CuII and AuIII Complexes with Glycoconjugated Dithiocarbamato Ligands for Potential Applications in Targeted Chemotherapy

Article abstract of DOI:10.1002/cmdc.201900226

This work is focused on the synthesis, characterization, and preliminary biological evaluation of bio-conjugated Au^III and Cu^II complexes with the aim of overcoming the well-known side effects of chemotherapy by improving the selective accumulation of an anticancer metal payload in malignant cells. For this purpose, carbohydrates were chosen as targeting agents, exploiting the Warburg effect that accounts for the overexpression of glucose-transporter proteins (in particular GLUTs) in the phospholipid bilayer of most neoplastic cells. We linked the dithiocarbamato moiety to the C1 position of three different monosaccharides: d-glucose, d-galactose, and d-mannose. Altogether, six complexes with a 1:2 metal-to-ligand stoichiometry were synthesized and in vitro tested as anticancer agents. One of them showed high cytotoxic activity toward the HCT116 colorectal human carcinoma cell line, paving the way to future in vivo studies aimed at evaluating the role of carbohydrates in the selective delivery of whole molecules into cancerous cells.

Full text of DOI:10.1002/cmdc.201900226

DOI: 10.1002/cmdc.201900226
Full Papers
CuII and AuIII Complexes with Glycoconjugated
Dithiocarbamato Ligands for Potential Applications in
Targeted Chemotherapy
Nicolꢀ Pettenuzzo,[a, b] Leonardo Brustolin,[a, b] Elisa Coltri,[a] Alberto Gambalunga,[c]
Federica Chiara,[c] Andrea Trevisan,[c] Barbara Biondi,[d] Chiara Nardon,[a] and
Dolores Fregona*[a]
This work is focused on the synthesis, characterization, and
preliminary biological evaluation of bio-conjugated AuIII and
CuII complexes with the aim of overcoming the well-known
side effects of chemotherapy by improving the selective accu-
mulation of an anticancer metal payload in malignant cells. For
this purpose, carbohydrates were chosen as targeting agents,
exploiting the Warburg effect that accounts for the overexpres-
sion of glucose-transporter proteins (in particular GLUTs) in the
phospholipid bilayer of most neoplastic cells. We linked the di-
thiocarbamato moiety to the C1 position of three different
monosaccharides: d-glucose, d-galactose, and d-mannose. Al-
together, six complexes with a 1:2 metal-to-ligand stoichiome-
try were synthesized and in vitro tested as anticancer agents.
One of them showed high cytotoxic activity toward the
HCT116 colorectal human carcinoma cell line, paving the way
to future in vivo studies aimed at evaluating the role of carbo-
hydrates in the selective delivery of whole molecules into can-
cerous cells.
Introduction
Cancer is the second-leading cause of death after cardiovascu-
lar diseases and was responsible for an estimated 9.6 million
deaths in 2018. Globally, about one in six deaths is due to
cancer.[1,2] Although several strategies have been developed to
fight cancer, killing neoplastic cells without affecting healthy
cells is still a challenge.
functionalized with monosaccharides to obtain selective deliv-
ery of the cytotoxic metal payload into the tumor tissue. Dis-
covered by Warburg in 1924, cancer cells require a larger
supply of nutrients, in particular carbohydrates, than healthy
cells, due to a series of metabolic mutations.[9–12] Most neoplas-
tic cells therefore overexpress glucose-transporter proteins, in
particular those belonging to the GLUT family, in their cell
membrane, and this phenomenon is correlated with poor pa-
tient prognosis.[13–15]
Within this lively field of research, a wide range of metal-
based compounds have been designed and tested in the last
decades to achieve the highest antiblastic activity along with a
good toxicological profile.[3] To enhance drug selectivity, re-
searchers exploit specific target proteins on the cell’s phospho-
lipid bilayer, which are overexpressed in neoplastic cells, by
conjugating the drug to the corresponding antibody or small
biomolecule.[4–8] In particular, this paper discloses the synthesis
and preliminary biological studies of coordination compounds
In recent years, this histopathological behavior was exploit-
ed by a few research groups, in particular those of Gao[16–20]
and Lippard,[21,22] who conjugated oxaliplatin to functionalized
monosaccharides at different positions of the pyranose ring
and tested the obtained PtII complexes for anticancer proper-
ties with promising results.
In this work, we combine some AuIII and CuII dithiocarba-
mates (DTCs) with various carbohydrates. The metal centers
AuIII and CuII were chosen owing to their well-known low-mi-
cromolar cytotoxic effects, when complexed with DTCs[23,24] or
other appropriate ligands,[25,26] against a wide panel of human
tumor cell lines. In this regard, different mechanisms of action
have been recognized in the literature over the last years: for
example, while reactivity toward the selenoenzyme thioredoxin
reductase, aquaporins, proteasome, or PARP-1 has been detect-
ed for AuIII compounds,[27–31] induced oxidative DNA or mito-
chondrial damage has been observed for CuII counterparts.[26,32]
DTCs are bidentate anions forming very stable coordination
compounds.[33] The strong chelating effect can prevent ligand
substitutions, thus favoring the application of this class of
complexes for biomedical purposes.
[a] Dr. N. Pettenuzzo, Dr. L. Brustolin, E. Coltri, Dr. C. Nardon, Prof. D. Fregona
Department of Chemical Sciences (DISC), University of Padova, Via Marzolo
1, 35131, Padova (Italy)
E-mail: dolores.fregona@unipd.it
[b] Dr. N. Pettenuzzo, Dr. L. Brustolin
Department of Surgical, Oncological and Gastroenterological Sciences
(DISCOG), University of Padova, Via Giustiniani 2, 35128, Padova (Italy)
[c] Dr. A. Gambalunga, Dr. F. Chiara, Prof. A. Trevisan
Department of Cardio-Thoraco-Vascular Sciences and Public Health (DCTV),
University of Padova, Via Giustiniani 2, 35128, Padova (Italy)
[d] Dr. B. Biondi
Institute of Biomolecular Chemistry, Padova Unit, CNR, Via Marzolo 1,
35131, Padova (Italy)
Supporting information for this article can be found under:
https://doi.org/10.1002/cmdc.201900226.
ChemMedChem 2019, 14, 1 – 12
1
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Concerning the nature of the selected monosaccharide (d-
glucose, d-galactose, and d-mannose), d-glucose is the most
common in the body and the last two diastereomers show a
similar affinity for GLUT proteins.[34–37] Three different DTC li-
gands were successfully complexed to the AuIII or CuII metal
centers. It is worth highlighting that the best method to conju-
gate the monosaccharide to the dithiocarbamato moiety
turned out the glycosylation reaction, discovered by Fisher and
optimized during the last century.[38–40]
for d-glucose and d-galactose drives the selective formation of
b-anomeric products.[41] Conversely, in the case of d-mannose,
an a-mannoside is obtained because of the axial position of
the acetyl group at C2.[41] Although the reaction is rapid and
highly stereoselective, many byproducts are formed in solution
due to the reactivity of the carbenium ion, which can interact
with other nucleophiles (e.g., water and acetate anion). A very
low yield, ranging from 10 to 28%, was recorded in all cases
after the glycosylation step. Conversely, the deacetylation and
the hydrogenation processes (second and third synthetic
steps, respectively) provide the unprotected glycosides almost
quantitatively.[41,45] During the process optimization we ob-
served that if column purification is performed only after the
second step instead of immediately after the glycosylation re-
action, higher yields can be obtained.
Results and Discussion
Synthesis and characterization of the ligands
The first step of the synthesis of the dithiocarbamato ligands
conjugated with d-glucose, d-galactose and d-mannose
(Figure 1) is the glycosylation reaction between the corre-
During the dithiocarbamate synthesis (Scheme 1 (iv),
Figure 1), we observed that the nucleophilic attack of the gly-
cosyl-conjugated secondary amine toward CS2 in water is
slower (6 h) than the same process involving less sterically hin-
dered alkyl amines as starting material (usually 1–2 h, for ex-
ample, piperidine).[24,33]
All the NMR spectra of the ligands were acquired in
[D6]DMSO and are provided in the Supporting Information.
This aprotic solvent allows the detection of resonances related
to the alcoholic functions of the monosaccharides. Moreover,
spectra recorded in [D6]DMSO show better solved peaks in the
2.9–4.9 ppm range. In contrast, acquisitions in D2O or MeOD
are characterized by closer peaks, thus hampering a proper in-
tegration process.
The 1H NMR spectra of the glycosyl-functionalized ligands
are characterized by the presence of several multiplets. Nota-
bly, 11 of the 18 total protons (i.e., the methynic protons of
the pyranose ring, the methylenic protons of the ethylic
spacer, and the methylenic protons at C6 as shown in Figure 1)
are diastereotopic, and each of them gives rise to a multiplet
in different regions of the spectrum. In this regard, a one-di-
1
mensional H NMR investigation does not provide sufficient in-
formation to unambiguously assign every resonance to the
corresponding proton set.
Figure 2 (top) shows the 2D 1H–1H COSY spectrum of the
ligand GluDTC, useful to better attribute each proton within
the structure (the corresponding 1D spectrum is shown in
Figure 2, at bottom, and the resonances are listed in Table 1).
The 2D spectra recorded for the d-galactoside and d-manno-
side derivatives are reported in the Supporting Information.
