| Literature DB >> 28655912 |
Joren Guillaume1, Jing Wang2, Jonas Janssens1, Soumya G Remesh2, Martijn D P Risseeuw1, Tine Decruy3,4, Mathy Froeyen5, Dirk Elewaut3,4, Dirk M Zajonc2,3, Serge Van Calenbergh6.
Abstract
Invariant Natural Killer T-cells (iNKT-cells) are an attractive target for immune response modulation, as upon CD1d-mediated stimulation with KRN7000, a synthetic α-galactosylceramide, they produce a vast amount of cytokines. Here we present a synthesis that allows swift modification of the phytosphingosine side chain by amidation of an advanced methyl ester precursor. The resulting KRN7000 derivatives, termed α-galactosylsphingamides, were evaluated for their capacity to stimulate iNKT-cells. While introduction of the amide-motif in the phytosphingosine chain is tolerated for CD1d binding and TCR recognition, the studied α-galactosylsphingamides showed compromised antigenic properties.Entities:
Mesh:
Substances:
Year: 2017 PMID: 28655912 PMCID: PMC5487351 DOI: 10.1038/s41598-017-04461-7
Source DB: PubMed Journal: Sci Rep ISSN: 2045-2322 Impact factor: 4.379
Figure 1Structures of α-GalCer (1) and OCH (2) and known phytosphingosine-modified α-GalCer derivatives 3 and 4.
Figure 2Model of the binding of 5a to mouse CD1d. The 5a carbons are depicted in green, while the Tyr73.A carbons are colored in yellow. The possible hydrogen bond with the amide oxygen atom is highlighted. Ribbons are colored as skyblue for the CD1d (chain A) and the mouse NKT TCR chains are colored in grey.
Overview of the synthesized α-galactosylsphingamides.
| R1 | R1 = C7H15 | R1 = C11H23 | R1 = C15H31 | R1 = C19H39 | R1 = C25H51 |
|---|---|---|---|---|---|
| R2 | |||||
| C9H19 | 5a | ||||
| (CH2)2Ph | 5b | ||||
| (CH2)4Ph | 5c | ||||
| (CH2)6Ph | 5d | 6d | 7d | 8d | 9d |
| (CH2)8Ph | 5e | ||||
| Ph- | 5f | 9f | |||
| Ph- | 5g | ||||
| ((CH2)2O)2C2H5 | 5h |
Figure 3Retrosynthesis of the target α-galactosylsphingamides.
Figure 4Reagents and conditions: (a) (i) Et3N, H2O, MeOH, RT, 4d; (ii) NaH, BnBr, DMF, RT, overnight, 94% (over 2 steps); (b) 4 m H2SO4, DMF, 0 °C → RT, overnight, 85%; (c) methyl (triphenylphosphoranylidene)acetate, toluene, 85 °C, 6 h, 79%; (d) NiCl2∙6 H2O, NaBH4, MeOH/THF, 0 °C, 1 h, 94%; (e) PPh3, diethylazodicarboxylate, diphenylphosphorylazide, THF, −20 °C → RT, 7 h, 95%; (f) (i) ZnCl2, AcOH/Ac2O, 3.5 h, RT; (ii) MeOH, NaOMe, pH 10, RT, overnight, 67% over 2 steps; (g) TMSOTf, THF, −30 °C, 2 h, 67%; (h) AlMe3, amine, CH2Cl2, reflux, overnight, 56%-85%; (i) (i) PMe3, H2O, THF, RT, 7 h; (ii) EDC, octanoic acid, CH2Cl2, RT, overnight, 51%–69% over 2 steps; (j) H2, Pd black, MeOH, CHCl3, RT, overnight, 37%–58%.
Figure 5IL-4 and IFN-γ secretion, measured at respectively 4 h and 16 h, after intraperitoneal injection of 5 µg of the glycolipids in mice. Data for one individual experiment using 8 mice for each glycolipid.
Figure 6Structures of reference compounds PBS-25 (22) and 23.
Figure 7Real-time TCR-binding kinetics to mCD1d-presented ligands. Each curve represents the TCR binding sensorgram to CD1d-glycolipid complexes at a different TCR concentration after reference substraction (binding response to CD1d without added glycolipid). Sensorgrams (top) indicate similar lipid loading levels for 5d-e. Single cycle TCR kinetics were measured for PBS-25 and 23. Here, TCR with increasing concentrations (3-fold) was sequentially added to CD1d-glycolipid with a final dissociation step of 10 min. Colored curves represent actual binding responses and black curves the fitted data. Kinetic values derived from the sensorgrams are listed below.
Figure 8IFN-γ and IL-4 secretion, measured at 16 h and 4 h respectively, after intraperitoneal injection of 5 µg of the glycolipids in mice. Data for one individual experiment using 8 mice for each glycolipid.
Figure 9(A,B) Electron density map of 5d and 23 with CD1d. Spacer lipid (palmitic acid) is depicted in orange. (C,D) H-bond interactions between 5d and 23 (yellow) with CD1d (grey). (E,F) Glycolipid ligand presentation shown in “TCR view” from top with molecular surface of CD1d in grey and ligands as yellow sticks.
Figure 10Comparison between 5d (green) and 23 (grey)presented by CD1d.
Data collection and refinement statistics of crystal structures of the CD1d-23 and CD1d-5d complexes.
| Data collection statistics | CD1d-23 | CD1d-5d |
|---|---|---|
| Space group | P212121 | P212121 |
| Cell dimension | ||
| | 42.2, 106.2, 106.9 | 42.2, 107.5, 109.6 |
| a, b, g (°) | 90, 90, 90 | 90, 90, 90 |
| Resolution range (Å) [outer shell] | 45–1.85 [1.92–1.85] | 40–1.75 [1.81–1.75] |
| No. reflections | 40,100 | 51,068 |
| Rmeas (%) | 13.3 [97.9] | 6.9 [58.1] |
| Rpim (%) | 5.5 [41.3] | 3.3 [27.5] |
| Multiplicity | 5.4 [5.2] | 4.3 [4.3] |
| Average I/sI | 17.1 [2.5] | 35.8 [3.1] |
| Completeness (%) | 95.2 [98.6] | 99.7 [100.0] |
|
| ||
| No. atoms | 3,435 | 3,385 |
| Protein | 2,992 | 2,929 |
| Ligand (spacer/antigen) | 18/46 | 18/46 |
| Carbohydrate | 56 | 52 |
| Water | 323 | 340 |
| R/Rfree (%) | 21.4/24.3 | 20.3/22.1 |
| Ramachandran plot (%) | ||
| Favored | 97.5 | 98.3 |
| Allowed | 100 | 100 |
| R.m.s. deviations | ||
| Bonds (Å) | 0.008 | 0.008 |
| Angles (°) | 1.37 | 1.33 |
| B-factors (Å2) | ||
| Protein | 24.4 | 31.8 |
| Spacer/Lipid | 40.8/29.4 | 52.5/60.7 |
| Carbohydrate | 44.7 | 46.8 |
| Water | 31.0 | 39.7 |