Synthesis and biological evaluation of α-galactosylceramide (KRN7000) and isoglobotrihexosylceramide (iGb3)

Article abstract of DOI:10.1016/j.bmcl.2006.01.040

Glycoceramides can activate NKT cells by binding with CD1d to produce IFN-γ, IL-4, and other cytokines. An efficient synthetic pathway for α-galactosylceramide (KRN7000) was established by coupling a protected galactose donor to a properly protected ceramide. During the investigation, it was discovered that when the ceramide was protected with benzyl groups, only β-galactosylceramide was produced from the glycosylation reaction. In contrast, the ceramide with benzoyl protecting groups produced α-galactosylceramide. Isoglobotrihexosylceramide (iGb3) was prepared by glycosylation of Galα1-3Galβ1-4Glc donor with 2-azido-sphingosine in high yield. Biological assays on the synthetic KRN7000 and iGb3 were performed using human and murine iNKT cell clones or hybridomas.

Full text of DOI:10.1016/j.bmcl.2006.01.040

Bioorganic & Medicinal Chemistry Letters 16 (2006) 2195–2199
Synthesis and biological evaluation of a-galactosylceramide
(KRN7000) and isoglobotrihexosylceramide (iGb3)
Chengfeng Xia,a,b Qingjia Yao,a,b Jens Schumann,c Emmanuel Rossy,c Wenlan Chen,b
¨
Lizhi Zhu,a,b Wenpeng Zhang,a Gennaro De Liberoc and Peng George Wanga,b,
*
aDepartment of Biochemistry, The Ohio State University, Columbus, OH 43210, USA
bDepartment of Chemistry, The Ohio State University, Columbus, OH 43210, USA
cExperimental Immunology, Department of Research, University Hospital Basel, CH-4031 Basel, Switzerland
Received 17 December 2005; revised 10 January 2006; accepted 11 January 2006
Available online 2 February2006
Abstract—Glycoceramides can activate NKT cells by binding with CD1d to produce IFN-c, IL-4, and other cytokines. An efficient
synthetic pathway for a-galactosylceramide (KRN7000) was established by coupling a protected galactose donor to a properly pro-
tected ceramide. During the investigation, it was discovered that when the ceramide was protected with benzyl groups, only b-ga-
lactosylceramide was produced from the glycosylation reaction. In contrast, the ceramide with benzoyl protecting groups produced
a-galactosylceramide. Isoglobotrihexosylceramide (iGb3) was prepared by glycosylation of Gala1-3Galb1-4Glc donor with 2-azido-
sphingosine in high yield. Biological assays on the synthetic KRN7000 and iGb3 were performed using human and murine iNKT
cell clones or hybridomas.
Ó 2006 Elsevier Ltd. All rights reserved.
OH
O
HO
HO
OH
O
a-Galactosylceramides (a-GalCers) were discovered
from an extract of a marine sponge, Agelas mauritianus,
by the Pharmaceutical Division of the Kirin Brewery
Company in 1993 and were named agelasphins
(AGLs).1 Although a number of b-galactosylceramides
have been reported, AGLs were the first reported galac-
tosylceramides having an a-galactosyl linkage. All these
compounds were active substances in prolonging the life
span of mice when intraperitoneally inoculated with B16
mouse melanoma cells.2 Therefore, various analogues
were synthesized to explore for candidates with similar
antitumor activity and that were readily available from
large-scale chemical synthesis. Among these com-
pounds, AGL-582 emerged as the most desirable candi-
date for clinical application and was designated as
KRN7000 (Fig. 1).3 Further studies revealed that a-Gal-
Cers could stimulate natural killer T (NKT) cells via
presentation by CD1d. The NKT cells can regulate a
variety of microbial, allergic, autoimmune, and tumor
conditions through the rapid secretion of interleukin-4
(IL-4), interferon-c (INF-c), and other cytokines and
HO
HO
C14H29
O
C25H51
O
C25H51
NH
OH
O
OH
O
OH
NH OH
HO
HO
O
O
O
C13H27
O
HO
OH
OH
iGb3
OH
KRN7000
OH
Figure 1. Structures of KRN7000 and iGb3.
chemokines.4 Because a-GalCers were either separated
from a marine sponge or chemically synthesized, they
are not the natural products of mammalian cells. Evi-
dences showed that during the development of NKT
cells from the mainstream T cell precursor pool to ma-
ture NKT cells, some endogenous glycolipid antigens
were essential for this process by presentation of
CD1d.5 Recently, Zhou et al. demonstrated that isogl-
obotrihexosylceramide (iGb3) was the primary endoge-
nous agonist ligand for NKT cells.6 Therefore, there
has been high demand for both KRN7000 and
iGb3 for immunology research and for preclinical
investigation.
Several methods for synthesis of KRN7000 have been
reported, which involved a crucial galactosylation reac-
tion with different galactose donors and lipid accep-
tors.7–10 Herein we used the active ‘armed’ perbenzyl-
Keywords: Immuno stimulator; Natural killer T cells; Glycoceramide;
Glcosylation.
*
Corresponding author. Tel.: +1 614 292 9884; fax: +1 614 688
3106; e-mail: wang.892@osu.edu
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.01.040

2196
C. Xia et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2195–2199
O
C25H51
O
C25H51
NH2 OBn
OH
OBn
NBoc
NBoc
NH2 OH
a
b
HO
NH OH
NH OBz
b, c
O
O
a
HO
C14H29
C14H29
C14H29
C14H29
HO
TrtO
OBn
C14H29
C14H29
OH
OBn
OH
1
2
3
OH
8
OBz
9
OBn
O
BnO
BnO
7
c
β-GalCer
d
e
O
C25H51
NH OBn
C14H29
O
C25H51
BnO
OTCA
HO
OBn
O
BnO
BnO
5
O
C25H51
OBn
O
O
C25H51
NH OBz
C14H29
BnO
BnO
NH OBn
e
f, g
NH OBz
O
d
KRN7000
C14H29
BnO
HO
BnO
O
C14H29
OBn
OBn
OBz
6
4
OBz
11
10
Scheme 1. Reagents and conditions: (a) NaH, BnCl, DMF, 87%; (b)
TFA-H2O (20:1), 85%; (c) C25H51CO2H, EDCI, THF, 91%; (d)
TMSOTf, Et2O-THF (5:1), ꢀ23 °C, 57%; (e) H2 (20 psi), 10% Pd/C,
MeOH, 93%.
Scheme 2. Reagents and conditions: (a) C25H51COSu, THF, 94%; (b)
TrtCl, DMAP, pyridine, 40 °C; (c) BzCl, DMAP, pyridine, 89% for
two steps; (d) pTsOH, MeOH, 92%; (e) 5, TMSOTf, Et2O-THF (5:1),
ꢀ23 °C, 59%; (f) NaOMe, MeOH, 94%; (g) H2 (20 psi), 10% Pd/C,
MeOH, 91%.
galactoside 511 as a donor. We first used a ceramide 4
with 3,4-dihydroxyl groups protected by benzyl groups
as an active ‘armed’ acceptor. As shown in Scheme 1,
the partially protected phytosphingosine 112 was sub-
jected to NaH and then benzyl chloride in dry DMF
to protect the two secondary hydroxy groups with ben-
zyl group. Removal of the ketal and Boc with trifluoro-
acetic acid afforded the amine 3. Then it was reacted
with fatty acid under the condensation with 1-ethyl-3-
(3-dimethylamino-propyl) carbodiimide (EDCI) to fur-
nish the ceramide 4. The ceramide 4 was subjected to
glycosylation with donor 5 with TMSOTf as an activa-
tor under ꢀ23 °C. Surprisingly, the b-glycosylated prod-
uct 6 was separated as a single glycosylation product.
The anomeric configuration was further confirmed by
NMR spectrums of b-GalCer which were consistent
with reported data.13
remaining two hydroxyl groups without work-up to af-
ford 9. The trityl was removed in methanol with the
presence of p-toluenesulfonic acid. Then the ceramide
10 was subjected to glycosylation with donor 5 under
activator TMSOTf. The glycoceramide 11 was isolated
as the sole product which had the desired a-configura-
tion. Then the benzoyl groups on sphingosine and the
benzyl groups on sugar were removed by saponification
in anhydrous methanol with NaOMe and hydrogena-
tion under catalysis of 10% Pd/C, respectively, to pro-
duce KRN7000.14
It is well known that the glycosylation outcome depends
on the delicate balance of donor/acceptor pair reactivity.
Herein we found out that the ‘armed’ sugar donor with
‘armed’ lipid acceptor produced a b-configuration and
‘armed’ sugar donor with ‘disarmed’ lipid acceptor pro-
duced an a-configuration. It is remarkable to see that
the concept of ‘armed’ and ‘disarmed’ even works
between a sugar donor and lipid acceptor.
We hypothesized that the formation of b-anomeric
product was due to the high reactivity of the acceptor.
While the donor was activated by TMSOTf, the armed
acceptor attacked the donor in an SN2 fashion with a
transition state A (Fig. 2), before the full development
of oxacarbonium intermediate, resulting in b-configura-
tion production. We then predicted that a less reactive
acceptor would favor a-glycosylation through a fully
developed oxacarbonium intermediate.
Our synthesis of iGb3 started with the chemical synthe-
sis of Gala1-3Galb1-4Glc reported by our laboratory
(see Scheme 3).15,16 The known 1-benzyl-peracetyl lac-
tose 12 was deacetylated to generate 1-benzyl lactose
13. The regioselective monoalkylation of the C3 hydro-
xy group of the galactose unit with p-methoxybenzyl
chloride (PMBCl) was achieved through activating this
hydroxy group with dibutylstannylene acetal. Original-
ly, acetyl group was used to protect the remaining hy-
droxy groups on compound 13. In the glycosylation of
the trisaccharide donor with the lipid, however, only
an orthoester was obtained. Efforts for transforming this
orthoester to glycoside by treatment with tin chloride or
other Lewis acids failed. The bulky pivaloyl group,
which could significantly restrain the formation of the
orthoester, was then employed to give 15. Oxidative
cleavage of the PMB group by DDQ released the 30-
OH to afford the disaccharide acceptor 16. The glycosyl-
ation of acceptor 16 with perbenzyl phenyl thiogalacto-
side donor 22 was carried out under activation by NIS/
TfOH to afford the protected trisaccharide 17 in a yield
of 92%. At this juncture, the benzyl groups have to be
switched to acetyl groups to avoid the reduction of the
double bond on the lipid during hydrogenation.
Thus, we switched the protecting groups on the cera-
mide from benzyl to benzoyl (as ceramide 10) to
decrease the acceptor reactivity (Scheme 2). The com-
mercially available phytosphingosine was condensed
with hexacosanic acid N-hydroxysuccinimide ester to
give the amide 8.10 Then the primary hydroxy group
was selectively protected with trityl chloride in pyridine.
When TLC showed that the starting material disap-
peared, benzoyl chloride was added to protect the
OBn
O
BnO
B
A
Acceptor
CCl3
OBn
O
BnO
BnO
δ
δ
BnO
BnO
O
BnO
Acceptor
NHTMS
Figure 2. Transition state of glycosylation. (A) The armed acceptor
attacked donor via SN2 pathway; (B) the disarmed acceptor reacted
with fully formed oxacarbonium donor.

C. Xia et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2195–2199
2197
OAc
O
OH
OPiv
O
OH
OH
OAc
O
OH
OPiv
O
AcO
AcO
HO
PMBO
PivO
PMBO
HO
HO
c
a
b
O
O
O
O
O
O
O
O
OBn
OBn
OBn
AcO
AcO
OBn
HO
HO
PivO
OAc
OH
PivO
OH
OH
OBn
OAc
OH
PivO
12
13
BnO
14
15
OAc
O
OPiv
PivO
OPiv
h,i
O
OPiv
e
f, g
OPiv
OPiv
d
OPiv
O
OPiv
OPiv
O
BnO
AcO
O
O
HO
O
O
PivO
OPiv
OBn
BnO
AcO
O
O
O
O
O
PivO
OPiv
17
OBn
PivO
OAc
PivO
PivO
PivO
PivO
16
18
OAc
OAc
OAc
O
AcO
AcO
AcO
AcO
AcO
AcO
O
C25H51
NH
O
O
OPiv
OPiv
OPiv
OPiv
j
k, l
OPiv
OPiv
O
OPiv
OPiv
O
OPiv
O
N3
O
AcO
AcO
AcO
O
O
O
PivO
O
PivO
O
O
O
O
O
C13H27
O
C13H27
PivO
PivO
PivO
PivO
PivO
PivO
PivO
OTCA
OPMB
OPMB
C13H27
19
OH
20
21
HO
HO
OBn
O
N3
O
C25H51
NH
BnO
BnO
OH
O
OH
m, n
OH
HO
SPh
O
HO
O
HO
O
OPMB
O
O
C13H27
BnO
HO
HO
22
23
OH
iGb3
Scheme 3. Reagents and conditions: (a) NaOMe, MeOH, quantitive; (b) Bu2SnO, MeOH, reflux, then PMBCl, TBAI, toluene, 70 °C, 74%; (c) PivCl,
DMAP, pyridine, 70 °C, 86%; (d) DDQ, DCM/H2O, 93%; (e) 22, NIS, TfOH, DCM, ꢀ20 °C, 91%; (f) H2 (20 psi), 10% Pd/C, MeOH, 97%; (g) Ac2O,
pyridine, 93%; (h) BnNH2, THF; (i) CCl3CN, DBU, DCM, 75% for two steps; (j) 23, TMSOTf, DCM, ꢀ20 °C, 82%; (k) PPh3, PhH/H2O, 50 °C;
(l) C25H51CO2H, EDCI, THF, 89% for two steps; (m) 10% TFA in DCM, 87%; (n) NaOMe, MeOH, reflux, 92%.
Selective exposure of the anomeric hydroxy group was
achieved by treatment with benzylamine in THF. It
was then converted to trichloroacetimidate 19, which
was ready for glycosylation with lipid 23.
Although Savage used an amide compound as an accep-
tor for glycosylation, the yield is around 45%.6 It is well
known that the amide has remarkable effects on the gly-
cosylation. So we used azide 2317 instead of amide for
the acceptor. It was also found that if the 3-OH on
sphingosine was protected with an ester, the acyl group
would shift to the amine while in reduction of the
azide.18 PMB is an excellent protecting group which
can be cleavaged by oxidation or in acidic condition.
The glycosylation of Schmidt’s type donor 19 and lipid
23 was performed at ꢀ20 °C for 2 h in 82% yield. The
azide was reduced to an amine by triphenylphosphine
with stoichiometric amount of water, then subjected to
amide formation with cerotic acid and EDCI to afford
protected iGb3 21. Cleavage of the PMB proceeded
smoothly in 10% of trifluoroacetic acid in dichlorometh-
ane. Finally, all the ester groups were removed by reac-
tion with sodium methoxide in refluxing methanol for
24 h to yield iGb3 in 92% yield.19
Liquid chromatography analysis coupled to mass spec-
trometry was performed to control the purity and confirm
the mass of the synthetic compounds. After injection of
large amounts of iGb3 (40 pmol), the base peak chro-
matogram and the ESI-MS spectra revealed highly pure
compounds (Fig. 3A). The m/z 1186.7 and 1146.5 ions
were attributed to the synthetic iGb3 by MS/MS spectra
analysis. Low energy CID experiments performed on
the sodiated adduct precursor ion allowed a partial struc-
tural analysis of iGb3 (Fig. 3B). The characteristic neutral
Figure 3. (A) Base peak chromatogram of iGb3 (40 pmol). The main
chromatographic peak has been attributed to the sodiated form of
iGb3 (m/z 1186.7). (Inset) Positive-ion mode spectra of the iGb3
standard. The m/z 1146.5 ion has been attributed to the dehydrated
iGb3 by MS/MS analysis. (B) ESI-MS/MS spectra of the parent ion at
m/z 1186.7 ([M+Na]+). * represents the sodiated daughter ions. (Inset)
Structure and fragmentation of iGb3 attributed from the ESI-MS/MS
spectra. The nomenclature of Costello et al., expanded by Adams and
Ann, has been used.
loss of the sugar groups (Y ions) confirmed the loss of the
3 hexose groups. Fragmentation of the sodiated forms
gave low intensity peaks and little structural information.
Replacement of sodium by lithium was then performed

2198
C. Xia et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2195–2199
because it provides more informative structural data.
Analysis in the presence of excess lithium was performed
by direct infusion and generated similar fragmentation
patterns (data not shown). These investigations con-
firmed the ceramide moiety as a d18:1/C26:0.
reported in detail. We performed T cell stimulation as-
says using human and murine iNKT cell clones or
hybridomas, as previously described.6 DMSO and meth-
anol were used to solubilize iGb3 and stimulation exper-
iments were performed at a final concentration of 1%
DMSO or methanol, a concentration which is not toxic
to cells. iGb3 stimulated human iNKT cells to release
interleukin-4 (IL-4) and interferon-c (IFN-c) when pre-
sented by hCD1d-transfected human B lymphoblastoid
cells (C1R) expressing high levels of hCD1d (Fig. 4A).
The stimulatory capacity of iGb3 strongly depended
on the solvent used. Only iGb3 dissolved in methanol
was able to stimulate the iNKT clone. IL-4 and IFN-c
release could be observed already at iGb3 concentra-
tions of 5 lg/mL and increased at higher doses in a
dose-dependent manner. Control a-GalCer was active
at doses as low as 10 pg/mL. Hence, iGb3 is less efficient
than the very strong agonist a-GalCer in stimulating the
used iNKT cell clone. The stimulatory capacity of iGb3
was then tested using C1R-hCD1d transfectants and
peripheral blood mononuclear cell (PBMC)-derived hu-
man dendritic cells (DC), that is, professional antigen-
presenting cells (APC) directly generated from fresh hu-
man blood. As shown in Figure 4B, iGb3 employed at
20 lg/mL in the presence of 1% methanol induced the
release of both IL-4 and IFN-c from human iNKT cells.
The extent of stimulation was lower when DC was used
as APC, likely because of the significantly lower expres-
sion levels of CD1d on DC as compared to C1R-hCD1d
transfectants (not shown). Finally, we also employed
iGb3, dissolved in methanol, in a murine iNKT stimula-
tion system using bone marrow-derived mouse DC as
APC and a murine iNKT hybridoma releasing interleu-
kin-2 (IL-2) as responder cells. iGb3 clearly stimulated
the murine iNKT hybridoma already at a concentration
of 0.2 lg/mL (Fig. 4C).
Since there has been some controversy to the reported
activities of iGb3, here the bioassays on iGb3 were
In summary, we developed an efficient synthetic route
for the synthesis of KRN7000 by controlling the galact-
osidic configuration using different protecting groups of
the ceramide. We also synthesized iGb3 from Gala1-
3Galb1-4Glc and 2-azido-sphingosine in high yield. Bio-
logical assays showed KRN7000 and iGb3 could stimu-
late both human and murine iNKT cells to produce
cytokines.
Figure 4. (A) Stimulation of a human Va24i NKT cell clone (JS7) by
human CD1d-transfected C1R cells (C1R-hCD1d) in the presence of
different concentrations of a-GalCer or iGb3. iGb3 was solubilized
using methanol or DMSO, both at final concentration of 1% (v/v) in
the stimulation assay. Diagrams show the mean release of IL-4 (upper
panel) or IFN-c (lower panel) to cell culture supernatant after 40 h,
determined by ELISA, as pg/mL SD. (B) iGb3 (solubilized in
methanol) was used to stimulate the human Va24i NKT cell clone
JS63. C1R-hCD1d cells or human monocyte-derived DC were used as
APC. iGb3 was solubilized in methanol. The bar charts show the mean
release of IL-4 (left diagram) or IFN-c (right diagram) in culture
supernatants (ng/mL SD) in the presence of iGb3 (20 lg/mL) or
methanol alone. Final methanol concentration was 1% (v/v). (C)
Stimulation of the mouse Va14i NKT hybridoma FF1 by murine bone
marrow-derived DC in the presence of different concentrations of
iGb3. iGb3 was solubilized in methanol. The diagram shows the mean
release of IL-2 (pg/mL SD) in cell culture supernatants determined
by ELISA.
Acknowledgments
This project was supported by Ohio State University to
Peng George Wang and by SNF (Grant 3100A0-
109918) to Gennaro De Libero.
References and notes
1. Natori, T.; Koezuka, Y.; Higa, T. Tetrahedron Lett. 1993,
34, 5591.
2. Notari, T.; Morita, M.; Akimoto, K.; Koezuka, Y.
Tetrahedron 1994, 50, 2771.
3. 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.

C. Xia et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2195–2199
2199
4. Brigl, M.; Brenner, M. Annu. Rev. Immunol. 2004, 22, 817.
5. Kronenberg, M. Annu. Rev. Immunol. 2005, 23, 877.
6. Zhou, D.; Mattner, J.; Cantu, C., III; Schrantz, N.; Yin, N.;
Gao, Y.; Sagiv, Y.; Hudspeth, K.; Wu, Y.; Yamashita, T.;
Teneberg, S.; Wang, D.; Proia, R. L.; Levery, S. B.; Savage,
P. B.; Teyton, L.; Bendelac, A. Science 2004, 306, 1786.
7. Morita, M.; Natori, T.; Akimoto, K.; Osawa, T.; Fukus-
ima, H.; Koezuka, Y. Bioorg. Med. Chem. Lett. 1995, 5,
699.
1H), 4.66-4.61 (m, 2H), 4.53 (m, 1H), 4.49 (t, J = 5.8 Hz,
1H), 4.42–4.36 (m, 4H), 4.31–4.29 (m, 1H), 2.43 (t,
J = 7.4 Hz, 2H), 2.23 (m, 1H), 1.92–1.86 (m, 2H), 1.82-
1.77 (m, 2H), 1.63 (m, 1H), 1.44–1.18 (m, 66H), 0.88–0.83
(m, 6H); 13C NMR (125 MHz, pyridine-d5) d 174.56,
102.81, 77.96, 74.29, 72.88, 72.26, 71.58, 69.95, 63.94,
52.78, 38.08, 35.60, 33.40, 31.64, 31.42, 31.31, 31.27, 31.20,
31.14, 31.08, 30.89, 27.78, 27.66, 24.20, 15.54; HRMS
([M+Na]+) 880.7240 calcd. 880.7212.
8. Figueroa-Perez, S.; Schmidr, R. R. Carbohydr. Res. 2000,
328, 95.
15. Janczuk, A. J.; Zhang, W.; Andreana, P. R.; Warrick, J.;
Wang, P. G. Carbohydr. Res. 2002, 337, 1247.
9. Plettenberg, O.; Bodmer-Narkevitch, V.; Wong, C. H.
J. Org. Chem. 2002, 67, 4559.
16. Zhang, W.; Wang, J.-Q.; Li, J.; Yu, L.-B.; Wang, P. G.
J. Carbohydr. Chem. 1999, 18, 1009.
10. Song, S.; Lee, T.; Jung, S.; Kim, D. Synthesis 2004, 6, 847.
11. Alessandro, D.; Alberto, M.; Alessandro, M. J. Org.
Chem. 1999, 64, 933.
12. Shirota, O.; Nakanishi, K.; Berova, N. Tetrahedron 1999,
55, 13643.
17. Taniguchi, M.; Seino, K-i.; Nakayama, T. Nat. Immunol.
2003, 4, 1164.
18. Rai, A. N.; Basu, A. Org. Lett. 2004, 6, 2861.
19. iGb3: 1H NMR (500 MHz, pyridine-d5): d 8.31 (d,
J = 8.0 Hz, 1H), 7.53 (br, 1H), 7.32 (br, 1H), 6.98 (br,
1H), 6.66 (br, 1H), 6.58 (br, 1H), 6.52 (br, 1H), 6.42 (br,
1H), 6.38 (br, 1H), 6.36 (br, 1H), 6.09 (br, 1H), 6.01 (dd,
J = 15.5, 6.0 Hz), 5.91 (dt, J = 15.3, 6.5 Hz, 1H), 5.76 (br,
1H), 5.64 (d, J = 3.8 Hz, 1H), 5.05–5.00 (m, 2H), 4.87 (d,
J = 12 Hz, 1H), 4.81–4.75 (m, 3H), 4.72–4.69 (m, 1H), 4.65
(m, 1H), 4.56 (m, 1H), 4.54–4.49 (m, 3H), 4.47–4.39 (m,
4H), 4.32–4.29 (m, 1H), 4.24–4.20 (m, 3H), 4.15 (dd,
J = 9.5, 2.5 Hz, 1H), 4.05–4.00 (m, 2H), 3.86–3.84 (m, 1H),
2.42 (t, J = 7.0 Hz, 2H), 2.06 (q, J = 6.5 Hz, 2H), 1.82 (m,
2H), 1.39–1.21 (m, 66H), 0.87 (t, J = 6.5 Hz, 3H), 0.86 (t,
J = 7.0 Hz, 3H); 13C NMR (100 MHz, pyridine-d5): d
174.20, 133.54, 133.16, 106.37, 98.59, 83.00, 81.06, 77.58,
77.47, 77.40, 75.68, 73.74, 73.60, 72.52, 71.73, 71.40, 71.30,
66.88, 63.13, 62.92, 62.73, 55.81, 50.56, 37.85, 33.65, 33.04,
33.02, 31.40, 30.94, 30.92, 30.89, 30.81, 30.70, 30.63, 30.54,
30.51, 27.31, 23.83, 15.18; HRMS ([M+Na]+) 1186.8136,
calcd. 1186.8168.
13. b-GalCer: 1H NMR (500 MHz, pyridine-d5) d 8.49 (d,
J = 8.5 Hz, 1H), 5.14 (m, 1H), 4.89 (d, J = 7.5 Hz, 1H),
4.18 (dd, J = 10.5, 5.5 Hz, 1H), 4.53 (d, J = 2.0 Hz, 1H),
4.49–4.43 (m, 2H), 4.41 (d, J = 6.0 Hz, 1H), 4.36 (t,
J = 6.0 Hz, 1H), 4.20 (t, J = 6.4 Hz, 1H), 4.12 (dd, J = 9.4,
2.7 Hz, 1H), 4.02 (t, J = 5.8 Hz, 1H), 2.40 (t, J = 7.3 Hz,
2H), 2.20 (m, 1H), 1.93-1.89 (m, 2H), 1.81–1.76 (m, 2H),
1.67 (m, 1H), 1.43-1.14 (m, 66H), 0.88–0.83 (m, 6H); 13C
NMR (125 MHz, pyridine-d5) d 174.44, 107.22, 77.98,
76.65, 76.28, 73.62, 73.54, 71.67, 71.10, 63.23, 53.19, 37.82,
34.42, 33.05, 33.04, 31.25, 31.10, 30.94, 30.86, 30.83, 30.76,
30.68, 30.55, 27.59, 27.30, 23.86, 15.21. Data can be found
in: Morita, M.; Natori, T.; Akimoto, K.; Osawa, T.;
Fukusima, H.; Koezuka, Y. Bioorg. Med. Chem. Lett.
1995, 5, 699.
14. KRN7000: 1H NMR (500 MHz, pyridine-d5) d 8.43 (d,
J = 8.6 Hz, 1H), 5.55 (d, J = 3.5 Hz, 1H), 5.25–5.21 (m,