Stability and toxicity of tris-tolyl bismuth(v) dicarboxylates and their biological activity towards Leishmania major

Article abstract of DOI:10.1039/c5dt03335g

A series of 29 tris-tolyl bismuth(v) di-carboxylato complexes of composition [Bi(Tol)₃(O₂CR)₂] involving either ortho, meta or para substituted tolyl ligands have been synthesized and characterised. Of these 15 were assessed for their toxicity towards Leishmania promastigotes and human fibroblast cells, with ten then being subsequently assessed against parasite amastigotes. The carboxylate ligands are drawn from a series of substituted and biologically relevant benzoic acids which allow a comparison with earlier studies on [BiPh₃(O₂CR)₂] and analogous Sb(v) [SbAr₃(O₂CR)₂] (Ar = Ph and Tol) complexes. Twelve complexes have been structurally characterized by single crystal X-ray diffraction and shown to adopt a typical trigonal bipyramidal geometry in which the three tolyl ligands occupy the equatorial plane. NMR studies on two illustrative examples indicate that the complexes are stable in D₂O and DMSO but only have a half-life of 1.2 hours in culture medium, with glucose being a contributing factor in decomposition and reduction to Bi(Tol)₃. Despite their short lifetime many complexes show significant toxicity towards promastigotes at low concentration (<6 μM) and at that concentration provide for good selectivity indices (parasite vs. mammalian cells), for example 114 for [Bi(o-Tol)₃(O₂CC₆H₃(2-OH,5-C₆H₃(2,4-F₂)))₂] and 838 for [Bi(m-Tol)₃(O₂CC₆H₄(2-OAc))₂]. Best activity and selectivity is observed with complexes containing o- and m-tolyl ligands, and it appears the primary influence on fibroblast toxicity is the Ar ligand while the carboxylate influences promastigote toxicity. The complexes are less effective in vitro against the parasite amastigotes, where longer incubation times and harsher chemical and biological environments are encountered in the assay. Nevertheless, there were some statistically relevant differences at 1 μM against the positive controls with the best performing complexes being [Bi(o-Tol)₃(O₂CC₆H₄(2-EtO))₂] and [Bi(m-Tol)₃(O₂CC₆H₄(2-OAc))₂].

Full text of DOI:10.1039/c5dt03335g

Dalton
Transactions
View Article Online
View Journal
PAPER
Stability and toxicity of tris-tolyl bismuth(V)
dicarboxylates and their biological activity towards
Cite this: DOI: 10.1039/c5dt03335g
Leishmania major†
a
a
b,c
a
Yih Ching Ong, Victoria L. Blair, Lukasz Kedzierski, Kellie L. Tuck and
a
Philip C. Andrews*
A series of 29 tris-tolyl bismuth(V) di-carboxylato complexes of composition [Bi(Tol)
either ortho, meta or para substituted tolyl ligands have been synthesized and characterised. Of these
5 were assessed for their toxicity towards Leishmania promastigotes and human fibroblast cells, with ten
3 2 2
(O CR) ] involving
1
then being subsequently assessed against parasite amastigotes. The carboxylate ligands are drawn from a
series of substituted and biologically relevant benzoic acids which allow a comparison with earlier studies
3 2 2 3 2 2
on [BiPh (O CR) ] and analogous Sb(V) [SbAr (O CR) ] (Ar = Ph and Tol) complexes. Twelve complexes
have been structurally characterized by single crystal X-ray diffraction and shown to adopt a typical tri-
gonal bipyramidal geometry in which the three tolyl ligands occupy the equatorial plane. NMR studies on
two illustrative examples indicate that the complexes are stable in D
of 1.2 hours in culture medium, with glucose being a contributing factor in decomposition and reduction
to Bi(Tol) . Despite their short lifetime many complexes show significant toxicity towards promastigotes
at low concentration (<6 μM) and at that concentration provide for good selectivity indices (parasite
vs. mammalian cells), for example 114 for [Bi(o-Tol) (O CC (2-OH,5-C (2,4-F ))) ] and 838 for
Bi(m-Tol) (O CC (2-OAc)) ]. Best activity and selectivity is observed with complexes containing
2
O and DMSO but only have a half-life
3
3
2
6
H
3
6
H
3
2
2
[
3
2
6
H
4
2
o- and m-tolyl ligands, and it appears the primary influence on fibroblast toxicity is the Ar ligand while the
carboxylate influences promastigote toxicity. The complexes are less effective in vitro against the parasite
amastigotes, where longer incubation times and harsher chemical and biological environments are
encountered in the assay. Nevertheless, there were some statistically relevant differences at 1 μM against
Received 28th August 2015,
Accepted 24th September 2015
DOI: 10.1039/c5dt03335g
the positive controls with the best performing complexes being [Bi(o-Tol)
[Bi(m-Tol) (O CC (2-OAc)) ].
3 2 6 4 2
(O CC H (2-EtO)) ] and
www.rsc.org/dalton
3
2
6
H
4
2
5
years with over 20 000 deaths annually, and 310 million
Introduction
2
people are currently at risk of infection.
Leishmaniasis is a group of diseases caused by the Leishmania
For more than 70 years the frontline treatment for leishma-
parasite and transmitted via infected female sandflies. Visceral niasis has been pentavalent antimonals; sodium stibogluco-
3
leishmaniasis (VL), the most serious form of the disease, nate (Pentostam; SSG)
and meglumine antimoniate
4
affects the internal organs and without treatment is normally (Glucantime), which are readily available and cost-effective.
fatal within two years. The World Health Organisation esti- However, they have severe and toxic side effects, and due to
1
mates that 300 000 cases of VL have been reported in the past their parenteral mode of administration over a 28-day period
5
compliance can be low and areas of drug resistance have
6
,7
appeared, such as Bihar in India.
Alternative treatments
8
9
such as liposomal Amphotericin B and Miltefosine are avail-
able, however their prohibitive cost makes access to the drug
difficult for the majority of patients. Furthermore, liposomal
Amphotericin B is still administered intravenously, and Milte-
fosine, the only orally available drug has a narrow therapeutic
a
School of Chemistry, Monash University, Clayton, Melbourne, VIC 3800, Australia.
E-mail: phil.andrews@monash.edu
b
Walter and Eliza Institute of Medical Research, Parkville, Melbourne, VIC 3052,
Australia
c
Department of Medical Biology, University of Melbourne, Parkville 3010, Australia
9
window and is teratogenic. These challenges mean that there
†
Electronic supplementary information (ESI) available. CCDC 1420787–1420797.
For ESI and crystallographic data in CIF or other electronic format see DOI:
0.1039/c5dt03335g
is a need for alternative, low cost, safe oral treatments with
reduced toxicity and side effects.
1
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 1 Tris-tolylbismuth(V) dicarboxylato complexes 1–3, 5–11 and Bi(Tol)
3
, 4 and 12 studied for anti-leishmanial activity on promastigotes and
amastigotes.
Pentavalent antimony itself is relatively non-toxic and DMSO but to be unstable in culture medium, undergoing oxi-
1
8
serves as a pro-drug which is reduced by trypanothione, dative decomposition to reform and precipitate BiPh
3
.
gluthathione or thiol-dependent reductase, TDR1 at pH 5 to However, the decomposition process was slow enough that the
1
0,11
the active Sb(III) species within the macrophage.
to a decrease in buffering ability and imbalance in thiol redox
potential as Sb(III) induces the efflux of trypanothione and found that it was not those derived from SbPh
glutathione out of the cell and their respective disulfides to be the most promising, but complexes in which Ar = m- or
This leads complex was still able to exert substantial toxic effects.
In our extensive study on [SbAr (O CR) ] complexes we
which proved
3
2
2
3
1
2
13
accumulate inside cells. It is suggested that the redistribu- p-tolyl. Therefore, to establish a much closer comparison we
tion of the Sb(III) species out of the parasite is responsible for undertook an assessment of the tolyl based complexes
1
0
the toxic effects.
In seeking more lipophilic alternatives to SSG we have 29 organometallic tris-tolylbismuth(V) dicarboxylate complexes
shown that several organometallic Sb(V) dicarboxylato com- [Bi(Tol) (O CR) ] (where Ar = o-, m-, p-Tol) and their sub-
plexes of general form [SbAr (O CR) ] are promising leads, sequent solubility and biological activity against Leishmania
and here report on the synthesis and characterization of
3
2
2
3
2
2
being highly effective against both promastigotes and amasti- major promastigotes and human fibroblast cells. The most
gotes at very low concentrations (0.5–3.5 µM) while maintain- promising complexes were subsequently assessed on L. major
1
3
ing low toxicity towards human fibroblasts. Considering the amastigotes. Ten complexes and two starting compounds,
periodic relationship of antimony and bismuth, and the low shown in Fig. 1 below, were found to have good anti-leishma-
1
4
systemic toxicity of bismuth in humans, we extended our nial activity with low fibroblast toxicity.
1
5–17
18
explorations to both Bi(III)
and Bi(V) based compounds
as promising candidates for future anti-leishmanial drugs, tar-
geting decreased toxicity, reduced side effects, and greater
potential for oral delivery. To that end we recently reported the
Results and discussion
chemistry and biological activity of a series of [BiPh
3 2 2
(O CR) ]
complexes towards L. major promastigotes and human fibro- The target [Bi(Tol) (O CR) ] (o-, p- or m-tolyl) complexes were
3
2
2
1
8
blasts, showing that while [BiPh (O CR) ] complexes show formed through an oxidative addition reaction between
3
2
2
good activity against promastigotes, they were much less stable Bi(Tol) and the chosen benzoic acid promoted by hydrogen
3
1
3
than their Sb(V) analogues, and were also non-selectively peroxide in a 1 : 2 : 1 stoichiometric ratio in diethyl ether at
1
9
toxic against the fibroblasts, even at low concentrations. Using room temperature. After 30 minutes each reaction mixture
one complex, [BiPh (O CC (m-OH) ] as an illustrative was filtered and the resulting colourless solutions allowed
example, we showed the complexes to be stable in water and to stand overnight. This resulted in the isolation of a suite of
3
2
6
H
3
2 2
)
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Scheme 1 General synthesis of tris-aryl bismuth(V) dicarboxylato
complexes.
[
(
Bi(Tol)
Scheme 1).
The composition of the final products as [Bi(Tol)
3
(O
2
CR)
2
]
complexes as microcrystalline solids
3
(O
2
CR)
2
]
Fig. 2 Molecular structure of [Bi(o-Tol) (dif) ], 3 showing thermal ellip-
3
2
1
13
was confirmed through H and
melting point, FT-IR spectrometry, mass spectrometry and
elemental analysis. In addition, the diflunisal (difH) derived
C NMR spectroscopy, soids at 50% probability. Hydrogen atoms have been omitted for clarity.
Selected bond lengths(Å) and bond angles (°): Bi(1)–C(15) 2.208(2);
Bi(1)–C(1) 2.216(2); Bi(1)–C(8) 2.223(2); Bi(1)–O(4) 2.2764(15); Bi(1)–O(1)
.2936(15); Bi(1)–O(5) 3.0740(16); Bi(1)–O(2) 3.0813(17); C(15)–Bi(1)–
2
3 2
complex, [Bi(o-Tol) (dif) ], 3, was analysed by single crystal
C(1) 110.62(8);C(15)–Bi(1)–C(8) 116.29(8);C(1)–Bi(1)–C(8) 132.94(8);
O(4)–Bi(1)–O(5) 46.22(5); O(1)–Bi(1)–O(2) 46.02(5).
X-ray diffraction. Full analytical details for 1–3 and 5–11 are
given in the Experimental section while for all other complexes
are provided in the ESI† (Table 1).
Depending on the solubilities of the final products, NMR
spectra were collected in either CDCl or D -DMSO at room
3 6
X-Ray crystallography
temperature. The chemical shifts and integrals all corres- Of the 29 complexes synthesised crystals suitable for single
ponded with formation of the target [Bi(Tol) (O CR) ] com- crystal X-ray diffraction studies were obtained for 11 of them:
3
2
2
plexes. All complexes showed low frequency shifts in the ortho, [Bi(o-Tol)
meta and para proton resonances of the tolyl ligand compared Tol) (O CC
to the corresponding protons in Bi(Tol) . The –OH signal [Bi(m-Tol) (O
(2-Me,3-Cl)))
NMR spectra of the parent acids between [Bi(p-Tol) (O CC (2-OEt))
1.61–13.67 ppm, was also absent indicating deprotonation NO )) ] 27, [Bi(p-Tol) (O CC
and complexation of the carboxylates to the Bi atom.
Tol) (O CC (2-NHC (2-Me,3-Cl)))
Compared to the parent acids which have their carboxylic isostructural and discussion on 3 is provided here as an
3
(dif)
(2-EtO))
CC (2-OH,5-Cl))
] 22, [Bi(p-Tol)
26,
(2-OH,5-Cl))
] 31. The complexes are
2
] 3, [Bi(m-Tol)
] 8, [Bi(m-Tol)
] 19, [Bi(m-Tol)
(O CC (2-OMe))
[Bi(p-Tol) (O CC
] 28 and [Bi(p-
3
(O
2
CC
6
H
3
(2,5-OH))
(O CC (4-NO
(O CC
2
] 6, [Bi(m-
)) ] 9,
(2-
] 25,
(4-
H
H
3
2
6
4
2
3
2
6
4
2 2
H
H
3
3
2
6
3
2
3
2
6
4
corresponding to the –COOH of the carboxylic acid, observed NHC
6
H
3
2
3
2
6
H
4
2
1
in the
1
H
3
2
6
H
4
2
]
3
2
6
H
4
H
3
2
2
3
2
6
2
3
2
6
H
4
6
H
3
2
−
1
acid carbonyl stretches in the range of 1650–1700 cm , all the illustrative example. Figures depicting the solid-state structures
complexes show a shift to a lower wavenumber for their of the other complexes along with a summary of their bond
carboxylate carbonyl asymmetric and symmetric stretches to lengths and crystallographic data are provided as ESI.†
1
−1
1
540–1650 cm− and 1300–1420 cm . This indicates the
deprotonation of the –COOH functionality to form the [Bi(Tol)
CR)
] complex. The difference between the symmetric and shown in Fig. 2 the complex adopts a typical distorted trigonal
asymmetric shifts are less than 200 cm for all the complexes. bipyramidal geometry. The tolyl ligands sit in the equatorial
Suitable single crystals of [Bi(o-Tol)
from a THF/n-hexane mixture through slow evaporation. As
(dif) ], 3 were obtained
3 2
3
-
(O
2
2
−
1
2
0
According to Deacon and Phillips, this indicates that the plane with a propeller-like orientation, while the carboxylates
carboxylate ligands adopt a bidentate chelating mode to the sit in the axial positions. All the Bi–C bond lengths are similar;
bismuth centre, and this trend was observed also in the solid Bi(1)–C(15) 2.208(2), Bi(1)–C(1) 2.216(2), Bi(1)–C(8) 2.223(2).
1
8
state structures of the [BiPh
3
(O
2
CR)
2
] complexes.
However, one of the C–Bi–C trigonal angles is more obtuse
Table 1 Synthesis of [Bi(Tol)
3
(O
2 2
CR) ] complexes 1–3, 5–11 with their melting points and yields
Complex
Carboxylic acid (code)
Ar
R
Mp (°C)
Yield (%)
1
2
3
5
6
7
8
9
2-Ethoxybenzoic acid
4-Nitrobenzoic acid
Diflunisal (difH)
3,5-Dimethylbenzoic acid
3,5-Dihydroxybenzoic acid
2-Methoxybenzoic acid
2-Ethoxybenzoic acid
4-Nitrobenzoic acid
Flufenamic acid
o-Tol
o-Tol
o-Tol
m-Tol
m-Tol
m-Tol
m-Tol
m-Tol
m-Tol
m-Tol
2-EtOC
6
H
4
126–128
146–148
153–154
173–176
170 (dec.)
145–146
144–145
170–172
143–147
129–132
39.0
44.6
75.1
51.5
64.6
47.6
46.2
41.9
66.4
76.9
4-NO
2-OH,5-C
3,5-Me
3,5-(OH)
2-MeOC
2-EtOC
4-NO
2-NH(C
2-OAcC
2
C
6
H
4
H
H
6
3
(2,4-F
2
)C
6
H
3
2
C
6
3
2 6 3
C H
6 4
H
6
H
4
2
C
6
H
4
1
1
0
1
6
H
4
3
-3-CF )C
6 4
H
Aspirin (aspH)
6 4
H
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
1
8
than the rest (cf. 132.94(8) vs. 110.62(8) and 116.29(8)), leading stand the stability of the complex in culture medium. While
to a distortion towards square pyramidal based geometry. The the complex was found to be stable in water and DMSO,
oxygen atoms of the carboxylate moieties are bound to the decomposition was found to occur in DMEM culture media to
3
bismuth centre in a bidentate fashion confirming observation precipitate the starting reagent BiPh as a white precipitate.
in the IR spectrum. The second carboxyl (CvO) oxygen atoms, A subsequent NMR study showed that the complex was stable
O(2) and O(5) of the carboxylate ligands, have longer Bi–O dis- to phosphate (or other anion) exchange and did not react with
tances, 3.0740(16) Å and 3.0813(17) Å, as compared to the two the amino acids in solution, noting that the key amino acid
shorter Bi–O bonds 2.2764(15) Å and 2.2936(15) Å. For this primed for oxidation, cysteine, is already present in DMEM in
1
8
complex, the presence of an ortho-OH functionality on the its oxidised state.
phenyl ring of the carboxylate ligand forms a hydrogen bond Similar studies were therefore conducted on [Bi(o-
to the oxygen of the carboxylate, and may contribute to the Tol) (dif) ], 3 as a representative example of the Bi(V)-tolyl
3
2
longer Bi–O bond length.
This structure for the [Bi(Tol)
similar to that observed for the [BiPh (O CR) ] complexes.
series of complexes. To assess stability in cell culture medium,
H NMR spectral data was collected at t = 0 h and 48 h. Con-
ditions similar to those used in the biological assay were used,
1
3
(O
2
CR)
2
] analogues are
1
8
3
2
2
For example, the analogous [BiPh
distorted trigonal bi-pyramidal structure.
