Highly Cytotoxic Bioconjugated Gold(I) Complexes with Cysteine-Containing Dipeptides

Article abstract of DOI:10.1002/chem.201501458

Several gold(I) complexes with cysteine-containing dipeptides have been prepared starting from cystine by coupling different amino acids and using several orthogonal protections. The first step is the reaction of cystine, where the sulfur centre is protected as disulfide, with Boc₂O in order to protect the amino group, followed by coupling of an amino acid ester; finally the disulfide bridge is broken with mercaptoethanol to afford the dipeptide derivative. Further reaction with [AuCl(PPh₃)] gives the gold-dipeptide-phosphine species. Starting from these formally gold(I) thiolate-dipeptide phosphine complexes with the general formula [Au(SR)(PR₃)] different structural modifications, such as change in the type of the amino protecting group, the type of phosphine, the number of gold(I) atoms per molecule, or the use of a non-proteinogenic conformationally restricted amino acid ester, were introduced in order to evaluate their influence in the biological activity of the final complexes. The cytotoxic activity, in vitro, of these complexes was evaluated against different tumour human cell lines (A549, MiaPaca2 and Jurkat). The complexes show an outstanding cytotoxic activity with IC₅₀ values in the very low micromolar range. Structure-activity relationship studies from the complexes open the possibility of designing more potent and promising gold(I) anticancer agents. Striking with gold: Gold(I) complexes with cysteine-containing dipeptides were prepared starting from cystine by coupling different amino acids, and using several orthogonal protections. The complexes show excellent cytotoxic activity with IC₅₀ values in the very low micromolar range. The structural changes in the parent compound led to the synthesis of the most effective compound in the cell lines tested, which is the complex with two AuPPh₃⁺ fragments coordinated to the Boc-Cys-Gly-OMe peptide (see scheme).

Full text of DOI:10.1002/chem.201501458

DOI: 10.1002/chem.201501458
Full Paper
&
Medicinal Chemistry
Highly Cytotoxic Bioconjugated Gold(I) Complexes with Cysteine-
Containing Dipeptides
Alejandro GutiØrrez,[a] Isabel Marzo,[b] Carlos Cativiela,[c] Antonio Laguna,[a] and
M. Concepción Gimeno*[a]
Abstract: Several gold(I) complexes with cysteine-containing
dipeptides have been prepared starting from cystine by cou-
pling different amino acids and using several orthogonal
protections. The first step is the reaction of cystine, where
the sulfur centre is protected as disulfide, with Boc2O in
order to protect the amino group, followed by coupling of
an amino acid ester; finally the disulfide bridge is broken
with mercaptoethanol to afford the dipeptide derivative.
Further reaction with [AuCl(PPh3)] gives the gold-dipeptide-
phosphine species. Starting from these formally gold(I) thio-
late–dipeptide phosphine complexes with the general for-
mula [Au(SR)(PR3)] different structural modifications, such as
change in the type of the amino protecting group, the type
of phosphine, the number of gold(I) atoms per molecule, or
the use of a non-proteinogenic conformationally restricted
amino acid ester, were introduced in order to evaluate their
influence in the biological activity of the final complexes.
The cytotoxic activity, in vitro, of these complexes was evalu-
ated against different tumour human cell lines (A549, MiaPa-
ca2 and Jurkat). The complexes show an outstanding cyto-
toxic activity with IC50 values in the very low micromolar
range. Structure–activity relationship studies from the com-
plexes open the possibility of designing more potent and
promising gold(I) anticancer agents.
have been prepared.[3] In addition, the mechanistic studies
show that in general DNA is not the main target and interac-
tions with several enzimes,[4] and more specifically the seleno-
enzime thioredoxine reductase,[5] proteasome,[6] kinases,[7]
among others have been reported as the modes that finally
lead to apoptosis through a mitochondrial pathway.
Introduction
Metal-based drugs are an important class of compounds in
medicinal chemistry, employed clinically for the treatment of
different diseases.[1] Cisplatin and the following generations of
platinum-based drugs are among the most widely used che-
motherapeutic agents.[2] However, their effectiveness is still re-
stricted by several clinical problems, such as a limited spec-
trum of activity, development of resistance and undesired toxic
side-effects. The use of different metal compounds with differ-
ent biological properties and targets has emerged as an alter-
native strategy to overcome these limitations.
Diverse approaches have been used to prepare gold(I) and
gold(III) drugs. For gold(I) the synthesis has been focused on
the use of auranofin analogues[8] or the coordination to gold
of several ligands mainly phosphines,[9] and carbenes.[10] For
gold(III), cyclometallated complexes or coordination to the
metallic centre of bypiridine, dithiocarbamate or porphyrin li-
gands have been successfully employed in order to prepare
new anticancer gold drugs.[11]
Gold compounds have been employed since the last century
for the treatment of rheumatoid arthritis and in recent decades
some gold(I) and gold(III) complexes with promising biological
activities as anticancer, antimicrobial, fungicidal, anti-HIV, or in
the treatment of asthma or parasitic diseases, among others,
Surprisingly and in spite of the great number of reported
gold complexes which show activity in biological systems, the
number of gold species with biomolecules such as amino acids
and peptides is very scarce. Peptide research on drug design
and drug discovery is one of the most promising fields in the
development of new drugs. Peptide sequences are constitu-
ents of larger proteins, where they are responsible for molecu-
lar recognition and biological activities.[12]
[a] Dr. A. GutiØrrez, Prof. A. Laguna, Prof. M. C. Gimeno
Departamento de Química Inorgµnica, Instituto Síntesis Química y Catµlisis
HomogØnea (ISQCH), CSIC-Universidad de Zaragoza
Pedro Cerbuna 12, 50009 Zaragoza (Spain)
E-mail: gimeno@unizar.es
There are only a few reports of reactions of gold(I) and
mainly gold(III) with amino acids such as cysteine and methio-
nine.[13] Gold(I) peptides with the cysteine-modified neuropep-
tide encephalin[14] and by gold(I) azide cycloaddition reactions
with alkynyl peptides have been reported by Metzler-Nolte
and co-workers.[15] The latter type of compounds has shown to
overcome cisplatin resistance in a p53-mutant cancer cell line.
