Synthetic analogues of the manzamenones and plakoridines which inhibit DNA polymerase

Article abstract of DOI:10.1016/j.bmcl.2006.03.005

An array of novel analogues of the marine oxylipins, the manzamenones and plakoridines, have been prepared in divergent fashion using an approach modelled on a biogenetic theory. Many of the target compounds show potent inhibition of DNA polymerases α and β and human terminal deoxynucleotidyl transferase (TdT).

Full text of DOI:10.1016/j.bmcl.2006.03.005

Bioorganic & Medicinal Chemistry Letters 16 (2006) 2877–2881
Synthetic analogues of the manzamenones and plakoridines
which inhibit DNA polymerase
Jeremy R. Doncaster,a Laura L. Etchells,a Neil M. Kershaw,a Ryoichi Nakamura,b
Hazel Ryan,a Ryo Takeuchi,b Kengo Sakaguchi,b
Ali Sardariana, and Roger C. Whiteheada,*
aThe School of Chemistry, The University of Manchester, Oxford Road, Manchester M13 9PL, UK
bDepartment of Applied Biological Science, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan
Received 7 February 2006; revised 1 March 2006; accepted 2 March 2006
Available online 24 March 2006
Abstract—An array of novel analogues of the marine oxylipins, the manzamenones and plakoridines, have been prepared in
divergent fashion using an approach modelled on a biogenetic theory. Many of the target compounds show potent inhibition
of DNA polymerases a and b and human terminal deoxynucleotidyl transferase (TdT).
Ó 2006 Elsevier Ltd. All rights reserved.
Marine sponges of the genus Plakortis are a rich source
of structurally diverse natural products with a variety of
bioactivities including anti-cancer, anti-bacterial and
anti-fungal properties.1 Some time ago, we became
interested in a family of oxidized fatty acid derivatives
(oxylipins) isolated from Japanese Plakortis sponges
by the research group of Jun’ichi Kobayashi. Members
of this family of compounds, which include untenone
A (3),2 the manzamenones (e.g., manzamenone A
(5)),3a–c plakevulin A (6)4a,b and the plakoridines (9,
10)3c,5 are characterized by the common structural fea-
tures of at least one fully saturated C16 alkyl chain
and at least one methoxycarbonyl group. It was origi-
nally observed by Kobayashi that many members of
the family possessed structures which could be derived
biosynthetically from (E)/(Z)-methyl-3,6-dioxo-4-doco-
senoate 1/2. Prompted by this observation, we have
described a plausible biosynthetic pathway which inter-
relates 1 and/or 2 with the various oxylipins (Scheme 1)
and we have also provided synthetic chemical evidence
in support of our proposal.6a–e
Untenone A and manzamenone A are efficient inhibitors
of DNA polymerases a and b (pol a and pol b) as well as
human terminal deoxynucleotidyl transferase (TdT)
(IC50 values ranging between 1.9 and 57 lM).7 The bio-
logical potency of 3 and 5 has provided stimulus for
investigations into the synthesis and evaluation of other
compounds belonging to this structural class. Herein, we
describe the preparation of novel analogues of the
manzamenones and the plakoridines using a divergent
synthetic approach modelled on our biogenetic theory,
together with the results of bioassays of the target
compounds against pol a, pol b and human TdT.
The ultimate aim of our research was to prepare a range
of analogues of the manzamenones (general structure
13) and the plakoridines (general structure 14) using
an approach modelled on our biogenetic theory
(Scheme 2).
Our first goal towards this end was the synthesis of
(E)- and (Z)-enediones of type 11/12, which was accom-
plished in expedient fashion from the methyl ester of
2-furan acetic acid 15 (Scheme 3). Thus, Friedel–Crafts
acylation of 15 with the appropriate acid chloride fol-
lowed by ketone reduction using the Huang-Minlon
modification of the Wolff–Kishner conditions8a,b gave
5-alkyl-furan-2-yl-acetic acids 16 in moderate yields.
Subsequent esterification under either acid-catalysed
conditions or in the presence of DCCI gave ester
derivatives 17.
Keywords: Oxylipins; Natural products; DNA polymerase; Biomimetic
synthesis.
*
Corresponding author. Tel.: +44 161 275 7856; fax: +44 161 275
4939; e-mail: roger.whitehead@manchester.ac.uk
Present address: The Department of Chemistry, The College of
Sciences, The University of Shiraz, Shiraz 71454, Iran.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.03.005

2878
J. R. Doncaster et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2877–2881
O
H3CO2C
H33C16
dehydrative
dimerisation
retro-
C16H33
CO2CH3
H
CO2CH3
reduction
HO
OH
Dieckmann
H33C16
HO
O
H33C16
C16H33
H33C16
CO2CH3
6
CO2CH3
H
4
3
O
plakevulin A
aldol
HO2C
O
5
untenone A
CO2CH3
manzamenone A
O
H33C16
CO2CH3
Mannich
O
O
Δ
4,5 : E
4,5 : Z
1
2
HO
CO2CH3
O
CO2CH3
R
'internal
redox'
O
H33C16
CO2CH
R
Δ
3 Michael
O
H33C16
R
O
N
H33C16
HN
N
C6H4OH
7
C6H4OH
C6H4OH
8
R = C3H7: plakoridine A 9
R = C15H31: plakoridine B 10
Scheme 1.
approach to compounds 13 which closely mirrors the pro-
posed biogenesis of the manzamenones involves retro-
Dieckmann reaction of tri-cyclic adducts 20 mediated
by different nucleophiles R3H. Herein, we report for the
first time that dehydrative dimerisation of cyclopente-
nones 18 to give adducts 20 can be achieved using triflu-
oroacetic anhydride (TFAA) as dehydrating agent: the
success of this reagent being a consequence of the absence
of nucleophilic entities in the reaction mixtures which
would otherwise quench the reactive species 20. Disap-
pointingly, exposure of 20 to a range of O and N centred
nucleophiles gave manzamenone analogues 13 in general-
ly poor yields with formation of alternative diastereoiso-
mers (believed to be 2-epi-13), being competitive with the
desired outcome. Formation of the latter was found
most significant following incomplete removal of trifluo-
roacetic acid (TFA) from samples of 20 prior to nucleo-
philic quenching and is believed to be a consequence of
reduced facial selectivity in protonation of the intermedi-
ate enols/enolates formed during the retro-Dieckmann
reactions.
O
5
R1
CO2R2
4
R2O2C
R1
O
Δ
Δ
4,5 : Z 11
4,5 : E 12
R1
H
HO
6
CO2R2
9
O
2
3
4
5
3
2
1
6
CO2R2
8
1
4
5
7
R1
R3
N
H
R4
13
R3OC
O
14
Scheme 2. Compound 13: R1 = linear alkyl; R2 = linear or branched
alkyl; R3 = hydroxy, alkoxy or alkylamino. Compound 14: R1 and
R2 = linear alkyl; R3 = linear alkyl or aryl; R4 = terminally substituted
alkyl.
Conditions were then developed for the controlled oxi-
dation of the furan ring of 17 to give either 11 or 12.6e
Treatment of 17 with bromine in methanol,9 followed
by hydrolysis of the resulting bis-acetals under mildly
acidic conditions,10 gave (Z)-enediones 11 which exist
predominantly in solution as the cyclic hemi-ketal tauto-
mers 110. Alternatively, exposure of 17 dissolved in a
mixture of acetone and water (5:1) to one equivalent
of bromine at ꢀ20 °C gave (E)-enediones 12,11 which ex-
ist in solution almost exclusively as enol tautomers 120.6e
The expediency of the synthetic route to manzamenone
analogues of type 13 has allowed investigations into the
effect of core structural changes on the enzyme inhibito-
ry activity of this class of compounds. For example,
cyclopropyl compound 23 was synthesized in three
steps, and in moderate yield, from 43-O-methylmanz-
amenone A (21) (Scheme 5):6d a more direct approach
to 23 utilising dimethylsulfoxonium methylide (Corey’s
reagent12) for incorporation of the cyclopropane ring
was, unfortunately, unsuccessful.
Mildly basic treatment of hemiketals 110 resulted in fac-
ile aldol cyclisation to give untenone analogues 18 from
which the target array of manzamenone analogues 13
could be prepared (Scheme 4).
Our preferred approach to the synthesis of compounds 13
involved initial preparation of manzamenone A ana-
logues 19 using previously described procedures.6a–d Sub-
sequent derivatisation of the free carboxyl group of 19
using standard coupling procedures then provided the
target compounds 13. An intrinsically more attractive
A selection of our target compounds of type 13 were
assayed against pol a, pol b and human TdT using the
R1
R1
CO2R2
CO2R2
CO2R2
iv), v)
O
O
O
HO
i), ii)
iii)
11
12
11'
CO2CH3
CO2H
CO2R2
vi)
R1
R1
O
O
O
O
OH
R1
R1
CO2R2
15
16
17
12'
O
O
Scheme 3. Reagents and conditions; (i) RCOCl, SnCl4, CH2Cl2, ꢀ5 °C, 84–98%; (ii) H2NNH2, NaOH, HOCH2CH2OH, D, 50–96%; (iii) R2OH,
AmberliteÒ IR 120 (H), D or R2OH, DCCI, CH2Cl2, 0 °C, 75–98%; (iv) Br2, CH3OH, Na2CO3, 67–96%; (v) 0.005 M H2SO4, H2O, dioxane, rt;
(vi) Br2, acetone, H2O, ꢀ20 °C to ꢀ10 °C, 56–61%.

J. R. Doncaster et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2877–2881
2879
R2O2C
R1
H
R1
9
2
3
1
6
CO2R2
8
7
4
R2O2C
R1
5
R2O2C
R1
ii) or iii)
O
R1
O
R1
O
iv)
H
H
H
H
19
O
HO2C
R1
HO
R1
CO2R2
i)
CO2R2
CO2R2
O
vi)
v)
O
11'
HO
CO2R2
R1
H
13
CO2R2
R3OC
R3OC
R1
18
2-epi-13
20
CO2R2
O
Scheme 4. Reagents and conditions; (i) NaHCO3 (aq), dioxane, rt, 1 h, 65–70%; (ii) 40–70 °C, neat, 1–6 day, 48–94%; (iii) scandium
trisdodecylsulfate (0.1 equiv.), H2O, 25 °C, 1–24 h, 70–80%; (iv) R3H, DCCI, CH2Cl2, 41–57%; (v) TFAA, CDCl3, rt, 24 h; (vi) for ester derivatives:
R3H, rt, 24 h, 15–63% from 18; for amide derivatives: R3H, CH2Cl2, rt, 24 h, 10–15% from 18.
group at C5 seemed to be important for good levels of
enzyme inhibition.
C16H33
H3CO2C
H33C16
H3CO2C
H33C16
C16H33
H
H
N
N
i)
CO2CH3
CO2CH3
The second phase of our research has involved the prep-
aration and bioassay of novel analogues of the plakori-
dines with general structure 14. These compounds have
been prepared from enols 120 in one simple synthetic
operation, using an approach modelled on the biogenet-
ic theory.6e Thus, storage of CDCl3 solutions of 120 and
imines 2614 at rt for prolonged periods of time gave
plakoridine analogues 14, via intermediate diketones
27, in acceptable yields (Scheme 6).
H
21
O
OCH3
H3CO2C
H3CO2C
H3CO2C
22
ii), iii)
C16H33
H
H33C16
CO2CH3
H
O
H3CO2C
23
Scheme 5. Reagents and conditions; (i) CH2N2, Et2O, rt, 94%; (ii)
neat, 125 °C, 4 h, 57%; (iii) HCl, Et2O, rt, 24 h, 56%.
Small quantities of the diastereoisomers 3-epi-14 were
also generated in these reactions which could be separat-
ed from the major product by careful chromatography.
methods previously described by Mizushina et al.13a–c
The results of the assays are presented in Table 1.
Regarding the structural features which influence the ex-
tent of inhibition of the polymerases, a number of con-
clusions can be reached: (i) lengthy alkyl chains and the
presence of a conjugated enone system are both prereq-
uisites for potent inhibition; (ii) a free carboxylic acid
group attached to C5 is not a requirement for good
activity and ester moieties at this position are well toler-
ated; (iii) inversion of relative configuration at C2 has
little effect on the extent of enzyme inhibition. With
regard to inhibition of human TdT, all structural
alterations resulted in a reduction of potency compared
with manzamenone A and, in particular, lengthy
pendant alkyl chains as well as a free carboxylic acid
The results of bioassays of the novel plakoridine
analogues against DNA polymerases a/b and human
TdT are presented in Table 2. All of the pyrrolidine
derivatives are good to moderate inhibitors of the poly-
merases, although generally less active than the manz-
amenone analogues. Potency is sensitive to even a
small reduction in the length of the alkyl chain R1 and
is insensitive to inversion of the stereochemistry at C3
as well as variation of the ring substituents R3 and R4:
the only exception to the latter generalization being
the analogue derived from tryptamine (entry 6) which
showed good potency and slight selectivity for DNA
polymerase b. None of the analogues displayed
significant inhibition of human TdT.
Table 1. Inhibition of pol a, pol b and human TdT by manzamenone analogues
Compound
R1
R2
R3
IC50 (lM)
Pol b
Pol a
TdT
2.5
>50c
8.6
1
C
16H33
CH3
CH3
CH3
iPr
OH
OH
1.9
>50c
1.9
3.2
>50c
2.3
2
C2H5
C12H25
C16H33
3
OH
OH
4
3.3
6.9
4.4
8.7
20
>12.5c
39.7
5
C
16H33
CH3
CH3
CH3
CH3
HN(CH2)2Ph
OC2H5
OC2H5
OCH3
6
C16H33
C16H33
C16H33
2.3
3.5
>12.5c
1.9
2.4
>12.5c
7a
8b
48.4
>12.5c
a Inverted stereochemistry at C2.
b Cyclopropyl compound 23.
c Poor solubility at higher concentrations did not allow accurate assessment of IC50
.

2880
J. R. Doncaster et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2877–2881
O
OH
R1
CO2R2
R1
i)
CO2R2
CO2R2
CO2R2
R3
HO
2
CO2R2
4
O
HO
3
O
O
O
ii)
12'
R4
5
12
+
O
O
1
R3
R1
R1
R3
N
R4
R1
N
R4
14
7
R4
O
N
6
+
N
R4
27
R3
NH2
H
R3
3-epi-14
24
25
26
Scheme 6. Reagents and conditions; (i) H2O, rt, 3 h, 90–97%; (ii) CDCl3, rt, 4–28 d, 26–40%.
Table 2. Inhibition of pol a, pol b and human TdT by plakoridine analogues
Compound
R1
R2
R3
R4
IC50 (lM)
Pol b
Pol a
TdT
1
C16H33
C16H33
C12H25
C12H25
C12H25
C12H25
CH3
CH3
CH3
CH3
CH3
CH3
C2H5
C6H5
C2H5
C2H5
C2H5
C6H5
Ph(CH2)2
Ph(CH2)2
Ph(CH2)2
PhCH2
8.9
7.9
8.7
8.8
22
>25b
>50b
>50b
>50b
>50b
>50b
2
3
25
4
5a
19.9
19.7
21
30
PhCH2
3-indolyl(CH2)2
27.1
10
6
a Inverted stereochemistry at C3.
b Poor solubility at higher concentrations did not allow accurate assessment of IC50
.
Acknowledgments
Table 3. Inhibition of pol a and pol b by enedione 1, untenone A (3)
and plakevulin A (6)
We acknowledge, with thanks, the EPSRC for funding
(J.R.D., L.L.E., N.M.K., H.R. and A.S.) and we are
also very grateful to Professor Gareth Morris of The
University of Manchester for invaluable advice concern-
ing the NMR analysis of several of our target com-
pounds. We also thank Dr. Masaharu Takemura of
the Functional Genomics Institute at Mie University
for the generous donation of DNA polymerase a.
IC50(lM)
Pol a
Pol b
(E)-Enedione 1
Untenone A (3)
Plakevulin A (6)
3.7
4.3
66
8.8
57
179
The results reported here for the array of novel
analogues of the manzamenones and plakoridines indi-
cate that the more potent inhibitors of DNA polymerases
a and b share three common structural features: at least
one long saturated alkyl chain, an a,b-unsaturated
carbonyl moiety and at least one carboxyl group. Unte-
none A (3) also possesses these three features and is a po-
tent inhibitor of pol a,7 whereas plakevulin A (6) lacks an
a,b-unsaturated carbonyl moiety and is only a moderate
inhibitor of pol a and a poor inhibitor of pol b.4b Prompt-
ed by these observations, we have assayed (E)-methyl-
3,6-dioxo-4-docosenoate (1), the putative biosynthetic
precursor of the natural plakoridines, against the poly-
merases. This compound also possesses the three key
structural features and, pleasingly, was found to be a po-
tent inhibitor of both pol a and pol b, showing a slight
selectivity for inhibition of the former (Table 3).
References and notes
1. Rahm, F.; Hayes, P. Y.; Kitching, W. Heterocycles 2004,
64, 523.
2. Ishibashi, M.; Takeuchi, S.; Kobayashi, J. Tetrahedron
Lett. 1993, 34, 3749.
3. (a) Tsukamoto, S.; Takeuchi, S.; Ishibashi, M.; Kobay-
ashi, J. J. Org. Chem. 1992, 57, 5255; (b) Kobayashi, J.;
Tsukamoto, S.; Takeuchi, S.; Ishibashi, M. Tetrahedron
1993, 49, 5955; (c) Takeuchi, S.; Kikuchi, T.; Tsukamoto,
S.; Ishibashi, M.; Kobayashi, J. Tetrahedron 1995, 51,
5979.
4. (a) Tsuda, M.; Endo, T.; Perpelescu, M.; Yoshida, S.;
Watanabe, K.; Fromont, J.; Mikami, Y.; Kobayashi, J.
Tetrahedron 2003, 59, 1137; (b) Saito, F.; Takeuchi, R.;
Kamino, T.; Kuramochi, K.; Sugawara, F.; Sakaguchi,
K.; Kobayashi, S.; Tsuda, M.; Kobayashi, J. Tetrahedron
Lett. 2004, 45, 8069.
In conclusion, we have prepared a range of novel ana-
logues of the marine oxylipins, the manzamenones
and the plakoridines, using an approach modelled on
a biogenetic theory. We have found that many of the
analogues show strong inhibition of DNA polymerases
a and b and human TdT with the plakoridine analogues
being generally less potent than the manzamenones.
The data indicate that three structural features are key
for good inhibition of the polymerases: at least one long
saturated alkyl chain, an a,b-unsaturated carbonyl
moiety and at least one carboxyl group. The results of
previous SAR studies of untenone A analogues are in
accord with this observation.
5. Takeuchi, S.; Ishibashi, M.; Kobayashi, J. J. Org. Chem
1994, 59, 3712.
6. (a) Al-Busafi, S.; Drew, M. G. B.; Sanders, T.; Whitehead,
R. C. Tetrahedron Lett. 1998, 39, 1647; (b) Al-Busafi, S.;
Whitehead, R. C. Tetrahedron Lett. 2000, 41, 3467; (c) Al-
Busafi, S.; Doncaster, J. R.; Drew, M. G. B.; Regan, A. C.;
Whitehead, R. C. J. Chem. Soc., Perkin Trans. 1 2002,
476; (d) Doncaster, J. R.; Ryan, H.; Whitehead, R. C.
Synlett 2003, 651; (e) Etchells, L. L.; Sardarian, A.;
Whitehead, R. C. Tetrahedron Lett. 2005, 46, 2803.
7. Saito, F.; Takeuchi, R.; Kamino, T.; Kuramochi, K.;
Sugawara, F.; Sakaguchi, K.; Kobayashi, S. Bioorg. Med.
Chem. Lett. 2004, 14, 1975.

J. R. Doncaster et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2877–2881
2881
8. (a) Huang-Minlon J. Am. Chem. Soc. 1946, 68, 2487; (b)
Huang-Minlon J. Am. Chem. Soc. 1949, 71, 3301.
9. Levisalles, J. Bull. Soc. Chim. Fr. 1957, 997.
10. Williams, P. D.; LeGoff, E. J. Org. Chem. 1981, 46, 4143.
11. Jurczak, J.; Pikul, S. Tetrahedron Lett. 1985, 26, 3039.
12. Golobolov, Y. G.; Nesmeyanov, A. N.; Lysenko, V. P.;
Boldeskul, I. E. Tetrahedron 1987, 43, 2609.
13. (a) Mizushina, Y.; Tanaka, N.; Yagi, H.; Kurosawa,
T.; Onoue, M.; Seto, H.; Horie, T.; Aoyagi, N.;
Yamaoka, M.; Matsukage, A.; Yoshida, S.; Sakaguchi,
K. Biochim. Biophys. Acta 1996, 1308, 256; (b)
Mizushina, Y.; Yagi, H.; Tanaka, N.; Kurosawa, T.;
Seto, H.; Katsumi, K.; Onoue, M.; Ishida, H.; Iseki,
A.; Nara, T.; Morohashi, K.; Horie, T.; Onomura, Y.;
Narusawa, M.; Aoyagi, N.; Takami, K.; Yamaoka, M.;
Inoue, Y.; Matsukage, A.; Yoshida, S.; Sakaguchi, K.
J. Antibiot. 1996, 49, 491; (c) Mizushina, Y.; Yoshida,
S.; Matsukage, A.; Sakaguchi, K. Biochim. Biophys.
Acta 1997, 1336, 509.
14. Simion, A.; Simion, C.; Kanda, T.; Nagashima, S.;
Mitoma, Y.; Yamada, T.; Mimura, K.; Tashiro, M.
J. Chem. Soc., Perkin Trans. 1 2001, 2071.