Modification of the duocarmycin pharmacophore enables CYP1A1 targeting for biological activity

Article abstract of DOI:10.1039/c1cc15638a

The identification of an agent that is selectively activated by a cytochrome P450 (CYP) has the potential for tissue specific dose intensification as a means of significantly improving its therapeutic value. Towards this goal, we disclose evidence for the

Full text of DOI:10.1039/c1cc15638a

View Online / Journal Homepage / Table of Contents for this issue
As featured in:
Showcasing research from the Institute of Cancer
Therapeutics, University of Bradford
Modification of the Duocarmycin Pharmacophore enables
CYP1A1 Targeting for Biological Activity
CYP-mediated activation of agents based on the anticancer
duocarmycins to address their intrinsic toxicity. The mechanism is
exemplified by CYP1A1 mediated oxidative spirocyclisation of a
chloromethylindoline pharmacophore resulting in
DNA-alkylation and cell death. The DNA damage signalling cascade
can be depicted by confocal laser scanning microscope green
imaging of immunostained ꢀH2AX, a marker for foci of DNA double
strand breaks against a red background of total DNA staining using
See Laurence H. Patterson et al.,
Chem. Commun., 2011, 47, 12062.
DRAQ5TM
.
www.rsc.org/chemcomm
Registered Charity Number 207890

Dynamic Article Links
ChemComm
View Online
Cite this: Chem. Commun., 2011, 47, 12062–12064
www.rsc.org/chemcomm
COMMUNICATION
Modification of the duocarmycin pharmacophore enables CYP1A1
targeting for biological activityw
Klaus Pors,a Paul M. Loadman,a Steven D. Shnyder,a Mark Sutherland,a
Helen M. Sheldrake,a Meritxell Guino,b Konstantinos Kiakos,c John A. Hartley,c
Mark Searceyb and Laurence H. Patterson*a
Received 12th September 2011, Accepted 4th October 2011
DOI: 10.1039/c1cc15638a
The identification of an agent that is selectively activated by a
cytochrome P450 (CYP) has the potential for tissue specific dose
intensification as a means of significantly improving its therapeutic
value. Towards this goal, we disclose evidence for the pathway of
activation of a duocarmycin analogue, ICT2700, which targets
CYP1A1 for biological activity.
not encompass the phenolic hydroxyl group in their structural
architectures.5 We hypothesise that de-hydroxylated (de-OH)
chloromethylindolines possess a pharmacophore capable of
undergoing oxidation by CYPs to generate the ultrapotent
cytotoxin. Essentially, site specific oxidation restores the
hydroxyl group as one route to the spirocyclisation mechanism
necessary for DNA alkylation concomitant with potent cell
killing (Fig. 1a).
The cytochromes P450 (CYPs) are a superfamily of mixed
function oxidases of which CYP1-4 subfamily members are
unique in their ability to oxidise xenobiotics including drugs.
There is now growing evidence that CYP1A1, 1B1, 2J2, 2S1
and 2W1 are over-expressed in many human tumour types.1
The presence of certain CYPs may reflect a resistance mechanism
by diminishing the pharmacological activity of anticancer
drugs whilst specific CYPs can also modulate cell proliferation
by the formation or conversion of endogenous signalling
molecules.2 The potential for CYP-selective metabolism of
xenobiotics coupled to their broad substrate specificity provides
a unique opportunity to design drugs whose activity is dependent
on a critical functional group that can be unmasked or restored
by CYP oxidation selectively in tumour tissue. Several clinically
used cancer chemotherapeutics, notably the DNA alkylating
oxazaphosphorines and nitrosoureas are known to be metabolised
by CYPs to reactive intermediates primarily in the liver.3
We have synthesised a chloromethylindoline (ICT2700)
and investigated the potential for this agent as a candidate
for the development of ultrapotent cytotoxins that are inactive
until CYP-activated in tumour tissue. This communication is
the first demonstration of a pharmacophore based on the
duocarmycin scaffold whose biological activity is dependent
on metabolic activation by CYP1A1.
