Azinomycin bisepoxides containing rigid aromatic linkers: synthesis, cytotoxicity and DNA interstrand cross-linking activity

Article abstract of DOI:10.1016/j.tetlet.2009.03.123

The synthesis of a series of azinomycin bisepoxides containing rigid linkers is achieved through two different strategies. Double copper-mediated C-N bond formation under Buchwald-type conditions can be realised but yields are poor with fully intact epoxi

Full text of DOI:10.1016/j.tetlet.2009.03.123

Tetrahedron Letters 50 (2009) 3648–3650
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Azinomycin bisepoxides containing rigid aromatic linkers: synthesis,
cytotoxicity and DNA interstrand cross-linking activity
Matthew J. Finerty a, John P. Bingham b, John A. Hartley b, Michael Shipman a,
*
a Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, UK
b CRUK Drug-DNA Interactions Research Group, UCL Cancer Institute, Paul O’Gorman Building, University College London, 72 Huntley Street, London WC1E 6BT, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of a series of azinomycin bisepoxides containing rigid linkers is achieved through two dif-
ferent strategies. Double copper-mediated C–N bond formation under Buchwald-type conditions can be
realised but yields are poor with fully intact epoxide substrates. The bisepoxides are made in improved
yields through the simultaneous formation of two amide bonds. Bioassays reveal that 5, containing a 1,3-
diaminobenzene linker, is a potent in vitro DNA interstrand cross-linking agent with significant cytotox-
Received 15 January 2009
Revised 9 March 2009
Accepted 16 March 2009
Available online 21 March 2009
icity in the NCI 60-human cancer cell panel (GI50 = 0.15 lM).
Ó 2009 Elsevier Ltd. All rights reserved.
DNA interstrand cross-linking (ISC) agents such as cisplatin,
chlorambucil, and melphalan represent an extremely important
class of clinical cancer chemotherapeutics.1 Currently considerable
effort is focused on the identification of new medicines that oper-
ate by this mode of action with improved therapeutic profiles.2
Azinomycins A (1) and B (2) are structurally unique natural prod-
ucts that possess significant in vivo antitumour activity and appear
to act by disruption of cellular DNA replication by ISC formation
(Fig. 1).3,4 Unfortunately, the 1-azabicyclo[3.1.0]hexane core of
these compounds is unstable, making them unsuitable for develop-
ment as potential anti-cancer therapeutics.
In considering routes to bisepoxides 4–6, we were attracted to
the idea of effecting two concomitant Cu-mediated C–N bond-
forming reactions between epoxyamide subunits and 1,2- 1,3- or
1,4-dihalobenzenes.7,8 At the outset of this work, we were con-
cerned about the compatibility of the electrophilic epoxide ring
and with the conditions commonly employed for amide N-aryla-
tion.8 Nevertheless, we felt this approach merited investigation
as if successful, the wide availability of aromatic and heteroaro-
matic dihalides would allow large numbers of bisepoxides to be
generated rapidly and screened in a highly divergent manner. To
explore this strategy, (2S,3S)-7 was prepared according to pub-
lished methods,9 and reacted with 0.5 equiv of 1,4-diiodobenzene
Previously, we have demonstrated that dimeric structures
based upon the epoxide domain of the azinomycins are highly po-
tent ISC agents, that display much improved chemical stability.5,6
For example, bisepoxide
3 induces appreciable amounts of
O
O
O
N
H
N
in vitro DNA ISC in the low nM range, and displays good cytotoxic-
ity in the NCI 60 human cancer cell panel (GI50 = 31 nM) (Fig. 1).5b
Thus far, all the bisepoxides studied have incorporated flexible
linkers between the epoxide subunits. We reasoned that by reduc-
ing the number of rotatable bonds within the linker, we might re-
duce the entropic cost of binding to the DNA backbone, and hence
increase potency. To test this hypothesis, we targeted the synthesis
of bisepoxides 4–6 based upon a simple disubstituted aromatic nu-
cleus. The precise distance and spatial disposition of the reacting
epoxide centres being fine tuned by careful choice of the substitu-
tion pattern around the aromatic ring. In this Letter, we describe
two different synthetic approaches to this new compound class,
and preliminary biological findings with these more rigid
derivatives.
MeO
X
1 azinomycin A (X = CH2);
O
N
H
2 azinomycin B (X = C=CHOH)
O
O
AcO
Y
Me
HO
O
O
O
O
H
Linker
X
N
N
H
X
O
O
O
O
bisepoxide
Linker
X
Y
Y
3
4a
4b
Me
OMe
H
Me
H
OMe
5
Me
OMe
OMe
6
Me
* Corresponding author. Tel.: +44 247 652 3186; fax: +44 247 652 4112.
E-mail address: m.shipman@warwick.ac.uk (M. Shipman).
Figure 1.
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.03.123

M. J. Finerty et al. / Tetrahedron Letters 50 (2009) 3648–3650
3649
Table 1
O
O
O
Synthesis of bisepoxides 4–6 by amide bond formation
O
O
H
N
i
NH2
Methoda
Product
Yield (%)
50b
Ar
O
O
Ar
O
Entry
1
Diamine
Epoxide
10%
O
Ar
O
N
H
7
4a
10
A
4a
H2N
NH2
NH2
O
where Ar = 1-naphthyl
O
2
3
4
11
11
11
A
A
B
4b
5
59
34
46
H2N
H2N
H
N
i
HO
O
NH2
HO
NH2
63%
O
OH
N
O
H
8
9
H2N
H2N
NH2
5
Scheme 1. Reagents and conditions: 1,4-diiodobenzene (0.5 equiv), CuI (1 equiv),
DMEDA (2 equiv), K2CO3 (2 equiv), K3PO4 (2 equiv), dioxane, 50 °C, 22 h.
NH2
5
11
A
6
0c
in the presence of stoichiometric amounts of CuI and N,N0-di-
methyl-ethylenediamine (DMEDA) (Scheme 1). Bisepoxide 4a
was isolated in 10% yield as a single diastereomer from this reac-
tion. However, the efficiency of this coupling was very poor with
no starting amide or monoamidated products recovered, suggest-
ing that 7 is incompatible with the reaction conditions. Evidence
in support of this supposition was obtained by reacting amide
(2S)-8,10 devoid of the epoxide ring, under identical conditions.
In this instance, bisarylation proceeded smoothly providing 9 in a
much improved yield (Scheme 1). Although bisamide 9 might be
converted into 4a by stereocontrolled epoxidation and acylation,
the need for additional synthetic steps discouraged us from further
pursuing N-arylation as a route to these compounds.
a
Method A: EDCI, HOBt, DMF, 0 °C?rt, 20 h; method B: HATU, DIEA, DMF, 24 h,
rt.
b
0.5 equiv of diamine used.
c
Only formation of monoepoxide detected by ES–MS.
is similar when the data from mean cytotoxicities (LC50
)
(3 > 5 ꢀ 4b > 4a) or total growth inhibitions (TGI) (3 > 5 > 4b > 4a)
are compared.
The DNA interstrand cross-linking activities of these bisepox-
ides were determined with an agarose gel assay using the
pBR322 linearised plasmid.12 The P 50-end labelled duplex was
32
incubated with 3 (0.01–1000 nM) for 1 h at 37 °C prior to denatur-
ing with alkali and subsequent gel electrophoresis. The extent of
cross-linking was determined by quantifying the relative amounts
of double-stranded and single-stranded DNA by storage phosphor-
image analysis. The efficiency of ISC formation follows the order
3 > 5 > 4b > 4a (Table 2). This trend directly correlates with the or-
der of in vitro cytotoxicity suggesting that the biological activity of
these agents is related to their ability to form DNA ISC. In all cases,
the extent of DNA ISC formation was shown to be dependent on
bisepoxide concentration with increasing ISC at higher drug con-
centrations (data not shown).
Several features of the biological data merit further comment. It
is clear that use of these aromatic linkers reduces potency in com-
parison to related systems based on flexible hydrocarbon linkers (5
vs 3). This suggests that the most accessible conformations of 4 and
5 are not optimal for reaction with the nucleophilic N-7 guanine
atoms on the complementary DNA strands which most likely pro-
duce the ISC.5 The observation that 5 is more active than 4b is con-
As an alternative strategy, we explored coupling of o-, m- or p-
diaminobenzene with the appropriately substituted epoxy acid.
Benzyl esters 10 and 11 were prepared as their (2S,3S)-enantio-
mers using published methods.9 Hydrogenolysis and coupling with
0.5 equiv of the appropriate diaminobenzene using either EDCI/
HOBt or HATU as activating agent effected amide bond formation
(Scheme 2 and Table 1). Using this approach, bisepoxide 4a could
be produced in much improved yield (50% cf 10% in Scheme 1).
Other rigid bisepoxides, namely 4b and 5, were prepared using
the same approach.11 One limitation was noted, in that sterically
congested 6 could not be formed from 1,2-diaminobenzene with
only monoacylation being observed.
The new bisepoxides were evaluated in vitro against the NCI 60
human tumour cell lines derived from nine cancer types (leukae-
mia, melanoma and cancers of the lung, colon, brain, ovary, breast,
prostate, and kidney). All of them exhibited appreciable cytotoxic-
ity. Mean values for GI50, LC50 and TGI for bisepoxides 4a, b and 5
are given in Table 2, together with those obtained previously for
3.5b The relative order of anticancer activities of these dimers is
3 > 5 > 4b > 4a as deduced from the mean values of 50% growth
inhibition (GI50) taken across all 60 cell lines. This order of potency
sistent with earlier observations that
a three carbon linker
between the epoxide units is optimal for biological activity.5b
Interestingly, bisepoxide 4b displayed significantly higher cytotox-
icity and ISC activity than its counterpart 4a lacking the MeO and
Me substituents on the naphthalene ring.13
In conclusion, we have explored the synthesis of a series of azi-
nomycin-like bisepoxides containing aromatic linkers using two
O
O
X
OBn
O
10 (X, Y = H);
O
11 (X = OMe, Y = Me)
Table 2
Y
Cytotoxicity and DNA ISC activity data
i. H2, 10% Pd/C, EtOAc, 1.5-3 h
ii. Diamine (0.3 eq), DMF
Bisepoxide
GI50
(
lM)
Cytotoxicitya
TGI (
lM)
DNA ISC
method A or B (see Table 2).
LC50
(
lM)
activityb (%)
O
4a
4b
5
4.4
0.39
0.15
74
62
62
3.2
25
1.7
28
56
O
O
O
H
H
8.7
3.4
0.21
X
N
Ar
N
X
O
O
3c
0.031
100
O
O
a
Measured against the NCI panel of 60 cancer cell lines.
At 100 nM. See text for details.
Data taken from Ref. 5b.
Y
Y
b
c
Scheme 2.

3650
M. J. Finerty et al. / Tetrahedron Letters 50 (2009) 3648–3650
5. (a) Hartley, J. A.; Hazrati, A.; Hodgkinson, T. J.; Kelland, L. R.; Khanim, R.;
Shipman, M.; Suzenet, F. Chem. Commun. 2000, 2325–2326; (b) LePla, R. C.;
Landreau, C. A. S.; Shipman, M.; Hartley, J. A.; Jones, G. D. D. Bioorg. Med. Chem.
Lett. 2005, 15, 2861–2864.
6. For an alternative approach, see: Casely-Hayford, M. A.; Pors, K.; James, C. H.;
Patterson, L. H.; Hartley, J. A.; Searcey, M. Org. Biomol. Chem. 2005, 3, 3585–
3589.
7. Strieter, E. R.; Bhayana, B.; Buchwald, S. L. J. Am. Chem. Soc. 2009, 131, 78–88
and references cited therein.
8. Bisamidation of 1,4-diiodobenzene under Cu catalysis has been observed, see:
Toto, P.; Gesquière, J.-C.; Deprez, B.; Willand, N. Tetrahedron Lett. 2006, 47,
1181–1186.
alternative strategies. The best yields were obtained by coupling
1,3- or 1,4-diaminobenzene with the requisite epoxy acids using
EDCI/HOBt or HATU as activators. The new bisepoxides displayed
good ISC efficiencies and appreciable in vitro cytotoxicities
although they were less potent than structurally related com-
pounds containing flexible hydrocarbon linkers. Future work will
focus on identifying the origin of these differences in potency,
and on designing new anti-cancer agents based upon this motif.
Acknowledgements
9. Bryant, H. J.; Dardonville, C. Y.; Hodgkinson, T. J.; Hursthouse, M. B.; Malik, K.
M. A.; Shipman, M. J. Chem. Soc., Perkin Trans.
1 1998, 1249–1255 and
references cited therein.
This work was supported financially by the University of War-
wick and CRUK (C2259/A3083). We thank the EPSRC National Mass
Spectrometry Service Centre for mass measurements, EPSRC (EP/
C007999/1) for the provision of additional mass spectrometers
and the NCI anti-cancer screening programme for the in vitro cyto-
toxicity data.
10. Amide (2S)-8 was prepared from (2S)-benzyl 2-hydroxy-3-methylbut-3-
enoate (Ref. 9) by heating in a sealed tube with 7 M ammonia in methanol
(40 °C, 16 h, 79%).
11. Selected spectroscopic data for 4a: mp = 144–144.5 °C. [
a]D +3.8 (c 1.1, CHCl3); IR
(film) 3338, 2974, 1693, 1611, 1543, 1127 cmꢁ1 1H NMR (400 MHz; CDCl3)
;
8.94 (2H, d, J 8.5 Hz, ArH), 8.36 (2H, dd, J 7.3, 1.2 Hz, ArH), 8.08 (2H, d, J 8.0 Hz,
ArH), 7.99 (2H, s, NH), 7.91 (2H, d, J 8.0 Hz, ArH), 7.64 (2H, m, ArH), 7.58–7.52
(8H, m, ArH), 5.33 (2H, s, H-2), 3.10 (2H, d, J 4.5 Hz, H-4), 2.87 (2H, d, J 4.5 Hz,
H-4), 1.60 (6H, s, CH3); 13C NMR (100 MHz; CDCl3) 165.9 (C), 164.6 (C), 134.2
(CH), 133.8 (C), 133.7 (C), 131.4 (C), 130.8 (CH), 128.6 (CH), 128.2 (CH), 126.5
(CH), 125.68 (C), 125.58 (CH), 124.5 (CH), 120.7 (CH), 76.2 (C-2), 56.2 (C-3),
53.7 (C-4), 17.6 (CH3). HRMS (ES+) calculated for C38H36N3O8 requires 662.2497
References and notes
1. For reviews, see: (a) Noll, D. M.; McGregor Mason, T.; Miller, P. S. Chem. Rev.
2006, 106, 277–301; (b) Schärer, O. D. ChemBioChem 2005, 6, 27–32; (c) Rajski,
S. R.; Williams, R. M. Chem. Rev. 1998, 98, 2723–2795.
[M+NH4þ]; found 662.2508; 4b: mp = 157.5–160.5 °C.
[
a
]
D = +33.8 (c 1.1,
CHCl3); IR (film) 3337, 2976, 1694, 1611, 1543, 1191 cmꢁ1
;
1H NMR (400 MHz;
CDCl3) 8.61 (2H, m, ArCH), 8.00 (2H, s, NH), 7.96 (2H, d, J 2.8 Hz, ArCH), 7.49–
7.48 (6H, m, ArCH), 7.36–7.34 (4H, m, ArCH), 5.32 (2H, s, H-2), 3.98 (6H, s,
ArOCH3), 3.08 (2H, d, J 4.4 Hz, H-4), 2.85 (2H, d, J 4.4 Hz, H-4), 2.67 (6H, s,
ArCH3), 1.59 (6H, s, CH3); 13C NMR (100 MHz; CDCl3) 165.7 (C), 164.5 (C), 155.8
(C), 134.3 (C), 133.7 (C), 133.2 (C), 128.0 (C), 127.8 (CH), 126.8 (C), 125.2 (CH),
123.7 (CH), 122.1 (CH), 120.6 (CH), 108.4 (CH), 76.3 (C-2), 56.2 (C-3), 55.5
(CH3), 53.5 (C-4), 20.1 (CH3), 17.7 (CH3). HRMS (ES+) calculated for C42H44N3O10
requires 750.3021 [M+NH4þ]; found 750.3019; 5: mp = 131–133 °C.
2. For recent illustrative examples, see: (a) Song, Z.; Weng, X.; Weng, L.; Huang, J.;
Wang, X.; Bai, M.; Zhou, Y.; Yang, G.; Zhou, X. Chem. Eur. J. 2008, 14, 5751–
5754; (b) Tiberghien, A. C.; Evans, D. A.; Kiakos, K.; Martin, C. R. H.; Hartley, J.
A.; Thurston, D. E.; Howard, P. W. Bioorg. Med. Chem. Lett. 2008, 18, 2073–2077;
(c) Duan, J.-X.; Jiao, H.; Kaizerman, J.; Stanton, T.; Evans, J. W.; Lan, L.; Lorente,
G.; Banica, M.; Jung, D.; Wang, J.; Ma, H.; Li, X.; Yang, Z.; Hoffman, R. M.;
Ammons, W. S.; Hart, C. P.; Matteucci, M. J. Med. Chem. 2008, 51, 2412–2420;
(d) Weng, X.; Ren, L.; Weng, L.; Huang, J.; Zhu, S.; Zhou, X.; Weng, L. Angew.
Chem., Int. Ed. 2007, 46, 8020–8023; (e) Woo, S.; Jung, J.; Lee, C.; Kwon, Y.; Na, Y.
Bioorg. Med. Chem. Lett. 2007, 17, 1163–1166.
3. For reviews, see: Hodgkinson, T. J.; Shipman, M. Tetrahedron 2001, 57, 4467–
4488; Casely-Hayford, M.; Searcey, M.. In Small Molecule DNA and RNA Binders;
Demeunynck, M., Bailly, C., Wilson, W. D., Eds.; Wiley-VCH: Weinheim, 2003;
Vol. 1, pp 676–696.
4. (a) Lown, J. W.; Majumdar, K. C. Can. J. Biochem. 1977, 55, 630–635; (b)
Armstrong, R. W.; Salvati, M. E.; Nguyen, M. J. Am. Chem. Soc. 1992, 114, 3144–
3145; (c) Fujiwara, T.; Saito, I.; Sugiyama, H. Tetrahedron Lett. 1999, 40, 315–
318; (d) Hartley, J. A.; Hazrati, A.; Kelland, L. R.; Khanim, R.; Shipman, M.;
Suzenet, F.; Walker, L. F. Angew. Chem., Int. Ed. 2000, 39, 3467–3470; (e)
Coleman, R. S.; Perez, R. J.; Burk, C. H.; Navarro, A. J. Am. Chem. Soc. 2002, 124,
13008–13017; (f) LePla, R. C.; Landreau, C. A. S.; Shipman, M.; Jones, G. D. D.
Org. Biomol. Chem. 2005, 3, 1174–1175; (g) Kelly, G. T.; Liu, C.; Smith, R.;
Coleman, R. S.; Watanabe, C. M. H. Chem. Biol. 2006, 13, 485–492.
[a] ;
D = +56.4 (c 1.2, CHCl3); IR (film) 3349, 2934, 1684, 1614, 1540 cmꢁ1 1H
NMR (400 MHz; CDCl3) 8.60 (2H, m, ArH), 8.13 (2H, s, NH), 7.94 (2H, d, J 2.5 Hz,
ArCH), 7.80 (1H, s, ArCH), 7.44 (2H, d, J 2.5 Hz, ArCH), 7.38–7.33 (6H, m,
6 ꢂ ArCH), 7.23–7.19 (1H, m, ArCH), 5.30 (2H, s, H-2), 3.95 (6H, s, ArOCH3), 3.09
(2H, d, J 4.4 Hz, H-4), 2.84 (2H, d, J 4.4 Hz, H-4), 2.66 (6H, s, ArCH3), 1.58 (6H, s,
CH3); 13C NMR (100 MHz; CDCl3) 165.7 (C), 164.7 (C), 155.7 (C), 137.6 (C),
134.3 (C), 133.1 (C), 129.5 (CH), 127.9 (C), 127.7 (CH), 126.8 (C), 125.2 (CH),
123.7 (CH), 122.1 (CH), 116.2 (CH), 111.5 (CH), 108.5 (CH), 76.3 (C-2), 56.2 (C-
3), 55.5 (CH3), 53.4 (C-4), 20.0 (CH3), 17.7 (CH3). HRMS (ES+) calculated for
C42H44N3O10 requires 750.3021 [M+NH4þ]; found 750.3025.
12. Hartley, J. A.; Berardini, M. D.; Souhami, R. L. Anal. Biochem. 1991, 193, 131–
134.
13. The importance of the naphthalene MeO– and Me– groups has been noted
previously, see: (a) Landreau, C. A. S.; LePla, R. C.; Shipman, M.; Slawin, A. M.;
Hartley, J. A. Org. Lett. 2004, 6, 3505–3509; (b) Ref. 4f.