The thiol-based reduction of Bi(V) and Sb(V) anti-leishmanial complexes

Article abstract of DOI:10.1016/j.jinorgbio.2021.111470

Low molecular weight thiols including trypanothione and glutathione play an important function in the cellular growth, maintenance and reduction of oxidative stress in Leishmania species. In particular, parasite specific trypanothione has been established as a prime target for new anti-leishmania drugs. Previous studies into the interaction of the front-line Sb(V) based anti-leishmanial drug meglumine antimoniate with glutathione, have demonstrated that a reduction pathway may be responsible for its effective and selective nature. The new suite of organometallic complexes, of general formula [MAr₃(O₂CR)₂] (M = Sb or Bi) have been shown to have potential as new selective drug candidates. However, their behaviour towards the critical thiols glutathione and trypanothione is still largely unknown. Using NMR spectroscopy and mass spectrometry we have examined the interaction of the analogous Sb(V) and Bi(V) organometallic complexes, [SbPh₃(O₂CCH₂(C₆H₄CH₃))₂] S1 and [BiPh₃(O₂CCH₂(C₆H₄CH₃))₂] B1, with the trifluoroacetate (TFA) salt of trypanothione and L-glutathione. In the presence of trypanothione or glutathione at the clinically relevant pH of 4–5 for Leishmania amastigotes, both complexes undergo facile and rapid reduction, with no discernible difference. However, at a higher pH (6–7), the complexes behave quite differently towards glutathione. The Bi(V) complex is again reduced rapidly but the Sb(V) complex undergoes slow reduction over 8 h (t_1/2 = 54 min.) These results give the first insights into why the highly oxidising Bi(V) complexes display low selectivity in their cytotoxicity towards leishmanial and mammalian cells, while the Sb(V) complexes show good selectivity.

Full text of DOI:10.1016/j.jinorgbio.2021.111470

Journal of Inorganic Biochemistry 221 (2021) 111470
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
The thiol-based reduction of Bi(V) and Sb(V) anti-leishmanial complexes
,
,
Rebekah N. Duffina 1, Liam J. Stephensb 1, Victoria L. Blairb, Lukasz Kedzierskic, Philip
,
*
C. Andrewsb
a Faculty of Pharmacy and Pharmaceutical Sciences, Drug delivery, disposition and dynamics, Monash University, Parkville, Melbourne, VIC 3053, Australia
b School of Chemistry, Monash University, Clayton, Melbourne, VIC 3800, Australia
c Faculty of Veterinary and Agricultural Sciences, The Peter Doherty Institute for Infection and Immunity, 792 Elizabeth Street, Melbourne, VIC 3000, Australia
A R T I C L E I N F O
A B S T R A C T
Keywords:
Low molecular weight thiols including trypanothione and glutathione play an important function in the cellular
growth, maintenance and reduction of oxidative stress in Leishmania species. In particular, parasite specific
trypanothione has been established as a prime target for new anti-leishmania drugs. Previous studies into the
interaction of the front-line Sb(V) based anti-leishmanial drug meglumine antimoniate with glutathione, have
demonstrated that a reduction pathway may be responsible for its effective and selective nature. The new suite of
organometallic complexes, of general formula [MAr3(O2CR)2] (M = Sb or Bi) have been shown to have potential
as new selective drug candidates. However, their behaviour towards the critical thiols glutathione and trypa-
nothione is still largely unknown. Using NMR spectroscopy and mass spectrometry we have examined the
interaction of the analogous Sb(V) and Bi(V) organometallic complexes, [SbPh3(O2CCH2(C6H4CH3))2] S1 and
[BiPh3(O2CCH2(C6H4CH3))2] B1, with the trifluoroacetate (TFA) salt of trypanothione and L-glutathione. In the
presence of trypanothione or glutathione at the clinically relevant pH of 4–5 for Leishmania amastigotes, both
complexes undergo facile and rapid reduction, with no discernible difference. However, at a higher pH (6–7), the
complexes behave quite differently towards glutathione. The Bi(V) complex is again reduced rapidly but the Sb
(V) complex undergoes slow reduction over 8 h (t1/2 = 54 min.) These results give the first insights into why the
highly oxidising Bi(V) complexes display low selectivity in their cytotoxicity towards leishmanial and
mammalian cells, while the Sb(V) complexes show good selectivity.
Trypanothione
Glutathione
Antimony
Bismuth
Bioinorganic
Leishmaniasis
1. Introduction
Pentostam™ (sodium stibogluconate) has a parasiticidal effect on axenic
amastigotes [10].