The crystal structure of a gold(I) complex with the human glu-
[b] Dr. I. Marzo
Departamento de Bioquímica y Biología Molecular
Universidad de Zaragoza, Pedro Cerbuna 12, 50009 Zaragoza (Spain)
[c] Prof. C. Cativiela
Departamento de Química Orgµnica, Instituto Síntesis Química y Catµlisis
HomogØnea (ISQCH), CSIC-Universidad de Zaragoza
Pedro Cerbuna 12, 50009 Zaragoza (Spain)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201501458.
Chem. Eur. J. 2015, 21, 11088 – 11095
11088
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tathione reductase[16] and also a gold protein with the AuPEt3
fragment bonded to a histidine rest[17] have been reported,
and could give interesting insights on the biological function
of gold compounds. We have previously reported on the syn-
thesis and functionalisation of gold(I)-phosphine-nicotinic acid
thiolate with amino acids providing complexes with very good
cytotoxic activity.[18] Additionally, gold(III) peptide derivatives
with histidine residues[19] and dithiocarbamate functionalised
dipeptides have also been described.[20]
tecting groups for each functional unit (amino, carboxylic acid
and thiol) of the dipeptide.[22]
In the cystine the sulfur atom is already protected as disul-
fide, so the first step consisted of the protection of the amino
group using Boc (tert-butoxycarbonyl) according to previously
described conditions (Scheme 1). The success of the reaction
1
was easily confirmed by H NMR spectroscopy, and the spec-
trum shows the appearance of the signals corresponding to
the Boc group at d=6.79 (carbamate proton) and 1.36 ppm
(alkyl protons). The following step was the coupling of the
amino acid ester through the free carboxylic acid of (Boc-Cys-
OH)2 through standard solution peptide chemistry (employing
PyBOP as activating agent and DIPEA as base), and subsequent
flash chromatography yielded 1a–6a derivatives in good
yields. Six proteinogenic amino acid esters from glycine, ala-
nine, valine, phenylalanine, methionine or proline were select-
ed to prepare the dipeptide ligands. Each amino acid differs in
the side chain group, which determine properties such as lipo-
philicity, polarity, reactivity or steric hindrance, and could have
influence in the exchange reactions that are supposed to give
the final complexes with other biomolecules in the organism,
In this context and with the aim of preparing more active
and selective gold drugs, we propose the formation of gold(I)
complexes with dipeptides. The introduction of peptides in the
complexes might decrease the undesired toxic side-effects
(they are biocompatible ligands) and they could serve as good
carriers for the delivery of the gold atom to the biological
target.[21] A major problem for cancer chemotherapeutics is
crossing the cell membrane, and a strategy to overcome this is
to make use of the peptide-based delivery systems, which can
transport small peptides and peptide-like drugs to the cells.
These peptide transporters have over-expressed receptors in
some types of tumours, which could be a target for gold–pep-
tide compounds, providing them with higher levels
of internalization and subsequent selectivity.[20]
Here we report on an efficient and general method
for preparing dipeptides containing
a cysteine
moiety and different proteinogenic amino acids,
using different orthogonal protections. The coordina-
tion of these dipeptides to a gold(I) phosphine frag-
ment gives novel and stable gold–peptide bioconju-
gates, the cytotoxic activities of which were studied.
Several structural modifications of these peptide–
gold-phosphine complexes were performed in order
to determine which factor could improve their activi-
ty. Consequently, changes in the phosphine ligand,
the amino protecting group, the number of coordi-
+
nated AuPPh3 units, or other modifications, such as
Scheme 1. Synthesis of the gold(I) dipeptide complexes 1–6.
the introduction of a conformationally restricted non-
proteinogenic amino acid (Oic) in the peptide were
performed.
Cytotoxic activity in vitro against different tumour cell lines
was analysed, showing outstanding cytotoxicity in all cases
with IC50 values in the very low micromolar range.
affecting to the final activity, selectivity or biodistribution of
the complex.
