Abstract

Glycosphingolipids (GSLs) from the Sphingomonadaceae family of bacteria have been reported to be potent stimulators of natural killer T cells. These glycolipids include mono-, tri- and tetraglycosylceramides. Here we have prepared the GSL-1 to GSL-4 series of glycolipids and tested their abilities to stimulate natural killer T cells. Among these glycolipids, only GSL-1 (Compound 1) is a potent stimulator. Using a series of synthetic diglycosylceramides, we show that oligoglycosylceramides from Sphingomonadaceae are not effectively truncated to GSL-1 in lysosomes in antigen-presenting cells, possibly because the higher-order GSLs are poor substrates for lysosomal acyltransfer enzymes.

Introduction

The innate immune system is central to controlling microbial growth, and a key aspect of this system is continuous surveillance for compounds that indicate the presence of microbes. For example, recognition and inflammatory responses to lipid A (Compound 2) are mediated through specific binding of this glycolipid by the receptors CD14 and TLR4 on monocytes, macrophages and other antigen-presenting cells (APCs)¹. Although many well-studied organisms produce lipid A, several Gram-negative bacteria do not secrete this glycolipid^{2, 3, 4}, raising the question: does the innate immune system survey for the presence of glycolipids from these bacteria? Among Gram-negative bacteria that do not produce lipid A, the best-studied outer membrane components are -glycosylceramides from the Sphingomonadaceae family of bacteria^{2, 5}. This family includes bacteria to which humans are commonly exposed, and human biliary cirrhosis has been correlated to infections with these organisms⁶.

Structures of glycolipids from Sphingomonas spp. have been proposed on the basis of spectroscopic measurements². GSL-1 is a monoglycosylceramide, and GSL-3 (Compound 3) and GSL-4 (Compound 4) both incorporate glucosamine along with one or two additional sugars (Fig. 1). Most of these GSLs use C₁₈ sphinganine, but a few of them incorporate C₂₁ sphinganine containing a cis double bond or cyclopropyl group.

Figure 1: Structures of potential NKT-cell agonists.

Shown are structures of glycolipids from the Sphingomonadaceae family of bacteria, an -galactosylceramide isolated from the sponge A. mauritianus, the synthetic NKT cell agonists KRN7000 and PBS57, endogenous antigen iGb3, and diglycosylceramides used to observe lysosomal processing of glycolipids.

Full size image (142 KB)

We⁷, along with other groups of researchers^{8, 9}, have shown that heat-killed Sphingomonas spp. and GSL-1 from Sphingomonas spp. are potent stimulators of natural killer T (NKT) cells. In addition, it has been reported NKT cells are stimulated by GSL-4 isolated from Sphingomonas⁹. To establish that bacteria outside the Sphingomonadaceae family are also active, we showed that NKT cell stimulating properties extend to other non-lipid A–producing bacteria in the Alphaproteobacteria subclass⁷.

NKT cells are regulatory T cells with a restricted repertoire of T cell receptors, and consequently this cell type is thought to be 'preprogrammed' to recognize a limited set of antigens¹⁰. Thus, NKT cells have been classified as part of the innate immune system. Stimulation of NKT cells through presentation of glycolipids by the CD1d protein on APCs results in a release of cytokines that influence responses of other aspects of the immune system, and NKT cells have been implicated in responses to numerous disease states^{11, 12}. The considerable impact of NKT cells on immune responses has prompted efforts to understand the types of glycolipid antigen that can stimulate these cells. Much of the initial work with NKT cells focused on antitumor responses to a glycolipid, KRN7000 (Compound 5), derived from a natural product isolated from the marine sponge Agelas mauritianus¹³ (for an example, (Compound 6), see structure in Fig. 1). 'Natural' antigens for NKT cells have since been described^{7, 8, 9, 14, 15, 16}, and include both endogenous and exogenous glycolipids.

Many variants of KRN7000 have been prepared and tested for NKT cell stimulatory activity¹⁷. From these studies, influences of structural variation on CD1d presentation and NKT cell stimulation have been determined. For example, sugars appended on KRN7000 can influence CD1d presentation and NKT cell stimulation: diglycosylceramides, including -Gal-(1-2)--GalCer (Compound 7), -Gal-(1-4)--GalCer (Compound 8) and -GalNAc-(1-4)--GalCer (Compound 9; Fig. 1), stimulate NKT cells only if they are truncated by glycosidases to give KRN7000 (refs. 15,18). Typically, glycolipids are transported to lysosomes in APCs, where they are exposed to a collection of glycosidases. -Glycosylceramides with small molecules appended at C6'' are tolerated by CD1d and the corresponding T cell receptor and do not need to be truncated to allow stimulation of NKT cells¹⁹.

The structural similarities of KNR7000 and GSL-1 suggest that structural variations should have comparable effects on NKT cell stimulation, and it would be expected that GSL-3 and GSL-4 would have to be truncated to GSL-1 to stimulate NKT cells. To determine whether GSL-3 and GSL-4 are stimulatory, and to avoid possible contamination of these glycolipids with GSL-1 isolated from Sphingomonas, we have synthesized these three glycolipids, compared their structures to those of isolated glycolipids, and determined their ability to stimulate cytokine release from NKT cells. This effort has confirmed the proposed structures of GSL-3 and GSL-4. We have also prepared 'GSL-2' (Compound 10) (Fig. 1) to determine the extent to which the carbohydrates of the GSLs are recognized by NKT cells.

Of critical importance in understanding responses to GSLs is determining the ability of glycosidases in the lysosome to process (that is, truncate) GSL-3 and GSL-4 to GSL-1. We therefore prepared the GSL-KNR7000 hybrids -GlcNH₂-(1-4)--GalCer (Compound 11) and -GlcNAc-(1-4)--GalCer (Compound 12) (Fig. 1) to probe both for lysomal glycosidases that can remove -glucosamine in the context of -GalCer and for acyltransfer enzymes that can acylate the amine in the former glycolipid.

To prepare the GSLs, we used a procedure that allowed the generation of all three GSLs found in Sphingomonadaceae and GSL-2, with an emphasis on convergent synthesis. The synthesis of GSL-1 has been reported⁸, and synthesis of the higher-order GSLs required both a consideration of convergence and efficient manipulation of the protecting group. The synthesis was based on known carbohydrate coupling procedures^{20, 21, 22}; however, glucuronic acid is a poor glycosyl donor and acceptor, and its reactivity influenced the synthesis of all of the GSLs. As in all syntheses of oligosaccharides, control and verification of anomeric stereochemistry were central in the design of the synthesis. To maximize the convergence of the synthesis of GSL-4 and to ensure that the anomeric configurations were correctly installed, appropriately protected GSL-2 was prepared and then the distal disaccharide was incorporated. This final coupling involved a primary alcohol and provided reasonable yields and stereocontrol late in the synthesis.

Preparation of GSL-1 and GSL-2 is shown in Scheme 1. Glucose with a p-methoxybenzyl group at C4 was oxidized to the glucuronic acid and converted to the methyl ester, giving Compound 14. We found that a t-butyldimethylsilyl ether at C4 did not completely withstand the oxidation conditions and that a p-methoxybenzyl ether complicated later steps in the synthesis. Consequently, it was necessary to exchange the C4 protecting groups at this stage to give Compound 15. Coupling with the azido version of sphingosine (Compound 16) gave a mixture of anomers, and only after the azide had been reduced and coupled with S-2-hydroxytetradecanoic acid (giving Compound 18) were the anomers separated. The poor glycosyl donor properties of Compound 15 made it necessary to use azide Compound 16, which is a better acceptor than the preformed ceramide. Deprotection of Compound 18 gave GSL-1. Coupling of Compound 18 with Compound 19 gave the anomer in good yield, and subsequent deprotection gave GSL-2.

Scheme 1: Preparation of GSL-1 and GSL-2.

The reagents were as follows (yields in parentheses). (a) BIAB, TEMPO, CH₂Cl₂, H₂O; CH₂N₂, Et₂O (43%). (b) DDQ, CH₂Cl₂; H₂O, TBSCl, imidazole, DMF (81%). (c) Dimethyl(methylthio)sulfonium triflate, DTBMP, 4 Å MS, CH₂Cl₂ (mixture of anomers; 67%). (d) H₂S, pyridine, S-2-hydroxytetradecanoyl chloride, pyridine (20% of the anomer). (e) HF, CH₃CN, H₂O (87%). (f) MeONa, H₂O, THF; Pd/C, H₂, THF, MeOH (40%). (g) Diphenylsulfoxide, Tf₂O, tri-t-butylpyrimidine, 3 Å MS, CH₂Cl₂ (70%). (h) MeONa, H₂O, THF; Pd/C, H₂, THF, MeOH (22%). Ac, acetyl; Bn, benzyl; PMB, p-methoxybenzyl; TBS, t-butyldimethylsilyl.

Full size image (60 KB)

The synthesis of GSL-3 followed a pathway similar to that used in the synthesis of GSL-4 (see Supplementary Scheme 1 online). The key convergent step in the synthesis of GSL-4 (Scheme 2) was the coupling of Compound 23 and Compound 27. It was necessary to reduce the azide selectively and to protect the resulting amine to avoid material losses late in the synthesis, and we found that coupling of mannose with Compound 25 gave the best anomeric selectivity with perbenzoylated mannose. The benzoyl groups tended, however, to slow the subsequent glycosylation. We therefore exchanged the ester groups on Compound 26 to give Compound 27 before the next coupling. The key convergent step occurred with good yield and anomeric selectivity (only a trace of the -anomer was detected). To facilitate removal of the -anomer, after reductive deprotection the material was peracylated to give Compound 28, which was amenable to chromatography. Hydrolysis of the ester was followed by deprotection of the amine, producing GSL-4 in good yield.

Scheme 2: Preparation of GSL-4.

The reagents were as follows (yields in parentheses). (a) Diphenylsulfoxide, Tf₂O, tri-t-butylpyrimidine, CH₂CH₂ (64%). (b) H₂S, pyridine, Et₃N; di-t-butyldicarbonate (82%). (c) HF-pyridine, THF (89%). (d) TMSOTf, 4 Å MS, CH₂Cl₂ (94%). (e) MeONa, MeOH, Ac₂O, DMAP, pyridine (94%). (f) NBS, acetone, H₂O (77%). (g) Diphenylsulfoxide, Tf₂O, tri-t-butylpyrimidine, 3 Å MS, CH₂Cl₂ (82%). (h) Pd/C, H₂, MeOH, THF; Ac₂O, DMAP, pyridine (55%). (i) MeONa, MeOH, H₂O, THF; trifluoroacetic acid, CH₂Cl₂ (66%). Ac, acetyl; Bn, benzyl; Boc, t-butyloxycarbonyl; Bz, benzoyl; Piv, pivaloyl; TBDPS, t-butyldiphenylsilyl.

Full size image (108 KB)

GSL-4 can be readily isolated from Sphingomonas paucimobilis², and we directly compared isolated material to synthetic GSL-4. As noted above, the structure of GSL-4 has been investigated spectroscopically², and synthesis of the proposed structure and favorable comparison to isolated material provided additional evidence in favor of this proposed structure. A challenge associated with the isolation of GSL-4 is that GSL-1 is present and difficult to separate completely from GSL-4. Nevertheless, after careful chromatography, we isolated GSL-4 in pure form. The isolated material contained primarily the ceramide shown in Figure 1. Also present, however, was a GSL-4–containing ceramide derived from a longer-chained (C21) sphinganine with a double bond. Resonances from this double bond appeared as minor resonances in the proton NMR spectrum of isolated GSL-4. Neglecting these minor resonances, the proton NMR spectra of isolated and synthetic GSL-4 were indistinguishable (within the error range of the spectrometer; Supplementary Fig. 1 online). Particular attention was paid to resonances from the protons on anomeric carbons, because the chemical shifts of these resonances are characteristic of the configuration and identity of the sugar. The carbon spectra of isolated and synthetic GSL-4 were indistinguishable, and TLC retention factors were identical.

Preparation of the GSL-KRN7000 hybrids - GlcNH₂-(1-4)--GalCer and -GlcNAc-(1-4)--GalCer (Fig. 1) followed a pathway similar to that giving GSL-2 (Scheme 3). That is, an appropriately protected form of -GalCer was generated (Compound 32) and coupled with the glucosamine derivative Compound 19 (Scheme 1), yielding Compound 33. Deprotection and reduction gave -GlcNH₂-(1-4)--GalCer. Peracylation, followed by ester hydrolysis, gave -GlcNAc-(1-4)--GalCer.

Scheme 3: Preparation of GSL-KRN7000 hybrids -GlcNH₂-(1-4)--GalCer and -GlcNAc-(1-4)--GalCer.

The reagents were as follows (yields in parentheses). (a) Diphenylsulfoxide, Tf₂O, TTBP, 3 Å MS, CH₂Cl₂ (62%). (b) TsOH, MeOH, CH₂Cl₂; TBSCl, imidazole, CH₃CN (79%). (c) Diphenylsulfoxide, Tf₂O, TTBP, 3 Å MS, Compound 19, CH₂Cl₂ (51%). (d) HF, pyridine, THF (86%). (e) Na°, NH₃, THF (56%). (f) Ac₂O, DMAP, pyridine; MeONa, MeOH, THF (48%). Ac, acetyl; Bn, benzyl; Ph, phenyl; TBS, t-butyldimethylsilyl.

Full size image (42 KB)

The collection of synthetic GSLs, isolated GSL-4 and GSL-KRN7000 hybrids allowed observation of both the NKT cell–stimulatory properties of these glycolipids and requirements for lysosomal glycolipid processing. NKT cell stimulation was observed by using a mouse hybridoma, a human NKT cell line, splenocytes from B6 mice and NKT cells isolated from human plasma^{7, 23}. In each of these experiments, APCs were present; thus, not only were the glycolipids presented by these cells, but they were also exposed to the glycosidases present in the lysosomes of these cells.

The stimulatory properties of the GSLs were compared to those of PBS57 (Compound 34), a surrogate for KRN7000 (ref. 24), by using a mouse NKT cell hybridoma (DN32.D3) (Fig. 2a). In initial experiments with GSL-4 isolated from S. paucimobilis, low stimulation of cytokine production was observed at relatively high concentrations of the glycolipid (100 ng/ml). Careful purification, however, yielded a compound that stimulated only at higher concentrations. The stimulatory properties of GSL-1 would make a sample of GSL-4 contaminated with even a trace amount of GSL-1 appear to be stimulatory. Notably, synthetic GSL-4, GSL-3 and GSL-2 did not stimulate the mouse NKT cell hybridoma (data for GSL-2 and GSL-3 not shown). Macrophages and dendritic cells were used to verify that the result was not APC dependent. Similarly, only GSL-1 stimulated human NKT cells (Fig. 2b) when dendritic cells and peripheral blood lymphocytes (PBLs) were used as APCs. In the experiments with the human NKT cell line, we used PBS57 and iGb3 (ref. 15; Compound 35) as references. Modest stimulation was observed with GSL-1 and iGb3 when dendritic cells were used as APCs. With PBLs as APCs, GSL-1 was nearly as potent as PBS57. With splenocytes from B6 mice and with NKT cells isolated from human plasma, a similar result was obtained: PBS57 and GSL-1 caused both interferon- (IFN-) and interleukin 4 (IL-4) release, but GSL-3 and GSL-4 gave no response (Supplementary Figs. 2 and 3 online).

Figure 2: Only GSL-1 and PBS57 cause significant stimulation of NKT cells.

Two types of NKT cell were stimulated with the following glycolipids: GSL-1, squares; GSL-4 (synthetic), circles; GSL-4 (isolated), triangles; PBS57, diamonds; iGb3, crosses. (a) DN32.D3 cells (mouse NKT cell hybridomas). Filled symbols, dendritic cells as APCs; open symbols, macrophages as APCs. (b) Human NKT cells. Filled symbols, dendritic cells as APCs; open symbols, PBLs as APCs. Experiments were run in triplicate; error bars represent 1 s.d.

Full size image (76 KB)

Many different types of lysosomal glycosidases are present in APCs; consequently, it is surprising that GSL-2 is not truncated to GSL-1, yielding a stimulatory glycolipid. Similarly, GSL-3 and GSL-4 might also be expected to be truncated to GSL-1. -Mannosidase²⁵ and -glycosidase A²⁶ are known to be active in lysosomes; thus, both GSL-3 and GSL-4 should be truncated to GSL-2. In higher organisms, amino sugars are generally acylated at nitrogen, and the glucosamines in GSL-2, GSL-3 and GSL-4 are non-acylated. We therefore reasoned that the glycosidase necessary to cleave glucosamine from GSL-2 might be absent or inactive in lysosomes. In addition, lysosomes contain an acyltransfer enzyme that acylates the amines of the terminal -glucosamine in heparin²⁷. It is not clear why the amine group in GSL-2 (and GSL-3 and GSL-4) is not acylated and subsequently truncated to GSL-1. Lysosomal trafficking and transport of glycosylceramides is highly dependent on lipid transport proteins^{28, 29}, and it is possible that, while bound to lipid transfer proteins, GSL-2 is not an effective substrate for acyltransfer enzymes.

To verify that glucosamine is not acylated and cleaved from glycosylceramides in lysosomes, we examined the stimulatory properties of -GlcNH₂-(1-4)--GalCer (Fig. 1). Because we have shown that other, similarly (1-4)-substituted, diglycosylceramides (that is, -Gal-(1-4)--GalCer and -GalNAc-(1-4)--GalCer)¹⁵ are truncated to -GalCer and stimulate NKT cells, lack of stimulation by -GlcNH₂-(1-4)--GalCer would provide evidence of the inability of the APC to acylate and cleave glucosamine from GSL-2. As a control, we prepared -GlcNAc-(1-4)--GalCer (Fig. 1), which would be the product of acylation of -GlcNH₂-(1-4)--GalCer. -N-Acetylglucosaminidase is a lysosomal enzyme (whose absence results in Sanfilippo syndrome)³⁰, which should cleave the N-acetylglucosamine from -GlcNAc-(1-4)--GalCer to yield -GalCer. Comparison of the NKT cell stimulatory properties of -GlcNH₂-(1-4)--GalCer and -GlcNAc-(1-4)--GalCer demonstrated that -GlcNH₂-(1-4)--GalCer is non-stimulatory, whereas -GlcNAc-(1-4)--GalCer is stimulatory, albeit less than PBS57 (Fig. 3). This result supports the conclusion that glycolipids containing glucosamine are inefficient substrates for the lysosomal acyltransfer enzyme.

Figure 3: GlcNAc from -GlcNAc-(1-4)--GalCer is truncated by lysomal processing to generate a stimulatory glycolipid.

DN32.D3 cells were stimulated with the indicated glycolipids by using dendritic cells as APCs. Experiments were run in triplicate; error bars represent 1 s.d.

Full size image (46 KB)

That GSL-1 is a potent stimulator of cytokine production by NKT cells and the other GSLs are non-stimulatory suggests that synthesis of these higher-order GSLs might be a mechanism of immune evasion. We have not determined the influence of stress on GSL1/GSL4 ratios, but this ratio might be exploited by bacteria to avoid detection.

The CD1d-NKT cell system provides a mechanism for surveillance for bacterial glycolipids. An understanding of the glycolipids recognized by this system is essential for identification of the types of organism that stimulate immune responses and the potential contributions of these organisms to immune disorders. Among bacterial glycolipids shown to stimulate NKT cells directly, those from the Sphingomonadaceae family are the most potent. The GSLs include relatively complex oligoglycosylceramides, and the synthesis of this series of glycolipids has provided both confirmation of their structure and the opportunity to determine stimulatory properties with homogeneous materials. Our observation that, among the GSLs, only GSL-1 is a potent stimulator suggests that the oligosaccharides in the higher GSLs are not directly presented by CD1d and recognized by NKT cells. Our studies with the GSL-KRN7000 hybrids demonstrated that the higher GSLs are not truncated to GSL-1, because GSL-2 is not a substrate for acyltransfer enzymes in lysosomes in APCs. Taken together, these results define the extent to which NKT cells respond to different GSLs and the reason why higher-order GSLs are not converted to GSL-1 in lysosomes. These results not only add to our understanding of NKT cell responses but also may lead to improved antigens for these immunoregulatory cells.

Acknowledgments

We acknowledge financial support from the National Institutes of Health (NIAID AI053725). J.M. is a Cancer Research Institute fellow and was supported by a grant from the Lupus Research Institute. A.B. is a Howard Hughes Medical Institute Investigator.

Competing interests statement:

The authors declare no competing financial interests.

Received 7 May 2007; Accepted 30 June 2007; Published online 29 July 2007.

References

1. Miller, S.I., Ernst, R.K. & Bader, M.W. LPS, TLR4 and infectious disease diversity. Nat. Rev. Microbiol. 3, 36–46 (2005). | Article | PubMed | ISI | ChemPort |
2. Kawahara, K., Kuraishi, H. & Zähringer, U. Chemical structure and function of glycosphingolipids of Sphingomonas spp. and their distribution among members of the -4 subclass of Proteobacteria. J. Ind. Microbiol. Biotechnol. 23, 408–413 (1999). | Article | PubMed | ISI | ChemPort |
3. Amano, K. et al. Deficiency of peptidoglycan and lipopolysaccharide components in Rickettsia tsutsugamushi. Infect. Immun. 55, 2290–2292 (1987). | PubMed | ISI | ChemPort |
4. Lin, M. & Rikihisa, Y. Ehrlichia chaffeensis and Anaplasma phagocytophilum lack genes for lipid A biosynthesis and incorporate cholesterol for their survival. Infect. Immun. 71, 5324–5331 (2003). | Article | PubMed | ISI | ChemPort |
5. Naka, T. et al. A novel sphingoglycolipid containing galacturonic acid and 2-hydroxy fatty acid in cellular lipids of Sphingomonas yanoikuyae. J. Bacteriol. 182, 2660–2663 (2000). | Article | PubMed | ISI | ChemPort |
6. Selmi, C. et al. Patients with primary biliary cirrhosis react against a ubiquitous xenobiotic-metabolizing bacterium. Hepatology 38, 1250–1257 (2003). | Article | PubMed | ISI | ChemPort |
7. Mattner, J. et al. Both exogenous and endogenous glycolipid antigens activate NKT cells during microbial infections. Nature 434, 525–528 (2005). | Article | PubMed | ISI | ChemPort |
8. Kinjo, Y. et al. Recognition of bacterial glycosphingolipids by natural killer T cells. Nature 434, 520–525 (2005). | Article | PubMed | ISI | ChemPort |
9. Sriram, V., Du, W., Gervay-Hague, J. & Brutkiewicz, R.R. Cell wall glycosphingolipids from Sphingomonas paucimobilis are CD1d-specific ligands for NKT cells. Eur. J. Immunol. 35, 1692–1701 (2005). | Article | PubMed | ISI | ChemPort |
10. Bendelac, A., Savage, P.B. & Teyton, L. Biology of NKT cells. Annu. Rev. Immunol. 25, 297–336 (2007). | Article | PubMed | ISI | ChemPort |
11. Yu, K.O.A. & Porcelli, S.A. The diverse functions of CD1d-restricted NKT cells and their potential for immunotherapy. Immunol. Lett. 100, 42–55 (2005). | Article | PubMed | ISI | ChemPort |
12. Van Kaer, L. -Galactosylceramide therapy for autoimmune diseases: prospects and obstacles. Nat. Rev. Immunol. 5, 31–42 (2005). | Article | PubMed | ISI | ChemPort |
13. Morita, M. et al. Structure-activity relationship of -galactosylceramides against B16-bearing mice. J. Med. Chem. 38, 2176–2187 (1995). | Article | PubMed | ISI | ChemPort |
14. Fischer, K. et al. Mycobacterial phosphatidylinositol mannoside is a natural antigen for CD1d-restricted T cells. Proc. Natl. Acad. Sci. USA 101, 10685–10690 (2004). | Article | PubMed | ChemPort |
15. Zhou, D. et al. Lysosomal glycosphingolipid recognition by NKT cells. Science 303, 523–527 (2004). | Article | PubMed | ISI | ChemPort |
16. Kinjo, Y. et al. Natural killer T cells recognize diacylglycerol antigens from pathogenic bacteria. Nat. Immunol. 7, 978–986 (2006). | Article | PubMed | ISI | ChemPort |
17. Savage, P.B., Bendelac, A. & Teyton, L. Glycolipids for natural killer T cells. Chem. Soc. Rev. 35, 771–779 (2006). | Article | PubMed | ISI | ChemPort |
18. Prigozy, T.I. et al. Glycolipid antigen processing for presentation by CD1d molecules. Science 291, 664–667 (2001). | Article | PubMed | ISI | ChemPort |
19. Zhou, X.T. et al. Synthesis and NKT cell stimulating properties of fluorophore- and biotin-appended 6''-amino-6''-deoxy-galactosylceramides. Org. Lett. 4, 1267–1270 (2002). | Article | PubMed | ISI | ChemPort |
20. Andersson, F., Fuegedi, P., Garegg, P. & Nashed, M. Synthesis of 1,2-cis-linked glycosides using dimethy(methylthio)sulfonium triflate as promoter and thioglycosides as glycosyl donors. Tetrahedr. Lett. 27, 3919–3922 (1986). | ISI | ChemPort |
21. Garcia, B.A., Poole, J.L. & Gin, D.Y. Direct glycosylations with 1-hydroxy glycosyl donors using trifluoromethanesulfonic anhydride and diphenyl sulfoxide. J. Am. Chem. Soc. 119, 7597–7598 (1997). | Article | ISI | ChemPort |
22. Schmidt, R.R. & Michael, J. Einfache synthese von - und -O-glycosylimidaten. Herstellung von glykosiden und disacchariden. Angew. Chem. Int. Edn. Engl. 19, 731–732 (1980). | Article |
23. Lee, P.T., Benlagha, K., Teyton, L. & Bendelac, A. Distinct functional lineages of human V-24 natural killer T cells. J. Exp. Med. 195, 637–641 (2002). | Article | PubMed | ISI | ChemPort |
24. Liu, Y. et al. A modified -galactosyl ceramide for staining and stimulating natural killer T cells. J. Immunol. Methods 312, 34–39 (2006). | Article | PubMed | ISI | ChemPort |
25. Roces, D.P. et al. Efficacy of enzyme replacement therapy in -mannosidosis mice: a preclinical animal study. Hum. Mol. Genet. 13, 1979–1988 (2004). | Article | PubMed | ISI | ChemPort |
26. Lusis, A.J. & Paigen, K. Properties of mouse -galactosidase. Biochim. Biophys. Acta 437, 487–497 (1976). | PubMed | ISI | ChemPort |
27. Meikle, P.J., Whittle, A.M. & Hopwood, J.J. Human acetyl-coenzyme A: -glucosaminide N-acetyltransferase: kinetic characterization and mechanistic interpretation. Biochem. J. 308, 327–333 (1995). | PubMed | ISI | ChemPort |
28. Zhou, D. et al. Editing of Cd1-bound lipid antigens by endosomal lipid transfer proteins. Science 303, 523–527 (2004). | Article | PubMed | ISI | ChemPort |
29. Sagiv, Y. et al. Cutting edge: impaired glycosphingolipids trafficking and NKT cell development in mice lacking Niemann-Pick type C1 protein. J. Immunol. 177, 26–30 (2006). | PubMed | ISI | ChemPort |
30. Zhao, H.G., Li, H.H., Bach, G., Schmidtchen, A. & Neufeld, E.F. The molecular basis of Sanfilippo syndrome type B. Proc. Natl. Acad. Sci. USA 93, 6101–6105 (1996). | Article | PubMed | ChemPort |

1. Department of Chemistry and Biochemistry, C100 BNSN, Brigham Young University, Provo, Utah 84602, USA.
2. Department of Pathology, 57 East 7th Street, University of Chicago, Chicago, Illinois 60637, USA.
3. Department of Melanoma Medical Oncology, University of Texas M.D. Anderson Cancer Center, 7455 Fannin Street, Houston, Texas 77054, USA.
4. Immunology Department, 10550 North Torrey Pines Road, Scripps Research Institute, La Jolla, California 92037, USA.
5. These authors contributed equally to this work.

Correspondence to: Paul B Savage¹ e-mail: paul_savage@byu.edu

MORE ARTICLES LIKE THIS

These links to content published by NPG are automatically generated.

NPG Journals

by Subject Area

• Chemistry

  • Chemistry
  • Drug discovery
  • Biotechnology
  • Materials
  • Methods & Protocols

• Clinical Practice & Research

  • Cancer
  • Cardiovascular medicine
  • Dentistry
  • Endocrinology
  • Gastroenterology & Hepatology
  • Methods & Protocols
  • Pathology & Pathobiology
  • Urology

• Earth & Environment

  • Earth sciences
  • Evolution & Ecology

• Life sciences

  • Biotechnology
  • Cancer
  • Development
  • Drug discovery
  • Evolution & Ecology
  • Genetics
  • Immunology
  • Medical research
  • Methods & Protocols
  • Microbiology
  • Molecular cell biology
  • Neuroscience
  • Pharmacology
  • Systems biology

• Physical sciences

  • Physics
  • Materials

by A - Z Index