Leishmaniasis is a devastating tropical disease caused by parasites of
the Leishmania species [1]. Millions of people are currently at risk of
developing the disease, with upwards of 40,000 fatalities due to a lack of
adequate treatment options [2–4]. Front-line treatments have relied
almost exclusively on the use of the pentavalent antimonials sodium
stibogluconate and meglumine antimoniate since the beginning of the
second World War [5]. Despite this, little is known and understood
about their mechanism of action and selective toxicity. Our current
understanding is focused on a reductive pathway as the primary
mechanism of action, with Sb(V) acting as a pro-drug where it is bio-
chemically reduced in vivo to the more bioactive Sb(III) form [6]. This
hypothesis is further supported by the apparent lack of toxicity of Sb(V),
and high toxicity of Sb(III) complexes to both stages of the parasite of
different Leishmania species [7–9]. Conversely, it has been observed that
The redox chemistry associated with current Sb(V) treatments occurs
preferentially in the low pH environment of the amastigote (pH 4–5),
and is believed to be facilitated by low molecular weight thiols such as
glutathione and trypanothione, the latter being parasite specific (Fig. 1)
[11,12]. Unlike most organisms, Leishmania species can conjugate the
tripeptide glutathione, and the polyamine spermidine to form trypano-
thione. This compound, which is the most abundant low molecular
weight thiol inside the Leishmania species, is essential for the growth and
survival of the parasite, providing a unique intracellular reducing
environment. As with glutathione in mammalian cells, trypanothione
can act as a redox scavenger and anti-oxidant, and hence is a crucial part
of the defensive mechanisms employed by Leishmania parasites in pre-
venting oxidative stress. The fact that trypanothione, and its related
enzymes, are not present in mammalian cells makes them attractive
* Corresponding author.
E-mail address: phil.andrews@monash.edu (P.C. Andrews).
1
Equal contribution from authors.
https://doi.org/10.1016/j.jinorgbio.2021.111470
Received 14 February 2021; Received in revised form 7 April 2021; Accepted 24 April 2021
Available online 29 April 2021
0162-0134/Crown Copyright © 2021 Published by Elsevier Inc. All rights reserved.

R.N. Duffin et al.
Journal of Inorganic Biochemistry 221 (2021) 111470
targets for current and future anti-leishmanial drugs. Indeed, Sun and
co-workers have demonstrated the rapid reduction of Sb(V) (as sodium
stibogluconate) to Sb(III) under both mildly acidic and neutral condi-
tions by trypanothione [12,13]. In contrast, the reduction of this com-
pound is negligible in the presence of the mammalian analogue
glutathione, under the same conditions. In a separate study, the same
group has shown that Sb(III) (as antimony tartrate) can form a novel
complex with trypanothione, through binding of the two sulfur atoms on
the cysteine residues, and the oxygen of a water molecule to the metal
centre [14]. In addition to their high affinity for trypanothione, the
enzyme responsible for keeping it in its reduced form, trypanothione
reductase, is also a major target site for current and future anti-
leishmanial drugs, including Sb(III) complexes. [15] [16,17].The solid
state crystal structure of Sb(III) with reduced trypanothione reductase
has been obtained, revealing the ability of the metal to directly coor-
dinate to the catalytic centre of the enzyme, and in particular with the
Cys52, Cys57, Thr335 and His461’ residues [17].
investigations into the chemical behaviour of these organometallic
complexes in the presence of either glutathione or trypanothione.
Herein, we present a NMR spectroscopic study on the interaction of two
previously synthesised tris-phenyl M(V) acetato complexes (M = Sb or
Bi); [SbPh3(O2CCH2R)2] S1 and [BiPh3(O2CCH2R)2] B1 (R = p-tolyl),
with trypanothione and L-glutathione, in an attempt to shed light on
their stability, redox activity, and reactivity [24].
2. Experimental
2.1. General
L-Glutathione was purchased from Oakwood chemicals without the
need for further purification. p-Tolyl acetic acid, 70% tert-butyl hydro-
peroxide (Luperox ™) and triphenyl antimony were purchased from
Sigma Aldrich. All remaining solvents and reagents were purchased
from either Sigma Aldrich or Merck. 1H NMR spectra were recorded on a
Bruker Avance DRX600 spectrophotometer (600 MHz), chemical shifts
were references to the appropriate deuterated solvent, D2O or d6-DMSO.
Melting point analysis was conducted in open end capillary tubes, on a
digital Stuart Scientific melting point apparatus SMP10. High-resolution
ESI mass spectroscopy (high-res ESI MS) was run on a Micromass Plat-
form QMS spectrometer, with an electrospray source and a cone voltage
of 35 eV.
Various mechanistic studies for pentavalent antimonials have also
been conducted using glutathione. Frezard et al. studied the interaction
of Glucantime™ (meglumine antimoniate) with glutathione at different
temperatures and pH values, mimicking both mammalian physiological
conditions and that of the parasite [11], concluding that GSH promotes
the reduction of Sb(V)/Sb(III) at an acidic pH, with elevated tempera-
tures increasing the rate [11]. This study also concluded that the
reduction should not occur in the neutral pH of the host cell cytosol
despite a high concentration of GSH, but could be facilitated in more
acidic macrophage organelles [11,18]. Sun et al. determined Sb(III) to
have a high binding affinity to glutathione, indicating that thiols were
the preferential biological target for Sb(III) [19].
2.2. Synthesis of trifluoroacetate (TFA) salt of trypanothione
The synthesis of the TFA-salt of trypanothione (TSH) was achieved
using methods as first detailed by Antoniou et al. [31]. However, in
contrast to this publication, that used the highly toxic diazomethane as a
precursor, the key diester intermediate 1 was obtained in three-steps
using much more benign reagents (Fig. 2). The full characterization of
the intermediates obtained in the synthesis of diester 1 are available in
the supporting information (SI Figs. S1 –S20). Briefly, to the commer-
cially available L-glutathione (2.0 g, 6.5 mmol) in DCM (15 mL) was
added trityl chloride (2.0 g, 7.0 mmol) portion-wise. The resulting so-
lution was stirred overnight, before being concentrated under vacuo to
yield a white solid that was taken forward to the next step of the syn-
thesis without the need for further purification. Methanol (60 mL) was
then used to dissolve this compound (S-trityl protected glutathione 2
(4.0 g, 7.3 mmol)), before a catalytic amount of p-TSA (277 mg, 1.5
mmol) was added. The reaction was stirred at reflux overnight before
the resulting solution was concentrated under vacuo. The crude white
solid that ensued was taken forward to the next step of the synthesis
without the need for further purification. In the final step to obtain
diester intermediate 1, S-trityl protected glutathione ester 3 (1.6 g, 2.8
mmol) was first dissolved in DCM (40 mL), before being cooled to 0 ◦C.
Once cooled, a solution of trityl chloride (850 mg, 3.0 mmol) and trie-
thylamine (0.77 mL, 5.5 mmol) in DCM was added dropwise. A clear
Further studies by Frezard et al. focused on other biologically rele-
vant thiols such as cysteinyl-glycine, cysteine, as well as trypanothione.
Reduction was monitored using both complexed Sb(V), as meglumine
antimoniate, and ‘free’ antimoniate as K[Sb(OH)6]) [20]. Similar to
their previous study, they noted that the rate of Sb(V/III) reduction
increased at lower pH and higher temperatures, and concluded that
complexed vs un-complexed antimoniate underwent reduction at
different rates [20]. However, studies by Yan et al. determined that the
reduction of meglumine antimoniate by glutathione and other
mammalian moieties were too slow to be biologically significant and
that the parasite specific trypanothione would constitute the greatest
reductive activity [12].
Recent work into the design of new antimony and bismuth based
anti-leishmanial drugs has focused on organometallic carboxylates; as a
way of decreasing toxicity while increasing compound lipophilicity to
allow for possible oral delivery and uptake through the stomach,
something not possible with the current hydrophilic Sb(V) drugs
[21–30]. Despite promising in vitro efficacy and cell selectivity, the
stability of these [MAr3(O2CR)2] (M = Sb or Bi) complexes has so far
only been studied in NMR solvents and in cell media. There have been no
Fig. 1. Chemical structures of glutathione and trypanothione.
2

R.N. Duffin et al.
Journal of Inorganic Biochemistry 221 (2021) 111470
Fig. 2. Alternative synthesis to diester 1; a key intermediate required in the synthesis of trypanothione.
glassy solid was obtained following the reaction time, and work-up
procedure as detailed by Antoniou et al. [31].The synthesis of N-boc
protected spermidine [32] and its subsequent involvement in the syn-
thesis of trypanothione followed methods previously detailed in the
literature. All structural characterization of the intermediates isolated
during this synthesis matched those previously reported in the literature
[31,32].
SbPh3. The decomposition graph was obtained using Microsoft excel and
the corresponding half-life derived from the logarithmic equation
produced.
2.5. Mass spectrometry
For the studies conducted at low pH, aliquots were taken directly
from the NMR tubes, diluted and then analysed by high-res ESI mass
spectroscopy. For the glutathione study conducted at a higher pH (6–7),
both complexes were reacted with an excess of glutathione in a 1:1 mix
of degassed H2O:EtOH. The mixture was heated overnight before
filtration. Both the solid and filtrate were analysed in degassed H2O on a
Micromass Platform QMS spectrometer.
The formation of TFA-trypanothione was confirmed by 1H, 13C and
19F NMR spectroscopy, all of which matched the spectra of those pre-
viously reported in the literature (see SI, Figs. S1–S4 for details). The
formation of TFA-trypanothione was also confirmed by high-res ESI MS,
whilst the purity of the compound, was analysed by RP-HPLC on a C-8
column using a linear gradient from 5% to 40% B over 20 min, where A
= 0.1% TFA in water and B = 0.1% TFA in MeCN, at a flow rate 0.9 mL/
min and monitored at 214 nm: tR = 8.31 min.