1
As common features, the H and 13C NMR spectra of com-
pounds 1a–5a show, in addition to all the expected resonan-
ces for the molecule, the signals belonging to the new amide
bond in the range d=8.16–7.40 ppm and d=170.5–
169.2 ppm, respectively. Compound 6a appears as a mixture
of rotamers (ratio 1:0.2) as a consequence of the conformation-
ally restricted structure of proline, which possesses a five-mem-
bered ring (pyrrolidine) in its structure, giving a more complex
spectrum. Selective reduction of the disulfides of 1a–6a em-
ploying a strong reducing thiol (b-mercaptoethanol)[23] in soft
basic media (DIPEA) afforded the desired dipeptide thiol li-
gands 1b–6b. Six different dipeptides containing cysteine
groups have been prepared according to the synthetic route
show in Scheme 1: Boc-Cys-Gly-OMe (1b), Boc-Cys-Ala-OMe
(2b), Boc-Cys-Val-OMe (3b), Boc-Cys-Phe-OMe (4b), Boc-Cys-
Results and Discussion
Gold(I) dipeptide phosphine complexes
In our aim of preparing gold complexes with cysteine-contain-
ing dipeptides we thought of using as precursor the previously
described complex, Ac-Cys(AuPPh3)-OH, in which there is an
acid moiety that could couple with other amino acid esters, in
a similar manner as we have reported earlier for the gold-phos-
phine complexes with the nicotinic acid thiolate.[18a] However,
the preparation of the desired complexes was not possible by
this route, probably because of solubility reasons. Then, a new
synthetic strategy was selected, which consists of the use of
commercially available cystine with different orthogonal pro-
1
Met-OMe (5b) and Boc-Cys-Pro-OMe (6b). The H NMR spectra
Chem. Eur. J. 2015, 21, 11088 – 11095
www.chemeurj.org
11089
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
present the resonance for the thiol group as a doublet of dou-
blets or as a multiplet in the range d=1.46-1.90 ppm, while
the resonances corresponding to Cb protons appear each one
with a characteristic shape as doublet of doublet of doublets
(diastereotopic protons). The IR spectra also present the ab-
sorption bands corresponding to the thiol group in the range
2562–2580 cmÀ1 as weak bands. The dipeptides 1b–6b are val-
uable ligands that easily react with [AuCl(PPh3)] employing
K2CO3 as a base, to give the desired complexes 1–6 in high
yields after chromatographic purification. Complexes 1–6, just
as all the complexes reported in this work, were fully character-
Structural modifications
The structure of complexes 1–6 described above, allows the in-
troduction of several structural modifications with the purpose
of obtaining information on how the changes in the structure
affect the biological properties of the complexes, and thus we
can stablish a structure–activity relationship (SAR), and can
obtain compounds with better activity, selectivity and pharma-
ceutical profile. The selected structural modifications per-
formed were: 1) alteration of the phosphine (7), 2) alteration of
the amino protecting group (8), 3) coupling of a non-proteino-
genic conformationally restricted amino acid ester (9), and
4) alteration of the charge and number of gold atoms per mol-
ecule (10–11). All of these modifications could influence the
polarity, electronic or steric properties of the complex, activity
or selectivity, and alter the biodistribution or reactivity of the
final complex towards other biomolecules in the organism.
The different modifications carried out are summarized in
Figure 2.
1
ized by H, 13C and 31P NMR spectroscopy, IR spectroscopy and
MS spectrometry. The 1H NMR spectra show the resonances
corresponding to the dipeptide and triphenylphosphine pro-
tons. The proton for the thiol group has disappeared, which
agrees with coordination of the sulfur to gold affording the de-
sired complexes. Remarkably, as a common feature, the reso-
nances of the Cb protons (diastereotopic protons) are simpli-
fied and each one appears as a doublet of doublets at approxi-
mately d=3.60 and 3.25 ppm, strongly downfield shifted rela-
tive to the starting dipeptide. The 31P{1H} NMR spectra present
a unique resonance for all of the complexes in the range d=
36–38 ppm, which agrees with similar gold(I) complexes, with
thiolates and phosphines as ligands previously reported, and is
3–5 ppm downfield shifted relative to the starting gold(I) com-
plex. The 13C{1H} NMR APT spectra show the resonances for the
dipeptide and triphenylphosphine moieties. Again, the signals
belonging to Cb appear downfield shifted after complex for-
mation (in the range of 30–35 ppm). The assignment of the
1
1
resonances was made with the H-13C HSQC and H-13C HMBC
experiments. The IR spectra present, among others, the ab-
sorptions for the amide, ester and aromatic unit and also the
disappearance of the absorptions for the thiol group. The mass
spectra (HRMS ESI+) present for all the compounds the molec-
ular peak [M+Na]+ and [M+H]+. The different gold(I) com-
plexes with the dipeptides are shown in Figure 1.
Figure 2. Structural modifications introduced in the complexes.
Preparation of the complex with a different phosphine
ligand (7)
The phosphine PPh2Py was selected to prepare the analogue
of complex 1. Because the type of phosphine coordinated to
gold(I) has a great influence in the activity of the final complex,
determining factors such as reactivity, behaviour in exchange
reactions or biodistribution, the change for the related phos-
phine with a pyridine moiety PPh2Py was employed as ancillary
ligand, which could bind further to other metals, interact with
other biomolecules or alter the lipophilicity of the complex.
Treatment of Boc-Cys-Gly-OMe (1b) with [AuCl(PPh2Py)] afford-
ed complex 7 in high yield after chromatographic purification
(Scheme 2).
1
The H and 13C{1H} NMR spectra show the expected resonan-
ces from all the protons and carbon atoms, which are essen-
tially the same as those observed for the analogous complex
Figure 1. Gold(I) dipeptide complexes 1–6.
Scheme 2. Synthesis of complex 7.
Chem. Eur. J. 2015, 21, 11088 – 11095
www.chemeurj.org
11090
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1, and will not be discussed further here. The only exceptions
are the signals that belong to the pyridine, for which a correct
assignment could be realized by 2D NMR spectroscopy (1H-1H
shows, as the most remarkable features, the appearance of the
signal of the thiol group at d=1.70 ppm as a doublet of dou-
blets and the diastereotopic Cb protons appears each one
with the characteristic shape as a doublet of doublet of dou-
blets. The IR spectrum clearly shows a band corresponding to
the absorption of the SH group at 2579 cmÀ1. The last step
consisted of the deprotonation of the thiol and coordination
of the corresponding thiolate to the gold(I) centre as described
above to give the desired complex 8 in pure form and high
yield. Complex 8 is also analogous to 1 and consequently in
the spectroscopic data the same features are observed: the
1H NMR spectrum shows the disappearance of the thiol proton
and the Cb protons resonance appears downfield shifted and
with a more simplified shape. In the 13C NMR spectrum the
signal belonging to Cb is also shifted downfield, and in the IR
spectrum the thiol band also disappears after the formation of
the complex, which agrees with the proposed structure.
COSY, H-13C HSQC and H-13C HMBC) since these resonances
and those of the phenyl rings appear at similar chemical shift.
