jaouen, Springer, Berlin, 2010, vol. 32, pp. 247–285.
15 B. E. Mann, Organometallics, 2012, 31, 5728–5735.
16 F. Zobi, Future Med. Chem., 2013, 5, 175–188.
17 U. Schatzschneider, Br. J. Pharmacol., 2014, DOI: 10.1111/
bph.12688.
18 R. Motterlini, P. Sawle, S. Bains, J. Hammad, R. Alberto,
R. Foresti and C. J. Green, FASEB J., 2004, 18, 284–286.
19 T. S. Pitchumony, B. Spingler, R. Motterlini and R. Alberto,
Chimia, 2008, 62, 277–279.
20 T. S. Pitchumony, B. Spingler, R. Motterlini and R. Alberto,
Org. Biomol. Chem., 2010, 8, 4849–4954.
21 L. A. P. Antony, T. Slanina, P. Sebej, T. Solomek and
P. Klan, Org. Lett., 2013, 15, 4552–4555.
22 P. Peng, C. Wang, Z. Shi, V. K. Johns, Y. Ma, J. Oyer,
A. Copik, R. Igarashi and Y. Liao, Org. Biomol. Chem., 2013,
11, 6671–6674.
X-ray crystallography
A single crystal of 15, 17, or 18 was immersed in a film of per-
fluoropolyether oil, mounted on a glass fiber and transferred
to a stream of cold nitrogen of the diffractometer. Diffraction
data were collected on an Oxford Diffraction system using a
Sapphire2-CCD detector for 17 and a Multiwire Proportional
counter for 15 and 18 at 110–113 K and using graphite mono-
chromated Mo-Kα radiation (λ = 0.71073 Å). The final cell con-
stants were obtained from a least squares fit of a subset of a
few thousand strong reflections. The program CrysAlis CCD
was used for data collection. CrysAlis RED and CrysAlis CCD
software were used for cell refinements and data reduction.
The program SADABS was used to account for the absorp-
tion.78 The OLEX 2 program was used for refinement and
artwork of the structure.79 Crystallographic data (excluding
structure factors) have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publications
No. CCDC 976263, 976264, and 976267.
23 T. R. Johnson, B. E. Mann, I. P. Teasdale, H. Adams,
R. Foresti, C. J. Green and R. Motterlini, Dalton Trans.,
2007, 1500–1508.
24 S. Romanski, B. Kraus, U. Schatzschneider, J. Neudörfl,
S. Amslinger and H.-G. Schmalz, Angew. Chem., Int. Ed.,
2011, 50, 2392–2396.
Acknowledgements
This work was supported by the Deutsche Forschungsge-
meinschaft (DFG) within the project FOR 630 “Biological
8676 | Dalton Trans., 2014, 43, 8664–8678
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
25 S. Romanski, B. Kraus, M. Guttentag, W. Schlundt, 49 P. Govender, S. Pai, U. Schatzschneider and G. Smith,
H. Rücker, A. Adler, J.-M. Neudörfl, R. Alberto, Inorg. Chem., 2013, 52, 5470–5478.
S. Amslinger and H.-G. Schmalz, Dalton Trans., 2012, 41, 50 N. E. Brückmann, M. Wahl, G. J. Reiß, M. Kohns,
13862–13875.
W. Wätjen and P. C. Kunz, Eur. J. Inorg. Chem., 2011, 4571–
26 S. Romanski, H. Rücker, E. Stamellou, M. Guttentag,
4577.
J. Neudörfl, R. Alberto, S. Amslinger, B. Yard and 51 U. Hasegawa, A. J. van der Vlies, E. Simeoni, C. Wandrey
H.-G. Schmalz, Organometallics, 2012, 31, 5800–5809.
and J. A. Hubbell, J. Am. Chem. Soc., 2010, 132, 18273–
27 S. Botov, E. Stamellou, S. Romanski, M. Guttentag,
18280.
