The Importance of Heterolepticity in Improving the Antibacterial Activity of Bismuth(III) Thiolates

Article abstract of DOI:10.1002/ejic.201600076

Five mixed thiolatobismuth(III) complexes [BiPh(5-MMTD)₂{4-MMT(H)}] (1), [Bi(1-MMTZ)₂{(PYM)(PYM(H))₂}] (2), [Bi(MBT)₂(5-MMTD)] (3), [Bi(4-BrMTD)₃{2-MMI(H)}] (4) and [Bi(1-MMTZ)₂{1-MMTZ(H)}(

Full text of DOI:10.1002/ejic.201600076

DOI: 10.1002/ejic.201600076
Full Paper
Antibacterial Activity
The Importance of Heterolepticity in Improving the
Antibacterial Activity of Bismuth(III) Thiolates
Ahmad Luqman,[a] Victoria L. Blair,[a] Rajini Brammananth,[b] Paul K. Crellin,[b]
Ross L. Coppel,[b] and Philip C. Andrews*[a]
Abstract: Five mixed thiolatobismuth(III) complexes [BiPh- coccus (VRE), Enterococcus faecalis and Escherichia coli showed
(5-MMTD)2{4-MMT(H)}] (1), [Bi(1-MMTZ)2{(PYM)(PYM(H))2}] (2), that mixed thiolato complexes containing the anionic thiazole-
[Bi(MBT)2(5-MMTD)] (3), [Bi(4-BrMTD)3{2-MMI(H)}] (4) and [Bi- based ligands MBT and 4-BrMTD are most effective. The mixed
(1-MMTZ)2{1-MMTZ(H)}(2-MMI){2-MMI(H)2}] (5) were synthesised thiolato complexes [Bi(MBT)2(5-MMTD)] (3) having thiazole- and
from imidazole-, thiazole-, thiadiazole-, triazole-, tetrazole- and thiadiazole- and [Bi(4-BrMBT)3{2-MMI(H)}] (4) containing thiaz-
pyrimidine-based heterocyclic thiones. Four of these complexes ole- and imidazole-based ligands proved to be more efficient,
1–4 were synthesized from BiPh3, while complex 5 was ob- with low minimum inhibitory concentrations of 1.73 and
tained from Bi[4-(MeO)Ph]3. Complexes 1–5 were structurally 3.45 μ
characterised by XRD. Evaluation of the antibacterial properties 2.20 μ
M
for 3 against VRE and E. faecalis, respectively, and
M
for 4 against M. smegmatis and E. faecalis. All complexes
against Mycobacterium smegmatis, Staphylococcus aureus, Meth- showed little or no toxicity towards mammalian COS-7 cell lines
icillin-resistant S. aureus (MRSA), Vancomycin-resistant Entero- at 20 μg mL–1.
Introduction
ato and/or thione ligands on the same Bi atom in the same
complex.
Recently, we have demonstrated that bismuth(III) thiolato com-
plexes derived from a variety of five-membered N-heterocyclic
thiones have powerful antibacterial properties against, among
others, methicillin-resistant Staphylococcus aureus (MRSA) and
Vancomycin-resistant Enterococcus (VRE).[1–3] The most potent
complexes proved to be heteroleptic monophenyl bis-thiolato
complexes, [BiPh(SR)2], several of which incorporated further
neutral thione ligands, [BiPh(SR)2(HSR)2]. These were often sev-
eral orders of magnitude more active than their tris-thiolatobis-
muth(III) analogues [Bi(SR)3]. This is well illustrated by the series
of complexes derived from 4-methyl-4H-1,2,4-triazole-3-thione
[4-MTT(H)]: [BiPh(4-MTT)2{4-MTT(H)}2] has a minimum inhibitory
Heteroleptic, or mixed-ligand, systems are important both in
the development of new metallodrugs, since they can support
incremental and intrinsic changes in the chemophysical nature
of metal complexes, and in understanding metal-based biologi-
cal functions (e.g., metalloenzymes) and deriving biomimetic
systems therefrom.[4] The majority of work has focused on het-
eroleptic transition metal complexes, since they include the ma-
jority of biologically essential metals and have proven to be
highly adaptable in structure and function.[5] Complexes of the
main group metals are less studied. The s-block metals are ubiq-
uitous and essential but highly ionic and form labile complexes.
The p-block metals are often toxic, but continue to find applica-
tion in medicine, infection control and medical imaging.[6,7] Al-
though main group metals lack some of the diversity in oxid-
ation states of the transition metals, they often have more pre-
dictable ligand bonding modes, geometries and structures.
concentration (MIC) towards S. aureus of 3.4 μ
M
, [BiPh(4-MTT)2]
[1]
9.7 μM, and [Bi(4-MTT)3] 180.1 μM.
From a structure–activity
perspective, this suggest that either the Ph group is crucial in
improving the bactericidal activity or, more generally, it is the
presence of two or three different ligands on the bismuth atom
which is the key to improving the bactericidal effects. To probe
these effects further we have now prepared and assayed a
greater diversity of heteroleptic BiIII complexes, limiting the role
of the Ph ligand and constructing them mostly with various
thiolate/thione ligands. Thus, although the complexes are still
derived from N-heterocyclic thiones, they change from being
nominally [BiPh(SR)2] to comprising two or three different thiol-
Our main focus is on bismuth, largely because of its apparent
anomalous status as a heavy metal with good environmental
credentials, and the fact it also displays good antimicrobial
properties with low systemic toxicity in humans.[8–10] Due to
the thiophilic nature of bismuth, biological targets are predomi-
nantly proteins and peptides rich in the S-based amino acids
cysteine and methionine.[11,12] The high lability of Bi–S bonds[13]
coupled with the hydrolytic and thermodynamic stability of bis-
muth thiolates has made them attractive targets in biological
and medicinal chemistry.[14,15] In recent years they have found
application as bactericidal and fungicidal agents,[16–18] and have
been assessed and proposed as effective agents for combating
[a] School of Chemistry, Monash University,
Clayton, Melbourne, VIC 3800, Australia
E-mail: phil.andrews@monash.edu
[b] Department of Microbiology, Monash University,
Clayton, Melbourne, VIC 3800, Australia
Eur. J. Inorg. Chem. 2016, 2738–2749
2738
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
wound infection,[19] as anticancer agents and for retarding bio-
film growth on surfaces.[20]
Several heteroleptic thiolato complexes of bismuth have
been reported previously. For example, Whitmire and co-work-
ers reported [Bi6(pydc)8(Hypdc)2(tu)8] and [{Bi2(pydc)3(tsc)2-
(H2O)2}]·3H2O incorporating thiourea (tu), thiosemicarbazide
(tsc) and 2,6-pyridinedicarboxylic acid (Hpydc) ligands,[21] and
[Bi(SC6F5)3(SPPh3)] and [Bi(SC6F5)3{S=C(NHMe)2}3] were reported
by Norman et al.[22] However, only for the large number of
bis(diorganodithiocarbamato) bismuth(III) complexes incorpo-
rating diorganodithiophosphate and alkylenedithiophosphate
ligands developed by Chauhan and co-workers has the antimic-
robial activity of such complexes been assessed. The majority
of these BiIII complexes showed increased inhibitory activity to-
wards a range of bacteria compared to those of the parent
dithiocarbamates and ancillary ligands.[23]
Figure 1. Heterocyclic thiones employed in the synthesis of heteroleptic bis-
muth(III) complexes 1–5.
Results and Discussion
We now describe the synthesis and characterisation of five
heteroleptic bismuth(III) complexes: [BiPh(5-MMTD)2{4-MMT-
(H)}] (1), [Bi(1-MMTZ)2{(PYM)(PYM(H))2}2] (2), [Bi(MBT)2(5-
MMTD)] (3), [Bi(4-BrMTD)3{2-MMI(H)}] (4), [Bi(1-MMTZ)2(2-MMI)-
{1-MMTZ(H)}{2-MMI(H)2}] (5), all prepared from various reactions
between BiPh3 and a variety of heterocyclic thiones (Figure 1):
2-MMI(H) = 1-methyl-1H-imidazole-2-thione, 4-MMT(H) = 4-
methyl-4H-1,2,4-triazole-3-thione, 1-MMTZ(H) = 1-methyl-1H-
tetrazole-5-thione, 5-MMTD(H) = 5-methyl-1,3,4-thiadiazole-2-
thione, PYM(H) = pyrimidine-2-thione, MBT(H) = 4-phenylthi-
azole-2-thione and Br-MBT(H) = 4-(4-bromophenyl)thiazole-2-
thione. The complexes were fully characterised by NMR and
FTIR spectroscopy and mass spectrometry, and their solid-state
structures were authenticated by single-crystal XRD. Their anti-
bacterial activity was assayed against a range of bacteria: Myco-
bacterium smegmatis, Staphylococcus aureus, MRSA, VRE, Entero-
coccus faecalis and Escherichia coli, and their toxicity to mamma-
lian cells was assessed by using COS-7 monkey liver cells.
The problem in targeting heteroleptic systems in which the li-
gands are anionic or neutral forms of similar classes of com-
pound is the difficulty in controlling the reaction conditions
such that the synthetic outcomes are reproducible. One possi-
bility is simply generating from any reaction mixture a series of
complexes that reflect a statistical distribution of ligands. Even
when using the correct stoichiometry for any target complex,
ligand lability can easily affect the outcome of this distribution.
From a thermodynamic perspective, strictly controlled reaction
conditions and allowing any solution to reach equilibrium be-
come critical. To explore this we studied the impact of the bis-
muth reagent, solvent, reaction stoichiometry and heating
method (reflux vs. microwave). To decrease the number of reac-
tion variables, BiPh3 was generally employed with toluene as
the reaction medium.
From a large range of reactant permutations, several reac-
tions generated reproducible outcomes with the composition
Scheme 1. Synthesis of heteropleptic bismuth(III) thiolato complexes 1–5.
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2739
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
of the complex formed consistent with the applied reactant and single-crystal XRD, all of which were consistent with each
stoichiometry, sometimes independent of the ratio of reactants other. NMR spectra were collected for all complexes in
used. The complexes 1–5 reported here are those for which all [D6]DMSO. The discussion below focuses mainly on the diag-
1
analytical data were consistent with and confirmed by single- nostic aspects of the H NMR spectra, that is, the consistency
crystal XRD. These are shown in Scheme 1 and summarised of ligand form and quantity with the crystal structure. Full ana-
below.
lytical details are provided in the Experimental Section. All anal-
yses were conducted on crystalline materials.
1H NMR Spectroscopy
Synthesis
The chemical shifts and associated integrals in the 1H NMR
spectrum of 1 indicates that it is composed of a residual Ph
ligand, two 5-MMTD thiolato ligands and one neutral 4-MMT(H)
ligand: [BiPh(5-MMTD)2{4-MMT(H)}]. The 4-MMT(H) and 5-MMTD
ligands are identified by singlets at δ = 3.41 and 2.57 ppm,
respectively, for the methyl protons. The singlet at δ = 8.38 ppm
can be assigned to the proton on the triazole ring (δ =
8.32 ppm in the parent compound). The phenyl protons give
peaks at 8.91 (o), 7.80 (m) and 7.41 (p) ppm, which are all shifted
to lower frequency relative to BiPh3 [δ = 7.76 (o), 7.38 (m) and
7.31 (p) ppm]. The disappearance of the signal of the NH proton
of the thiadiazole ring at δ = 12.46 ppm indicates conversion
from thione to thiolate on deprotonation and binding to the
bismuth(III) centre. The neutral triazole NH signal appears at δ =
13.67 ppm.
The monophenylbismuth(III) thiolato complex [BiPh(5-MMTD)2-
{4-MMT(H)}] (1) was prepared by treating BiPh3 with 5-MMTD(H)
and 4-MTT(H) in a 1:2:1 ratio in refluxing toluene for 12 h
(Scheme 1, a). From the resulting solution, which was allowed
to stand for 10 d at room temperature, a batch of yellow crys-
tals appeared. Repeating the same reaction under microwave
irradiation for 10 min at 115 °C improved the yield of crystalline
material to 75 %. Different stoichiometries (2:1:1, 2:2:1 or 1:1:1)
always resulted in complex 1, though the yield and purity dif-
fered.
Complex [Bi(1-MMTZ)2{(PYM)(PYM(H))2}] (2) was synthesised
in a similar manner by the reaction of BiPh3 with 3 equiv. of
PYM(H) and 2 equiv. of 1-MMTZ(H). Various PYM(H):1-MMTZ-
(H):BiPh3 ratios (2:1:1 or 2:2:1) always resulted in complex 2 as
the main product (Scheme 1, b). The resulting solution was al-
lowed to stand for 10 d at room temperature and yielded or-
ange crystals (61 %). From the analogous reaction with micro-
wave heating the yield of crystals was 65 %.
The chemical shifts and integrals of the ligands in complex
2
indicate that its composition is [Bi(1-MMTZ)2{(PYM)-
(PYM(H))2}]. The methyl protons of the 1-MMTZ tetrazole thiol-
ato ligand appear at δ = 3.80 ppm, which is only marginally
shifted downfield compared to the parent thione (δ =
3.78 ppm). The doublet at δ = 8.56 ppm and the triplet at δ =
7.27 ppm correspond to the protons of the pyrimidine ring,
which are shifted downfield compared to the parent thione
(8.27 and 6.83 ppm respectively).
A 1:2:1 reaction of BiPh3, MBT(H) and 5-MMTD(H) (Scheme 1,
c) in toluene at reflux resulted in crystalline material, which ana-
lysed as [Bi(MBT)2(5-MMTD)] (3). The orange crystals of 3 depos-
ited as the hot toluene solution cooled to room temperature
over 5 h in a yield of 69 %. A similar yield of 72 % was obtained
by conducting the analogous reaction with microwave heating
for 10 min.
1
In the H NMR spectrum of [Bi(MBT)2(5-MMTD)] (3) the ab-
sence of peaks for the NH protons of the parent thiones MBT(H)
or 5-MMTD(H) (normally found at ca. 13.1 and 12.6 ppm, respec-
tively) suggests that both ligands exist as thiolate anions. The
relative integrals gave a 2:1 ratio for MBT/5-MMTD, which thus
reflects the reaction stoichiometry. The CH3 protons of the thia-
diazole 5-MMTD ligand resonate at almost the same frequency
as those in the parent thione (2.54 vs. 2.52 ppm), while the
singlet corresponding to the ring proton in the MBT anion ap-
pears at δ = 7.50 ppm, shifted from 8.53 ppm in the spectrum
of the thiazole thione. Only the signals of the ortho protons in
the Ph moiety of the MBT ligand shift significantly, and move
to 7.86 ppm from 7.42 ppm in MBT(H).
Complex [Bi(4-BrMTD)3{2-MMI(H)}] (4) was synthesised by
treating BiPh3 with 4-BrMTD(H) and 2-MMI(H), in 1:3:1 ratio in
refluxing toluene for 7 h (Scheme 1, d). The resulting solution
was allowed to stand for 7 d at room temperature and yielded
orange crystals in 58 % yield. From the analogous reaction with
microwave heating in toluene at 115 °C for 10 min, a 65 % yield
of crystals was obtained after 7 d.
In contrast to 1–4, [Bi(1-MMTZ)2(2-MMI){1-MMTZ(H)}{2-MMI-
(H)2}] (5) was synthesised by using Bi(Ph-4-OMe)3 as the aryl
bismuth precursor rather than BiPh3 (Scheme 1, e). Reactions of
BiPh3 with thiones 2-MMI(H) and 1-MMTZ(H) only resulted in
mixtures with monophenyl bismuth complexes as the main
constituent. The more reactive Bi[4-(MeO)Ph]3 was therefore
used and provided access to the target tris-thiolato species.
Heating a 1:3:3 mixture of Bi[4-(MeO)Ph]3/2-MMI(H)/1-MMTZ(H)
in toluene to reflux or under microwave irradiation gave orange
solutions from which crystals of 5 were deposited after two
weeks in yields of 71 and 74 % respectively.
Complex 4 exhibits resonances for both 2-MMI(H) and 4-
BrMTD, integrating in a ratio of 1:3. In the parent thiones, the
chemical shifts of the NH protons in 2-MMI(H) and 4-BrMTD(H)
appear at δ = 13.76 and 12.00 ppm respectively. Only the signal
1
at δ = 12.03 ppm remains in the H NMR spectrum of 4, that
is, the 4-BrMTD ligand is anionic and 2-MMI(H) remains neutral.
Signals at 7.72 (o- and m-Ph) and 7.39 (thiazole CH) ppm are
assigned to protons of 4-BrMTD, while the protons on the imid-
azole ring of 2-MMI(H) appear as doublets at δ = 7.04 and
6.87 ppm.
Characterisation
1
The compositions and structures of 1–5 were confirmed by H
In [Bi(1-MMTZ)2(2-MMI){1-MMTZ(H)}{2-MMI(H)2}] (5), integra-
and 13C NMR and FTIR spectroscopy, ESI-MS, elemental analysis tion of the proton signals indicated an overall ratio of the tetraz-
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2740
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ole to imidazole ligands of 1:1. One difficulty is that the diag-
1
nostic signal for NH in the tetrazole is not observable in the H
NMR spectrum of the parent thione 1-MMTZ(H) or in that of 5.
However, the NH signal of 2-MMI(H) is found at δ = 12.03 ppm
in the spectra of the parent thione and 5. Further, the backbone
CH protons resonate at δ = 7.06 and 6.85 ppm in the neutral
ligand and at δ = 7.06 and 6.88 ppm in the thiolate form. For
the tetrazole ligand the shift of the CH3 peak on changing from
the neutral form to the anion is marginal (3.91 vs. 3.89 ppm).
Thus, for complex 5 the composition was only really confirmed
by XRD.
XRD Studies
Crystals of complex 1 grew over several weeks from DMSO solu-
tion and were analysed by single-crystal XRD and authenticated
as [BiPh(5-MMTD)2{4-MMT(H)}]2. Complex 1 crystallises in the
monoclinic space group P21/n. The molecular structure is
shown in Figures 2 and 3, and selected bond lengths and an-
gles are listed in Table 1.
Figure 3. Dimeric structure of 1 with thermal ellipsoids shown at 50 % proba-
bility. Hydrogen atoms (except NH) have been omitted for clarity. Symmetry
operator ′ = 2 – x, –y, –z.
Table 1. Selected bond lengths [Å] and angles [°] for 1.