3. Results and discussion
2.3. Synthesis of [SbPh3(O2CCH2(C6H4CH3))2], S1 and
[BiPh3(O2CCH2(C6H4CH3))2], B1
Complexes S1 and B1, [MPh3(O2CCH2(C6H4CH3))2] (S1, M = Sb; B1,
M = Bi) (Fig. 3), were synthesised and purified as previously described
Both complexes were prepared as previously described [24]. Briefly:
0.5 mmol of BiPh3 or SbPh3 was dissolved in toluene, and to this solution
two equivalents of 70% tBuOOH was added. 1.0 mmol of p-tolylacetic
acid was then added and the reaction stirred for approximately an hour.
The solution was filtered and left to stand yielding both complexes as
crystalline solids. The complexes were identified by 1H NMR and
melting point analysis [24].
2.4. 1H NMR spectroscopic study
Both complexes were made into 10 mM stocks in 1 mL of d6-DMSO.
100 μL of this solution was placed into an NMR tube and the contents
frozen in liquid nitrogen at ꢀ 80 ◦C. To this frozen tube, 3.5 equivalents
of trypanothione, or 7 equivalents of glutathione in 400 μL of 1:1 D2O/
DMSO mix was added and the contents frozen once again. For the low
pH studies, spectra were recorded one after another for upwards of 1 h,
and every 2 h thereafter. For the higher pH glutathione study, spectra
were recorded every half hour for Sb and one after the other for upwards
of an hour for Bi. The half-life of the Sb complex was then calculated by
proton integration ratio of the original complex to the formation of
Fig. 3. Chemical structures of complexes [SbPh3(O2CCH2(C6H4CH3))2], S1 and
[BiPh3(O2CCH2(C6H4CH3))2] B1 analysed in this study.
3

R.N. Duffin et al.
Journal of Inorganic Biochemistry 221 (2021) 111470
[23]. The composition of the crystalline products was confirmed using
1H NMR spectroscopy and melting point analysis, showing them to be
the desired complexes.
The 1H NMR spectrum acquired immediately after mixing revealed
that both S1 and B1 are rapidly reduced by trypanothione at this
particular pH. Both spectra demonstrate that the Sb(V) and Bi(V) com-
plexes are immediately reduced, with the presence of SbPh3 (7.29–7.34
ppm) and BiPh3 (7.66, 7.35 and 7.27 ppm), and the parent p-tolylacetic
acid (7.07, 3.47 and 2.20 ppm) clearly visible in their respective 1H NMR
spectra (Figs. 4 and 5). In addition, the β-protons of the cysteine residue
of trypanothione at 2.70 and 2.77 ppm (see SI Figs. S1–S4 for NMR
spectrum of trypanothione) are reduced in intensity, with a new reso-
nance detected at approximately 3.10 ppm. The resonance relating to
3.1. Reduction of complexes B1 and S1 with TFA-trypanothione (pH
4–5)
Our original studies on these two complexes showed B1 to have little
selectivity for the Leishmania parasite cells over mammalian cells, and
that S1 was much more selective [24]. As such, we anticipated that the
Bi complex, B1, would display higher reactivity towards both trypano-
thione and glutathione, originating from the significant difference in
reduction potential between Bi(V/III) and Sb(V/III) of 2.00 and 0.605 V
respectively [33].
the α-proton of the cysteine residue is also clearly reduced in intensity,
with a new resonance seen at approximately 4.30 ppm. Both these
phenomena are indicative of the formation of the oxidised form of try-
panothione, as well as the possible formation of a metal (Sb(III) or Bi
(III)) trypanothione species, as was previously demonstrated by Sun
et al. [12–14]
Though the promastigote form of the parasite is the initial source of
infection, this form is unable to proliferate in mammalian hosts and only
spends a small amount of time in the bloodstream before being taken up
by macrophages pending an immune response. The clinically relevant
form of the parasite is the amastigote, and it is this which is the target of
interest for novel anti-leishmania drugs, as it hides and proliferates
within the immune cells of the mammalian host [17,33]. In vivo, the
amastigote residues in an acidic environment of ca. pH 4–5.5 [18,34]. To
mimic this in our study, we prepared the TFA salt of trypanothione,
which gives a final pH of between 4 and 5, and expected this compound
to have the same chemical effect on the organometallic complexes as
trypanothione itself.
In order to confirm the presence of MPh3, p-tolylacetic acid, oxidised
trypanothione, and any related metal trypanothione complex, aliquots
were taken from respective NMR tubes and analysed by high-res ESI-MS.
In both cases, the presence of oxidised trypanothione was clearly iden-
tified (722.30619 – [M + H]+) (see SI Fig. S17), however no other
1
species that were present in the H NMR spectra, were detected using
this method.