R. Alberto, J. Neudörfl, B. Yard and H.-G. Schmalz, Organo- 52 G. Dördelmann, H.
metallics, 2013, 32, 3587–3594. U. Schatzschneider, Inorg. Chem., 2011, 50, 4362–4367.
28 P. C. Kunz, H. Meyer, J. Barthel, S. Sollazzo, A. M. Schmidt 53 G. Dördelmann, T. Meinhardt, T. Sowik, A. Krüger and
Pfeiffer, A. Birkner and
and C. Janiak, Chem. Commun., 2013, 49, 4896–4898.
29 U. Schatzschneider, Eur. J. Inorg. Chem., 2010, 1451–1467.
30 U. Schatzschneider, Inorg. Chim. Acta, 2011, 374, 19–23.
31 R. D. Rimmer, A. E. Pierri and P. C. Ford, Coord. Chem.
Rev., 2012, 256, 1509–1519.
U. Schatzschneider, Chem. Commun., 2012, 48, 11528–
11530.
54 M. Ma, H. Noei, B. Mienert, J. Niesel, E. Bill, M. Muhler,
R. A. Fischer, Y. Wang, U. Schatzschneider and N. Metzler-
Nolte, Chem.–Eur. J., 2013, 19, 6785–6790.
32 V. P. Lorett-Velasquez, T. M. A. Jazzazi, A. Malassa, 55 C. R. Bertozzi and E. M. Sletten, Angew. Chem., Int. Ed.,
H. Görls, G. Gessner, S. H. Heinemann and
M. Westerhausen, Eur. J. Inorg. Chem., 2012, 1072–1078.
33 R. Kretschmer, G. Gessner, H. Görls, S. H. Heinemann and
M. Westerhausen, J. Inorg. Biochem., 2011, 105, 6–9.
34 M. A. Gonzalez, N. L. Fry, R. Burt, R. Davda, A. Hobbs and
P. K. Mascharak, Inorg. Chem., 2011, 50, 3127–3134.
35 C. S. Jackson, S. Schmitt, Q. P. Dou and J. J. Kodanko,
Inorg. Chem., 2011, 50, 5336–5338.
36 H. Pfeiffer, T. Sowik and U. Schatzschneider, J. Organomet.
Chem., 2013, 734, 17–24.
37 R. D. Rimmer, H. Richter and P. C. Ford, Inorg. Chem.,
2010, 49, 1180–1185.
2009, 48, 6974–6998.
56 R. K. V. Lim and Q. Lin, Chem. Commun., 2010, 46, 1589–
1600.
57 J. C. Jewett and C. R. Bertozzi, Chem. Soc. Rev., 2010, 39,
1272–1279.
58 P. Thirumurugan, D. Matosiuk and K. Jozwiak, Chem. Rev.,
2013, 113, 4905–4979.
59 C. Bischof, T. Joshi, A. Dimri, L. Spiccia and
U. Schatzschneider, Inorg. Chem., 2013, 52, 9297–9308.
60 B. W. Michel, A. R. Lippert and C. J. Chang, J. Am. Chem.
Soc., 2012, 134, 15668–15671.
61 J. Wang, J. Karpus, B. S. Zhao, Z. Luo, P. R. Chen and
C. He, Angew. Chem., Int. Ed., 2012, 51, 9652–9656.
62 L. Yuan, W. Lin, L. Tan, K. Zheng and W. Huang, Angew.
Chem., Int. Ed., 2013, 52, 1628–1630.
38 A. E. Pierri, A. Pallaoro, G. Wu and P. C. Ford, J. Am. Chem.
Soc., 2012, 134, 18197–18200.
39 J. S. Ward, J. M. Lynam, J. W. B. Moir, D. E. Sanin,
A. P. Mountford and I. J. S. Fairlamb, Dalton Trans., 2012, 63 A. J. Atkin, J. M. Lynam, B. E. Moulton, P. Sawle,
41, 10514–10517.
R. Motterlini, N. M. Boyle, M. T. Pryce and I. J. S. Fairlamb,
Dalton Trans., 2011, 40, 5755–5761.
