Enantiopure hydroxylactones from D-xylose. A novel approach to the enantiodivergent synthesis of (+)- and (-)-muricatacin suitable for the preparation of 7-oxa analogues

Article abstract of DOI:10.1016/j.tetlet.2003.09.168

A new route towards enantiopure hydroxylactones 3 and ent-3, the final chiral precursors in an enantiodivergent synthesis of (+)- and (-)-muricatacin, has been developed starting from D-xylose.

Full text of DOI:10.1016/j.tetlet.2003.09.168

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 8897–8900
Enantiopure hydroxylactones from
D
-xylose. A novel approach
to the enantiodivergent synthesis of (+)- and (−)-muricatacin
suitable for the preparation of 7-oxa analogues
Velimir Popsavin,* Ivana Krstic´ and Mirjana Popsavin
Department of Chemistry, Faculty of Sciences, University of Novi Sad, Trg D. Obradovic´a 3, 21000 Novi Sad,
Serbia and Montenegro
Received 5 August 2003; revised 8 September 2003; accepted 19 September 2003
Abstract—A new route towards enantiopure hydroxylactones 3 and ent-3, the final chiral precursors in an enantiodivergent
synthesis of (+)- and (−)-muricatacin, has been developed starting from
© 2003 Elsevier Ltd. All rights reserved.
D-xylose.
Hydroxylactones of type 1 (Scheme 1) are naturally
occurring 5-hydroxyalkylbutan-4-olides that show dif-
ferent biological activity depending on the length of the
alkyl chain and absolute configuration of the stereocen-
ters. One such molecule that has attracted considerable
attention since its isolation from the seeds of Anona
muricata1 is muricatacin (5-hydroxy-4-heptadecano-
lide), an acetogenin derivative that shows cytotoxic
activity against certain human tumor cell lines. Interest-
ingly, the isolated sample was a mixture of enantiomers
4a and 4b with the (−)-(R,R)-isomer 4b being predomi-
nant (ee, ca. 25%). Both (+)- and (−)-muricatacin show
the same antitumor activity.1,2 The biological activity of
muricatacin and other related compounds, has stimu-
lated significant interest in the synthesis of 5-hydroxy-
alkylbutan-4-olides. Many syntheses of (+)- and/or
(−)-muricatacin from various non-carbohydrate precur-
sors have been reported,2,3 along with a number of
carbohydrate based approaches,4,5 most being target
oriented. Herein we report on a novel general approach
to an enantiodivergent synthesis of (+)- and (−)-muri-
catacin from
D-xylose, that is suitable for elaboration
to a variety of 7-oxa analogues.
As outlined in Scheme 1, (+)-muricatacin 4a might be
prepared by a sequence that will ensure the introduc-
tion of the C-2 and C-3 stereocenters of
D-xylose into
the target structure 4a via the aldehydo-lactone 2. It
was further assumed that the key intermediate 2 should
be available from a suitably protected
D-xylose deriva-
tive through the following steps: (i) ‘CH2CO2R’—intro-
duction at C-1 followed by g-lactonization, and (ii)
Scheme 1. (a) Naturally occurring 5-hydroxyalkylbutan-4-olides (conventional numbering); (b) enantiodivergent strategy for
preparation of (+)- and (−)-muricatacin by chirality transfer from -xylose (sugar numbering).
D
Keywords: 5-hydroxyalkylbutan-4-olides;
D-xylose; muricatacin; enantiodivergent synthesis; Wittig reaction.
* Corresponding author. Tel.: +381-21-350-122; fax: +381-21-454-065; e-mail: popsavin@ih.ns.ac.yu
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.09.168

8898
V. Popsa6in et al. / Tetrahedron Letters 44 (2003) 8897–8900
oxidative glycol cleavage of the C4ꢀC5 bond. An alter-
native sequence that involves (iii) ‘CH2CO2R’—elonga-
tion at C-5 followed by g-lactonization, and (iv) C1ꢀC2
glycol cleavage, should provide access to the aldehydo-
gave the corresponding lactol 7. Wittig olefination of 7
with (carbomethoxymethylidene)triphenylphosphorane
in DMF took place stereospecifically to afford the
(E)-unsaturated ester 8 (83%) as the only isolable
product. The E-selectivity of this step was essential,
because it is well known that similar (Z)-a,b-unsatu-
rated esters rapidly undergo a sequential lactonization/
Michael ring-closure process.8 Catalytic hydrogenation
of 8 over PtO2 in ethanol yielded the corresponding
saturated ester 9, which upon treatment with sodium
methoxide in methanol furnished the hydroxylactone
10 in 82% yield. Oxidative cleavage of the diol func-
tionality in 10 was achieved by treatment with NaIO4-
impregnated wet silica in dichloromethane, whereby
the aldehydo-lactone 29 was obtained. In the light of
its stereochemical features the molecule 2 fully corre-
sponds to the chiral lactone core of (+)-muricatacin 4a,
and to the related 5-hydroxyalkylbutan-4-olides of type
1.
lactone ent-2 bearing the C3ꢀC4 chiral segment of
D-
xylose. It was also assumed that both 2 and ent-2 can
be converted to the targets 4a and 4b by a Wittig
elongation/catalytic reduction process. Alternatively,
the chiral synthons 2 and ent-2 may be first converted
to the dihydroxylactones 3 and ent-3 and finally to the
targets 4a and 4b through a known two-step sequence.5
The syntheses of 2 and ent-2 are summarized in
Scheme 2. The preparation of 2 started with 5-O-ben-
zoyl-1,2-O-cyclohexylidene-a-
D
-xylofuranose 5 which
was readily available from
D
-xylose in three steps.6,7
Treatment of 5 with benzyl bromide in DMF, in the
presence of NaH as a base, followed by removal of the
cyclohexylidene protective group with dilute acetic acid
Scheme 2. Reagents and conditions: (a) BnBr, NaH, DMF, 0°C“rt, 1.5 h, 74%; (b) 7:3 AcOH–H2O, reflux, 5.5 h, 85%; (c)
Ph3P:CHCO2Me, DMF, 60–70°C, 3.5 h, 83%; (d) H2/PtO2, EtOH, rt, 16 h for 8, 76% of 9, 19 h for 13, 92% of 14; (e) (i) NaOMe,
MeOH, rt, 1.5 h, (ii) 2:1 TFA–H2O, rt, 10 min, 82%; (f) aq. NaIO4, silica gel, CH2Cl2, rt, 1 h, 89% of 2, 52% of 17; (g) NaOMe,
MeOH, 0°C“rt, 1 h, 86%; (h) DMSO, DCC, anh. H3PO4, Py, rt, 5.5 h, 76% (83% based on recovered 11); (i) Ph3P:CHCO2Me,
CH2Cl2, rt, 2 h, 97%; (j) 1:1 AcOH–H2O, reflux, 1.5 h, 58% of 15 (64% based on recovered 14), 7.5% of 16; (k) 2:1 TFA–H2O,
rt, 2 h, 77% from 15; (l) [Ph3PCH2(CH2)9Me]+ Br−, LiHMDS, THF, −78°C“rt, 20 h for 12, 3 days for 2; 43% of 19, 6.5% of 20.

V. Popsa6in et al. / Tetrahedron Letters 44 (2003) 8897–8900
8899
Scheme 3. Reagents and conditions: (a) (i) aq. NaIO4, silica gel, CH2Cl2, rt, 1 h, (ii) NaBH4, MeOH, 0°C“rt, 1.5 h, (iii) 2:1
TFA–MeOH, 0°C“rt, 1.5 h, 64% (73% based on recovered 2); (b) H2–Pd/C, EtOAc, rt, 19 h, 69%; (c) (i) NaBH4, MeOH,
0°C“rt, 2 h, (ii) 2:1 TFA–MeOH, 0°C“rt, 1.5 h, (iii) H2–Pd/C, rt, 19 h, 71%.
The 5-O-benzoyl-3-O-benzyl-1,2-O-cyclohexylidene-a-
-xylofuranose 6 was conveniently used as an intermedi-
mixture. All our attempts to improve the yield of 20 were
unsuccessful and afforded only the elimination product
21. We were therefore forced to find an alternative
methodology for elaboration of the muricatacin side
chain. According to our plan (Scheme 1), conversion of
both 2 and ent-2 to the corresponding diols 3 and ent-3
(Scheme 3) represents a possible alternative route for
completion of the synthesis.
D
ate for the preparation of ent-2. Treatment of 6 with
sodium methoxide in methanol afforded the primary
alcohol 11 in 86% yield (64% from 5). However, when
the last two steps (5“6“11) were carried out as an
one-pot procedure, the intermediate 11 was obtained in
89% overall yield with respect to 5. Oxidation of the
primary hydroxyl group in 11 gave 12, which was further
treated with (carbomethoxymethylidene)triphenylphos-
phorane in dry dichloromethane to afford the expected
unsaturated ester 13 as a 2:1 mixture of the correspond-
ing Z- and E-isomers. Catalytic hydrogenation of 13,
followed by hydrolytic removal of the cyclohexylidene
protective group, gave a 58% yield of the corresponding
lactol 15 (64% based on recovered 14), accompanied with
a small amount of the carboxylic acid 16 (7.5%). Oxida-
tive cleavage of purified diol 15 with sodium periodate
on silica afforded the formate 17, which upon treatment
with aqueous trifluoroacetic acid yielded the g-lactone
ent-2,10 with the absolute configuration of both stere-
ocenters corresponding to (−)-muricatacin 4b. Spectral
data (1H and 13C NMR) and physical constants of ent-2
were in good agreement with values recorded for its
enantiomer 2.
The preparation of 3 began with the synthesis of the
primary alcohol 22 from dihydroxy-lactone 10. Oxidative
cleavage of the terminal diol in 10 provided the aldehydo-
lactone 2, which was isolated in pure form after the usual
work-up and used in the next step without further
purification. Subsequent reduction of crude 2 with
sodium borohydride gave the expected primary alcohol
22 along with an equal amount of ester 23. The mixture
was not separated (except for characterization purposes),
but was further treated with trifluoroacetic acid to
complete the lactonization of 23 to 22. The intermediate
22 was thus obtained in a 64% overall yield with respect
to the starting compound 10 (73% based on recovered 2).
Catalytic hydrogenolysis of 22 (10% Pd/C) furnished the
known diol 3, a key intermediate in the synthesis of
conformationally constrained analogues of diacyl-
glycerol.12 The 1H and 13C NMR spectral data (Table 1)
and the optical rotation13 of 3 thus obtained were in
reasonable agreement with reported values.12 Compound
3 can be converted to (+)-muricatacin according to the
reported procedure.5 The intermediate ent-2 was con-
verted to the dihydroxylactone ent-3 by using an one-pot
procedure that involved a reduction of the aldehyde
group (NaBH4 in MeOH) in ent-2, followed by subse-
quent hydrogenolytic removal of the benzyl ether protec-
tive group (10% Pd/C) under the acidic conditions (2:1
TFA–MeOH). This procedure provided the desired inter-
mediate ent-3 in 71% overall yield. The 1H and 13C NMR
spectral data (Table 1), as well as the value of optical
rotation14 for ent-3 were fully consistent with those
reported previously.5 Since (−)-muricatacin 4b has
already been synthesized from diol ent-3,5 the prepara-
tion of ent-3 formally represents a novel synthesis of
With the requisite chirons 2 and ent-2 in hand, we next
focused on their C11-elongation in order to elaborate the
muricatacin side chain. According to the initial plan,
Wittig olefination of 2 and ent-2 with the appropriate
C11ꢀylide should enable us to resolve this problem.
However, in order to avoid wasting valuable intermedi-
ates (2 and ent-2), the Wittig reaction was first explored
on aldehyde 12 as a model compound. Thus, aldehyde
12 was reacted with ylide 18 (generated in situ from
undecyltriphenylphosphonium bromide and LiHMDS in
THF at −78°C),11 to give an acceptable yield of the
corresponding (Z)-olefin 19 (43%), as the only isolable
product. The aldehydo-lactone 2 under the same reaction
conditions also gave the desired unsaturated derivative
20 but in only 6.5% yield. Traces of the elimination
product 21 were also isolated from a complex reaction
(−)-4b from
D-xylose.

8900
V. Popsa6in et al. / Tetrahedron Letters 44 (2003) 8897–8900
Table 1. NMR spectral data for compounds 3 and ent-3
Compound
Solvent
CDCl3
Methanol-d4 2.53
CDCl3
CDCl3
lH (ppm)
lC (ppm)
Reference
2×H-2
2×H-3
H-4
H-5 and
2×H-6
C-1
C-2
C-3
C-4
C-5
C-6
3
2.57
2.26
2.25
2.28
2.23
4.59
4.68
4.60
4.57
3.65
3.60
3.75
3.69
178.21
180.44
177.31
178.00
28.45
29.34
28.31
28.40
23.90
24.68
23.89
23.90
80.77
81.79
80.62
80.70
73.56
74.53
73.42
73.50
63.27
63.77
63.28
63.30
This work
Ref. 12
This work
Ref. 5
ent-3
2.60
2.52
In conclusion, a new and flexible strategy for the syn-
thesis of enantiopure 5-hydroxyalkylbutan-4-olides by
mann, W.; Scholz, G.; Bunte, G.; Wolff, C.; Peters, E. M.;
Peters, K.; Von Schnering, H. G. Liebigs Ann. Chem. 1992,
1069–1080; (q) Marshall, J. A.; Welmaker, G. S. Synlett
1992, 537–538; (r) Figade`re, B.; Haramange, J. C.; Lau-
rens, A.; Cave´, A. Tetrahedron Lett. 1991, 32, 7539–7542;
(s) Scholz, G.; Tochtermann, W. Tetrahedron Lett. 1991,
32, 5535–5538.
chirality transfer from
D-xylose has been developed.
The synthetic pathway that provided an access to (+)-
and (−)-5,6-dihydroxy-4-hexanolides 3 and ent-3 for-
mally represents a new enantiodivergent synthesis of
(+)- and (−)-muricatacin. This approach is potentially
useful for the preparation of hitherto unknown 7-oxa
(+)- and (−)-muricatacin analogues 24a and 24b (via 22
and ent-22).
4. (a) Chadrasekhar, M.; Chandra, K. L.; Singh, V. K.
Tetrahedron Lett. 2002, 43, 2773–2775; (b) Popsavin, V.;
Grabez, S.; Popsavin, M.; Petrovic´, J. Carbohydrate Lett.
2000, 3, 411–418; (c) Yoon, S.-H.; Moon, H.-S.; Hwang,
S.-K.; Choi, S. R.; Kang, S. K. Bioorg. Med. Chem. 1998,
6, 1043–1049; (d) Rassu, G.; Pinna, L.; Spanu, P.; Zanardi,
F.; Battistini, L.; Casiraghi, G. J. Org. Chem. 1997, 62,
4513–4517; (e) Gravier-Pelletier, C.; Saniere, M.; Charvet,
I.; Le Merrer, Y.; Depezay, J.-C. Tetrahedron Lett. 1994,
35, 115–118; (f) Kang, S. K.; Cho, H. S.; Sim, H. S.; Kim,
B. K. J. Carbohydr. Chem. 1992, 11, 807–812.
Acknowledgements
This work was supported by a research grant from the
Ministry of Science, Technologies and Development of
the Republic of Serbia (Grant No. 1896).
5. Saniere, M.; Charvet, I.; Le Merrer, Y.; Depezay, J.-C.
References
Tetrahedron 1995, 51, 1653–1662.
6. Popsavin, M.; Popsavin, V.; Vukojevic´, N.; Miljkovic´, D.
Collect. Czech. Chem. Commun. 1994, 59, 1884–1888.
7. Szeja, W. Pol. J. Chem. 1980, 54, 1323–1325.
8. Prakash, K. R. C.; Rao, S. P. Tetrahedron 1993, 49,
1505–1510.
1. Reiser, M. J.; Kozlowski, J. F.; Wood, K. V.; McLaughlin,
J. L. Tetrahedron Lett. 1991, 32, 1137–1140.
2. Cave´, A.; Chaboche, C.; Figade`re, B.; Haramange, J. C.;
Laurens, A.; Peyrat, J. F.; Pichon, M.; Szlosek, M.;
Cotte-Lafitte, J.; Quero, A. M. Eur. J. Med. Chem. 1997,
32, 617–623.
9. Selected data for 2: mp 93–94°C (from CH2Cl2–hexane),
[h]D +135.8 (c 1.05 in CHCl3); 1H NMR (CDCl3): l
3. (a) Rangavan, S.; Joseph, S. C. Tetrahedron: Asymmetry
2003, 14, 101–105; (b) Carda, M.; Rodriguez, S.; Gonzalez,
F.; Castillo, E.; Villanueva, A.; Marco, J. A. Eur. J. Org.
Chem. 2002, 2649–2655; (c) Baylon, C.; Prestat, G.; Heck,
M.-P.; Mioskowski, C. Tetrahedron Lett. 2000, 41, 3833–
3835; (d) Couladouros, E. A.; Mihou, A. P. Tetrahedron
Lett. 1999, 40, 4861–4862; (e) Solladie, G.; Hanquet, G.;
Izzo, I.; Crumbie, R. Tetrahedron Lett. 1999, 40, 3071–
3074; (f) Szlosek, M.; Franck, X.; Figade`re, B.; Cave´, A.
J. Org. Chem. 1998, 63, 5169–5172; (g) Chang, S.-W.;
Hung, C.-Y.; Liu, H.-H.; Uang, B. J. Tetrahedron: Asym-
metry 1998, 9, 521–529; (h) Quayle, P.; Rahman, S.;
Herbert, J. Tetrahedron Lett. 1995, 36, 8087–8088; (i) van
Aar, M. P. M.; Thijis, L.; Zwanenburg, B. Tetrahedron
1995, 51, 11223–11234; (j) Bonini, C.; Federici, C.; Rossi,
L.; Righi, G. J. Org. Chem. 1995, 60, 4803–4812; (k)
Somfai, P. J. Chem. Soc., Perkin Trans. 1 1995, 817–819;
(l) Marshall, J. A.; Welmaker, G. S. J. Org. Chem. 1994,
59, 4122–4125; (m) Makabe, H.; Tanaka, A.; Oritani, T.
Biosci., Biotechnol. Biochem. 1993, 57, 1028–1029; (n)
Saiah, M.; Bessodes, M.; Antoniakis, K. Tetrahedron Lett.
1993, 34, 1597–1598; (o) Wang, Z. M.; Zhang, X. L.;
Sharpless, K. B.; Sinha, S. C.; Sinha-Bagchi, A.; Keinan,
E. Tetrahedron Lett. 1992, 33, 6407–6410; (p) Tochter-
1.92–2.68 (m, 4H, 2×CH2), 3.78 (dd, 1H, J4,5=2.7, J5,6=
1.5 Hz, H-5), 4.58 (d, 1H, Jgem=11.9 Hz, PhCH-a),
4.74–4.85 (m, 2H, H-4 and PhCH-b), 7.33 (m, 5H, Ph), 9.68
(d, 1H, H-6); 13C NMR (CDCl3): l 23.63 and 27.77
(2×CH2), 73.53 (PhCH2), 78.88 (C-4), 84.02 (C-5), 128.20,
128.52, 128.69 and 136.18 (Ph), 176.56 (C-1), 201.57 (C-6).
10. Selected data for ent-2: mp 93–94°C (from CH2Cl2–hex-
1
ane), [h]D −129.8 (c 1.0 in CHCl3); H NMR (CDCl3): l
1.90–2.65 (m, 4H, 2×CH2), 3.76 (dd, 1H, J4,5=2.7, J5,6
=
1.5 Hz, H-5), 4.60 (d, 1H, Jgem=11.9 Hz, PhCH-a),
4.72–4.84 (m, 2H, H-4 and PhCH-b), 7.37 (m, 5H, Ph), 9.69
(d, 1H, H-6); 13C NMR (CDCl3): l 23.53 and 27.72
(2×CH2), 73.45 (PhCH2), 78.84 (C-4), 83.94 (C-5), 128.19,
128.42, 128.61 and 136.16 (Ph), 176.57 (C-1), 201.84 (C-6).
11. Markidis, T.; Kokotos, G. J. Org. Chem. 2001, 66,
1919–1923.
12. Lee, J.; Lewin, N. E.; Acs, P.; Blumberg, P. M.; Marquez,
V. E. J. Med. Chem. 1996, 39, 4912–4919.
13. Compound 3 (syrup): [h]D +57.99 (c 2.01 in MeOH); Ref.
12: [h]D +58.18 (c 2.03 in MeOH); 1H and 13C NMR data
in Table 1.
14. Compound ent-3 (syrup): [h]D −40.03 (c 2.07 in MeOH);
Ref. 5: [h]D −43.3 (c 0.9 in MeOH); 1H and 13C NMR data
in Table 1.