Figure 1. Glycoconjugated dithiocarbamato ligands synthesized in this work.
sponding pentaacetylated glycosyl donor and the carboxyben-
zyl-protected 2-N-methylaminoethanol (Cbz-2-N-methylami-
noethanol) as glycosyl acceptor in dry CH2Cl2 (Scheme 1).[41]
This step undergoes the activating process promoted by the
Lewis acid BF3·Et2O.[42] In particular, the role of the Lewis acid is
to favor the leaving of the acetyl group at the anomeric
carbon atom with the subsequent formation of a carbenium
ion.[43,44] Contrary to Fischer’s glycosylation, in which unpro-
tected carbohydrates are used, in the case of acetylated sugars
the carbonyl oxygen atom of the acetyl group at the C2 posi-
tion can stabilize the carbenium intermediate. When this stabi-
lizing group is present, the reaction becomes highly stereose-
lective thanks to the assisted stabilization of the adjacent
acetyl group that prevents axial nucleophilic attack of the gly-
cosyl acceptor. In particular, the equatorial substituent at C2
1
As shown in the 1D H NMR spectrum of the GluDTC ligand
(Figure 2 bottom), the groups of protons in a position with re-
spect to the dithiocarbamato moiety resonate at 3.40 and at
about 4.20 ppm for the N-methyl and the N-methylene pro-
tons, respectively. These values are in agreement with data re-
ported for other dithiocarbamates bearing similar substituents
surrounding the dithiocarbamic nitrogen atom (for example,
the sarcosine-DTC ligand).[46]
The protons of the hydroxy groups of the pyranose ring
couple with the adjacent methynic protons and give rise, in
the case of GluDTC, to three doublets (protons 7, 8, 9) and a
triplet (proton 10), at 4.95, 4.87, 4.83, and 4.50 ppm, respective-
&
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
2
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Scheme 1. Reagents and conditions: (i) BF3·Et2O, dry CH2Cl2, 08C!RT, 16 h; (ii) NaOMe, MeOH, RT, 16 h, column chromatography; (iii) H2/Pd, MeOH/EtOAc, RT,
2 h; (iv) CS2, KOH, H2O, 08C, 6 h, lyophilization; (v) [AuIII(PPh3)Cl3], CH2Cl2/MeOH/H2O (10:10:1 v/v), RT (byproducts: [AuI(PPh3)Cl], PPh3=O, KCl, HCl); (vi)
CuCl2·2H2O, acetone/MeOH (1:1 v/v), RT (byproduct: KCl).
ly (Table 1). Generally, the resonances of the equatorial protons
in the pyranose ring are shifted to lower fields relative to the
axial protons.[47,48]
the hydroxylic protons resonate at higher chemical shifts in
the case of GluDTC, in which all the hydroxy groups are in an
equatorial configuration.
In particular, all methynic protons in the b-d-glucoside
ligand experience the most shielding effect relative to the axial
counterparts of the b-d-galactoside and a-d-mannoside deriva-
The FTIR spectra of all dithiocarbamato ligands were collect-
ed in the medium (4000–600 cmꢀ1) wavenumber domain and
are reported in the Supporting Information (Figures S3, S6, S9).
The diagnostic absorptions are listed in Table 2 and are com-
pared with those of the dimethyl dithiocarbamato molecule,
which, similarly to glycoside analogues, bears two linear alkyl
substituents in a position with respect to the NꢀCSS moiety.
Two main IR regions are taken into account when character-
izing DTC derivatives, both ligand salts and metal–DTC com-
plexes. First is the spectral region 1100–1550 cmꢀ1, in which
the n(NꢀCSS) band can be found and its position depends
strongly on the extent of double bond character of the NꢀC
bond.[24] In the case of a single-bond form, the band is located
1
tives (except for the H at C1). For example, the proton at the
position C2 of the former resonates at highest field (2.93 ppm)
with a difference of 0.66 ppm with respect to the equatorial
proton at the same position of the ManDTC (3.59 ppm). The
same is true for the axial proton at C2 of GalDTC relative to
the equatorial proton of ManDTC. Likewise, the equatorial C4
proton in GalDTC resonates at lower fields with respect to the
corresponding protons of the other derivatives.
The shielding effect occurring for the methynic protons is
accompanied by deshielding for the alcoholic protons. In fact,
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
3
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Table 2. Main IR absorption peaks of the dimethyl dithiocarbamate and
the three glycosylated ligands.
Vibrations[a]
IR [cmꢀ1
]
Dimethyl
DTC–Na+
GluDTC
GalDTC
ManDTC
n(NꢀCSS) (s)
1486
962
–
–
–
1481
953
1077–1033
898
2926
1424
1481
952
1077–1043
892
2931-2890
1422
1377
1480
955
1083–1055
882
2929
1413
nas(SꢀCꢀS) (m)
nas(CꢀO) (s)
n(OꢀCꢀO) (w)
n(CꢀH sp3) (m)
d(CꢀH sp3) (m)
–
1377
1377
1353
n(CꢀC) (m)
–
1286
1259
1210
3392
1287
1258
1210
3365
1280
1258
1210
3393
n(OH) (br)
–
[a] s: strong, m: medium, w: weak, and br: broad.
at 1250–1350 cmꢀ1, whereas it is found at higher wavenum-
bers (1640–1690 cmꢀ1) if the DTC ligand contains a basically
double C=N bond (thioureidic form).[24,49]
The NꢀCSS band can be recognized at about 1480 cmꢀ1 for
all the three synthesized ligands, denoting a pronounced
double bond character. As a comparison, the same vibrational
frequency for the dimethyl dithiocarbamate is at 1486 cmꢀ1
,
pointing out an excellent correlation between the simplest ali-
phatic ligand and the glycosyl-functionalized ligands.
The band associated with the asymmetric stretching na(CSS),
detectable in the region ranging from 800 to 1100 cmꢀ1, is
found at 952–955 cmꢀ1 for the ligands under investigation,
with excellent correlation with respect to the same absorption
1
Figure 2. Top: COSY 2D H–1H NMR ([D6]DMSO, 600 MHz, 258C) spectrum of
for the dimethyl dithiocarbamate at 962 cmꢀ1 [50–52]
.
GluDTC (O-ethyl, N-methyl, b-d-glucopyranosyl dithiocarbamato, potassium
salt), reported as an example. Bottom: stacked H NMR spectra of GluDTC
and AuGlu. On passing from the free ligand to the complex, the most char-
acteristic deshielding and shielding effects are highlighted with red arrows.
1
Other areas of the spectrum provide valuable information
regarding the carbohydrate moiety, in particular the “sugar
region” (1200–950 cmꢀ1) and the “anomeric region” (950–
750 cm-[1]).[53,54]
Table 1. 1H NMR resonances of the synthesized glycosylated DTC ligands
in [D6]DMSO at 258C.
The overlapping intense bands of CꢀO and CꢀC stretching
vibrations in glycosidic bonds and pyranosic ring predominate
in the former region as shown in Figures S3, S6, S9 and
Table 2. In contrast, the “anomeric region” contains weaker
bands of more diagnostic skeletal vibrations sensitive to the
nature of the substituent in the acetal group and the anomeric
structure. In particular, based on published data, O-glycosides
in b-anomeric form give rise to a weak band related to the Oꢀ
CꢀO stretching between 950 and 890 cmꢀ1, whereas a-anom-
ers present a similar absorption at lower wavenumbers (from
890 to 750 cmꢀ1).[54] These bands are visible in the FTIR spectra
of the synthesized ligands, confirming the stereoselectivity of
the glycosylation reaction that led to the formation of b-anom-
ers for GluDTC and GalDTC derivatives (n(OCO)=898 and
892 cmꢀ1), and a-stereoisomers in the case of ManDTC
(n(OCO)=882 cmꢀ1).[54,55] However, the assignment for this vi-
bration mode is not unambiguously determined in literature,
as bands for other types of vibrations (not clearly understood)
also occur in the same region (Table 3).
Protons[a]
d [ppm]
GluDTC
GalDTC
ManDTC
1 (CꢀH backbone)
2 (CꢀH backbone)
3 (CꢀH backbone)
4 (CꢀH backbone)
5 (CꢀH backbone)
6 (CꢀH backbone)
4.14 (d)
2.93 (m)
3.11 (m)
3.06 (m)
3.05 (m)
4.10 (d)
3.25 (m)
3.25 (m)
3.62 (m)
3.31 (m)
4.61 (s)
3.59 (m)
3.30 (m)
3.38 (m)
3.45 (m)
3.65, 3.44 (m)
3.46–3.52 (m)
3.63, 3.45 (m)
7 (C2ꢀOH)
8 (C3ꢀOH)
9 (C4ꢀOH)
10 (C6ꢀOH)
4.83 (d)
4.95 (d)
4.87 (d)
4.50 (t)
4.64 (m)
4.80 (m)
4.31 (d)
4.56 (t)
4.67 (d)
4.65 (d)
4.52 (d)
4.38 (t)
11 (OꢀCH2 spacer)
12 (NꢀCH2 spacer)
13 (NꢀCH3 spacer)
3.93, 3.67 (m)
4.23 (m)
3.90, 3.67 (m)
4.22 (t)
3.75, 3.63 (m)
4.21 (m)
3.40 (s)
3.40 (s)
3.40 (s)
[a] Numbered as shown in Figure 1.
&
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
4
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Examples of CuII dithiocarbamato compounds with a M/L
stoichiometry other than 1:2 have never been reported on the
basis of this synthetic route. However, unstable dimeric com-
plexes of the type [CuIICl(DTC)]2 with two chlorides as bridging
ligands can be obtained by redox reaction between a CuI pre-
cursor and thiuram disulfide in an aprotic solvent.[58] In this re-
search field, Vecchio and co-workers studied the copper(II) be-
havior in solution in the presence of various Cu/DTC stoichio-
metries.[59]
Table 3. Proton resonances of the glycosylated ligand in the AuIII [1:2]Cl
complexes in [D6]DMSO at 258C.
Protons[a]
d [ppm]
AuGlu
AuGal
AuMan
1 (CꢀH backbone)
2 (CꢀH backbone)
3 (CꢀH backbone)
4 (CꢀH backbone)
5 (CꢀH backbone)
6–7 (CꢀH backbone)
4.20 (t)
4.15 (d)
4.66 (s)
3.59 (m)
3.19–3.22 (m)
3.34–3.38 (m)
3.43–3.46 (m)
3.66, 3.43 (m)
2.94–2.98 (m)
3.09–3.17 (m)
3.00–3.06 (m)
3.09–3.17 (m)
3.68, 3.43 (m)
3.20–3.27 (m)
3.20–3.27 (m)
3.63 (m)
3.33–3.37 (m)
3.48–3.55 (m)
All reactions with glycosylated DTCs were carried out in
polar solvent by adding the ligands to CuCl2·2H2O (Scheme 1
(vi)). Due to high hydrophilicity, the purification of the copper(-
II) derivatives from reaction byproducts, such as KCl and the
unreacted ligand, was difficult to achieve. Thus, C18 column
chromatography followed the synthesis of the complexes.
8–9 (OꢀCH2 spacer)
10–11(NꢀCH2 spacer)
12 (NꢀCH3 spacer)
3.88–4.06 (m)
3.88–4.06 (m)
3.43 (s)
3.86–4.05 (m)
3.86–4.05 (m)
3.43 (s)
3.37–3.97 (m)
4.04, 3.75 (m)
3.43 (s)
13 (C6ꢀOH)
14 (C4ꢀOH)
15 (C3ꢀOH)
16 (C2ꢀOH)
4.54 (t)
5.06 (d)
5.09 (d)
5.01 (d)
4.58 (t)
4.40 (d)
4.90 (d)
4.76 (d)
4.52 (t)
4.80–4.82 (m)
4.65 (d)
Characterization of dithiocarbamato AuIII and CuII complexes
4.80–4.82 (m)
[a] Numbered as shown in Figure 1.
All the obtained compounds were fully characterized by
1
means of H NMR, FTIR, UV/Vis, ESI-MS and elemental analysis
1
(see Supporting Information for H NMR and FTIR spectra). The
1H NMR spectra of all the AuIII complexes (600 MHz, 298 K)
were acquired in [D6]DMSO (Supporting Information) compar-
ing their resonances with those of the corresponding ligands
(an example is shown in Figure 2, bottom). In particular, upon
coordination the protons are generally characterized by down-
field shifts caused by the electron-withdrawing effect of the
electronegative metal. Only the H12 protons of the spacer are
affected by a shielding effect of about 0.2 ppm.
Concerning the CuII derivatives, a general broadening of the
resonances is observed when performing NMR analyses. This
effect, associated with the paramagnetic nature of the metal
center (d9 electronic configuration), appears as hyperfine shift
(d) to NMR signals and the shortening of both nuclear longitu-
dinal (T1) and transverse (T2) relaxation times.[60] The shifts and
the signal broadening, observed in particular for the closest
protons of the ethylic spacer (H12 and H13), result in resonances
with low intensities overlapping with other peaks of the carbo-
hydrate backbone. Therefore, these spectra (Figure S15 shows
an example) proved not so useful for the characterization of
the complexes and other diagnostic methods were exploited
in this work.
Synthesis of bis-dithiocarbamato AuIII complexes
After the synthesis and characterization of the three DTC li-
gands, the next step was the complexation reaction with AuIII
centers. A total of three AuIII complexes were synthesized with
a 1:2 metal-to-ligand stoichiometric ratio (hereinafter referred
to as [1:2]Cl).
The novel class of dithiocarbamato ligands explored in this
study possess four hydroxy groups that can interact with the
metal center, obtaining several byproducts. As a consequence,
a new synthetic pathway, which permits synthesis of the
[1:2]Cl compounds in good purity, was designed and opti-
mized in this work (Scheme 1).[25,56]
The reaction starts from a CH2Cl2 solution containing the tri-
phenylphosphino trihalo AuIII derivative [AuIIIPPh3Cl3] as re-
agent, and the desired product is formed quantitatively upon
the addition of one equivalent of dithiocarbamato ligand, dis-
solved in a methanol/water mixture (10:1 v/v) (Scheme 1, (v),
Figure S1). From a mechanistic point of view, it is possible to
suppose that coordination of the first DTC ligand to the AuIII
sphere promotes the formation of an instable intermediate of
the type [AuIIICl(DTC)(PPh3)]+. On one hand, the chelation of a
second dithiocarbamate leads to the desired product as a chlo-
ride salt. On the other, release of the triphenylphosphine is fa-
vored by its participation in a redox equilibrium (oxidation to
PPh3O) paralleled with the reduction of a second equivalent of
the [AuIIICl3(PPh3)] reagent, forming the stable [AuICl(PPh3)]
complex.
Regarding the FTIR characterization (Table 4), the coordina-
tion of the metal center strongly affects the SSCꢀN stretching
frequency, which is upshifted by about 80 and 35 cmꢀ1 relative
to that of the free ligands in all AuIII- and CuII-based com-
pounds, respectively. For all three CuII–DTC complexes, a
strong band is observed at about 1515 cmꢀ1, pointing out a
weaker NꢀCSS bond than that recorded for the Au counter-
parts.[61,62] For all complexes the CSS stretching mode cannot
be clearly evaluated instead, due to the presence of strong
bands related to the backbone vibrations which overlap the
na(CꢀS) absorption usually occurring at about 940–
1060 cmꢀ1.[51,63,64]
Synthesis of bis-dithiocarbamato CuII complexes
The copper ion in the +2 oxidation state can coordinate two
DTC molecules, for instance, by starting from the CuCl2·2H2O
salt, thus letting the chlorides out and yielding a neutral tetra-
coordinated and more stable product with a 1:2 metal-to-DTC
ligand ratio.[51,57]
Interestingly, all the spectra of the synthesized complexes
show a diagnostic absorption at about 890 cmꢀ1, recognized
as n(OCO) vibration of the glycosidic linkage. In fact, as de-
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
5
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Table 4. Main IR absorption bands of the AuIII complexes and CuII glycosylated compounds.
Complex
IR [cmꢀ1
]
AuGlu
AuGal
AuMan
CuGlu
CuGal
CuMan
n(NꢀCSS)
nas(CꢀO)
1565
1076
897
554
486
386
–
1564
1072
891
528
480
386
–
1564
1054
880
528
481
384
–
1514
1077
897
565
291
–
1516
1072
898
542
267
–
1516
1057
884
533
293
–
n(OꢀCꢀO)
ns(CSS)
n(CS) + d(SCS)
na(MꢀS)
ns(MꢀS)
357
352
364
scribed previously, these bands can be exploited to distinguish
between the a and b anomers.[54]
Moving to far FTIR, for the AuIII compounds the absorption
in the range 370–440 cmꢀ1 is ascribed to the AuꢀS asymmetric
stretching associated with a bidentate symmetric coordina-
tion.[24,52]
The negative values found for AuGlu and CuGlu account for
the observed high water solubility, and the AuIII compounds
are more hydrophilic than the copper(II) counterparts. Intrigu-
ingly, a difference of logP (as absolute value) of 2.7 was detect-
ed between the CuGlu and its analogue with piperidine dithio-
carbamate as a ligand.
The ns(MꢀS) in the copper complexes is set at about
358 cmꢀ1, in agreement with published data.[65]
Stability of the complexes in aqueous media
For all the characterized compounds, the vibrational modes
related to CSS-ring deformation (combined n(CS) + d(SCS) vi-
brational modes) occur at about 480 cmꢀ1 and 280 cmꢀ1 for
the gold and copper derivatives, respectively.[52,65]
The new dithiocarbamato AuIII and CuII complexes were also
characterized by UV/Vis spectrophotometry (see section below
about stability in aqueous media and the Supporting Informa-
tion) as well as by ESI-MS and elemental analysis (see Experi-
mental Section) in order to confirm the expected stoichiometry.
To determine the stability of the complexes in a medium suita-
ble for biological tests, a sequence of UV/Vis spectra was col-
lected over 72 h after dissolving the synthesized AuIII and CuII
derivatives in saline solution (NaCl, 0.9% w/v).
The electronic spectra over time of AuGlu and CuGlu, as
model compounds, are shown in Figure 3. The spectrum of
AuGlu displays two intense bands at 270 nm and 312 nm relat-
!
ed to the p* p intra-ligand NCS and SCS transitions, similar
to 1:2 AuIII dithiocarbamato analogues reported in the litera-
ture.[52,56,70] Moreover, as chloride salt, it is stable over 72 h
under physiological conditions. The UV/Vis spectrum of CuGlu
shows different bands. The absorption at 266 nm is ascribable
LogP measurements
The lipophilicity of a molecule can be represented as the loga-
rithm of the n-octanol/water partition coefficient and often
strongly correlates with its pharmacological activity and toxici-
ty.[66] The biphasic solvent system n-octanol/water is commonly
accepted in the scientific community, as it mimics well the
water/phospholipid membrane interface.[67] The results, pre-
sented as the mean of at least three independent measure-
ments, are presented in Table 5. Considering the similarity of
the three glycosylated ligands, only the values for the O-gluco-
syl derivatives are reported herein as an example. For compa-
rative purposes, the logP values of the bis-(piperidine-DTC) CuII
complexes and d-glucose are also listed in Table 5.
!
to the overlap between the p* p absorption of the conjugat-
!
ed NCSS group and the intra-ligand p* p transition located
mainly in the CSS moiety.[71,72] Moreover, the broad band at
428 nm is related to the d–d transitions derived from the dis-
torted symmetry (D2h or C2v, e=11930 Lmolꢀ1 cmꢀ1), the theo-
retical explanation of which is reported elsewhere.[73]
To sum up, it is important to highlight that all the synthe-
sized complexes show optimal stability in aqueous solution
over 72 h, as no substantial variations of the main bands are
detectable for each of them (see Supporting Information Figur-
es S31–S34 for the UV/Vis spectra over time of the AuIII- and
CuII-galactosyl and mannosyl derivatives). However, an intracel-
lular dissociation of the complexes cannot be excluded.[74]
Table 5. LogP values of AuGlu and CuGlu determined at 258C by UV/Vis
measurements.[a]
In vitro cytotoxic activity evaluation
Complex
LogP
To identify the best compound for future studies, all synthe-
sized complexes and the reference drug cisplatin were tested
against the HCT116 human colorectal carcinoma cell line, and
the results are summarized in Table S1. The new water-soluble
AuIII complexes are inactive, with IC50 values higher than
50 mm. On the other hand, the CuGlu complex shows cytotoxic
activity at low micromolar concentrations, in contrast to the
CuGal and CuMan derivatives.
