Drug delivery with a calixpyrrole-trans -Pt(II) complex

Article abstract of DOI:10.1021/ja307791j

A meso-p-nitroaniline-calix[4]pyrrole derivative trans-coordinated to a Pt(II) center was synthesized and its structure solved by X-ray analysis. Adenosine monophosphate (AMP) was used as a model compound to evaluate the potential for the assisted delivery of the metal to the DNA nucleobases via the phosphate anion-binding properties of the calix[4]pyrrole unit. An NMR investigation of the kinetics of AMP complexation in the absence of an H-bonding competing solvent (dry CD₃CN) was consistent with this hypothesis, but we could not detect the interaction of the calix[4]pyrrole with phosphate in the presence of water. However, in vitro tests of the new trans-calixpyrrole- Pt(II) complex on different cancer cell lines indicate a cytotoxic activity that is unquestionably derived from the coexistence of both the trans-Pt(II) fragment and the calix[4]pyrrole unit.

Full text of DOI:10.1021/ja307791j

Article
pubs.acs.org/JACS
Drug Delivery with a Calixpyrrole−trans-Pt(II) Complex
Grazia Cafeo,† Grazia Carbotti,‡ Angela Cuzzola,∥ Marina Fabbi,‡ Silvano Ferrini,‡ Franz H. Kohnke,*,†
Georgia Papanikolaou,† Maria Rosaria Plutino,⊥ Camillo Rosano,‡ and Andrew J. P. White§
†Dipartimento di Scienze Chimiche, Universita
‡IRCCS A.O.U. San Martino-IST, Istituto Nazionale per la Ricerca sul Cancro, Largo R. Benzi 10, 16132 Genoa, Italy
̀
di Messina, viale F. Stagno D’Alcontres 31, 98166 Messina, Italy
∥Dipartimento di Chimica e Chimica Industriale, Universita
̀
di Pisa, via Risorgimento 35, 56126 Pisa, Italy
⊥Istituto per lo Studio dei Materiali Nanostrutturati, ISMN - CNR, OU Palermo, c/o Dipartimento di Scienze Chimiche, Universita
degli Studi di Messina, Viale F. Stagno d’Alcontres 31, 98166 Messina, Italy
§Chemical Crystallography Laboratory, Department of Chemistry, Imperial College London, South Kensington, London SW72AZ
United Kingdom
̀
S
* Supporting Information
ABSTRACT: A meso-p-nitroaniline-calix[4]pyrrole derivative trans-coordinated
to a Pt(II) center was synthesized and its structure solved by X-ray analysis.
Adenosine monophosphate (AMP) was used as a model compound to evaluate
the potential for the assisted delivery of the metal to the DNA nucleobases via
the phosphate anion-binding properties of the calix[4]pyrrole unit. An NMR
investigation of the kinetics of AMP complexation in the absence of an H-
bonding competing solvent (dry CD3CN) was consistent with this hypothesis,
but we could not detect the interaction of the calix[4]pyrrole with phosphate in
the presence of water. However, in vitro tests of the new trans-calixpyrrole−
Pt(II) complex on different cancer cell lines indicate a cytotoxic activity that is
unquestionably derived from the coexistence of both the trans-Pt(II) fragment
and the calix[4]pyrrole unit.
inactive, has been re-examined by the discovery of a number of
trans-Pt(II) derivatives that were found to be effective and/or
even complementary where cis-[PtCl2(NH3)2] resistance was
observed.5,8 Extensive research in this area has thus
demonstrated that the nature of the ligands, along with the
cis-/trans-stereochemistry, plays a crucial role in the cytotoxic
activity of Pt(II) complexes.
Calixpyrroles are a class of molecular receptors that have
gained considerable interest due to their ability to bind anions9
and to act as ditopic (ion-pair) receptors10 as well as hosts for
neutral molecules11 that can accept NH hydrogen bonds.12 In
1998, Sessler and co-workers13 reported that SiO2 stationary
phase functionalized with calix[4]pyrrole units could be used in
the HPLC separation of oligonucleotides and peptides with
advantages over sapphyrin-modified silica gels, thus indicating
the ability of the calix[4]pyrrole unit to recognize these anionic
biomolecules.
INTRODUCTION
■
Traditional cancer chemotherapy is severely hampered by the
toxic side effects of the drugs used for this purpose; hence, the
search for ways to achieve selective drug delivery to the
diseased tissues is a topic of considerable current activity. A
number of novel drug-delivery methods have emerged over
recent years that are based on the achievements of supra-
molecular chemistry and which include the use of dendrimers.1
Passive targeting by the EPR effect2 has been exploited using
drug−nanoparticles and drug−polymer conjugates,3 and active
targeting has also been pursued, using chemical moieties that
are preferentially recognized by cancerous cells as part of the
drug’s structure, so that these moieties can act as ‘homing’
devices.4 The achievement of effective cancer regression using
lower drug doses is also one of the objectives of selective drug-
delivery systems.
