RCAI-56, a carbocyclic analogue of KRN7000: its synthesis and potent activity for natural killer (NK) T cells to preferentially produce interferon-γ

Article abstract of DOI:10.1016/j.tetlet.2007.03.074

RCAI-56 (3), a carbocyclic analogue of KRN7000 (1) was synthesized through an efficient coupling of a carba-α-d-galactose derivative 11 with cyclic sulfamidate derivative 13 of phytosphingosine to give 15. Carbasugar derivative 11 was prepared by starting

Full text of DOI:10.1016/j.tetlet.2007.03.074

Tetrahedron Letters 48 (2007) 3343–3347
RCAI-56, a carbocyclic analogue of KRN7000: its synthesis
and potent activity for natural killer (NK) T cells to
preferentially produce interferon-c
Takuya Tashiro,a Ryusuke Nakagawa,a Takatsugu Hirokawa,b Sayo Inoue,a
Hiroshi Watarai,a Masaru Taniguchia and Kenji Moria,*
aLaboratory for Immune Regulation, RIKEN Research Center for Allergy and Immunology, Hirosawa 2-1, Wako-shi,
Saitama 351-0198, and Yokohama Institute, Suehiro-cho 1-7-22, Tsurumi-ku, Yokohama-shi, Kanagawa 230-0045, Japan
bMolecular Modeling and Design Team, Computational Biology Research Center, National Institute of Advanced Industrial
Science and Technology (AIST), Aomi 2-42, Koto-ku, Tokyo 135-0064, Japan
Received 23 February 2007; revised 12 March 2007; accepted 15 March 2007
Available online 19 March 2007
Abstract—RCAI-56 (3), a carbocyclic analogue of KRN7000 (1) was synthesized through an efficient coupling of a carba-a-D-gal-
actose derivative 11 with cyclic sulfamidate derivative 13 of phytosphingosine to give 15. Carbasugar derivative 11 was prepared by
starting from methyl a-D-galactopyranoside (4), employing Pd(II)-catalyzed Ferrier rearrangement as the key step. RCAI-56 (3) is a
potent stimulant of NKT cells in vivo to induce the production of Th1 biased cytokines such as interferon-c in mice. According to
the docking model of CD1d-3 complex, its stabilization energy is approximately at the same level as that of the CD1d-1 complex.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
HO
HO
OH
O
OH
In 1995, researchers at Kirin Brewery Co. developed an
anticancer drug candidate KRN7000 (a-GalCer, 1,
Fig. 1)1 through the modification of the structures of
agelasphins, which had been isolated in 1993 as antican-
cer sphingolipids from the extract of an Okinawan mar-
ine sponge, Agelas mauritianus.2,3 These sphingolipids
exhibited antitumor activity in vivo in mice and human.
It has been shown that 1 is a ligand to make a complex
with CD1d protein, a glycolipid presentation protein on
the surface of the antigen presenting cells of the immune
system.4 Natural killer (NK) T cells are activated by rec-
ognition of CD1d-1 complex with their invariant Va14
antigen receptors, and release both helper T(Th)1 and
Th2 types of cytokines in large quantities at the same
time.5 Th1 type cytokines such as interferon(IFN)-c
mediate protective immune functions like tumor rejec-
tion, whereas Th2 type cytokines such as interleu-
kin(IL)-4 mediate regulatory immune functions to
HO
O
(CH2)12Me
HN
OH
(CH2)23Me
O
KRN7000 (α-GalCer, 1)
HO
HO
OH
O
OH
HO
(CH2)12Me
HN
OH
(CH2)23Me
O
α-C-GalCer (2)
Figure 1. Structure of KRN7000 (1) and a-C-GalCer (2).
ameliorate autoimmune diseases. Th1 and Th2 type
cytokines can antagonize each other’s biological ac-
tions.6 Because of this antagonism, use of 1 for clinical
therapy was unsuccessful. To circumvent this problem,
*
Corresponding author. Tel.: +81 48 462 1339; fax: +81 48 467
9381; e-mail: kjk-mori@arion.ocn.ne.jp
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.03.074

3344
T. Tashiro et al. / Tetrahedron Letters 48 (2007) 3343–3347
many groups are trying to develop new analogues of 1,
which induce NKT cells to produce only Th1 or Th2
type cytokines (see reviews7–9).
of the CD1d-1 complex is placed out of the hydrophobic
pocket of CD1d. Bioassay of analogue 3 would enable
us to examine whether the pyranose O atom is impor-
tant for the recognition by NKT cells or not.
In 2003, Franck, Tsuji and their co-workers reported
that their synthetic a-C-galactosylceramide (a-C-Gal-
Cer, 2) caused an enhanced Th1 type response
in vivo.10 Derivative 2 was developed by replacing the
glycosidic O atom at the galactose-ceramide linkage to
a CH2 group, and three different syntheses of 2 have al-
ready been reported.11–13 Why analogue 2 induces NKT
cells to release Th1 biased cytokines? It is proposed that
2 is stable against a-galactosidase in vivo, and therefore
CD1d-2 complex can stimulate NKT cells for a longer
period than CD1d-1 can do, causing Th1 biased
response.8
As described below, we synthesized 3, coined a code
name RCAI-56 to it, and found it to be a potent activa-
tor of mouse NKT cells to produce interferon-c. It must
be added that, an independent and unsuccessful attempt
of Chung and co-workers to provide the carbocyclic
analogues of 1 was published very recently.16
2. Results and discussion
2.1. Retrosynthetic analysis
According to the X-ray crystallographic analysis of
CD1d-1 complex, the glycosidic O atom of 1 makes a
hydrogen bonding with the a2 helix of CD1d (mouse:
Thr156; human: Thr154).14,15 Based on this result, we
began our attempt to synthesize carbocyclic analogue
3 (Scheme 1), with the linking O atom and hence the sta-
ble ether structure. The new analogue 3 might make a
more stable complex with CD1d than 2, and might
induce NKT cells to release Th1 biased cytokines.
Whereas the O atom of the pyranose ring makes no
hydrogen bonding with CD1d, the galactose part of 1
Scheme 1 shows the retrosynthetic analysis of RCAI-56
(3). It can be prepared from carba-a-D-galactose part A,
phytosphingosine part B, and cerotic acid. We envisage
that Bittman’s cyclic sulfamidate B17 is useful as an elec-
trophile to attack the sodium alkoxide derived from A.
Carba-a-D-galactose derivative A will be prepared from
the known cyclohexanone C, which can be synthesized
from galacto-hexenopyranoside D18 via Pd(II)-catalyzed
Ferrier rearrangement developed by Ikegami.19 The cyc-
lic sulfamidate B is to be prepared from commercially
available phytosphingosine.
2.2. Synthesis of carba-a-D-galactose derivative 11
HO
HO
OH
HO
Scheme 2 summarizes the synthesis of carba-a-D-galact-
ose derivative 11. We adopted a method similar to the
one developed for the synthesis of a carba-a-D-glucose
derivative by Pohl and his co-workers.20 The key reac-
OH
O
(CH2)12Me
HN
OH
(CH2)23Me
O
HO
HO
BnO
OH
O
5 steps
cf. lit.18
RCAI-56 (3)
5 mol% PdCl2
O
BnO
1,4-dioxane, H2O (2:1)
HO
BnO
OMe
OMe
60 °C, 3 h
methyl α-D-galactopyranoside
(4)
5 (= D)
BnO
BnO
O
OBn
HO2C(CH2)24Me
cerotic acid
O
TBSOTf
BnO
BnO
O
O2S
O
O
2,6-lutidine
Tebbe reagent
BnO
NBn
BnO
OH
BnO
(CH2)12Me
pyridine, THF
–40 °C~r.t., 1 h
(88%)
CH2Cl2
BnO
BnO
A
B
OTBS
OH
–20 °C, 1 h
6 (= C)
(89%, 2 steps)
7
BnO
OH
O
BH3•THF
THF, r.t., 2.5 h;
NaH, BnBr
(n-Bu)4NI
BnO
BnO
BnO
OH
HO
(CH2)12Me
BnO
5
BnO
BnO
NH2 OH
OH
then 3 N NaOH aq.
30% H2O2 aq.
0 °C, 1 h
DMF, THF
r.t., 12 h
BnO
BnO
OTBS
OTBS
phytosphingosine
C
8
9
(76%)
BnO
BnO
BnO
BnO
BnO
BnO
OBn
OBn
O
TBAF
THF
r.t., 1 d
(68%, two steps)
BnO
BnO
BnO
OH
OMe
OTBS
D
10
11 (= A)
Scheme 1. Retrosynthetic analysis of RCAI-56 (3).
Scheme 2. Synthesis of carba-a-D-galactose derivative 11.

T. Tashiro et al. / Tetrahedron Letters 48 (2007) 3343–3347
3345
tion was Ikegami’s modification of the Ferrier rear-
rangement (5!6).19 The known building block 5
(=D), was prepared from commercially available methyl
a-D-galactopyranoside (4) through five steps in a high
overall yield.18 The cyclohexane framework of ketone
6 (=C) was constructed by Pd(II)-catalyzed Ferrier rear-
rangement of 5.19 The free hydroxy group of the result-
ing 6 was protected as tert-butyldimethylsilyl(TBS) ether
by treatment with tert-butyldimethylsilyl triflate
(TBSOTf) at ꢀ20 °C to give TBS ether 7 (89%, two
steps). This was then converted to alkene 8 by methyl-
enation with Tebbe reagent (88%). Alkene 8 was sub-
jected to hydroboration-oxidation by treatment with
borane-THF and NaOH aq/H2O2 to furnish alcohol 9
with the desired b-configuration at C-5 (76%). Benzyla-
tion of the free hydroxy group of 9 yielded 10. Removal
of the TBS group of 10 with tetra-n-butylammonium
fluoride (TBAF) furnished the desired tetrabenzylated
carba-a-D-galactose derivative 11 (=A) in 68% yield
(two steps).21
2.3. Synthesis of RCAI-56 (3), the carbocyclic analogue
of KRN7000
Completion of the synthesis of RCAI-56 (3) is illustrated
in Scheme 3. The carba-a-D-galactose derivative 11 was
coupled with cyclic sulfamidate 13 (=B), which was pre-
pared from phytosphingosine (12) by Bittman’s proto-
col,17 to give amine 15 in 76% yield via acid hydrolysis
of sodium sulfamidate intermediate 14. Removal of
the acetonide protective group of amine 15 afforded
aminodiol 16 (97%). Hydrogenolysis of all of the benzyl
groups of 16 was followed by acylation of the resulting
17 with cerotyl chloride (18)13,22 to give RCAI-56 (3) as
a colorless solid, mp 147–149 °C, in 60% yield (two
steps).23 The overall yield of 3 was 6.7% based on methyl
a-D-galactopyranoside (4) through 20 steps.
OH
O
6 steps
lit.17
O
HO
(CH2)12Me
O
O2S
NBn
NH2 OH
(CH2)12Me
13 (= B)
12
2.4. Results of bioassay
BnO
BnO
OBn
BnO
Concentrations of IFN-c and IL-4 in the sera of mice
were measured after their treatment with either 1 or 3.
As shown in Figure 2, RCAI-56 (3) induced Th1 biased
cytokine production in vivo. Indeed, in comparison with
KRN7000 (1), RCAI-56 (3) brought about remarkable
increase in the production of IFN-c with concomitant
decrease in IL-4 release. The structural differences
between 3 and 1 are (i) the lack of the pyranose O atom
and (ii) the replacement of the labile glycosidic linkage
of 1 with a stronger ether linkage. Comparison of the
OBn
O
BnO
OH
11
BnO
O
BnO
O
N
NaH, DMF, THF
70 °C, 20 h
(CH2)12Me
NaO3S
Bn
14
BnO
OBn
BnO
O
BnO
20% H2SO4 aq.
p-TsOH·H2O
O
O
Et2O
0 °C, 15 min
(76%)
MeOH, CH2Cl2
reflux, 10 h
(97%)
BnHN
15
(ng/mL)
(CH2)12Me
serum IFN-γ
35
KRN7000 (1) 0.2 μg
30
RCAI-56 (3) 0.2 μg
25
BnO
BnO
OBn
20
15
10
5
OH
10% Pd-C
BnO
O
(CH2)12Me
cyclohexene
1 N HCl aq.
MeOH
BnHN
OH
16
0
reflux, 6 h
0
6
12
18
24
30
36
42
48
HO
HO
OH
HO
time (h)
ClCO(CH2)24Me
OH
18
(pg/mL)
3000
O
(CH2)12Me
serum IL-4
Et3N, THF, MeOH, CHCl3
r.t., 20 h
H2N
17
OH
KRN7000 (1) 0.2 μg
RCAI-56 (3) 0.2 μg
2500
2000
1500
1000
500
(60%, two steps).
HO
HO
OH
OH
HO
O
(CH2)12Me
HN
OH
(CH2)23Me
0
0
O
3
6
9
12
RCAI-56 (3)
Scheme 3. Synthesis of RCAI-56 (3).
time (h)
Figure 2. The concentrations of IFN-c or IL-4 in serum in mice.24

3346
T. Tashiro et al. / Tetrahedron Letters 48 (2007) 3343–3347
References and notes
1. 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, 2176–2187.
2. Natori, T.; Koezuka, Y.; Higa, T. Tetrahedron Lett. 1993,
34, 5591–5592.
3. Natori, T.; Morita, M.; Akimoto, K.; Koezuka, Y.
Tetrahedron 1994, 50, 2771–2776.
4. Kawano, T.; Cui, J.; Koezuka, Y.; Toura, I.; Kaneko, Y.;
Motoki, K.; Ueno, H.; Nakagawa, R.; Sato, H.; Kondo,
E.; Koseki, H.; Taniguchi, M. Science 1997, 278, 1626–
1629.
5. Wu, D.; Xing, G.-W.; Poles, M. A.; Horowitz, A.; Kinjo,
Y.; Sullivan, B.; Bodmer-Narkevitch, V.; Plettenburg, O.;
Kronenberg, M.; Tsuji, M.; Ho, D. D.; Wong, C.-H. Proc.
Natl. Acad. Sci. U.S.A. 2005, 102, 1351–1356.
6. Rissoan, M.-C.; Soumelis, V.; Kadowaki, N.; Grouard,
G.; Briere, F.; Malefyt, R. de W.; Liu, Y.-J. Science 1999,
283, 1183–1186.
Figure 3. The docking model of mCD1d-KRN7000 (1, purple) and
mCD1d-RCAI-56 (3, yellow).
7. Savage, P. B.; Teyton, L.; Bendelac, A. Chem. Soc. Rev.
2006, 35, 771–779.
8. Franck, R. W.; Tsuji, M. Acc. Chem. Res. 2006, 39, 692–
701.
9. Berkers, C. R.; Ovaa, H. Trends Pharmacol. Sci. 2005, 26,
252–257.
bioactivities in vivo of a-C-GalCer (2) with 1 or 3 will be
reported elsewhere.
10. Schmieg, J.; Yang, G.; Franck, R. W.; Tsuji, M. J. Exp.
Med. 2003, 198, 1631–1641.
2.5. Docking model of mouse(m)CD1d-RCAI-56 complex
11. Yang, G.; Schmieg, J.; Tsuji, M.; Franck, R. W. Angew.
Chem., Int. Ed. 2004, 43, 3818–3822.
12. Chen, G.; Schmieg, J.; Tsuji, M.; Franck, R. W. Org. Lett.
2004, 6, 4077–4080.
13. Wipf, P.; Pierce, J. G. Org. Lett. 2006, 8, 3375–3378.
14. Zajonc, D. M.; Cantu, C., III; Mattner, J.; Zhou, D.;
Savage, P. B.; Bendelac, A.; Wilson, I. A.; Teyton, L. Nat.
Immunol. 2005, 6, 810–818.
15. Koch, M.; Stronge, V. S.; Shepherd, D.; Gadola, S. D.;
Mathew, B.; Ritter, G.; Fersht, A. R.; Besra, G. S.;
Schmidt, R. R.; Jones, E. Y.; Cerundolo, V. Nat. Immunol.
2005, 6, 819–826.
Comparison between the computational docking mod-
els25 of mCD1d-KRN7000(1) and mCD1d-RCAI-56(3)
complexes are depicted in Figure 3. As can be seen in
the figure, the binding conformations of 1 and 3 are
not so different. The ligand binding score (Glide Score)26
of the CD1d-ligand complexes were also calculated as
ꢀ18.97 for 1 and ꢀ19.12 for 3. Presence of the O atom
between the carba-sugar and the ceramide of 3 therefore
keeps the structure of mCD1d-3 complex almost as
same as that of the complex with 1. This means that
the stability of the connecting ether linkage together
with the absence of the pyranose O atom must have
caused the remarkable Th1 bias of RCAI-56 (3) in its
ability to induce cytokine release.
16. Yu, S.-H.; Park, J.-J.; Chung, S.-K. Tetrahedron: Asym-
metry 2006, 17, 3030–3036.
17. Sun, C.; Bittman, R. J. Org. Chem. 2004, 69, 7694–7699.
18. Das, S. K.; Mallet, J.-M.; Sinay, P. Angew. Chem., Int. Ed.
¨
1997, 36, 493–496.
19. Iimori, T.; Takahashi, H.; Ikegami, S. Tetrahedron Lett.
1996, 37, 649–652.
3. Conclusion
20. Ko, K.-S.; Zea, C. J.; Pohl, N. L. J. Am. Chem. Soc. 2004,
126, 13188–13189.
1
21. The H, and 13C NMR spectra of 11 were identical with
In conclusion, we developed RCAI-56 (3), a new carba-
a-D-galactose analogue of KRN7000. RCAI-56 (3) is a
remarkably potent inducer of Th1 biased cytokine pro-
duction in vivo. Further studies are in progress to clarify
the structural requirements for a glycosphingolipid
ligand in controlling the ratio of Th1/Th2 responses.
those reported by Chung.16
22. Heidelberg, T.; Martin, O. R. J. Org. Chem. 2004, 69,
2290–2301.
23
1
23. Mp 147–149 °C; ½aꢁD +27.8 (c 0.32, pyridine); H NMR
of (400 MHz, pyridine-d5, 25 °C): d = 8.43 (d,
3
J = 8.4 Hz, 1H, NH), 6.85–6.82 (m, 1H, OH), 6.37 (d,
J = 6.4 Hz, 1H, OH), 6.31–6.28 (m, 1H, OH), 6.07 (d, J =
5.2 Hz, 1H, OH), 6.00–5.98 (m, 1H, OH), 5.97 (t,
J = 5.4 Hz, 1H, OH), 5.21–5.18 (m, 1H, 2H), 4.69 (br s,
1H, 400-H), 4.50 (dd, J = 10, 4.0 Hz, 1H, 1-Ha), 4.47–4.43
(m, 1H, 200-H), 4.34–4.18 (m, 5H, 3-, 4-, 100-, 300-H, 600-Ha),
4.26 (dd, J = 10, 5.2 Hz, 1H, 1-Hb), 4.00 (ddd-like,
J = 9.6, 5.4, 4.8 Hz, 1H, 600-Hb), 2.51–2.42 (m, 1H, 500-
H), 2.44 (t, J = 7.6 Hz, 2H, 20-H2), 2.33–2.24 (m, 1H, 5-
Ha), 2.14–2.06 (m, 1H, 5a00-Ha), 2.00 (br t, J = 13 Hz, 1H,
5a00-Hb), 1.98–1.84 (m, 2H, 5-Hb, 6-Ha), 1.82 (quint.-like,
J = 7.6 Hz, 2H, 30-H2), 1.76–1.67 (m, 1H, 6-Hb), 1.50–1.17
(m, 66H, 7-, 8-, 9-, 10-, 11-, 12-, 13-, 14-, 15-, 16-, 17-, 40-,
50-, 60-, 70-, 80-, 90-, 100-, 110-, 120-, 130-, 140-, 150-, 160-,
Acknowledgements
We are grateful to Dr. K. Seino and Mr. H. Kamijuku
at St. Marianna University School of Medicine for their
preliminary contribution in bioassay. We thank Ms. K.
Utsunomiya for her technical support. Our thanks are
due to Professor H. Watanabe and Dr. K. Ishigami
(the University of Tokyo), and also Drs. K. Fuhshuku
and M. Shiozaki (RIKEN Research Center for Allergy
and Immunology) for their helpful comments.

T. Tashiro et al. / Tetrahedron Letters 48 (2007) 3343–3347
3347
170-, 180-, 190-, 200-, 210-, 220-, 230-, 240-, 250-H2), 0.85 (t,
J = 7.2 Hz, 6H, 18-, 260-H3) ppm.
24. The serum samples were obtained at indicated times after
the injection of KRN7000 (1) or RCAI-56 (3). Concen-
tration was measured by ELISA (Cytometric Bead Array
system; BD Bioscience).
25. KRN7000 (1) and RCAI-56 (3) molecules were docked
into the X-ray structure of mCD1d (PDB code: 1Z5L)14
using Glide 4.0 (Schro¨dinger, LLC). Initial structures of 1
and 3 were built based on co-crystallized ligands PBS
and R16 in mCD1d. Hydrogen-bonding interaction
between the ligands and the mCD1d and their visualiza-
tion were generated by MacPyMOL (DeLano Scientific,
LLC).
26. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T.
A.; Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E.
H.; Shelley, M.; Perry, J. K.; Shaw, D. E.; Francis, P.;
Shenkin, P. S. J. Med. Chem. 2004, 47, 1739–1749.