AuGlu
CuGlu
[Cu(pipeDTC)2][24]
cisplatin [PtII(NH3)2Cl2][68]
d-glucose[69]
ꢀ1.9ꢁ0.1
ꢀ1.2ꢁ0.1
+1.5ꢁ0.1
ꢀ2.4
ꢀ2.82ꢁ0.1
[a] Data are reported as the meanꢁSD; each experiment was repeated at
least three times. Data for the other reference compounds were obtained
from published sources, as indicated.
&
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
6
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Table 6. Cytotoxicity values of the CuGlu complex toward HCT116 cancer
cells, alone or in the presence of a GLUT1 or SGLT inhibitor, or a competi-
tor (d-glucose).
Complex
IC50 [mm][a]
DMEM[b]
Phlorizin[c]
EDG[d]
d-Glucose[e]
2.0ꢁ0.1
CuGlu
2.7ꢁ0.4
2.6ꢁ0.2
2.1ꢁ0.2
[a] Determined at 72 h, 8000 cells per well. Data are reported as the
meanꢁSD; each experiment was repeated at least three times, and each
concentration tested in at least three replicates. [b] CuGlu alone; in low-
glucose DMEM. [c] SGLT inhibitor, present at 1 mm. [d] GLUT1 inhibitor,
present at 10 mm. [e] Competitor, present at 50 mm.
The data highlight no remarkable change of the antiprolifer-
ative activity after the addition of the two inhibitors or the
competitor, in contrast to what was observed for similar metal-
lodrugs by Gao and co-workers, who studied a new class of ox-
aliplatin-based sugar monoconjugates.[16] The observed behav-
ior could be associated with the presence of two hydrophilic
carbohydrate units per anticancer Cu-based molecule.
Conclusions
Compared with the work of Vecchio and collaborators,[59]
wherein the authors studied CuII complexation by a series of
glucosylated DTC ligands in solution, in this study a new class
of dithiocarbamato complexes containing
a carbohydrate
moiety as cancer-targeting agent were disclosed. After the
design phase, all the compounds were synthesized, character-
ized and tested in vitro as potential metallodrugs. In particular,
the synthetic conditions were optimized for three different
monosaccharides, and the desired complexes were obtained in
good yield and purity. The compounds involve two sugar units
linked through bulky alkyl amines, but, from a biological point
of view, the first experiments highlighted no targeting effect of
the carbohydrate functionalization.
Figure 3. UV/Vis analysis of AuGlu (top, 30 mm) and CuGlu (bottom, 20 mm)
dissolved in saline solution at 258C over 72 h.
Scientific evidence suggests that HCT116 neoplastic cells
overexpress GLUT-1 in their phospholipid bilayer.[75] The col-
lected results indicate that the AuIII complexes, CuGal and
CuMan, are not good substrates for glucose transporter pro-
teins. Moreover, cell uptake through the phospholipid bilayer
via passive transport is unlikely due to their high hydrophilicity
and the ionic nature of the gold-based compounds.
Afterward, to check whether CuGlu ([CuII(DTC-b-d-glucose)2])
can selectively target one or more glucose transporters,
HCT116 cells were exposed for 72 h to two different GLUT in-
hibitors, being added together with CuGlu to the cell culture
medium in selected wells. The first is O-ethylidene-a-d-glucose
(EDG), used as a GLUT-1 inhibitor at a concentration of
10 mm,[22,76] and the second is phlorizin dihydrate, which irre-
versibly binds the active site of SGLT (sodium-dependent glu-
cose co-transporter) proteins at 1 mm.[16,77] Furthermore, to
evaluate the competitiveness of CuGlu (as potential GLUT sub-
strate) with other carbohydrates, d-glucose was selected as
competitor and added to the medium at a concentration of
50 mm in the presence of CuGlu. The results of the mitochon-
drial viability tests in the presence of the inhibitors and com-
petitor are listed in Table 6.
Among all, the CuGlu compound showed an interesting IC50
value toward the HCT116 human colorectal carcinoma cell line.
Remarkably, the copper complexes CuGal and CuMan, bearing
two diastereomers of d-glucose, do not show any cytotoxic
properties. The next step will be in vivo evaluations of the role
of the carbohydrate to further determine whether active trans-
port is involved in the mechanism of action of the CuGlu com-
plex.
Experimental Section
General experimental methods: All reagents and solvents for the
syntheses were purchased from commercial sources and used as
received. Elemental analyses were carried out at the microanalysis
laboratory (Chemistry Department, University of Padova) with a
Carlo Erba 1108 CHNS-O micro-analyzer. ESI-MS spectra were re-
corded on a Mariner Perspective Biosystem instrument, setting a
5 kV ionization potential and a 20 mLminꢀ1 flow rate. A mixture of
coumarin and 6-methyltryptophan was used as a standard in posi-
tive ionization mode. Samples were dissolved in methanol, and the
same solvent was used as eluent. ESI-MS spectra were processed
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
7
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
by Data Explorer software. Infrared spectra were recorded at room
temperature under N2 atmosphere, in solid KBr by a spectropho-
tometer Nicolet Nexus 5SXC instrument in the range 4000–
400 cmꢀ1 (32 scansions, 2 cmꢀ1 resolution) and in Nujol between
two polyethylene tablets by a spectrophotometer Nicolet Nexus
870 in the range 600–50 cmꢀ1 (200 scansions, 4 cmꢀ1 resolution).
1H NMR one-dimensional spectra were recorded in CDCl3,
[D4]MeOH, or [D6]DMSO at 298 K on a Bruker Avance DMX 600
spectrometer equipped with a TXI-5 mm x,y,z-field gradient probe-
head, with Topspin 1.3 software package, operating in Fourier
transform. Typical acquisition parameters (1H: 600.13 MHz):
128 transients (for bidimensional acquisition: 512 transients of
32 scans/block), spectral width 7.5 kHz, 2 k data points and a delay
time of 7.0 s. Chemical shifts were referred to solvent peak. Peak
multiplicity is described as follows: s (singlet), d (doublet), dd (dou-
blet of doublets), ddd (doublet of doublets of doublets), t (triplet)
and m (multiplet). For the synthesis of fully acetylated b-d-glucose
1 and b-d-galactose 1’, the same procedure as that reported by
Yamada and co-workers was carried out.[78]
yl)carbamate (1.2 mmol), BF3·Et2O (5.0 mmol) was added dropwise
under stirring in inert atmosphere. The clear yellow mixture was
maintained at 08C for 1 h and overnight at RT. Then, the solution
was firstly neutralized by the dropwise addition of saturated aque-
ous NaHCO3, transferred to a separatory funnel and then washed
with NaHCO3 (2ꢂ30 mL), HCl(aq) 1n (3ꢂ30 mL) and brine (3ꢂ
30 mL). The combined organic fractions were dried with Na2SO4, fil-
tered and the solvent removed by rotatory evaporation, obtaining
a yellow oil that was used in the next step without further purifica-
tion.
General procedure for deacetylation (4, 4’, 4’’): The syrupy resi-
due obtained in the previous step was dissolved in dry MeOH
(20 mL), transferred in a 50 mL round-bottom flask, and excess
NaOMe was added at RT. The solution was then kept under inert
atmosphere overnight. After completion of the reaction as moni-
tored by TLC, the mixture was neutralized with Amberlite IR-120
H+ resin, filtered, and dried in rotatory evaporator to obtain a
yellow oil. The crude product was then re-dissolved in EtOAc, the
solution was filtered, and the pure glycosylated product was puri-
fied by means of flash column chromatography on silica, obtaining
a clear oil (eluent gradient: EtOAc 100%, then EtOAc/MeOH 85:15).
Penta-O-acetyl d-mannopyranoside (1’’): Dry pyridine (50 mL) and
acetic anhydride (40 mL) were added to d-mannose (30 mmol) in a
100 mL round-bottom flask. The mixture was kept at RT for four
days. In the meantime, the carbohydrate slowly went into solution.
Then, the mixture was diluted with EtOAc (100 mL) and transferred
into a separatory funnel. At this point deionized water (50 mL) was
added, the organic fraction was separated and successively
washed three times with deionized water (50 mL), 1m HCl(aq) (3ꢂ
25 mL), saturated aqueous NaHCO3 (3ꢂ25 mL) and with an aque-
ous 5% w/v solution of CuSO4 (5ꢂ25 mL) to remove excess pyri-
dine from the organic layer. All the obtained organic fractions were
combined, dried, and the solvent was evaporated obtaining a clear
viscous oil. In one circumstance, white crystalline needles of the
pure product were obtained by triturating the syrup with diethyl
General procedure for Cbz cleavage: The deacetylated glycoside
was dissolved in EtOAc/MeOH mixture (2:1, 30 mL). Successively,
H2 was bubbled in the presence of Pd/C (10% w/w) for 2 h under
stirring in a 50 mL round-bottom flask equipped with a gas bub-
bler. The reaction was monitored by means of TLC using an etha-
nol solution of ninhydrin for the colorimetric detection of the
amine. After completion, the mixture was filtered over a pad of
Celite, and the solvent was evaporated to dryness, obtaining a
clear oil that solidified in a fluffy white solid when dried by
vacuum pump.