Again, the 31P{1H} NMR spectra present a unique resonance
corresponding to the phosphorus of the 2-diphenylphosphino
pyridine ligand, the chemical shift is observed at d=37.0 ppm,
a value similar to that observed for 1. IR and MS data agree
with the proposed structure.
1
1
Preparation of the complex with different amino protecting
group (8)
Boc (tert-butoxycarbonyl) was employed as the amino protect-
ing group in the preparation of the complexes 1–6. In this
case, the Z (benzyloxycarbonyl) group was selected in order to
evaluate how this change affects the biologic activity. Al-
though in both cases the amino group is protected as a carba-
mate, the change introduced could affect the polarity or ex-
change reactions with other biomolecules in the organism,
which would be important for the activity and selectivity of
the drug. The [Au(Z-Cys-Gly-OMe)(PPh3)] complex 8 was pre-
pared following a similar synthetic route to that used in the
preparation of complexes 1–6 (Scheme 3).
Preparation of the complex with different coupled amino
acid ester (9)
The type of amino acid ester has influence in the ability of the
carrier to deliver the gold(I) to its target. In fact, in a previous
work we observed that functionalisation with proline usually
gives better cytotoxic complexes compared with other amino
acids.[18b] In this case, we decided
to couple to the cystine deriva-
tive the octahydroindole amino
acid ester (Oic), a non-proteino-
genic bicyclic proline ana-
logue.[24] This amino acid has
been employed in the design of
peptides with interesting phar-
maceutical profiles, mainly be-
cause conformationally restricted
amino acids help to fix confor-
mations and have influence in
the interactions with receptors
in the organism. Moreover, the
Scheme 3. Synthesis of complex 8.
dipeptides
containing
this
moiety could present a higher
resistance to hydrolysis by di-
Again, orthogonal protection of the different functional
groups in the molecule was necessary. In the first place, start-
ing from the disulfide 1a, cleavage of Boc amino protecting
group was carried out employing standard conditions by treat-
ment with 3m HCl/AcOEt solution. Introduction of the Z pro-
tecting group was successfully achieved by treatment with
a convenient source as Z-Cl, employing DIPEA as a base and
DMAP as catalyst. The intermediate 8b was obtained in mod-
erate yield after chromatographic purification. Discussion of
spectroscopic data (1H and 13C NMR spectra) is similar to that
described previously for the analogue 1b with the exception
that there are signals belonging to the benzyl group instead of
the tert-butyl group as can be found in the Supporting Infor-
mation. Disulfide reduction was carried out in the same way as
verse enzymes. For all of these reasons, the preparation of
a complex bearing this moiety is expected to improve the ac-
tivity and selectivity. The synthetic route employed to prepare
the complex 9 is shown in the Scheme 4.
As can be observed, this synthetic route is essentially the
same as described previously for the preparation of the dipep-
tide complexes 1–6 and will not be further discussed here. In
this case, the coupling of another amino ester type was carried
out. Similar to complex 6 bearing the proline moiety, complex
9 appears as a mixture of rotamers in the NMR spectra. This
complex possesses a high number of protons and carbon
atoms, many of them showing similar chemical shifts. Never-
theless, all the signals could be correctly assigned by the reali-
1
zation of 2D NMR experiments (1H-1H COSY, H-13C HSQC and
1
described above. Again, the H NMR spectrum of dipeptide 8c
1H-13C HMBC).
Chem. Eur. J. 2015, 21, 11088 – 11095
www.chemeurj.org
11091
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
probably due to more steric hindrance. In all cases we ob-
served the expected resonances for the protons belonging to
1
the dipeptide and triphenylphosphine. In the H NMR spectra
of complexes 10 and 11, the carbamate and amide protons,
and all the protons belonging to the dipeptide appear more
downfield shifted than observed for complex 1. Integration of
aromatic protons corresponding to triphenylphosphine ligands
also confirms the coordination of additional gold(I) triphenyl-
phosphine moieties. The 19F NMR spectra show a single signal
at d=À77.8 ppm that belongs to the triflate counter anion in
both cases. However, the most notable feature is the chemical
shift for the phosphorus atoms in the 31P NMR spectra, which
appears at d=33.2 ppm (dinuclear complex) and d=32.3 ppm
(trinuclear complex), that are strongly up-field shifted (4 and
5 ppm, respectively) compared with 1. The mass spectra
(ESI+) presents the molecular peak [M]+.
Scheme 4. Synthesis of complex 7.
Preparation of the complexes with higher number of gold
atoms per molecule (10 and 11)
Diverse studies indicate that in gold(I) complexes with thiolates
and phosphines as ligands the cytotoxicity is due mainly to
the metallic centre; whereas the phosphine ligand allows the
gold(I) atom to cross the cell membrane, the thiolate takes
part in exchange ligand reactions with other biomolecules.
Keeping this in mind, it is logical to think that adding more
gold(I) atoms per molecule will increase the cytotoxicity of the
complex, as we observed previously.[18b] Moreover, there is
a strong interest in the synthesis of lipophilic and cationic
compounds which seem to be able to cross cell membranes
and selectively accumulate in mitochondria,[25] just where it ap-
pears that gold(I) complexes exert their therapeutic effect, by
inhibiting diverse enzymes as for example thioredoxin reduc-
tase (TrX). Moreover, further coordination of gold(I) atoms to
sulfur could have a strong influence on the transference reac-
tions of the metal to other biomolecules, determining its trans-
port, biodistribution or enzyme-inhibition ability. For all of
these reasons, we carried out the synthesis of the dinuclear
(10) and trinuclear (11) cationic complexes derived from the
mononuclear complex 1 by treatment with the high electro-
philic complex [Au(OTf)(PPh3)] generated in situ (Scheme 5).
