Synthesis and?antitumour activity of?new muricatacin and?goniofufurone analogues

Article abstract of DOI:10.1016/j.ejmech.2006.06.008

A divergent approach to the 7-oxa (-)-muricatacin analogue 2, the corresponding (+)-enantiomer ent-2 and the furanolactone 3 is reported starting from d-xylose. The resulting lactones have shown a potent and selective in vitro cytotoxicity against certain

Full text of DOI:10.1016/j.ejmech.2006.06.008

European Journal of Medicinal Chemistry 41 (2006) 1217–1222
http://france.elsevier.com/direct/ejmech
Preliminary communication
Synthesis and antitumour activity
of new muricatacin and goniofufurone analogues
a,
a
a
a
*
Velimir Popsavin , Bojana Srećo , Ivana Krstić , Mirjana Popsavin ,
Vesna Kojićb, Gordana Bogdanovićb
a Department of Chemistry, Faculty of Sciences, University of Novi Sad, Trg D. Obradovića 3, 21000 Novi Sad, Serbia
b Institute of Oncology Sremska Kamenica, Institutski put 4, 21204 Sremska Kamenica, Serbia
Received 20 February 2006; revised and accepted 19 June 2006
Available online 07 August 2006
Abstract
A divergent approach to the 7-oxa (–)-muricatacin analogue 2, the corresponding (+)-enantiomer ent-2 and the furanolactone 3 is reported
starting from D-xylose. The resulting lactones have shown a potent and selective in vitro cytotoxicity against certain human neoplastic cell lines.
© 2006 Elsevier Masson SAS. All rights reserved.
Keywords: Muricatacin analogues; Wittig reaction; Goniofufurone analogue; Bioisostere; Antitumour lactones
1. Introduction
vious work, we report herein on the synthesis and antitumour
activity of the hydroxylactones 2, ent-2 and 3. The enantiomers
2 and ent-2 were designed as possible muricatacin bioisosteres,
while the furanolactone 3 represents a conformationally con-
strained one-carbon homologue of ent-2. Compound 3 might
be also considered as a non-styryl analogue of goniofufurone
(4), a naturally occurring cytotoxic lactone that was isolated
from the stem bark of Goniothalamus giganteus (Annonaceae)
[21–23].
The plant family Annonaceae has for a long time aroused a
considerable interest from a biomedicinal point of view, mainly
due to its bioactive polyketide constituents [1]. Within the
Annonaceae, the tropical plant Annona muricata has given
rise to the isolation of muricatacin, an acetogenin hydroxylac-
tone that inhibits proliferation of certain human tumour cell
lines [2]. Remarkable, the natural product is comprised of
(–)-(4R,5R)-5-hydroxy-4-heptadecanolyde (1) and its (+)-
(4S,5S) enantiomer (ent-1; Scheme 1), with former predominat-
ing (ee, ca. 25%). Both enantiomers showed essentially the
same antitumour activity, and not surprisingly they were the
objects of numerous synthetic efforts [3]. A number of murica-
tacin stereoisomers and analogues have also been synthesized
[4–16], but only a few were evaluated for their antitumour
activity [13–16]. As a part of our ongoing project in the synth-
esis of new hydroxylactones as potential antitumour agents
from monosaccharides [17–20], we have recently disclosed a
novel general approach to the enantiodivergent synthesis of 1
and ent-1 from D-xylose, which is suitable for elaboration to a
variety of 7-oxa-analogues [20]. As an extension of our pre-
2. Results and discussion
2.1. Chemistry
To synthesize the target compounds we developed two syn-
thetic routes starting from D-xylose. The first route follows the
previously described enantiodivergent approach [20] adopted
to complete the synthesis of both 2 and ent-2. The second
route is a more efficient way to prepare the enantiomer 2,
which also provides an access to the furanolactone 3, via an
appropriate divergent intermediate.
The enantiodivergent synthesis of 2 and ent-2 is shown in
Scheme 2. The common starting material was the cyclohexyli-
dene derivative 5 readily available from D-xylose in three steps
[18,24]. By using a five-step sequence recently developed in
our laboratory [20], compound 5 was converted to the lactol
* Corresponding author.
E-mail address: popsavin@ih.ns.ac.yu (V. Popsavin).
0223-5234/$ - see front matter © 2006 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2006.06.008

1218
V. Popsavin et al. / European Journal of Medicinal Chemistry 41 (2006) 1217–1222
Scheme 1. (–)-Muricatacin (1), (+)-muricatacin (ent-1), (+)-goniofufurone (4) and the target analogues 2, ent-2 and 3.
Scheme 2. Reagents and conditions: (a) aq. NaIO4, silica gel, CH2Cl2, r.t., 0.5 h, (b) NaBH4, MeOH, 0 °C, 1.5 h, (c) 2:1 TFA/H2O, r.t., 1.5 h, 50% from 6; (d)
C10H21Br, Ag2O, AgOTf, Et2O, reflux, 7.5 h, 71% of 10, 80% of ent-10; (e) H2–Pd/C, EtOAc, r.t., 19 h, 87% of 2, 82% of ent-2.
6 in 42% overall yield. The intermediate 6 was subsequently
converted to the hydroxylactone 9 by using a newly developed
one-pot procedure that involved a previous oxidative cleavage
of the diol 6 with sodium periodate on silica, sodium borohy-
dride reduction of resulting aldehyde 7 to diol 8, followed by a
subsequent acid promoted lactonisation of the intermediate 8.
This procedure provided the key intermediate 9 in 50% overall
yield (from 6). Compound 9 readily reacted with an excess of
decyl bromide and silver oxide in ether, in the presence of a
catalytic amount of silver triflate, to give the 7-O-decyl deriva-
tive 10 in 71% yield. Hydrogenolytic removal of benzyl ether
protective group in 10 furnished 2 in 87% yield. Enantiopure
hydroxylactone ent-9, easily obtainable from 5 by a recently
developed method from our laboratory [20], represents the
key chiral intermediate for preparation of target ent-2. Thus,
O-alkylation of ent-9, followed by a subsequent hydrogenolytic
removal of benzyl protective group, under the conditions simi-
lar to those already used for the conversion of 9 to 2, afforded
the (+)-muricatacin analogue ent-2.
silver oxide in refluxing ether, in the presence of a catalytic
amount of silver triflate, to afford the corresponding 5-O-
decyl derivative 12 in 57% yield. Cleavage of the cyclohexyli-
dene functionality in 12 with aq. AcOH gave the correspond-
ing lactol 13 in 84% yield. The product 13 was mainly the β
anomer since its chloroform solution mutarotated to a positive
1
equilibrium value {[α]D = −3.4 → +5.6 (71 h)}. The H NMR
spectrum recorded 48 h after dissolution of the crystalline sam-
ple 13 confirmed the presence of both α- and β-anomers, with
the former predominating (α/β = 5:2). A periodate diol clea-
vage in 13, followed by in situ trapping of the resulting alde-
hyde 13a with Ph3P=CHCOMe, under the Taylor conditions
[25] gave the expected unsaturated ester 14 as 1:1 mixture of
corresponding E- and Z-isomers. The mixture was not sepa-
rated, but was further hydrogenated over 10% Pd/C. Hydroge-
nolytic removal of the benzyl ether protective group was fol-
lowed by a partial migration of the formyl group from O-5 to
O-4, whereupon the expected saturated ester 15 was obtained,
along with an equal amount of the 4-O-formyl regioisomer 16.
The mixture of 15 and 16 was treated with aqueous trifluoroa-
cetic acid to afford target 2 in 31% overall yield with respect to
the starting compound 5 (from seven steps).
By using the former multistep sequences, compound 5 was
converted to the (+)-enantiomer ent-2 in an overall yield of
16%, while the (–)-enantiomer 2 was obtained in an overall
yield of 13% with respect to starting compound 5. In order to
improve the yield of 2 we have envisioned an alternative route
to target 2 via the lactol 13 (Scheme 3). Compound 13 repre-
sents a possible divergent intermediate in the synthesis of both
analogues 2 and 3. The primary alcohol 11 easily obtainable
from 5 in 89% yield [20] was treated with decyl bromide and
After preparation of (+)- and (–)-muricatacin analogues 2
and ent-2, we were also interested in obtaining the lactone 3.
In fact, 3 is a conformationally constrained one-carbon homo-
logue of ent-2, but at the same time it might be considered as a
non-styryl analogue of (+)-goniofufurone (4). A potent and
selective cytotoxicity of certain dephenyl goniofufurone deri-

V. Popsavin et al. / European Journal of Medicinal Chemistry 41 (2006) 1217–1222
1219
Scheme 3. Reagents and conditions: (a) C10H21Br, Ag2O, AgOTf, Et2O, reflux, 56 h, 57%; (b) (i) 7:3 AcOH/H2O, reflux, 5 h; (ii) 0.1 M NaOMe/MeOH, r.t., 0.5 h,
84%; (c) aq. NaIO4, silica gel, CH2Cl2, r.t., 1 h; (d) Ph3P=CHCO2Me, CH2Cl2, r.t., 1 h; (e) H2–Pd/C, EtOH, r.t., 19 h; (f) 2:1 TFA/H2O, r.t., 1 h, 72% of 2 (from 13);
(g) Ph3P=CHCO2Me, MeOH, 0 °C → r.t., 25 h, 50%; (h) H2–Pd/C, EtOH, r.t., 20 h, 86%.
vatives [17] prompted us to prepare the lactone 3 and to eval-
uate its antitumour activity. We have assumed that the bicyclic
core of 3 might be accessible by a spontaneous lactonisation
and Michael ring-closure accompanying the Z-selective Wittig
reaction of a stabilised ylide with the furanose lactol 13, under
the reaction conditions similar to those earlier reported [17,26].
Thus, the lactol 13 was treated with Ph3P=CHCO2Me in dry
methanol to give the expected furanolactone 17 in 50% yield.
A minor amount of the corresponding E-enoate (<10%) was
also obtained, however this side-product could not be isolated
in pure form, since it had similar chromatographic properties as
accompanying impurities. Hydrogenolytic removal of the ben-
zyl protective group in 17 (10% Pd/C) gave an 86% yield of
target 3.
As outlined in Table 1, (–)-muricatacin analogue 2 was
inactive against the HL60 and Raji malignant cells, but showed
a significant cytotoxic activity against K562, Jurkat and HT-29
cell lines. The activity of this compound towards the Jurkat
leukaemic T cells was comparable to that observed for doxor-
ubicin. However, compound 2 was found to be significantly
more active against the K562 and HT-29 cell lines, being
approximately 4.5- and sixfold more potent than doxorubicin,
respectively. The (+)-enantiomer ent-2 showed a remarkable
cytotoxicity towards the K562 and HT-29 cell lines, being
almost two and threefold, respectively, more potent than dox-
orubicin. This analogue was also active against Jurkat cells, but
it was more than eightfold less active with respect to the refer-
ence compound (DOX). Conversely, the furanolactone 3 has
shown a potent cytotoxic activity against the HL60 and Raji
malignant cells, comparable to that observed for the reference
compound, doxorubicin. Such a cytotoxicity profile of 3 nicely
complements the observed biological effects of 2 and ent-2,
since these analogues were devoid of any significant cytotoxi-
city against both HL60 and Raji cells. Finally, neither of the
lactones 2, ent-2 and 3 exhibited any antiproliferative activity
towards the normal foetal lung MRC-5 cells.
2.2. Antitumour activity
Compounds 2, ent-2 and 3 were evaluated for their in vitro
cytotoxicity against human myelogenous leukaemia K562,
human promyelocytic leukaemia HL60, human T cell leukae-
mia (Jurkat), human colon adenocarcinoma HT-29, human
Burkitt’s lymphoma (Raji cells) and normal foetal lung MRC-
5 cells. Cytotoxic activity was evaluated by using the modified
MTT assay [27], after exposure of cells to the test compounds
for 24 h. The doxorubicin (DOX) was used as a reference com-
pound.
In summary, we have synthesized two novel muricatacin
analogues 2 and ent-2, as well as a new non-styryl analogue
of goniofufurone 3 starting from D-xylose. The targets 2, ent-2
and 3 have shown a significant antiproliferative activity against
Table 1
In vitro cytotoxicity of 2, ent-2, 3 and DOX
Compounds
IC50 (μM)a
K562
0.18
0.45
> 100
0.82
HL60
> 100
> 100
4.14
Jurkat
1.95
10.80
> 100
1.28
HT-29
0.086
0.18
> 100
0.51
Raji
MRC-5
> 100
> 100
> 100
0.32
2
> 100
79.19
7.37
ent-2
3
DOX
2.96
6.45
a
IC50 is the concentration of compound required to inhibit the cell growth by 50% compared to an untreated control.

1220
V. Popsavin et al. / European Journal of Medicinal Chemistry 41 (2006) 1217–1222
certain human neoplastic cells, but were completely inactive
towards the normal MRC-5 cell line. Both enantiomers 2 and
ent-2 were inactive against HL60 and Raji cells, but showed a
potent cytotoxic activity against K562, Jurkat and HT-29 cell
lines. Their activities are particularly notable in the case of
K562 and HT-29 cells being twofold up to sixfold higher
than that of doxorubicin. On the contrary, the furanolactone 3
showed a strong cytotoxicity towards the HL60 and Raji neo-
plastic cells, but was devoid of any cytotoxicity against the
K562, Jurkat and HT-29 cell lines. To the best of our knowl-
edge, none of muricatacin analogues were hitherto reported to
exhibit such a potent and selective antitumour activity. More-
over, the presented divergent approach is suitable for the synth-
esis of variety of muricatacin or goniofufurone oxa-analogues
and therefore, it may be of use in search for new antineoplastic
agents derived from the parent molecules.
80.51 (C-5), 127.79, 127.86, 128.40, 128.50 and 137.71 (Ph),
177.36 (C=O); CI MS: m/z 237 [M + H]+.
3.1.2. General procedure for the preparation of muricatacin 7-
oxa-analogues
To a stirred solution of 9 or ent-9 (1 eq) in dry Et2O
(0.3 mmol/ml) were added successively Ag2O (2.5 eq), AgOTf
(0.25 eq) and alkyl bromide (5 eq). The mixture was stirred
under reflux for 6–7 hours, than filtered and evaporated. The
residue was purified by flash column chromatography (CH2Cl2
→ 9:1 CH2Cl2/EtOAc) to give pure intermediates 10 (71%) or
ent-10 (80%). To a stirred solution of 10 or ent-10 (1 eq) in
EtOAc (0.1 mmol/ml) was added 10% Pd/C (0.18 g/mmol of
10 or ent-10). The suspension was hydrogenated at r.t. for 19 h,
than filtered through a Celite pad and washed with EtOAc. The
combined filtrates were evaporated and the residue purified by
flash chromatography (CH2Cl2 → 9:1 CH2Cl2/EtOAc).
3. Experimental section
3.1.2.1. 6-O-Decyl-2,3-dideoxy-D-treo-hexono-1,4-lactone (2).
Yield: 62% (from 9). Selected data: m.p. 42 °C (from
CH2Cl2/hexane); [α]D = −33.0 (c 0.8); H NMR (CDCl3): δ
3.1. Chemistry
1
Melting points (m.p.) were determined on a Büchi 510
apparatus and were not corrected. Optical rotations were mea-
sured on a P 3002 (Krüss) polarimeter in chloroform solutions.
NMR spectra were recorded on a Bruker AC 250 E instrument
and the chemical shifts (δ-scale) are expressed in ppm values
downfield from tetramethylsilane. Chemical ionisation mass
spectra were recorded on Finnigan-MAT 8230 spectrometer
with isobutane as a reagent gas. TLC was performed on DC
Alufolien Kieselgel 60 F254 (E. Merck). Flash column chroma-
tography was performed using Kieselgel 60 (0.040–0.063, E.
Merck). All organic extracts were dried with anhydrous
Na2SO4. Organic solutions were concentrated in a rotary eva-
porator under diminished pressure at a bath temperature below
35 °C.
0.85 (t, 3 H, J = 6.4 Hz, CH3), 1.13–1.63 (m, 16 H, 8 × CH2
from side chain), 2.24 (m, 2 H, 2 × H-3), 2.39–2.72 (m, 2 H,
Jgem = 16.9 Hz, 2 × H-2), 2.87 (d, 1 H, exchangeable with
D2O, J5,OH = 5.2 Hz, OH), 3.44 (t, 2 H, J = 6.6 Hz, H-8),
3.51 (d, 2 H, J5,6 = 6.1 Hz, 2 × H-6), 3.78 (m, 1 H, J4,5 = 3.4,
J5,6 = 6.1, J5,OH = 5.2 Hz, H-5), 4.56 (td, 1 H, J3,4 = 7.0,
J4,5 = 3.2 Hz, H-4); 13C NMR (CDCl3): δ 13.99 (CH3),
22.56, 23.74, 25.96, 28.27, 29.20, 29.34, 29.45, 29.47 and
31.78 (8 × CH2 from side chain, C-2 and C-3), 71.12 (C-6),
71.69 (C-8), 71.94 (C-5), 79.79 (C-4), 177.53 (C-1); CI MS:
m/z 573 (2M+ + H), 287 (M+ + H).
3.1.2.2. 6-O-Decyl-2,3-dideoxy-L-treo-hexono-1,4-lactone (ent-
2). Yield: 66% (from ent-9). Selected data: m.p. 42 °C (from
CH2Cl2/hexane); [α]D = +33.0 (c 0.8); IR, NMR and mass
spectral data were identical to those recorded for the enantio-
mer 2.
3.1.1. 5-O-benzyl-2.3-dideoxy-D-threo-hexono-1,4-lactone (9)
To a stirred solution of lactol 6 (0.1022 g, 0.34 mmol) in
CH2Cl2 (1.5 ml), was added Kieselgel 60 (0.063–0.2 mm;
0.7 g) and 0.65 M aq. NaIO4 (0.7 ml, 0.45 mmol). The mixture
was vigorously stirred at room temperature (r.t.) for 0.5 h, fil-
tered and the precipitate washed with CH2Cl2. The combined
organic solutions were dried and evaporated to afford the crude
aldehyde 7 as a colourless oil. To a cooled (0 °C) and stirred
solution of crude 7 in MeOH (1 ml) was added NaBH4
(0.013 g, 0.34 mmol) and the mixture was stirred at 0 °C for
1.5 h. To the solution was added TFA (1.4 ml) and the result-
ing mixture was stirred for 2 h at r.t. and than co-evaporated
with toluene (2 × 20 ml). Flash column chromatography (4:1
CH2Cl2/EtOAc) of the residue gave pure 9 (0.0407 g, 50%)
as a colourless oil, [α]D = −23.1 (c, 1.05); 1H NMR (CDCl3): δ
1.87–2.32 (m, 2 H, 2 × H-3), 2.36–2.85 (m, 3 H, 2 × CH-2 and
OH), 3.50 (m, 1 H, J5,6a = 4.9, J5,6b = 5.3, J4,5 = 5.0 Hz, H-5),
3.70 (dd, 1 H, J6a,6b = 11.6 Hz, H-6a), 3.79 (dd, 1H, H-6b),
4.65 and 4.73 (2 × d, 2 H, Jgem = 11.6 Hz, PhCH2), 4.70 (m,
1 H, H-4), 7.28–7.45 (m, 5 H, Ph); 13C NMR (CDCl3): δ 24.08
(C-3), 28.22 (C-2), 60.93 (C-6), 72.99 (PhCH2), 80.22 (C-4),
3.1.3. 3-O-Benzyl-1,2-O-cyclohexylidene-5-O-decyl-α-D-
xylofuranose (12)
To a solution of 11 (0.9556 g, 2.98 mmol) in dry Et2O
(19.5 ml) were added successively Ag2O (1.7292 g,
7.46 mmol), AgOTf (0.1905 g, 0.74 mmol) and 1-
bromodecane (1.55 ml, 7.45 mmol). The mixture was stirred
under reflux for 56 h, than filtered and evaporated. The residue
was purified by flash chromatography (19:1 hexane/EtOAc →
9:1 hexane/EtOAc), to give pure 12 (0.7847 g, 57%) as a col-
ourless oil, [α]D = −25.2 (c, 1.56); 1H NMR (CDCl3): δ 0.90 (t,
3 H, J = 6.3 Hz, Me), 1.08–1.82 (m, 26 H, 8 × CH2 from side
chain and 5 × CH2 from C6H10), 3.48 (m, 2 H, OCH2 from side
chain), 3.68 (dd, 1 H, J5a,5b = 9.8, J4,5a = 6.2 Hz, H-5a), 3.73
(dd, 1 H, J4,5b = 6.3 Hz, H-5b), 4.00 (d, 1 H, J3,4 = 3.1 Hz, H-
3), 4.38 (td, 1 H, H-4), 4.55 and 4.69 (2 × d, 2 H,
Jgem = 12.0 Hz, PhCH2), 4.60 (d, 1 H, J1,2 = 3.8 Hz, H-2),
5.94 (d, 1 H, H-1), 7.26–7.40 (m, 5 H, Ph); 13C NMR
(CDCl3): δ 14.0 (Me), 22.58, 23.52, 23.80, 24.85, 26.04,

V. Popsavin et al. / European Journal of Medicinal Chemistry 41 (2006) 1217–1222
1221
29.22, 29.40, 29.47, 29.50, 29.61, 31.80, 35.78 and 36.41
(8 × CH2 from side chain and 5 × CH2 from C6H10), 67.78
(C-5), 71.65 (OCH2 from side chain), 71.96 (PhCH2), 79.07
(C-4), 81.88 (C-3), 81.98 (C-2), 104.54 (C-1), 112.21 (Cq from
C6H10), 127.41, 127.68, 128.30 and 137.68 (Ph); CI MS: m/z
461 (M+ + H).
0.5 H, J2,3 = 12.0 Hz, H-2, Z-isomer), 6.15 (dd, 0.5 H,
J2,3 = 15.8 Hz, J2,4 = 5.9 Hz, H-2, E-isomer), 6.16 (m, 0.5 H,
H-3, Z-isomer), 6.86 (m, 0.5 H, H-3, E-isomer), 7.21–7.40 (m,
5 H, Ph), 8.10 (s, 1 H, CHO). 13C NMR (CDCl3): δ 13.99
(Me), 22.56, 25.92, 29.21, 29.32, 29.34, 29.40, 29.46 and
31.78 (8 × CH2 from side chain), 51.39 and 51.62 (OMe),
68.21 and 68.63 (C-6), 71.38, 71.59, 71.74 and 71.87
(PhCH2 and C-8), 72.92 and 73.56 (C-4 and C-5, Z-isomer),
73.02 (C-5, E-isomer), 76.19 (C-4, E-isomer), 122.98 (C-2, Z-
isomer), 123.78 (C-2, E-isomer), 127.69, 127.73, 127.79,
127.84, 128.22, 128.35, 129.63, 137.17 and 137.58 (Ph),
143.29 (C-3, E-isomer), 145.80 (C-3, Z-isomer), 160.18 and
160.43 (CHO),165.74 and 166.0 (C-1). A solution of 14
(0.1495 g, 0.34 mmol) in EtOH (3 ml) was hydrogenated
over 10% Pd/C (0.0718 g) for 19.5 h at r.t. The suspension
was filtered through a Celite pad and washed with EtOAc.
The combined filtrates were evaporated and the residue puri-
fied by flash column chromatography (7:3 CH2Cl2/Et2O), to
afford an inseparable 1:1 mixture of regioisomers 15 and 16
3.1.4. 3-O-Benzyl-5-O-decyl-D-xylofuranose (13)
A solution of 12 (0.8189 g, 1.78 mmol) in 70% aq. AcOH
was refluxed for 5 h and evaporated by co-distillation with
toluene. The residue was dissolved in dry MeOH (1 ml), and
treated with 0.1 M NaOMe in MeOH (1 ml, 0.1 mmol) at r.t.
for 0.5 h. The mixture was first neutralised with 1 M AcOH in
MeOH (0.1 ml, 0.1 mmol) and than co-evaporated with
toluene. Flash column chromatography (4:1 CH2Cl2/EtOAc)
of the residue gave pure 13 (0.5722 g, 84%) as a colourless
solid. Recrystallisation from CH2Cl2/hexane gave colourless
crystals, m.p. 68–70 °C; [α]D = −3.4 → +5.6 (c, 1.1; 71 h);
1
1
anomeric ratio (from H NMR): α/β = 5:2; H NMR (CDCl3 +
D2O): δ 0.89 (t, 3 H, J = 6.4 Hz, Me), 1.20–1.65 (m, 16 H,
8 × CH2 from side chain), 3.46 (m, 2 H, OCH2 from side
chain), 3.61 (d, 2 H, J4,5 = 5.7 Hz, 2 × H-5α), 3.65 (dd, 0.4 H,
J4,5a = 5.3, J5a,5b = 10.4 Hz, H-5aβ), 3.71 (dd, 0.4 H,
J4,5b = 4.0 Hz, 5bβ), 4.00 (m, 1 H, H-3), 4.19 (dd, 1 H,
J1,2 = 4.3, J2,3 = 2.9 Hz, H-2α), 4.25 (d, 0.4 H, J2,3 = 2.1 Hz,
H-2β), 4.38 (m, 0.4 H, H-4β), 4.44 (m, 1 H, H-4α), 4.54 and
4.70 (2 × d, 2 H, Jgem = 11.6 Hz, PhCH2α), 4.58 and 4.68
(2 × d, 0.8 H, Jgem = 11.6 Hz, PhCH2β), 5.13 (bs, 0.4 H, H-
1β), 5.49 (d, 1 H, J1,2 = 4.3 Hz, H-1α), 7.29–7.41 (m, 5 H, Ph);
13C NMR (CDCl3 + D2O): δ 14.05 (Me), 22.62, 25.98, 26.03,
29.26, 29.40, 29.43, 29.46, 29.49, 29.52, 29.55 and 31.84
(8 × CH2 from side chain), 69.19 (C-5β), 69.44 (C-5α),
71.77, 71.80 and 71.84 (PhCH2 and OCH2 from side chain),
72.59 (C-2α), 77.26 (C-4α), 79.40 (C-2β), 79.79 (C-4β), 83.52
(C-3α), 95.96 (C-1α), 103.37 (C-1β), 127.48, 127.65, 127.67,
127.94, 128.33, 128.45, 137.40 and 137.83 (Ph); CI MS: m/z
381 (M+ + H).
1
(0.0963 g, 81%) as a colourless oil. H NMR (CDCl3, selected
data): δ 2.78 and 3.20 (2 × d, 1 H each, 2 × OH), 3.40 and 3.42
(m, 4 H, 2 × H-8), 3.65 (s, 3 H, OMe), 5.00 and 5.09 (2 × m, 1
H each, CH-OCHO and CH-OH), 8.10 and 8.16 (2 × s, 1 H
each, CHO). 13C NMR (CDCl3, selected data): δ 13.98 (Me),
51.59 and 51.63 (OMe), 70.21 and 70.94 (C-OH), 72.95 and
73.92 (C-OCHO), 160.66 and 160.53 (OCHO), 173.15 and
174.02 (C-1). A solution of purified mixture 15 and 16
(0.0953 g, 0.275 mmol) in a mixture of TFA (1.3 ml) and
water (0.6 ml) was stirred at r.t. for 1 h and than co-
evaporated with toluene (2 × 20 ml). Flash chromatography
(9:1 CH2Cl2/EtOAc) of the residue, followed by recrystallisa-
tion from a mixture of CH2Cl2/hexane, gave pure 2 (0.0766 g,
97%). Physical constants and spectral data of 2 thus obtained
were in excellent agreement with those presented in the proce-
dure, see Section 3.1.2.1.
3.1.6. 3,6-Anhydro-5-O-benzyl-7-O-decyl-2-deoxy-D-ido-
heptono-1,4-lactone (17)
3.1.5. Alternative procedure for the preparation of 2
To a stirred and cooled (0 °C) solution of 13 (0.1098 g,
0.29 mmol) in dry MeOH (3 ml) was added Ph3P=CHCOMe
(0.1035 g, 0.31 mmol). After stirring the mixture at r.t. for 1 h,
an additional amount of Ph3P=CHCOMe (0.0867 g,
0.26 mmol) was added in one portion. The mixture was stirred
at r.t. for 25 h and than evaporated. Flash chromatography (1:1
Et2O/hexane) of the residue gave pure 17 (0.0582 g, 50%) as a
To a stirred solution of 13 (0.1980 g, 0.52 mmol) in CH2Cl2
(6 ml) were added Kieselgel 60 (0.063–0.2 mm; 1.56 g) and
0.65 M aq. NaIO4 (1.04 ml, 0.68 mmol) and the mixture was
vigorously stirred at r.t. for 2 h. To the suspension was than
added Ph3P=CHCO2Me (0.3840 g, 1.15 mmol) and the stirring
at r.t. was continued for an additional 1 h. The mixture was
filtered and the liquid phase was evaporated. Flash column
chromatography (CH2Cl2), of the residue gave pure 14
(0.2089 g, 92%) as an inseparable 1:1 mixture of the corre-
1
colourless oil, [α]D = +13.1 (c, 1.0); H NMR (CDCl3): δ 0.89
(t, 3 H, J = 6.3 Hz, Me), 1.21–1.57 (m, 16 H, 8 × CH2 from
side chain), 2.70 (m, 2 H, 2 × H-2), 3.46 (m, 2 H, 2 × H-9),
3.64 (d, 2 H, 2 × H-7), 4.20 (d, 1 H, J5,6 = 4.1 Hz, H-5), 4.26
(m, 1 H, H-6), 4.59 and 4.70 (2 × d, 2 H, Jgem = 11.9 Hz,
CH2Ph), 4.92 (d, 1 H, J3,4 = 4.7 Hz, H-4), 4.98 (m, 1 H, H-
3), 7.27–7.42 (m, 5 H, Ph); 13C NMR (CDCl3): δ 14.07 (Me),
22.63, 26.06, 29.26, 29.42, 29.52, 29.54, 29.60 and 31.84
(8 × CH2 from side chain), 35.96 (C-2), 68.50 (C-7), 71.79
(C-9), 72.68 (CH2Ph), 76.77 (C-3), 79.58 (C-6), 81.45 (C-5),
85.45 (C-4), 127.68, 128.09, 128.52 and 137.13 (Ph), 175.28
(C-1); CI MS: m/z 405 (M+ + H).
1
sponding Z- and E/isomers. H NMR (CDCl3): δ 0.88 (t, 3 H,
J = 6.3 Hz, Me), 1.16–1.60 (m, 16 H, 8 × CH2 from side
chain), 3.40 (m, 2 H, 2 × H-8), 3.52 (dd, 0.5 H, J5,6a = 5.7,
J6a,6b = 10.6 Hz, H-6a, E-isomer), 3.61 (dd, 0.5 H,
J5,6b = 4.5 Hz, H-6b, E-isomer), 3.63 (d, 1 H, J = 4.6 Hz,
2 × H-6, Z-isomer), 3.71 and 3.76 (2 × s, 3 H each, OMe),
4.30 (td, 0.5 H, J3,4 = 5.9, J4.5 = 6.1, J2,4 = 5.9 Hz, H-4, E-
isomer), 4.44–4.66 (m, 4 H, 2 × PhCH2), 5.18 (m, 0.5 H, H-
5, E-isomer), 5.33 (m, 0.5 H, H-4 and H-5, Z-isomer), 5.99 (d,

1222
V. Popsavin et al. / European Journal of Medicinal Chemistry 41 (2006) 1217–1222
3.1.7. 3,6-Anhydro-7-O-decyl-2-deoxy-D-ido-heptono-1,4-
lactone (3)
Acknowledgments
This work was supported by a research grant from the Min-
istry of Science and Environment Protection of the Republic of
Serbia (Grant No. 142005). The authors are grateful to Mr.
Dejan Djoković (Faculty of Chemistry, University of Belgrade,
S&M) for recording the CI mass spectra.
A solution of 17 (0.1508 g, 0.37 mmol) in dry EtOH (3 ml)
was hydrogenated over 10% Pd/C (0.0755 g) at r.t. for 20 h.
After the same workup as described above (Procedure 3.1.2),
followed by chromatographic purification on a column of flash
silica (4:1 Et2O/hexane → Et2O), pure 3 (0.1005 g, 86%) was
obtained as
a colourless solid. Recrystallisation from
References
CH2Cl2/hexane gave colourless crystals, m.p. 59–60 °C,
[α]D = +35.4 (c, 0.45); 1H NMR (CDCl3): δ 0.86 (t, 3 H,
J = 6.4 Hz, Me), 1.10–1.65 (m, 16 H, 8 × CH2 from side
chain), 2.64 (d, 1 H, J2a,2b = 18.7, H-2a), 2.75 (dd, 1 H,
J2a,2b = 18.7, J2b,3 = 5.4 Hz, H-2b), 3.50–3.59 (m, 2 H,
2 × H-9), 3.85 (m, 2 H, 2 × H-7), 4.10 (m, 1 H, H-6), 4.25
(bs, 1 H, exchangeable with D2O, OH), 4.50 (d, 1 H,
J5,6 = 3.1 Hz, H-5), 4.85 (d, 1 H, J3,4 = 4.2 Hz, H-4), 5.01
(dd, 1 H, J2b,3 = 5.4, J3,4 = 4.2 Hz, H-3); 13C NMR (CDCl3): δ
14.02 (Me), 22.58, 25.87, 29.20, 29.28, 29.34, 29.43 and 31.78
(8 × CH2 from side chain), 36.00 (C-2), 69.41 (C-7), 72.49 (C-
9), 75.76 (C-5), 76.49 (C-3), 78.59 (C-6), 88.19 (C-4), 175.45
(C-1); CI MS: m/z 629 (2M+ + H), 315 (M+ + H).
[1] A. Bermejo, B. Figadère, M.-C. Zafra-Polo, I. Barrachina, E. Estornell,
D. Cortes, Nat. Prod. Rep. 22 (2005) 269–303.
[2] M.J. Reiser, J.F. Kozlowski, K.V. Wood, J.L. McLaughlin, Tetrahedron
Lett. 32 (1991) 1137–1140.
[3] B. Dhotare, A. Chattopadhyay, Tetrahedron Lett. 46 (2005) 3103–3105
(and references cited therein).
[4] H. Konno, N. Hiura, M. Yanaru, Heterocycles 57 (2002) 1793–1797.
[5] J.M. Andres, N. de Elena, R. Pedrosa, A. Perez-Encabo, Tetrahedron
Asymmetry 12 (2001) 1503–1509.
[6] M. Pichon, J.C. Jullian, B. Figadère, A. Cavé, Tetrahedron Lett. 39
(1998) 1755–1758.
[7] S.W. Chang, C.Y. Hung, H.H. Liu, B.J. Uang, Tetrahedron Asymmetry 9
(1998) 521–529.
[8] G. Rassu, L. Pinna, P. Spanu, F. Zanardi, L. Battistini, G. Casiraghi, J.
Org. Chem. 62 (1997) 4513–4517.
[9] A. Gypser, M. Peterek, H.D. Scharf, J. Chem. Soc., Perkin Trans. 1
(1997) 1013–1016.
3.2. In vitro antitumour assay
[10] I. Baussanne, O. Schwardt, J. Royer, M. Pichon, B. Figadère, A. Cavé,
Antitumour activity was evaluated by the tetrazolium colori-
metric MTT assay. The assay is based on cleavage of the 3-
(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide
(MTT), to formazan by mitochondrial dehydrogenases in
viable cells. Exponentially growing cells were harvested,
counted by trypan blue exclusion and plated into 96-well
microtitar plates (Costar) at optimal seeding density of
5 × 103 cells per well to assure logarithmic growth rate
throughout the assay period. Viable cells were plated in a
volume of 90 μl per well, and preincubated in complete med-
ium at 37 °C for 24 h to allow cell stabilisation prior to the
addition of substances. Tested compounds, at twice the
required final concentration, in growth medium (10 μl per
well) were added to all wells except to the control ones and
microplates were incubated for 24 h. The wells containing
cells without tested compounds were used as control. Three
hours before the end of incubation period 10 μl of MTT solu-
tion was added to each well. MTT was dissolved in medium at
5 mg/ml and filtered to sterilise and remove a small amount of
insoluble residue present in some batches of MTT. Acidified 2-
propanol (100 μl of 0.04 M HCl in 2-propanol) was added to
each well and mixed thoroughly to dissolve the dark blue crys-
tals. After a few minutes at r.t. to ensure that all crystals were
dissolved, the plates were read on a spectrophotometer plate
reader (Multiscan MCC340, Labsystems) at 540/690 nm. The
wells without cells containing complete medium and MTT
only acted as blank. The compound cytotoxicity was expressed
as the IC50 (50% inhibitory concentration).
Tetrahedron Lett. 38 (1997) 2259–2262.
[11] M.P.M. Vanaar, L. Thijs, B. Zwanenburg, Tetrahedron 51 (1995) 11223–
11234.
[12] M. Saiah, M. Bessodes, K. Antonakis, Tetrahedron Lett. 34 (1993)
1597–1598.
[13] S.H. Tsai, P.C. Hsieh, L.L. Wei, H.F. Chiu, Y.C. Wu, M.J. Wu, Tetra-
hedron Lett. 40 (1999) 1975–1976.
[14] S.H. Yoon, H.S. Moon, S.K. Kang, Bull. Kor. Chem. Soc. 19 (1998)
1016–1018.
[15] S.H. Yoon, H.S. Moon, S.K. Hwang, S.R. Choi, S.K. Kang, Bioorg.
Med. Chem. 6 (1998) 1043–1049.
[16] A. Cavé, C. Chaboche, B. Figadère, J.C. Harmange, A. Laurens, J.F.
Peyrat, M. Pichon, M. Szlosek, J. CotteLafitte, A.M. Quero, Eur. J.
Med. Chem. 32 (1997) 617–623.
[17] V. Popsavin, S. Grabež, M. Popsavin, I. Krstić, V. Kojić, G. Bogdano-
vić, V. Divjaković, Tetrahedron Lett. 45 (2004) 9409–9413.
[18] V. Popsavin, S. Grabež, I. Krstić, M. Popsavin, D. Djoković, J. Serb.
Chem. Soc. 68 (2003) 795–804.
[19] V. Popsavin, S. Grabež, M. Popsavin, J. Petrović, Carbohydr. Lett. 3
(2000) 411–418.
[20] V. Popsavin, I. Krstić, M. Popsavin, Tetrahedron Lett. 44 (2003) 8897–
8900.
[21] X.P. Fang, J.E. Anderson, C.J. Chang, P.E. Fanwick, J.L. McLaughlin, J.
Chem. Soc., Perkin Trans. 1 (1990) 1655–1661.
[22] H.B. Mareyala, M. Joe, Curr. Med. Chem. Anti-Cancer Agents 1 (2001)
293–300.
[23] M.A. Blazquez, A. Bermejo, M.C. Zafra-Polo, D. Cortes, Phytochem.
Anal. 10 (1999) 161–170.
[24] W. Szeja, Pol. J. Chem. 54 (1980) 1323–1325.
[25] H.S. Outram, S.A. Raw, R.J.K. Taylor, Tetrahedron Lett. 43 (2002)
6185–6187.
[26] K.R.C. Prakash, S.P. Rao, Tetrahedron 49 (1993) 1505–1510.
[27] D.A. Scudiero, R.H. Shoemaker, K.D. Paull, A. Monks, S. Tierney, T.H.
Nofziger, M.J. Currens, D. Seniff, M.R. Boyd, Cancer Res. 48 (1988)
4827–4833.