2-(N-methylamino)ethyl-b-d-glucopyranoside (5): Yield: 25%;
1H NMR (600 MHz, MeOD): d=2.36 (s, 3H, CH3), 2.84–2.86 (m, 2H,
CH2NH), 3.15 (dd, 1H, C2-H), 3.21–3.24 (m, 2H, C4-H, C5-H), 3.31
(dd, 1H, C3-H), 3.58–3.63 (m, 2H, C6-HA, CHAHBCH2NH), 3.81 (dd,
1H, C6-HB), 3.88 (ddd, 1H, CHAHBCH2NH), 4.22 ppm (d, 1H, C1-H);
elemental analysis calcd (%) for C9H19NO6 (MW=237.25 gmolꢀ1): C
45.56, H 8.07, N 5.90, found: C 44.77, H 8.70, N 5.55.
1
ether and cooling. Yield: 86%. H NMR (600 MHz, CDCl3, mixture of
anomers): signals of b-anomer: d=1.98 (s, 3H), 2.07 (s, 3H), 2.08 (s,
3H), 2.15 (s, 3H), 2.19 (s, 3H), 3.99–4.05 (m, 1H), 4.07 (dd, 1H), 4.26
(dd, 1H), 5.23 (dd, 1H), 5.31–5.34 (m, 2H), 6.06 ppm (d, 1H); signals
of a-anomer: d=1.98 (s, 3H), 2.03 (s, 3H), 2.07 (s, 3H), 2.15 (s, 3H),
2.16 (s, 3H), 3.79 (ddd, 1H), 4.11 (dd, 1H), 4.28 (dd, 1H), 5.11 (dd,
1H), 5.27 (t, 1H), 5.46 (dd, 1H), 5.84 ppm (d, 1H).
2-(N-methylamino)ethyl-b-d-galactopyranoside (5’): Yield: 28%;
1H NMR (600 MHz, MeOD): d=2.36 (s, 3H, CH3), 2.80 (ddd, 1H,
CH2NH), 2.84 (ddd, 1H, CH2NH), 3.45 (dd, 1H, C3-H), 3.49 (ddd, 1H,
C5-H), 3.52 (dd, 1H, C2-H), 3.61 (ddd, 1H, CHAHBCH2NH), 3.69 (dd,
1H, C6-HA), 3.73 (dd, 1H, C6-HB), 3.80 (dd, 1H, C4-H), 3.91 (ddd, 1H,
CHAHBCH2NH), 4.21 ppm (d, 1H, C1-H); elemental analysis calcd (%)
for C9H19NO6 (MW=237.25 gmolꢀ1): C 45.56, H 8.07, N 5.90, found:
C 45.30, H 8.62, N 5.81.
Benzyl (2-hydroxyethyl-N-methyl)carbamate (2): To an ice-cold
freshly distilled dichloromethane solution (100 mL) containing 2-(N-
methylamino)ethanol (1 mmol) was added, portion-wise and under
stirring, N-benzyloxy carbonyloxysuccinimide (Cbz-OSu, 2 mmol)
and triethylamine (2 mmol). Two hours later, the mixture was
warmed at room temperature and left under vigorous stirring for
72 h. After the completion of the reaction (monitored with TLC by
means of ninhydrin colorimetric detection), the solution was dilut-
ed with deionized water (50 mL) and transferred to a separatory
funnel. The organic fraction was separated and successively
washed with HCl(aq) 1n (3ꢂ25 mL), NaHCO3 (sat) (3ꢂ25 mL) and
brine (2ꢂ25 mL). The combined organic fractions were dried with
Na2SO4, filtered and the solvent evaporated, obtaining a crude mix-
ture in the form of viscous oil. The product was then re-dissolved
in EtOAc and purified by means of flash column chromatography
on silica gel (eluent: EtOAc/ligroin, 7:3 v/v, Rf: 0.26). The pure prod-
uct was isolated as a clear oil. Yield: 85%. 1H NMR (CDCl3,
600 MHz): d=2.99 (s, 3H, CH3), 3.43 (m, 2H, CH2OH), 3.54 (br, 1H,
OH), 3.72 (m, 2H, CH2N), 5.13 (s, 2H, CH2Ph), 7.35 ppm (m, 5H, Ph).
2-(N-methylamino)ethyl-a-d-mannopyranoside (5’’): Yield: 22%;
1H NMR (600 MHz, MeOD): d=2.36 (s, 3H, CH3), 2.82–2.86 (m, 2H,
CH2NH), 3.48 (ddd, 1H, CHAHBCH2NH), 3.56 (ddd, 1H, C5-H), 3.63 (t,
1H, C4-H), 3.73 (dd, 1H, C6-HA), 3.74 (dd, 1H, C3-H), 3.79 (ddd, 1H,
CHAHBCH2NH), 3.86 (dd, 1H, C2-H), 3.86 (dd, 1H, C6-HB), 4.80 ppm
(d, 1H, C1-H); elemental analysis calcd (%) for C9H19NO6 (MW=
237.25 gmolꢀ1): C 45.56, H 8.07, N 5.90, found: C 45.20, H 8.25, N
5.34.
General procedure for the synthesis of the glycosylated dithio-
carbamato salt: To a cold aqueous solution (10 mL) containing the
free amine (0.5 mmol), was added KOH (1 equiv), followed by
excess CS2. After 6 h of stirring, Ar was bubbled for a few minutes
to eliminate the excess carbon disulfide still present. Then, the so-
lution was frozen in an CO2(s)/acetone bath and lyophilized, leading
General procedure for O-glycosylation (3, 3’, 3’’): To a cold, fresh-
ly distilled dichloromethane solution (70 mL) containing fully acety-
lated carbohydrate (1.0 mmol) and benzyl (2-hydroxyethyl-N-meth-
&
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
8
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
to the formation of a yellow viscous oil. The oil was re-dissolved in
anhydrous methanol and precipitated with diethyl ether to obtain
a white flocculent solid that was collected and dried. The product, as
potassium salt, is not hygroscopic and stable for several days at RT.
Bis(methyl-(2-(b-d-glucopyranosyl)ethyl)dithiocarbamato)gol-
d(III) chloride (AuGlu): Yield: 75%; mp: 608C; H NMR ([D6]DMSO,
1
600 MHz): d=4.20 (d, 1H), 2.94–2.98 (m, 1H), 3.09–3.17 (m, 1H),
3.00–3.06 (m, 1H), 3.09–3.17 (m, 1H), 3.68 (m, 1H), 3.43 (m, 1H),
3.88–4.06 (m, 2H), 3.88–4.06 (m, 2H), 3.43 (s, 3H), 4.54 (t, 1H), 5.06
(d, 1H), 5.09 (d, 1H), 5.01 ppm (d, 1H); medium FTIR (KBr): v˜ =1565
(na, NꢀCSS), 1076 (na, CO), 987 (na, CSS), 897 cmꢀ1 (n, OCO), far FTIR
(Nujol): v˜ =554 (ns, CSS), 486 (na, Au-S), 386 cmꢀ1 (ns, Au-S); ESI-MS
m/z: [MꢀCl]+ 821.09 (821.08), [5+H]+ 238.14 (238.13); elemental
analysis calcd (%) for C20H36AuClN2O12S4 (MW=857.19 gmolꢀ1): C
28.02, H 4.23, N 3.37, S 14.96, found: C 28.16, H 4.31, N 3.22, S
14.90.
Potassium methyl-(2-(b-d-glucopyranosyl)ethyl) dithiocarbamate
(GluDTC): Yield: 65%; 1H NMR ([D6]DMSO, 600 MHz): d=2.93 (m,
1H, C2-H), 3.05 (m, 2H, C4-H, C5-H), 3.11 (m, 1H, C3-H), 3.40 (s, 3H,
N-CH3), 3.44, 3.65 (m, 2H, C6-HA-B), 3.67, 3.93 (m, 2H, O-CH2), 4.14
(d, 1H, C1-H), 4.20–4.26 (m, 2H, N-CH2), 4.50 (t, 1H, C6-OH), 4.87 (d,
1H, C4-OH), 4.83 (d, 1H, C2-OH), 4.95 ppm (d, 1H, C3-OH); medium
FTIR (KBr): v˜ =2926 (na, CꢀH), 1481 (na, NꢀCSS), 953 (na, CSS),
898 cmꢀ1 (n, OCO); ESI-MS m/z: [2M]ꢀ 623.11 (623.12), [DTC-CS2]ꢀ
236.11 (236.11), [MꢀCH2OH]ꢀ 282.12 (282.05); elemental analysis
calcd (%) for C10H18KNO6S2 (MW=351.48 gmolꢀ1): C 34.17, H 5.16,
N 3.99, S 18.25, found: C 33.83, H 5.57, N 3.90, S 17.94.
Bis(methyl-(2-(b-d-galactopyranosyl)ethyl)dithiocarbamato)gol-
1
d(III) chloride (AuGal): Yield: 71%; mp: 608C; H NMR ([D6]DMSO,
600 MHz): d=4.15 (d, 1H), 3.20–3.27 (m, 1H), 3.20–3.27 (m, 1H),
3.63 (m, 1H), 3.33–3.37 (m, 1H), 3.48–3.55 (m, 2H), 3.86–4.05 (m,
2H), 3.86–4.05 (m, 2H), 3.43 (s, 3H), 4.58 (t, 1H), 4.40 (d, 1H), 4.90
(d, 1H), 4.76 ppm (d, 1H); medium FTIR (KBr): v˜ =1564 (na, NꢀCSS),
1072 (na, CO), 954 (na, CSS), 891 (n, OCO), far FTIR (Nujol): v˜ =528
(ns, CSS), 480 (na, Au-S), 386 cmꢀ1 (ns, Au-S); ESI-MS m/z: [MꢀCl]+
821.09 (821.08), [5’+H]+ 238.13 (238.13); elemental analysis calcd
(%) for C20H36AuClN2O12S4 (MW=857.19 gmolꢀ1): C 28.02, H 4.23, N
3.37, S 14.96, found: C 27.97, H 4.10, N 3.31, S 14.51.
Potassium
methyl-(2-(b-d-galactopyranosyl)ethyl)dithiocarba-
1
mate (GalDTC): Yield: 61%; H NMR ([D6]DMSO, 600 MHz): d=3.25
(m, 2H, C2-H, C3-H), 3.31 (m, 1H, C5-H), 3.40 (s, 3H, N-CH3), 3.49
(m, 2H, C6-HA-B), 3.62 (m, 1H, C4-H), 3.67, 3.90 (m, 2H, O-CH2), 4.10
(d, 1H, C1-H), 4.21–4.24 (m, 2H, N-CH2), 4.31 (d, 1H, C4-OH), 4.56 (t,
1H, C6-OH), 4.64 (d, 1H, C3-OH), 4.80 ppm (d, 1H, C2-OH); medium
FTIR (KBr): v˜ =2931–2890 (na, CꢀH), 1481 (na, NꢀCSS), 952 (na, CSS),
892 cmꢀ1 (n, OCO), ESI-MS m/z: [2M]ꢀ 623.11 (623.12), [DTC-CS2]ꢀ
236.11 (236.11), [MꢀCH2OH]ꢀ 282.12 (282.05); elemental analysis
calcd (%) for C10H18KNO6S2 (MW=351.48 gmolꢀ1): C 34.17, H 5.16,
N 3.99, S 18.25, found: C 32.04, H 6.55, N 2.98, S 18.55.
Bis(methyl-(2-(a-d-mannopyranosyl)ethyl)dithiocarbamato)gol-
d(III) chloride (AuMan): Yield: 70%; mp: >508C; 1H NMR
([D6]DMSO, 600 MHz): d=4.66 (d, 1H), 3.59 (m, 1H), 3.19–3.22 (m,
1H), 3.34–3.38 (m, 1H), 3.43–3.46 (m, 1H), 3.66 (m, 1H), 3.43 (m,
1H), 3.37–3.97 (m, 2H), 3.75 (m, 1H), 4.04 (m, 1H), 3.43 (s, 3H),
4.52 (t, 1H), 4.80–4.82 (d, 1H), 4.65 (d, 1H), 4.80–4.82 ppm (d, 1H);
medium FTIR (KBr): v˜ =1564 (na, NꢀCSS), 1054 (na, CO), 973 (na,
CSS), 880 cmꢀ1 (n, OCO), far FTIR (Nujol): v˜ =528 (ns, CSS), 481 (na,
Au-S), 384 cmꢀ1 (ns, Au-S); ESI-MS m/z, [MꢀCl]+ found (calc.):
821.09 (821.08); elemental analysis calcd (%) for C20H36AuClN2O12S4
(MW=857.19 gmolꢀ1): C 28.02, H 4.23, N 3.37, S 14.96, found: C
28.12, H 3.98, N 3.19, S 15.04.
Potassium
methyl-(2-(a-d-mannopyranosyl)ethyl)dithiocarba-
mate (ManDTC): Yield: 74%; 1H NMR ([D6]DMSO, 600 MHz): d=
3.30 (m, 1H, C3-H), 3.38 (m, 2H, C4-H), 3.40 (s, 3H, N-CH3), 3.45 (m,
1H, C5-H), 3.45, 3.63 (m, 2H, C6-HA-B), 3.63, 3.75 (m, 2H, O-CH2),
4.16–4.26 (m, 2H, N-CH2), 4.38 (t, 1H, C6-OH), 4.52 (d, 1H, C4-OH),
4.61 (d, 1H, C1-H), 4.65 (d, 1H, C3-OH), 4.67 ppm (d, 1H, C2-OH);
medium FTIR (KBr): v˜ =2929 (na, CꢀH), 1480 (na, NꢀCSS), 955 (na,
CSS), 882 cmꢀ1 (n, OCO); ESI-MS m/z: [2M]ꢀ 623.11 (623.12), [DTC-
CS2]ꢀ 236.11 (236.11); elemental analysis calcd (%) for C10H18KNO6S2
(MW=351.48 gmolꢀ1): C 34.17, H 5.16, N 3.99, S 18.25, found: C
33.69, H 5.40, N 3.97, S 18.52.
General procedure for the synthesis of CuII bis-dithiocarbamato
complexes: In a 50 mL round-bottom flask CuCl2·2H2O (0.25 mmol)
was dissolved in acetone (15 mL) under stirring. At this point, the
glycosylated dithiocarbamato ligand as potassium salt (0.5 mmol),
previously dissolved in dry MeOH (15 mL), were added dropwise to
the light-green acetone solution. The mixture initially turned
darker, due to the coordination between excess CuII ions with the
hydroxy groups of the ligand, and then brown, with a small
amount of a clear precipitate (KCl). The solution was filtered, the
solvent volume decreased by half by rotatory evaporation and a
brown flocculent solid was collected after precipitation with diethyl
ether. Then, the product was re-dissolved in Milli-Q H2O and puri-
fied by reversed-phase column chromatography on C18-functional-
ized silica (eluent H2O-CH3CN, with a gradient from 9:1 to 8:2 in
20 min with 0.05% TFA), obtaining a brown amorphous powder
after the lyophilization step. All the three products were soluble in
water, DMSO, MeOH and slightly in EtOH.
[Au(PPh3)Cl3] (6): [AuI(PPh3)Cl] (1 mmol), prepared as reported else-
where,[79] were dissolved in CH2Cl2 (15 mL) in a 50 mL two-necked
round-bottom flask, equipped with a rubber septum. At this point,
excess chlorine gas was bubbled through a rubber cannula. The
mixture, initially colorless, gradually turned bright green–yellow,
denoting the formation of related AuIII complex [AuIII(PPh3)X3] in so-
lution.[80] After monitoring the completion of the reaction by TLC,
the chlorine flow was then stopped, and Ar was bubbled through
for 10 min to remove excess halogen. The complex was not isolat-
ed and was used directly within a few minutes.
General procedure for the synthesis of AuIII bis-dithiocarbamato
chloride complexes: To a 50 mL round-bottom flask containing a
dichloromethane solution of [AuIII(PPh3)X3] (0.1m, 10 mL) was
added the glycoconjugated DTC ligand (1 mmol), previously dis-
solved in a MeOH/H2O mixture (10:1 v/v, 11 mL), at room tempera-
ture. The immediately formed yellow solid was filtered, washed
with dichloromethane (5ꢂ10 mL) and dried in vacuum pump in
the presence of P4O10. Then the product was transferred into a
25 mL round-bottom flask, re-dissolved in cold dry MeOH, to sepa-
rate the residual KCl by filtration, and precipitated with diethyl
ether, obtaining a yellow flocculent solid. This last step was repeat-
ed several times until no more KCl was observed at the bottom of
the flask after the re-dissolution of the complex in dry methanol.
All the three products were soluble in H2O, MeOH, and DMSO.
Bis(methyl-(2-(b-d-glucopyranosyl)ethyl)dithiocarbamato)cop-
per(II) (CuGlu): Yield: 61%; mp: 1568C (dec.); medium FTIR (KBr):
v˜ =1514 (na, NꢀCSS), 1077 (na, CO), 990 (na, CSS), 897 cmꢀ1 (n,
OCO), far FTIR (Nujol): v˜ =565 (ns, CSS), 357 (na, Cu-S), 291 cmꢀ1 (ns,
Cu-S); ESI-MS m/z: [M]+ 687.05 (687.04), [5+H]+ 238.14 (238.13);
elemental analysis calcd (%) for C20H36Cu2N2O12S4 (MW=
688.31 gmolꢀ1): C 34.90, H 5.27, N 4.07, S 18.63, found: C 34.08, H
5.25, N 3.64, S 18.42.
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
9
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Bis(methyl-(2-(b-d-galactopyranosyl)ethyl)dithiocarbamato) cop-
per(II) (CuGal): Yield: 80%; mp: 1478C (dec.); medium FTIR (KBr):
v˜ =1516 (na, NꢀCSS), 1072 (na, CO), 960 (na, CSS), 898 cmꢀ1 (n,
OCO), far FTIR (Nujol): v˜ =542 (ns, CSS), 360 (na, Cu-S), 290 cmꢀ1 (ns,
Cu-S); ESI-MS m/z: [M]+ 687.05 (687.04), [5’+H]+ 238.14 (238.13);
elemental analysis calcd (%) for C20H36Cu2N2O12S4 (MW=
688.31 gmolꢀ1): C 34.90, H 5.27, N 4.07, S 18.63. Found: C 34.50, H
5.21, N 3.22, S 18.48.
bation medium was removed after 72 h from the treated wells,
and 100 mL per well of a 10% resazurin solution in LG-DMEM
were added and incubated for 2 h at 378C. The control cells
were those treated with cell culture medium. The obtained cell
viabilities were plotted against the compound concentration to
determine the IC50 value (concentration of the test agent induc-
ing 50% decrease in cell number relative to control cell cul-
tures). The final IC50 values and their standard deviations were
evaluated from the data coming from at least three independ-
ent experiments. For comparison purposes, the cytotoxicity of
cisplatin (dissolved in saline solution) was evaluated under the
same experimental conditions. In this regard, it should be noted
that the experiments were performed over a period of 72 h, as
this is the time necessary for cisplatin to effect its activity. To
perform experiments with inhibitors, a procedure similar to
those reported by Gao and Lippard were followed. O-Ethyli-
dene-a-d-glucose was dissolved in 999 mL of cell culture
medium at a concentration 10 mm prior to addition of the com-
plex.[22] Conversely, phlorizin dihydrate, due to its low solubility,
was previously dissolved in DMSO at a concentration of 1m.
Successively, 1 mL of the solution was dissolved in 999 mL of cell
culture medium (LG-DMEM), to yield the following final concen-
trations of 1 mm of the inhibitors in each well.[16] This procedure
results in a DMSO concentration of 0.01% v/v in the growth
medium, which has no effect on cell viability.[81] The competitor
d-glucose was dissolved in culture medium at a concentration
of 50 mm prior to drug addition.
Bis(methyl-(2-(a-d-mannopyranosyl)ethyl)dithiocarbamato) cop-
per(II) (CuMan): Yield: 88%; mp: 1938C (dec.); medium FTIR (KBr):
v˜ =1516 (na, NꢀCSS), 1057 (na, CO), 960 (na, CSS), 884 cmꢀ1 (n,
OCO), far FTIR (Nujol): v˜ =533 (ns, CSS), 364 (na, Cu-S), 293 cmꢀ1 (ns,
Cu-S); ESI-MS m/z: [M]+ 687.04 (687.04), [5’’+H]+ 238.14 (238.13);
elemental analysis calcd (%) for C20H36Cu2N2O12S4 (MW=
688.31 gmolꢀ1): C 34.90, H 5.27, N 4.07, S 18.63, found: C 34.64, H
5.25, N 3.63, S 18.11.
LogP measurements: The measurements were carried out by
UV/Vis techniques recording the electronic spectra of the
aqueous solution of the desired compound before (C0) and
after mixing it with a pre-saturated octanol solution for 2 h (C1)
at RT. The evaluation of the corresponding n-octanol/water
partition coefficient was carried out exploiting the following
equation, using the absorbance value at the maximum absorp-
tion wavelength of the spectrum [Eq. (1)]:
Coctanol
Cwater
C0 ꢀ C1
ð1Þ
logP ¼ log
¼ log
C0
In vitro viability studies: HCT116 cells (American Type Cul-
ture Collection, ATCC) were cultured in 75 cm2 cell culture flasks
in low-glucose Dulbecco’s modified Eagle’s medium (LG-DMEM),
with addition of fetal bovine serum (FBS, 10%), l-glutamine
(5 mm), streptomycin (100 mgmLꢀ1), and penicillin (100 UmLꢀ1)
(Sigma–Aldrich), and incubated at 378C in a 5% CO2 controlled
atmosphere. For the cytotoxicity assay, the medium was re-
moved from the flask, and the cells washed with 6 mL of PBS,
and then shaken in the presence of 1 mL of trypsin (Sigma–Al-
drich), with 3 min incubation. LG-DMEM was successively added,
and the obtained cellular suspension seeded in 96-well micro-
plates (8ꢂ103 cells per well) in the growth medium (200 mL) and
incubated at 378C in a 5% CO2 atmosphere for 24 h to allow
cell adhesion, prior to drug testing. The ionic AuIII complexes of
the type [1:2]Cl and the CuII glycosylated derivatives, which are
all water soluble, were dissolved in saline solution (NaCl 0.9%
w/v). All compounds were tested against the HCT116 cells at
various micromolar concentrations to obtain dose–response
plots. Briefly, the metal derivatives were dissolved in saline at
concentrations of 50, 25, 10, 5, 2, 1, and 0.5 mm. Successively,
each solution (1 mL) was dissolved in 999 mL of cell culture
medium (LG-DMEM), to yield the following final concentrations
of metal complex: 50, 25, 10, 5, 2, 1, and 0.5 mm. After prepara-
tion of the 96-well plates, containing 8000 cells each, the
medium was removed and replaced with fresh medium contain-
ing the compounds to be studied at increasing concentrations.
Triplicate conditions were established for each treatment, and
six independent experiments were carried out for each com-
pound. Mitochondrial cell viability was evaluated by means of
resazurin assay. To perform the assay, the entire volume of incu-
Acknowledgements
This work was supported financially by T.R.N.–Imballaggi Logis-
tics Services (PhD fellowship to N.P., www.trnimballaggi.it/en/),
the University of Padua (postdoctoral fellowship to C.N., “Asseg-
no di Ricerca Senior” area scientifica di ateneo no. 03—Scienze
chimiche), and the L’Orꢀal Foundation–UNESCO (postdoctoral fel-
lowship to C.N., www.forwomeninscience.com/en/fellowships).
Conflict of interest
The authors declare no conflict of interest.
Keywords: carbohydrates · copper · gold · dithiocarbamates ·
metal complexes
[1] “Key Facts on Cancer”, World Health Organization (WHO), http://
www.who.int/news-room/fact-sheets/detail/cancer, September 12, 2018.
[2] “Cancer Tomorrow: A tool that predicts the future cancer incidence and
mortality burden worldwide from the current estimates in 2018 up
until 2040”, International Agency for Research on Cancer (IARC), http://
gco.iarc.fr/tomorrow/home, 2018.
[3] a) S. Monro, K. L. Colꢃn, H. Yin, J. Roque III, P. Konda, S. Gujar, R. P.
Thummel, L. Lilge, C. G. Cameron, S. A. McFarland, Chem. Rev. 2019, 119,
797–828; b) X. Wang, X. Wang, S. Jin, N. Muhammad, Z. Guo, Chem.
Rev. 2019, 119, 1138–1192.
[4] E. C. Calvaresi, P. J. Hergenrother, Chem. Sci. 2013, 4, 2319–2333.
[5] I. R. Vlahov, C. P. Leamon, Bioconjugate Chem. 2012, 23, 1357–1369.
˝
[6] E. I. Vrettos, G. Mezo, A. G. Tzakos, Beilstein J. Org. Chem. 2018, 14, 930–
954.
&
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
10
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
[7] W. X. Ren, J. Han, S. Uhm, Y. J. Jang, C. Kang, J.-H. Kim, J. S. Kim, Chem.
Commun. 2015, 51, 10403–10418.
[46] L. Ronconi, C. Maccato, D. Barreca, R. Saini, M. Zancato, D. Fregona,
Polyhedron 2005, 24, 521–531.
[47] A. Nickon, M. A. Castle, R. Harada, C. E. Berkoff, R. O. Williams, J. Am.
Chem. Soc. 1963, 85, 2185–2186.
[48] J. B. Lambert, J. E. Goldstein, J. Am. Chem. Soc. 1977, 99, 5689–5693.
[49] D. A. Brown, W. K. Glass, M. A. Burke, Spectrochim. Acta Part A 1976, 32,
145–147.
[8] S. Xing-Dong, K. Xia, H. Shu-Fen, C. Jia-Xi, S. Jing, C. Bing-Bing, Z. Jin-
Wu, M. Zong-Wan, Eur. J. Med. Chem. 2017, 138, 246–254.
[9] O. Warburg, K. Posener, E. Negelein, Biochem. Z. 1924, 152, 319–344.
[10] O. Warburg, Science 1956, 123, 309–314.
[11] J. Zheng, Oncol. Lett. 2012, 4, 1151–1157.
[50] K. J. Cavell, J. O. Hill, R. J. Magee, J. Inorg. Nucl. Chem. 1979, 41, 1277–
1280.
[12] R. J. DeBerardinis, N. S. Chandel, Sci. Adv. 2016, 2, e1600200.
[13] R. A. Medina, G. I. Owen, Biol. Res. 2002, 35, 9–26.
[14] Z. Tan, C. Yang, X. Zhang, P. Zheng, W. Shen, Oncotarget 2017, 8,
60954–60961.
[15] J. Yang, J. Wen, T. Tian, Z. Lu, Y. Wang, Z. Wang, X. Wang, Y. Yang, Onco-
target 2017, 8, 11788–11796.
[16] P. Liu, Y. Lu, X. Gao, R. Liu, D. Zhang-Negrerie, Y. Shi, Y. Wang, S. Wanga,
Q. Gao, Chem. Commun. 2013, 49, 2421–2423.
[17] H. Li, X. Gao, R. Liu, W. Yu, M. Zhang, Z. Fu, Y. Mi, Y. Wang, Z. Yao, Q.
Gao, Eur. J. Med. Chem. 2015, 101, 400–408.
[51] L. Giovagnini, S. Sitran, M. Montopoli, L. Caparrotta, M. Corsini, C. Rosani,
P. Zanello, Q. P. Dou, D. Fregona, Inorg. Chem. 2008, 47, 6336–6343.
[52] F. Forghieri, C. Preti, L. Tassi, G. Tosi, Polyhedron 1988, 7, 1231–1237.
[53] R. Varma, S. Y. Kulkarni, C. I. Jose, V. S. Pansare, Carbohydr. Res. 1984,
133, 25–32.
[54] R. S. Tipson, Infrared Spectroscopy of Carbohydrates: a Review of the Lit-
erature, National Bureau of Standards Monographs, 1966, p. 13–14.
[55] S. Synytsya, M. Novak, Ann. Transl. Med. 2014, 2, 17–31.
[56] P. T. Beurskens, J. A. Cras, J. G. M. Van der Linden, Inorg. Chem. 1970, 9,
475–479.
[18] Q. Mi, Y. Ma, X. Gao, R. Liu, P. Liu, Y. Mi, X. Fu, Q. Gao, J. Biomol. Struct.
Dyn. 2016, 34, 2339–2350.
[57] A. R. Hendrickson, R. L. Martin, N. M. Rohde, J. Am. Chem. Soc. 1976, 98,
2115–2119.
[58] A. R. Hendrickson, R. L. Martin, D. Taylor, J. Chem. Soc. Chem. Commun.
1975, 843–844.
[19] M. Wu, H. Li, R. Liu, X. Gao, M. Zhang, P. Liu, Z. Fu, J. Yang, D. Zhang-
Negrerie, Q. Gao, Eur. J. Med. Chem. 2016, 110, 32–42.
[20] X. Gao, S. Liu, Y. Shi, Z. Huang, Y. Mi, Q. Mi, J. Yang, Q. Gao, Eur. J. Med.
Chem. 2017, 125, 372–384.
[59] R. Turnaturi, V. Oliveri, M. Viale, M. Monticone, G. Vecchio, ChemPlu-
sChem 2015, 80, 1786–1792.
[21] M. Patra, S. G. Awuah, S. J. Lippard, J. Am. Chem. Soc. 2016, 138, 12541–
12551.
[60] N. N. Murthy, K. D. Karlin, I. Bertini, C. Luchinat, J. Am. Chem. Soc. 1997,
119, 2156–2162.
[61] K. Nakamoto, J. Fujita, R. A. Condrate, Y. Morimoto, J. Chem. Phys. 1963,
39, 423–427.
[62] C. N. R. Rao, R. Venkataraghavan, Spectrochim. Acta 1962, 18, 541–547.
[63] R. Kellner, G. S. Nikolov, N. Trendafilova, Inorg. Chim. Acta 1984, 84,
233–239.
[22] M. Patra, T. C. Johnstone, K. Suntharalingam, S. J. Lippard, Angew. Chem.
Int. Ed. 2016, 55, 2550–2554; Angew. Chem. 2016, 128, 2596–2600.
[23] C. Nardon, D. Fregona, Curr. Top. Med. Chem. 2016, 23, 360–380.
[24] L. Brustolin, C. Nardon, N. Pettenuzzo, N. Zuin Fantoni, S. Quarta, F.
Chiara, A. Gambalunga, A. Trevisan, L. Marchiꢀ, P. Pontisso, D. Fregona,
Dalton Trans. 2018, 47, 15477–15486.
ˇ
ˇ
ˇ
[25] M. Altaf, A. A. Isab, J. Vanco, Z. Dvorꢄk, Z. Trꢄvnꢅcek, H. Stoeckli-Evans,
[64] F. Bonati, R. Ugo, Organomet. Chem. 1967, 10, 257–268.
[65] R. Kellner, G. S. Nikolov, J. Inorg. Nucl. Chem. 1981, 43, 1183–1188.
[66] P. D. Leeson, B. Springthorpe, Nat. Rev. Drug Discovery 2007, 6, 881–890.
[67] F. Esteves, C. Moutinho, C. Matos, J. Liposome Res. 2013, 23, 83–93.
[68] T. Marzo, S. Pillozzi, O. Hrabina, J. Kasparkova, V. Brabec, A. Arcangeli, G.
Bartoli, M. Severi, A. Lunghi, F. Totti, C. Gabbiani, A. G. Quiroga, L. Mes-
sori, Dalton Trans. 2015, 44, 14896–14905.
[69] M. F. Mazzobre, M. V. Romꢇn, A. F. Mourelle, H. R. Corti, Carbohydr. Res.
2005, 340, 1207–1211.
[70] P. T. Beurskens, H. J. A. Blaauw, J. A. Cras, J. J. Steggerda, Inorg. Chem.
1968, 7, 805–810.
[71] C. K. Jørgensen, J. Inorg. Nucl. Chem. 1952, 24, 1571–1585.
[72] F. Takami, S. Wakahara, T. Maeda, Tetrahedron Lett. 1971, 12, 2645–2648.
[73] S. Choi, E. R. Menzel, J. R. Wasson, J. Inorg. Nucl. Chem. 1977, 39, 417–422.
[74] D. J. Berry, R. T. M. de Rosales, P. Charoenphun, P. J. Blower, Mini Rev.
Med. Chem. 2012, 12, 1174–1183.
RSC Adv. 2015, 5, 81599–81607.
[26] J. C. Garcꢅa-Ramos, A. G. Gutiꢆrrez, A. Vꢇzquez-Aguirre, Y. Toledano-
MagaÇa, A. L. Alonso-Sꢇenz, V. Gomez-Vidales, M. Flores-Alamo, C.
Mejꢈa, L. Ruiz-Azuara, BioMetals 2017, 30, 43–58.
[27] B. Bertrand, A. Citta, I. L. Franken, M. Picquet, A. Folda, V. Scalcon, M. P.
Rigobello, P. Le Gendre, A. Casini, E. Bodio, J. Biol. Inorg. Chem. 2015, 20,
1005–1020.
[28] M. N. Wenzel, A. F. Mꢃsca, V. Graziani, B. Aikman, S. R. Thomas, A. De Al-
meida, J. A. Platts, N. Re, C. Coletti, A. Marrone, G. Soveral, A. Casini,
Inorg. Chem. 2019, 58, 2140–2148.
[29] G. Boscutti, C. Nardon, L. Marchiꢀ, M. Crisma, B. Biondi, D. Dalzoppo, L.
Dalla Via, F. Formaggio, A. Casini, D. Fregona, ChemMedChem 2018, 13,
1131–1145.
[30] M. F. Tomasello, C. Nardon, V. Lanza, G. Di Natale, N. Pettenuzzo, S. Sal-
maso, D. Milardi, P. Caliceti, G. Pappalardo, D. Fregona, Eur. J. Med.
Chem. 2017, 138, 115–127.
[75] P. Lopez-Serra, M. Marcilla, A. Villanueva, A. Ramos-Fernandez, A. Palau,
L. Leal, J. E. Wahi, F. Setien-Baranda, K. Szczesna, C. Moutinho, A. Marti-
nez-Cardus, H. Heyn, J. Sandoval, S. Puertas, A. Vidal, X. Sanjuan, E. Mar-
tinez-Balibrea, F. Vinals, J. C. Perales, J. B. Bramsem, T. F. Ørntoft, C. L. An-
dersen, J. Tabernero, U. McDermott, M. B. Boxer, M. G. Vander Heiden, J.
Pablo Albar, M. Esteller, Nat. Commun. 2014, 5, 3608.
[76] S. C. Rumsey, O. Kwon, G. W. Xu, C. F. Burant, I. Simpson, M. Levine, J.
Biol. Chem. 1997, 272, 18982–18989.
[77] A. Malhotra, S. Kudyar, A. K. Gupta, R. P. Kudyar, P. Malhotra, Int. J. Appl.
Basic Med. Res. 2015, 5, 161–163.
[78] N. Michihata, Y. Kaneko, Y. Kasai, K. Tanigawa, T. Hirokane, S. Higasa, H.
Yamada, J. Org. Chem. 2013, 78, 4319–4328.
[79] N. C. Baenziger, W. E. Bennett, D. M. Soboroff, P. F. O’Donnell, J. R. Doyle,
Polyhedron 1998, 17, 2379–2410.
[80] A. Leyva, X. Zhang, A. Corma, Chem. Commun. 2009, 4947–4949.
[81] G. Da Violante, N. Zerrouk, I. Richard, G. Provot, J. C. Chaumeil, P.
Arnaud, Biol. Pharm. Bull. 2002, 25, 1600–1603.
[31] C. Nardon, N. Pettenuzzo, D. Fregona, Curr. Med. Chem. 2016, 23, 3374–
3403.
[32] Z. Molphy, D. Montagner, S. S. Bhat, C. Slator, C. Long, A. Erxleben, A.
Kellett, Nucleic Acid Res. 2018, 46, 9918–9931.
[33] G. Hogarth, Mini Rev. Med. Chem. 2012, 12, 1202–1215.
[34] D. Deng, P. Sun, C. Yan, M. Ke, X. Jiang, L. Xiong, W. Ren, K. Hirata, M.
Yamamoto, S. Fan, N. Yan, Nature 2015, 526, 391–396.
[35] D. Deng, N. Yan, Protein Sci. 2016, 25, 546–558.
[36] A. M. Navale, A. N. Paranjape, Biophys. Rev. Lett. 2016, 8, 5–9.
[37] M. Mueckler, B. Thorens, Mol. Aspects Med. 2013, 34, 121–138.
[38] E. Fischer, Ber. Dtsch. Chem. Ges. 1893, 26, 2400–2412.
[39] W. Koenigs, E. Knorr, Ber. Dtsch. Chem. Ges. 1901, 34, 957–981.
[40] B. Helferich, K. F. Wedemeyer, Liebigs Ann. 1949, 563, 139–145.
ˇ
[41] R. Sardzꢅk, G. T. Noble, M. J. Weissenborn, A. Martin, S. J. Webb, S. L.
Flitsch, Beilstein J. Org. Chem. 2010, 6, 699–703.
[42] G. Magnusson, G. Noori, J. Dahmen, T. Frejd, T. Lave, Acta Chem. Scand.
B. 1981, 35, 213–216.
[43] P. O. Adero, T. Furukawa, M. Huang, D. Mukherjee, P. Retailleau, L. Bohꢉ,
D. Crich, J. Am. Chem. Soc. 2015, 137, 10336–10345.
[44] L. Bohꢉ, D. Crich, Nat. Chem. 2016, 8, 99–100.
[45] B. Ren, M. Wang, J. Liu, J. Ge, X. Zhang, H. Dong, Green Chem. 2015, 17,
1390–1394.
Manuscript received: April 11, 2019
Revised manuscript received: April 29, 2019
Accepted manuscript online: && &&, 0000
Version of record online: && &&, 0000
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
11
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

FULL PAPERS
N. Pettenuzzo, L. Brustolin, E. Coltri,
A. Gambalunga, F. Chiara, A. Trevisan,
B. Biondi, C. Nardon, D. Fregona*
&& – &&
CuII and AuIII Complexes with
Glycoconjugated Dithiocarbamato
Ligands for Potential Applications in
Targeted Chemotherapy
Energy-hungry cells, beware: This
work is focused on the synthesis, char-
acterization, and preliminary anticancer
activity evaluation of carbohydrate-con-
jugated AuIII and CuII dithiocarbamato
complexes, with the aim of improving
the selective accumulation of the metal
payload into malignant cells by exploit-
ing the well-known Warburg effect.
&
ChemMedChem 2019, 14, 1 – 12
www.chemmedchem.org
12
ꢁ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!