Bi(1)–C(1)
Bi(1)–S(1)
Bi(1)–S(4)
Bi(1)–S(2)
2.267(6)
2.8507(16)
2.7772(16)
2.6763(17)
1.709(6)
1.741(6)
1.717(6)
2.777(5)
2.777(5)
86.66(16)
87.4(3)
C(1)–Bi(1)–S(4)
C(1)–Bi(1)–S(1)
S(2)–Bi(1)–S(4)
S(2)–Bi(1)–S(1)
S(4)–Bi(1)–S(1)
C(7)–S(1)–Bi(1)
C(10)–S(2)–Bi(1)
C(13)–S(4)–Bi(1)
C(1)–Bi(1)–N(6)′
N(6)′-Bi(1)–S(1)
S(2)–Bi(1)–N(6)′
86.42(16)
88.303(4)
93.81(5)
96.39(5)
167.16(5)
97.65(19)
96.0(2)
S(1)–C(7)
S(2)–C(10)
S(4)–C(13)
Bi(1)–N(6)′1
N(6)–Bi(1)′
C(1)–Bi(1)–S(2)
C(10)–S(3)–C(11)
93.7(2)
91.92(19)
80.78(12)
176.92(12)
S(2)–C(10) 96.02(2) and Bi(1)–S(4)–C(13), 93.7(2)°] are compara-
ble to those of previously synthesized complexes (av. 96.13°).[1]
Dimerisation of two monomeric units takes place through
a bismuth–bridging nitrogen interaction (Bi(1)–N(6)′ 2.77(5) Å;
Figure 3), which increases the coordination number of the bis-
muth(III) centre to five and results in an overall distorted
square-based pyramidal coordination geometry and formation
of a central eight-membered Bi2S2N2C2 ring.
Figure 2. Molecular structure of 1 with thermal ellipsoids shown at 50 %
probability. Hydrogen atoms (except NH) have been omitted for clarity.
Complex 1 is a dimeric, homometallic complex containing
two deprotonated molecules of 5-MMTD, a neutral 4-MTT(H)
molecule and one phenyl group (Figure 2). The Bi–S bond
lengths [Bi(1)–S(1) 2.8507(16), Bi(1)–S(2) 2.6763(17) and Bi(1)–
S(4) 2.7772(16) Å] are comparable to those of the complex
[BiPh(4-MMT)2{4-MMT(H)}2][24] containing triazole moieties (av.
Orange needle crystals suitable for X-ray diffraction studies
2.8055 Å) and longer than those of thiadiazole-containing com- were obtained from a toluene solution of 2 after 7 d. The com-
plex [Bi2(5-MMTD)4]∞ (av. 2.642 Å).[1] The molecular structure plex crystallizes in the triclinic space group P1¯. The asymmetric
reveals a trigonal-pyramidal geometry in which the phenyl unit is shown in Figure 4, and selected bond lengths and angles
group lies perpendicular to the BiS3 plane. The structure is sta- are listed in Table 2. The geometry around the bismuth(III) cen-
bilised by intramolecular hydrogen bonds between H(3N) of the tre is distorted pentagonal-bipyramidal with a coordination
4-MTT(H) and N(4) of the 5-MMTD ligand [H(3N)–N(4), number of seven. The tetrazole-containing ligand binds prima-
1.93(3) Å]. The Bi–S–C angles [Bi(1)–S(1)–C(7) 97.65(19)°, Bi(1)– rily through its S– centre, which confirms the predominance of
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2741
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
the thiol form. This is in contrast to pyrimidine-based ligands, lengths are in the range of 2.902(16)–3.047(15) and 2.542(5)–
which mainly exist in the thione form.
2.621(5) Å, respectively. Four hydrogen atoms on N(2), N(4), N(5)
and N(6) were located by residual density in the electron-den-
sity difference maps and placed in calculated positions. Each
hydrogen atom has half-occupancy due to disorder, so in total
they only contribute +2 to balancing the overall charge on the
bismuth centre.
Needle-like crystals of 3 were obtained after two weeks by
slow evaporation of a toluene solution. Single-crystal analysis
revealed the formula to be [Bi(MBT)2(5-MMTD)]. The complex
crystallizes in the monoclinic space group P21/c. The asymmet-
ric unit is shown in Figure 5 with selected bond lengths and
angles listed in Table 3. In the asymmetric unit the mixed thiol-
ato complex 3 is composed of one 5-MMTD and two MBT che-
lating ligands bonded covalently to the bismuth(III) centre
through the deprotonated thiol groups and datively through
the N atoms in the thiazole and thiadiazole moieties. The Bi–S
bond lengths [Bi(1)–S(1) 2.599(14), Bi(1)–S(3) 2.630(15) and
Bi(1)–S(5) 2.581(16) Å] are similar to those of the tris-thiolatobis-
muth(III) complex [Bi(MBT)3] (av. 2.591 Å),[2] the thiadiazole-con-
taining complex [Bi2(5-MMTD)4]∞ (av. 2.642 Å)[1] and
[Bi(SC6F5)3(SPPh3)] (av. 2.581 Å). The C–S [S(1)–C(1) 1.726(6),
S(5)–C(19) 1.728(6) and S(3)–C(9) 1.687(6) Å] and C–N bond
lengths [N(1)–C(1) 1.310(7), N(2)–C(10) 1.315(7) and N(3)–C(19)
1.301(8) Å] have single-bond character and are comparable
with the reported values [S–C 1.699(14) and C–N
1.309(12) Å].[25] All the ligands form four-membered metallacy-
clic rings due to chelation through the N and S donor atoms
with small bite angles [S(1)–Bi(1)–N(1) 59.35(10), S(3)–Bi(1)–N(2)
58.37(10) and S(5)–Bi(1)–N(3) 58.70(12)°].
Figure 4. Molecular structure of 2 with thermal ellipsoids shown at 50 %
probability. Four hydrogen atoms are shown, but each has only 0.5 occu-
pancy, so in total they only contribute +2 to balancing the overall charge on
the bismuth(III) centre.
Table 2. Selected bond lengths [Å] and angles [°] for 2.
Bi(1)–N(1)
Bi(1)–N(3)
Bi(1)–S(1)
Bi(1)–S(2)
Bi(1)–S(3)
Bi(1)–S(4)
Bi(1)–S(5)
S(1)–C(1)
S(2)–C(5)
S(3)–C(9)
S(4)–C(13)
S(5)–C(15)
N(3)–C(5)
N(5)–C(9)
N(1)–C(1)
2.542(5)
2.621(5)
N(1)–Bi(1)–S(1)
N(1)–Bi(1)–S(2)
N(1)–Bi(1)–S(4)
N(1)–Bi(1)–S(5)
N(3)–Bi(1)–S(1)
N(3)–Bi(1)–S(2)
N(3)–Bi(1)–S(4)
N(3)–Bi(1)–S(5)
C(1)–S(1)–Bi(1)
C(5)–S(2)–Bi(1)
C(9)–S(3)–Bi(1)
C(13)–S(4)–Bi(1)
C(15)–S(5)–Bi(1)
S(1)–Bi(1)–S(3)
S(2)–Bi(1)–S(3)
57.10(10)
136.96(10)
72.26(12)
83.48(10)
137.75(10)
56.65(10)
88.51(11)
83.30(11)
81.21(18)
83.57(19)
108.22(19)
102.51(18)
90.17(18)
103.19(5)
82.26(5)
2.9123(17)
2.9029 (16)
3.0472(15)
2.6472(15)
2.8223(15)
1.708(6)
1.715(5)
1.687(6)
1.722(6)
1.734(6)
1.364(7)
1.360(7)
1.358(6)
The Bi–S bond lengths of the deprotonated tetrazole thiol-
ates [Bi(1)–S(4) 2.647(15), Bi(1)–S(5), 2.822(15) Å] are comparable
to those of tetrazole-containing complex [BiPh(1-MMTZ)2-
{1-MMTZ(H)}] (av. 2.72 Å).[24] The C–S [S(4)–C(13) 1.722(3), S(5)–
C(15) 1.734(6) Å] and C–N bond lengths [N(7)–C(13) 1.347(7),
N(11)–C(15),1.330(7) Å] show single-bond character. Two of the
three pyrimidine-containing ligands chelate the Bi centre
through their N and S atoms with formation of two four-mem-
bered metallacyclic, rings while the third pyrimidine molecule
binds to the bismuth(III) centre in a monodentate mode involv-
Figure 5. Asymmetric unit of [Bi(MBT)2(5-MMTD)] 3 with thermal ellipsoids
shown at 50 % probability. Hydrogen atoms have been omitted for clarity.
ing only the
S
atom. The C(9)–S(3)–Bi(1) bond angle
Dimerisation of two monomeric units of 3 occurs through a
[108.22(19)°] of the S-bonded pyrimidine-containing ligand is long interaction between the bismuth centre and π electrons
larger than the C(1)–S(1)–Bi(1) [81.21(18)°] and C(5)–S(2)–Bi(1) of an adjacent aromatic ring of MBT (Figure 6). This increases
[83.57(19)°] bond angles of the chelating ligands. Thus, the the coordination number of the bismuth(III) centre to seven and
binding mode of the pyrimidine-2-thione can be judged by gives an overall distorted pentagonal-bipyramidal coordination
comparing these C–S–Bi bond angles. The Bi–S and Bi–N bond geometry.
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2742
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 3. Selected bond lengths [Å] and angles [°] for 3.
Bi(1)–N(1)
Bi(1)–N(2)
Bi(1)–N(3)
Bi(2)–N(4)
Bi(2)–N(5)
Bi(2)–N(6)
Bi(2)–N(7)
Bi(1)–S(1)
Bi(1)–S(3)
Bi(1)–S(5)
Bi(2)–S(11)
Bi(2)–S(9)
Bi(2)–S(7)
S(1)–C(1)
2.804(5)
2.879(5)
2.864(5)
7.089(6)
2.957(5)
2.847(5)
2.881(5)
2.5994(14)
2.6303(15)
2.5814(16)
2.5771(15)
2.5975(15)
2.6008(15)
1.726(6)
N(3)–C(19)
N(4)–C(20)
1.301(8)
1.304(9)
S(1)–Bi(1)–N(1)
S(3)–Bi(1)–N(2)
S(5)–Bi(1)–N(3)
S(7)–Bi(2)–N(5)
S(9)–Bi(2)–N(6)
S(11)–Bi(2)–N(7)
C(1)–S(1)–Bi(1)
C(10)–S(3)Bi(1)
C(19)–S(5)–Bi(1)
C(22)–S(7)–Bi(2)
C(31)–S(9)–Bi(2)
C(40)–S(11)–Bi(2)
59.35(10)
58.37(10)
58.70(12)
57.84(10)
59.06(10)
58.55(11)
87.66(19)
89.27(18)
89.6(2)
91.13(19)
88.65(19)
89.59(19)
Figure 7. Molecular structure of 4 with thermal ellipsoids show at 50 % proba-
bility. Selected hydrogen atoms removed for clarity.
Table 4. Selected bond lengths [Å] and angles [°] for 4.
Bi(1)–S(1)
Bi(1)–S(3)
Bi(1)–S(5)
Bi(1)–S(7)
S(1)–C(1)
S(3)–C(10)
S(5)–C(19)
S(7)–C(28)
N(1)–C(1)
N(2)–C(10)
N(3)–C(19)
N(4)–C(28)
Br(1)–C(7)
2.5899(10)
2.6692(11)
2.5890(11)
3.0227(13)
1.739(4)
1.735(4)
1.734(4)
1.720(5)
1.306(5)
1.319(5)
1.314(5)
1.335(6)
1.908(4)
90.88(14)
90.07(15)
92.01(15)
99.50(15)
92.15(4)
97.75(4)
77.91(4)
94.94(4)
100.20(4)
124.7(4)
123.5(3)
115.2(3)
127.6(4)
C(1)–S(1)–Bi(1)
C(10)–S(3)–Bi(1)
C(19)–S(5)–Bi(1)
C(28)–S(7)–Bi(1)
S(1)–Bi(1)–S(3)
S(1)–Bi(1)–S(7)
S(5)–Bi(1)–S(1)
S(5)–Bi(1)–S(3)
S(5)–Bi(1)–S(7)
N(2)–C(10)–S(3)
N(1)–C(1)–S(1)
N(3)–C(19)–S(6)
N(5)–C(28)–S(7)
Figure 6. Dimeric [Bi(MBT)2(5-MMTD)]2 3 with thermal ellipsoids shown at
50 % probability. Hydrogen atoms have been omitted for clarity.
Single crystals of complex 4 grew over a few weeks from
toluene solution and were authenticated as [Bi(4-BrMBT)3{2-
MMI(H)}]. The molecular structure of complex 4 is shown in
Figure 7 with selected bond lengths and angles listed in Table 4.
The complex crystallises in the triclinic space group P1¯ with a
four-coordinate trigonal-pyramidal geometry. Complex 4 has
one neutral 2-MMI(H) and three deprotonated 4-BrMTD ligands.
The binding of the 4-BrMTD ligand is primarily through its S–
Orange crystals suitable for X-ray diffraction studies were ob-
centre. The Bi–S bond lengths of the 4-BrMTD ligand lie in the tained from a toluene solution of 5 after 6–7 d. The complex
range of 2.589(10)–2.669(11) Å and are comparable with those crystallizes in the triclinic space group P1. The molecular struc-
¯
of the tris-thiolato complex [Bi(4-BrMBT)3] (av. 2.601 Å)[2] and ture was established by X-ray crystallography and with the help
[Bi(2-SC5H3NSiMe)3] (av. 2.626 Å).[26] The C–S bond lengths of of NMR spectroscopy; hydrogen atoms were located by residual
4-BrMTD range from 1.734(5) to 1.739(4) Å, which are longer electron density and placed in calculated positions (Figure 8).
than the thionic C–S bond length of 1.720(5) Å in 2-MMI(H). Selected interatomic parameters are summarised in Table 5.
The C–N bond lengths of the 4-BrMBT ligands [N(1)–C(1) Complex 5 contains one neutral and two tetrazole thiolate and
1.306(5), N(2)–C(10) 1.319(5) and N(3)–C(19) 1.314(5) Å] are two neutral and one imidazole thiolate ligands. The geometry
shorter than those of neutral 2-MMI(H) [N(4)–C(28) 1.335(5) Å]. around the bismuth(III) centre is a distorted octahedron with a
The Bi–S bond length of 2-MMI(H) [Bi(1)–S(7) 3.0227(13) Å] is coordination number of six. The binding of two 1-MMTZ and
comparable to the reported values for coordinate dative Bi←S one 2-MMI ligand is primarily through the deprotonated sulfur
bonds (ca. 3.0 Å) and is longer than those of 4-BrMBT (Table 4). atom of the thiol group, while the remaining ligands bind in a
These somewhat different Bi–S, C–S and C–N bond lengths sup- neutral fashion. The Bi–S bond lengths for the three thiolato
port the different coordination modes of 4-BrMBT and 2-MMI(H) ligands [Bi(1)–S(2) 2.775(2), Bi(1)–S(6) 2.711(2) and Bi(1)–S(5)
in the complexes.
2.791(2) Å] are comparable with those of [BiPh(1-MMTZ)2{1-
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2743 © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
MMTZ(H)}2] (av. 2.724 Å) but slightly shorter than those of crystal owing to the possibility of the three calculated NH pro-
[BiPh(2-MMI)2{2-MMI(H)}2] (av. 2.812 Å).[24] In the neutral 2- tons on the neutral ligands being located on any three of the
MMI(H) ligands, the Bi–S bond lengths of 2.969(2) and heterocyclic ligands at any one time. The structure is stabilised
3.015(2) Å are significantly longer than those in 5, which are in by intramolecular hydrogen bonding [H(9N)–N(13) 1.93(11),
the range of 2.9173(12)–2.9476(10) Å. In the neutral 1-MMTZ(H) H(11N)–N(5) 1.99(12) and H(4N)–N(17) 2.01(11) Å, which lie in
ligand, the Bi–S bond length is slightly shorter than that in the range of typical hydrogen bonds].
[BiPh(1-MMTZ)2{1-MMTZ(H)}2] (av. 2.934 Å).[24] Thus, they consti-
tute an average between typical covalent Bi–S (ca. 2.6 Å)[27,28]
and coordinate dative Bi←S (ca. 3.0 Å) bonding modes.[29,30]
Antibacterial Studies
This may be due to averaging of the six bonds throughout the
All the parent thiones and their BiIII complexes were tested
against six different strains of bacteria: M. smegmatis, S. aureus,
MRSA, VRE, E. faecalis and E. coli. For comparison, the activity
of BiPh3 was also assessed. The solvent used for the preparation
of sample solutions of required concentration was DMSO. Iso-
niazid (isonicotinyl hydrazide), a first-line drug for the treatment
of tuberculosis, and vancomycin (clinical MIC = 2 μg/mL,
1.38 μM
) were used as controls.[31–34]
The MIC is the minimal concentration of a drug that still
completely inhibits bacterial growth in culture. The in vitro ac-
tivities of all the compounds are listed in Table 6. Most of these
heterocyclic, mixed thiolatobismuth complexes demonstrated
significant antimicrobial activity against gram-positive bacteria,
that is M. smegmatis, S. aureus, MRSA, VRE and E. faecalis.
Complexes 1–4 proved to be highly active against of all the
gram-positive bacteria tested. Bismuth compounds[35–38] tend
to be less effective against gram-negative bacteria such as E.
coli, presumably because of difficulties caused by the double
membrane. However, some complexes do show good activity,
and in this study complexes 1, 3 and 4 showed activity at rela-
tively low MICs.
Figure 8. Molecular structure of 5 with thermal ellipsoids shown at 50 %
probability. Hydrogen atoms (except NH) have been omitted for clarity.
Table 5. Selected bond lengths [Å] and angles [°] for 5.
Complex 1 contains one phenyl, two anionic thiadiazole thi-
olato ligands and one neutral triazole thione ligand, and shows
good activity towards all the gram-positive bacteria with MIC ≤
Bi(1)–S(1)
Bi(1)–S(2)
Bi(1)–S(3)
Bi(1)–S(5)
Bi(1)–S(4)
Bi(1)–S(6)
S(1)–C(1)
S(2)–C(3)
S(3)–C(5)
S(4)–C(9)
S(5)–C(13)
S(6)–C(15)
N(4)–C(1)
2.693(2)
2.775(2)
3.015(2)
2.791(2)
2.969(2)
2.711(2)
1.743(10)
1.718(10)
1.698(9)
1.708(12)
1.714(10)
1.704(10)
1.305(11)
N(9)–C(5)
N(13)–C(13)
N(11)–C(9)
N(17)–C(15)
C(1)–S(1)–Bi(1)
C(3)–S(2)–Bi(1)
C(5)–S(3)–Bi(1)
C(9)–S(3)–Bi(1)
C(13)–S(5)–Bi(1)
C(15)–S(6)–Bi(1)
S(1)–Bi(1)–S(6)
S(1)–Bi(1)–S(2)
S(6)–Bi(1)–S(2)
1.351(12)
1.335(13)
1.332(13)
1.336(13)
103.2(3)
99.0(3)
105.6(3)
108.2(19)
98.8(3)
110.3(3)
2.693(2)
93.04(8)
75.94(8)
7.49 μ
of 13.4 μ
M
. It also shows good activity towards E. coli with an MIC
. The most closely related complexes with which to
M
judge the impact of the various ligands are our previously de-
scribed bis- and tris-thiolato complexes [BiPh(5-MMTD)2] and
[Bi(5-MMTD)3].[1] Notably, [Bi(5-MMTD)3] performs poorly with
an MIC of 66.23 μM for S. aureus, MRSA and VRE and 165.56 μM
for M. segamatis, E. faecilis and E. coli. Surprisingly, the
heteroleptic complex [BiPh(5-MMTD)2] is even more ineffective
towards MRSA and VRE (180.23 μM) but shows better activity
Table 6. Antibacterial activities of mixed-ligand bismuth(III) thiolate complexes 1–5 against M. smegmatis, S. aureus, MRSA, VRE, E. coli and E. faecalis.
MIC [μg/mL (μ )]
M
M. smegmatis
S. aureus
MRSA
VRE
E. faecalis
E. coli
Isoniazid
1 (7.29)
>100
>100
>100
>100
>100
4-MMT(H)
>100
5-MMTD(H)
>100
1-MMTZ(H)
>100
PYM(H)
>100
MBT(H)
>100
4-BrMTD(H)
>100
1
2
3
4
5
2.5 (3.74)
2.5 (3.25)
>100 (>138)
2.5 (2.20)
10 (13.5)
5.0 (7.49)
5.0 (6.50)
5.0 (6.90)
5.0 (4.40)
>100 (>135)
5.0 (7.49)
10 (13.0)
5.0 (6.90)
5.0 (4.40)
>100 (>135)
2.5 (3.74)
5.0 (6.50)
1.25 (1.73)
5.0 (4.40)
>100 (>135)
2.5 (3.74)
5.0 (6.50)
2.5 (3.45)
2.5 (2.20)
10 (13.5)
10 (15.0)
>100 (>130)
20 (27.6)
10 (8.81)
>100 (>135)
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2744
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
towards S. aureus (18.23 μ
(2.73 μ ) and E. coli (36.46) μ
ligand alone is not particularly good at facilitating antimicrobial ligand, as the MICs for [BiPh(MBT)2] and [Bi(MBT)3] are 127.3
activity and that heteroleptic complexes generally perform bet- and 149.1 μ respectively.
M
), M. smegamatis (4.56 μ
M
), E. faecilis VRE (both 171.07 μ
M
). The activity of the complex towards E.
M
M
. This suggests that the 5-MMTD coli (27.6 μ
M) is almost certainly predicated on the 5-MMTD
M
ter. The introduction of the thione ligand 4-MMT(H) both main-
tains and increases the bactericidal activity towards the gram-
positive bacteria and E. coli. The complex [BiPh(4-MMT)2{4-
MMT(H)}2], which also contains the triazole in its neutral form,
is highly active against all the bacteria tested, with MIC ranging
Similarly, complex 4, containing thiazole- and imidazole-
based ligands, also shows excellent activity with MIC values of
4.40 μ
E. faecalis. This level of activity towards S. aureus and MRSA is
comparable with that of vancomycin (1.38 μ
).[31] In compari-
M towards S. aureus, MRSA and VRE, and 2.20 μM towards
M
from 0.34 μM for VRE to 13.40 μM for E. coli. Related complexes
son with the tris-thiolato parent complex [Bi(4-BrMTD)3] the ef-
fect of introducing the neutral 2-MMI(H) ligand on the Bi atom
is significant. The MIC for [Bi(4-BrMTD)3] against S. aureus is
with the ligand in its anionic form, 4-MMT–, follow the typical
pattern of losing effectiveness as the number of this ligand in-
creases. As such, the MICs for [BiPh(4-MMT)2] range from
9.78 μ
M
, and that against MRSA is ten times higher (97.8 μ
M).
9.72 μ
smegmatis), and they become significantly worse for [Bi(4-
MMT)3] (180.13 μ for MRSA, VRE, E. coli and 18.13 μ for S.
aureus, E. faecalis), except towards M. smegmatis (9.07 μ ).
M (MRSA, S. aureus, E. faecalis) to 19.44 μM (E. coli, VRE, M.
Against the other four bacteria the MIC is 39.14 μ
M
. The mono-
phenyl bis-thiolato complex [BiPh(4-BrMTD)] is significantly
more active towards all the bacteria tested than [Bi(4-BrMTD)3],
though not as potent as 4, except in the case of S. aureus
M
M
M
(1.21 μ
M
).[1] For comparison, the MIC for [BiPh(4-BrMTD)] to-
Complex 2 contains two thiolato anions (1-MMTZ) derived
from 1-methyl-1H-tetrazole-5-thione, and also incorporates one
anionic (PYM) and two neutral ligands of pyrimidine-2-thione
[PYM(H)] in its thione form. The complex demonstrates good
[2]
wards both MRSA and VRE was 6.06 μ
M
.
Interestingly, com-
plexes containing the imidazole ligand as an anion, [BiPh(2-
MMI)2] and [Bi(2-MMI)3], and also in its neutral form, [BiPh(2-
MMI)2{2-MMI(H)}2], are poor antibacterial agents.[1] None of the
complexes are particularly active towards S. aureus, MRSA, VRE
and E. coli, and all three complexes have MICs greater than
100 μg mL–1 for each of these bacteria. Thus, the dominant li-
gand in 4 is the thiazole 4-BrMTD, and this further supported by
comparing the activity of 4, [BiPh(4-BrMTD)] and [Bi(4-BrMTD)3]
towards E. coli, for which they have MICs of 8.81, 48.5 and
activity towards S. aureus, VRE and E. facealis at 6.50 μ
M and
moderately good activity towards MRSA at 13.0 μ . This is con-
M
sistent with our previous study that demonstrated that the het-
eroleptic bis-tetrazole thiolato complexes [BiPh(1-MMTZ)2{1-
MMTZ(H)}2] and [BiPh(1-MMTZ)2] are significantly more active
towards all the bacteria (ranges of 1.33–3.34 μM and 4.84–
9.68 μ , respectively, for gram-positive bacteria) than the ho-
M
39.14 μM, respectively.
moleptic tris-thiolato complex [Bi(1-MMTZ)3], which shows very
Complex 5 is the most ineffective of the five complexes.
Structurally it is similar to 2 in that it contains two thiolato
anions (1-MMTZ) derived from 1-methyl-1H-tetrazole-5-thione.
However, the only similarities are in the activity of the two com-
little activity towards S. aureus, MRSA, VRE and E. coli (all
180 μ
wards M. smegmatis (1.33–9.07 μ
with 2 (3.13 μ ).
M
).[1] However, all three complexes are highly active to-
M
), which is largely consistent
M
plexes towards M. smegmatis (3.13 and 13.5 μ
E. faecelis (6.25 and 13.5 μ , respectively) and their inactivity
towards E. coli (> 130 μ ). Surprisingly, unlike 1–4 it shows no
activity towards S. aureus, MRSA or VRE at the upper concentra-
tion tested (MIC > 135 μ ). This suggests the coligands, one
M, respectively),
While many metal complexes containing the neutral PYM(H)
ligand have been structurally characterised, only one complex
containing the anionic PYM ligand has been characterised.[37]
The only antibacterial studies associated with metal complexes
of this ligand in any form were performed on a series of Cu and
Ag complexes of form [M(PPh3)n{PYM(H)}m]X (M = Cu; X = BF4;
n = 3, m = 1; n = m = 2M = Ag; X = NO3; n = 3, m = 1; n = 2,
m = 1).[39] The Ag complexes were more active towards S. au-
reus than the Cu complexes (8.0 vs. 24 μg/mL), and their activi-
ties were similar to that of complex 2 (5.0 μg/mL). The Ag com-
plexes are also active towards E. coli (7.5 μg/mL), but 2 and the
Cu complexes are not, and this suggests that PYM(H) is not the
determining factor in this.
M
M
M
anionic (2-MMI) and two neutral ligands [2-MMI(H)] of 1-methyl-
1H-imidazole-2-thione, have a dramatic effect. Previous BiIII
complexes we have studied containing either of these ligands
have demonstrated generally poor antibacterial properties (MIC
≥ 135 μ
moleptic.[1] The exceptions were towards M. smegmatis (13.5
and 19.5 μ for {BiPh(2-MMI)2[2-MMI(H)]2} and [BiPh(2-MMI)2],
respectively) and E. faecalis (13.5 and 9.75 μ for {BiPh(2-
M), irrespective of whether they are heteroleptic or ho-
M
M
MMI)2[2-MMI(H)]2} and [BiPh(2-MMI)2], respectively). Therefore,
this may contribute to the effectiveness of 5 towards M. smeg-
matis and E. faecalis (both 13.5 μM).
Complex 3, having thiazole and thiadiazole moieties, shows
excellent activity towards S. aureus, MRSA, VRE and E. faecalis
with MICs of 6.90, 6.90, 1.73 and 3.45 μM, respectively. These
MICs align closely with those observed for the two analogous
thiazole-based complexes studied previously, [Bi(MBT)3] and
[BiPh(MBT)2].[2] The sole exception is the activity of [Bi(MBT)3]
towards VRE, for which the MIC was significantly higher
Mammalian Cell Toxicity
Although a strong antibacterial activity is essential for potential
new therapeutic agents, the compound must have low toxicity
towards human or animal eukaryotic cells. Toxicity assays of all
(59.6 μM). Considering the role of the 5-MMTD ligand, it is inter-
esting that the previously described complex [BiPh(5-MMTD)2] the complexes were conducted on cultured COS-7 cell lines.
showed good activity towards M. smegmatis (4.56 μ ) and E. These revealed the complexes to be non-toxic to mammalian
faecalis (2.73 μ ) but was largely ineffective against MRSA and cells at 20 μg/mL (Figure 9).
M
M
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2745
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ether. Ethanol, DMSO and toluene were used as solvents for recrys-
tallisation. Diethyl ether and THF were dried prior to use with an
M-Braun-SPS-800 solvent purification system and stored over mo-
lecular sieves (4 Å). Microwave irradiation was carried out in a CEM
Discoverer Microwave oven with 0–300 W power, model number
9080, maximum current 6.38 A with 50/60 MHz frequency. NMR
spectra of complexes were recorded with a Bruker DPX 400 spec-
trometer in [D6]DMSO with TMS as internal standard at room tem-
perature. Elemental analysis (C, H and N) was performed by the
Campbell Microanalytical lab, Department of Chemistry, University
of Otago, Dunedin, New Zealand. Melting points were determined
with a Stuart Scientific SMP3 melting point apparatus. Mass spectra
were recorded with Micromass Platform Electrospray Mass Spec-
trometer. The atom-labelling scheme used for NMR chemical shift
assignments is shown in Figure 10.
Crystallographic Data: Crystallographic data of compounds 1–3
was collected at the MX1 beamline comprising a Φ goniostat with
an ADSC Quantum 210r detector at the Australian Synchrotron,
Melbourne, Victoria, Australia. All data were collected at 100 K,
maintained by using an open flow of nitrogen. The software used
for collection and reduction of the data was BlueIce,[40] and data
were processed with XDS.[41] Crystallographic data of compounds
4 and 5 were collected with an Oxford Gemini Ultra equipped with
an Oxford Cryosystems 700 Cryostream [123(1) K]. Data were col-
Figure 9. Toxicity assays of compounds 1–5, isoniazid (INH) and ethambutol
(EMB) against cultured COS-7 cell lines.
Conclusions
Five new heteroleptic bismuth(III) thiolato complexes 1–5 were
synthesised and fully characterised. Crystals were grown from
DMSO (1) or toluene solution (2–5), and allowed their solid-
state structures to be characterised by single-crystal XRD. Com-
plexes 2, 4 and 5 are monomeric, and the geometries around
the central bismuth atom are distorted pentagonal-bipyramidal,
trigonal-pyramidal and distorted octahedral, respectively. In
contrast, complex 1 adopts a trigonal-pyramidal geometry at
the bismuth atom and extends into a dimeric structure through
long intermolecular Bi–N and Bi–S interactions. Dimerisation of
two monomeric units of 3 occurs through a long interaction
between the bismuth centre and π electrons of an adjacent
aromatic ring to give an overall distorted pentagonal-pyramidal
coordination geometry. Both NMR studies and the crystallo-
graphic data reveal the different binding modes of the ligands.
Analysis of the antibacterial activity of the complexes against
M. smegmatis, S. aureus, MRSA, VRE, E. faecalis and E. coli dem-
onstrated that heteroleptic BiIII complexes are almost always
more effective than their homoleptic analogues. From our study
the thiadiazole (5-MMTD) and thiazole ligands (MBT and 4-
BrMTD) appear to be particularly effective, while the tetrazole
ligand 1-MMTZ results in more modest activity. The imidazole
[2-MMI, 2-MMI(H)] and pyrimidine ligands [PYM, PYM(H)] appear
to be the least effective. All the complexes showed poor activity
towards the gram-negative bacterium E. coli but, more promis-
ingly, showed little or no toxicity towards mammalian COS-7
cell lines at 20 μg mL–1.
lected with graphite-monochromated Mo-Kα radiation (λ
=
0.71073 Å) and processed with the CrysAlisPro 1.171.34.3 soft-
ware;[42] Lorentzian, polarisation and absorption corrections (multi-
scan) were applied. All compounds 1–5 were solved and refined
with SHELX-97.[43] All non-hydrogen atoms were refined with aniso-
tropic thermal parameters, unless otherwise indicated. Hydrogen
atoms were placed in calculated positions by using a riding model
with d(C–H) = 0.95–0.98 Å and Uiso(H) = xUiso(C), x = 1.2 or 1.5,
unless otherwise indicated. For compound 5 the data were mod-
elled as a merohedral twin and solved by using the twin data proc-
essing in CrysAlsiPro, and the final structure was refined by using
HKLF 5 format.
CCDC 1436829 (for 1), 1436830 (for 2), 1436831 (for 3), 1436832
(for 4), and 1436833 (for 5) contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre.
Antimicrobial Activity: Mycobacterium smegmatis was cultured in
Middlebrook 7H9/7H10 medium (Difco) for 3 d at 37 °C. Escherichia
coli was cultured overnight at 37 °C in Luria–Bertani medium. Staph-
ylococcus aureus, MRSA, Enterococcus faecalis and VRE were cultured
in brain–heart infusion medium (Oxoid) overnight at 37 °C.
In initial activity screens, filter paper discs were soaked with 2 μL
of 5 mg/mL compound, and then placed on the surface of agar
plates that had been pre-spread with a culture of the bacterial
strain of interest. Following incubation, the plates were examined
for the presence of zones of inhibition of bacterial growth around
the discs. Compounds that were negative in this assay were not
examined further, and MIC assays were performed for all positive
compounds.
This all suggests that future endeavours should focus on de-
veloping and assaying a broader range of heteroleptic thiadi-
azole- and thiazole-based BiIII complexes as the most promising
targets for this class of complex.
To determine MICs, liquid cultures of bacteria were prepared con-
taining serial dilutions of the compounds, from 80 to 0.25 μg/mL.
After incubation, the cultures were examined to determine the min-
imum concentration of compound that caused total inhibition of
growth; this value was recorded as the MIC. The antimycobacterial
drugs isoniazid and ethambutol were included as controls.
Experimental Section
BiPh3 was synthesised by a standard Grignard metathesis reaction
in which BiCl3 was treated with MgPhBr in dried diethyl ether at
0 °C, and subsequently recrystallised from ethanol. PhBiCl2 was syn-
thesised by mixing BiPh3 and BiCl3 in a 2:1 ratio in dried diethyl
Cell Toxicity Assays: Toxicities against eukaryotic cells were as-
sayed by using an MTT-based in vitro toxicology assay kit (Sigma-
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2746
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 10. Complexes 1–5 with labelling corresponding to NMR chemical shift assignments.
Aldrich; cat. no. TOX1-1KT), according to the manufacturer's instruc- (0.043 g, 0.37 mmol) in toluene (12 h reflux) afforded yellow crystals
1
tions. Compounds were added to cultured COS-7 (monkey kidney
derived) cells at 10 and 20 μg/mL for 2 h prior to addition of the
of 1 (0.10 g, 62 %), m.p. 232–234 °C. H NMR (400 MHz, [D6]DMSO):
δ = 13.67. (s, 1 H, NH), 8.90 (d, 2 H, Ha), 8.38 (s, 1 H, He), 7.80 (t, 2
tetrazolium dye MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl- H, Hb), 7.41 (t, 1 H, Hc), 3.43 (s, 3 H, Hg), 2.57 (s, 6 H, Hj) ppm. 13C
tetrazolium bromide]. OD690 measurements were expressed as a
NMR (100 MHz, [D6]DMSO): δ = 164.7 (Ci), 143.0 (Ce), 139.3 (Ca),
percentage the DMSO-only negative controls. All assays were per- 132.8 (Cb), 127.7 (Cc), 31.2 (Cg), 15.4 (Cj) ppm. FTIR: ν = 3096 (b),
˜
formed in triplicate. Isoniazid- and ethambutol-treated COS-7 cells
were included as controls.
1567 (s), 1505 (s), 1473 (s), 1425 (m), 1348 (m), 1197 (s), 1155 (m),
1052 (m), 1031 (s), 766 (s), 722 (m), 691 (s), 679 (m) cm–1. ESI-MS+
(solvent: DMSO/MeOH, 35 eV): m/z = 741 (20) [BiPhL2(L′H)-
(H2O)3(Na)]+, 687 (70) [BiPhL2(L′H)(Na)]+, 571 (7) [BiPhL2(Na)]+. ESI-
MS– (solvent: DMSO/MeOH, 35 eV): m/z (%) = 777 (10) [BiPhL2(L′H)-
(DMSO)(Cl)]–, 681 (45) [BiPhL2(L′H)(OH)]–. (L = 5-MMTD and L′ = 4-
MMT). BiC15H16N7S5 (663.6): calcd. C 27.15, H 2.58, N 15.86; found
C 27.45, H 2.31, N 15.55.
General Procedures (GP)
GP1: Microwave heating: All reactions were conducted with one
equivalent of BiPh3 plus one and/or two equivalents of the chosen
thione(s). Crystalline BiPh3 was ground together with the thione(s)
and placed in an Emery microwave vial followed by the addition of
toluene. The sample was then heated in the microwave reactor. On
completion, all volatile substancess were removed and the solid
product obtained was washed with a small amount of cold toluene,
ethanol, diethyl ether or acetone to remove unconverted BiPh3 and
thiones(s). The solid was then taken up in DMSO or toluene and
allowed to crystallise over 7–14 d. All reactions were carried out
300 W.
Synthesis of [Bi(1-MTTZ)2{(PYM)(PYM(H)2)}] (2)
Microwave-Assisted Synthesis: A mixture of pyrimidine-2-thione
(0.12 g, 1.06 mmol) and triphenylbismuth (0.15 g, 0.33 mmol) in
toluene (4 mL) was stirred constantly at room temperature for
about 5 min according to GP1. The colourless solution obtained
was than mixed with the solution of 1-methyl-1H-tetrazole-5-thione
(0.06 g, 0.53 mmol) in toluene (2 mL) and irradiated at 100 °C for
8 min. The resulting solution was allowed to stand for 10 d at room
temperature. The orange crystals obtained were filtered off, washed
with toluene and acetone and air-dried. These were identified as 2,
yield 0.15 g (65 %).
GP2: Synthesis by Conventional Reflux: Crystalline BiPh3 and the
chosen thione(s) were placed in a small round-bottomed flask with
toluene (25 mL) as solvent and the mixture heated to reflux for 4–
15 h. The toluene was removed in vacuo and the remaining solid
washed with a small amount of toluene, ethanol, diethyl ether or
acetone to remove unconverted BiPh3 and/or thione(s) prior to dry-
ing in air. The solid was then taken up in DMSO or toluene and
allowed to crystallise over 7–14 d.
Conventional Method: The general procedure GP2 with triphenyl-
bismuth (0.15 g, 0.33 mmol), pyrimidine-2-thione (0.12 g,
1.06 mmol), and 1-methyl-1H-tetrazole-5-thione (0.06 g, 0.53 mmol)
in toluene (6 h reflux) afforded orange crystals of 2 (0.12 g, 61 %),
m.p. 201 °C. 1H NMR (300 MHz, [D6]DMSO): δ = 8.56 (d, 3J = 6.20 Hz,
6 H, Ha), 7.27 (t, 3J = 6.20 Hz, 3 H, Hb), 3.80 (s, 6 H, He) ppm. 13C
Heteroleptic Thiolato Bismuth(III) Complexes
Synthesis of [BiPh(5-MMTD)2{4-MMT(H)}] (1)
NMR (100 MHz, [D6]DMSO): δ = 156.5 (Ca), 32.2 (Ce) ppm. FTIR: ν =
˜
Microwave-Assisted Synthesis: A mixture of 5-methyl-1,3,4-thiadi-
azole-2-thione (0.10 g, 0.75 mmol) and triphenylbismuth (0.17 g,
0.37 mmol) in toluene (4 mL) was stirred constantly at room tem-
perature for five minutes according to GP1. The yellow solution
obtained was then mixed with a solution of 4-methyl-4H-1,2,4-tri-
azole-3-thione (0.043 g, 0.37 mmol) in toluene (2 mL) and was irra-
diated at 115 °C for 10 min. The yellow crystals of 1 thus obtained
were washed with acetone and diethyl ether, yield 0.12 g (75 %).
3064 (m), 1597 (m), 1537 (m), 1464 (s), 1417 (s), 1373 (s), 1316 (s),
1267 (s), 1207 (m), 1035 (s), 979 (s), 756 (s), 735 (s), 700 (m), 655
(m) cm–1. ESI-MS+ (solvent: DMSO/MeOH, 35 eV): m/z (%) = 737 (20),
[Bi(L)2(L′3H2)(H2O)3(Na)]+ (L = 1-MTTZ and L′ = PYM). BiC14H14N10S4
(659.6): calcd. C 24.81, H 2.21, N 22.31; found C 24.26, H 2.31, N
22.23.
Synthesis of [Bi(MBT)2(5-MMTD)] (3)
Conventional Method: The general procedure GP2 with triphenyl- Microwave-Assisted Synthesis: A mixture of 4-phenylthiazole-2-
bismuth (0.17 g, 0.37 mmol), 5-methyl-1,3,4-thiadiazole-2-thione thione (0.14 g, 0.72 mmol) and triphenylbismuth (0.15 g, 0.36 mmol)
(0.10 g, 0.75 mmol), and 4-methyl-4H-1,2,4-triazole-3-thione in toluene (4 mL) was stirred constantly at room temperature for
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2747
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
about 5 min. The solution obtained was than mixed with a solution
ole-5-thione (0.15 g, 1.29 mmol) and 1-methyl-1H-imidazole-2-thi-
of 5-methyl-1,3,4-thiadiazole-2-thione (0.05 g, 0.36 mmol) in tolu- one (0.14 g, 1.29 mmol) in toluene (4 h reflux) afforded 5 as orange
ene (2 mL) and irradiated at 112 °C for 10 min. The yellow crystals crystals (0.12 g, 71 %).
thus obtained were washed with toluene and diethyl ether and
identified as 3, yield 0.09 g (72 %).
1
M.p. 125 °C. H NMR (400 MHz, [D6]DMSO): δ = 12.03 (s, 2 H, NH′),
7.04 (d, 3 H, Ha), 6.85 (d, 3 H, Hb), 3.89 (s, 9 H, He), 3.43 (s, 9 H, Hc)
ppm. 13C NMR (100 MHz, [D6]DMSO): δ = 120.3 (Cb), 114.5 (Ca), 33.4
Conventional Method: The general procedure GP2 with triphenyl-
bismuth (0.15 g, 0.36 mmol), 4-phenylthiazole-2-thione (0.14 g,
0.72 mmol) and 5-methyl-1,3,4-thiadiazole-2-thione (0.05 g,
0.36 mmol) in toluene yielded yellow crystals of 3 (0.08 g, 69 %).
(Ce), 33.3 (Cc) ppm. FTIR: ν = 3102 (m), 3058 (m), 1571 (s), 1505 (m),
˜
1437 (s), 1417 (s), 1372 (s), 1306 (s), 1269 (s), 1073 (s), 1032 (m), 973
(s), 776 (s), 737 (s), 703 (m), 690 (m), 670 (m) cm–1. ESI-MS+ (solvent:
DMSO/MeOH, 35 eV): m/z (%) = 934 (50) [Bi(L3H)(L′3H2)(H2O)2(H)]+,
209 (20) [Bi]+ (L = 1-MMTZ and L′ = 2-MMI). BiC18H27N18S6 (896.9):
calcd. C 24.10, H 3.03, N 28.11; found C 24.15, H 3.21, N 28.43.
1
M.p. 170 °C (decomp.). H NMR (400 MHz, [D6]DMSO): δ = 7.72 (d,
4 H, Hb), 7.55 (s, 2 H, Hf), 7.25 (m, 6 H, Hc,d), 2.54 (s, 3 H, Hj) ppm.
13C NMR (100 MHz, [D6]DMSO): δ = 152.2 (Ce), 141.9 (Ch), 133.4 (Ca),
128.6 (Cc), 127.9 (Cd), 126.1 (Cb), 113.6 (Cf), 15.9 (Cj) ppm. FT-IR: ν =
˜
2943 (s), 1632 (s), 1458 (s), 1270 (m), 953 (m), 762 (s) cm–1. ESI-MS+
(solvent: DMSO/MeOH, 35 eV): m/z (%) = 860 (10) [PhBiL2-
(L′)(DMSO)(H2O)(K)]+, 842 (18) [BiL2(L′)(DMSO)(K)]+, 762 (18) [BiL2-
(L′)(H2O)2(H)]+ (L = 5-MMTD and L′ = MBT). ESI-MS– (solvent: DMSO/
MeOH, 35 eV): m/z 995 (10) [BiL2(L′)(DMSO)3(Cl)]–, 892 (20) [BiL2-
(L′)(DMSO)(H2O)3(Cl)]–. BiC21H15N4S6 (724.7): calcd. C 34.80, H 2.09,
N 7.73; found C 34.25, H 2.31, N 7.65.
Acknowledgments
The authors thank Monash University, Clayton, Melbourne and
the Australian Research Council (DP110103812) for financial
support. The authors also wish to thank Associate Professor
Benjamin Howden (Austin Health) for supplying the MRSA and
VRE strains for antimicrobial activity testing and Charles Ma for
culturing the COS-7 cells.
Synthesis of [Bi(4-BrMTD)3{2-MMI(H)}] (4)
Microwave-Assisted Synthesis: A mixture of triphenylbismuth
(0.08 g, 0.17 mmol) and 4-(4-bromophenyl)thiazole-2-thione (0.14 g,
0.51 mmol) in toluene (4 mL) was stirred at room temperature for
5 min according to GP1. The solution obtained was than mixed
with the solution of 1-methyl-1H-imidazole-2-thione (0.02 g,
0.17 mmol) in toluene (2 mL) and irradiated at 115 °C for 10 min.
The orange crystals of 4 that precipitated were collected by filtra-
tion, washed with toluene and diethyl ether and dried in air, yield
0.13 g (65 %).
Keywords: Bismuth · S ligands · Heteroleptic complexes ·
Drug design · Medicinal chemistry · Antibiotics
[1] A. Luqman, V. L. Blair, R. Brammananth, P. K. Crellin, R. L. Coppel, P. C.
Andrews, Chem. Eur. J. 2014, 20, 14362.
[2] A. Luqman, V. L. Blair, R. Brammananth, P. K. Crellin, R. L. Coppel, L. Kedzi-
erski, P. C. Andrews, Eur. J. Inorg. Chem. 2015, 725.