64 S. McLean, B. E. Mann and R. K. Poole, Anal. Biochem.,
2012, 427, 36–40.
40 M. A. Gonzalez, M. A. Yim, S. Cheng, A. Moyes, A. J. Hobbs
and P. K. Mascharak, Inorg. Chem., 2012, 51, 601–608.
41 M. A. Gonzalez, S. J. Carrington, N. L. Fry, J. L. Martinez
and P. K. Mascharak, Inorg. Chem., 2012, 51, 11930– 65 R. Motterlini, J. E. Clark, R. Foresti, P. Sarathchandra,
11940.
B. E. Mann and C. J. Green, Circ. Res., 2002, 90, e17–
e24.
66 M. A. Gonzalez, M. A. Yim, S. Cheng, A. Moyes, A. J. Hobbs
and P. K. Mascharak, Inorg. Chem., 2011, 51, 601–608.
42 S. J. Carrington, I. Chakraborty and P. K. Mascharak, Chem.
Commun., 2013, 49, 11254–11256.
43 M. A. Gonzalez, S. J. Carrington, I. Chakraborty,
M. M. Olmstead and P. K. Mascharak, Inorg. Chem., 2013, 67 W. Huber, R. Linder, J. Niesel, U. Schatzschneider,
52, 11320–11331.
44 W. Huber, R. Linder, J. Niesel, U. Schatzschneider,
B. Spingler and P. C. Kunz, Eur. J. Inorg. Chem., 2012, 2012,
3140–3146.
B. Spingler and P. C. Kunz, Eur. J. Inorg. Chem., 2012, 3140– 68 D. L. Reger, J. D. Elgin, R. F. Semeniuc, P. J. Pellechia and
3146. M. D. Smith, Chem. Commun., 2005, 4068–4070.
45 F. Mohr, J. Niesel, U. Schatzschneider and C. W. Lehmann, 69 D. L. Reger, B. Reinecke, M. D. Smith and R. F. Semeniuc,
Z. Anorg. Allg. Chem., 2012, 638, 543–546. Inorg. Chim. Acta, 2009, 362, 4377–4388.
46 J. Niesel, A. Pinto, H. W. Peindy N’Dongo, K. Merz, I. Ott, 70 G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb,
R. Gust and U. Schatzschneider, Chem. Commun., 2008,
1798–1800.
A. Nudelman, B. M. Stoltz, J. E. Bercaw and K. I. Goldberg,
Organometallics, 2010, 29, 2176–2179.
47 H. Pfeiffer, A. Rojas, J. Niesel and U. Schatzschneider, 71 C. A. Parker, Proc. R. Soc. London, Ser. A, 1953, 220, 104–
Dalton Trans., 2009, 4292–4298. 116.
48 J. B. Matson, M. J. Webber, V. K. Tamboli, B. Weber and 72 C. G. Hatchard and C. A. Parker, Proc. R. Soc. London, Ser.
S. I. Stupp, Soft Matter, 2012, 8, 6689–6692.
A, 1956, 235, 518–536.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8664–8678 | 8677

View Article Online
Paper
Dalton Transactions
73 H. J. Kuhn, S. E. Braslavskyi and R. Schmidt, Pure Appl. 77 F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005,
Chem., 1989, 61, 187–210. 7, 3297–3305.
74 E. A. Jaffe, R. L. Nachman, C. G. Becker and C. R. Minick, 78 G. M. Sheldrick, Universität Göttingen, Germany,
J. Clin. Invest., 1973, 52, 2745–2756.
75 F. Neese, WIREs Comput. Mol. Sci., 2011, 2, 73–78.
76 A. Schaefer, H. Horn and R. Ahlrichs, J. Chem. Phys., 1992,
97, 2571–2577.
2006.
79 O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard
and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339–
341.
8678 | Dalton Trans., 2014, 43, 8664–8678
This journal is © The Royal Society of Chemistry 2014