DOI: 10.1002/cmdc.201100355
A Heterodimeric Glucuronide Prodrug for Cancer Tritherapy: the Double
Role of the Chemical Amplifier
Marion Grinda,[a] Jonathan Clarhaut,[b, c] Isabelle Tranoy-Opalinski,[a] Brigitte Renoux,[a] Arnaud Monvoisin,[b]
Laurent Cronier,[b] and Sꢀbastien Papot*[a]
Most anticancer drugs suffer from poor selectivity leading to
severe side effects due to action against normal as well as dis-
eased tissue. Thus, the development of drug carriers designed
to deliver potent therapeutic agents exclusively at the tumor
site has emerged as one of the great challenges in medicinal
research.[1] One promising targeting strategy relies on the use
of nontoxic prodrugs that can be activated by an enzyme nat-
urally overexpressed in the tumor microenvironment.[2] In this
approach, enzymatic prodrug activation is followed by the re-
lease of the parent drug thereby restoring its antitumor activi-
ty selectively in malignant tissues. Several enzyme-responsive
prodrugs have already been evaluated in vivo with encourag-
ing results.[3] However, the relatively slow action and low con-
centration of all tumor-associated enzymes discovered so far
represent the “Achilles’ heel” of such a targeting strategy.
These unfavorable enzymatic parameters limit the amount of
drug liberated in targeted tissues and therefore the efficacy of
the treatment. To overcome this drawback, Shabat,[4] de G-
root[5] and McGrath[6] simultaneously introduced the concept
of self-immolative dendrimers allowing the release of several
drug units after a single triggering event, thanks to an efficient
chemical amplification process. In an elegant study, the Shabat
group also developed heterodimeric[7] and heterotrimeric[8] sys-
tems designed to deliver highly toxic drug cocktails with a
single enzymatic activation step.
matic event into a double drug release. Second, in the course
of this process, the amplifier is converted into a third highly
toxic species that, combined with the two other drugs, leads
to selective and potent cancer tritherapy.
The device 1 was designed to simultaneously target two es-
tablished agents: the widely used doxorubicin and the well
known histone deacetylase inhibitor MS-275. As illustrated in
Scheme 1b, this heterodimeric prodrug includes a nitrobenzyl-
phenoxy carbamate linker between the glucuronide trigger
and amplifier unit. With this design, the enzyme substrate is lo-
cated at a substantial distance from the two bulky drugs to
allow easy recognition of 1 by b-glucuronidase. Thus, enzyme-
catalyzed cleavage of prodrug 1 should result in the release of
phenol intermediate 2, which induces the release of aniline 3
through a 1,6-elimination process. Once turned on, the amplifi-
er first causes the expulsion of doxorubicin via a 1,4-elimina-
tion followed by spontaneous decarboxylation. Addition of
water to ortho-azaquinone methide 6 then generates aniline 7
thereby permitting the release of MS-275 along with the for-
mation of derivative 8. As azaquinone methides are potential
alkylating species,[10] we anticipated that 8 could also be toxic
toward cancer cells thus playing the role of a third antitumor
agent.
We chosen b-glucuronidase as the triggering enzyme since
it has been detected in high levels in a wide range of malig-
nancies, such as breast, lung, colon and ovarian carcinomas, as
well as melanomas.[11] Tietze[12] was one of the first researchers
to propose this tumor-specific enzyme as a target for selective
therapy using nontoxic glucuronide prodrugs in the course of
a prodrug monotherapy (PMT[11b]). Since then, several glucuro-
nide prodrugs[13] have been evaluated in vivo demonstrating
superior efficacy compared to standard chemotherapy.[14] The
efficiency of this approach is, however, limited by the reduced
turnover of b-glucuronidase in the tumor microenvironment.
Indeed, the optimal pH for b-glucuronidase activity is around
4, whereas the pH of tumor extracellular media is 6–7. In this
context, the use of novel targeting devices such as 1 should
circumvent this problem through the release of several drug
units after a single enzymatic hydrolysis.
Herein, we present the novel drug delivery device 1 de-
signed for the selective targeting of three different cytotoxic
agents (Scheme 1a). This system is composed of five distinct
units: an enzymatic trigger, a self-immolative linker[9] and two
potent anticancer drugs articulated around a chemical amplifi-
er. Enzymatic cleavage of the trigger–linker bond generates in-
termediate 2, which then falls apart spontaneously giving rise
to anticancer activity. In this study, we demonstrated for the
first time that the chemical amplifier can play two crucial roles
in the efficiency of this targeting strategy. First, the amplifier is
responsible for signal amplification transforming a single enzy-
[a] Dr. M. Grinda, Dr. I. Tranoy-Opalinski, Dr. B. Renoux, Dr. S. Papot
Laboratoire de Synthꢀse et Rꢁactivitꢁ des Substances Naturelles
UMR-CNRS 6514, Universitꢁ de Poitiers
The synthesis of prodrug 1 was carried out starting from al-
cohol 9,[15] previously described in the literature (Scheme 2).
First, protecting groups of the carbohydrate moiety were
modified via a five-step strategy to yield fully allyl-protected
glucuronyl derivative 10. Protected in this way, the glucuronide
can be entirely deprotected in the course of a one-step proce-
dure under mild conditions at the end of the synthesis.[16]
Treatment of benzyl alcohol 10 with 4-nitrophenyl chlorofor-
mate and pyridine afforded activated carbonate 11 (84%). Ani-
line 12[17] was introduced chemoselectively via nucleophilic
4 rue Michel Brunet, BP 633, 86022 Poitiers (France)
[b] Dr. J. Clarhaut, Dr. A. Monvoisin, Dr. L. Cronier
Institut de Physiologie et Biologie Cellulaires
UMR-CNRS 6187, Universitꢁ de Poitiers
1 rue Georges Bonnet, BP 633, 86022 Poitiers (France)
[c] Dr. J. Clarhaut
INSERM CIC 0802
2 rue de la Milꢁtrie, CHU de Poitiers, 86021 Poitiers (France)
Supporting information for this article is available on the WWW under
ChemMedChem 2011, 6, 2137 – 2141
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2137