The target compound ICT2700 was synthesised using de-OH
starting materials and procedures similar to those previously
described for the synthesis of seco-CPI-MI (ICT2740).6 The
route to preparation of ICT2700 is outlined in Supporting
Information (Scheme S1w). In brief, the synthesis involved the
common key steps: reaction of Boc-protected nitrogen with
1,3-dichloropropene afforded a vinylchloride, which upon
treatment with Bu3SnH (or TTMSS) and AIBN provided
the chloromethylpyrroloindoline subunit via 5-exo-trig radical
cyclisation. Removal of the Boc-protection under acidic condi-
tions was followed by EDC-mediated coupling to 5-methoxy-
indole-2-carboxylic acid to provide the target compound
ICT2700.
The phenol-containing seco-compounds of the natural products
CC-1065 and the duocarmycins are recognized as ultrapotent
cytotoxins.4 Their mechanism of action involves spirocyclisation
of the deep-embedded chloromethylindoline fragment to trigger
production of an N3-adenine covalent adduct upon binding of the
minor groove of DNA. Nature’s evolution of the duocarmycins
may have been derived from ancestral precursors, which did
Given the high level of genomic instability and mutations
seen in tumours, resistance mechanisms are likely to be complex
and multifactorial, allowing the cancer cell many escape routes
to survival.7 Accordingly, a duocarmycin prodrug must not
only demonstrate conversion by CYP metabolism from an
inactive form to an ultrapotent compound but once activated
also evade drug resistant mechanisms associated with natural
product efflux and DNA repair from alkylating agents. To
explore this we investigated chloromethylpyrroloindolines
ICT2700 and ICT2740 in both wild-type and resistant A2780
ovarian cancer cell lines. The results demonstrate that ICT2740,
the seco-chloromethylpyrroloindolone, was cytotoxic to A2780
cells (IC50 o 1 nM) whereas ICT2700 was at least 1000-fold
less cytotoxic (Fig. 1c). Furthermore, it was demonstrated that
a Institute of Cancer Therapeutics, University of Bradford,
West Yorkshire BD7 1DP, UK.
E-mail: L.H.Patterson@Bradford.ac.uk
b Cancer Research UK Drug-DNA Interactions Research Group,
UCL Cancer Institute, London, WC1E 6BT, UK
c School of Pharmacy, University of East Anglia, Norwich, Norfolk
NR4 7TJ, UK
w Electronic supplementary information (ESI) available: Synthetic
methods and characterization data of ICT2700. Experimental proce-
dures and figures MRM analysis of CYP1A1, CYP1A1 bactosome
metabolism, thermal cleavage and EROD assays. See DOI: 10.1039/
c1cc15638a
c
12062 Chem. Commun., 2011, 47, 12062–12064
This journal is The Royal Society of Chemistry 2011

View Online
Fig. 1 Pathway of activation of ICT2700, DNA alkylation and chemosensitivity data. (a1) CYP1A1-mediated metabolism generates the seco-
compound ICT2740. (a2) ICT2740 spirocyclises to produce an activated cyclopropane product.5d (a3) The cyclopropapyrroloindolone binds in a
sequence-selective manner to cause DNA damage. (a4) The DNA adducts generated by CYP1A1-mediated activation causes selective cell kill. (b)
The proposed pathway of CYP1A1 activation is supported by experimental evidence (inserted autoradiograph); Drug–DNA incubations were
carried out at 37 1C for 2 h. ICT2700M is a metabolite extract from incubation of ICT2700 with CYP1A1/NADPH bactosomes for 30 min at
37 1C. (c) The cytotoxicity of compounds was measured against a panel of ovarian A2780 carcinoma as well as CHO cells lacking or expressing
CYP1A1. Cells were treated for 96 h with ICT2700 or ICT2740 in a suitable concentration range. All data represent the mean of three independent
experiments. For full experimental conditions, see Supplementary Information.
the cytotoxicity of ICT2740 was not significantly affected by
an elevated expression of the P-glycoprotein efflux pump or
defect in DNA MMR repair present in doxorubicin (A2780AD)
and cisplatin (A2780/cp70) resistant cancer cell lines, respectively.
Distribution of ICT2700 and ICT2740 into A2780 cells was
similar (Fig. 2), indicating that the high differential activity
between ICT2700 and ICT2740 is not due to poor cellular
uptake of ICT2700. Accordingly, the absence of a para-directing
phenolic hydroxyl group in the pharmacophore of ICT2700 is
consistent with an inability to undergo closure to the cyclopro-
papyrroloindolone necessary for DNA alkylation and attendant
cell damage.8
in the transfected cells but not wild-type cells (Fig. S1w).
Oxidative O-deethylation of ethoxyresorufin as a test substrate
for CYP1A1 catalytic activity confirmed that CYP1A1 was
functionally active in the CHO CYP1A1 cells but not the CHO
wild-type (Fig. S2w).
An experiment was performed to see if the observed cyto-
toxicity of ICT2700 in CYP1A1-transfected CHO cells
could be correlated with CYP1A1-mediated metabolism and
DNA damage. Incubation of ICT2700 with CYP1A1 bacto-
somes produced an extract (ICT2700M) that was analysed by
LC/MS/MS to comprise unconverted ICT2700 (85.7%),
ICT2740 (3.3%), spirocyclised cyclopropapyrroloindolone
product CPI-MI (1.7%) and two mono-hydroxylated meta-
bolites (9.1%), measured as percent total peak AUC’s (Fig. 3).
The latter two metabolites both possessed a molecular weight
consistent with retention of the de-hydroxy chloromethylindo-
line moiety and hence not considered to contribute to DNA
alkylation and cytotoxicity.
Next we investigated the effect of ICT2700 on CHO cells
stably transfected with functional CYP1A1 and showed this
compound to be cytotoxic at pM concentrations. In comparison,
ICT2700 was at least 5 orders of magnitude less potent in
wild-type CHO cells in which CYP1A1 was not expressed
(Fig. 1c). In the absence of a specific CYP1A1 antibody, we
used MRM analysis to show CYP1A1 protein was expressed
Next we probed purine-N3 alkylation with the incubate
ICT2700M. Using a thermal cleavage assay,9 an autoradiograph
of a known DNA sequence revealed a prominent ICT2700-
induced thermal cleavage site around a cluster of adenines
within the sequence 86950-AAAAA-30
865
(Fig. 1b) and minor
alkylation sites at base pairs 764 and 772 (Fig. S3w). These
adenine-rich sequences are reported as a consensus binding
regions of both adozelesin10 and (+)-CC-1065.11 In line with
our hypothesis, ICT2700M was also shown to alkylate these
adenine sites albeit at a higher concentration than the intrinsically
active seco-compound ICT2740. In contrast, ICT2700 was not
observed to alkylate DNA under the conditions investigated,
consistent with deactivation of the duocarmycin pharmacophore
as a consequence of the removal of OH group in the embedded
chloromethylindoline fragment.
Fig. 2 Concentrations of ICT2700 or ICT2740 in A2780 cells or
tissue culture medium (TCM) following 15 min incubation of 10 mM
or 100 mM of ICT2700 or ICT2740 with 3 ꢀ 106 cells in 500 mL cell
culture medium.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 12062–12064 12063

View Online
Notes and references
1 (a) G. I. Murray, W. T. Melvin, W. F. Greenlee and M. D. Burke,
Annu. Rev. Pharmacol., 2001, 41, 297–316; (b) H. Chang, J. M. Su,
C. C. Huang, L. C. Liu, C. H. Tsai, M. C. Chou and P. Lin, Cancer
Detect. Prev., 2005, 29, 562–569; (c) P. Gibson, J. H. Gill, P. A. Khan,
J. M. Seargent, S. W. Martin, P. A. Batman, J. Griffith, C. Bradley,
J. A. Double, M. C. Bibby and P. M. Loadman, Mol. Cancer Ther.,
2003, 2, 527–534; (d) P. Lin, H. Chang, W. L. Ho, M. H. Wu and
J. M. Su, Lung Cancer, 2003, 42, 255–261; (e) M. C. McFadyen,
M. E. Cruickshank, I. D. Miller, H. L. McLeod, W. T. Melvin,
N. E. Haites, D. Parkin and G. I. Murray, Br. J. Cancer, 2001, 85,
242–246; (f) M. C. McFadyen, W. T. Melvin and G. I. Murray, Br. J.
Cancer, 2004, 91, 966–971; (g) S. Haas, C. Pierl, V. Harth, B. Pesch,
S. Rabstein, T. Bruning, Y. Ko, U. Hamann, C. Justenhoven,
H. Brauch and H. P. Fischer, Int. J. Cancer, 2006, 119, 1785–1791;
(h) J. G. Jiang, C. L. Chen, J. W. Card, S. Yang, J. X. Chen,
X. N. Fu, Y. G. Ning, X. Xiao, D. C. Zeldin and D. W. Wang,
Cancer Res., 2005, 65, 4707–4715; (i) M. Kumarakulasingham,
P. H. Rooney, S. R. Dun-das, C. Telfer, W. T. Melvin, S. Curran
and G. I. Murray, Clin. Cancer Res., 2005, 11, 3758–3765;
(j) S. T. Saarikoski, H. A. Wikman, G. Smith, C. H. Wolff and
K. Husgafvel-Pursiainen, J. Histochem. Cytochem., 2005, 53,
549–556; (k) M. Karlgren, A. Gomez, K. Stark, J. Svard,
C. Rodriguez-Antona, E. Oliw, M. L. Bernal, S. Ramon y Cajal,
I. Johansson and M. Ingelman-Sundberg, Biochem. Biophys. Res.
Commun., 2006, 341, 451–458; (l) D. Downie, M. C. McFadyen,
P. H. Rooney, M. E. Cruickshank, D. E. Parkin, I. D. Miller,
C. Telfer, W. T. Melvin and G. I. Murray, Clin. Cancer Res., 2005,
11, 7369–7375; (m) M. A. Rieger, R. Ebner, D. R. Bell, A. Kiessling,
J. Rohayem, M. Schmitz, A. Temme, E. P. Rieber and B. Weigle,
Cancer Res., 2004, 64, 2357–2364.
Fig. 3 Metabolite identification from CYP1A1 bactosomes incubation.
A typical chromatograph demonstrating the appearance of hydroxylated
metabolite ICT2740 (m/z 396.2) and its spirocyclised toxic metabolite
(m/z 360.1) following a 30 min incubation of ICT2700 (m/z 380.1) with
CYP1A1/NADPH bactosomes. Products A and B are mono-hydroxyl-
ated products, which m/z indicate intact chloromethyl fragment with no
loss to spirocyclisation.
In summary, synthetic modifications of seco-duocarmycins
have been the focus of addressing the unmet clinical potential
of the (+)-CC-1065 and duocarmycin family of agents, which
have failed to show an acceptable therapeutic index in clinical
trials.4 Previous studies have focused on inactivating the
alkylating pharmacophore of the seco-duocarmycins with
electron-withdrawing groups including carbamate derivatisation
at the phenolic hydroxyl group or by substitution with func-
tional groups with potential for bioreduction.4,11,12 Although
good differential activity and improved pharmacokinetics in
pre-clinical models have been observed, the clinical trials with
carzelesin and KW-2189 have not corroborated the potential
benefit of carbamate prodrug design.4
2 (a) M. C. McFadyen, W. T. Melvin and G. I. Murray, Mol. Cancer
Ther., 2004, 3, 363–371; (b) C. Rodriguez-Antona and
M. Ingelman-Sundberg, Oncogene, 2006, 25, 1679–91.
3 (a) L. Yu and D. J. Waxman, Drug Metab. Dispos., 1996, 24,
1254–1262; (b) L. H. Patterson and G. I. Murray, Curr. Pharm.
Des., 2002, 8, 1335–1347.
4 (a) D. L. Boger and D. S. Johnson, Angew. Chem., Int. Ed. Engl.,
1996, 35, 1438–1474; (b) S. E. Wolkenburg and D. L. Boger, Chem.
Rev., 2002, 102, 2477; (c) N. Ghosh, H. M. Sheldrake, M. Searcey
and K. Pors, Curr. Top. Med. Chem., 2009, 9, 1494–1524.
5 K. S. MacMillan and D. L. Boger, J. Am. Chem. Soc., 2008, 130,
16521–16523.
6 (a) D. L. Boger, R. J. Wysocki and T. Ishizaki, J. Am. Chem. Soc.,
1990, 112, 5230–5240; (b) D. L. Boger, W. Y. Yun and
B. R. Teegarden, J. Org. Chem., 1992, 57, 2873–2876;
(c) D. L. Boger, C. W. Boyce, R. M. Garbaccio and M. Searcey,
Tetrahedron Lett., 1998, 39, 2227–2230; (d) D. L. Boger,
A. Santillan, M. Searcey, S. R. Brunette, S. E. Wolkenberg,
M. P. Hedrick and Q. Jin, J. Org. Chem., 2000, 65, 4101–4111;
(e) D. L. Boger, C. W. Boyce, R. M. Garbaccio and
J. A. Goldberg, Chem. Rev., 1997, 97, 787–828.
7 R. D. Baird and S. B. Kaye, Eur. J. Cancer, 2003, 39, 2450–61.
8 M. Ichimura, T. Ogawa, K. Takahashi, A. Mihara, I. Takhashi
and H. Nakano, Oncol. Res., 1993, 5, 165–171.
9 K. Kiakos, A. Sato, T. Asao, P. J. McHugh, M. Lee and
J. A. Hartley, Mol. Cancer Ther., 2007, 6, 2708–2718.
10 L. H. Hurley, V. L. Reynolds, D. H. Swenson, G. L. Petzold and
T. A. Scahill, Science, 1984, 226, 843–844.
11 J. P. Lajiness, W. M. Robertson, I. Dunwiddie, M. A. Broward,
G. A. Vielhauer, S. J. Weir and D. L. Boger, J. Med. Chem., 2010,
53, 7731–7738.
Our strategy is to harness the unique capability of selected
CYPs to affect regioselective aryl oxidation in order to generate
an ultrapotent and non-reversible active (hydroxyl) metabolite.
Towards this aim we have identified a de-activated pharmaco-
phore based on the duocarmycins with potential to undergo
tissue specific oxidative activation in which CYP1A1 is expressed.
The importance of this study lies in the potential to use CYP-
mediated activation of agents based on the ultrapotent cyto-
toxic duocarmycins to address their intrinsic toxicity with the
promise of overcoming the lack of therapeutic index observed
in clinical studies.
This work was supported by Yorkshire Cancer Research
(grant B207 and Programme Grant respectively). M.S and
H.M.S thank Heptagon and EPSRC (RCUK Academic
Fellowship) respectively and J.A.H. thanks Cancer Research
UK (Grant C2259/A9994). We are grateful to the late
Dr Thomas Friedberg for CHO CYP1A1 cells. We thank
Mrs P. A. Cooper, Ms N. Harris, Mr D. Healey for technical
assistance, the EPSRC NMSSC for HRMS measurements and
acknowledge Dr J. H. Gill for valuable discussions.
12 M. Tercel, G. J. Atwell, S. Yang, A. Ashoorzadeh, R. J. Stevenson,
K. J. Botting, Y. Gu, S. Y. Mehta, W. A. Denny, W. J. Wilson and
F. B. Prujin, Angew. Chem., Int. Ed., 2011, 50, 2606–2609.
c
12064 Chem. Commun., 2011, 47, 12062–12064
This journal is The Royal Society of Chemistry 2011