Chemical synthesis and cytotoxicity of some azinomycin analogues devoid of the 1-azabicyclo[3.1.0]hexane subunit

Article abstract of DOI:10.1016/S0960-894X(99)00663-0

A series of compounds related to the left-hand domain of the azinomycins have been made and evaluated for cytotoxic activity against a small panel of human tumour cell lines. The epoxide ring is shown to be essential for biological activity. Cytotoxicity is also shown to be sensitive to changes in the substitution pattern on the aromatic ring and the amide group. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(99)00663-0

Bioorganic & Medicinal Chemistry Letters 10 (2000) 239±241
Chemical Synthesis and Cytotoxicity of some Azinomycin
Analogues Devoid of the 1-Azabicyclo[3.1.0]hexane Subunit
Timothy J. Hodgkinson, a Lloyd R. Kelland, b Michael Shipman a,* and Franck Suzenet a
aSchool of Chemistry, University of Exeter, Stocker Rd, Exeter, Devon EX4 4QD, UK
bCRC Centre for Cancer Therapeutics, Institute of Cancer Research, Sutton, Surrey SM2 2NG, UK
Received 7 October 1999; accepted 22 November 1999
AbstractÐA series of compounds related to the left-hand domain of the azinomycins have been made and evaluated for cytotoxic
activity against a small panel of human tumour cell lines. The epoxide ring is shown to be essential for biological activity. Cyto-
toxicity is also shown to be sensitive to changes in the substitution pattern on the aromatic ring and the amide group. # 2000
Elsevier Science Ltd. All rights reserved.
In 1986, azinomycins A 1 and B 2 were isolated from
the culture broths of Streptomyces griseofuscus S42227
and were found to exhibit potent in vitro cytotoxic
activity and signi®cant in vivo anti-tumour activity (Fig.
1).1±4 Armstrong et al.5 have shown that azinomycin B
causes interstrand cross-links in the major groove of
duplex DNA by initial alkylation at guanine and sub-
sequent reaction at a second purine residue two bases
along on the complimentary strand of the DNA duplex.
Further studies by Fujiwara et al.6 have revealed more
details concerning the chemical events which lead to
these interstrand cross-links. Along with the azinomy-
cins, an additional metabolite 3 devoid of the 1-aza-
bicyclo[3.1.0]hexane ring system was isolated from the
culture broths of S. griseofuscus S42227.1±3 It was ori-
ginally concluded that this material lacked anti-bacterial
or anti-tumour activity,3 however further studies by
Terashima7 have indicated that this compound pos-
sesses strong cytotoxic activity against P388 murine
leukemia [IC50 (mg/mL) 0.0036].7 In other work, azino-
mycin±lexitropsin hybrid molecules (e.g. 4) containing
the left hand domain of the azinomycins have been
shown to possess DNA-cleaving activity.8 From these
and other studies, it appears that the epoxide domains
of compounds 1±4 play a key role in their biological
activity. In this Letter, we describe the synthesis and
cytotoxic activity of a range of compounds related to
epoxide 1 in which systematic modi®cations have been
made to this structural motif. These studies provide
some new insights into the nature of the interactions
between the epoxide domains of agents 1±4 and DNA.
A series of primary amides 7a±e possessing di€erent
aromatic chromophores were made from homochiral
epoxide (2S,3S)-5 to evaluate how the nature of the
aromatic group in¯uences cytotoxicity. These com-
pounds were made by acylation of the secondary hy-
droxyl group of 59,10 and subsequent conversion of the
Figure 1.
*Corresponding author. Tel.: +44-1392-263469; fax: +44-1392-
263434; e-mail: m.shipman@exeter.ac.uk
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(99)00663-0

240
T. J. Hodgkinson et al. / Bioorg. Med. Chem. Lett. 10 (2000) 239±241
Scheme 1. Method A: (i) H2, Pd/C, MeOH; (ii) NH4OH, Et3N, HOBt, PyBOP, DMF. Method B: (i) H2, Pd/C, MeOH; (ii) 4-methoxybenzlamine,
Et3N, HOBt, PyBOP, DMF; (iii) CAN, MeCN/H2O.
Scheme 2. Reagents and conditions: (i) Et2Zn, CH2I2, CH2Cl2, 87%; (ii) 1-naphthoyl chloride, Et3N, DMAP, CH2Cl2, 91%; (iii) H2, Pd/C, MeOH;
(iv) 4-methoxybenzylamine, Et3N, HOBt, PyBOP, DMF, 93% from 9; (v) CAN, MeCN/H2O, 73%.
benzyl ester into the primary amide (Scheme 1). Two
di€erent methods were used for this later conversion,
although we have found that the more direct approach
is generally higher yielding (i.e. Method A). In addition,
(2R,3R)-3 was made in an identical fashion to our
published route to (2S,3S)-3 by using AD-mix-b in the
asymmetry inducing dihydroxylation step.9,10
have determined that epoxide (2S,3S)-3 is highly cyto-
toxic and has potency comparable to the established
anti-tumour agents mitomycin C and cisplatin (entry 1,
cf. entries 11 and 12). The enantiomer of this com-
pound, (2R,3R)-2, was found to be slightly less potent
indicating some level of chiral recognition (entry 1, cf.
entry 2). The epoxide ring is clearly important for cyto-
toxic activity as cyclopropane 10 was essentially inactive
(entry 4, cf. entry 8). The potency of these compounds is
highly dependent on the nature of the aromatic residue
(entry 1, cf. entries 3±7). Even a very subtle change from
the methyl to ethyl ether at C-3 of the naphthalene
residue results in a substantial loss of potency (entry 1,
cf. entry 7). It is also notable that epoxide 7d containing
the 2-quinoxaloyl group, an established DNA inter-
calating group,14 was completely inactive.15 Addition-
ally, it is interesting that we observed a loss of potency
when the amide functionality was further substituted
(entry 1 cf. entries 9 and 10).
In accordance with the postulated mechanism of action
of the azinomycins,5,6 we imagined that the cytotoxic
activity of metabolite 3 might arise from monoalkyla-
tion of DNA by the reactive epoxide moiety. To test this
hypothesis, (2S)-10 was made in which the epoxide ring
was replaced by the isosteric but quite inert cyclopro-
pane ring. This was achieved by cyclopropanation11 of
homochiral allylic alcohol (2S)-89,10 followed by con-
version to (2S)-10 along similar lines to that described
above (Scheme 2). From the chemical standpoint, it is
notable that no concomitant cleavage of the cyclopro-
pane ring was observed during the hydrogenation step.
In the case of four representative compounds ((2S,3S)-3
(2R,3R)-3, (2S,3S)-7b and (2S)-10), we have obtained
more direct evidence that these compounds induce
cytotoxicity by alkylating DNA. This was accomplished
by undertaking DNA binding-sequence speci®city stu-
dies using the taq polymerase stop assay.16 In these
experiments, good correlation was observed between
DNA binding and cell growth inhibition potency.
Cyclopropyl amide 10 did not produce any taq stop
sites over background DMSO control levels at up to
50 mM. The other three compounds produced binding
anities which paralled their potency in the cytotoxicity
tests. Further studies directed towards gaining greater
To study the role of the amide substitution pattern,
secondary amides 12 and 13 were made from (2S,3S)-
1110 by selective cleavage of the benzyl ester followed by
coupling with glycine ethyl ester and (S)-valine ethyl
ester respectively (Scheme 3).
With these compounds in hand, our attention turned to
assessing their biological activity. All the compounds
were tested for in vitro cytotoxic activity against a small
panel of human tumour cell lines {A2780, A2780cisR,12
CH1, SKOV-3 (all ovarian) and HT29 (colon)} (Table
1).13 Consistent with Terashima's observations,7 we
Scheme 3. Reagents and conditions: (i) H2, Pd/C, MeOH; (ii) EtO2CCH2NH2.HCl, Et3N, HOBt, PyBOP, DMF; 89% from 11;
(iii) (S)-EtO2CCH(iPr)NH2.HCl, Et3N, HOBt, PyBOP, DMF, 85% from 11.

T. J. Hodgkinson et al. / Bioorg. Med. Chem. Lett. 10 (2000) 239±241
241
Table 1.
assigned, see Moran E. J.; Armstrong, R. W. Tetrahedron
Lett. 1991, 32, 3807.
5. Armstrong, R. W.; Salvati, M. E.; Nguyen, M. J. Am.
Chem. Soc. 1992, 114, 3144.
6. Fujiwara, T.; Saito, I.; Sugiyama, H. Tetrahedron Lett.
1999, 40, 315.
7. Hashimoto, M.; Matsumoto, M.; Yamada, K.; Terashima,
S. Tetrahedron Lett. 1994, 35, 2207.
8. Shishido, K.; Haruna, S.; Iitsuka, H.; Shibuya, M. Hetero-
cycles 1998, 49, 109.
9. Bryant, H. J.; Dardonville, C. Y.; Hodgkinson, T. J.; Ship-
man, M.; Slawin, A. M. Z. Synlett. 1996, 973.
10. 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.
11. Charette, A. B.; Lebel, H. J. Org. Chem. 1995, 60, 2966.
12. A2780cisR is a subline of A2780 possessing acquired
resistance (ca. 10-fold) to the commonly used DNA cross-
linking reagent cisplatin.
Entry Compound
Cytotoxic activitya against:
A2780 A2780cisR CH1 SKOV-3 HT29
1
2
3
4
5
6
7
8
(2S,3S)-3
(2R,3R)-3
(2S,3S)-7a
(2S,3S)-7b
(2S,3S)-7c
(2S,3S)-7d >2 5
(2S,3S)-7e
(2S)-10
<0.05
0.058
20.5
0.44
0.39
0.076 <0.05
0.155 0.062
>25 >25
1.25
2.3
>25
0.33
0.61
>25
2.15
2.6
>2 5
4.3
>2 5
1.1
1.35
0.55
0.17
5.1
13.0
>2 5
3.5
>2 5
>2 5
1.4
>2 5
>2 5
12.0
>2 5
1.8
>2 5
0.34
9
(2S,3S)-12
(2S,3S)-13
Mitomycin C
Cisplatin
0.83
0.56
0.06
3.0
0.26
0.315
0.04
0.1
7.3
2.95
0.2
4.4
2.25
2.25
0.04
1.7
10
11
12
0.35
0.05
0.3
a96-h incubation, IC50 (mM) values.
13. Cytotoxicity assay: Cells were added to 96-well microtitre
plates at 5Â103/well and allowed to attach overnight. The
compounds were dissolved immediately before use in DMSO
(at 20 mM) and diluted in tissue culture growth medium prior
to adding to the cell lines at a range of ®nal concentrations
(from 100 mM down to 2.5 nM). Drugs remained in contact
with the cells throughout a 96-h incubation period before
assessing for growth inhibitory e€ects using the Sulforhod-
amine B (SRB) assay. For a description of this SRB assay, see
Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon,
J.; Vistica, D.; Warren, J.; Bokesch, H.; Kennedy, S.; Boyd,
M. R. J. Nat. Cancer Inst. 1990, 82, 1107.
insight into the molecular mechanism of action of these
and other azinomycin analogues are ongoing and these
results will be disclosed in due course.
Acknowledgements
We gratefully acknowledge the ®nancial support pro-
vided by the Cancer Research Campaign (to T. J. H.
and F. S.). We are indebted to the EPSRC National
Mass Spectrometry Centre for performing some of the
mass spectral measurements and the EPSRC Chemical
Database Service at Daresbury.17
14. For example, see Wang, A. H.-J.; Ughetto, G.; Quigley, G.
J.; Hakoshima, T.; van der Marel, G. A.; van Boom J. H.;
Rich, A. Science 1984, 225, 1115.
15. This lack of biological activity may be attributed to poor
transport through the cell membrane. While we have not
directly measured the lipophilicity of this compound, we have
noted that it displays poor solubility in organic solvents com-
pared with the other analogues.
16. The technique involved incubation of these compounds
with plasmid DNA of known sequence (2 h at 37 ꢀC). Cispla-
tin was run as a control in these experiments. For a description
of this assay, see Ponti, M.; Forrow, S. M.; Souhami, R. L.;
D'Incalci, M.; Hartley, J. A. Nucl. Acid Res. 1991, 19, 2929.
17. Fletcher, D. A.; McMeeking, R. F.; Parkin, D. J. Chem.
Inf. Comp. Sci. 1996, 36, 746.
References and Notes
1. Nagaoka, K.; Matsumoto, M.; Ono, J.; Yokoi, K.; Ishizeki,
S.; Nakashima, T. J. Antibiot. 1986, 39, 1527.
2. Yokoi, K.; Nagaoka, K.; Nakashima, T. Chem. Pharm.
Bull. 1986, 34, 4554.
3. Ishizeki, S.; Ohtsuka, M.; Irinoda, K.; Kukita, K.-I.;
Nagaoka, K.; Nakashima, T. J. Antibiot. 1987, 40, 60.
4. Azinomycin B has been shown to be identical to carzino-
philin whose structure had previously been incorrectly