3.2. Reduction of complexes B1 and S1 with glutathione (pH 4–5)
In our investigation, both complexes were first dissolved in d6-DMSO
at a stock solution concentration of 10 mM. Sonication was used to assist
the dissolution of the bismuth complex [BiPh3(O2CCH2(C6H4CH3))2],
B1. A known concentration of B1 and S1 were then placed into an NMR
tube and the contents frozen. To this, a solution of the TFA-
trypanothione salt (3.5 equivalents), at pH 4–5 in D2O/d6-DMSO
(1:1), was added to the frozen tube and the contents immediately frozen
once again to inhibit any reaction from occurring. On melting and
mixing of the components, the kinetics of Sb(V) and Bi(V) reduction by
the TFA-trypanothione were studied by 1H NMR spectroscopy.
The determination that S1 and B1 are rapidly reduced (time ‘0’) in
the presence of trypanothione at pH 4–5, provides support for the
mechanism by which both Sb(V) and Bi(V) exert a toxic effect on the
amastigotes. The remaining question is that of selectivity (selectivity
index for S1 is >9, no toxicity towards mammalian cells was observed at
the highest concentration of 100 μM, and for B1 is 6.3, high degree of
both parasite and mammalian toxicity) and toxicity towards the
mammalian host cells. To answer this, we turned to study the impact of
glutathione as the dominant reducing thiol peptide in mammalian cells,
but of lower concentration in leishmanial cells. As was the case with
Fig. 4. 1H NMR spectra of complex [SbPh3(O2CCH2(C6H4CH3))2] S1 with an excess of TFA-trypanothione in d6-DMSO/D2O 1:1 mix at ‘0 min’. The formation of
SbPh3 and p-tolylacetic acid have been highlighted.
4

R.N. Duffin et al.
Journal of Inorganic Biochemistry 221 (2021) 111470
Fig. 5. 1H NMR spectra of complex [BiPh3(O2CCH2(C6H4CH3))2] B1 with an excess of TFA-trypanothione in d6-DMSO/D2O 1:1 mix at ‘0 min’. The formation of
BiPh3 and p-tolylacetic acid have been highlighted.
trypanothione, under the same conditions the 1H NMR spectra both
indicate that S1 and B1 are again rapidly reduced, with MPh3 and p-
tolylacetic acid observed in the respective NMR spectra at time point ‘0’
(see SI Figs. S7–S10).
Fig. 6. Overlay of 1H NMR spectra of [BiPh3(O2CCH2(C6H4CH3))2] B1, BiPh3 and B1 + GSH (t = “0”), highlighting the presence of both B1 and BiPh3 in the reaction
mixture at time point ‘0’. At t = “0”, peaks corresponding to BiPh3 are donated by blue squares whilst those corresponding to B1 are labelled with yellow stars. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
5

R.N. Duffin et al.
Journal of Inorganic Biochemistry 221 (2021) 111470
3.3. Stability of complexes B1 and S1 in acidic pH
react with DMSO to form GSSG over time [30] making it difficult to
accurately determine the ratio produced in the NMR. Previous studies by
Frezard et al. deduced that the Sb(V) meglumine antimonate reacted
with five equivalents of glutathione to give the Sb(GS)3 complex and the
disulfide GSSG. However, formation of Sb(GS)3 was not proven.
To determine whether Sb(GS)3 is formed on reduction of the
organometallic complex S1, a 1:3 mixture of S1 and glutathione was
reacted in degassed 1:1 EtOH/H2O. This suspension was heated to 60 ◦C
and allowed to stir for 24 h. On filtration, the captured solid and the
solid material obtained from the filtrate on removal of the ethanol and
water were analysed following dissolution in degassed D2O (see SI
Fig. S6). Comparison of the NMR spectra with that obtained for Sb(GS)3
by Sun et al. showed no formation of the Sb(GS)3 in the reduction of S1
[19]. Mass spectrometry on this sample and others prepared using
differing stoichiometries also showed no evidence of the formation of Sb
(GS)3. For compound S1, ESI-MS was however able to identify the free p-
tolylacetic acid, and residual complex S1 (see SI Fig. S16). In contrast,
the reaction of the bismuth complex B1 with glutathione did result in the
formation of Bi(GS)3 with the Bi(GS)+2 fragment observed at m/z
[821.1]+ (Fig. 9). This may indicate that there are significant differences
in their mammalian biochemistry [23,24].
To determine the impact solely of solution pH on the stability and
reaction kinetics, a D2O/DMSO solution of TFA at pH 4–5 was used to
dissolve complexes S1 and B1, and 1H NMR spectra recorded at set in-
tervals (0, 0.5, 1, 2, 4, 8, 16 and 24 h) for up to 24 h. The spectra showed
no changes in splitting or chemical shifts over the 24 h period, indicating
both complexes to be stable in the acidic solution (see SI Figs. S11-S14).
3.4. Reduction of complexes B1 and S1 with glutathione (pH 6–7)
Studies have shown that the reduction of Sb(V) complexes is highly
pH dependent. The pH of normal mammalian cells is close to 7.0 with
glutathione present in concentrations of 5–10 mM and predominantly in
the reduced (GSH) form. Having demonstrated that glutathione at a pH
of 4–5 can rapidly reduce both S1 and B1, the next part of the study was
to establish whether the same was true at neutral pH consistent with
healthy mammalian cells.
After following the same reaction procedure, the NMR spectra ob-
tained immediately after the addition of glutathione to the complexes
(time ‘0’), indicated that the bismuth complex B1 reacts at a much faster
rate than the antimony complex S1. Complex B1 was completely
reduced in less than 5 min, with the reaction occurring immediately on
mixing (Figs. 6 and S15).
The glutathione redox cycle is an efficient system for the detoxifi-
cation of oxidising species within mammalian cells [35,36]. GSH is
oxidised to GSSG by xenobiotic oxidising species, rendering them
harmless. GSSG is then reduced back to its anti-oxidant GSH form by
glutathione reductases, allowing GSH to further interact with potentially
damaging species [37,38]. As B1 was found to form the Bi(GS)3 com-
plex, it is possible this may contribute to the increased cytotoxicity that
was previously observed [23,24]. Bioavailable GSH would bind strongly
to the bismuth centre, disallowing interactions with oxidising species
and halting the redox cycle [32]. In contrast to this, previous studies
have indicated that the organometallic Sb(V) analogues exhibit little to
no activity towards the mammalian controls [23,24]. Though S1 induces
oxidation of GSH to GSSG (subsequently reducing to SbPh3 and the free
p-tolylacetic acid), there is no evidence of binding to GSH upon its
reduction. Therefore, the GSSG produced could undergo enzymatic
reduction back to GSH without inhibition of its biological processes.
This decomposition differs from the predicted interaction of
For S1 the reduction reaction was significantly slower. The decom-
position rate was calculated by measuring the changing ratio of a signal
belonging to S1, at 7.92 ppm, against an increasing signal from newly
appearing SbPh3, at 7.33 ppm. From this a half-life of 54 min. was
determined for the reduction of S1 by glutathione at pH 6–7 (Fig. 7). Of
interest though is the fact that the reduction of S1 is faster than that
observed by Frezard et al. for meglumine antimoniate [11].
As the concentration of S1 decreases, signals for SbPh3 and p-tolyl-
acetic acid appear (Fig. 8). At the conclusion of the reaction a new set of
signals are also present that correspond to the formation of oxidised
glutathione (GSSG), which was also observed in the spectrum for the
reduction of B1 (see SI Fig. S5).
As an excess of GSH (six equivalents) was used in each reaction the
determination of stoichiometry is difficult. GSH has been observed to
Fig. 7. 1H NMR decomposition graph of [SbPh3(O2CCH2(C6H4CH3))2] S1, with an excess of GSH in d6-DMSO:D2O. Readings were taken hourly on a 600 MHz Bruker
NMR spectrometer at 25 ◦C. A decreasing signal of the original complex was compared to an increasing signal of SbPh3 and a proton ratio obtained.
6

R.N. Duffin et al.
Journal of Inorganic Biochemistry 221 (2021) 111470
Fig. 8. 1H NMR spectra of complex [SbPh3(O2CCH2(C6H4CH3))2] S1 with an excess GSH in d6-DMSO/D2O 1:1 after 24 h. The formation of SbPh3, p-tolylacetic acid
and GSSG have been highlighted.
Fig. 9. Positive mode ESI-MS of the reaction of B1 with glutathione, highlighting the formation of Bi(Gs)3 through the identification of the fragment [Bi(Gs)2]+.
complexed Sb(V) by Frezard et al. Studies into the interaction of
meglumine antimoniate and un-complexed antimonate concluded that
the un-complexed antimoniate reduced at a rate that was 8-fold quicker
than the complexed meglumine antimoniate [20]. Frezard et al. theor-
ised that due to this, complexed antimony slows down the rate of
reduction by blocking interaction with thiol containing compounds.
Therefore, this slower reaction rate was attributed to the dissociation of
the antimony being the rate limiting step [20].
water soluble. These chemical differences may contribute to the differ-
ences in reduction. The absolute composition of meglumine antimoniate
may also contribute. Studies into structural characterization have
focused on a 1:2 Sb(V) to NMG compositional ratio as the predominant
form. However, studies by Frezard et al. have contested this composi-
tion, suggesting the drug is a mix of homoleptic complexes of varying
metal to ligand ratios; 1:1, 1:2, 2:2, and 2:3 [40]. The rate at which each
of these complexes undergoes reduction may differ as a consequence of
their differing composition and structural chemistry, and so the reduc-
tion rate for ‘meglumine antimoniate’ is an average. In contrast, S1 is a
pure heteroleptic complex with an invariant composition of a 1: 3: 2
ratio of Sb(V), phenyl , and carboxylate.
Nevertheless, the distinctly different chemical and physical proper-
ties of meglumine antimoniate as compared to complex S1 may
contribute to the large differences in reaction rate. Complex S1 is an
organometallic compound with three aryl rings and two carboxylate
ligands, the solubility of which in aqueous media is limited with a
calculated cLogP of 8.14 [39]. Conversely, being highly hydrophilic
meglumine antimoniate has a cLogP value of ꢀ 2.40 [39], and is highly
Regardless, the rate of reduction is substantially slower than for the
bismuth analogue B1 which provides an insight into why bismuth
complexes belonging to this family tend to display non-selective toxicity
7

R.N. Duffin et al.
Journal of Inorganic Biochemistry 221 (2021) 111470
towards the leishmanial and mammalian cells [23,24]. Studies into the
interaction of antimoniates with trypanothione have concluded that it
reduces Sb(V) at a much higher rate than GSH [15], and therefore it can
be deduced that the reduction of the Bi(V) analogue would be even more
facile. This reduction also occurs rapidly in the acidic conditions of the
amastigote and macrophage [41]. This substantial difference between
the reduction rate of Bi(V) versus Sb(V) organometallics by thiol rich
peptides may help to us to understand the significant differences in their
biological activity and selectivity in the treatment of Leishmaniasis.
org/10.1016/j.jinorgbio.2021.111470.
References
[1] J.A.R. Postigo, Int. J. Antimicrob. Agents 36 (2010) S62–S65.
´
[2] J. Alvar, I.D. Velez, C. Bern, M. Herrero, P. Desjeux, J. Cano, J. Jannin, M. den
Boer, W.L.C. Team, PLoS One 7 (2012), e35671.
[3] H.A. Fletcher, M. Chatterjee, A. Cooper, T. Hussell, P.M. Kaye, J. Prior, R. Reljic,
S. Vermaak, M. Vordermeier, A. Williams, F1000Res. 7 (2018) 485.
[4] S.M. Gossage, M.E. Rogers, P.A. Bates, Int. J. Parasitol. 33 (2003) 1027–1034.
[5] P. Rees, P. Kager, M. Keating, W. Hockmeyer, Lancet 316 (1980) 226–229.
[6] J. Chulay, L. Fleckenstein, D. Smith, Trans. Roy. Soc. Trop. Med. Hyg. 82 (1988)
69–72.
4. Conclusions
[7] J.C. Mottram, G.H. Coombs, Exp. Parasitol. 59 (1985) 151–160.
[8] W.L. Roberts, J.D. Berman, P.M. Rainey, Antimicrob. Agents Chemother. 39 (1995)
1234–1239.
This study has investigated the reduction of analogous organome-
tallic complexes of Sb(V) and Bi(V) in a range of pH values with try-
panothione, which is parasite specific, and its mammalian tripeptide
analogue glutathione. In the clinically relevant pH range of 4–5, both
[BiPh3(O2CCH2(C6H4CH3))2] (B1) [SbPh3(O2CCH2(C6H4CH3))2] (S1),
are rapidly reduced by trypanothione and glutathione. However, in the
presence of glutathione at higher pH values differing reduction rates are
observed. The highly oxidising bismuth complex B1 underwent rapid
reduction over a matter of a few minutes, whilst the reduction rate of the
antimony complex S1 was significantly slower, with complete degra-
dation taking approximately 8 h. A half-life of t = 54 min was calculated
for the logarithmic decay of complex S1 into SbPh3 and the parent p-
tolylacetic acid. Formation of the Sb(Gs)3 complex was absent for S1,
however was observed to form from B1, (Bi(GS)3), providing some
valuable insights into the chemical nature of the selective cytotoxicity
observed for this class of complex. This rate of reduction was observed to
be more rapid than that of the front-line anti-leishmanial drug meglu-
mine antimoniate. This study has established a significant difference in
the cellular chemistry between Bi(V) and Sb(V) triaryl carboxylates,
which can be linked to their distinctly different biological activity and
selectivity.
[9] D. Sereno, J.-L. Lemesre, Antimicrob. Agents Chemother. 41 (1997) 972–976.
[10] M. Ephros, A. Bitnun, P. Shaked, E. Waldman, D. Zilberstein, Antimicrob. Agents
Chemother. 43 (1999) 278–282.
´
[11] F. Frezard, C. Demicheli, C.S. Ferreira, M.A. Costa, Antimicrob. Agents Chemother.
45 (2001) 913–916.
[12] S. Yan, F. Li, K. Ding, H. Sun, J. Biol. Inorg. Chem. 8 (2003) 689–697.
[13] S. Yan, I.L.K. Wong, L.M.C. Chow, H. Sun, Chem. Commun. 2 (2003) 266–267.
[14] S. Yan, K. Ding, L. Zhang, H. Sun, Angew. Chem. Int. 39 (2000) 4260–4262.
[15] A.H. Fairlamb, A. Cerami, Annu. Rev. Microbiol. 46 (1992) 695–729.
[16] S. Wyllie, M.L. Cunningham, A.H. Fairlamb, J. Biol. Chem. 279 (2004)
39925–39932.
[17] P. Baiocco, G. Colotti, S. Franceschini, A. Ilari, J. Med. Chem. 52 (2009)
2603–2612.
[18] J. Alexander, D.G. Russell, Adv. Parasitol. 31 (1992) 175–254.
[19] H. Sun, S.C. Yan, W.S. Cheng, Eur. J. Biochem. 267 (2000) 5450–5457.
[20] C. dos Santos Ferreira, P.S. Martins, C. Demicheli, C. Brochu, M. Ouellette,
´
F. Frezard, Biometals 16 (2003) 441–446.
[21] B.L. Herwaldt, J.D. Berman, Am. J. Trop. Med. Hyg. 46 (1992) 296–306.
[22] M.I. Ali, M.K. Rauf, A. Badshah, I. Kumar, C.M. Forsyth, P.C. Junk, L. Kedzierski, P.
C. Andrews, Dalton Trans. 42 (2013) 16733–16741.
[23] R.N. Duffin, V.L. Blair, L. Kedzierski, P.C. Andrews, Dalton Trans. 47 (2018)
971–980.
[24] R.N. Duffin, V.L. Blair, L. Kedzierski, P.C. Andrews, J. Inorg. Biochem. 189 (2018)
151–162.
[25] Y.C. Ong, V.L. Blair, L. Kedzierski, P.C. Andrews, Dalton Trans. 43 (2014)
12904–12916.
[26] Y.C. Ong, V.L. Blair, L. Kedzierski, K.L. Tuck, P.C. Andrews, Dalton Trans. 44
(2015) 18215–18226.
List of abbreviations
[27] T. Iftikhar, M.K. Rauf, S. Sarwar, A. Badshah, D. Waseem, M.N. Tahir, A. Khan, K.
M. Khan, G.M. Khan, J. Organomet. Chem. 851 (2017) 89–96.
[28] R. Mushtaq, M.K. Rauf, M. Bolte, A. Nadhman, A. Badshah, M.N. Tahir,
M. Yasinzai, K.M. Khan, Appl. Organomet. Chem. 31 (2017) 1–8.
[29] L. Saleem, A.A. Altaf, A. Badshah, M.K. Rauf, A. Waseem, M. Danish, S.S. Azam, M.
N. Arshad, A.M. Asiri, S. Ahmad, Inorg. Chim. Acta 474 (2018) 148–155.
[30] S. Sarwar, T. Iftikhar, M.K. Rauf, A. Badshah, D. Waseem, M.N. Tahir, K.M. Khan,
G.M. Khan, Inorg. Chim. Acta 476 (2018) 12–19.
GSH
TSH
Glutathione
Trypanothione
Oxidised glutathione
GSSG
1H NMR Proton nuclear magnetic resonance
FT-IR
CLogP
GS
Fourier transform infrared
Partition co-efficient
[31] A.I. Antoniou, D.A. Pepe, D. Aiello, C. Siciliano, C.M. Athanassopoulos, J. Org.
Chem. 81 (2016) 4353–4358.
[32] M.A. Ilies, W.A. Seitz, B.H. Johnson, E.L. Ezell, A.L. Miller, E.B. Thompson, A.
T. Balaban, J. Med. Chem. 49 (2006) 3872–3887.
Complexed glutathione
[33] H. Sun, Biological Chemistry of Arsenic, Antimony and Bismuth, Wiley Online
Library, 2011.
Declaration of Competing Interest
[34] J. Alexander, A.R. Satoskar, D.G. Russell, J. Cell Sci. 112 (1999) 2993–3002.
[35] H.J. Forman, H. Zhang, A. Rinna, Mol. Asp. Med. 30 (2009) 1–12.
[36] A. Meister, M.E. Anderson, Annu. Rev. Biochem. 52 (1983) 711–760.
[37] J. Harlan, J. Levine, K. Callahan, B. Schwartz, L. Harker, J. Clin. Invest. 73 (1984)
706–713.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
[38] J.R. Babson, N.S. Abell, D.J. Reed, Biochem. Pharmacol. 30 (1981) 2299–2304.
[39] I.V. Tetko, J. Gasteiger, R. Todeschini, A. Mauri, D. Livingstone, P. Ertl, V.
A. Palyulin, E.V. Radchenko, N.S. Zefirov, A.S. Makarenko, J. Comp.Mol. Des. 19
(2005) 453–463.
Acknowledgements
´
[40] F. Frezard, P.S. Martins, M.C. Barbosa, A.M. Pimenta, W.A. Ferreira, J.E. de Melo,
The authors would like to thank the Australian Research Council
(DP170103624) and Monash University for financial support.
J.B. Mangrum, C. Demicheli, J. Inorg. Biochem. 102 (2008) 656–665.
[41] R. Ge, H. Sun, Acc. Chem. Res. 40 (2007) 267–274.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
8