Platinum complexes are a group of anticancer drugs with a
long history, cis-[PtCl2(NH3)2] being the first of a large number
of Pt(II) derivatives studied in this area, a few of which have
reached and are still in clinical use.5,6 Pt(II) complexes have
recently received renewed attention with the introduction of a
number of new-generation drugs containing carboxylate ligands
associated with planar bulky amines and/or N-donor hetero-
cycles.7 Moreover, the initial assumption that cis-stereo-
chemistry is an absolute prerequisite for anticancer activity,
fostered by the observation that the trans-[PtCl2(NH3)2] is
Inspired by the above-mentioned results on the separation of
oligonucleotides, we reasoned that, if an anticancer drug could
be associated with the calixpyrrole unit, then its delivery to
nucleobases (in DNA) could be improved as a result of the
interaction of the calix with the anionic phosphate units.14 With
these thoughts in mind, we prepared the aminophenyl
Received: August 6, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja307791j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 1. Synthetic Pathway to trans-5
derivative of calix[4]pyrrole 4 (Scheme 1) to be used as a
suitable ligand in Pt(II) coordination. Here we report our initial
studies of a trans-Pt(II)-calixpyrrole derivative with potential
application as an anticancer drug.
RESULTS AND DISCUSSION
■
meso-p-Aminophenyl derivative of calix[4]pyrrole 4 was
synthesized as described in Scheme 1. The method is similar
to that recently reported for a meso-ethyl-substituted analogue
of 3.15 Calix[4]pyrrole 3 was obtained by the same method
used for the cyclodimerization of 2,16 but in the presence of an
excess of pyrrole (5 equiv). Mixing equimolar amounts of
calix[4]pyrrole 4 and cis-[PtCl2(DMSO)2] 1 in CH3CN
resulted in the formation of pale-yellow crystals (mp 166−
168 °C).
The single-crystal structure of trans-5 shows the platinum to
have a distorted square planar coordination geometry [cis
angles in the range 87.23(11)−94.41(4)°] with two trans-
chloride ligands, the nitrogen of an aminophenyl moiety, and a
sulfur-bound DMSO molecule comprising the coordination
sphere (Figure 1). Hence, after the substitution, a configura-
tional isomerization with respect to 1 has occurred.17 The plane
of the aminophenyl ring is inclined by ∼106° to the mean plane
of the calix[4]pyrrole macrocycle [as defined by the four
quaternary carbon atoms C(6), C(14), C(22), and C(30)], and
the Pt(II) coordination plane is inclined by ∼110° to this
former plane such that the complex forms a U-shaped cleft.
An included DMSO solvent molecule is held within this cleft
by a series of N−H···O hydrogen-bonding interactions
involving the N−H groups of three of the four pyrrole rings
(interactions a−c in Figure 1). Whereas these three pyrrole
rings are inclined to the mean plane of the calix[4]pyrrole
macrocycle by between ∼47° and 61° with their N−H groups
pointing toward the trapped DMSO molecule, the fourth
[based on N(18)] is inclined by ∼73° with the N−H group
pointing away from the hydrogen-bonded DMSO molecule.
This N−H group is not involved in any significant
intermolecular interactions. There is an approach of one of
the methyl hydrogen atoms of the encapsulated DMSO
molecule to the O(40) oxygen atom of its coordinated
counterpart, but with an O···H separation of ∼2.56 Å any
interaction can only be weak. If one takes into account that
Figure 1. Crystal structure of trans-5. Selected bond lengths (Å) and
angles (deg). Pt−Cl(1) 2.2980(11), Pt−Cl(2) 2.3090(11), Pt−N(38)
2.095(4), Pt−S(40) 2.2132(11), Cl(1)−Pt−Cl(2) 176.02(4), Cl(1)−
Pt−N(38) 87.23(11), Cl(1)−Pt−S(40) 94.41(4), Cl(2)−Pt−N(38)
88.98(11), Cl(2)−Pt−S(40) 89.43(4), N(38)−Pt−S(40) 176.73(11).
The N−H···O hydrogen bonds have [N···O] and [H···O] separations
(Å), and [N−H···O] angles (deg) of (a) 2.878(5), 1.98, 178; (b)
3.041(6), 2.17, 164; (c) 3.215(6), 2.40, 150, respectively.
trans-5 crystallizes from a CH3CN solution (initial concen-
tration of the two reagents 17 mM) in which the only ‘available’
DMSO is the one displaced from the Pt(II) center, the
presence of the DMSO within the calixpyrrole moiety of trans-5
indicates a selectivity of over 1:103 (at least in the solid state)
with respect to acetonitrile, which is also known to form
hydrogen-bond complexes with calixpyrrole derivatives.18 The
other included solvent molecule (an acetonitrile molecule) is
B
dx.doi.org/10.1021/ja307791j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 2. Partial 1H MNR (500 MHz, CD2Cl2, 25 °C, 5 mM) spectra of 4 (a), and of the reaction mixture 4 + 1 (b−e) at the indicated times after
the addition of 1 equiv of 1. Selected signals are assigned using black squares and circles and white squares as shown in the reaction diagram.
1
linked to the outside of the complex by a hydrogen bond to the
aminophenol nitrogen atom N(38) with N···O and H···O
separations of 3.082(9) and ∼2.22 Å respectively, and an N−
H···O angle of 159°.
2.54 for the DMSO displaced from 1. The H NMR changes
described for the DMSO signals were consistent with the
spectral changes observed for the resonances pertaining to the
calixpyrrole unit. Upon formation of trans-5, the resonance for
the NH2 unit at δ 3.63 disappeared to give a broad singlet at δ
6.02 for the newly formed Pt-NH2 moiety. A new AA’BB’
system centered at δ 6.95 and 7.27 was formed, while the
resonances of the AA’BB’ system of the uncoordinated
calixpyrrole 4 at δ 6.55 and 6.74 disappeared. The general
downfield shift relative to the calixpyrrole protons clearly
indicated an happened Pt(II) coordination. On the other hand,
The formation of trans-5 was monitored by 1H MNR also in
CD2Cl2 solution (5 mM) at 25 °C where crystals were not
formed at this concentration (see Figure 2). The displacement
of DMSO from the Pt center was indicated by a reduction of
intensity of the signal at δ 3.50 for the bound DMSO in cis-
[PtCl2(DMSO)2] 1 and the appearance of two new signals: one
at δ 3.34 for the DMSO coordinated to the Pt(II) and one at δ
C
dx.doi.org/10.1021/ja307791j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 2. Synthetic Pathway to trans-8
the Pt-NH2 amino and the bound DMSO protons did not show
ization was accelerated on the addition of free DMSO to a
solution of trans-5.
3
the expected diagnostic coupling constant (2JPtH and JPtH
respectively) with 195Pt, even when lowering the magnetic
field strength (300 vs 500 MHz) or increasing the sample
temperature. The high molecular weight of trans-5 caused a
decrease of the Pt relaxation time (T1) via the chemical
shielding anisotropy (CSA), thereby resulting in an observed
loss of the two typical satellite peaks.19 Moreover, on standing,
the compound developed a new resonance at δ 3.42 that
increased gradually, while the resonance at δ 3.34 decreased by
a corresponding amount (the sum of the intensities for the two
signals remained constant). We assigned the new signal δ 3.42
to the DMSO unit of cis-5, formed by a rearrangement at the
Pt(II) center of the initially formed trans-5. Concomitantly, a
new AA’BB’ system appeared at δ 6.99 and 7.37, and an NH2
resonance at δ 5.80 increased in intensity over time, while the
resonances for the initially formed trans-5 decreased (Figure S2,
Supporting Information). Within 77 h, the ratio between the
initially formed and the rearranged isomers was 75:25. Even in
the presence of an excess of calixpyrrole 4, only one of the
DMSO ligands of 1 was displaced from the Pt(II) center.
The formation of one trans-Pt(II) coordination compound
that subsequently equilibrated with a rearranged cis-isomer
generated concern regarding the correlation between the
structure observed in the solid state and the stereochemical
assignments proposed for the species in solution. To address
this issue, a freshly prepared batch of the crystalline product
obtained from CH3CN was filtered from the crude mixture (the
sample had the same melting point as that analyzed by X-ray
diffraction), and it was immediately dissolved in CD2Cl2. The
1H NMR spectrum was recorded and found to be identical to
that of the coordination compound formed in the initial stages
of the reaction (i.e., before trans/cis isomerization equilibrium).
Free DMSO was also present in a 1:1 stoichiometric ratio with
respect to trans-5. On standing, the spectral changes described
above for the partial trans-5/cis-5 equilibration was observed.
However, all our attempts to isolate cis-5 as a pure compound
from CH3CN solutions were unsuccessful, probably because it
does not yield crystals and/or it equilibrates to trans-5 as this is
subtracted from the solution during its crystallization. Solutions
of trans-5 that did not contain free DMSO (displaced from 1,
see Supporting Information [SI]) were found to be configura-
tionally stable (over 24 h at 25 °C in either CD2Cl2 or
CD3CN). Thus, the presence of ‘free’ DMSO plays a crucial
role in the isomerization process described above. In fact, in
separate experiments we observed that the trans/cis isomer-
The DMSO molecule in the solid-state structure of trans-5
prompted us to investigate whether the calixpyrrole moiety
plays an accelerating role toward the formation of this
compound. In fact, the calixpyrrole unit could either assist
the departure of one DMSO molecule from the Pt(II) center in
1 through its complexation, or it could favor the approach of
the NH2 unit to the Pt(II) center by clamping to a bound
DMSO moiety of the metal complex 1. Coordination of DMSO
to the calix moiety of trans-5 could also accelerate the observed
trans/cis isomerization. Calixpyrroles are also capable of
complexing chloride.9−12 We therefore compared the formation
of trans-5 with those of dipyrromethane derivative 6 (Scheme 2,
Figure S3 in SI) and p-toluidine 7 (Figure S4 in SI), which both
contain the NH2 ligand but not the calixpyrrole moiety. The
kinetic study of the substitution reactions of 1 with ligands 4, 6,
and 7 (CD2Cl2 at 25 °C in a 1:1 molar ratio 10 mM) provided
second-order rate constants k2/s−1 M−1 (see SI) 0.311
( 0.006), 0.143 ( 0.004), and 0.261 ( 0.007) for trans-5,
trans-8, and trans-9, respectively. The observed t1/2 were 330,
750, 450 s, respectively (Figures S5−S7, SI).
1
The intensities of the diagnostic H NMR signals were used
to measure the progress of the reactions (see Figures S3−S4 in
SI). Stereochemical assignments were based on analogies of the
1H NMR resonances of these compounds with those of the
corresponding substructure(s) present in trans-5. In all cases
the trans/cis isomerization of the initially formed amino-Pt(II)
complexes were marginal until nearly quantitative displacement
of one mole of DMSO from 1 had occurred. However, while
trans-8 was sufficiently stable to be isolated as a pure isomer,
stereochemically pure 9 could not be isolated. Interestingly, in
the p-toluidine trans-9/cis-9 derivatives, characterized by a
lower molecular weight than trans-5/cis-5, the protons for the
amino unit and the DMSO directly bonded to the platinum(II)
center exhibited typical 195Pt satellites (60 vs 70 Hz, and 20 vs
25 Hz, see SI). The difference in 195Pt coupling constants for
the two species was in agreement with the slightly different
trans influence exerted by the dimethylsulfoxide/amino groups
(trans-9) and the chloride moieties (cis-9).20
The amino-ligands 4, 6, and 7 have similar electronic
requirements, but they have substantially different steric bulk. If
we compare rates for p-toluidine, 7, and dipyrromethane
derivative 6 (t1/2 = 450 and 750 s, respectively), we observe a
difference that is qualitatively consistent with the larger steric
hindrance of 6. However, the bulkier 4 has a t1/2 of 330 s, a
value that suggests an active role for the calixpyrrole. However,
D
dx.doi.org/10.1021/ja307791j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 3. Partial 1H NMR spectra (CD3CN/D2O 80:20 v/v, 1.5 mM, 25 °C) of (a) trans-5; (b) the reaction mixture of trans-5 and (TBA)2AMP 11
(1:1) 1.5 h after mixing; the signals reproduced in the inset with expanded vertical scale are consistent with the presence of the indicated m/z values
in the ESI-MS spectrum of this mixture; (c) (TBA)2AMP 11.
when the concentration of 4 was doubled, t1/2 for the formation
of trans-5 was 180 s (Figure S5b in SI). This further evidence of
second-order kinetics suggests that the formation of trans-5
does not occur within a supramolecular [4 + 1] complex.21
However, complexation of the displaced DMSO by the newly
formed trans-5 (and possibly by unreacted 4) would slow down
the back re-entry reaction of DMSO, thus producing the
observed acceleration compared with that in which trans-8,
lacking the assisting calixpyrrole cavity, is formed.
As our objectives included an evaluation of the biological
activity of trans-5, we tested its stability in the presence of two
species that are ubiquitous in the biological environment: water
and chloride. No significant spectral changes were observed in
D2O-saturated CD2Cl2 or in D2O/CD3CN (20:80, v/v)
solutions after over 4 h. Appreciable stability was also observed
in the presence of chloride [trans-5/TBACl 1:10, in D2O/
CD3CN (20:80, v/v), 50 mM in chloride, i.e. ∼5 times the
chloride concentration normally found in plasma].22 Under
these conditions only 30% of the Pt(II) in trans-5 was displaced
after 24 h. Although the formation of aqua species is believed to
be a key step for the cytotoxic activity of Pt complexes,8c we
were pleased with the observed stability of trans-5, as it
supports the assumption that trans-5 is, at least initially, the
chemical species that interacts with the cells in the in vitro tests
(vide infra). Similar considerations apply for the stability in the
presence of chloride.
Since DNA is a polyphosphate, we then tested the ability of 4
and of trans-5 to interact with TBAH2PO4 and (TBA)2HPO4.
Significant complexation-induced shifts were observed in the
1H NMR spectra (CD2Cl2) while trans-5 was not decomposed
by these anions. However, the measurement of association
constants was hampered by the extensive broadening of the
NMR resonances before an excess of salts was added (e.g. see
Figure S8 in SI for trans-5 + TBAH2PO4).
We arbitrarily selected adenosine monophosphate (AMP) as
a DNA model that could be used in a preliminary investigation
of the ability of trans-5 to deliver the Pt(II) drug to a
nucleobase,23 and we tested the ability of 4 to form a
1
supramolecular complex with the (TBA)2AMP 11. The H
NMR spectrum of a mixture of 4 and (TBA)2AMP 11 (1.7:1 in
CD3CN, Figure S9 in SI) shows a broadening of all the calix
resonances, especially for the pyrrole NH units that could not
be located, even at −40 °C. The formation of a complex was
also confirmed by a significant Δδ in the 31P{1H} NMR
spectrum (4.20 to 3.10 ppm). The reaction of trans-5 with
(TBA)2AMP was investigated by 1HNMR in CD3CN (1:1; 1.5
mM, 25 °C). On mixing the reagents, the pyrrole NH units
shifted and appeared as broad signals at higher ppm values.
E
dx.doi.org/10.1021/ja307791j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 1. In Vitro Cytotoxicity of Compounds 6 and 4 and of Their Pt(II) Derivatives trans-8 and trans-5, Compared with That
of trans- and cis-[PtCl2(NH3)2], Oxaliplatin, and Carboplatin in a Human Cancer Cell-Line Panel As Assessed by the MTT Cell
a
Viability Assay
b
e
b
b
c
c
cmpd
A2774
OVCAR3
SKOV3
MDA-MB-231
MCF7
d
6
>100
32 (8)
98 (60)
23 (3)
>100
>100
48 (13)
103 (41)
35 (13)
>100
>100
46 (16)
106 (42)
28 (17)
>100
>100
39 (8)
89 (12)
26 (12)
>100
>100
64 (51)
114 (31)
40 (27)
>100
d
d
trans-8
d
4
trans-5
d
trans-[PtCl2(NH3)2]
f
cis-[PtCl2(NH3)2]
5 (4)
6 (2)
4 (1)
5 (1)
18 (12)
1 (0.3)
89 (7)
f
oxaliplatin
1.3 (1)
54 (1)
0.5 (0.1)
56 (2)
0.7 (0.2)
73 (6)
1.7 (1)
63 (13)
f
carboplatin
a
b
Values are IC50, in μM, expressed as mean values ( SD) of at least three independent experiments (72 h drug exposure). Human ovarian cancer
c
d
e
cell line. Human breast cancer cell line. Concentration range from 10 to 100 μM/L. >100: no cytotoxic activity was detected within the
concentration range tested. Concentration range from 0.05 to 500 μM/L.
f
These changes are consistent with the formation of a complex
between the calixpyrrole moiety of trans-5 and the anionic
component of the nucleotide. All of the nucleotide resonances
were also considerably broadened. The substitution toward the
Pt(II) complex by the nucleobase was indicated by the
formation of uncoordinated 4, while all of the AMP protons
broadened, hampering any δ assignment (Figure S10 in SI).
When the same experiment was performed in CD3CN/D2O, a
number of signals could be detected for the adenine moiety
(Figure 3) consistent with a mixture of complexation products
suggesting the presence/contribution of a more complex
model, but we did not investigate this further (Figure S15 in
SI). Calixpyrroles have been shown to be able to bind anions
with significant strength even in the presence of water,25
although less effectively than in the absence of hydrogen-
bonding competing solvents. However, water may also produce
Pt(II) aqua species, and these would be reactive intermediates
producing ‘apparent’ first-order kinetic behavior. However,
trans-8 reacted with 11 with a first-order kinetic behavior, at
essentially the same rate in both dry CD3CN and CD3CN/D2O
(80:20 v/v) (Figures S16, S17 in SI, t1/2 1500 and 1400 s,
respectively). The dipyrromethane unit of trans-8 is expected to
bind phosphate considerably less effectively than the
calixpyrrole in trans-5, and for trans-8 there was no evidence
of an ‘intracomplex’ reaction of 11. Since an intramolecular
substitution could not take place, most probably in this case the
first-order kinetic could be ascribed to a solvolytic pathway.
Overall, the above kinetic studies indicate that the calixpyrrole
unit is playing a role in the nucleobase coordination by means
of trans-5 under dry conditions, and it might still contribute to
some extent in the presence of water.
Ligands associated with either trans- or cis-Pt(II) have been
shown to play a crucial role in the cytotoxic activity of this
metal.7 We therefore tested trans-5 and trans-8 for their
cytotoxicity toward a panel of human cancer cell lines (Table 1)
using a cell viability assay based on tetrazolium reduction. A
comparison between trans-5 and trans-8 was expected to define
the importance of the whole calixpyrrole moiety, free ligands 4
and 6 were used as ‘blanks’, trans- and cis-[PtCl2(NH3)2],
oxaliplatin, and carboplatin as benchmarks.8c The cytotoxic
activity of the different compounds and complexes was first
tested in dose−response experiments at different time points
(24, 48, and 72 h) of continuous exposure. The representative
experiments (see Figure S18 in SI) show that the cytotoxic
activity of trans-5 and trans-8 is already observed at 24 h and
reaches its maximal effect at 72 h. Interestingly, the cytotoxic
effect of trans-5 and trans-8 was detectable earlier than that of
the reference cytotoxic drug, cis-[PtCl2(NH3)2], which had
limited effect at 24 h. As reported in Table 1, A2774, OVCAR-
3, and SKOV-3 ovarian carcinoma cells were sensitive to the
trans-5 complex with an average IC50 of 23, 35, and 28 μM,
respectively (at 72 h), while trans-[PtCl2(NH3)2] showed no
toxicity in the concentration range tested (10−100 μM). trans-
5 was consistently more effective than carboplatin but less than
cisplatin and oxaliplatin. In addition, the calix[4]pyrrole
derivative 4 had very limited toxicity with IC50 close to 100
which were not isolated.22 However, additional evidence for
b
AMP coordination was provided by ESI-MS (negative ion
mode) of the crude mixture (Figure S11 in SI), which
contained signals at m/z values consistent with the presence of
[Pt(OH)2(CH3CN)AMP], [PtCl(OH)(CH3CN)AMP],
[PtCl2(CH3CN)AMP], and [PtCl(CH3CN)(AMP)2]. All of
these peaks exhibited an isotopic pattern consistent with the
indicated composition.
To investigate whether the calixpyrrole fragment played an
active role for the transfer of Pt(II) from trans-5 to the
nucleobase, the kinetic behavior of the reaction was analyzed,
and this could be fitted to a first-order model (the observed t1/2
differences, i.e. 690 and 670 s for trans-5:11 being 1:1 and 1:4,
respectively, were within experimental errors, see Figure S12 in
SI). These kinetic results are consistent with the coordination
of the nucleobase taking place within the preformed [trans-
5·11]complex. Although we could not measure the association
constant for this complex by NMR titration, it is evident that at
the concentration used for this study it is little dissociated.
Increasing the concentration of 11 cannot significantly shift the
equilibrium toward its formation any further, and if the reaction
occurs mostly within the [trans-5·11] complex, the reaction rate
is only marginally affected by changes in the concentration of
11. Additional evidence for the ‘intra complex’ occurrence of
the coordination is provided by the interference of TBAF,
because fluoride can effectively compete with phosphate
binding by the calixpyrrole unit (Figure S13 in SI).24 In fact,
the observed t1/2 for the reactions [(TBA)2AMP + trans-5
(1:1)] and [(TBA)2AMP + trans-5 + TBAF (1:1:1.1)], at 25
°C, 1.5 mM trans-5 in CD3CN, were 690 and 1200 s (see
Figure S14 in SI for the latter value).
Reactions of trans-5 with 11 in CD3CN/D2O (80:20 v/v)
could still be fitted to a first-order rate law, although they were
slower than the analogous reactions in dry CH3CN, with the
only effect that the increased concentration of 11 moved the
equilibrium toward the complete formation of the final species,
F
dx.doi.org/10.1021/ja307791j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
μM in the three cell lines. Complex trans-8 also showed
cytotoxic activity with IC50 of 32, 48, and 46 μM, respectively
whereas the dipyrromethane derivative 6 showed no toxicity.
Since DMSO is a ligand that does not promote the activity of
Pt(II) derivatives26 these data indicate that ligands 4 and 6
confer cytotoxic activity to the otherwise inert trans-
[PtCl2(NH3)2]. Similar results were obtained with breast
cancer cell lines MDA-MB-231 and MCF7 (Table 1).
To mimic the situation in vivo, where the drug exposure time
is limited by its clearance, cytotoxic activity was evaluated by a
short treatment (3 h) of the cells with the different compounds
and complexes, followed by cell culture for 72 h. As shown in
the representative experiment of Figure S19 in SI, trans-8 and
trans-5 had significant cytotoxic activity (IC50 of 61 and 66 μM,
respectively), which appeared more similar to that of cis-
[PtCl2(NH3)2] (IC50 = 56 μM) under these experimental
conditions.
DNA phosphate groups under the conditions and in the
presence of the complex chemical mixture that is used for cell
cultures. Nevertheless, our hypothesis is supported by a
molecular modeling and docking simulation (see SI and Figure
S21). According to this model, trans-5 forms hydrogen bonds
between the pyrrole rings of the calyx and two oxygen atoms of
a phosphate, consequently pointing the trans-Pt inside the
minor groove toward two consecutive adenine residues. The
results obtained to date indicate that the calixpyrrole unit has
considerable potential for the development of new anticancer
trans-Pt(II) drugs. In vitro trans-5 outperforms the widely used
carboplatin but lags behind oxaliplatin and the very toxic
cisplatin. A comparison with the compounds used as blanks
also indicates that trans-5 and trans-8 reach the biological site of
action (most likely DNA) before losing the ligands 4 and 6
respectively. In this study we found that calixpyrrole 4 and
dipyrromethane 6 have low cytotoxicity. We also observed that
dipyrromethane 6 may produce proliferative effects (see Figure
S18 in SI), and we plan additional studies to uncover the
origins of this property. To the best of our knowledge, the work
presented here is the first report on the biological activity of
calixpyrroles, and it also shows that this class of receptor
molecules, whose host guest chemistry was uncovered by the
seminal work of Sessler and co-workers,11 is still rich in
potential applications even after 16 years of research.
To gain more insight into the cytotoxic mechanisms of trans-
Pt complexes 5 and 8, we evaluated apoptosis by the annexin-
V/propidium iodide (annexin-V/PI) method in A2774 cells
after a 24-h treatment. As shown in Figure 4 (and Figure S20 in
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, spectroscopic data, and crystallo-
graphic data for trans-5. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
franz@unime.it
■
Notes
The authors declare no competing financial interest.
Figure 4. Flow-cytometric analysis of annexin-V/PI staining of A2774
ovarian cancer cells treated with the different compounds at their IC80
dose or their acetonitrile solvent for 24 h. Viable cells take up no
stains. Cells stained with PI alone are necrotic, whereas cells stained
with annexin V-FITC alone represent early apoptosis. Cells in the final
stages of apoptosis take up both stains. Data are expressed as the
percentage of cells in each category.
ACKNOWLEDGMENTS
■
We thank Prof Luigi Monsu Scolaro for helpful discussions,
̀
Miss Silvia Carbone for helping with the biological assays, Mr.
Gaetano Irrera (ISMN-CNR) for technical assistance, the
Consiglio Nazionale delle Ricerche (CNR), AIRC Grant 5955,
and the Compagnia di San Paolo di Torino and the Italian
Ministry of Health for financial support.
SI), both trans-8 and trans-5 complexes induced apoptosis in
ovarian cancer cell lines. In particular, trans-8 efficiently
induced an apoptotic response, whereas in samples treated
with trans-5 a considerable percentage of necrotic cells was also
detected.
The antitumor activity of cis-[PtCl2(NH3)2] is related to the
induction of apoptosis.27 However, both apoptosis and necrosis
have been found in the same population of cis-[PtCl2(NH3)2]-
treated cells,28 possibly as a result of cisplatin-induced damage
to the apoptotic program.29
REFERENCES
■
(1) (a) Venkataraman, S.; Hedrick, J. L.; Ong, Z. Y.; Yang, C.; Ee, P.
L. R.; Hammond, P. T.; Yang, Y. Y. Adv. Drug Delivery Rev. 2011, 63,
1228−1246. (b) Yang, K.; Weng, L.; Cheng, Y.; Zhang, H.; Zhang, J.;
Wu, Q.; Xu, T. J. Phys. Chem. B 2011, 115, 2185−2195. (c) Medina, S.
H.; El-Sayed, M. E. H. Chem. Rev. 2009, 109, 3141−3157. (d) Tekade,
R. K.; Kumar, P. V.; Jain, N. K. Chem. Rev. 2009, 109, 49−87.
(2) Fang, J.; Nakamura, H.; Maeda, H. Adv. Drug Delivery Rev. 2011,
63, 136−151.
(3) (a) Bei, D.; Meng, J.; Youan, B.-B. C. Nanomedicine (London,
United Kingdom) 2010, 5, 1385−1399. (b) Rao, J. ACS Nano 2008, 2,
1984−1986. (c) Haag, R.; Kratz, F. Angew. Chem., Int. Ed. 2006, 45,
1198−1215.
(4) (a) Teow, Y.; Valiyaveettil, S. Nanoscale 2010, 2, 2607−2613.
(b) Low, P. S.; Henne, W. A.; Doorneweerd, D. D. Acc. Chem. Res.
2008, 41, 120−129. (c) van Zutphen, S.; Reedijk, J. Coord. Chem. Rev.
2005, 249, 2845−2853. and references therein. (d) Ndinguri, M. W.;
Solipuram, R.; Gambrell, R. P.; Aggarwal, S.; Hammer, R. P.
CONCLUSIONS
■
This work was stimulated by the results of the kinetic studies
discussed above and by the idea that calixpyrrole 4 could target
the delivery of platinum drugs to DNA. We currently cannot
provide irrefutable proof that the observed cytoxic activity of
trans-5 can be explained by a mechanism involving the initial
formation of a complex between its calixpyrrole unit and the
G
dx.doi.org/10.1021/ja307791j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Bioconjugate Chem. 2009, 20, 1869−1878. (e) Mukhopadhyay, S.;
Barnes, C. M.; Haskel, A.; Short, S. M.; Barnes, K. R.; Lippard, S. J.
Bioconjugate Chem. 2008, 19, 39−49.
(5) Abu-Surrah, A. S.; Kettunen, M. Curr. Med. Chem. 2006, 13,
1337−1357. (b) van Zutphen, S.; Pantoja, E.; Soriano, R.; Soro, C.;
Tooke, D. M.; Spek, A. L.; den Dulk, H.; Brouwer, J.; Reedijk, J.
Dalton Trans. 2006, 1020−1023.
(6) For several examples of supramolecular chemistry used for the
‘drug delivery’ of Pt compounds see: (a) Kapp, T.; Dullin, A.; Gust, R.
Bioconjugate Chem. 2010, 21, 328−337. (b) Chen, G. J.; Kumar, J.;
Gregory, A.; Stenzel, M. H. Chem. Commun. 2009, 6291−6293.
(c) Malik, N.; Evagorou, E. G.; Duncan, R. Anticancer Drugs 1999, 10,
767−776.
(7) (a) Harper, B. W.; Krause-Heuer, A. M.; Grant, M. P.; Manohar,
M.; Garbutcheon-Singh, K. B.; Aldrich-Wright, J. R. Chem.Eur. J.
2010, 16, 7064−7077.
32−38. (c) Navarro, J. A. R.; Lippert, B. Coord. Chem. Rev. 1999, 185−
186, 653−667.
1
(24) The H NMR titration of trans-5 with TBAF resulted in initial
disappearance of the NH resonances which appeared as very broad
signals at high ppm values in a 1:1 mixture (Figure S13 [SI]), thus
hampering the measurement of a binding constant by this method.
(25) Buschmann, H.-J.; Schollmeyer, E.; Cleve, E.; Mutihac, L. Anal.
Lett. 2011, 44, 2703−2710.
(26) (a) Yi, Y. W.; Insoo, B. DNA Repair 2011, 10, 1084−1085.
(b) Fischer, S. J.; Benson, L. M.; Fauq, A.; Naylor, S.; Windebank, A. J.
NeuroToxicology 2008, 29, 444−452.
(27) Eastman, A. The Mechanism of Action of Cisplatin: From
Adducts to Apoptosis. In Cisplatin. Chemistry and Biochemistry of a
Leading Anticancer Drug; Lippert, B., Ed.; Wiley-VCH: Basel,
Switzerland, 1999; pp 111−134,
(28) Pestell, K. E.; Hobbs, S. M.; Titley, J. C.; Kelland, L. R.; Walton,
M. I. Mol. Pharmacol. 2000, 57, 503−511.
(29) Gonzalez, V. M.; Fuertes, M. A.; Alonso, C.; Perez, J. M. Mol.
Pharmacol. 2001, 59, 657−663.
(8) (a) Aris, S. M.; Farrell, N. P. Eur. J. Inorg. Chem. 2009, 1293−
1302. (b) Coluccia, M.; Natile, G. Anti-Cancer Agents Med. Chem.
2007, 7, 111−123. (c) Kelland, L. Nat. Rev. Cancer 2007, 7, 573−584.
(d) Radulovic, S.; Tesic, Z.; Manic, S. Curr. Med. Chem. 2002, 9,
1611−1618. (e) Natile, G.; Coluccia, M. Coord. Chem. Rev. 2001,
216−217, 383−410.
(9) (a) Gale, P. A.; Anzenbacher, P.; Sessler, J. L. Coord. Chem. Rev.
2001, 222, 57−102. (b) Gale, P. A.; Sessler, J. L.; Kral, V. Chem.
Commun. 1998, 1−8.
(10) Custelcean, R.; Delmau, L. H.; Moyer, B. A.; Sessler, J. L.; Cho,
W.-S.; Gross, D.; Bates, G. W.; Brooks, S. J.; Light, M. E.; Gale, P. A.
Angew. Chem., Int. Ed. 2005, 44, 2537−2542.
(11) Allen, W. E.; Gale, P. A.; Brown, C. T.; Lynch, V. M.; Sessler, J.
L. J. Am. Chem. Soc. 1996, 118, 12471−12472.
(12) Gale, P. A. Acc. Chem. Res. 2011, 44, 216−226.
(13) Sessler, J. L.; Gale, P. A.; Genge, J. W. Chem.Eur. J. 1998, 4,
1095−1099.
(14) There are several examples in the literature of molecules (e.g.,
intercalators) that were linked to Pt(II) derivatives to target DNA. See
for example Whan, R. M.; Messerle, B. A.; Hambley, T. W. Dalton
Trans. 2009, 932−939. For examples of anticancer activity of Pt(II)
complexes of aromatic amines see: Osa, T.; Hino, H.; Fujieda, S.;
Shimizu, F.; Arakawa, M. Chem. Pharm. Bull. 1986, 34, 1881−1887.
(15) Yang, W.; Yin, Z.; Wang, C.-H.; Huang, C.; He, J.; Zhu, Xi.;
Cheng, J.-P. Tetrahedron 2008, 64, 9244−9252.
(16) Bruno, G.; Cafeo, G.; Kohnke, F. H.; Nicolo, F. Tetrahedron
2007, 63, 10003−10010.
(17) For examples of similar cis/trans-isomerisations see: (a) Kong,
P.-C.; Rochon, F. D. Inorg. Chim. Acta 1982, 61, 269−271.
(b) Szuecova, L.; Travnicek, Z.; Popa, I.; Marek, J. Polyhedron 2008,
27, 2710−2720. (c) Guido, E.; D’Amico, G.; Russo, N.; Sicilia, E.;
Rizzato, S.; Albinati, A.; Romeo, A.; Plutino, M. R.; Romeo, R. Inorg.
Chem. 2011, 50, 2224−2239 and refs therein.
(18) Anzenbacher, P., Jr.; Jursikova, K.; Lynch, V. M.; Gale, P. A.;
Sessler, J. L. J. Am. Chem. Soc. 1999, 121, 11020−11021.
(19) (a) Pregosin, P. S. Platinum NMR Spectrosc. Annu. Rep. NMR
Spectrosc. 1986, 17, 285−349. (b) Lippert, B. In Progress in Inorganic
Chemistry; Lippard, S. J., Ed.; Wiley: New York, 1989; Vol. 37, pp 1−
97.
(20) (a) Monsu Scolaro, L.; Mazzaglia, A.; Romeo, A.; Plutino, M. R.;
̀
Castriciano, M.; Romeo, R. Inorg. Chim. Acta 2002, 330, 189−196.
(b) Cafeo, G.; Lo Passo, C.; Monsu Scolaro, L.; Pernice, I.; Romeo, R.
̀
Inorg. Chim. Acta 1998, 275−276, 141−149.
(21) Cacciapaglia, R.; Di Stefano, S.; Mandolini, L. Acc. Chem. Res.
2004, 37, 113−122.
(22) The normal concentration of chloride in plasma should be
between 8 and 16 mM; from http://www.medicalhealthtests.com/
blog/pathology-test/chloride-plasma.html.
(23) Guanine N7 fragment in DNA is known to be one major target
for cis-Pt(II) besides adenine N7; see: (a) Dijt, F. J.; Canters, G. W.;
den Hartog, J. H. J.; Marcelis, A. T. M.; Reedijk, J. J. Am. Chem. Soc.
1984, 106, 3644−3647. (b) Martin, R. B. Acc. Chem. Res. 1985, 18,
H
dx.doi.org/10.1021/ja307791j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX