.
Angewandte
Communications
0
species into Pt , but was capable of detecting cisplatin with
high specificity in a range of buffer conditions. Using
combinatorial library synthesis, Tae et al. also disclosed
II
a rhodamine–triazole probe that is capable of detecting Pt
species such as cisplatin through the ring opening of its
[
17]
spirolactam motif.
We reasoned that it would be possible to engineer
a fluorescent probe specifically designed for the detection of
II
Pt drugs in a cellular environment by taking advantage of the
IV
II
difference in reactivity between the Pt prodrug and its Pt
IV
congener. As Pt prodrug complexes are kinetically inert
because of their saturated coordination sphere and stable
electronic configuration, we could design functional groups
II
that would selectively interact with Pt species. Examples of
such functional groups include dithiocarbamates and dithio-
[
18]
carbamate esters. Indeed, in clinical trials, sodium dieth-
yldithiocarbamate (NaDDTC) was found to sequester excess
administered cisplatin, and to mitigate systemic toxicity
Figure 2. Molecular representation of Rho-DDTC; ellipsoids set at
50% probability.
[
19]
induced by the therapy.
Our strategy was to tether the
non-fluorescent spirolactam form of rhodamine B (RhoB) to
a recognition motif comprised of DDTC (Supporting Infor-
mation, Scheme S1). The recognition motif would bind Pt
II
2+
+
of Pt was high, as the use of other metal ions, such as Na ,
+
2+
2+
2+
2+
2+
3+
species through the dithiocarbonyl bond; thereby, the trans
K , Mg , Cu , Ni , Zn , Fe , and Fe , which play
important roles in biological systems, only led to baseline
values in the emission and absorption spectra (Figure S8).
Compared to Pt , the detection of cisplatin was slower under
similar reaction conditions and required at least 60 min before
a fluorescence signal could be measured. The selectivity for
cisplatin compared to other metal ions remained high, even at
elevated concentrations (Na, K, and Mg at ca. 15 mm), which
is essential for its application as an intracellular probe
(Figure S9).
To fully exploit Rho-DDTC as a probe for Pt drugs, we
sought to understand its underlying mechanism of action. We
postulated that aquation was essential for probe activation,
because cisplatin readily undergoes aquation to form reactive
aqua species. However, even after aquation was suppressed
with 1m NaCl, the probe could still be activated, which thus
suggests alternate pathways. Hence, we evaluated Rho-
DDTC against several cis-[PtA X ] scaffolds (A =
am(m)ine) to understand the role of the ligands in probe
activation (Table S2). Intriguingly, when chelating A ligands
were present, the Pt complex could not activate the probe.
To understand this observation, the reaction products of
K PtCl , cisplatin, carboplatin, and JM118 with Rho-DDTC
were analyzed by ESI-MS. The mass shifts indicated retention
of an ammine and a cyclohexylamine ligand on the Rho-
DDTC/cisplatin and Rho-DDTC/JM118 adducts, respec-
tively (Figure S10). JM118 exclusively formed adducts con-
taining the cyclohexylamine ligand (m/z 1057), whereas
adducts with the alternate ammine ligand (m/z 976) were
absent. Furthermore, the same adducts were formed from
cisplatin and carboplatin, which indicates that CBDCA was
displaced.
ligands are labilized for reactions with the spirolactam motif,
[
20]
and the fluorescence is selectively turned on.
2
+
The target probe Rho-DDTC was prepared in three steps
(
Scheme S1). RhoB was treated sequentially with POCl and
3
NH OH to obtain hydroxamate derivative 1. Reaction with
2
excess 1,2-dibromoethane in the presence of triethylamine as
a base yielded 2, which was quantitatively converted into
Rho-DDTC by a reaction with NaDDTC. The C2 spacer
between the Rho and DDTC motifs was employed to
facilitate positioning of bound Pt at the spirolactam upon
1
DDTC-binding. Rho-DDTC was fully characterized by H
1
3
and C NMR, ESI-MS, and single-crystal X-ray crystallo-
graphic analyses (see the Supporting Information). The spiro
form was confirmed in solution by the presence of the
1
3
C resonance of the spiro C atom at ca. 65 ppm (CDCl and
3
[
D ]DMSO), as well as in the solid state (Figure 2, C8). The
6
crystal structure shows that the thiocarbonyl group is oriented
towards the spirolactam, presumably to minimize steric
2
2
repulsion between the NEt substituent of the carbamate
2
II
and the xanthine ring, thus affording a suitable binding
configuration.
2
+
Detection of Pt using Rho-DDTC was investigated in
HEPES buffer (20 mm, pH 7.4, 30% EtOH). The probe was
stable over a large pH range (3.5–12) and hence compatible
with physiological conditions (Figure S4). When K PtCl was
2
4
2
4
2
+
added as the Pt source (80 mm), a visible change from
colorless to pink was observed after 5 min with a 60-fold
increase in absorbance at lmax = 565 nm, as observed by UV/
Vis spectroscopy (Figure S5). A strong fluorescence turn-on
response was also recorded at lem = 584 nm (l = 490 nm)
ex
2
+
with a 65-fold enhancement relative to the control. At Pt
concentrations between 10–80 mm, both absorbance and
fluorescence increased linearly (Figure S6). Analysis of
a Jobꢀs plot indicated maximum fluorescence intensity at
a mole fraction of 0.5, which suggests that Rho-DDTC
Taken together, the data suggested that the displacement
of an am(m)ine ligand is a critical step for Rho-DDTC
activation. We reasoned that, in the absence of aquation, Pt
complexes first bind to a reactive DDTC motif with
concomitant displacement of an ammine ligand (Scheme S2).
II
2
+
interacted with Pt in a 1:1 ratio (Figure S7). The selectivity
2
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!