Enantioselective synthesis of D-ribo-(2S,3S,4R)-C18-phytosphingosine using double stereodifferentiation

Article abstract of DOI:10.1016/S0040-4039(02)02732-6

An enantioselective synthesis of D-ribo-C₁₈-phytosphingosine as its tetraacetate derivative 10, starting from D-mannitol and employing the Sharpless asymmetric dihydroxylation reaction on allylic alcohol 6 as the key step, is described.

Full text of DOI:10.1016/S0040-4039(02)02732-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1035–1037
Enantioselective synthesis of
D
-ribo-(2S,3S,4R)-C18-phytosphingosine using double
stereodifferentiation
S. Vasudeva Naidu and Pradeep Kumar*
Division of Organic Chemistry: Technology, National Chemical Laboratory, Pune-411008, India
Received 18 September 2002; revised 22 November 2002; accepted 29 November 2002
Abstract—An enantioselective synthesis of D-ribo-C18-phytosphingosine as its tetraacetate derivative 10, starting from D-mannitol
and employing the Sharpless asymmetric dihydroxylation reaction on allylic alcohol 6 as the key step, is described. © 2003
Elsevier Science Ltd. All rights reserved.
Sphingolipids constitute a class of widely ranging natu-
ral products. Sphingosine, phytosphingosine and their
biosynthetic precursor, sphinganine are long chain
amino alcohols, generally possessing 18 or 20 carbon
atoms. They are building blocks of sphingolipids such
as sphingomyelins, glycosphingolipids, and phosphos-
phingolipids, which are important membrane con-
stituents playing vital roles in cell regulation and signal
transduction.1
of D-ribo-(2S,3S,4R)-C18-phytosphingosine from D-
mannitol and employing the Sharpless asymmetric
dihydroxylation as a key step.
The allylic alcohol 6 is envisaged as a chiral building
block from which
D-ribo-C18-phytosphingosine and
related analogs can be synthesized. The synthesis of
intermediate 6 starts from -mannitol, an inexpensive
D
Phytosphingosine was first detected in fungi,2a plants,2b
yeasts2c and other microorganisms.2d Thus, it was
found that its occurrence is not limited to plant tissues,
but it is also present in mammalian tissues, for example
in kidney,3a–c liver,3d uterus,3e intestine,3f skin,3g,h and
blood plasma.3i
D
-ribo-C18-Phytosphingosine ((2S,3S,4R)-2-aminoocta-
decane-1,3,4-triol) and its related C20-homologue are
widely distributed as amides of a-hydroxy long chain
acids in plant sphingolipids.4 Various methods for the
synthesis of phytosphingosines either racemic5 or enan-
tiomerically enriched6 have been described in literature.
As part of our research program aimed at developing
enantioselective synthesis of naturally occurring
lactones7a,b and amino alcohols,7c–f the Sharpless asym-
metric dihydroxylation was envisaged as a powerful
tool to the chiral dihydroxy compounds offering con-
siderable opportunity for synthetic manipulations.
Herein we report a new and enantioselective synthesis
Scheme 1. Reagents and conditions: (a) n-C13H27MgBr, CuI,
THF, 45°C to rt, 1 h, (86%); (b) BnBr, NaH, n-Bu4NI, THF,
rt, 4 h, (95%); (c) p-TSA, MeOH, rt, 32 h, (90%); (d) (i)
NaIO4 adsorbed on silica gel, DCM, rt, 30 min (95%); (ii)
(EtO)2P(O)CH2COOEt, LiBr, Et3N, THF, rt, overnight,
(89%); (e) DIBAL-H, Et2O, 0°C to rt, 2 h, (91%).
* Corresponding author. Tel.: +91-20-5893300, ext. 2050; fax:+91-20-
5893614; e-mail: tripathi@dalton.ncl.res.in
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02732-6

1036
S. V. Naidu, P. Kumar / Tetrahedron Letters 44 (2003) 1035–1037
the dihydroxylation of allylic alcohols and their
derivatives using OsO4 (stoichiometric or catalytic)
and NMO is reported to give only poor to moderate
diastereoselectivity.13
and readily available chiral pool material as illustrated
in Scheme 1.
D-Mannitol was first converted to
diepoxide 1 following a literature procedure.8
The opening of diepoxide 1 was carried out regioselec-
tively using tridecylmagnesium bromide in the pres-
ence of CuI to afford compound 2 in 86% yield
having [h]2D0=+21.33 (c 0.36, CHCl3). Protection of
hydroxyl groups in 2 with benzyl bromide furnished
the corresponding 2,5-O-dibenzylated product 3 in
essentially quantitative yield which on subsequent
deprotection of the isopropylidene group with a cata-
lytic amount of p-TSA afforded compound 4 in excel-
In order to achieve the synthesis of target compound
10 from 7a (Scheme 3), we required the transforma-
tion of the hydroxy group into azide at the C-2 posi-
tion. To this end protection of 7a as p-methoxybenzyl-
idene derivative was effected using 4-methoxybenzalde-
hyde dimethyl acetal in the presence of a catalytic
amount of p-TSA to afford a mixture of 1,3- and
1,2-benzylidene compounds in
a 19:1 ratio. The
lent yield. Oxidative cleavage of
4 with NaIO4
desired major 1,3-benzylidene compound 8 was sepa-
rated by silica gel column chromatography. Com-
pound 8 was then converted into an O-mesylated
derivative, which on treatment with sodium azide in
DMF furnished the azide 9 with the desired stereo-
chemistry at C-2. Deprotection of benzyl, cleavage of
1,3-benzylidene protecting group and reduction of
azide to amine were carried in one-pot reaction by
hydrogenation in the presence of 10% Pd/C in ethanol
followed by acetylation to furnish 10 in 69% yield.
The physical and spectroscopic data of 10 were in full
agreement with the literature data.6c
adsorbed on silica gel gave the corresponding aldehyde
in 95% yield, which was subsequently treated with
(EtO)2P(O)CH2COOEt under the Wittig–Horner reac-
tion conditions to afford the a,b-unsaturated ester 5 in
89% yield.
In the asymmetric dihydroxylation of an olefin, the
stereoselective outcome of the reaction is affected by
the presence of pre-existing chiral information in the
substrate. With a view to exploiting the concept of
double diastereoselection, the olefinic ester 5 was sub-
jected to the Sharpless asymmetric dihydroxylation
(SAD) conditions.9 However, the reaction proceeded
much more slowly with a poor diastereoselectivity
probably as a consequence of the electron-withdrawing
properties of the ester group.9 While the Sharpless
asymmetric dihydroxylation on the allylic alcohols
with different long alkyl chains has been exploited to a
large extent,10 the reaction on allylic alcohols having
chiral centers remains still unexplored. The enantio-
selectivity of an asymmetric dihydroxylation reaction
can be modulated by the size of the allylic substituent
and the configuration at the allylic position. Therefore,
in order to explore the possibility of achieving a good
diastereoselectivity, it was thought worthwhile to con-
vert ester 5 into the allylic alcohol 6 for a SAD reac-
tion. Thus, DIBAL-H reduction of ester 5 furnished
the corresponding allyl alcohol 611 in 91% yield which
was then subjected to the Sharpless asymmetric dihy-
droxylation reaction (Scheme 2). The results of double
diastereoselection are given in Table 1.
Scheme 2. Reagents and conditions: (a) ligand, OsO4,
MeSO2NH2, K3Fe(CN)6, K2CO3, t-BuOH:H2O (1:1), 24 h,
0°C ( 89–92%).
Table 1. Double stereodifferentiation in AD reaction of 6
Ligand
7a
7b
Yield (%)
(DHQD)2PHAL
(DHQ)2PHAL
9
1
1
2
92
89
The dihydroxylation of allyl alcohol 6 under the Sharp-
less asymmetric dihydroxylation conditions using
(DHQD)2PHAL ligand afforded the diastereomeric
mixture 7a:7b in a 9:1 ratio.12 The stereochemical purity
of 7a was easily enriched to 90% by recrystalli-
zation twice from acetone. The compound 7a was fully
characterized by analytical and spectroscopic data.11
The use of (DHQ)2 PHAL ligand under similar condi-
tions gave a diastereomeric ratio 7a:7b (1:2). The high
diastereoselectivity obtained in the former case could be
regarded as a matched case where the chirality informa-
tion of the reagent and the substrate probably act
synergistically while the poor degree of diastereo-
selection observed in the latter case may be because of
opposite influences of the chiral reagent and substrate
(mismatched case). It should be mentioned here that
Scheme 3. Reagents and conditions: (a) p-MeO-PhCH(OMe)2,
p-TSA, CH2Cl2, rt, overnight (70%); (b) (i) MeSO2Cl, Et3N,
DMAP (Cat), CH2Cl2, rt, overnight; (ii) NaN3, DMF, 80°C,
24 h (81%); (c) (i) Pd/C, H2, EtOH; (ii) Ac2O, Py, DMAP,
CH2Cl2, rt, 24 h (69%).

S. V. Naidu, P. Kumar / Tetrahedron Letters 44 (2003) 1035–1037
1037
Conclusion
hedron 1996, 52, 2187–2192; (c) Mulzer, J.; Brand, C.
Tetrahedron 1986, 42, 5961–5968; (d) He, L.; Byun, H.;
Bittmann, R. J. Org. Chem. 2000, 65, 7618–7626; (e)
Nakamura, T.; Shiozaki, M. Tetrahedron 2001, 57, 9087–
9092; (f) Yoda, H.; Oguchi, T.; Takabe, K. Tetrahedron:
Asymmetry 1996, 7, 2113–2116; (g) Matsumoto, K.;
Ebata, T.; Matsushita, H. Carbohydr. Res. 1995, 279,
93–106; (h) Murakami, T.; Minamikawa, H.; Hato, K.;
Nakahara, Y.; Ogawa, T. Tetrahedron Lett. 1994, 35,
745–748; (i) Nakamura, T.; Shiozaki, M. Tetrahedron
Lett. 1999, 40, 9063–9064; (j) Schmidt, R. R.; Maier, T.
Carbohydr. Res. 1988, 174, 169–179; (k) Martin, C.;
Prunck, W.; Bortolussi, M.; Bloch, R. Tetrahedron:
Asymmetry 2000, 11, 1585–1592.
In summary, a highly enantioselective synthesis of
D-
ribo-C18-phytosphingosine has been achieved from a
readily available carbohydrate precursor by using the
Sharpless asymmetric dihydroxylation procedure. The
concept of double diastereoselection was employed for
the first time on a chiral allylic alcohol in SAD reac-
tion. The merits of this synthesis are high diastereo-
selectivity and high yielding reaction steps. The other
isomer L-lyxo-C18-phytosphingosine can be synthesized
from S-diepoxide using the chiral ligand (DHQ)2PHAL
in the dihydroxylation step and following the reaction
sequence shown above.
7. (a) Pais, G. C. G.; Fernandes, R. A.; Kumar, P. Tetra-
hedron 1999, 55, 13445–13450; (b) Fernandes, R. A.;
Kumar, P. Tetrahedron: Asymmetry 1999, 10, 4349–4356;
(c) Fernandes, R. A.; Kumar, P. Tetrahedron: Asymmetry
1999, 10, 4797–4802; (d) Fernandes, R. A.; Kumar, P.
Eur. J. Org. Chem. 2000, 3447–3449; (e) Pandey, R. K.;
Fernandes, R. A.; Kumar, P. Tetrahedron Lett. 2002, 43,
4425–4426; (f) Fernandes, R. A.; Kumar, P. Tetrahedron
Lett. 2000, 41, 10309–10312.
Acknowledgements
S.V.N. thanks CSIR New Delhi for financial assistance.
We are grateful to Dr. M. K. Gurjar for his support
and encouragement. This is NCL Communication No.
6634.
8. Le Merrer, Y.; Dureault, A.; Greck, C.; Micas-Languin,
D.; Gravier, C.; Depezay, J. C. Heterocycles 1987, 25,
541–548.
9. (a) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B.
Chem. Rev. 1994, 94, 2483–2547; (b) Becker, H.; Sharp-
less, K. B. Angew. Chem., Int. Ed. Engl. 1996, 35, 448–
451.
References
1. For an excellent overview of sphingolipid functions, see:
Merril, A. H., Jr.; Sweeley, C. C. In Biochemistry of
Lipids, Lipoproteins and Membranes; Vance, D. E.;
Vance. J. E., Eds.; Elsevier Science: Amsterdam, 1996;
Chapter 12, pp. 309–339.
10. Zhang, Z.-B.; Wing, Z.-M.; Wang, Y. X.; Liu, H.-Q.; Lei,
G.-X.; Shi, M. J. Chem. Soc., Perkin Trans. 1 2000,
53–57.
2. (a) Oda, T. J. Pharm. Soc. Jpn. 1952, 72, 142–144; (b)
Carter, H. E.; Clemer, W. D.; Lands, W. M.; Muller, K.
L.; Tomizawa, H. H. J. Biol. Chem. 1954, 206, 613–623;
(c) Thorpe, S. R.; Sweeley, C. Biochemistry 1967, 6,
887–897; (d) Dearborn, D. G.; Smith, S.; Korn, E. D. J.
Biol. Chem. 1976, 251, 2976–2982.
11. Spectroscopic data for selected compounds: Compound
6: Colorless oil, [h]2D0=+24.89 (c 0.54, CHCl3); IR (neat):
wmax 3372, 2924, 2853, 2450, 1464 cm−1 1H NMR (200
;
3. (a) Karlsson, K. A. Acta Chem. Scand. 1964, 18, 2395–
2396; (b) Karlsson, K. A. Acta Chem. Scand. 1964, 18,
2397–2398; (c) Karlsson, K. A.; Samuelsson, B. E.; Steen,
G. O. Acta Chem. Scand. 1968, 22, 1361–1364; (d) Baren-
holz, Y.; Gatt, S. Biochem. Biophys. Res. Commun. 1967,
27, 319–324; (e) Takamatsu, K.; Mikami, M.; Kikuchi,
K.; Nozawa, S.; Iwamori, M. Biochim. Biophys. Acta
1992, 1165, 177–182; (f) Okabe, K.; Keenan, R. W.;
Schmidt, G. Biochem. Biophys. Res. Commun. 1968, 31,
137–143; (g) Wertz, P. W.; Miethke, M. C.; Long, S. A.;
Stauss, J. S.; Downing, D. T. J. Invest. Dermatol. 1985,
84, 410–412; (h) Schmidt, R. R. In Liposome Dermatics;
Braun-Falco, O.; Corting, H. C.; Msibach, H. I., Eds.;
Springer: Berlin, 1992; pp. 44–56; (i) Vance, D. E.;
Sweeley, C. C. J. Lipid Res. 1967, 8, 621–630.
4. (a) Carter, H. E.; Galanos, D. S.; Fujino, Y. Can. J.
Biochem. Physiol. 1956, 34, 320–333; (b) Grob, C. A. Rec.
Chem. Progr. 1957, 18, 55–68; (c) Stoffel, W. Chem. Phys.
Lipids 1973, 11, 318–324; (d) Rodd’s Chemistry of Carbon
Compounds, 2nd ed.; Elsevier: Amsterdam, 1983; Vol. 1E,
p. 394.
5. Sisido, K.; Hirowatari, N.; Tamura, H.; Kobata, H.;
Takagisi, M.; Isida, T. J. Org. Chem. 1970, 35, 350–353.
6. (a) Kobayashi, S.; Hayashi, T.; Kawasuji, T. Tetrahedron
Lett. 1994, 35, 9573–9576; (b) Lin, G.; Shi, Z. Tetra-
MHz, CDCl3): l 0.90 (t, J=6 Hz, 3H), 1.26 (m, 26H),
3.80 (dd, J=6, 8 Hz, 1H), 4.19 (d, J=6 Hz, 2H), 4.35 (d,
J=10 Hz, 1H), 4.57 (d, J=10 Hz, 1H), 5.54 (m, 1H),
5.81 (dt, J=6, 16 Hz, 1H), 7.34 (m, 5H); 13C NMR (50
MHz, CDCl3): l 14.07, 22.64, 25.36, 29.36, 29.66, 30.83,
31.90, 31.97, 35.65, 58.92, 63.00, 70.24, 127.36, 127.69,
128.31, 131.66, 132.61, 133.46, 138.12; MS (EI) m/z (%)
374 (M+). Anal. calcd for C25H42O2 (374.60): C, 80.15; H,
11.30. Found: C, 79.95; H, 11.31.
Compound 7a: white solid (mp 77°C); [h]2D0=−7.60
(c 0.86, CHCl3); IR: wmax 3422, 2926, 1458, 1370, 1215,
–
765, 668 cm−1
;
1H NMR (200 MHz, CDCl3): l 0.89 (t,
J=6 Hz, 3H), 1.27–1.30 (m, 26H), 3.32 (br s, 3H), 3.65
(s, 2H), 3.75–3.80 (m, 2H), 3.95–4.10 (m, 1H), 4.62 (s,
2H), 7.34 (m, 5H); 13C NMR (50 MHz CDCl3): l 14.03,
22.60, 25.17, 29.29, 29.62, 29.80, 30.57, 31.86, 65.02,
70.31, 72.77, 72.96, 76.38, 77.00, 77.66, 81.34, 127.80,
127.91 128.42, 138.13; MS (EI) m/z (%) 408 (M+), 393
(M+−15). Anal. calcd for C25H44O4 (408.6): C, 73.48; H,
10.85. Found C, 73.21; H, 10.52.
1
12. The diastereomeric ratio was determined by H and 13C
NMR spectral data.
13. Cha, J. K.; Christ, W. J.; Kishi, Y. Tetrahedron Lett.
1983, 24, 3943–3946.