3
(dif)
2
] complex also shows a however the signal-to-noise ratio for the highest concentration
comparison used in the in vitro biological assay (100 μM) was too low to
A
between the phenyl and tolyl analogues shows that the bond provide any meaningful data and, as such, a sample concen-
lengths; Bi–C (2.204(8) Å, 2.217(12) Å) and Bi–O (2.253(6) Å, tration of 1.0 mM was used. The culture medium was freeze-
3
.041(7) Å), are similar and are only slightly shorter than those dried and reconstituted in D
2 2
O to minimise the H O signal.
in [Bi(Tol) (dif) ]; Bi–C (2.208(3), 2.2218(3), 2.227(3) Å) and Complex 3 was added to the culture medium and the resulting
3
2
Bi–O (2.227(2), 2.293(2), 3.073(2), 2.081(2) Å). This indicates mixture allowed to stand for 24 hours, resulting in the for-
that the presence of an additional o-Me group on the phenyl mation of a white precipitate. The supernatant was then
ring has little influence on the structure of the complex. Con- decanted off, and the white precipitate dissolved in d -DMSO
6
1
versely, the analogous Sb–C bonds in the Sb(V) complex and analysed by H NMR. By comparing the chemical shifts
[SbPh
3 2 2 6 4 3 2
(O CCH C H (m-CH )) ] are 2.0949(14), 2.0974(14) and produced by the unknown white precipitate, it was deduced
2
.1105(13) Å, and the Sb–O bonds are 2.0880(9) Å and 2.1108 that the precipitate was the starting reagent Bi(o-Tol) . (Fig. 3)
3
(10) Å, with the carboxylate ligands binding in a monodentate This was confirmed by a comparison of the melting point
manner.
which showed the unknown white precipitate (mp. 130 °C)
was consistent with Bi(o-Tol) (mp. 132 °C), as opposed to 3
3
Stability studies
(
mp. 154 °C). This indicated that over time the Bi(V) complex 3
To determine whether the complexes are stable to atmospheric was being reduced.
moisture and gases, all complexes were exposed to air and
To determine the rate of decomposition, a H NMR spec-
1
screened by melting point over a period of six months. There trum of 3 in culture medium was recorded every 30 min for
was no change in the observed melting points over this time, 10 hours. Using the integration of one chemical shift from the
indicating the complexes have a high degree of stability in the culture medium at 4.4 ppm as a standard, the decrease in inte-
solid-state. Short-term stability in solution was evaluated gration of the ortho proton in 3 was measured. It was found
1
through the H NMR spectrum of each complex being col- that the integration fell from 1.000 (t = 0 h) to 0.0245 after
lected and compared at t = 0 and 48 h in either CDCl
DMSO. No change in the chemical shifts was observed.
3
or D
6
-
4.5 hours and then to 0.000 after 5 hours. The plot, shown in
Fig. 4, followed a one-phase exponential decay (R = 0.9955)
2
In our earlier studies on [BiPh (O CR) ] complexes we used with a half-life of 1.2 hours. This indicated that even though
3
2
2
[
BiPh
3
(O
2 6 3 2
CC H (2,5-OH))
] as an example to study and under- the complexes have a short half-life, they have potent anti-
1
Fig. 3
3 2 3
H NMR spectra comparison of unknown white precipitate with [Bi(o-Tol) (dif) ], 3 & Bi(o-Tol) in the aromatic region. Coloured dots indi-
cate the chemical shifts for the respective tolyl protons.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
3 2 6
Glucose was added to [Bi(m-Tol) (asp) ] 11 in d -DMSO in
1
a stoichiometric ratio of 2 : 1, and the H NMR spectrum
recorded at t = 0, 48 h, and 1 week, at room temperature
(
20 °C). Control studies showed that when the individual
species were left to stand in d -DMSO, there were no changes
6
in the spectrum. However, when both 11 and glucose were
added together, after 48 h significant changes in the chemical
shifts of 11 and glucose were observed (Fig. 5). Integrations of
the initial signals of 11 and glucose decreased, while signals
matching that of starting reagent Bi(m-Tol)
range of the aromatic region at t = 48 h, both complex 11 and
Bi(m-Tol) are present in solution. However after 1 week, only
the Bi(m-Tol) could be observed.
3
appeared. In the
3
3
This experiment was repeated, this time with 11 and
1
glucose stirred at 40 °C over 18 hours. The resulting H NMR
3 2
Fig. 4 Decomposition curve of [Bi(o-Tol) (dif) ], 3 in DMEM culture
spectrum resembled that of t = 48 h at 25 °C, where both 11
and Bi(m-Tol)3 were present, indicating that the rate of
decomposition of 11 increases at higher temperatures.
However, the rate of decomposition is not comparable with
that observed in the culture medium (5 h in culture medium
vs. 1 week in DMSO), suggesting that while glucose is a com-
ponent of the decomposition process, other, as yet unknown,
factors contribute to an accelerated rate of decay.
media at 25 °C with time.
parasitic activity while being non-toxic towards human fibro-
blast cells.
It is still not clear what is oxidised as the Bi(V) complex
reduces to BiAr . One of the principle constituents of culture
3
medium (DMEM and M199) is glucose, second only to NaCl in
concentration. Initial NMR spectra looking at the interaction
of 3 with glucose proved complicated, so to simplify the data
Biological activity
L. major promastigotes and human fibroblast cells. All com-
the aspirin (aspH) derivative [Bi(m-Tol) (asp) ], 11 was chosen plexes were first screened for solubility in DMSO (see ESI†).
3
2
as the illustrative Bi(V) complex for the study.
A total of fifteen Bi(V) complexes 1–3, 5–11, 25–29, their parent
1
Fig. 5 H NMR spectroscopic studies of [Bi(m-Tol)
3
(asp)
2 6
] 11 and glucose in d -DMSO, 25 °C at t = 0 h, t = 48 h, and t = 1 week. Dotted boxes are
drawn around the chemical signals of Bi(m-Tol)
3
and solid boxes are drawn around the corresponding tolyl chemical signals of 11.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
benzoic acids along with Bi(o-Tol)
3
, 4, Bi(m-Tol)
3
, 36 and Bi(p- concentration of 3.12 μM, promastigote viability was slightly
Tol) , 12 were assessed for their anti-parasitic activity on higher, ranging from 0.4–21.9%, with 100% fibroblast viability.
3
L. major promastigotes and their toxicity towards human fibro-
Compounds 1, 5, 7 and 8 on the other hand belong to the
blast cells. Insoluble complexes were excluded from the assay. first category (Fig. 6). They had promastigote viabilities and
The complexes were tested over concentration range of 48 nM fibroblast viabilities of 3.9–9% and 82–100% at 6.12 μM respect-
to 100 μM in culture media from a 10 mM DMSO stock. The ively, and 11.9–24% and 92.7–100% at 3.12 μM respectively.
resulting IC50 values are presented in Table 2. DMSO and At a lower concentration of 3.12 μM, the lowest threshold
Amphotericin B controls were set up at concentrations equi- reached by the complex in eradicating promastigotes was
valent to that of the compound concentrations (48 nM to 11.9%, while still maintaining a respectable >92% viability for
1
00 μM). DMSO showed no apparent effect on any of the pro- the fibroblasts. In comparison, the lower viability for the pro-
mastigotes or human fibroblast cells. Amphotericin B was the mastigotes at 6.12 μM also indicated that the fibroblasts were
reference anti-leishmanial reagent. Complexes with a suitable affected by the complex, giving the lowest fibroblast viability to
selectivity index (parasites vs. human cells) were taken forward be 82%. In this case, compound 1 sustained a constant 100%
into a study on their effectiveness towards amastigotes.
fibroblast viability while killing up to 91% of the promasti-
After 48 hours, the parent carboxylic acids showed no toxi- gotes at 6.12 μM, and 76% promastigotes at 3.12 μM (Fig. 6).
city towards the parasites or the fibroblasts. In contrast, all of
the bismuth complexes showed anti-promastigote activity.
Selectivity
Ideally, Bi(V) complexes would show excellent bioactivity Our previous studies focused on two areas: the Sb(V) com-
towards the promastigotes (low percentage viabilities) while pounds derived from carboxylates of both non-NSAID and
1
3
maintaining non-toxicity towards human fibroblasts (high per- NSAID classes, and the Bi(III) compounds which were derived
1
5
centage viability). This would be an indication that the from NSAID and non-NSAID classes. With the Sb(V) com-
complex would be less toxic and so would have decreased pounds, it was determined that there was a trend in which
harmful side effects. Most of the complexes tested also showed p-Tol and m-Tol analogues were more promising than the o-Tol
toxicity to human fibroblasts at concentrations above 12.5 μM. and Ph analogues by being more bioactive against both
1
3
However, at lower concentrations, below 10 μM, compounds L. major promastigotes and amastigotes. With the Bi(V) com-
–12 showed distinct discrimination between the promasti- pounds however, the trend was found to differ from that of the
gotes and fibroblasts. Some compounds in this group Sb(V) complexes.
1
performed well at one concentration, higher and lower concen-
Complexes derived from 4-nitrobenzoic acid were available
trations tended not to give as desirable fibroblast/promastigote for comparison across the entire tolyl series while compounds
viabilities. Other compounds were able to perform well over a derived from 2-ethoxybenozic acid were available for compari-
larger range of concentrations.
son across the aryl series of o-, m-, p-Tol and Ph.
Compounds 2, 3, 9, 10 and 11 belong to this latter category
Based on duplicate tests on fibroblasts and promastigotes,
(Fig. 6). At 6.12 μM, promastigote viability ranged from it became apparent that the primary influence on the non-
0
.0–6.3% and fibroblast viability of 83.8–100%. At a complex toxicity of the complex was based on the aryl group while the
Table 2 IC50 and selectivity ratios of Bi(Tol)
human fibroblasts
3
4, 36, 12 and their derived complexes 1–3, 5–11, 25–29 respectively with L. major promastigotes and
IC50 (± SEM) (µM)
Selectivity index
Complex
[
Bi(Tol)
3
L
2
]
Ar
L
Promastigotes
Fibroblasts
Selectivity ratio:
at 6.12 μM
1
2
3
4
5
6
7
8
9
1
1
3
1
2
2
2
2
2
o-Tol
O
O
O
2
2
2
CC
C C
6
H
4
(2-EtO)
0.93 (0.04)
0.69 (0.03)
0.57 (0.02)
0.85 (0.03)
1.10 (0.04)
1.11 (0.11)
0.93 (0.04)
0.89 (0.03)
0.75 (0.05)
0.76 (0.02)
0.76 (0.02)
20.99 (0.05)
0.98 (0.02)
0.84 (0.02)
1.04 (0.02)
1.04 (0.02)
1.77 (0.02)
0.89 (0.02)
17.13 (0.05)
14.21 (0.03)
16.13 (0.04)
11.02 (0.03)
9.86 (0.02)
11.19 (0.02)
12.02 (0.01)
11.52 (0.02)
13.77 (0.02)
19.81 (0.04)
11.96 (0.01)
460.20 (0.29)
28.38 (0.04)
5.55 (0.04)
5.76 (0.05)
5.82 (0.05)
7.82 (0.03)
4.48 (0.04)
18.4
20.6
28.3
13.0
9.0
10.0
12.9
12.9
18.4
26.1
15.7
21.9
29.0
6.6
11.3
42.4
114.1
1.2
10.8
2.0
20.7
21.0
39.9
60.0
838.0
1.3
2.1
6.0
3.5
4.6
3.8
3.7
6
H
4
(4-NO
2
)
CC
6
H
3
(2-OH,5-C H (2,4-F ))
6 3 2
—
m-Tol
p-Tol
O
O
O
O
O
O
O
2
CC
CC
CC
CC
CC
CC
CC
6
6
6
6
6
6
6
H
3
H
3
H
4
H
4
H
4
H
4
H
4
(3,5-Me
2
)
2
2
2
2
2
2
(3,5-(OH)
(2-MeO)
(2-EtO)
(4-NO )
2
(2-NH(C
(2-OAc)
2
)
0
1
2
2
5
6
7
8
9
6
4 3
H (3-CF )))
—
—
O
O
O
O
O
2
CC
CC
CC
CC
CC
6
6
6
6
6
H
3
H
3
H
4
H
4
H
4
(3,5-Me
2
)
2
2
2
2
(3,5-(OH)
(2-MeO)
(2-EtO)
2
)
5.5
5.6
4.4
5.0
2
(4-NO )
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Fig. 6 Activity of 1–12 on L. major promastigotes (●) and human fibroblasts (○) after 48 hours.
activity against the promastigote was based on the carboxylate the ratio of IC50 of the fibroblasts to the IC50 of the promasti-
ligand. By looking at their IC50 values, the general trend gotes. Compounds with a score ≥9.0 proceeded to further
of selectivity observed across the series was identified as o-Tol biological testing against L. major amastigotes. This excluded
≥
m-Tol > p-Tol ≥ Ph (Table 2).
[Bi(m-Tol) ] 36 and the [Bi(p-Tol) L ] complexes 25–29.
3 3 2
For the promastigotes, [Bi(o-Tol)
3
(O
2
CC
6
H
4
(4-NO
2
))
2
]
L. major amastigotes. Based on their selectivity index, com-
had the lowest IC50 (0.69 ± 0.03 µM), followed closely by pounds 1–12 were assessed for their activity against L. major
[Bi(m-Tol) (O CC H (4-NO )) ] (0.75 ± 0.04 µM), and then amastigotes, the clinically relevant form of the parasite.
3 2 6 4 2 2
[
Bi(p-Tol)
For the fibroblasts, the same trend was followed, with the in the presence 1 µM complex concentration, selected based
ortho analogue having the highest IC50 (14.21 ± 0.03 µM), fol- on IC50 values obtained from the promastigote assay (Fig. 7).
lowed very closely by meta (14.04 ± 0.02 µM) and para (4.48 ± The macrophage invasion assay presents a more challen-
.04 µM). This trend was also followed by compounds derived ging biological environment and demands more from the
3
(O
2
CC
6
H
4
(4-NO
2
))
2
] (0.89 ± 0.02 µM).
Amastigote-infected macrophages were incubated for 48 hours
0
from 2-ethoxybenzoic acid, [Bi(Tol) (O CC H (2-EtO)) ].
complexes than the promastigote assay. Compounds have to
3
2
6
4
2
For promastigotes, [Bi(o-Tol)
Tol) (O CC (2-EtO)) ] had similar IC50 values; (0.93 ± ment of phagolysosomes. Compounds 1, 2, 3, 9, and 11
.04 µM) and (0.89 ± 0.03 µM) respectively, followed by [Bi(p- showed anti-amastigote activity that resulted in statistically sig-
Tol) (O CC (2-EtO)) ] at (1.77 ± 0.02 µM), and lastly nificant differences when compared to the positive control,
BiPh (O CC (2-EtO))
For fibroblasts, a similar trend was followed as for the first gote activity. Differences although significant, were small and
set: the IC50 values are highest for the ortho analogues (17.13 ± do not appear physiologically relevant. Nevertheless, these
3 2 6 4 2
(O CC H (2-EtO)) ] and [Bi(m- traverse two membranes and be active in the acidic environ-
3
2
6
H
4
2
0
3
2
6
H
4
2
1
8
[
3
2
6
H
4
2
] at (2.29 ± 0.04 µM).
while compounds 5, 6, 7, 8 and 10 displayed no anti-amasti-
0
0
.05 µM), followed by the meta (11.63 ± 0.02 µM), para (0.89 ± data provide information regarding the structure–activity
1
8
.02 µM) and finally phenyl (3.84 ± 0.03 µM).
relationship. The more active and selective compounds were
] and [Bi(m-Tol) ] complexes, though
From this comparison of IC50 values across different aryl mostly the [Bi(o-Tol)
3
L
2
3 2
L
groups, it can be seen that the ortho and meta analogues the m-Tol complexes displayed a greater range of toxicities
were more selective and promising, while the p-Tol and the based on their ligand combination. Compared to earlier
Ph series were relatively similar in terms of activity on both assays performed with tris-arylantimony(V) dicarboxylates, the
parasites and fibroblasts, and were not as selective. These bismuth analogues were not as effective. Future complexes
complexes were therefore not included for future testing.
have to be designed to be more stable. The Bi(III) compounds
4 and Bi(p-Tol) 12, products of the in situ reduction
From the individual percentage viabilities at certain Bi(o-Tol)
3
3
complex concentrations, we analysed the data to obtain the of the Bi(V) complexes were tested at a 10-fold higher concen-
IC50 for each complex based on their activity against promasti- tration than the bismuth(V) complexes and showed some anti-
gotes and fibroblasts. A selectivity index was calculated from parasitic activity.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
bipyramidal geometry. The complexes are monomeric in the
solid state with a seven coordinate Bi(V) centre. The carboxylate
ligands adopt a bidentate binding mode to the Bi(V) atom, a
general feature across the series of compounds as confirmed
by FT-IR analysis.
Fifteen of the complexes were assessed for their anti-para-
sitic activity against L. major promastigotes and for cytotoxicity
against human fibroblasts. All complexes were highly active
against the promastigotes while the free acids proved to be
non-toxic at the highest tested concentration of 100 μM. The
primary influence on the non-toxicity of the complexes
appears to be based mainly on the aryl group while the activity
against the promastigote is based on the carboxylate ligand.
IC50 values showed the general trend of selectivity observed
across the series to be o-Tol ≥ m-Tol > p-Tol ≥ Ph.
To assess the stability of the Bi(V) complexes, a representa-
Fig. 7 Infected macrophages after 48 hours. Number of macrophages tive example, complex [Bi(o-Tol) (dif) ], 3 was studied in D O
3
2
2
infected with at least one amastigote was determined by microscopy of
fixed specimens in duplicates. Amphotericin B (AmpB) was used as a
reference compound at 0.5 μg mL− . DMSO control was at 1%. Error
bars indicate SEM, one-way ANOVA with Dunnett’s multiple comparison
and DMSO and in DMEM cell culture medium. Resulting
H NMR spectra showed the complex to be stable in D O and
2
1
1
DMSO, however in cell culture medium, [Bi(o-Tol) (dif) ] was
3
2
test was used to determine statistical significance between test samples observed to reduce to its starting compound Bi(o-Tol)
3
which
and non-treated control (+ve control).
precipitates out as an off-white solid. Further solution state
studies found the half-life of complex 3 to be 1.2 hours. While
our previous studies have shown little impact of amino acids,
phosphate or other anions on the decomposition of the Bi(V)
1
It is not surprising to see low or no activity in the macro- complexes, H NMR solution studies on the structurally
phage invasion assay. In the majority of cases promising simpler complex [Bi(m-Tol) (asp) ], 11, indicated that glucose,
3
2
results from promastigote testing cannot be extrapolated to present in high concentration in DMEM and M199, does
2
1
amastigote setting. The short half-life of compounds could plays a role in the decomposition though the rate is not as fast
be responsible for relatively low levels of observed activity, par- as that observed when the complex is in culture medium
ticularly if they can only exert most of their toxicity towards the meaning other, as yet unknown factors, contribute to the
amastigotes during the first few hours in culture medium.
It is also possible that the concentration tested (the IC50
reduction of the Bi compounds.
Despite the rapid decomposition rate of the Bi(V) complexes
value) was too low. Despite being sufficient to kill 50% of pro- in culture medium they are still active long enough to exert a
mastigotes in culture that concentration was too low to be was significant, and somewhat selective, toxic effect on the pro-
effective against amastigotes. It is reasonable to assume that mastigotes and fibroblasts. The decomposition products Bi
this is due to a number of coincident factors: an inability of (Tol) are relatively non-toxic to both.
3
the compounds to cross macrophage membranes because
Based on their selectivity index, ten complexes derived
of permeability constraints; inefficient uptake by parasites; mainly from Bi(o-Tol)3 and Bi(m-Tol)3 were further screened
rapid intracellular degradation; or inactivity in the more for amastigote and macrophage toxicity, representing the clini-
acidic environment. Moreover, non-optimized compounds cal form of the disease. In this assay the compounds did not
were tested and it is unlikely that they would exhibit potent perform as well reflecting the significantly more challenging
anti-leishmanial activities without further iterative rounds of biological environment, as well the inherent instability of the
medicinal chemistry optimisation. As such our ongoing efforts complexes under such conditions.
are aimed at structural modification to improve permeability
and stability, and to focus on compounds encapsulation as a
delivery system to the amastigotes.
Experimental section
All solvents used were purchased from Merck. For synthesis of
BiPh , solvents were dried prior to use via the MBraun-SPS-800
and stored over molecular sieves (4 Å) in a Schlenk flask under
3
Conclusion
A series of twenty-nine organometallic triarylbismuth(V) dicarb-
2
N . Tris-o-tolylbismuth, 4 and tris-p-tolylbismuth, 12 were syn-
2
2
oxylates of the form [Bi(Tol) (O CR) ] were synthesized and thesized via literature procedure. All other required chemi-
3
2
2
fully characterized. Eleven solid-state structures were deter- cals were purchased from Sigma-Aldrich and used as received.
mined by single crystal X-ray diffraction (3, 6, 8, 9, 19, 22, NMR spectra were recorded on a Bruker Avance DRX400
2
5–28, 31) showing the complexes adopt a typical trigonal spectrometer (400 MHz) with chemical shifts referenced to the
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
appropriate deuterated solvents. Infrared spectra were radiation (λ = 0.71070 Å) and processed using the CrysAlisProv
2
8
recorded on an Agilent Technologies Cary 360 FTIR spectro- 1.171.34.36 software; Lorentz, polarization and absorption
−
1
meter in the range 4000–500 cm . Melting points were deter- corrections (multi-scan) were applied. Crystallographic data for
mined in soda glass tubes on a digital Stuart Scientific melting compound 27 was obtained on a Bruker X8 APEXII CCD dif-
point apparatus SMP10. Mass spectrometry (ESI) was per- fractometer equipped with an OXFORD Cryosystems 700 Cryo-
formed on a Micromass Platform QMS spectrometer with an stream and cooled to 123(2) K. Data was collected with mono-
electrospray source and a cone voltage of 35 eV. CHN elemen- chromatic (graphite) MoK
tal analysis was performed by The Campbell Microanalytical processed using the Bruker Apex2 v2014.7-1 software;
Laboratory, Department of Chemistry, University of Otago in Lorentz, polarization and absorption corrections (multi-scan –
α
radiation (λ = 0.71065 Å) and
2
9
3
0
Dunedin, New Zealand.
SADABS) were applied. All compounds were solved and
3
1
refined with SHELX-97. All non-hydrogen atoms were refined
with anisotropic thermal parameters unless otherwise indi-
Biological assays
Cell viability assay. The Celltiter Blue Cell Viability Assay cated and hydrogen atoms were placed in calculated positions
Promega, Madison, WI, USA) was used for screening for anti- using a riding model with C–H = 0.95–0.98 Å and Uiso(H) =
(
leishmanial activity and toxicity. Compounds were dissolved in xUiso(C), x = 1.2 or 1.5 unless otherwise indicated. A summary
DMSO at 10 mmol L− working stock and diluted out in appro- of crystallographic data for 3 (CCDC 1420787) is provided
1
priate culture media. The assay was set up in duplicates in 96- below in the Experimental section while for compounds 6, 8,
6
well plates according to the manufacturer’s instructions. 10
9, 19, 22, 25–28, 31 (CCDC 1420788–1420797) it is provided in
5
−1
promastigotes per mL and 10 mL primary human fibro- the ESI.†
blasts were used. Cell viability was assessed by measuring fluo-
rescence at 550 nm excitation and 590 nm emission as per
2
3
General procedure
manufacturers’ instructions.
added to samples at the time of setting up the assay and the General synthetic procedure (GP1): stoichiometric equivalents
negative control (no cells) value was subtracted from all sub- of Bi(Tol) and benzoic acid (1 : 2) were each dissolved in 5 mL
sequent readings as a background value. The mean was calcu- of warm solvent and mixed. An excess of 30% H was added
The Celltiter Blue dye was
3
2 2
O
lated from duplicate readings. All readings were compared to and the mixture was stirred for 10 minutes and filtered. Crys-
the no-drug control and the percentage growth inhibition tals were subsequently obtained on allowing the filtrate to
was calculated. DMSO controls were included. All plates stand at room temperature overnight. Grignard reagents were
were assessed microscopically. The graphs shown in this paper synthesized under Schlenk conditions in Et O by the reaction
2
give the percentage of positive control versus concentration. of stoichiometric ratios of Mg turnings with 2-bromotoluene,
Macrophage invasion assay was performed as previously 3-bromotoluene and 4-bromotoluene respectively. The result-
2
4
described.
ing Grignard reagent was reacted with BiCl in a 3 : 1 stoichio-
3
Cell culture. Leishmania major virulent clone V121 was metric ratio to obtain the Bi(Tol)
derived from the L. major isolate LRC-L137 and maintained at solid.
3
product as a crystalline
2
6 °C in M199 medium supplemented with 10% (v/v) heat
Synthesis of tris-o-tolylbismuth bis(2-ethoxybenzoate), 1. Bi-
inactivated FBS (Trace Biosciences, NSW, Australia). The (o-Tol) (0.200 g, 0.41 mmol), 2-ethoxybenzoic acid (125 μL,
human primary fibroblast cells were cultured in Dulbecco’s 0.83 mmol) and 100 μL 30% H were reacted in warm
Modified Eagle’s Medium (DMEM) (Life Technologies) sup- diethyl ether according to GP. Yield: 39.0% (0.130 g);
2
5
3
2 2
O
1
plemented with 10% HI-FBS at 37 °C in 5% CO
2
.
3 2
mp. 126–128 °C; H NMR (400 MHz, (CD ) SO, 25 °C): δ = 8.34
Data analysis. Statistical analysis IC50 values were calcu- (3H, dd, J = 7.9 Hz, 1.1 Hz, Tol CH), 7.58 (6H, m, Tol CH), 7.47
lated in Graphpad Prism by non-linear regression (curve fit), (3H, td, J = 7.4 Hz, 1.0 Hz, Tol CH), 7.25 (2H, m, CH), 7.07 (2H,
dose–response inhibition, log(inhibitor) vs. normalised dd, J = 7.6 Hz, 1.8 Hz, CH), 6.92 (2H, d, J = 7.9 Hz, CH), 6.78
response, variable slope.
(2H, td, J = 7.4 Hz, 0.8 Hz, CH), 3.90 (4H, q, J = 6.9 Hz,
OCH CH ), 2.60 (9H, s, Tol CH ), 1.13 (6H, t, J = 7.0 Hz,
2
3
3
3
Crystallography
13
1
OCH
2
CH
); C{ H} (100 MHz, (CD
CH
lected at the MX1 beamline and compound 31 was collected (CH ), 133.2 (CH ), 130.9 (CH ), 130.7 (CH ), 129.5 (CH ),
3
)
2
SO, 25 °C): δ = 170.9
Crystallographic data of compounds 6, 22, 25, 26, 28 were col- (COO), 163.0 (BiC), 156.3 (COCH
2
3
), 141.5 (CCH ), 134.0
3
ar
ar
ar
ar
ar
on the MX2 beamline at the Australian Synchrotron, Mel- 128.3(CHar), 125.4 (CHar), 119.7 (CHar), 113.2 (CHar), 63.6
+
bourne, Victoria, Australia, operating at 17.4 keV (λ
=
(CH
2
), 22.9 (Tol CH
3
3
), 14.4 (CH ); MS ESI 208.9 [Bi], 391.0 [Bi-
+
+
.71080 Å) using an open flow of N cryostream and cooled to (o-Tol) ] , 647.1 [Bi(o-Tol) L] ; ESI 165.1 [L] , 721.2 [Bi(o-
2 2 3
−
−
0
1
2 2
73(2) K for MX1 and 100(2) K for MX2. The software used for Tol) L
]−; IR 2980 (w), 2931 (w), 2897 (w), 2115 (w). 1742 (w),
2
6
data collection and reduction of the data were BluIce and 1719 (w), 1646 (w), 1597 (sh), 1475 (m), 1447 (sh), 1382 (sh),
XDS. Crystallographic data for compounds 3, 8, 9, 19 was col- 1326 (s), 1297 (sh), 1269 (sh), 1232 (s), 1163 (sh), 1143 (sh),
27
lected on an OXFORD Gemini Ultra equipped with an 1111 (sh), 1041 (sh), 988 (sh), 922 (sh), 794 (m), 742 (s), 709
OXFORD Cryosystems 700 Cryostream and cooled to 123(2) (sh), 663 (sh); Elemental analysis [C39
H
39BiO
6
·2H
2
O] (848.28)
K. Data was collected with monochromatic (graphite) MoKα Calculated C 55.19 H 5.11 Found C 55.20 H 4.69.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
3
Synthesis of tris-o-tolylbismuth bis(4-nitrobenzoate), 2. 7.5 Hz, Tol p-CH), 7.13 (2H, s, p-CH), 2.35 (9H, s, Tol CH ), 2.27
1
3
1
Bi(o-Tol) (0.200 g, 0.41 mmol), 4-nitrobenzoic acid (0.139 g, (12H, s, CH3); C{ H} (100 MHz, (CD ) SO, 25 °C): δ = 171.8
3
3 2
0
.83 mmol) and 100 μL 30% H
2
O
2
were reacted in warm (COO), 160.2 (BiC), 141.2 (CCH
3
3 3
), 137.4 (CCH ), 133.4 (o-CH ),
diethyl ether according to GP. Yield: 44.6% (0.149 g); 132.4 (CCCO), 131.8 (CHar), 131.3 (CHar), 130.5 (CHar), 127.3
1
+
mp. 146–148 °C; H NMR (400 MHz, CDCl , 25 °C): δ = 8.44 (CH ), 21.5 (CH ), 20.7 (CH ); MS ESI 208.97 [Bi], 346.18 [2Bi-
3
ar
3
3
2
+
+
(
3H, dd, J = 1.1 Hz, 8.0 Hz, Tol CH), 8.11 (4H, d, J = 8.9 Hz, (m-Tol) + (m-Tol)H] , 391.03 [Bi(m-Tol)
CH), 7.93 (4H, d, CH), 7.56 (3H, dd, J = 1.0 Hz, 7.5 Hz, Tol CH), Tol)
.53 (3H, t, J = 7.0 Hz, Tol CH), 2.68 (9H, s, CH3); C{ H} 2917 (w), 2862 (w), 1574 (sh), 1553 (sh), 1472 (w), 1447 (w),
100 MHz, (CD SO, 25 °C): δ = 168.6 (COO), 161.9 (BiC), 149.6 1385 (sh), 1347 (s), 1309 (m), 1259 (m), 1208 (w), 1168 (w),
CCH ), 142.1 (CNO ), 140.0 (CCOO), 134.7 (CHar), 133.7 1092 (w), 1040 (w), 999 (w), 980 (sh), 869 (w), 811 (sh), 771 (sh),
2
] , 631.18 [Bi(m-
+
−
−
3
(L)] ; ESI 149.10 [L] ; IR 3503 (w), 3377 (br), 3051 (w),
1
3
1
7
(
3 2
)
(
3
2
(
2
CH ), 131.6 (CH ), 130.7 (CH ), 129.0 (CH ), 123.2 (CH ), 675 (sh); Elemental analysis [C H BiO ·H O] (798.28) Calcu-
ar
ar
ar
ar
ar
39 39
4
2
+
+
3.8 (CH
3
); MS ESI 208.9 [Bi], 391.1 [Bi(o-Tol)
2
] , 648.1 lated C 58.65 H 5.17 Found C 58.31 H 5.18.
+
−
]−; IR 2929 (br),
[Bi(o-Tol)
3
L] ; 166.0 [L] , 723.2 [Bi(o-Tol)
2
L
2
Synthesis of tris-m-tolylbismuth bis(3,5-dihydroxybenzoate),
2
1
861 (br), 1646 (sh), 1601 (sh), 1519 (sh), 1474 (m), 1449 (m), 6. Bi(m-Tol) (0.400 g, 0.83 mmol), 3,5-dihydroxybenzoic acid
3
405 (w), 1349 (m), 1292 (m), 1206 (sh), 1163 (sh), 1123 (0.256 g, 1.66 mmol) and 100 μL 30% H O were reacted in
2 2
(
(
(
sh), 1099 (sh), 999 (sh), 874 (sh), 819 (sh), 791 (sh), 742 warm diethyl ether according to GP. Yield: 64.6% (0.377 g);
1
sh), 719 (sh), 669 (m); Elemental analysis [C H BiN O ] mp. 170 °C (decomp.); H NMR (400 MHz, (CD ) SO, 25 °C):
3
5
29
2
8
3 2
3
814.60) Calculated C 51.61 H 3.59 N 3.44 Found C 51.47 δ = 9.40 (4H, s, OH), 7.96 (3H, s, Tol o-CH), 7.91 (3H, d, J =
3
H 3.63 N 3.31.
7.6 Hz, Tol o-CH), 7.61 (3H, t, J = 7.7 Hz, Tol m-CH), 7.37 (3H,
3
4
Synthesis of tris-o-tolylbismuth bis(2′,4′-difluoro-4-hydroxy- d, J = 7.5 Hz, Tol p-CH), 6.72 (3H, d, J = 2 Hz, o-CH), 6.31 (2H,
4
13
1
biphenyl-3-carboxylate), 3. Bi(o-Tol)
Diflunisal (0.208 g, 0.83 mmol) and 100 μL 30% H
reacted in warm diethyl ether according to GP. Yield: 75.1% 141.2 (CCOO), 134.4 (CCH ), 133.4 (CH ), 131.8 (CH ), 131.2
(
2
3
(0.200 g, 0.41 mmol), t, J = 2 Hz, p-CH), 2.36 (9H, s, Tol CH
3
); C{ H} (100 MHz,
2
O
2
were (CD SO, 25 °C): δ = 171.7 (COO), 160.1 (BiC), 158.2 (COH),
3 2
)
3
ar
ar
1
3 2 3
0.302 g); mp. 153–154 °C; H NMR (400 MHz, (CD ) SO, (CHar), 130.3 (CHar), 107.6 (CHar), 106.0 (CHar), 21.5 (CH ); MS
5 °C): δ = 11.69 (2H, s, OH), 8.36 (3H, d, J = 7.9 Hz, o-CHar), ESI 208.99 [Bi], 391.00 [Bi(m-Tol)
.72 (5H, m, CH ), 7.66 (3H, t, J = 7.4 Hz, m-CH ), 7.57 (3H, t, 725.13 [BiL] ; ESI 153.08 [L] ; IR 3511 (w), 3394 (w), 3157
+
+
+
2
] , 635.10 [Bi(m-Tol)
3
(L)] ,
2
+
−
−
7
ar
ar
J = 7.4 Hz, p-CHar), 7.53 (2H, d, J = 8.9 Hz, CHar), 7.46 (2H, q, (br), 2980 (w), 1569 (s), 1465 (m), 1465 (w), 1341 (sh), 1298 (s),
J = 6.9 Hz, CHar), 7.31 (2H, t, J = 10.4 Hz, CHar), 7.16 (2H, td, 1282 (m), 1202 (w), 1155 (sh), 1097 (w), 1005 (sh), 979 (sh), 954
J = 8.1 Hz, 1.8 Hz, CH ), 6.89 (2H, d, J = 8.6 Hz, CH ), 2.60 (w), 858 (sh), 811 (sh), 811 (sh), 785 (sh), 678 (sh); Elemental
ar
ar
1
3
1
(
9H, s, CH
3
); C{ H} (100 MHz, (CD
3
)
2
8 2 2
SO, 25 °C): δ = 172.6 analysis [C35H31BiO ·2H O·Et O] (898.28) Calculated C 52.12
(
(
1
1
(
COO), 161.2 (BiC), 160.4 (CF), 159.8 (CF), 158.1 (CCOO), 141.7 H 5.05 Found C 52.01 H 5.04.
CCH ), 134.8 (CH ), 134.3 (CH ), 134.0 (CH ), 132.0 (CH ), Synthesis of tris-m-tolylbismuth bis(2-methoxybenzoate),
31.3 ((F)CCHC(F)), 130.6 (CHar), 129.3 (CHar), 124.6 (i-Car), 7. Bi(m-Tol) (0.200 g, 0.41 mmol), 2-methoxybenzoic acid
23.8 (i-Car), 117.1 (CHar), 115.2 (COH), 112.1 (CHar), 104.5 (0.126 g, 0.83 mmol) and 100 μL 30% H were reacted in
3
ar
ar
ar
ar
3
2 2
O
+
+
CH ), 23.0 (CH ); MS ESI 208.9 [Bi], 391.1 [Bi(o-Tol) ] , 731.2 warm diethyl ether according to GP. Yield: 47.6% (0.153 g);
ar 3 2
+
−
−
1
[Bi(o-Tol)
3
L] ; ESI 249.0 [L] ; IR 3074 (br), 2951 (w), 1634 (m), mp. 145–146 °C; H NMR (400 MHz, (CD
3
)
2
SO, 25 °C): δ = 8.08
3
1
1
1
593 (m) 1561 (sh), 1509 (w), 1481 (sh), 1435 (sh), 1381 (sh), (3H, s, Tol o-CH), 8.00 (3H, d, J = 8 Hz, Tol o-CH), 7.61 (3H, t,
352 (sh), 1292 (m), 1245 (s), 1223 (sh), 1140 (sh), 1102 (sh),
033 (w), 967 (sh), 917 (w), 889 (sh), 851 (sh), 833 (sh), 743 d, J = 8 Hz, CH), 6.91 (2H, t, J = 7.4 Hz, p-CH), 3.80 (6H, s,
3 3 3 2
sh), 726 (sh), 703 (sh), 662 (sh); Elemental Analysis OCH ), 2.37 (9H, s, Tol CH ); C{ H} (100 MHz, (CD ) SO,
3
J = 7.7 Hz, Tol m-CH), 7.39 (7H, m, Tol p-CH + CH), 7.04 (2H,
3
3
1
3
1
(
[
C H BiF O ] (980.77) Calculated C 57.56 H 3.60 Found 25 °C): δ = 172.1 (COO), 159.6 (BiC), 157.6 (COCH ), 149.7
4
7
35
4
6
3
C 57.67 H 3.88.
(CCH3), 141.2 (CHar), 133.6 (CHar), 131.8 (CHar), 131.7 (CHar),
Crystal data for 3. C51
H
45BiF
4
O
7
; M
r
= 1054.85; triclinic; 131.1 (CHar), 130.4 (CHar), 130.1 (CHar), 123.9 (CCOO), 120.0
+
space group: P ˉ1 ; a = 12.2901(3), b = 12.4486(3), c = 15.1716(4); (CH ), 112.5 (CH ), 55.8 (OCH ), 21.5 (CH ); MS ESI 208.99
α = 76.296(2); β = 76.185(2); γ = 84.494(2); V = 2187.95(9) Å ; Z = [Bi], 391.00 [Bi(m-Tol)
, reflections collected/unique: 56 966/14 625 (Rint = 0.0449); [L] , 693.00 [Bi(m-Tol) L
2 2
R values (I > 2σ(I)) = 0.0265; wR(F ) values (I > 2σ(I)) = 0.0503; 2835 (w), 1601 (sh), 1581 (sh), 1542 (sh), 1457 (m), 1435 (sh),
ar
ar
3
3
3
+
+
−
2
] , 633.14 [Bi(m-Tol)
3
(L)] ; ESI 151.10
−
]−; IR 3045 (w), 3001 (w), 2915 (w),
2
2
1
2
R
1
values (all data) = 0.0355; wR(F ) values (all data) = 0.0539; 1352 (s), 1300 (w), 1268 (sh), 1249 (m), 1177 (sh), 1162 (sh),
GOF = 1.033; temperature = 123(2) K. 1142 (sh), 1097 (w), 1021 (w), 981 (sh), 857 (sh), 809 (sh), 774
Synthesis of tris-m-tolylbismuth bis(3,5-dimethylbenzoate), (sh), 750 (sh), 704 (sh), 659 (sh); Elemental analysis
. Bi(m-Tol) (0.200 g, 0.41 mmol), 3,5-dimethylbenzoic acid [C37 35BiO ] (784.66) Calculated C 56.64 H 4.50 Found C 56.22
0.125 g, 0.83 mmol) and 100 μL 30% H were reacted in H 4.53.
warm diethyl ether according to GP. Yield: 51.5% (0.165 g); Synthesis of tris-m-tolylbismuth bis(2-ethoxybenzoate), 8.
SO, 25 °C): δ = 7.99 Bi(m-Tol) (0.200 g, 0.41 mmol), 2-ethoxybenzoic acid
3H, s, Tol o-CH), 7.96 (3H, d, J = 8.1 Hz, Tol o-CH), 7.61 (3H, (0.125 μL, 0.83 mmol) and 100 μL 30% H were reacted in
t, J = 7.7 Hz, Tol m-CH), 7.47 (4H, s, o-CH), 7.36 (3H, d, J = warm diethyl ether according to GP. Yield: 46.2% (0.154 g);
5
(
3
H
6
2 2
O
1
mp. 173–176 °C; H NMR (400 MHz, (CD
3
)
2
3
3
(
2 2
O
3
3
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
1
mp. 144–145 °C; H NMR (400 MHz, (CD
3
)
2
SO, 25 °C): δ = 8.05
Synthesis of tris-m-tolylbismuth bis(2-(acetoxy)benzoate),
3
(
3H, s, Tol o-CH), 8.01 (3H, d, J = 7.7 Hz, Tol o-CH), 7.60 (3H, 11. Bi(m-Tol)3 (0.200 g, 0.41 mmol), aspirin (0.150 g,
t, J = 7.8 Hz, Tol m-CH),7.36 (7H, m, Tol p-CH + CH), 7.02 (2H, 0.83 mmol) and 100 μL 30% H O were reacted in warm
2 2
d, J = 8.3 Hz, CH), 6.90 (2H, t, J = 7.4 Hz, p-CH), 4.02 (4H, q, diethyl ether according to GP. Yield: 76.9% (0.265 g);
3
3
3
3
1
J = 6.9 Hz, OCH CH ), 2.35 (9H, s, Tol CH ), 1.28 (6H, t, mp. 129–132 °C; H NMR (400 MHz, (CD ) SO, 25 °C): δ = 7.92
2
3
3
3 2
1
3
1
3
2 3 3 2
OCH CH ); C{ H} (100 MHz, (CD ) SO, 25 °C): δ = 172.4 (3H, s, Tol o-CH), 7.88 (3H, d, J = 7.8 Hz, Tol o-CH), 7.75 (2H,
3
3
(
1
3
COO), 160.0 (BiC), 156.9 (i-Car), 141.0 (CCH ), 133.5 (CHar), d, J = 7.6 Hz, o-CH), 7.63 (3H, t, J = 7.6 Hz, Tol m-CH), 7.52
31.7 (CH ), 131.6 (CH ), 131.0 (CH ), 130.4 (CH ), 130.1 (2H, t, J = 7.7 Hz, p-CH), 7.41 (3H, d, J = 7.6 Hz, Tol p-CH),
ar ar ar ar
CHar), 124.2 (CCOO), 120.0 (CHar), 113.6 (CHar), 63.9 7.29 (2H, t, J = 7.3 Hz, asp m-CH), 7.08 (2H, d, J = 7.9 Hz, m-
3
3
3
3
(
(
+
13
1
OCH
2
+
CH
3
), 21.4 (CH
3
), 14.7 (OCH
2
CH
3
); MS ESI 391.2 [Bi(m- CH), 2.38 (9H, s, Tol CH
3
), 2.00 (6H, s, asp CH
3
); C{ H}
+
+
−
Tol) ] , 647.3 [Bi(m-Tol) (L)] , 835.4 [Bi(m-Tol) (L) + Na] ; ESI
65.0 [L] , 721.2 [Bi(m-Tol)
(150 MHz, (CD ) SO, 25 °C): δ = 170.3 (COO), 169.1 (BiC), 159.3
2
3
3
2
3 2
−
2 2 3
L ), 133.4 (CHar), 132.7 (CHar),
]−; IR 3052 (w), 2979 (w), 2921 (i-Car), 149.7 (i-Car), 141.6 (CCH
1
(
w), 1583 (sh), 1554 (sh), 1532 (w), 1472 (w), 1450 (sh), 1356 (s), 131.9 (CHar), 131.4 (CHar), 131.3 (CHar), 130.3 (CHar), 126.6 (i-
+
1
1
292 (w), 1269 (sh), 1238 (s), 1164 (sh), 1146 (sh), 1113 (sh), C ), 125.9 (CH ), 123.5 (CH ), 21.5 (CH ), 20.7 (CH ); MS ESI
ar ar ar 3 3
+
+
035 (sh), 980 (sh), 925 (sh), 853 (sh), 811 (sh), 795 (sh), 754 209.0 [Bi], 391.2 [Bi(m-Tol)
2
] , 661.3 [Bi(m-Tol)
+ Na] ; IR 2920 (w), 1755 (sh), 1624 (sh), 1592 (sh),
1553 (m), 1473 (m), 1448 (m), 1355 (sh), 1324 (sh), 1220 (sh),
Synthesis of tris-m-tolyl-bismuth bis(4-nitrobenzoate), 9. 1192 (sh), 1089 (sh), 1040 (w), 1012 (w), 980 (m), 954 (w), 919
Bi(m-Tol) (0.200 g, 0.41 mmol), 4-nitrobenzoic acid (0.138 g, (m), 878 (w), 850 (w), 809 (sh), 768 (sh), 748 (sh), 703 (sh), 672
.83 mmol) and 100 μL 30% H O were reacted in warm (sh); Elemental Analysis [C H BiO ·2H O] (876.23) Calculated
3
L] , 863.2 [Bi-
+
(sh), 704 (m), 667(sh); Elemental analysis [C39
H
39BiO
6
]
3 2
(m-Tol) L
(812.72) Calculated C 57.64 H 4.84 Found C 57.11 H 4.92.
3
0
2
2
39 35
8
2
diethyl ether according to GP. Yield: 41.9% (0.140 g); C 53.43 H 4.48 Found C 53.00 H 4.04.