Cytotoxic activity
The cytotoxicity of complexes 1–11 was tested against three
different human tumour cell lines: Jurkat (T-cell leukaemia), Mi-
aPaca2 (pancreatic carcinoma) and A549 (lung carcinoma), and
compared to the results with cisplatin.
Compounds 1–11 are not soluble in water, but they are solu-
ble in DMSO and in the DMSO/water mixtures used in the
tests, which contain a small amount of DMSO. We did not ob-
serve any precipitation of the complexes or metallic gold while
performing the tests. Their colourless [D6]DMSO solutions are
very stable at room temperature, as shown in the 1H NMR
spectra in which the signals remain the same for weeks. Cells
were exposed to different concentrations of each compound
for a total of 24 h. Using the colorimetric MTT viability assay
(MTT=3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyltetrazolium
bromide), the IC50 values (final concentration <0.5% DMSO)
were calculated from dose-response curves obtained by non-
linear regression analysis. IC50 values are concentrations of
a drug required to inhibit tumour cell proliferation by 50%,
compared to the control cells treated with DMSO alone. The
IC50 values for complexes 1–11 are collected in Table 1. These
values can be compared with those reported for cisplatin dis-
solved in water: IC50 at 24 h in A549, MiaPaca2 and Jurkat cells
are 114.2, 76.5 and 10.8 mm, respectively.[26]
1
Remarkably, the H NMR spectrum of complex 11 appears as
a mixture of rotamers in contrast with complexes 1 and 10,
Table 1. IC50 [mm] (24 h), with standard deviations, of complexes against
A549, MiaPaca2 and Jurkat cell lines.
Complex
A549
MiaPaca2
Jurkat
1
2
3
4
5
6
7
8
1.5Æ0.2
1.9Æ0.1
2.3Æ0.1
15.6Æ0.11
4.8Æ0.1
3.0Æ0.1
5.0Æ0.2
2.7Æ0.1
2.1Æ0.1
1.8Æ0.1
3.5Æ0.1
2.0Æ0.2
1.9Æ0.1
3.0Æ0.1
5.4Æ0.1
1.8Æ0.1
0.7Æ0.1
0.5Æ0.1
1.5Æ0.1
1.2Æ0.1
0.1Æ0.1
1.5Æ0.1
0.9Æ0.1
1.6Æ0.1
2.2Æ0.1
0.4Æ0.1
1.7Æ0.1
0.5Æ0.1
0.8Æ0.1
1.1Æ0.1
1.5Æ0.1
0.6Æ0.1
0.8Æ0.1
9
10
11
Scheme 5. Synthesis of the dinuclear (10) and trinuclear (11) complexes de-
rived from 1. i) [Au(OTf)(PPh3)], ii) 2 [Au(OTf)(PPh3)].
Chem. Eur. J. 2015, 21, 11088 – 11095
www.chemeurj.org
11092
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
All the synthesized complexes were active against the differ-
ent tumour cell lines in very low concentrations (low micromo-
lar range). Complexes 1–11 exhibit excellent antiproliferative
activities, with IC50 values ranging from 1.5 to 15.6 mm in A549
cells, 0.4 to 2.2 mm in Jurkat cells, and 0.1 to 5.4 mm in MiaPa-
ca2. The Jurkat and MiaPaca2 cell lines were the most sensitive
to our compounds, whereas A549 showed more resistance to
the complexes.
protecting group, the type of phosphine, the number of
+
AuPPh3 fragments coordinated to the sulfur centre, or the
use of non-proteinogenic conformationally restricted amino
acid ester in the peptide, were introduced in order to evaluate
their influence in the biological activity of the final complexes.
The cytotoxic activity, in vitro, of these complexes was evaluat-
ed against different tumour human cell lines (A549, MiaPaca2
and Jurkat). The complexes show excellent cytotoxic activity
with IC50 values in the very low micromolar range (as low as
0.1 mm). The structural changes made to the parent compound
led to the synthesis of the most effective compound in all the
The gold(I) dipeptide complexes (1–6) displayed very good
cytotoxicity in all the tumour cell lines. The type of the protei-
nogenic amino ester coupled to cysteine has some influence
on the cytotoxicity of the final complex: the complexes which
incorporate glycine or proline in their structure showed the
best IC50 values. The change of the type of phosphine ligand
coordinated to gold(I) (complex 7), the amino protecting
group (complex 8) or the amino acid ester coupled to cysteine
(complex 9) led to more potent complexes in the MiaPaca2
tumour cell line, although in the A549 resulted in a decrease
of the cytotoxicity of the complex in comparison with the re-
lated complex 1. The coordination of an additional [AuPPh3]+
fragment (complex 10) to 1 has a strong influence in the cyto-
toxicity, mainly in the MiaPaca2 and Jurkat tumour cell lines. In
particular, complex 10 was the most potent of all the series.
Surprisingly, coordination of two additional [AuPPh3]+ frag-
ments (complex 11) to 1 did not improve the potency of the
complex. Then, the use of a dipeptide containing glycine or
proline, or the coordination of two gold(I) centres to the sulfur
atom, giving a cationic complex, yielded the most potent cyto-
toxic complexes.
+
cell lines tested, which is the complex with two AuPPh3 frag-
ments coordinated to the Boc-Cys-Gly-OMe peptide.
Experimental Section
Instrumentation
C, H, and N analysis were carried out with a Perkin–Elmer 2400 mi-
croanalyzer. Mass spectra were recorded on a Bruker Esquire 3000
Plus, with the electrospray (ESI) technique and on a Bruker Micro-
flex (MALDI-TOF). 1H, 13C{H}, 31P{H} and 19F NMR, including 2D ex-
periments, were recorded at room temperature on a Bruker
Avance 400 spectrometer (1H, 400 MHz, 13C, 100.6 MHz) or on
a Bruker Avance II 300 spectrometer (1H, 300 MHz, 13C, 75.5 MHz),
with chemical shifts (d, ppm) reported relative to the solvent peaks
of the deuterated solvent.[21]
Starting materials
[AuCl(PPh3)] and [AuCl(PPh2Py)] were prepared according to pub-
lished procedures.[27] All other reagents were commercially avail-
able. Solvents were used as received without purification or drying.
