Synthesis and biological evaluation of pyrrolo[2,1-c][1,4]benzodiazepine (PBD) C8 cyclic amine conjugates

Article abstract of DOI:10.1016/j.bmcl.2003.12.017

We report examples of a series of novel pyrrolo[2,1-c][1,4]benzodiazepine (PBD) analogues 12-15 prepared from a common functionalized building block 11 that can be conveniently synthesized on a large scale and in optically pure form. Isoindoline analogue

Full text of DOI:10.1016/j.bmcl.2003.12.017

Bioorganic & Medicinal Chemistry Letters 14 (2004) 901–904
Synthesis and biological evaluation of
pyrrolo[2,1-c][1,4]benzodiazepine (PBD) C8
cyclic amine conjugates
Luke A. Masterson,a Stephen J. Croker,a Terence C. Jenkins,b
Philip W. Howarda,* and David E. Thurstona,*
aCancer Research UK Gene Targeted Drug Design Research Group, The School of Pharmacy,
University of London, 29/39 Brunswick Square, London WC1 1AX, UK
bYorkshire Cancer Research Laboratory of Drug Design, Tom Connors Cancer Research Centre,
University of Bradford, All Saints Road, Bradford, West Yorkshire BD7 1DP, UK
Received 6 August 2003; revised 31 October 2003; accepted 3 December 2003
Abstract—We report examples of a series of novel pyrrolo[2,1-c][1,4]benzodiazepine (PBD) analogues 12–15 prepared from a
common functionalized building block 11 that can be conveniently synthesized on a large scale and in optically pure form.
Isoindoline analogue 15 is the most cytotoxic agent in this series, has the highest DNA-binding affinity, and shows significant
activity in the in vivo hollow fibre assay.
# 2004 Elsevier Ltd. All rights reserved.
The pyrrolo[2,1-c][1,4]benzodiazepines (PBDs, e.g., 1a–c
in Fig. 1) are a group of antitumour agents which includes
the natural products anthramycin and DC-81 (1a).1 They
exert their cytotoxic effect by covalently bonding to the
exocyclic C2-NH2 of guanine residues within the minor
groove of DNA through their N10–C11 imine (or
equivalent carbinolamine) functionality1. Synthetic PBD
derivatives (e.g., 1b–c) are of interest as potential anti-
cancer drugs and also as gene targeting agents because
they bind to guanine bases in duplex DNA in a sequence-
selective manner with a preference for 50-Pu-G-Pu
motifs.1ꢀ3 Recent research in this area has focused on
improving and extending this sequence selectivity.4 Var-
ious approaches have been explored, including the cou-
pl0ing of two PBD units through A-C8/A0-C805ꢀ7 or C-C2/
A -C808 linkages, or by conjugation of PBDs to secondary
residues that possess complementary DNA sequence
selectivity such as cyclopropylbenzindole9,10 (e.g., CBI,
2 in Fig. 1) or distamycin/netropsin moieties.11ꢀ13
Figure 1. Generic structures of a pyrrolobenzodiazepine (PBD, 1a–c)
and cyclopropylbenzindole (CBI, 2).
has a carboxylate-terminated substituent at C8 and a
protecting group at N10 that can be removed at the
final synthetic stage to generate the DNA-interactive
N10–C11 imine/carbinolamine moiety. Previously-
reported building blocks have been prepared from
intermediates of type 10 that contain Troc ester/Alloc,9
methyl ester/Troc11,12 and methyl ester/Fmoc10 combi-
nations at the C8/N10-positions, respectively. We report
here a synthesis of building block 11 from intermediate
10 that utilizes benzyl and Boc protecting groups at C8/
N10, respectively. With this combination of protecting
groups, 11 can be produced efficiently on a large scale
using simple procedures and without recourse to air-
sensitive reagents. More importantly, stereochemistry is
maintained at the C11a position of the PBD.
For studies involving coupling to an amine or alcohol, a
building block such as 11 (Scheme 1) is required which
Keywords: DNA Binding; Pyrrolobenzodiazepines; PBDs; Sequence-
selective; Gene targeting.
* Corresponding authors. Tel.: +44-207-753-5932; fax: +44-207-753-
5935; e-mail: david.thurston@ulsop.ac.uk; phil.howard@ulsop.ac.uk
0960-894X/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.12.017

902
L. A. Masterson et al. / Bioorg. Med. Chem. Lett. 14 (2004) 901–904
Scheme 1. (a) Br(CH2)3OH, NaOH(aq), reflux, 5 h, 71%; (b) 70% HNO3, 94%; (c) PhCH2OH, cat. p-TsOH, toluene, 66%; (d) (i) (COCl)2/DMF-
CH2Cl2, 18 h; (ii) (2S)-(+)-pyrrolidinemethanol, Et3N, ꢀ50 ꢁC to rt, 18 h, 98% or EDCI/DMAP/DMF, (2S)-(+)-pyrrolidinemethanol, 55%; (e)
.
SnCl2 2H2O/MeOH, reflux, 1 h, 94%; (f) (Boc)2O, THF, reflux, 18 h, used without further purification; (g) pyridinium dichromate/CH2Cl2, 4 A
sieves, 1 h, 60% or DMSO, (COCl)2, Et3N, ꢀ40 ꢁC, 66%; (h) 10% Pd/C/H2, 16 psi/EtOH, 98%.
Scheme 2. (a) EDCI/DMAP/DMF, then pyrrolidine, piperidine, indoline or isoindoline; (b) 95% CF3COOH, ꢀ10 ꢁC.
We also report the use of this building block to synthe-
size four novel compounds 12–15 resulting from its
conjugation to two cyclic amines (pyrrolidine and piper-
idine) and two benzo-fused heterocycles (indoline and iso-
indoline), all of which may be considered as substructures
of the established DNA-binding CBI moiety (2, Fig. 1).
These compounds have been screened in ovarian A2780
and CH1 cell lines and their cisplatin-resistant counter-
parts (A2780cisR and CH1cisR, respectively), in SKOV-3
cells and also in the NCI 60-cell-line panel (data not
shown). The isoindoline analogue 15 was sufficiently
active to be progressed to the NCIs standard in vivo hol-
low fibre assay in which it achieved the significant total
score of 40 (20 is the minimum total score to demonstrate
antitumour activity according to NCI criteria).
coupling reaction was also effected but in lower yield
(55%) using 1-[3-(dimethylamino)propyl]-3-ethyl-carbo-
diimide hydrochloride (EDCI) and 4-(dimethyl-
amino)pyridine (DMAP).14 Reduction of the coupled
product 7 with tin(II) chloride in refluxing MeOH gave
the amine 8 as a single product, which was reacted with
di-t-butyl dicarbonate to give the Boc-protected amine
9. As an alternative to the Swern oxidation of 9 to the
benzyl ester-protected PBD (10) by the usual Fukuyama
approach,15 ring closure was achieved using pyridinium
dichromate and 4 A molecular sieves.16 This allowed the
oxidation to be carried out on a large scale (>60 g) with
relative ease, although final yields tended to be poorer
than for the Swern oxidation. After purification of 10 by
flash chromatography, removal of the benzyl ester pro-
tecting group by palladium-catalyzed hydrogenolysis
gave 11 in pure form.17 Throughout this 8-step synth-
esis, each product was obtained in excellent yield and
with only compound 10 requiring chromatographic
purification.
The key core building block 11 was synthesized by the
route shown in Scheme 1. Starting from commercially
available vanillic acid (3), reaction with 3-bromopropa-
nol gave the ether-alcohol 4. Simultaneous nitration and
oxidation with 90% nitric acid afforded the nitro di-acid
5 which, by chemoselective esterification with benzyl
alcohol and p-toluenesulphonic acid, afforded the nitro
monoester 6. Careful addition of the acid chloride gen-
erated from 6 to a solution of (2S)-(+)-pyrrolidine-
methanol and triethylamine in CH2Cl2 at ꢀ50 ꢁC gave
the coupled product 7 in excellent yield (98%). The
Building block 11 was conjugated to each of the four
amines using standard coupling methodology (EDCI/
DMAP)14 (Scheme 2). Pyrrolidine, piperidine and indo-
line were commercially available but isoindoline was
prepared from phthalimide using a literature method.18
Finally, the Boc protecting group was removed using

L. A. Masterson et al. / Bioorg. Med. Chem. Lett. 14 (2004) 901–904
903
Table 1. In vitro cytotoxicity (in human ovarian cell lines) and induced thermal stabilization of DNA melting for the novel PBD–amine conjugates
12–15 and PBDs 1b–c
Compd
Cytotoxicity (mM)a
Induced ÁTm (ꢁC)c after incubation at 37 ꢁC for
SKOV-3
A2780
A2780cisR
RFb
CH1
CH1cisR
RFb
0 h
4 h
18 h
12
13
14
15
1b
1c
5.4
23.5
5.8
1.45
1.70
0.46
1.5
3.2
1.55
0.23
0.064
0.17
4.3
14.5
4.9
0.94
0.155
0.48
2.9
4.5
3.2
4
2.4
2.8
1.4
4.9
1.5
0.24
0.082
0.145
1.85
7.9
3.0
0.42
0.11
0.145
1.3
1.6
2
2
1.3
1.0
0.5
0.8
0.7
1.0
0.2
0.2
0.9
0.9
1.0
1.1
0.4
0.4
1.2
1.2
1.2
1.3
0.6
0.4
a Concentration of agent required to inhibit cell growth by 50% compared with PBD-free controls after incubation for 96 h at 37 ꢁC.
b RF=resistance factor (IC50 cisplatin-resistant/parent) for A2780 or CH1 tumour cells.
c For a 5:1 molar ratio of duplex calf thymus (CT) DNA (100 mM in DNAp) and ligand (20 mM) in aqueous buffer (10 mM NaH2PO4/Na2HPO4+1
mM Na2EDTA, pH 7.00ꢃ0.01).7,20,21 All values areꢃ0.1 ꢁC (typicallyꢃ <0.06–0.08 ꢁC) from at least 3 determinations.
95% trifluoroacetic acid to provide the target com-
pounds 12–1519 in good yields. These molecules were
evaluated for in vitro cytotoxicity in selected ovarian cell
lines and in the NCI 60-cell-line panel (data not shown),
and for DNA duplex binding affinity using a thermal
denaturation assay with calf thymus (CT) DNA.7,20,21
Results from these assays are shown in Table 1.
In summary, the novel PBD-building block 11 reported
here can be synthesized in good yield on a large scale
(>60 g) and with maintenance of stereochemistry at the
C11a position using simple procedures with minimum
chromatography, and without recourse to air-sensitive
reagents. The potential use of 11 for coupling to amine-
or alcohol-functionalised substrates is illustrated by the
synthesis of the four cyclic amine conjugates 12–15. One
of these (15) has significant in vitro cytotoxicity and
DNA-binding affinity, and has promising antitumour
activity as revealed by the hollow fibre assay data. The
results of studies of 15 in human tumour xenograft
experiments will be reported elsewhere.
All new compounds were significantly cytotoxic in the
ovarian cell lines, with IC50 values ranging from 0.23–
23.5 mM (Table 1). Compound 15 had the consistently
lowest IC50 values of <1 mM, except in the SKOV-3 cell
line (1.45 mM). Similarly, in the NCI screen, 15 had an
average GI50 value of 2.14 mM across the 60-cell panel
and was significantly more cytotoxic than compounds
12–14 (average GI50 values=3.0, 72.4 and 19.2 mM,
respectively). This biological activity is reflected in the
thermal denaturation data (Table 1) which show that 15
develops 75% of its final helix-stabilizing effect to melt-
ing (i.e., ÁTm=1.3 ꢁC after 18 h) without prior DNA-
drug incubation at 37 ꢁC (i.e., ÁTm=1.0 ꢁC at t=0).
Compounds 12–14 effect a similar but slightly lower
ÁTm after 18 h incubation (i.e., 1.2 ꢁC), although lower
thermal stabilization is achieved at t=0 (i.e.,
ÁTm=0.5–0.8 ꢁC). One interpretation of this result is
that the isoindoline ring of 15 is connected to the PBD
moiety such that the fused rings align to be coplanar
with the A-ring of the PBD moiety. In this case the
whole conjugate should fit into the minor groove of
double-stranded DNA in a superior linear (and/or iso-
helical) fashion, thereby maximizing contact with the
DNA bases and improving covalent adduct stability.
Conversely, analogues 12 and 13 present fewer contacts
to the minor groove due to their smaller pyrrolidine and
piperidine rings. In the case of 14, the linkage to the
indoline unit should not allow a linear fit within the
DNA minor groove. Interestingly, although 15 did not
have an improved cytotoxicity profile compared to the
known C8-OMe (1b) or C8–OBn (1c) DC-81 analogues,
its relative induced ÁTm values were significantly higher
(by factors of ꢂ5-fold and 3.4-fold at t=0 and t=18 h,
respectively), supporting the notion that the isoindoline
ring confers improved contacts with the minor groove
floor and wall of the DNA host. Compound 15 also has
significant activity in the NCI standard hollow fibre
assay,22,23 with intraperitoneal (ip), subcutaneous (sc)
and total scores of 38, 2 and 40, respectively.
Acknowledgements
Cancer Research UK (C180/A1060) and Yorkshire
Cancer Research (programme grant to T.C.J.) are
thanked for financial support of this work. We
acknowledge Dr. Lloyd R. Kelland (Institute of Cancer
Research, Sutton, UK) for providing the in vitro cyto-
toxicity data. The National Cancer Institute (NCI) is
also acknowledged for providing the 60-cell line and
hollow fibre assay data.
References and notes
1. Thurston, D. E. Advances in the Study of Pyrrolo[2,1-
c][1,4]benzodiazepine (PBD) Antitumour Antibiotics. In
Molecular Aspects of Anticancer Drug-DNA Interactions
Vol 1; Neidle, S., Waring, M. J., Eds.; The Macmillan
Press Ltd.: London, 1993; p 54.
2. Reddy, B. S. P.; Damayanthi, Y.; Reddy, B. S. N.; Lown,
J. W. Anti-Cancer Drug Design 2000, 15, 225.
3. Kamal, A.; Laxman, N.; Ramesh, G.; Neelima, K.;
Kondapi, A. K. Chem. Commun. 2001, 437.
4. Thurston, D. E. Br. J. Cancer 1999, 80, 65.
5. Bose, D. S.; Thompson, A. S.; Ching, J. S.; Hartley, J. A.;
Berardini, M. D.; Jenkins, T. C.; Neidle, S.; Hurley, L. H.;
Thurston, D. E. J. Am. Chem. Soc. 1992, 114, 4939.
6. Thurston, D. E.; Bose, D. S.; Thompson, A. S.; Howard,
P. W.; Leoni, A.; Croker, S. J.; Jenkins, T. C.; Neidle, S.;
Hartley, J. A.; Hurley, L. H. J. Org. Chem. 1996, 61, 8141.
7. Gregson, S. J.; Howard, P. W.; Hartley, J. A.; Brooks, N. A.;
Adams, L. J.; Jenkins, T. C.; Kelland, L. R.; Thurston, D. E.
J. Med. Chem. 2001, 44, 737.
8. Gregson, S. J.; Howard, P. W.; Thurston, D. E. Bioorg.
Med. Chem. Lett. 2003, 13, 2277.

904
L. A. Masterson et al. / Bioorg. Med. Chem. Lett. 14 (2004) 901–904
9. Tercel, M.; Stribbling, S. M.; Sheppard, H.; Siim, B. G.;
Wu, K.; Pullen, S. M.; Botting, K. J.; Wilson, W. R.; Denny,
W. A. J. Med. Chem. 2003, 46, 2132.
64.75, 60.0, 56.1, 46.5, 34.0, 28.75, 28.2, 23.0; MS-ES:
100%); [a]D25=+103.8ꢁ (c=0.92,
437.3 ([M+H]ꢄ+
CHCl3).
,
10. Zhou, Q.; Duan, W. H.; Simmons, D.; Shayo, Y.;
Raymond, M. A.; Dorr, R. T.; Hurley, L. H. J. Am.
Chem. Soc. 2001, 123, 4865.
11. Baraldi, P. G.; Cacciari, B.; Guiotto, A.; Leoni, A.;
Romagnoli, R.; Spalluto, G.; Mongelli, N.; Howard,
P. W.; Thurston, D. E.; Bianchi, N.; Gambari, R. Bioorg.
Med. Chem. Lett. 1998, 8, 3019.
12. Baraldi, P. G.; Balboni, G.; Cacciari, B.; Guiotto, A.;
Manfredini, S.; Romagnoli, R.; Spalluto, G.; Thurston,
D. E.; Howard, P. W.; Bianchi, N.; Rutigliano, C.;
Mischiati, C.; Gambari, R. J. Med. Chem. 1999, 42, 5131.
13. Damayanthi, Y.; Reddy, B. S. P.; Lown, J. W. J. Org.
Chem. 1999, 64, 290.
18. Gawley, R. E.; Chemburkar, S. R.; Smith, A. L.;
Anklekar, T. V. J. Org. Chem. 1988, 53, 5381.
19. Data for 15: 1H NMR (CDCl3): d 7.66 (d, 2H, J=4.4 Hz),
7.48 (s, 1H), 7.30 (s, 4H), 6.89 (s, 1H), 4.48 (t, 3H, J=6.6
Hz), 3.84 (s, 3H), 3.81–3.69 (m, 2H), 3.61–3.51 (m, 1H),
2.97 (p, 5H, J=6.9 Hz), 2.32–2.28 (m, 2H), 2.30–2.01 (m,
4H); 13C NMR (CDCl3): d 169.3, 164.7, 162.5, 150.5,
147.6, 140.6, 127.9, 127.6, 123.0, 122.6, 120.4, 111.5,
110.5, 65.6, 56.1, 53.7, 52.2, 50.6, 46.7, 34.2, 29.6, 24.2;
MS (EI) m/z (relative intensity): 419 ([M]+ꢄ, 21), 416 (2),
415 (2), 246 (10), 245 (3), 231 (3), 174 (4); HRMS calcd
for C24H25O4N3=419.1845, found 419.1821.
20. Jones, G. B.; Davey, C. L.; Jenkins, T. C.; Kamal, A.;
Kneale, G. G.; Neidle, S.; Webster, G. D.; Thurston, D. E.
Anti-Cancer Drug Design 1990, 5, 249.
14. Boger, D. L.; Fink, B. E.; Hedrick, M. P. J. Am. Chem.
Soc. 2000, 122, 6382.
15. Fukuyama, T.; Liu, G.; Linton, S. D.; Lin, S. C.; Nishino,
H. Tetrahedron Lett. 1993, 34, 2577.
21. McConnaughie, A. W.; Jenkins, T. C. J. Med. Chem.
1995, 38, 3488.
16. Herscovici, J.; Antonakis, K. J. Chem. Soc., Chem.
Commun. 1980, 561.
17. Data for 11: H NMR (400 MHz, CDCl3) d 7.24 (s, 1H),
22. Hollingshead, M. G.; Alley, M. C.; Camalier, R. F.;
Abbott, B. J.; Mayo, J. G.; Malspeis, L.; Grever, M. R.
Life Sciences 1995, 57, 131.
1
6.69 (s, 1H), 5.59 (d, 1H, J=9.5 Hz), 4.31 (t, 2H, J=6.4
Hz), 3.89 (s, 3H), 3.8–3.7 (m, 1H), 3.6–3.52 (m, 1H),
3.52–3.4 (m, 1H), 2.91 (t, 2H, J=6.3 Hz), 2.2–1.95 (m,
4H), 1.38 (s, 9H); 13C NMR (100 MHz, CDCl3) d 174.5,
167.2, 149.3, 148.7, 129.2, 126.4, 115.4, 110.9, 85.6, 82.1,
23. Hollingshead, M. G.; Plowman, J.; Alley, M. C.; Mayo,
J. G.; Sausville, E. A. The Hollow Fiber Assay. Relevance
of Tumor Models for Anticancer Drug Development. In
Contrib. Oncoldate; Fiebig, H. H., Burger, A. M., Eds.;
Karger: Basal, 1999; p 109–120.