1
mp. 170–172 °C; H NMR (400 MHz, (CD
3
)
2
SO, 25 °C): δ = 8.15
3
4
(
8H, br. dd, J = 8.6 Hz, J = 3.0 Hz, o-CH, m-CH), 8.02 (3H, s,
Tol o-CH), 7.98 (3H, d, J = 8.0 Hz, Tol o-CH), 7.65 (3H, t, J =
3
3
Acknowledgements
3
7
.7 Hz, Tol m-CH), 7.40 (3H, d, J = 7.5 Hz, Tol p-CH), 2.37 (9H,
1
3
1
s, Tol CH3); C{ H} (100 MHz, (CD ) SO, 25 °C): δ = 158.3 The authors thank the Australian Research Council, Monash
3 2
(BiC), 149.5 (CNO
2
), 147.6 (CCH
3
), 141.7 (CCOO), 133.6 (CHar), University and the Walter and Eliza Institute of Medical
1
32.2 (CHar), 131.6 (CHar), 130.7 (CHar), 130.6 (CHar), 123.5 Research for financial support, and Dr Chris Tonkin (WEHI)
+
+
(
CH ), 21.5 (CH ); MS ESI 209.0 [Bi], 391.1 [Bi(m-Tol) ] , for the supply of human primary fibroblast cells.
ar
3
2
+
−
−
−
6
3 2 2
48.3 [Bi(m-Tol) (L)] ; ESI 166.0 [L] , 723.1 [Bi(m-Tol) L ] ;
IR 2920 (w), 1622 (m), 1583 (sh), 1523 (sh), 1467 (w), 1406 (sh),
331 (s), 1162 (w), 1130 (sh), 1099 (sh), 1011 (sh), 981 (sh), 876
1
References
(sh), 831 (sh), 777 (sh), 720 (sh), 676 (sh); Elemental analysis
[
C
35
H
29BiN ] (814.60) Calculated C 51.61 H 3.59 N 3.44
2
O
8
1 L. Kedzierski, A. Sakthianandeswaren, J. M. Curtis,
P. C. Andrews, P. C. Junk and K. Kedzierska, Curr. Med.
Chem., 2009, 16, 599–614.
Found C 51.18 H 3.31 N 3.29.
Synthesis of tris-m-tolylbismuth bis(2-{[3-trifluoromethyl)-
phenyl]amino}benzoate), 10. Bi(m-Tol)
flufenamic acid (0.233 g, 0.83 mmol) and 100 μL 30% H O
3
(0.200 g, 0.41 mmol),
2 WHO, Leishmaniasis, http://www.who.int/leishmaniasis/
en/, Accessed August 2015.
2
2
were reacted in warm diethyl ether according to GP. Yield:
3 W. L. Roberts and P. M. Rainey, Antimicrob. Agents
Chemother., 1993, 37, 1842–1846.
1
6
6.4% (0.284 g); mp. 143–147 °C; H NMR (400 MHz,
(
CD ) SO, 25 °C): δ = 9.57 (2H, s, NH), 8.03 (3H, s, o-CH ), 7.97
4 W. L. Roberts, W. J. McMurray and P. M. Rainey, Antimicrob.
Agents Chemother., 1998, 42, 1076–1082.
5 F. Frezard, C. Demicheli and R. R. Ribeiro, Molecules, 2009,
14, 2317–2336.
6 Ashutosh, S. Sundar and N. Goyal, Med. Microbiol., 2007,
56, 143–153.
3
2
ar
(
(
(
3H, d, J = 7.9 Hz, o-CHar), 7.88 (2H, d, J = 7.5 Hz, CHar), 7.59
3H, t, J = 7.7 Hz, CHar), 7.49 (2H, t, J = 7.7 Hz, CHar), 7.35
10H, m, CH ), 7.26 (5H, m, CH ), 6.85 (2H, t, CH ), 2.26 (9H,
ar
ar
ar
1
3
1
s, CH
1
3
); C{ H} (100 MHz, (CD
3
)
2
SO, 25 °C): δ = 159.4 (BiC),
42.4 (C(NH), 141.6 (C(NH)), 133.6 (CHar), 132.5 (CHar), 132.0
(
1
(
CH ), 131.4 (CH ), 130.5 (CH ), 130.4 (CH ), 130.3 (CH ),
7 S. Sundar, Trop. Med. Int. Health, 2001, 6, 849–854.
8 S. Sundar and J. Chakravarty, J. Global Infect. Dis., 2010, 2,
159–166.
9 S. Sundar and P. L. Olliaro, Ther. Clin. Risk Manage., 2007,
3, 733–740.
ar
ar
ar
ar
ar
30.1 (CHar), 124.9 (i-Car), 123.1 (i-Car), 122.5 (CHar), 119.2
+
3
CHar), 117.9 (CHar), 115.2 (CHar), 21.2 (CH ); MS ESI 209.0
+
+
−
[Bi], 391.2 [Bi(m-Tol) ] , 762.3 [Bi(m-Tol) L] ; ESI 236.0
2
3
−
−
[L-COO] , 280.0 [L] ; IR 3256 (br), 3058 (w), 2922 (w), 1583 (sh),
1
1
561 (m), 1513 (m), 1462 (sh), 1462 (w), 1352 (sh), 1333 (sh), 10 S. Yan, F. Li, K. Ding and H. Sun, J. Biol. Inorg. Chem.,
272 (s), 1214 (m), 1160 (sh), 1116 (sh), 1067 (sh), 998 (sh), 979 2003, 8, 689–697.
(
(
sh), 931 (sh), 902 (w), 860 (sh), 796 (sh), 748 (sh), 696 (sh), 665 11 X. Wang and H. Sun, in Comprehensive Inorganic Chemistry
6 2 4 2
sh); Elemental analysis [C49H39BiF N O ·2H O] (1078.28) Cal-
II, ed. J. R. Poeppelmeier, Elsevier, Amsterdam, 2nd edn,
2013, pp. 975–986.
culated C 54.55 H 4.02 N 2.60 Found C 54.72 H 3.65 N 2.71.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
1
1
2 S. Wyllie, M. L. Cunningham and A. H. Fairlamb, J. Biol. 22 D. H. R. Barton, N. Y. Bhatnagar, J.-P. Finet and
Chem., 2004, 279, 39925–39932. W. B. Motherwell, Tetrahedron, 1986, 42, 3111–3122.
3 M. I. Ali, M. K. Rauf, A. Badshah, I. Kumar, C. M. Forsyth, 23 J. Mikus and D. Steverding, Parasitol. Int., 2000, 48, 265–
P. C. Junk, L. Kedzierski and P. C. Andrews, Dalton Trans.,
013, 42, 16733–16741.
4 H. L. Seng and E. R. T. Tiekink, in Comprehensive Inorganic
Chemistry II, ed. J. R. Poeppelmeier, Amsterdam, 2nd edn,
269.
2
24 K. Lackovic, J. P. Parisot, N. Sleebs, J. B. Baell, L. Debien,
K. G. Watson, J. M. Curtis, E. Handman, I. P. Street and
L. Kedzierski, Antimicrob. Agents Chemother., 2010, 54,
1712–1719.
1
2
013, pp. 951–974.
1
1
1
5 P. C. Andrews, R. Frank, P. C. Junk, L. Kedzierski, I. Kumar 25 E. Handman, R. E. Hocking, G. F. Mitchell and
and J. G. MacLellan, J. Inorg. Biochem., 2011, 105, 454–461.
6 P. C. Andrews, P. C. Junk, L. Kedzierski and R. M. Peiris,
Aust. J. Chem., 2013, 66, 1297–1305.
T. W. Spithill, Mol. Biochem. Parasitol., 1983, 7, 111–
126.
26 T. M. McPhillips, S. E. McPhillips, H.-J. Chiu, A. E. Cohen,
A. M. Deacon, P. J. Ellis, E. Garman, A. Gonzalez,
N. K. Sauter, R. P. Phizackerley, S. M. Soltis and P. Kuhn,
J. Synchrotron Radiat., 2002, 9, 401–406.
7 P. C. Andrews, V. L. Blair, R. L. Ferrero, P. C. Junk,
L. Kedzierski and R. M. Peiris, Dalton Trans., 2014, 43,
1
279–1291.
1
1
2
2
8 Y. C. Ong, V. L. Blair, L. Kedzierski and P. C. Andrews, 27 W. Kabsch, J. Appl. Crystallogr., 1993, 26, 795–800.
Dalton Trans., 2014, 43, 12904–12916.
28 CrysAlisPro v1.171.34.36, Oxford Diffraction Ltd (Agilent
9 V. V. Sharutin, O. K. Sharutina and V. S. Senchurin,
Russ. J. Inorg. Chem., 2013, 58, 1470–1474.
Technologies), Oxfordshire, UK, 2010.
29 Bruker Apex2 v2014.7-1, Bruker AXS, Madison, US,
0 G. B. Deacon and R. J. Phillips, Coord. Chem. Rev., 1980, 33,
2014.
2
27–250.
30 G. M. Sheldrick, SADABS v2.30, University of Gottingen,
1 M. Vermeersch, R. I. da Luz, K. Toté, J.-P. Timmermans,
2002.
P. Cos and L. Maes, Antimicrob. Agents Chemother., 2009, 31 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
3, 3855–3859. logr., 2008, 64, 112–122.
5
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015