Effects of lipid chain lengths in α-galactosylceramides on cytokine release by natural killer T cells

Article abstract of DOI:10.1021/ja045385q

Glycolipid presentation by CD1 proteins has emerged as an important aspect of antigen recognition, and presentation of α-glycosylceramides by CD1d to natural killer T cells has become a central focus in understanding how glycolipid presentation can influence immune responses. An α-galactosylceramide containing relatively long lipid chains has been the subject of intense study because, when presented by CD1d to natural killer T cells, it stimulates the release of both proinflammatory and immunomodulatory cytokines. Using an efficient synthesis of α-galactosylceramides, we have prepared a series of glycolipids in which the lipid chain lengths have been incrementally varied. The responses of natural killer T cells to these glycolipids have been determined, and we have found that truncation of the phytosphingosine lipid chain increases the relative amounts of immunomodulatory cytokines released. In similar fashion, the length of the acyl chain in α-galactosylceramides influences cytokine release profiles. Copyright

Full text of DOI:10.1021/ja045385q

Published on Web 10/02/2004
Effects of Lipid Chain Lengths in r-Galactosylceramides on Cytokine Release
by Natural Killer T Cells
Randal D. Goff,§ Ying Gao,§ Jochen Mattner,† Dapeng Zhou,† Ning Yin,§ Carlos Cantu, III,‡
Luc Teyton,‡ Albert Bendelac,† and Paul B. Savage*,§
Brigham Young UniVersity, ProVo, Utah 84602, UniVersity of Chicago, Chicago, Illinois 60637, and
Scripps Research Institute, La Jolla, California 92037
Received July 30, 2004; E-mail: paul_savage@byu.edu
Antigen presentation to T cells plays a central role in adaptive
and innate immunity.1 While the responses to antigen presentation
can be complex, the process of antigen binding, presentation, and
recognition by T cell receptors is fundamentally a series of
molecular recognition events. In recent years glycolipid presentation
by CD1 proteins has emerged as an important aspect of antigen
Figure 1. Structures of R-galactosyl ceramides.
recognition.2 For example, a member of the CD1 family of proteins,
CD1b, presents glycolipids from mycobacteria (e.g., Mycobacterium
tuberculosis) to T cells,3 which in turn release proinflammatory
cytokines.4 Another CD1 protein, CD1d, has been shown to present
R-glycosylceramides (e.g., 1 in Figure 1) to a subset of T cells
termed natural killer T cells (NKT cells),5 which are primed to
produce and release an array of cytokines.6 This class of glycolipids
was identified from screening of marine natural products for
antitumor activity.7 Preliminary structure-activity studies focusing
on antitumor properties suggested that an acyl chain of ap-
proximately 26 carbons and a phytosphingosine base containing
18 carbons gave the most active compound.7b
In vitro and in vivo studies with NKT cells indicate that an array
of cytokines is released in response to 1 presented by CD1d6 in-
cluding proinflammatory (e.g., interferon-γ (INF-γ)) and immuno-
modulatory cytokines (e.g., interleukin 4 (IL-4)). Release of
proinflammatory cytokines is believed to be responsible for the
antitumor,7 antiviral,8 antibacterial9 effects of 1. On the other hand,
immunomodulatory cytokines produced by NKT cells in response
to 1 can attenuate proinflammatory responses and thereby delay or
prevent the onset of some autoimmune diseases in animal models
including type 1 diabetes10 and animal encephalomyelitis11 (a model
of multiple sclerosis). In some cases, the opposing responses of
NKT cells to 1 (proinflammatory vs immunomodulatory) offset one
another, resulting in no effect of 1 on disease states in animal
models.12 To fully utilize the responses of NKT cells medicinally,
it would be advantageous to bias their responses to glycolipids to
either a proinflammatory or an immunomodulatory response.
Little information is available about how R-galactosylceramide
structure relates to the cytokine release profiles of NKT cells. It
has been suggested that the length of the phytosphingosine chain
can influence cytokine release.13 Glycolipid 2b (Figure 1) was
shown to increase the amount of IL-4 relative to that of INF-γ
released by NKT cells and thereby to offer additional protection
against animal encephalomyelitis as compared to 1. However, the
extent to which glycolipid chain lengths influence cytokine release
profiles from NKT cells has not been established.
Scheme 1. Synthesis of Glycosyl Acceptors 8a-fa
a Reagents (yields in parentheses): a) Ph3PdCH(CH2)nCH3, THF (60-
63%). b) OsO4, NMO, t-BuOH, H2O (51-72% 5, 25-28% 6). c) HCl,
THF; MeOH; HO2C(CH2)nCH3, DCC, NHS, THF; Ac2O, Et3N, DMAP,
THF (45-47%). d) MeONa, MeOH; TBSCl, imidazole, THF; Ac2O, Et3N,
DMAP, THF; HF, THF (50-57%).
and we have found that further truncation of the phytosphingosine
lipid chain increases the IL-4 vs INF-γ bias of released cytokines.
In similar fashion, the length of the acyl chain in R-galactosyl-
ceramides influences cytokine release profiles.
We prepared R-galactosylceramides via a route that differs from
those reported by other groups7b,14 but in our hands proved to be
very efficient. Phytosphingosines were derived from Garner’s
aldehyde (Scheme 1). Olefins 4a-c were prepared as described
by Berova and co-workers15 who generated phytosphingosines from
analogous alkenes using prepared phytosphingosines from alkenes
analogous to 4a-c using Sharpless’ asymmetric dihydroxylation
procedure (giving a 1-to-1 mixture of the D-ribo (5) and L-arabino
(6) diols). We found that use of osmium tetraoxide with NMO alone
gave a readily separable 2-to-1 mixture of diols 5a-c and 6a-c.
Deprotection of 5a-c gave the corresponding phytosphingosines,
which were coupled with varied carboxylic acids followed by
peracylation to give 7a-f. Acylation generating 7a-f was per-
formed to simplify purification of these amphiphilic compounds.
A protection-deprotection strategy was used to generate 8a-f.
Galactosyl donor 10 was prepared in few steps from commercially
available anhydrogalactose 9 (Scheme 2). We found that use of
this galactosyl donor offered advantages over perbenzylated donors
in that it allowed incremental deprotection of the galactosylcer-
amides and better purification strategies. Coupling of donor 10 with
ceramides 8a-f gave predominantly the R-anomers contaminated
with small amounts of the â-anomers. Separation of these anomers
with the full array of protecting groups intact proved to be very
difficult, and removal of all of the protecting groups from 11a-f
Using an efficient synthesis of R-galactosylceramides, we have
prepared a series of glycolipids in which the lipid chain lengths
have been incrementally varied (2a-c and 3a-c, Figure 1). The
responses of NKT cells to these glycolipids have been determined,
§ Brigham Young University.
† University of Chicago.
‡ Scripps Research Institute.
9
13602
J. AM. CHEM. SOC. 2004, 126, 13602-13603
10.1021/ja045385q CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2. Synthesis of Glycosyl Donor 10 and Coupling to
Acceptors 8a-f to Generate 2a-c and 3a-ca
concentrations and with mouse NKT cells verified that IL-4/INF-γ
ratios were independent of glycolipid concentration (see Supporting
Information).
From studies of 1 and 2b, Oki et al.17 have suggested that
truncation of the phytosphingosine chain of 1 leads to a less stable
complex with CD1d and that INF-γ release by NKT cells requires
longer stimulation by CD1d-glycolipid complexes than IL-4
release. The results obtained with 2a-c and 3a-c are consistent
with this model. Furthermore, results with 3a and mouse NKT cells
suggest that loss of too much lipid results in complexes with CD1d
too unstable to interact well with NKT cell receptors.
The profound influence that glycolipid-mediated stimulation of
NKT cells has on the immune system has been demonstrated in
the antitumor effects of 1 and the ability of 1 to inhibit the onset
of autoimmunity. The fact that these are two different and
presumably opposing responses to 1 suggests that 1 may not be an
optimal ligand for generating proinflammatory or immunomodu-
latory responses. Our results suggest that there is a correlation
between lipid chain length and cytokine release profiles, and the
chain-shortened glycolipids bias cytokine release toward an im-
munomodulatory response. This property of glycolipids 2a and 3a
may prove useful in altering immune responses and in the treatment
of certain autoimmune diseases.
a Reagents (yields in parentheses): a) NaH, BnBr, TBAI, DMF (95%).
b) Ac2O, TFA (90%). c) HBr, AcOH, CH2Cl2(75%). d) AgOTf, 8a-f, 4Å
mol sieves, CH2Cl2(45-62%). e) H2, Pd/C, EtOAc, EtOH; MeONa, MeOH
(84-90%).
Acknowledgment. Financial support from the National Institutes
of Health (NIAID) is gratefully acknowledged.
Figure 2. Ratios of cytokines released from cell populations containing
NKT cells stimulated with the glycolipids indicated. Standard errors from
multiple experiments are indicated. (A) Cytokines produced by spleen cells
obtained from the following strains of mice: ([) B6 and (9) BALB/c
stimulated with 2a-c. (B) Cytokines produced by spleen cells obtained
from the following strains of mice: ([) B6 and (9) BALB/c stimulated
with 3b-c. (C) (b) Cytokines produced by a human NKT cell line
stimulated by 3a-c.
Supporting Information Available: Experimental procedures for
preparation of 2a-c and 3a-c; results from NKT cell labeling with
glycolipid-CD1d tetramers; comparison of cytokine ratios from varied
concentrations of glycolipids; amounts of IL-4 and INF-γ produced in
response to each glycolipid. This material is available free of charge
via the Internet at http://pubs.acs.org.
yielded an inseparable mixture of anomers. However, deprotection
of the C2′′ hydroxyl group (on galactose) allowed ready removal
of the trace amounts of the â-anomers. Subsequent removal of the
remaining acetate groups yielded R-galactosylceramides 2a-b and
3a-c.
Presentation of glycolipids to NKT cells in vivo requires antigen-
presenting cells that express CD1d. For one series of assays mouse
spleen cells were used because this cell population contains both a
relatively high number of NKT cells and antigen presenting cells.
To verify that cytokine release in response to 2a-c and 3a-c was
not strain dependent we measured cytokine release from splenocytes
from two strains of mice. In addition, a human NKT cell line was
used, along with antigen presenting cells,16 to assay cytokine release.
Cytokine concentrations were measured using ELISA.
The focus of the cytokine release assays was to compare relative
amounts of INF-γ and IL-4 released in response to 2a-c and 3a-c
as a measure of the bias of the NKT cells toward either proinflam-
matory (INF-γ) or immunomodulatory (IL-4) responses. Ratios of
IL-4 to INF-γ were calculated and plotted vs phytosphingosine and
acyl chain lengths (Figure 2).
Truncation of the phytosphingosine chain of 1 results in release
of greater amounts of IL-4 by NKT cells in both mouse cell lines
(Figure 2A). A similar effect was observed with human NKT cells
(data not shown). This effect extends to relatively short chain
lengths. Notably, the ratio of IL-4 to INF-γ released by NKT cells
is also influenced by the length of the acyl chain (Figure 2, B and
C), with shorter chain lengths increasing this ratio. The exception
is with mouse NKT cells and 3a (Figure 2B). Glycolipid 3a caused
release of relatively small amounts of both cytokines, possibly due
to its inability to form a stable complex with T cell receptors (see
Supporting Information). In contrast to data from mouse NKT cells,
3a caused release of both cytokines from human NKT cells with a
bias toward IL-4 (Figure 2C). Assays performed at varied glycolipid
References
(1) Pier, G. B.; Lyczak, J. B.; Wetzler, L. M., Eds. Immunology, Infection,
and Immunity; ASM Press: Washington, DC, 2004.
(2) Brigl, M.; Brenner, M. B. Annu. ReV. Immunol. 2004, 22, 817-890.
(3) Moody, D. B.; Reinhold; B. B.; Guy, M. R.; Beckman, E. M.; Frederique,
D. E.; Furlong, S. T.; Ye, S.; Reinhold: V. N.; Sieling, P. A.; Modlin, R.
L.; Besra, G. S.; Porcelli, S. A. Science 1997, 278, 283-287.
(4) Rosat, J.-P.; Grant, E. P.; Beckman, E. M.; Dascher, C. C.; Sieling, P.
A.; Frederique, D.; Modlin, R. L.; Porcelli, S. A.; Furlong, S. T.; Brenner,
M. B. J. Immunol. 1999, 162, 366-371.
(5) Kawano, T.; Cui, J.; Koezuka, Y.; Toura, I.; Kaneko, Y.; Motoki, K.;
Ueno, H.; Nakagawa, R.; Sato, H.; Kondo, E.; Koseki, H.; Tanguchi, M.
Science 1997, 278. 1626-1629.
(6) Stetson, D. B.; Mohrs, M.; Reinhardt, R. L.; Baron, J. L.; Wang, Z.-E.;
Gapin L.; Kronenberg, M.; Locksley, R. M. J. Exp. Med. 2003, 198,
1069-1076.
(7) (a) Natori, T.; Koezuka, Y.; Higa, T. Tetrahedron Lett. 1993, 34, 5591-
5592. (b) 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.
(8) Grubor-Bauk, B.; Simmons, A.; Mayrhofer, G.; Speck, P. G. J. Immunol.
2003, 170, 1430-1434.
(9) Nieuwenhuis, E. E.; Matsumoto, T.; Exley, M.; Schleipman, R. A.;
Glickman, J. Bailey, D. T.; Corazza, N.; Colgan, S. P.; Onderdonk, A.
B.; Blumberg, R. S. Nat. Med. 2002, 8, 588-593.
(10) Hong, S.; Wilson, M. T.; Serizawa, I.; Wu, L.; Singh, N.; Naidenko, O.
V.; Miura, T.; Haba, T.; Scherer, D. C.; Wei, J.; Kronenberg, M.; Koezuka,
Y.; Van Kaer, L. Nat. Med. 2001, 7, 1052-1056.
(11) Singh, A. K.; Wilson, M. T.; Hong, S.; Olivarer-Villagomez, D.; Du, C.;
Stanic, A. K.; Joyce, S.; Sriram, S.; Koezuka, Y.; Van Kaer, L. J. Exp.
Med. 2001, 194, 1801-1811.
(12) Pa´l, E.; Tabira, T.; Kawano, T.; Taniguchi, M.; Miyake, S.; Yamamura,
T. J. Immunol. 2001, 166, 662-668.
(13) Miyamoto, K.; Miyake, S.; Yamamura, T. Nature 2001, 413, 531-534.
(14) (a) Figueroa-Pe´rez, S.; Schmidt, R. R. Carbohydr. Res. 2000, 328, 95-
102. (b) Plettenburg, O.; Bodmer-Narkevitch, Wong, C.-H. J. Org. Chem.
2002, 67, 4559-4564.
(15) Shirota, O.; Nakanishi, K.; Berova, N. Tetrahedron 1999, 55, 13643-
13658.
(16) Lee, P. T.; Benlagha, K.; Teyton, L.; Bendelac, A. J. Exp. Med. 2002,
195, 637-641.
(17) Oki, S.; Chiba, A.; Yamamura, Y.; Miyake, S. J. Clin. InVest. 2004, 113,
1631-1640.
JA045385Q
9
J. AM. CHEM. SOC. VOL. 126, NO. 42, 2004 13603