Comparison of the activity of this family of gold–peptide de-
rivatives with the anticancer drug cisplatin, shows that these
complexes displayed much more in vitro cytotoxic activity. In
relation with other gold(I)–peptide species our complexes have
much lower IC50 values, although comparison with the same
cell lines is not possible. The activity is also higher than in the
gold(I)–nicotinic acid thiolate functionalised with amino acids
previously described by us.[14b] Nevertheless, the very low
values obtained in these complexes in resistant cell lines such
as A549 and MiaPaca2 measured after 24 h, represent out-
standing values, which makes them very promising metallo-
drug candidates for continuing their evaluation and studying
their mechanism of action.
General procedure for coupling amino acid esters: synthesis
of compounds 1a–6a and 9a (Procedure A)
To a solution of (Boc-Cys-OH)2 (1 mmol) in anhydrous DMF (5 mL)
was added the corresponding amino ester hydrochloride
(2.4 mmol), PyBOP (2.2 mmol) and DIPEA (6.6 mmol). The mixture
was stirred for 48 h under argon atmosphere at room temperature.
Then, the resultant clear solution was diluted with AcOEt (100 mL)
and washed with water (5”60 mL). The organic phase was dried
over anhydrous MgSO4, filtered off and evaporated to dryness. The
crude of the reaction was purified by column chromatography on
silica gel using as eluent a mixture of ethyl acetate/hexane (1:1).
Conclusions
General procedure for reducing disulfides to thiols: synthe-
sis of compounds 1b–6b, 8c and 9b (Procedure B)
A series of gold(I) complexes with cysteine-containing dipepti-
des have been prepared starting from cystine by coupling dif-
ferent amino acids, and using several orthogonal protections.
In these molecules the gold centre is coordinated directly to
the dipeptide, to the sulfur of the cysteine as thiolate. These
novel derivatives with biologically relevant molecules could de-
liver the gold centre selectively to the tumour cells because
they can act as peptidomimetics and target peptide delivery
systems, which have over-expressed receptors in tumour cells.
In these formal gold(I) thiolate–dipeptide phosphine com-
plexes, with the general formula [Au(SR)(PR3)], different struc-
tural modifications such as change in the type of the amino
To a solution of the corresponding disulfide (Boc-Cys-aa-OMe)2
(1 mmol) in dry CH2Cl2 (50 mL), b-mercaptoethanol (4 mmol) and
DIPEA (4 mmol) were added. The mixture was stirred for 48 h
under argon atmosphere at room temperature. Then, the solvent
was evaporated under reduced pressure and the crude of reaction
redissolved in AcOEt (100 mL) and washed with an aqueous satu-
rated solution of KHSO4 (3”40 mL) and a saturated aqueous solu-
tion of NaCl (3”40 mL). The organic phase was dried over anhy-
drous MgSO4, filtered off and evaporated to dryness. The crude of
the reaction was purified by column chromatography on silica gel
using as eluent a mixture of acetone/hexane (1:1).
Chem. Eur. J. 2015, 21, 11088 – 11095
www.chemeurj.org
11093
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[3] a) E. R. T. Tiekink, Inflammopharmacology 2008, 16, 138; b) I. Ott, Coord.
Chem. Rev. 2009, 253, 1670; c) A. Casini, L. Messori, Curr. Top. Med.
Chem. 2011, 11, 2647; d) B. Bertrand, A. Casini, Dalton Trans. 2014, 43,
4209; e) M. Navarro, Coord. Chem. Rev. 2009, 253, 1619; f) C. M. Che,
R. W. Y. Sun, Chem. Commun. 2011, 47, 9554.
[4] K. P. Bhabak, B. J. Bhuyan, G. Mugesh, Dalton Trans. 2011, 40, 2099.
[5] a) P. T. Barnard, S. J. Berners-Price, Coord. Chem. Rev. 2007, 251, 1889;
b) A. Casini, C. Hartinger, C. Gabbiani, E. Mini, P. J. Dyson, B. K. Keppler,
L. Messori, J. Inorg. Biochem. 2008, 102, 564.
General procedure for the synthesis of gold(I) dipeptide
complexes 1–9 (Procedure C)
To a solution of the corresponding dipeptide derivative Boc-Cys-
aa-OMe (1 mmol) in acetone (10 mL) an excess of K2CO3 (2 mmol)
was added. The mixture was stirred for 5 min. Then, [AuCl(PPh3)] or
[AuCl(PPh2Py)] (1 mmol) in acetone (10 mL) was added and the re-
action mixture was stirred at room temperature for 24 h. After this
time, the reaction mixture was filtered over Celite and the clear so-
lution was evaporated under reduced pressure. Complexes 1–9
were purified by column chromatography of silica gel using as
eluent a mixture of AcOEt/hexane (1:1).
[6] L. Dalla Via, C. Nardon, D. Fregona, Future Med. Chem. 2012, 4, 525.
[7] M. Stallings-Mann, L. Jamieson, R. P. Regala, C. Weens, N. R. Murray, A. P.
Fields, Cancer Res. 2006, 66, 1767.
[8] a) T. M. Simon, D. H. Kunishima, G. J. Vibert, A. Lorber, Cancer Res. 1981,
41, 94; b) I. Ott, X. Quian, Y. Xu, D. H. V. Vlecker, I. J. Marques, D. Kubutat,
J. Will, W. S. Sheldrick, P. Jesse, A. Prokop, C. P. Bagowski, J. Med. Chem.
2009, 52, 763; c) E. Vergara, A. Casini, F. Sorrentino, O. Zava, E. Cerrada,
M. P. Rigobello, A. Bindolli, M. Laguna, P. J. Dyson, ChemMedChem 2010,
5, 96; d) D. T. Hill, A. A. Isab, D. E. Griswold, M. J. DiMartino, E. D. Matz,
A. L. Figueroa, J. E. Wawro, C. DeBrosse, W. M. Reiff, R. C. Elder, B. Jones,
J. W. Webb, C. F. Shaw III, Inorg. Chem. 2010, 49, 7663.
[9] a) S. J. Berners-Price, G. R. Girard, D. T. Hill, B. M. Sutton, P. S. Jarrett, L. F.
Faucette, R. K. Johnson, C. K. Mirabelli, P. J. Sadler, J. Med. Chem. 1990,
33, 1386; b) C. Santini, M. Pellei, G. Papini, B. Morresi, R. Galassi, S. Ricci,
F. Tisato, M. Porchia, M. P. Rigobello, V. Gandin, C. Marzano, J. Inorg. Bio-
chem. 2011, 105, 232; c) J. Fernµndez-Gallardo, B. T. Ellie, F. J. Sulzmaier,
M. Sanau, J. W. Ramos, M. Contel, Organometallics 2014, 33, 6669.
[10] a) M. M. Jellicoe, S. J. Nichols, B. A. Callus, M. V. Baker, P. J. Barnard, S. J.
Berners-Price, J. Whelan, G. C. Yeoh, A. Filipovska, Carcinogenesis 2008,
29, 1124; b) E. Schuh, C. Pflüger, A. Citta, A. Folda, M. P. Rigobello, A.
Bindoli, A. Casini, F. Mohr, J. Med. Chem. 2012, 55, 5518; c) L. Oehninger,
R. Rubbiani, I. Ott, Dalton Trans. 2013, 42, 3269; d) W. Liu, R. Gust,
Chem. Soc. Rev. 2013, 42, 755; e) R. Rubbiani, I. Kitanovic, H. Alborzinia,
S. Can, A. Kitanovic, L. A. Onambele, M. Stefanopoulou, Y. Geldmacher,
W. S. Sheldrick, G. Wolber, A. Prokop, S. Wçlfl, I. Ott, J. Med. Chem. 2010,
53, 8608; f) A. Citta, E. Schuh, F. Mohr, A. Folda, M. L. Massimino, A. Bind-
oli, A. Casini, M. P. Rigobello, Metallomics 2013, 5, 1006; g) F. Hacken-
berg, H. Muller-Bunz, R. Smith, W. Streciwilk, X. Zhu, M. Tacke, Organo-
metallics 2013, 32, 5551.
[11] a) G. Marcon, S. Carotti, M. Coronnello, L. Messori, E. Mini, P. Orioli, T.
Mazzei, M. A. Cinellu, G. Minguetti, J. Med. Chem. 2002, 45, 1672;
b) C. M. Che, R. W. Y. Sun, C. B. Ko, N. Zhu, H. Sun, Chem. Commun.
2003, 1718; c) L. Ronconi, C. Marzano, P. Zanello, M. Corsini, C. Macca,
A. Trevisan, D. Fregona, J. Med. Chem. 2006, 49, 1648; d) M. Coronnello,
E. Mini, B. Caciagli, M. A. Cinellu, A. Bindoli, C. Gabbiani, L. Messori, J.
Med. Chem. 2005, 48, 6761; e) J.-J. Zhang, W. Lu, R. W.-Y. Sun, C.-M. Che,
Angew. Chem. Int. Ed. 2012, 51, 4882; Angew. Chem. 2012, 124, 4966.
[12] a) N. Sewald, H. Jakubke, Peptides: Chemistry and Biology, Wiley-VCH,
Weinheim, 2002; b) N. L. Benoiton, Chemistry of Peptide Synthesis, CRC,
Boca Raton, 2006.
[13] a) D. H. Brown, G. C. McKinley, W. E. Smith, J. Chem. Soc. Dalton Trans.
1978, 199; b) A. A. Isab, P. J. Sadler, J. Chem. Soc. Dalton Trans. 1982,
135; c) G. Lewis, C. F. Shaw III, Inorg. Chem. 1986, 25, 58; d) R. C. Elder,
W. B. Jones, R. Floyd, Z. Zhao, J. G. Dorsey, K. Tepperman, Met.-Based
Drugs 1994, 1, 363; e) J. Lenke, A. Pinto, P. Nichoff, V. Vasylyeva, N. Met-
zler-Nolte, Dalton Trans. 2009, 7063; f) Y. Hashimoto, K. Tsuge, T. Konno,
Chem. Lett. 2010, 39, 601; g) T. Zhou, C. T. Lum, S. S.-Y. Chui, C.-M. Che,
Angew. Chem. Int. Ed. 2013, 52, 2930; Angew. Chem. 2013, 125, 3002.
[14] J. Caddy, U. Hoffmanns, N. Metzler-Nolte, Z. Naturforsch. Sect. B 2007,
62, 460.
Detailed synthetic procedures for the individual compounds can
be found in the Supporting Information.
Cell culture
Jurkat (leukaemia) and MiaPaca2 (pancreatic carcinoma) cell lines
were maintained in RPMI 1640, while A549 (lung carcinoma) were
grown in DMEM (Dulbecco’s Modified Eagle’s Medium). Both
media were supplemented with 5% foetal bovine serum (FBS),
200 UmLÀ1 penicillin, 100 mgmLÀ1 streptomycin and 2 mm l-gluta-
mine. Medium for A549 cells was also supplemented with 2.2 gLÀ1
Na2CO3, 100 mgmLÀ1 pyruvate and 5 mL non-essential amino acids
(Invitrogen). Cultures were maintained in a humidified atmosphere
of 95% air/5% CO2 at 378C. Adherent cells were allowed to attach
for 24 h prior to addition of compounds.
Cytotoxicity assay by MTT
The MTT assay was used to determine cell viability as an indicator
for cell sensitivity to the complexes. Exponentially growing cells
were seeded at a density of approximately 1”105 cellsmLÀ1 for the
adherent cell lines (A549, MiaPaca2) or 5”104 cellsmLÀ1 (Jurkat), in
a 96-well flat-bottomed microplate and 24 h later they were incu-
bated for 24 h with the compounds. The complexes were dissolved
in DMSO and tested in concentrations ranging from 0.1 to 25 mm
and in quadruplicate. Cells were incubated with our compounds
for 24 h at 378C. 10 mL of MTT (5 mgmLÀ1) was added and plates
were incubated for 1–3 h at 378C. Finally, 100 mL per well iPrOH
(0.05m HCl) was added. The optical density was measured at
490 nm using a 96-well multiscanner autoreader (ELISA). The IC50
was calculated by nonlinear regression analysis using Origin soft-
ware (Origin Software, Electronic Arts, Redwood City, California,
USA).
Acknowledgements
The authors thank the Ministerio de Economía y Competitivi-
dad (MINECO, FEDER CTQ2013-48635-C2-1-P, CTQ2013-40855-
R) and DGA-FSE (E40 and E77) for financial support. A.G.
thanks the CSIC for a Jae-Predoc fellowship.
[15] S. D. Kçster, H. Alborzinia, S. Can, I. Kitanovic, S. Wçlfl, R. Rubbiani, I.
Ott, P. Riesterer, A. Prokop, K. Merza, N. Metzler-Nolte, Chem. Sci. 2012,
3, 2062.
[16] S. Urig, K. Fritz-Wolf, R. RØau, C. Herold-Mende, K. Tóth, E. Davioud-
Charvet, K. Becker, Angew. Chem. Int. Ed. 2006, 45, 1881; Angew. Chem.
2006, 118, 1915.
Keywords: bioconjugation · cytotoxic activity · dipeptides ·
gold · medicinal chemistry
[17] J. Zou, P. Taylor, J. Dornan, S. P. Robinson, M. D. Walkinshaw, P. J. Sadler,
Angew. Chem. Int. Ed. 2000, 39, 2931; Angew. Chem. 2000, 112, 3054.
[18] a) A. GutiØrrez, J. Bernal, M. D. Villacampa, C. Cativiela, A. Laguna, M. C.
Gimeno, Inorg. Chem. 2013, 52, 6473; b) A. GutiØrrez, L. Gracia-Fleta, I.
Marzo, C. Cativiela, A. Laguna, M. C. Gimeno, Dalton Trans. 2014, 43,
17054.
[1] a) Metals in Medicine (Ed.: J. C. Dabrowiak), Wiley-VCH, Weinheim, 2009;
b) P. C. A. Bruijnincx, P. J. Sadler, Curr. Opin. Chem. Biol. 2008, 12, 197;
c) G. Gasser, N. Metzler-Nolte, Curr. Opin. Chem. Biol. 2012, 16, 84.
[2] Cisplatin: Chemistry and Biochemistry of a Leading Anticancer Drug (Ed.:
B. Lippert), Wiley-VCH, Weinheim, 1999.
Chem. Eur. J. 2015, 21, 11088 – 11095
www.chemeurj.org
11094
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[19] a) M. Wienken, B. Lippert, E. Zangrando, L. Randaccio, Inorg. Chem.
1992, 31, 1983; b) S. L. Best, T. K. Chattopadhyay, M. I. Djuran, R. A.
Palmer, P. J. Sadler, I. Sóvµgó, K. Vµrnagy, J. Chem. Soc. Dalton Trans.
1997, 2587; c) S. Carotti, G. Marcon, M. Marussich, T. Mazzei, L. Messori,
E. Mini, P. Orioli, Chem.-Biol. Interact. 2000, 125, 29; d) U. Rychlewska, B.
[23] M. J. Bowman, J. Chmielewski, Bioorg. Med. Chem. 2009, 17, 967.
[24] F. J. Sayago, M. I. Calaza, A. I. JimØnez, C. Cativiela, Tetrahedron 2009, 65,
5174.
[25] J. S. Modica-Napolitano, J. R. Aprille, Adv. Drug Delivery Rev. 2001, 49, 63.
[26] M. Frik, A. Martínez, B. T. Ellie, O. Gonzalo, D. Ramírez de Mingo, M.
Sanau, R. Sµnchez-Delgado, T. Sadhukha, S. Prabha, J. W. Ramos, I.
Marzo, M. Contel, J. Med. Chem. 2014, 57, 9995.
ˇ
´
ˇ ´
´
´
Warzajtis, B. “. Glisic, M. D. Zivkovic, S, Rajkovic, M. I. Djuran, Dalton
Trans. 2010, 39, 8906.
[20] M. N. Kouodom, L. Roncani, M. Colegato, C. Nardon, L. Marchiò, Q. P.
Dou, D. Aldinucci, F. Formaggio, D. Fregona, J. Med. Chem. 2012, 55,
2212.
[27] R. Unson, R. Laguna, Inorg. Synth. 1982, 21, 71
[21] A. F. A. Peacock, G. A. Bullen, L. A. Gethings, J. P. Williams, F. H. Kriel, J.
Coates, J. Inorg. Biochem. 2012, 117, 298.
Received: April 14, 2015
[22] X. Zhu, K. Pachamuthu, R. R. Schmidt, J. Org. Chem. 2003, 68, 5641.
Published online on June 25, 2015
Chem. Eur. J. 2015, 21, 11088 – 11095
www.chemeurj.org
11095
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim