Synthesis of a ceramide sphingolipid as a potential sex pheromone of the hair crab Erimacrus isenbeckii using butane-2,3-diacetal desymmetrised glycolic acid building blocks

Article abstract of DOI:10.1055/s-2005-862361

A stereoselective synthesis of a ceramide sphingolipid as a potential sex pheromone of the hair crab Erimacrus isenbeckii is reported using diastereoselective alkylation and aldol reactions of butane-2,3-diacetal (BDA) desymmetrised glycolic acid building

Full text of DOI:10.1055/s-2005-862361

LETTER
481
Synthesis of a Ceramide Sphingolipid as a Potential Sex Pheromone of the Hair
Crab Erimacrus isenbeckii Using Butane-2,3-diacetal Desymmetrised Glycolic
Acid Building Blocks
Synthesis of
a
Cera
a
mide
S
phin
r
golipid
r
fromDe
e
symmetrise
n
d
G
lycolic
A
cid J. Dixon,1 Steven V. Ley,* Sophie Lohmann, Tom D. Sheppard2
Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, CB2 1EW, UK
Fax +44(1223)336442; E-mail: svl1000@cam.ac.uk
Received 17 December 2004
OMe
OMe
Abstract: A stereoselective synthesis of a ceramide sphingolipid as
a potential sex pheromone of the hair crab Erimacrus isenbeckii is
reported using diastereoselective alkylation and aldol reactions of
butane-2,3-diacetal (BDA) desymmetrised glycolic acid building
blocks as the key synthetic steps.
O
O
O
O
O
O
OMe
OMe
2
3
Key words: acetals, chiral auxiliaries, pheromones, sphingolipids,
stereoselective synthesis
Figure 2 Enantiomeric butane-2,3-diacetal desymmetrised chiral
glycolate equivalents.
We have previously demonstrated the use of lactones 2
and 3 in a total synthesis of the phytotoxic agent Herbaru-
min II.9
The ceramides shown in Figure 1 were isolated by Fuset-
ani et al. from water containing post-moult female hair
crabs of the species Erimacrus isenbeckii.3 A sponge
soaked in the ceramide mixture was shown to induce
guard and copulatory behaviour in male hair crabs. Close-
ly related compounds have been previously reported
which are phospholipase A2 inhibitors4 or exhibit cyto-
toxicity against tumour cells in mice.5 The ceramide
mixture was subsequently synthesised by Fusetani and co-
workers from galactose pentaacetate and 9-bro-
mononanol,6 and the pheromone 1 was later synthesised
by Masuda et al. from 12-bromododecanol and the Garner
aldehyde.7
We intended to construct the hydroxyacid fragment 4,
using a stereoselective alkylation of glycolate 2 with allyl
bromide 5, followed by introduction of the long side chain
via a Wittig olefination with phosphonium salt 6, after
oxidative cleavage of the alkene (Scheme 1). The anti-
diol motif in amine 7 was to be constructed by an aldol
reaction between glycolate 3 and the suitably protected
serine-derived aldehyde 8.10
OH
O
OH
N
H
OH
8
15
R
O
OH
OH
OH
amide bond formation
1
N
H
n
m
OH
OH
R = H, Me
m = 14–18
n = 8–10
Wittig
TBDMSO
H2N
O
1 - R = Me, m = 15, n = 8
OTBDMS
Figure 1 Ceramides isolated from water containing post-moult
female hair crabs.
OH
15
8
OH
TBDMSO
homologation
4
selective glycolate
alkylation
selective glycolate
aldol
7
We planned to synthesise ceramide 1 using the butane-
2,3-diacetal (BDA) desymmetrised glycolic acid building
blocks developed in our group for the asymmetric synthe-
sis of a-hydroxycarbonyl compounds.8 The enantiomeric
lactones 2 and 3 (Figure 2) are readily available in multi-
gram quantities from 1-chloropropane-2,3-diol8a,f or from
mannitol or ascorbic acid8h and undergo highly diastereo-
selective reactions with a wide range of electrophiles.
OMe
OMe
TBDMSO
Bn2N
O
O
O
Br
O
H
O
OMe
5
O
OMe
2
8
3
O
I
PPh3
Br
I
12
5
6
9
Scheme 1 Retrosynthetic analysis of ceramide 1.
SYNLETT 2005, No. 3, pp 0481–0484
1
6.
0
2.
2
0
0
5
Advanced online publication: 04.02.2005
DOI: 10.1055/s-2005-862361; Art ID: D37704ST
© Georg Thieme Verlag Stuttgart · New York

482
D. J. Dixon et al.
LETTER
According to our experience, this should represent a
matched aldol reaction between the two chiral units 3 and
8, leading to a single diastereomeric product.8b,g The
amine fragment 7 would then be obtained by conversion
of the lactone functionality into an aldehyde and homolo-
gation with diiodide 9.
OMe
OMe
a
O
O
O
O
O
OMe
O
OMe
13
2
b
OMe
The required long chain phosphonium salt 6 was synthe-
sised from commercially available pentadecanolide 10
(Scheme 2). One-pot DIBALH reduction of 10 and Wittig
olefination gave alcohol 11 in excellent yield. Bromina-
tion of the resulting alcohol 11 proceeded smoothly with
triphenylphosphine dibromide11 to give bromide 12 in
99% yield, and displacement with triphenylphosphine
gave the phosphonium salt 6 in quantitative yield.
OMe
O
O
c
O
O
O
OMe
O
O
OMe
H
15
12
14
d
O
OMe
e
O
OH
O
O
17
OH
OH
a
16
O
16
OMe
O
4
12
11
Scheme
3
a) LHMDS (0.95 equiv), THF, –78 °C, then
10
CH2=CHCH2Br, –45 °C to r.t., 82%; b) O3, CH2Cl2, –78 °C, then
Me2S (2 equiv); c) n-BuLi (1.8 equiv), 6 (0.9 equiv), THF, 26% from
6; d) H2, Rh/alumina, MeOH, 87%; e) TFA, H2O, 100%.
b
PPh3
Br
Br
c
12
12
thesis of alcohol 19 constitutes an extremely concise route
to a protected form of the side chain of the phosphatase in-
hibitor calyculin A, with the correct relative (but opposite
absolute) stereochemical features.14
6
12
Scheme 2 a) DIBALH (1.5 equiv), PhMe, –78 °C, then MeOH (0.5
equiv), then n-BuLi (2 equiv), i-BuPPh3Br (2 equiv), THF, –78 °C to
r.t., 83%; b) PPh3Br2 (2.6 equiv), pyridine, MeCN, 99%; c) PPh3 (1.0
equiv), neat, 120 °C, 100%.
Deprotection of the acid labile protecting groups, fol-
lowed by global TBS protection of the alcohols gave the
tris-TBS ether 21 in 93% yield over the two steps. The es-
ter was reduced to the alcohol 22 with DIBALH in 62%
yield, then oxidised to aldehyde 23 in good yield under
Swern conditions.
With the salt 6 in hand, lactone 2 was alkylated stereose-
lectively with allyl bromide 5 to give the highly crystal-
line alkene 13 in very good yield (Scheme 3).8a,f Oxidative
cleavage of alkene 13 to the aldehyde 14 proved problem-
atic under a variety of conditions (O3, OsO4 and NaIO4,
etc.), almost certainly due to the unstable nature of 14.
Consequently, a one-pot oxidative cleavage and in situ
Wittig olefination was carried out to yield dialkene 15 di-
rectly, albeit in a moderate 26% yield. Reduction of the
double bonds with hydrogen and catalytic rhodium on
alumina gave the lactone 16 in 87% yield, and straight-
forward acidic deprotection of the BDA group provided
the desired a-hydroxyacid 4 in quantitative yield.
Attempts at Wittig olefination of aldehyde 23 with n-de-
cyltriphenylphosphonium bromide were unsuccessful
under a variety of conditions, yielding only unreacted
starting aldehyde. A report in the literature of a similarly
unreactive aldehyde15 indicated that a Takai olefination16
might be successful. The required diiodide 9 was synthe-
sised from decanal 24, following the procedure of Stern-
hell and Pross (Scheme 5).17 The subsequent olefination
between diiodide 9 and aldehyde 23 finally yielded the de-
sired alkene 26 in a moderate 33% yield. Hydrogenation
of the alkene, with concomitant removal of the benzyl
groups proceeded in good yield to give the known serino-
lipid 7.7
N,N-Dibenzyl protected aldehyde 8 (Scheme 4) was se-
lected as a suitable coupling partner for the aldol reaction
with lactone 3. The aldehyde was readily obtained by
LiBH4 reduction in diethyl ether–methanol12 of commer-
cially available protected serine 17, followed by Swern
oxidation. Due to the sensitivity of aldehyde 8,10 it was
prepared and used immediately without chromatographic
purification.
NBn2
NBn2
NBn2
b
a
H
OMe
TBDMSO
OH
TBSDMO
O
TBDMSO
O
17
18
8
The aldol reaction proceeded smoothly with lactone 3 un-
der standard conditions8b,g to yield the alcohol 19 as a sin-
gle crystalline diastereoisomer (Scheme 5). The relative
and absolute stereochemistry was readily confirmed at
this point by single crystal X-ray diffraction.13 The syn-
Scheme 4 a) LiBH4 (3.9 equiv), MeOH, Et2O, 0 °C to r.t., 95%; b)
DMSO (4.5 equiv), (COCl)2 (2.3 equiv), Et3N (9 equiv), CH2Cl2, –78
°C, 97%.
Synlett 2005, No. 3, 481–484 © Thieme Stuttgart · New York

LETTER
Synthesis of a Ceramide Sphingolipid from Desymmetrised Glycolic Acid
483
NBn2
RO
1
OR
3
OMe
OMe
TBDMSO
TBDMSO
a
2
4
5
OH
O
O
a
9
O
O
TBDMSO
9
HO
19
NH
OR
2'
O
O
OMe
OMe
NH2 OTBDMS
18
3
O
5
27 R = TBDMS
R = H
b
b
1
Scheme 6 a) 4 (1 equiv), EDC·HCl (15 equiv), HOBt (3 equiv),
CH2Cl2, 97%; b) TBAF, THF, 78%.
TBDMSO
Bn2N
O
OH
O
c
TBDMSO
OMe
HO
OMe
OTBDMS
Bn2HN Cl OH
asymmetric synthesis of a-hydroxyacid 4, avoiding the
need for enzymatic resolution at the final stage,7 or sepa-
ration of diastereoisomers after amide bond formation
with the serinolipid 5.6 We are currently investigating fur-
ther reactions of the chiral glycolate equivalents 2 and 3
and their application to natural product synthesis.
21
20
d
TBSDMO
TBDMSO
O
e
TBDMSO
OH
TBDMSO
H
Bn2N
OTBDMS
Bn2N
23
OTBDMS
Acknowledgment
22
We would like to thank Dr J. E. Davies for the X-ray crystallogra-
phic work. We would like to acknowledge Pfizer Global Research
and Development and the Novartis Research fellowship (to SVL),
the EPSRC (TDS), and the foreign office of France Lavoisier
Fellowship (SL) for their financial support.
f
TBDMSO
TBDMSO
7
7
g
TBDMSO
TBDMSO
NH2 OTBDMS
Bn2N
OTBDMS
References
7
26
(1) New address: D. J. Dixon, Department of Chemistry,
University of Manchester, Oxford Road, Manchester, M13
9PL, UK.
I
NH2
O
N
h
i
(2) New address: T. D. Sheppard, Chemistry Department,
University College London, Chirstopher Ingold
Laboratories, 20 Gordon Street, London, WC1H 0AJ, UK.
(3) Asai, N.; Fusetani, N.; Matsunaga, S.; Sasaki, J. Tetrahedron
2000, 56, 9895.
(4) Loukaci, A.; Bultel-Ponce, V.; Longeon, A.; Guyot, M. J.
Nat. Prod. 2000, 63, 799.
(5) Morita, M.; Motoki, K.; Akimoto, K.; Natori, T.; Sakai, T.;
Sawa, E.; Yamaji, K.; Koezuka, Y.; Kobayashi, E.;
Fukushima, H. J. Med. Chem. 1995, 38, 2176.
(6) Asai, N.; Fusetani, N.; Matsunaga, S. J. Nat. Prod. 2001, 64,
1210.
H
I
H
6
6
6
24
25
9
Scheme 5 a) LHMDS (1.05 equiv), THF, –78 °C, then 8 (1.1
equiv), 90%; b) AcCl, MeOH; c) TBSOTf (5.2 equiv), 2,6-lutidine (7
equiv), CH2Cl2, 0 °C to r.t., 93% over two steps; d) DIBALH (2.9
equiv), CH2Cl2, 0 °C, 62%; e) DMSO (3.8 equiv), (COCl)2 (1.9
equiv), Et3N (16.8 equiv), –78 °C to r.t., 87%; f) 9 (3 equiv), CrCl2
(12 equiv), DMF, THF, 33%; g) H2, Pd/C, EtOH, 82%; h) hydrazine
hydrate (7.9 equiv), 100 °C, 100%; i) Et3N, I2, CCl4, 13%.
(7) Masuda, Y.; Yoshida, M.; Mori, K. Biosci., Biotechnol.,
Biochem. 2002, 66, 1531.
We were able to improve on the literature procedure for
coupling amine 7 with the hydroxyacid fragment 4, by us-
ing unprotected a-hydroxyacid 4 directly in the amide
bond formation, rather than the corresponding tert-bu-
tyldimethylsilyl ether7 or acetate.6 Amide coupling with
N-ethyl-N¢-(3-dimethylaminopropyl)carbodiimide (EDC)
and 1-hydroxybenzotriazole (HOBt) proceeded in a grati-
fying 97% yield, and deprotection of the remaining silyl
ethers gave the sex pheromone 1 in 78% yield
(Scheme 6). The spectral data for 1 were in agreement
with literature values.6,7,18
(8) (a) Diez, E.; Dixon, D. J.; Ley, S. V. Angew. Chem. Int. Ed.
2001, 40, 2906. (b) Dixon, D. J.; Ley, S. V.; Polara, A.;
Sheppard, T. Org. Lett. 2001, 3, 3749. (c) Dixon, D. J.; Ley,
S. V.; Rodriguez, F. Angew. Chem. Int. Ed. 2001, 40, 4763.
(d) Dixon, D. J.; Ley, S. V.; Rodriguez, F. Org. Lett. 2001,
3, 3753. (e) Dixon, D. J.; Guarna, A.; Ley, S. V.; Polara, A.;
Rodriguez, F. Synthesis 2002, 1973. (f) Ley, S. V.; Diez, E.;
Dixon, D. J.; Guy, R. T.; Michel, P.; Nattrass, G. L.;
Sheppard, T. D. Org. Biomol. Chem. 2004, 2, 3608.
(g) Ley, S. V.; Dixon, D. J.; Guy, R. T.; Palomero, M. A.;
Polara, A.; Rodriguez, F.; Sheppard, T. D. Org. Biomol.
Chem. 2004, 2, 3618. (h) Michel, P.; Ley, S. V. Synthesis
2003, 1598.
In conclusion, we have demonstrated that diastereoselec-
tive alkylation and aldol reactions of lactones 2 and 3 are
readily scaleable and practically applicable to asymmetric
synthesis. The synthesis of pheromone 1 represents the
most efficient synthesis to date (9 steps in longest linear
sequence from lactone 3, 9% overall yield), and the first
(9) (a) Diez, E.; Dixon, D. J.; Ley, S. V.; Polara, A.; Rodriguez,
F. Synlett 2003, 1186. (b) Diez, E.; Dixon, D. J.; Ley, S. V.;
Polara, A.; Rodriguez, F. Helv. Chim. Acta 2003, 86, 3717.
(10) Laib, T.; Chastanet, J.; Zhu, J. P. J. Org. Chem. 1998, 63,
1709.
(11) Sandri, J.; Viala, J. Synth. Commun. 1992, 22, 2945.
Synlett 2005, No. 3, 481–484 © Thieme Stuttgart · New York

484
D. J. Dixon et al.
LETTER
(12) Soai, K.; Ookawa, A. J. Org. Chem. 1986, 51, 4000.
(13) The X-ray structural data for this compound have been
deposited with the Cambridge Crystallographic Data Centre,
reference number CCDC 257805. These data can be
obtained free of charge on the web at www.ccdc.cam.ac.uk/
data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12, Union Road,
Cambridge CB2 1EZ, UK; fax: +44(1223)336033.
(14) Koskinen, A. M. P.; Chen, J. S. Tetrahedron Lett. 1991, 32,
6977.
(18) Physical data for pheromone 1, (2S,2¢R,3S,4S)-2-(2¢-
hydroxy-21¢-methyl-docosanoylamino)-1,3,4-
pentadecanetriol: [a]D22 +12.5 (c 0.16, CHCl3–MeOH, 1:1)
{lit. [a]D28 +14 (c 0.70, CHCl3–MeOH, 1:1), see ref.7]. IR
(film): nmax = 3377 (O-H, N-H), 2917 (C-H), 2850 (C-H),
2480, 1639 (C=O), 1620, 1471, 1049 (C-O) cm–1. 1H NMR
(600 MHz, CDCl3–CD3OD, 1:1): d = 0.82 [6 H, d, J = 6.6
Hz, CH(CH3)2], 0.85 (3 H, t, J = 6.6 Hz, CH3), 1.06–1.60 [55
H, br m, 27 × CH2, CH(CH3)2], 1.60–1.69 (1 H, m, H-5),
1.70–1.80 (1 H, m, H-5), 3.50–3.53 (2 H, m, H-4, H-2¢), 3.71
(1 H, dd, J = 11.3, 4.8 Hz, H-1), 3.76 (1 H, dd, J = 11.3, 4.8
Hz, H-1), 4.00 (1 H, dd, J = 7.7, 3.7 Hz, H-3), 4.06–4.10 (1
H, m, H-2). 13C NMR (125 MHz, CDCl3–CD3OD, 1:1): d =
13.10, 21.70, 22.00, 24.50, 25.20, 26.80, 27.30, 28.70,
28.88, 28.93, 29.01, 29.03, 29.10, 29.12, 29.30, 31.30,
31.90, 33.90, 38.40, 50.90, 60.30, 71.20, 71.60, 74.60,
175.17. Found (ES): [M – Na]+ 650.5729, C38H77NO5
requires 650.5694.
(15) Masaki, Y.; Yoshizawa, K.; Itoh, A. Tetrahedron Lett. 1996,
37, 9321.
(16) Okazoe, T.; Takai, K.; Utimoto, K. J. Am. Chem. Soc. 1987,
109, 951.
(17) Pross, A.; Sternhell, S. Aust. J. Chem. 1970, 23, 989.
Synlett 2005, No. 3, 481–484 © Thieme Stuttgart · New York