Asymmetric synthesis of d - Ribo -phytosphingosine from 1-tetradecyne and (4-Methoxyphenoxy)acetaldehyde

Article abstract of DOI:10.1021/jo100707d

(Figure presented) An asymmetric synthesis of d-ribo-phytosphingosine (1) was achieved by utilizing the ProPhenol (12)-catalyzed alkynylation of unsaturated aldehyde 8 to afford allylic propargylic alcohol (S)-6 followed by asymmetric epoxidation and opening of propargylic epoxy alcohol anti-5 with NaN₃/NH₄Cl. Deprotection and reduction of the resulting acyclic azide 3 then gave 1. Alkyne-azide 3 was subjected to an intramolecular click reaction, generating a bicyclic triazole, which was found to have unexpected vicinal coupling constants. Application of the advanced Mosher method verified the configurations of the three contiguous stereogenic centers of 1. An alkynyl azide analogue of 1, which may be useful as a glycosyl acceptor in the synthesis of α-galactosylceramide derivatives, was also readily prepared by this route.

Full text of DOI:10.1021/jo100707d

pubs.acs.org/joc
Asymmetric Synthesis of D-ribo-Phytosphingosine from 1-Tetradecyne
and (4-Methoxyphenoxy)acetaldehyde
Zheng Liu, Hoe-Sup Byun, and Robert Bittman*
Department of Chemistry and Biochemistry, Queens College of The City University of New York, Flushing,
New York 11367-1597
robert.bittman@qc.cuny.edu
Received April 14, 2010
An asymmetric synthesis of D-ribo-phytosphingosine (1) was achieved by utilizing the ProPhenol
(12)-catalyzed alkynylation of unsaturated aldehyde 8 to afford allylic propargylic alcohol (S)-6
followed by asymmetric epoxidation and opening of propargylic epoxy alcohol anti-5 with NaN3/
NH4Cl. Deprotection and reduction of the resulting acyclic azide 3 then gave 1. Alkyne-azide 3 was
subjected to an intramolecular click reaction, generating a bicyclic triazole, which was found to have
unexpected vicinal coupling constants. Application of the advanced Mosher method verified the
configurations of the three contiguous stereogenic centers of 1. An alkynyl azide analogue of 1, which
may be useful as a glycosyl acceptor in the synthesis of R-galactosylceramide derivatives, was also
readily prepared by this route.
Introduction
PHS 1-phosphate,3b,4 inositol phosphorylceramide,5 and
KRN7000 (2, the R-anomer of galactosylceramide, an im-
munostimulant of invariant natural killer T (iNKT) cells)
(Chart 1).6
Because of the biological importance of PHS, there has
been considerable interest in the synthesis of 1 and its stereo-
isomers.7 The construction of the three contiguous stereo-
genic centers poses an interesting and demanding challenge.
Historically, natural chiral pools have played a significant
role in their syntheses, but asymmetric reactions have
emerged as a more favorable strategy for reasons of chirality
economy and efficiency. Recently, the catalytic asymmetric
alkynylation reaction developed by Trost et al. has been used
(2S,3S,4R)-D-ribo-Phytosphingosine (4-D-hydroxysphin-
ganine, PHS, 1) is distributed ubiquitously, including in
membranes of fungi, plants, bacteria, marine organisms,
and mammalian tissues.1 In addition to its structural role
in membranes, 1 regulates cellular growth2 and mediates
the heat stress response of yeast.3 Moreover, 1 serves as a
metabolic precursor of important lipid mediators such as
(1) (a) Karlsson, K.-A.; Martensson, E. Biochim. Biophys. Acta 1968, 152,
230–233. (b) Carter, H. E.; Hirschberg, C. B. Biochemistry 1968, 7, 2296–
2300.
(2) Veerman, E. C. I.; Valentijn-Benz, M.; van’t Hof, W.; Nazmi, K.; van
Marle, J.; Amerongen, A. V. N. Biol. Chem. 2010, 391, 65–71.
(3) (a) Jenkins, G. M.; Richards, A.; Wahl, T.; Mao, C.; Obeid, L.;
Hannun, Y. J. Biol. Chem. 1997, 272, 32566–32572. (b) Skrzypek, M. S.;
Nagiec, M. M.; Lester, R. L.; Dickson, R. C. J. Bacteriol. 1999, 181, 1134–
1140. (c) Meier, K. D.; Deloche, O.; Kajiwara, K.; Funato, K.; Riezman, H.
Mol. Biol. Cell 2006, 17, 1164–1175.
(4) (a) Kim, M. K.; Park, K. S.; Lee, H.; Kim, Y. D.; Yun, T.; Bae, Y. S.
Exp. Mol. Med. 2007, 39, 185–194. (b) Cowart, L. A.; Shotwell, M.; Worley,
M. L.; Richards, A. J.; Montefusco, D. J.; Hannun, Y. A.; Lu, X. Mol. Syst. Biol.
2010, 6, article no. 349.
(5) (a) Smith, S. W.; Lester, R. C. J. Biol. Chem. 1974, 249, 3395–3405. (b)
Rhome, R.; Del Poeta, M. Annu. Rev. Microbiol. 2009, 63, 119–131. (c)
€
Bjorkbom, A.; Ohvo-Rekilae, H.; Kankaanpaeae, P.; Nyholm, T. K. M.;
Westerlund, B.; Slotte, J. P. Biochim. Biophys. Acta 2010, 1798, 453–460.
(6) For reviews of KRN7000 and its analogues, see: (a) Franck, R. W.;
Tsuji, M. Acc. Chem. Res. 2006, 39, 692–701. (b) Savage, P. B.; Teyton, L.;
Bendelac, A. Chem. Soc. Rev. 2006, 35, 771–779. (c) Zheng, L.; Ye, X.-S.
J. Chin. Pharm. Sci. 2008, 17, 263–271.
4356 J. Org. Chem. 2010, 75, 4356–4364
Published on Web 06/07/2010
DOI: 10.1021/jo100707d
r
2010 American Chemical Society

Liu et al.
JOCFeatured Article
CHART 1. Structures of D-ribo-Phytosphingosine and
R-Galactosylceramide (KRN7000)
SCHEME 1. Retrosynthetic Plan
as a key step in the syntheses of natural products.8 As an
extension of our previous studies on the synthesis of
phytosphingolipids7c,e,9 and of glycolipid-based iNKT cell
agonists,10 we describe here a stereocontrolled synthesis of 1
via a sequence of catalytic alkynylation and Sharpless asym-
metric epoxidation (SAE)11 reactions to generate the inter-
mediate chiral propargylic epoxy alcohol 5 which was
converted to 3 by ring-opening attack by azide ion. Although
the ring-opening was expected to result in an inversion of
configuration at C-2, the stereochemical outcome required
verification because of a previous report of unexpected reten-
tion of configuration during a Ti(O-i-Pr)2(N3)2-mediated
epoxide ring-opening reaction.12 To determine the relative
configurations in azido diol 3, we used an intramolecular
click reaction13 to form rigid bicyclic triazole 15, which,
however, proved to have unusual vicinal coupling constants.
The configurations of the three contiguous stereogenic cen-
ters of 1 were instead determined by application of the
advanced Mosher method.14
alcohols (ceramides).15,16 Azido alkynyl alcohol 4 was read-
ily obtained by the synthetic procedure described here. Thus,
the route to 1 described herein may be used to prepare a
glycosyl acceptor for the preparation of 2 and other galacto-
sylceramide derivatives.6,17
Results and Discussion
Outline of the Synthetic Plan. As illustrated in Scheme 1,
we envisaged 1 and 4 to be accessible from azido diol 3. The
2S,3S configuration in 3 can be generated by SAE followed
by opening of the resulting epoxy alcohol anti-5 with NaN3/
NH4Cl. SAE of (S)-6 under kinetic resolution conditions can
improve the diastereomeric excess of anti-5. Although the
enantiomeric excess of (S)-6 will not affect the subsequent
stereoselectivity, it plays a key role in maximizing the yield of
SAE. The enantiomeric excess of (S)-6 can be enhanced via
catalytic alkynylation of enal 8 with alkyne 7. Asymmetric
alkynylation of R,β-unsaturated aldehydes has been used in
the synthesis of many complicated molecules with high
efficiency18 but often requires stoichiometric or catalytic
titanium in addition to zinc. Trost and co-workers have
recently simplified and expanded the scope of this reaction.8a
We have employed the Trost protocol to make 6 starting with
enal 8, which was prepared from commercially available
aldehyde 9 using our recently developed two-step HWE/
AlH3 reduction protocol.19 This synthetic route may permit
access to other stereoisomers of 1 because the catalytic
alkynylation and SAE reactions can be used to generate
Interestingly, 2-azido alcohols have been found to be more
favorable glycosyl acceptors than the corresponding 2-amido
(7) (a) For a review of PHS syntheses prior to 2002, see: Howell, A. R.;
Ndakala, A. J. Curr. Org. Chem. 2002, 6, 365–391. For additional references
on subsequent syntheses of 1, see: (b) Lu, X.; Byun, H.-S.; Bittman, R. J.
Org. Chem. 2004, 69, 5433–5438. (c) Singh, O. V.; Kampf, D. J.; Han, H.
Tetrahedron Lett. 2004, 45, 7239–7242. (d) Lu, X.; Bittman, R. Tetrahedron
Lett. 2005, 46, 3165–3168. (e) Lombardo, M.; Capdevila, M. G.; Pasi, F.;
Trombini, C. Org. Lett. 2006, 8, 3303–3305. (f) Jeon, J.; Shin, M.; Yoo, J. W.;
Oh, J. S.; Bae, J. G.; Jung, S. H.; Kim, Y. G. Tetrahedron Lett. 2007, 48, 1105–
1108. (g) Chang, C.-W.; Chen, Y.-N.; Adak, A. K.; Lin, K.-H.; Tzou,
D.-L. M.; Lin, C.-C. Tetrahedron 2007, 63, 4310–4318. (h) Abraham, E.;
Candela-Lena, J. I.; Davies, S. G.; Georgiou, M.; Nicholson, R. L.; Roberts,
ꢀ
ꢀ
P. M.; Russell, A. J.; Sanchez-Fernandez, E. M.; Smith, A. D.; Thomson,
J. E. Tetrahedron: Asymmetry 2007, 18, 2510–2513. (i) Kim, S.; Lee, N.; Lee,
S.; Lee, T.; Lee, Y. M. J. Org. Chem. 2008, 73, 1379–1385. (j) Abraham, E.;
Brock, E. A.; Candela-Lena, J. I.; Davies, S. G.; Georgiou, M.; Nicholson,
ꢀ
R. L.; Perkins, J. H.; Roberts, P. M.; Russell, A. J.; Sanchez-Fernandez,
E. M.; Scott, P. M.; Smith, A. D.; Thomson, J. E. Org. Biomol. Chem. 2008, 6,
ꢀ
ꢀ
´
1665–1673. (k) Llaveria, J.; Dıaz, Y.; Matheu, M. I.; Castillon, S. Org. Lett.
2009, 11, 205–208.
(8) For reactions with R,β-unsaturated aldehydes, see: (a) Trost, B. M.;
Weiss, A. H.; von Wangelin, A. J. J. Am. Chem. Soc. 2006, 128, 8–9. (b) Trost,
B. M.; Weiss, A. H. Org. Lett. 2006, 8, 4461–4464. (c) Trost, B. M.; Weiss,
A. H. Angew. Chem., Int. Ed. 2007, 46, 7664–7666. (d) Trost, B. M.; Sieber,
J. D.; Qian, W.; Dhawan, R.; Ball, Z. T. Angew. Chem., Int. Ed. 2009, 48,
5478–5481.
(9) He, L.; Byun, H.-S.; Bittman, R. J. Org. Chem. 2000, 65, 7618–7626.
(10) Lu, X.; Song, L.; Metelitsa, L. S.; Bittman, R. ChemBioChem 2006, 7,
1750–1756.
(15) (a) Schmidt, R. R.; Zimmermann, P. Angew. Chem., Int. Ed. Engl.
1986, 25, 725–726. (b) Polt, R.; Szabo, L.; Treiberg, J.; Li, Y.; Hruby, V. J.
J. Am. Chem. Soc. 1992, 114, 10249–10258.
(16) For the use of azido sphingosine analogues for the synthesis of
β-glycosylsphingolipids, see: (a) Bittman, R. Chem. Phys. Lipids 2004, 129,
111–131. (b) Lankalapalli, R. S.; Baksa, A.; Liliom, K.; Bittman, R.
ChemMedChem 2010, 5, 682–686.
(17) (a) Veerapen, N.; Brigl, M.; Garg, S.; Cerundolo, V.; Cox, L. R.;
Brenner, M. B.; Besra, G. S. Bioorg. Med. Chem. Lett. 2009, 19, 4288–4291.
(b) Leung, L.; Tomassi, C.; Van Beneden, K.; Decruy, T.; Trappeniers, M.;
Elewaut, D.; Gao, Y.; Elliott, T.; Al-Shamkhani, A.; Ottensmeier, C.;
Werner, J. M.; Williams, A.; Van Calenbergh, S.; Linclau, B. ChemMedChem
2009, 4, 329–334. (c) Chen, W.; Xia, C.; Wang, J.; Thapa, P.; Li, Y.; Nadas, J.;
Zhang, W.; Zhou, D.; Wang, P. G. J. Org. Chem. 2007, 72, 9914–9923. (d)
Ding, N.; Li, C.; Liu, Y.; Zhang, Z.; Li, Y. Carbohydr. Res. 2007, 342, 2003–
2013.
(11) For a review of SAE, see: Katsuki, T; Martin, V. S. Org. React. 1996,
48, 1–299.
(12) By comparing the synthetic final product with authentic samples of
other diastereoisomers, Pu and Franck found that retention took place
during the Ti(O-i-Pr)2(N3)2-mediated opening of a primary epoxy alcohol.
The reason for retention is unclear. (a) Pu, J.; Franck, R. W. Tetrahedron
2008, 64, 8618–8629. (b) Pu, J. Ph.D. Dissertation, The City University of New
York, 2006.
(13) Pearson, W. H.; Bergmeier, S. C.; Chytra, J. A. Synthesis 1990, 156–
159.
(14) For the advanced Mosher method for determining the absolute
configuration by 1H-NMR spectroscopy, see: Ohtani, I.; Kusumi, T.;
Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991, 113, 4092–4096.
(18) For reviews of catalytic asymmetric organozinc additions to carbo-
nyl compounds, see: (a) Pu, L.; Yu, H. B. Chem. Rev. 2001, 101, 757–824. (b)
Pu, L. Tetrahedron 2003, 59, 9873–9886.
(19) Liu, Z.; Gong, Y.; Byun, H.-S.; Bittman, R. New J. Chem. 2010, 34,
470–475.
J. Org. Chem. Vol. 75, No. 13, 2010 4357

Liu et al.
SCHEME 3. Synthesis of Epoxy Alcohols anti-5 and syn-5
JOCFeatured Article
SCHEME 2. Synthesis of Allylic Propargylic Alcohols (S)-6
and (R)-6
not. However, to our surprise, ProPhenol zinc alkynylation
of 8 with 1-tetradecyne catalyzed by (S,S)-12 provided
(E)-enynol (R)-6 with a high enantioselectivity (85% ee).25
The enantiomeric excess and absolute configuration of
E-enynols (S)-6 and (R)-6 were determined by preparing
the (R)- and (S)-MTPA esters and analyzing their 1H NMR
spectra by the subtraction protocol of the advanced Mosher
method (see the Supporting Information).14
Epoxidation of (E)-Enynol 6. As shown in Schemes 1 and 3,
we intended to use SAE as one of the key steps to build the
other two stereogenic centers from enynols (S)- and (R)-6.
Under Sharpless kinetic resolution conditions, epoxida-
tion of (S)-6 (60% ee) with cumene hydroperoxide (CHP)
in the presence of substoichiometric amounts of catalysts
(D-(-)-DIPT, Ti(OPr-i)4) and 4 A molecular sieves gave
epoxy alcohol anti-5 in good yield (68%; the maximum yield
was 75%). After the unreacted substrate was removed by
chromatography, the desired propargylic epoxy alcohol anti-
5 was obtained in high diastereomeric excess (93% de). In
contrast, epoxidation of (R)-6 (85% ee) under the same
conditions led to syn-5 in a much lower diastereomeric excess
(35% de), indicating that the R configuration eroded the
diastereomeric induction dominated by the catalyst. This
mismatched effect brought about by the acetylenic moiety
may be more pronounced than that of the alkyl group.
Opening of Epoxy Alcohol anti-5. As shown in Scheme 4,
opening of epoxy alcohol anti-5 with NaN3/NH4Cl26 pro-
vided azido diol 3 in 61% yield, together with other uni-
dentified compounds. Although this reaction was not very
clean, alternative methods, such as Ti(O-i-Pr)2(N3)2,27 were
not attempted because in our prior experience this reagent
gave rise to many unidentified compounds,10,28 including the
Ti-catalyzed semipinacol rearrangement product of R-hy-
droxy epoxides.29
the opposite configurations by use of their counterpart
ligands.
Asymmetric Synthesis of Enynol (S)-6. On the basis of the
retrosynthetic analysis depicted in Scheme 1, the first target,
the conjugated (E)-enynol (S)-6, can be prepared by coupling
of enal 8 with 7 by catalytic alkynylation. As shown in
Scheme 2, allylic alcohol 11 was prepared from (4-methoxy-
phenoxy)acetaldehyde (9)20 via a two-step HWE/reduction
protocol.19 Previously, because it is difficult to achieve high
E-selectivity by the HWE olefination reaction, (E)-R,β-un-
saturated ester 10 was obtained by nucleophilic substitution
of alkyl 4-bromocrotonate with 4-methoxyphenol.21 HWE
reaction of aldehyde 9 with triethyl phosphonoacetate in the
presence of K2CO3 in H2O/2-PrOH (1:1) produced ester 10
in 86% yield and high E-selectivity (E/Z = 28:1). Reduction
of ester 10 by AlH3 (generated from LAH and n-BuBr in
THF and used in situ)19 provided allylic alcohol 11 in 88%
yield. Oxidation of allyl alcohol 11 with PCC gave (E)-R,β-
unsaturated enal 8 in 58% yield.
Chiral allylic propargylic alcohols have been accessible
with high ee either by reduction of the corresponding ketone
with a stoichiometric amount of pinanylborane22 or by
lipase-catalyzed resolution of racemic allylic propargylic
alcohols.23 In our hands, alkynylation of enal 8 with 1-tetra-
decyne catalyzed by ProPhenol ligand (R,R)-1224 reproduc-
ibly provided a high yield of the desired (E)-enynol (S)-6
(86%), but the enantiomeric excess was only moderate at
best (60% ee). Under degassed reaction conditions, the
chemical yield was improved, but the enantioselectivity was
Investigation of the Stereochemistry of the Azido Diols from
syn- and anti-5 on the Basis of J Values of Bicyclic Derivatives.
To verify the stereochemistry of azido diols 3, propargylic
epoxy alcohol anti-5 was converted to bicyclic dihydroxy
triazole 14 via a copper-free intramolecular click reaction in
refluxing toluene for 48 h (Scheme 4).13 Diol 14 was then
converted to diacetate 15. However, when syn-5 (which has a
low de) was subjected to the same reaction sequence, we
obtained a mixture of diastereoisomer 150 [4S,5S,6S] and ent-
(20) Piggott, M. J.; Wege, D. Aust. J. Chem. 2003, 56, 691–702.
(21) He, L.; Byun, H.-S.; Smit, J.; Wilschut, J.; Bittman, R. J. Am. Chem.
Soc. 1999, 121, 3897–3903.
(22) Midland, M. M.; McDowell, D. C.; Hatch, R. L.; Tramontano, A.
J. Am. Chem. Soc. 1980, 102, 867–869.
(23) Burgess, K.; Jennings, L. D. J. Am. Chem. Soc. 1991, 113, 6129–6139.
(24) Trost, B. M.; Ito, H. J. Am. Chem. Soc. 2000, 122, 12003–12004.
(25) The reason for the different ee values is not clear, but the ee of the
ligands appear to differ based on our determination of their specific rota-
tions: (S,S)-12: [R]25 þ38.7 (c 0.63, CHCl3), [lit.24 [R]25 þ49.8 (c 3.0,
-40.9 (c 0.65, CHCl3), [Sigma-Aldrich catalog: [R]25D -50 (c 1.0, CHCl3)].
4358 J. Org. Chem. Vol. 75, No. 13, 2010
(26) Behrens, C. H.; Ko, S. Y.; Sharpless, K. B.; Walker, F. J. J. Org.
Chem. 1985, 50, 5687–5696.
(27) Caron, M.; Carlier, P. R.; Sharpless, K. B. J. Org. Chem. 1988, 53,
5185–5187.
(28) Liu, Z.; Byun, H.-S.; Bittman, R. Unpublished results.
(29) Snape, T. J. Chem. Soc. Rev. 2007, 36, 1823–1842.
D
D
CHCl3), Sigma-Aldrich catalog: [R]25D þ50 (c 1.0, CHCl3)]; (R,R)-12: [R]25
D

Liu et al.
JOCFeatured Article
SCHEME 4. Opening of anti-5 and Unexpected Coupling
Constants in Bicyclic Triazole Derivative 15
1
FIGURE 1. Partial H NMR spectra of 15 (top) and ent-15/150
(ratio 1:1.2, bottom).
CHART 2. Reported Coupling Constants in Similar Bicyclic
Systems30a,b
15 (the enantiomer of 15, (4S,5R,6R)) in a ratio of 1.2:1
(Scheme 4). The partial 1H NMR spectra of triazoles 15 and
150 are reported in Figure 1.
In 2005, Kim et al. prepared bicyclic triazoles 16a-d from
chiral aziridines (Chart 2).30a The coupling constant in the cis
relationship is 5.5-5.6 Hz, whereas that in the trans is
3.3-4.3 Hz. Although H-4 in diol 14 gave a doublet signal
(J = 5.5 Hz) and H-6 gave a quartet signal (J = 3.4 Hz), H-5
demonstrated an abnormal doublet of doublets (4.3 and 5.0
Hz), probably because of intramolecular hydrogen bonding
between the hydroxy groups. The coupling constants of 14
match those of 16a-d; therefore, these data may tentatively
confirm the relative configuration of the three contiguous
stereogenic centers. However, it must be pointed out that the
intramolecular hydrogen bonds in 14 and 16a-d are differ-
ent, which leads to uncertainty regarding the values of the
coupling constants. Furthermore, because of the absence of
the substituent at the 6 position, 16a-d may adopt different
conformations compared with 14. In 1988, based on cou-
pling constants,30b Ferris and Devades reported a con-
formational analysis in the pyrroloimidazole ring system
17a and 17b, which is very similar to our bicyclic triazole
15 (Chart 2). Since 17a and 17b were fully protected, the
impact of intramolecular hydrogen bonds is avoided. Their
investigation revealed that the value of the trans coupling
constant (J5,6 in 17) was 1.8 Hz, whereas the cis coupling
constants (J6,7 in 16 and 17, J5,6 in 16) were 6.0 and 7.2 Hz,
respectively.30b
The coupling constants of cis J4,5 and trans J5,6 in bicyclic
triazole 15 (Figure 1) were expected to differ largely. How-
ever, to our surprise, we found that they have two very close
J values (J4,5 = 5.8 Hz and J5,6 = 4.4 Hz). Furthermore, the
two J5,6 values of the trans coupling constants in 15 and 150
are markedly different (4.4 Hz in 15, 1.5 Hz in 150). On the
basis of this analysis, it is possible that one of the known
stereocontrolled steps did not proceed in the normal way.
Therefore, we decided to verify the course of the construction
of the three contiguous stereogenic centers by examining (1)
the configuration at C-4 in anti-5 to check which enynol
reacted [(S)-6 or (R)-6] in SAE, (2) the configurations at the
C-2 and C-3 positions in anti-5, and (3) the opening of anti-5
to verify the configuration at C-2 of 3.
(30) (a) Kim, M. S.; Yoon, H. J.; Lee, B. K.; Kwon, J. H.; Lee, W. K.;
Kim, Y.; Ha, H.-J. Synlett 2005, 2187–2190. (b) Ferris, J. P.; Devadas, B. J.
Org. Chem. 1987, 52, 2355–2361.
J. Org. Chem. Vol. 75, No. 13, 2010 4359

Liu et al.
JOCFeatured Article
Verification of the Sharpless Kinetic Resolution. According
to the empirical rule established by Sharpless and co-
workers, the Sharpless kinetic resolution of secondary allylic
alcohols favors the reaction in which the R enantiomer of the
racemic mixture forms the epoxide, while the S enantiomer
is recovered in an optically enriched form when D-(-)-DIPT
is used.31 For allylic propargylic alcohols, the S enan-
tiomer should react faster with D-(-)-DIPT because the
acetylenic moiety has a higher priority than the olefinic
group. Allylic propargylic alcohols have already been proved
to uphold the empirical rule,32 although the acetylenic
moiety would cause less steric crowding in the transition
states for SAE in comparison with alkyl groups. However,
the verification is limited to hydrocarbon substrates.32 In
order to confirm the Sharpless kinetic resolution in our case,
the (R)- and (S)-MTPA esters of anti-5 were prepared to
verify the configuration at C-4 (see the Supporting Infor-
mation). Analysis of the Δδ values of the protons indicates
that the reacted allylic propargylic alcohol was in fact (S)-6,
confirming the prediction made by the empirical rule.
Epoxide Configuration in anti-5. The absolute stereo-
chemistry of an epoxide in chiral epoxy alcohols is generally
assigned by the advanced Mosher method using (R)- and
(S)-MTPA esters of the corresponding ring-opened diol or
by established empirical mnemonics developed for different
asymmetric epoxidation strategies.33 Parker and Katsoulis32c
determined the absolute configuration of the epoxide in pro-
pargylic epoxy alcohols by converting the epoxide to a 1,3-diol
and analyzing the corresponding acetonide by the commonly
used [13C]-acetonide method developed by Rychnovsky et al.34
This method needs a two-step derivatization. We selected the
Et2AlCl-catalyzed cyclization of epoxytrichloroacetimidates35
to transfer the chiral information of the epoxide to the newly
formed secondary hydroxy group in an oxazoline or dihydroox-
azine (Scheme 5). Generally, cyclization takes place preferen-
tially at the more polarized center of the epoxide with complete
inversion of stereochemistry.35 As a result, after this transforma-
tion, the newly formed hydroxy group and C-4 oxygen will retain
the same relative configuration as that in the epoxy alcohol
between the epoxide and C-4 hydroxy group. By determining the
configuration of the newly formed hydroxy group, we can
assign the stereochemistry of the epoxide in anti-5. Reaction of
anti-5 with trichloroacetonitrile in the presence of Et2AlCl and
DBU gave the corresponding 2,3-epoxy-1-trichloroacetimidate,
SCHEME 5. Conversion of anti-5 to (S)- and (R)-18
which delivered oxazoline 13 in a two-step yield of 76%
(Scheme 5). The regiochemistry of cyclization was judged
1
by analysis of the H-1H COSY spectrum of 13 (see the
Supporting Information). Analysis of the (R)- and (S)-
MTPA esters of 13 ((R)-and (S)-18) revealed the anti
relationship between the C-2 hydroxy group and C-4 oxy-
gen (Figure 2), indicating that SAE of enynol (S)-6 followed
the normal prediction made by Sharpless et al. for allylic
alcohols bearing saturated alkyl groups.
FIGURE 2. Absolute stereochemistry determination of 13 via the
advanced Mosher method. Δδ values for the MTPA derivatives (S)-
18 and (R)-18 (Δδ = δS - δR ppm, 500 MHz).
Configuration of C-2. At this stage, the stereochemistry
of the opening of epoxy alcohol anti-5 (the first step of
Scheme 4) requires verification. For this purpose, we
planned to determine the configuration at C-2 of 3 by pre-
paring the (R)- and (S)-MTPA amides (Scheme 6). Reaction
of diol 3 with BnBr and NaH in the presence of a catalytic
amount of TBAI provided azide 19 in 63% yield. Several
methods were explored for the reduction of azide 19. We
found that azide 19 was smoothly converted to the corre-
sponding amine 20 by using 1,3-dithiopropane as the redu-
cing agent.36 In situ reaction of 20 with (R)- and (S)-MTPA
chlorides gave the (S)- and (R)-MTPA amides ((S)- and (R)-
21), respectively. Analysis of the two MTPA amides demon-
strated the syn relationship between the azide at C-2 and
oxygen at C-4, indicating that opening of epoxy alcohol anti-
5 took place by a simple SN2 inversion (Figure 3). Therefore,
the evidence showed that the construction of the three
contiguous stereogenic centers was correct. The unexpected
(31) Martin, V. S.; Woodard, S. S.; Katsuki, T.; Yamada, Y.; Ikeda, M.;
Sharpless, K. B. J. Am. Chem. Soc. 1981, 103, 6237–6240.
(32) (a) Rao, A. V. R.; Khrimian, A. P.; Krishna, P. R.; Yadagiri, P.;
Yadav, J. S. Synth. Commun. 1988, 18, 2325–2330. (b) Yadav, J. S.;
Radhakrishna, P. Tetrahedron 1990, 46, 5825–5832. (c) Parker, K. A.;
Katsoulis, I. A. Org. Lett. 2004, 6, 1413–1416.
(33) (a) Gao, Y.; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.; Masamune,
H.; Sharpless, K. B. J. Am. Chem. Soc. 1987, 109, 5765–5780. (b) Jacobsen,
E. N. Asymmetric Catalytic Epoxidation of Unfunctionalized Olefins. In Cata-
lytic Asymmetric Synthesis; Ojima, I., Ed.; Wiley-VCH: New York, 1993;
pp 159-202. (c) Johnson, R. A.; Sharpless, K. B. Catalytic Asymmetric
Epoxidation of Allylic Alcohols. In Catalytic Asymmetric Synthesis; Ojima,
I., Ed.; Wiley-VCH: New York, 1993; pp 103-158. (d) Wang, Z. X.; Tu, Y.;
Frohn, M.; Zhang, J. R.; Shi, Y. J. Am. Chem. Soc. 1997, 119, 11224–11235. (e)
Zhang, W.; Basak, A.; Kosugi, Y.; Hoshino, Y.; Yamamoto, H. Angew. Chem.,
Int. Ed. 2005, 44, 4389–4391.
(34) (a) Rychnovsky, S. D.; Rogers, B. N.; Richardson, T. I. Acc. Chem.
Res. 1998, 31, 9–17. (b) Evans, D. A.; Rieger, D. L.; Gage, J. R. Tetrahedron
Lett. 1990, 31, 7099–7100. (c) Rychnovsky, S. D.; Skalitzky, D. J. Tetra-
hedron Lett. 1990, 31, 945–948.
(35) (a) Hatakeyama, S.; Matsumoto, H.; Fukuyama, H.; Mukugi, Y.;
Irie, H. J. Org. Chem. 1997, 62, 2275–2279. (b) Liu, Z.; Byun, H.-S.; Bittman, R.
Org. Lett. 2010, 12, in press.
(36) (a) Bayley, H.; Standring, D. N.; Knowles, J. R. Tetrahedron Lett.
1978, 39, 3633–3634. (b) Sun, C.; Bittman, R. J. Org. Chem. 2004, 69, 7694–
7699.
4360 J. Org. Chem. Vol. 75, No. 13, 2010

Liu et al.
JOCFeatured Article
SCHEME 6. Conversion of Azido Diol 3 to (S)- and (R)-MTPA
Amide 21
FIGURE 3. Absolute stereochemistry determination of 20 by the
advanced Mosher method. Δδ values for the MTPA amides (S)-21
and (R)-21 (Δδ = δS - δR ppm, 500 MHz).
SCHEME 7. Completion of the Syntheses of 1 and 4 from 3 and
19, Respectively
coupling constants in 15 and 150 may result from the two
different conformations they adopted. This result indicates
the need for caution when coupling constants are used to
judge the relative configuration in bicyclic triazole and
related similar systems.
Completion of the Preparation of 1 and 4. As shown in
Scheme 7, deprotection of the PMP group by CAN in 3
followed by hydrogenation of the resulting triol 22 using
Pearlman’s catalyst (Pd(OH)2/C) in MeOH afforded 1. Its
1H and 13C NMR spectra and specific rotation matched the
previously reported data. The structure of 1 was further
confirmed by conversion to its tetraacetyl derivative 23.
In the synthesis of glycosylceramides, the choice of the
glycosyl acceptor is a critical consideration (together, of
course, with the selection of the glycosyl donor). A free
amino group at the C-2 position of the sphingolipid is not
a viable choice in the acceptor; moreover, the amide func-
tionality of ceramide is not suitable because it deactivates the
primary hydroxy group of the acceptor through unfavorable
hydrogen bonding interactions.15 Since an azide is appar-
ently devoid of hydrogen-bonding interactions with the
adjacent hydroxy group and can be readily converted to an
amide in two steps after the glycosidation reaction,16 we
decided to prepare 4. The saturated analogue of azido PHS 4
has been prepared from 1 by a tedious protecting group
manipulation involving the conversion of an amino group to
an azide by a diazo-transfer reaction.37 In contrast, 2-azido
carbinol 4, which is accessible from anti-5 via deprotection of
19, is an attractive alternative glycosyl acceptor because the
azido group is introduced into the phytosphingosine back-
bone at an early stage of the synthesis. Since modification of
the lipid chain length in R-galactosylceramide analogues
influences an array of cytokines release from activated iNKT
cells and demonstrates a profound relationship between
structure and activity,38 the synthetic route described here
allows modification of the chain length with ease.
aldehyde 919 and AlH3 reduction provided allylic alcohol 11,
and PCC oxidation afforded R,β-unsaturated aldehyde 8.
Catalytic alkynylation of 8 with 7 and SAE of (S)-6 followed
by regioselective NaN3/NH4Cl opening of the resulting
propargylic epoxy alcohol anti-5 delivered (2S,3S,4R) azido
diol 3 with the desired three contiguous stereogenic centers in
good yield and high stereoselectivity. Deprotection of 3 with
CAN and catalytic hydrogenation gave 1. A key intermedi-
ate, alkynyl-azido 3, can be readily converted to glycosyl
acceptor 4 via O,O-dibenzylation and CAN deprotection.
An advanced Mosher method clarified the unexpected values
of the coupling constants in bicyclic triazoles 15 and 150
generated by a copper-free intramolecular click reaction of 3
and confirmed the configurations of the stereogenic centers.
Experimental Section
(E)-4-(40-Methoxyphenoxy)-2-butenal (8). To a cooled, ra-
pidly stirred suspension of PCC (15.0 g, 69.6 mmol) and Celite
(16 g) in 250 mL of CH2Cl2 was added alcohol 11 (8.44 g, 43.5
mmol) in one portion. After the mixture had stirred for 3 h at rt,
the resulting dark mixture was diluted with 150 mL of Et2O.
Filtration through a pad of Florisil left a dark solid residue that
was washed with Et2O. The combined filtrates were concen-
trated, and the residue was purified by flash chromatography (a
gradient of hexane/EtOAc 6:1 to 3:1) to afford 8 (5.5 g, 58%) as
1
a slightly yellow solid: H NMR (400 MHz, CDCl3) δ 3.77 (s,
Conclusion
3H), 4.76 (dd, J = 1.9, 4.0 Hz, 2H), 6.46 (ddt, J = 15.8, 7.8, 1.9
Hz, 1H), 6.82-6.87 (m, 4H), 6.94 (dt, J = 15.8, 4.0 Hz, 1H), 9.62
(d, J = 7.8 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 55.7, 67.1,
114.7, 115.6, 132.2, 151.3, 151.9, 154.3, 193.0.
A stereocontrolled synthetic route to 1 from aldehyde 9
and 1-tetradecyne (7) has been developed. HWE reaction of
(4S,2E)-1-(40-Methoxyphenoxy)-2-octadecen-5-yn-4-ol ((S)-6).
A flame-dried, round-bottom flask was charged with the com-
mercially available ProPhenol ligand (R,R)-12 (1.0 g, 1.57 mmol),
alkyne 7 (9.2 g, 47.1 mmol), and 300 mL of toluene.
A solution of Me2Zn (39.3 mL, 1.2 M in toluene, 47.1 mmol)
was added rapidly via syringe. The reaction mixture was stirred
€
(37) Xia, C.; Schumann, J.; Emmanuel, R.; Zhang, Y.; Chen, W.; Zhang,
W.; De Libero, G.; Wang, P. G. J. Med. Chem. 2007, 50, 3489–3496.
(38) (a) Miyamoto, K.; Miyake, S.; Yamamura, T. Nature 2001, 413, 531–
534. (b) Goff, R. D.; Gao, Y.; Mattner, J.; Zhou, D.; Yin, N.; Cantu, C.;
Teyton, L.; Bendelac, A.; Savage, P. B. J. Am. Chem. Soc. 2004, 126, 13602–
13603.
J. Org. Chem. Vol. 75, No. 13, 2010 4361

Liu et al.
JOCFeatured Article
for 90 min at rt, and gas slowly evolved. A solution of R,β-
unsaturated aldehyde 8 (3.0 g, 15.6 mmol) in a minimal amount
of toluene was added via syringe over ∼10 s. The reaction
mixture was sealed and cooled to 4 °C for 4 days without
stirring. The reaction mixture was then slowly quenched
with aqueous saturated NH4Cl solution, and the layers were
separated. The aqueous layer was extracted with EtOAc (3 ꢀ
200 mL), and the combined organic layers were dried (Na2SO4),
filtered, and concentrated. Purification of the residue by flash
chromatography (a gradient of hexane/EtOAc 6:1 to 7:2) pro-
vided (S)-6 (5.2 g, 86%, 60% ee): 1H NMR (500 MHz, CDCl3) δ
0.88 (t, J = 7.1 Hz, 3H), 1.19-1.40 (m, 18H), 1.46-1.54 (m,
2H), 2.21 (dt, J = 2.0, 7.2 Hz, 2H), 2.45 (d, J = 4.3 Hz, 1H), 3.74
(s, 3H), 4.46-4.49 (m, 2H), 4.88-4.92 (m, 1H), 5.95 (ddt, J =
5.3, 15.4, 1.4 Hz, 1H), 6.08 (ddt, J = 1.2, 15.4, 5.3 Hz, 1H),
6.78-6.85 (m, 4H); 13C NMR (100 MHz, CDCl3) δ 14.0, 18.6,
22.6, 28.5, 28.8, 29.0, 29.3, 29.4, 29.5, 29.6, 31.8, 55.5, 62.3, 68.0,
78.8, 87.2, 114.5, 115.5, 127.1, 132.4, 152.5,þ153.7; ESI-HRMS
[M þ Na]þ calcd for m/z C25H38NaO3 409.2713, found
409.2717.
3.77 (s, 3H), 3.99 (dd, J = 11.5, 5.3 Hz, 1H), 4.26 (dd, J = 11.5, 2.7
Hz, 1H), 4.65-4.68 (m, 1H), 6.80-6.88 (m, 4H); 13C NMR (125
MHz, CDCl3) δ 14.1, 18.7, 22.7, 28.4, 28.9, 29.1, 29.3, 29.5, 29.63,
29.65, 31.9, 53.8, 55.7, 57.4, 60.8, 68.0, 76.0, 88.1, 114.6, 115.8,
þ
152.5, 154.2; ESI-HRMS [M þ NH4þ] calcd for C25H42NO4
420.3108, found 420.3114.
(S)-1-[(20R,30R)-30-[(400-Methoxyphenoxy)methyl]oxiran-20-yl]-
pentadec-2-yn-1-ol (syn-5). Compound syn-5 was prepared in 63%
yield and 35% de from (R)-6 (85% ee) according to the procedure
used to prepare anti-5: 1H NMR (500 MHz, CDCl3) δ 0.88 (t, J =
7.1 Hz, 3H), 1.22-1.40 (m, 18H), 1.47-1.55 (m, 2H), 2.22 (dt, J =
7.2, 1.9 Hz, 2H), 3.27 (dd, J = 2.2, 4.1 Hz, 1H), 3.40-3.43 (m,
1H), 3.77 (s, 3H), 3.98 (dd, J = 11.5, 5.3 Hz, 1H), 4.23 (dd, J =
11.5, 2.9 Hz, 1H), 4.38-4.42 (m, 1H), 6.81-6.88 (m, 4H); 13C
NMR (125 MHz, CDCl3) δ 14.1, 18.7, 22.7, 28.4, 28.9, 29.1, 29.3,
29.5, 29.62, 29.64, 31.9, 54.2, 55.7, 58.2, 60.4, 61.8, 76.9, 87.6,
114.6, 115.7, 152.5, 154.2.
(1S)-2-(40-Methoxyphenoxy)-1-[(40R,50R)-20-(trichloromethyl)-
40,50-dihydro-50-(tetradec-100-ynyl)oxazol-40-yl]ethanol (13). To an
ice-cold solution of anti-5 (30 mg, 74.5 μmol) in 1 mL of CH2Cl2
were added DBU (1.1 μL, 7.5 μmol) and trichloroacetonitrile
(15 μL, 149 μmol). After being stirred at 0 °C until no starting
material was observed on TLC, the reaction mixture was diluted
with CH2Cl2 (10 mL), quenched with saturated NH4Cl solution
(5 mL), and extracted with CH2Cl2. The combined organic
extracts were washed with saturated aqueous NH4Cl solution,
dried (Na2SO4), and concentrated. The residue was dissolved in
Et2O and passed through a short column packed with anhydrous
Na2SO4 and silica gel. Evaporation of Et2O gave a residue that
was dried under high vacuum (0.2 Torr, overnight) and used
directly in the subsequent cyclization reaction without further
purification.
(4R,2E)-1-(40-Methoxyphenoxy)-2-octadecen-5-yn-4-ol ((R)-
6). (R)-6 was prepared in 82% yield and 85% ee according to
the procedure used to prepare (S)-6. The 1H and 13C NMR
spectra were identical to those of (S)-6: ESI-HRMS [M þ Na]þ
calcd for C25H38NaO3þ 409.2713, found 409.2715.
General Preparation and Analysis of MTPA Esters or Amide.
The reactions were generally run on a 0.02-mmol scale. A
mixture of pyridine (4.0 equiv) and substrate (1.0 equiv) in
CH2Cl2 (0.6 mL) was treated with neat (R)- or (S)-MTPA
chloride (3.0 equiv). The solution was stored in a desiccator
until no starting material was observed by TLC. It is important
to monitor the reaction by TLC to ensure complete reaction
because kinetic resolution of an incomplete reaction may sig-
nificantly alter the ee or de measurements. The reaction mixture
was passed through a short plug of silica gel to remove polar
impurities, and the plug was washed with EtOAc/hexane (the
ratio made the Rf = 0.5). After the filtrate was concentrated, the
residue was dried under high vacuum (0.2 Torr, 1 h) and
dissolved in CDCl3. The (S)- and (R)-MTPA esters and amides
were prepared by using (R)- and (S)-MTPA chlorides, respec-
tively.
To an ice-cold solution of epoxy trichloroacetimidates in
CH2Cl2 (1 mL) was added Et2AlCl (37.3 μL, 1.0 M solution in
hexane, 37.3 μmol). After being stirred at rt until no starting
material was observed on TLC, the reaction mixture was
quenched with saturated aqueous NaHCO3 solution. The reac-
tion mixture was diluted with Et2O, washed with water, dried,
and evaporated. The residue was purified by flash chromatog-
raphy (hexane/EtOAc 5:1) to afford oxazoline 13 (31 mg, 76%):
1H NMR (500 MHz, CDCl3) δ 0.88 (t, J = 7.1 Hz, 3H), 1.21-
1.36 (m, 18H), 1.43-1.50 (m, 2H), 2.21 (dt, J = 1.9, 7.1 Hz, 2H),
2.48 (d, J = 4.8 Hz, 1H), 3.77 (s, 3H), 4.05 (dd, J = 6.0, 9.5 Hz,
1H), 4.12 (dd, J = 4.2, 9.5 Hz, 1H), 4.15-4.20 (m, 1H), 4.46 (dd,
J = 5.8, 7.0 Hz, 1H), 5.61 (dt, J = 1.9, 7.0 Hz, 1H), 6.81-6.90
(m, 4H); 13C NMR (100 MHz, CDCl3) δ 14.1, 18.8, 22.7, 28.2,
28.8, 29.1, 29.3, 29.5, 29.62, 29.63, 29.66, 31.9, 55.7, 69.5, 70.5,
74.9, 75.8, 76.1, 91.2, 114.7, 115.6, 152.3, 154.3, 162.7; ESI-
(R)-1-[(20R,30R)-30-[(400-Methoxyphenoxy)methyl]oxiran-20-yl]-
pentadec-2-yn-1-ol (anti-5). Molecular sieves (4 A, the amount is
not critical if the allyl propargyl alcohol, CH2Cl2, and cumene
hydroperoxide are predried) were added to a solution of D-(-)-
DIPT (363 mg, 1.55 mmol) in 50 mL of dry CH2Cl2. The mixture
was stirred at rt for 30 min before it was cooled to -40 °C. Ti(OPr-
i)4 (367 mg, 1.29 mmol) was added to the reaction mixture, which
was stirred for 30 min. After cumene hydroperoxide (590 mg, 80%
technical grade, 3.10 mmol) was added, the reaction mixture was
stirred for 30 min. A solution of (S)-6 (1.33 g, 3.45 mmol, 60% ee)
in a minimal amount of dry CH2Cl2 was added, and the reaction
mixture was sealed and stored at -20 °C without stirring for
3 days. An aqueous precooled (0 °C) solution of tartaric acid
(10 mL, 10% w/v) was added dropwise, and the mixture was
allowed to warm to rt over 1 h, after which time the solution
became transparent. The organic layer was separated, washed
with brine, and concentrated. The residue was dissolved in Et2O at
0 °C, and the solution was treated with a solution (4 mL) of 30%
w/v NaOH in saturated brine. The two-phase mixture was stirred
vigorously for 1 h at 0 °C. The phases were separated, the aqueous
layer was extracted with Et2O, and the combined organic layers
were dried over Na2SO4. The solvent was evaporated, and the
residue was purified by flash chromatography (hexane/EtOAc 5:1
to 4:1) to afford anti-5 (949 mg, 68% (maximum yield, 75%), 93%
þ
HRMS [M þ H]þ calcd for C27H38Cl3NO4 546.1939, found
546.1944.
(2S,3S,4R)-1-(40-Methoxyphenoxy)-2-azidooctadec-5-yne-3,4-
diol (3). To epoxy alcohol anti-5 (95 mg, 0.246 mmol) in 4.5 mL of
MeOH/H2O (8:1) were added NH4Cl (66 mg, 1.23 mmol) and
NaN3 (160 mg, 2.46 mmol). The reaction mixture was heated at
refluxfor 48 h. The reaction mixturewas allowedtocooltort, and
the solvents were evaporated. The residue was extracted with
CH2Cl2 (3 ꢀ 10 mL). The combined extracts were dried (Na2SO4)
and concentrated. Purification of the residue by flash chroma-
tography (elution with EtOAc/hexane, 5:1 to 3:1 to 5:2) afforded
3 (68 mg, 62%): [R]25D þ22.8 (c 1.2, MeOH); 1H NMR (CDCl3) δ
0.88 (t, J = 7.1 Hz, 3H), 1.19-1.33 (m, 16H), 1.33-1.41 (m, 2H),
1.48-1.56 (m, 2H), 2.25 (dt, J = 2.0, 7.1 Hz, 2H), 2.53 (br s, 1H),
2.60 (br s, 1H), 3.76-3.80 (m, 4H), 3.85-3.90 (m, 1H), 4.18 (dd,
J = 7.1, 10.0 Hz, 1H), 4.42 (dd, J = 3.3, 10.0 Hz, 1H), 4.67-4.71
(m, 1H), 6.82-6.91 (m, 4H);13CNMR (CDCl3) δ 14.1, 18.7, 22.7,
28.5, 28.9, 29.1, 29.3, 29.5, 29.6, 29.7, 31.9, 55.7, 62.2, 64.4, 69.1,
72.7, 76.1, 89.3, 114.7, 115.7, 152.3, 154.3;ESI-HRMS[Mþ Na]þ
calcd for C25H39N3NaO4þ 468.2833, found 468.2834.
1
de): H NMR (500 MHz, CDCl3) δ 0.88 (t, J = 7.1 Hz, 3H),
1.20-1.41 (m, 18H), 1.47-1.55 (m, 2H), 2.03 (br s, 1H), 2.22 (dt,
J = 7.2, 1.9 Hz, 2H), 3.30 (t, J = 2.5 Hz, 1H), 3.50-3.53 (m, 1H),
4362 J. Org. Chem. Vol. 75, No. 13, 2010

Liu et al.
(4R,5S,6S)-6-[(40-Methoxyphenoxy)methyl]-3-dodecyl-5,6-di-
JOCFeatured Article
(2S,3S,4R)-1-(40-Methoxyphenoxy)-3,4-bis(benzyloxy)octadec-
5-yn-2-amine (20). To a solution of 19 (23 mg, 37 μmol) in MeOH
(1 mL) were added Et3N (102 μL, 0.73 mmol) and 1,3-dithiopro-
pane (73 μL, 0.73 mmol). The reaction mixture was stirred over-
nightat50°C. The white precipitate was removed by filtration and
washed twice with MeOH. After the solvent was evaporated, the
residue was dried under high vacuum (0.2 Torr, overnight) and
used directly in the subsequent MTPA ester analysis without
further purification.
hydro-4H-pyrrolo[1,2-c][1,2,3]triazole-4,5-diol (14). A solution
of 20 mg (45 μmol) of diol 3 in 2 mL of toluene was stirred at
90 °C for 48 h. The solvent was evaporated, and the residue was
purified by column chromatography (elution with hexane/
EtOAc 1:1 to 2:3) to provide 14 (19 mg, 95%): [R]25 -20.8
D
(c 0.48, MeOH); 1H NMR (500 MHz, CDCl3) δ 0.88 (t, J = 7.0
Hz, 3H), 1.19-1.39 (m, 16H), 1.53-1.65 (m, 4H), 1.66-1.75 (m,
2H), 2.71 (t, J = 7.7 Hz, 2H), 3.75 (s, 3H), 4.42-4.46 (m, 1H),
4.48-4.53 (m, 1H), 4.73 (q, J = 3.4 Hz, 1H), 4.96-4.99 (m, 1H),
5.21 (d, J = 5.5 Hz, 1H), 6.72-6.80 (m, 4H); 13C NMR (100
MHz, CDCl3) δ 14.1, 22.7, 25.4, 29.1, 29.4, 29.62, 29.65, 29.68,
31.9, 55.7, 64.5, 64.8, 66.9, 77.9, 114.7, 115.9, 137.8, 143.3, 152.0,
154.6; ESI-HRMS [M þ H]þ calcd for C25H40N3O4þ 446.3013,
found 446.3006.
(2S,3S,4R)-2-Azidooctadec-5-yne-1,3,4-triol (22). Diol 3 (75
mg, 0.17 mmol) was dissolved in 2.5 mL of CH3CN/H2O (4:1) at
rt, and CAN (461 mg, 0.84 mmol) was added. The mixture was
stirred at rt until completion of the reaction as monitored by
TLC (about 1 h) and diluted with CHCl3. The resulting solution
was washed with H2O and brine. The organic layer was dried
(Na2SO4) and concentrated. The residue was purified by flash
chromatography (hexane/EtOAc 3:2) to afford triol 22 (52 mg,
(4R,5S,6S)-6-[(40-Methoxyphenoxy)methyl]-3-dodecyl-5,6-di-
hydro-4H-pyrrolo[1,2-c][1,2,3]triazol-4,5-diyl Acetate (15). To a
solution of 10 mg (22 μmol) of 14 in 1 mL of CH2Cl2 was added
100 μL (717 μmol) of Et3N and 50 μL (530 μmol) of Ac2O. The
solution was stirred overnight at rt. After the solvent was
removed, the residue was further dried under high vacuum
(0.7 Torr) for 1 h, dissolved in a minimal amount of CH2Cl2,
and filtered through a pad of silica gel in a buret. The pad was
rinsed with 10 mL of hexane/EtOAc 4:1. Concentration gave
diacetate 15 (11 mg, 98%) as a colorless syrup: 1H NMR (500
MHz, CDCl3) δ 0.88 (t, J = 7.2 Hz, 3H), 1.20-1.37 (m, 18H),
1.61-1.70 (m, 2H), 2.13 (s, 3H), 2.14 (s, 3H), 2.73 (t, J = 7.7
Hz, 2H), 3.75 (s, 3H), 4.48 (dd, J = 2.7, 10.4 Hz, 1H), 4.59 (dd,
J = 3.0, 10.4 Hz, 1H), 4.87 (dt, J = 4.3, 2.9 Hz, 1H), 6.05 (dd,
J = 4.3, 5.7 Hz, 1H), 6.31 (d, J = 5.8 Hz, 1H), 6.73-6.81 (m,
4H); 13C NMR (125 MHz, CDCl3) δ 14.1, 20.4, 20.5, 22.7,
25.3, 29.2, 29.3, 29.4, 29.57, 29.62, 29.64, 29.7, 31.9, 55.6, 62.2,
64.6, 66.4, 76.8, 114.6, 116.1, 134.6, 143.7, 151.7, 154.7, 169.5,
169.6.
(4S,5S,6S)-6-[(40-Methoxyphenoxy)methyl]-3-dodecyl-5,6-di-
hydro-4H-pyrrolo[1,2-c][1,2,3]triazol-4,5-diyl Acetate (150). Com-
pound 150 was prepared from syn-5 (35% de) according to the
same sequence used to convert anti-5 to 15 and was contaminated
with15 (ratio of 150/ent-15 = 1.2:1): 1H NMR (500 MHz, CDCl3)
δ 0.88 (t, J = 7.2 Hz, 3H), 1.20-1.37 (m, 18H), 1.61-1.70 (m,
2H), 2.14 (s, 3H), 2.15 (s, 3H), 2.69-2.72 (m, 2H), 3.76 (s, 3H),
4.51-4.53 (m, 2H), 4.72-4.75 (m, 1H), 5.85 (dd, J = 1.5, 1.9 Hz,
1H), 6.03 (d, J = 1.3 Hz, 1H), 6.73-6.81 (m, 4H); 13C NMR (100
MHz, CDCl3) δ 14.1, 20.4, 20.5, 22.7, 25.3, 29.2, 29.3, 29.4, 29.57,
29.62, 29.64, 29.7, 31.8, 55.6, 64.0, 67.0, 69.4, 84.3, 114.6, 115.8,
134.9, 143.6, 151.9, 154.5, 169.7, 169.8.
91%): [R]25 þ43.2 (c 0.5, MeOH); 1H NMR (500 MHz,
D
CD3OD) δ 0.90 (t, J = 7.1 Hz, 3H), 1.23-1.35 (m, 16H),
1.39-1.46 (m, 2H), 1.49-1.56 (m, 2H), 2.25 (dt, J = 2.0, 7.0
Hz, 2H), 3.52-3.57 (m, 2H), 3.69 (dd, J=7.2, 11.5 Hz, 1H),
3.96-4.00 (m, 1H), 4.45-4.47 (m, 1H); 13C NMR (100 MHz,
CDCl3) δ 14.1, 18.7, 22.7, 28.5, 28.9, 29.1, 29.3, 29.5, 29.62,
29.65, 31.9, 62.7, 63.9, 64.2, 73.7, 76.2, 89.3; ESI-HRMS [M þ
Na]þ calcd for C18H33N3O3Naþ 362.2414, found 362.2412.
D-ribo-Phytosphingosine (1). To a solution of 34 mg (0.10
mmol) of triol 22 in 5 mL of MeOH was added 11 mg (0.020
mmol) of 20% Pd(OH)2/C. The resulting suspension was de-
gassed three times and was stirred with a balloon filled with H2
overnight. The crude reaction mixture was filtered through a
short pad of Celite, which was washed with 30 mL of MeOH.
The combined filtrates were concentrated and purified by flash
chromatography (CHCl3/MeOH/concd NH4OH 130:25:4) to
afford 1 (20 mg, 63%) as a white solid. The product was
dissolved in a minimum volume of CHCl3 and passed through
a 0.45-μm filter to remove the suspended silica gel: mp 99-
101 °C (lit.9 mp 98.5-101.5 °C); [R]25D þ8.0 (c 0.8, C5H5N) [lit.9
[R]25 þ7.3 (c 0.9, C5H5N)]; H NMR (500 MHz, CD3OD) δ
1
D
0.89 (t, J = 7.1 Hz, 3H), 1.22-1.41 (m, 24H), 1.49-1.60 (m,
1H), 1.68-1.77 (m, 1H), 2.94-2.97 (m, 1H), 3.33 (dd, J= 5.4,
7.8 Hz, 1H), 3.47-3.52 (m, 1H), 3.55 (dd, J = 6.8, 10.9 Hz, 1H),
3.75 (dd, J = 4.1, 10.9 Hz, 1H); 13C NMR (125 MHz, CD3OD) δ
14.4, 23.8, 26.6, 30.5, 30.79, 30.82, 31.0, 33.1, 34.8, 55.8, 64.0,
þ
74.4, 76.4; ESI-HRMS [M þ Na]þ calcd for C18H39NNaO3
340.2822, found 340.2823.
D-ribo-Phytosphingosine Tetraacetate (23). Compound 23 was
prepared from 1 according to ref 9: [R]25D þ22.6 (c 0.7, CHCl3)
[lit.39 [R]20D þ21.9 (c 1.1, CHCl3)]; 1H NMR (500 MHz, CHCl3) δ
0.88 (t, J = 7.0 Hz, 3H), 1.16-1.39 (m, 24H), 1.57-1.72 (m, 2H),
2.03 (s, 1H), 2.05 (s, 6H), 2.09 (s, 3H), 4.00 (dd, J = 2.8, 11.7 Hz,
1H), 4.29 (dd, J = 4.7, 11.7 Hz, 1H), 4.44-4.51 (m, 1H), 4.93 (t,
J = 9.9, 2.8 Hz, 1H), 5.11 (dd, J = 2.8, 8.4 Hz, 1H), 6.06 (d, J =
9.4 Hz, 1H); 13C NMR (125 MHz, CHCl3) δ 14.1, 20.75, 20.78,
21.1, 22.7, 23.3, 25.5, 28.0, 29.28, 29.34, 29.5, 29.57, 29.61, 29.64,
29.66, 31.9, 47.5, 62.8, 71.8, 73.0, 169.8, 170.1, 170.9, 171.2.
(2S,3S,4R)-2-Azido-3,4-bis(benzyloxy)octadec-5-yn-1-ol (4).
Compound 19 (546 mg, 0.872 mmol) was dissolved in 25 mL
of CH3CN/H2O (4:1) at rt, and CAN (2.39 g, 4.36 mmol) was
added. The mixture was stirred at rt until completion, as
monitored by TLC (about 1 h), and was then diluted with
CHCl3. The resulting solution was washed with H2O and brine.
The organic layer was dried (Na2SO4) and concentrated. The
residue was purified by flash chromatography (elution with
hexane/EtOAc 10:1 to 6:1) to afford 349 mg (77%) of alcohol
4 as a colorless oil: 1H NMR (500 MHz, CDCl3) δ 0.88 (t, J = 7.1
(2S,3S,4R)-1-(40-Methoxyphenyl)-2-azido-3,4-benzyloxy-5-
octadecyne-1,3,4-triol (19). To a mixture of 222 mg (5.56 mmol)
of NaH (60%) and 620 mg (1.39 mol) of diol 3 in 10 mL of
freshly distilled THF were added 343 μL (6.95 mmol) of benzyl
bromide and 3 mg (8 μmol) of TBAI at rt. The mixture was
stirred at rt overnight and then was quenched with 5 mL of
MeOH. The reaction mixture was poured into a mixture of
ice and EtOAc. The organic layer was separated, washed with
aqueous 1 M HCl, water, saturated aqueous NaHCO3 solution,
and brine, and then dried (Na2SO4). The solvent was evapo-
rated, and the residue was purified by flash chromatography
(hexane/EtOAc 40:1) to afford 19 (546 mg, 63%) as a colorless
1
oil. H NMR (500 MHz, CDCl3) δ 0.88 (t, J = 7.1 Hz, 3H),
1.20-1.30 (m, 12H), 1.37-1.44 (m, 2H), 1.51-1.60 (m, 2H),
2.28 (dt, J = 1.6, 7.0 Hz, 2H), 3.77 (s, 3H), 3.82 (t, J = 4.9 Hz,
1H), 4.00-4.06 (m, 2H), 4.18-4.23 (m, 1H), 4.40-4.42 (m, 1H),
4.51 (d, J = 11.6 Hz, 1H), 4.65 (d, J = 11.4 Hz, 1H); 13C NMR
(100 MHz, CDCl3) δ 14.1, 18.9, 22.7, 28.6, 29.0, 29.2, 29.4,
29.55, 29.64, 29.66, 31.9, 55.7, 61.5, 68.3, 70.0, 70.8, 74.3, 75.6,
80.1, 89.1, 114.6, 115.6, 127.8, 128.0, 128.2, 128.3, 128.4, 137.6,
137.8, 152.5, 154.1; ESI-HRMS [M
C39H51N3O4Naþ 648.3772, found 648.3776.
þ
Na]þ calcd for
(39) Shirota, O.; Nakanishi, K.; Berova, N. Tetrahedron 1999, 55, 13643–
13658.
J. Org. Chem. Vol. 75, No. 13, 2010 4363

Liu et al.
JOCFeatured Article
Hz, 3H), 1.20-1.33 (m, 16H), 1.36-1.45 (m, 2H), 1.50-1.58 (m,
2H), 2.28 (dt, J = 1.9, 7.1 Hz, 2H), 2.32 (br s, 1H), 3.72-3.81 (m,
3H), 3.81-3.87 (m, 1H), 4.37-4.40 (m, 1H), 4.50 (d, J = 11.8 Hz,
1H), 4.63 (d, J = 11.4 Hz, 1H), 4.81 (d, J = 11.4 Hz, 1H), 4.86 (d,
J = 11.8 Hz, 1H), 7.26-7.38 (m, 10H); 13C NMR (125 MHz,
CDCl3) δ 14.1, 18.8, 22.7, 28.6, 28.9, 29.1, 29.3, 29.5, 29.60, 29.63,
31.9, 62.1, 63.0, 69.6, 70.7, 73.9, 75.3, 80.8, 89.3, 127.85, 127.89,
128.0, 128.1, 128.38, 128.39, 137.3, 137.5; ESI-HRMS [M þ Na]þ
calcd for C32H45N3NaO3þ 542.3353, found 542.3356.
Acknowledgment. This work was supported in part by
National Institutes of Health Grant No. HL-083187. Z.L.
acknowledges a Doctoral Student Research Grant from the
CUNY Graduate Center. We thank Dr. William F. Berkowitz
for helpful comments in the preparation of the manuscript.
Supporting Information Available: Copies of 1H NMR and
13C NMR spectra of all new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
4364 J. Org. Chem. Vol. 75, No. 13, 2010