Conventional Method: The general procedure GP2 with triphenyl-
bismuth (0.08 g, 0.17 mmol), 4-(4-bromophenyl)thiazole-2-thione
(0.14 g, 0.51 mmol) and 1-methyl-1H-imidazole-2-thione (0.02 g,
0.17 mmol) in toluene (7 h reflux) afforded 4 as orange crystals
(0.12 g, 60 %).
[3] A. Luqman, V. L. Blair, R. Brammananth, P. K. Crellin, R. L. Coppel, P. C.
Andrews, Eur. J. Inorg. Chem. 2015, 4935.
[4] V. Shivankar, Y. Gaikwad, H. Mulla, R. Patil, L. Gavali, IOSR J. Appl. Chem.
2012, 26.
[5] A. Gupta, R. K. Sharma, R. Bohra, V. K. Jain, J. E. Drake, M. B. Hursthouse,
M. E. Light, J. Organomet. Chem. 2003, 678, 122.
[6] N. P. E. Barry, P. J. Sadler, Chem. Commun. 2013, 49, 5106.
[7] P. Chellan, P. J. Sadler, Philos. Trans. R. Soc. London Ser. A 2015, 373,
20140182.
1
M.p. 145 °C (decomp.). H NMR (400 MHz, [D6]DMSO): δ = 12.03 (s,
1 H, NH), 7.72 (m, 12 H, Hbc), 7.39 (s, 3 H, Hf), 7.04 (d, 1 H, Hj), 6.87
(d, 1 H, Hi), 3.41 (s, 3 H, Hk) ppm. 13C NMR (100 MHz, [D6]DMSO):
δ = 167.2 (Cg), 150.9 (Ce), 131.4 (Cc), 127.4 (Cb), 120.6 (Ci), 114.4 (Cj),
[8] R. Mohan, Nat. Chem. 2010, 2, 336.
[9] R. Ge, Z. Chen, Q. Zhou, Metallomics 2012, 4, 239.
[10] N. Yang, H. Sun, Encyclia Encyclopedia of Environmental Health 2011, 404.
[11] P. Sadler, H. Li, H. Sun, Coord. Chem. Rev. 1999, 185, 689.
[12] R. Ge, H. Sun, Acc. Chem. Res. 2007, 40, 267.
[13] N. Yang, H. Sun, Coord. Chem. Rev. 2007, 251, 2354.
[14] D. M. Keogan, D. M. Griffith, Molecules 2014, 19, 15258.
[15] J. A. R. Salvador, S. A. C. Figueredo, R. M. A. Pinto, S. M. Silvestre, Future
Med. Chem. 2012, 4, 1495.
[16] L. Dawara, R. V. Singh, J. Coord. Chem. 2011, 64, 931.
[17] J. A. Lessa, D. C. Reis, J. G. Da Silva, L. T. Paradizzi, N. F. da Silva, M. de F. A.
Carvalho, S. A. Siqueira, H. Beraldo, Chem. Biodiversity 2012, 9, 1955.
[18] C.-L. Wu, P. Domenico, D. J. Hassett, T. J. Beveridge, A. R. Hauser, J. A.
Kazzaz, Am. J. Respir. Cell Mol. Biol. 2002, 26, 731.
[19] J. P. Folsom, B. Baker, P. S. Stewart, J. Appl. Microbiol. 2011, 111, 989.
[20] Halwani, S. Hebert, Z. E. Suntres, R. M. Lafreni, A. O. Azghani, A. Omri,
Int. J. Pharm. 2009, 373, 141.
113.4 (Cf) ppm. FTIR: ν = 3059 (m), 1651 (s), 1595 (s), 1473 (m), 1441
˜
(m), 1272 (s), 1180, 1072 (s), 1051 (s), 774 (m), 718 (s), 686 (m), 667
(m) cm–1. ESI-MS+ (solvent: DMSO/MeOH, 35 eV): m/z (%) = 1156
(27) [Bi(L)3(L′H)(H2O)(H)]+(79Br isotope), 1158 (25) [Bi(L)3(L′H)-
(H2O)(H)]+(81Br isotope). ESI-MS– (solvent: DMSO/MeOH, 35 eV): m/z
1172 (50) [Bi(L)3(L′H)(H2O)(OH)] – (79Br isotope), (L = 4-BrMBT and
L′ = 2-MMI). BiC31H21Br3N5S7 (1136.6): calcd. C 32.76, H 1.86, N 6.16;
found C 32.51, H 1.89, N 5.93.
Synthesis of [Bi(1-MMTZ)2{1-MMTZ(H)}(2-MMI){2-MMI(H)2}] 5
Microwave-Assisted Synthesis: A mixture of tris-(2-methoxy-
phenyl)bismuth (0.22 g, 0.43 mmol) and 1-methyl-1H-I-tetrazole-5-
thione (0.15 g, 1.29 mmol) in toluene (4 mL) was stirred constantly
at room temperature for about 5 min. The colourless solution ob-
tained was than mixed with a solution of 1-methyl-1H-imidazole-2-
thione (0.14 g, 1.29 mmol) in toluene (2 mL) and irradiated at 100 °C
for 6 min. The resulting solution was allowed to stand for two weeks
at room temperature. The orange crystals of 5 which formed were
collected by filtration, washed with toluene and dried in air, yield
0.13 g (74 %).
[21] V. Stavila, K. H. Whitmire, I. Rusakova, Chem. Mater. 2009, 21, 5456.
[22] L. J. Farrugia, F. J. Lawlor, N. C. Norman, J. Chem. Soc., Dalton Trans. 1995,
1163.
[23] H. P. S. Chauhan, N. M. Shaikh, U. P. Sing, Appl. Organomet. Chem. 2006,
20, 142.
[24] A. Luqman, V. L. Blair, A. M. Bond, P. C. Andrews, Angew. Chem. 2013,
125, 7388; Angew. Chem. Int. Ed. 2013, 52, 7247.
Conventional Method: The general procedure GP2 with tris(2-
methoxyphenyl)bismuth (0.22 g, 0.43 mmol), 1-methyl-1H-I-tetraz-
[25] D. R. Perez, S. H. Tarulli, O. V. Quinzani, J. Dristas, R. Faccio, L. Suescun, A.
Mombru, Z. Anorg. Allg. Chem. 2007, 633, 1066.
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2748
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[26]
[27]
[28]
[29]
[30]
[36] G. Tegos, S. R. Stermitz, O. Lomovskaya, K. Lewis, Antimicrob. Agents Che-
mother. 2002, 46, 3133.
[37] T. Kotani, D. Nagai, K. Asahi, H. Suzuki, F. Yamao, N. Kataoka, T. Yagura,
Antimicrob. Agents Chemother. 2005, 2729.
[38] D. M. Keogan, D. M. Griffith, Molecules 2014, 19, 15258–15297.
[39] G. K. Batsala, V. Dokorou, N. Kourkoumelis, M. J. Manos, A. J. Tasiopoulos,
T. Mavromoustakos, M. Simčič, S. Golič-Grdadolnik, S. K. Hadjikakou, In-
org. Chim. Acta 2012, 382, 146.
[40] 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, J. Syn-
chrotron Radiat. 2002, 9, 401.
E. Block, G. Ofori-Okai, H. Kang, J. W. Zubieta, Inorg. Chem. 1991, 30,
4784.
Asato, R. K. Kamamuta, Y. Akamine, T. Fukami, R. Nukada, M. Mikuriya, S.
Deguchi, Y. Yokota, Bull. Chem. Soc. Jpn. 1997, 70, 639.
P. C. Andrews, G. B. Deacon, W. R. Jackson, M. Maguire, N. Scott, B. W.
Skelton, A. H. White, J. Chem. Soc., Dalton Trans. 2002, 4634.
H. Suzuki, N. Komatsu, T. Ogawa, T. Murafuji, T. Ikegami, Y. Matano, Or-
ganobismuth Chemistry, Elsevier, 2001.
H. Suzuki, Y. Matano, in: Organobismuth Chemistry, Elsevier, Amsterdam,
2001, p. 133.
[31] T. M. File Jr., Infect. Dis. Clin. Prac. 2011, 19, 207.
[41] W. Kabsch, J. Appl. Crystallogr. 1993, 26, 795.
[32] S. J. Hal, D. L. Paterson, Antimicrob. Agents Chemother. 2011, 55, 405.
[33] E. Y. Choi, J. W. Huh, C.-M. Lim, Y. Koh, S.-H. Kim, S.-H. Choi, Y. S. Kim, M.-
N. Kim, S.-B. Hong, Intensive Care Med. 2011, 37, 639.
[34] S. Satola, F. Lessa, S. Ray, S. Bulens, R. Lynfield, W. Schaffner, G. Dumyati,
J. Nadle, J. Patel, J. Clin. Microbiol. 2011, 49, 1583.
[42] CrysAlisPro, v. 1.171.34.36, Oxford Diffraction Ltd., Agilent Technologies,
Oxfordshire, UK, 2010.
[43] SHELX-97, G. Sheldrick, Institute for Inorganic Chemistry, University of
Göttingen, Germany, 1996, 65.
[35] P. Domenico, L. Baldassarri, P. E. Schoch, K. Kaehler, M. Sasatsu, B. A.
Cunha, Antimicrob. Agents Chemother. 2001, 45, 1417.
Received: January 27, 2016
Published Online: April 24, 2016
Eur. J. Inorg. Chem. 2016, 2738–2749
www.eurjic.org
2749
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim