Artigo Acesso aberto Revisado por pares

Galactose-modified iNKT cell agonists stabilized by an induced fit of CD1d prevent tumour metastasis

2011; Springer Nature; Volume: 30; Issue: 11 Linguagem: Inglês

10.1038/emboj.2011.145

ISSN

1460-2075

Autores

Sandrine Aspeslagh, Yali Li, Esther Dawen Yu, Nora Pauwels, Matthias Trappeniers, Enrico Girardi, Tine Decruy, Katrien Van Beneden, Koen Venken, Michael Drennan, Luc Leybaert, Jing Wang, Richard W. Franck, Serge Van Calenbergh, Dirk M. Zajonc, Dirk Elewaut,

Tópico(s)

T-cell and B-cell Immunology

Resumo

Article6 May 2011free access Galactose-modified iNKT cell agonists stabilized by an induced fit of CD1d prevent tumour metastasis Sandrine Aspeslagh Sandrine Aspeslagh Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Yali Li Yali Li Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA Search for more papers by this author Esther Dawen Yu Esther Dawen Yu Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA Search for more papers by this author Nora Pauwels Nora Pauwels Faculty of Pharmaceutical Sciences, Laboratory for Medicinal Chemistry, Ghent University, Ghent, Belgium Search for more papers by this author Matthias Trappeniers Matthias Trappeniers Faculty of Pharmaceutical Sciences, Laboratory for Medicinal Chemistry, Ghent University, Ghent, Belgium Search for more papers by this author Enrico Girardi Enrico Girardi Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA Search for more papers by this author Tine Decruy Tine Decruy Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Katrien Van Beneden Katrien Van Beneden Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Koen Venken Koen Venken Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Michael Drennan Michael Drennan Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Luc Leybaert Luc Leybaert Department of Basic Medical Sciences-Physiology Group, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium Search for more papers by this author Jing Wang Jing Wang Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA Search for more papers by this author Richard W Franck Richard W Franck Department of Chemistry, Hunter College of CUNY, New York City, NY, USA Search for more papers by this author Serge Van Calenbergh Serge Van Calenbergh Faculty of Pharmaceutical Sciences, Laboratory for Medicinal Chemistry, Ghent University, Ghent, Belgium Search for more papers by this author Dirk M Zajonc Corresponding Author Dirk M Zajonc Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA Search for more papers by this author Dirk Elewaut Corresponding Author Dirk Elewaut Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Sandrine Aspeslagh Sandrine Aspeslagh Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Yali Li Yali Li Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA Search for more papers by this author Esther Dawen Yu Esther Dawen Yu Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA Search for more papers by this author Nora Pauwels Nora Pauwels Faculty of Pharmaceutical Sciences, Laboratory for Medicinal Chemistry, Ghent University, Ghent, Belgium Search for more papers by this author Matthias Trappeniers Matthias Trappeniers Faculty of Pharmaceutical Sciences, Laboratory for Medicinal Chemistry, Ghent University, Ghent, Belgium Search for more papers by this author Enrico Girardi Enrico Girardi Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA Search for more papers by this author Tine Decruy Tine Decruy Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Katrien Van Beneden Katrien Van Beneden Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Koen Venken Koen Venken Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Michael Drennan Michael Drennan Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Luc Leybaert Luc Leybaert Department of Basic Medical Sciences-Physiology Group, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium Search for more papers by this author Jing Wang Jing Wang Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA Search for more papers by this author Richard W Franck Richard W Franck Department of Chemistry, Hunter College of CUNY, New York City, NY, USA Search for more papers by this author Serge Van Calenbergh Serge Van Calenbergh Faculty of Pharmaceutical Sciences, Laboratory for Medicinal Chemistry, Ghent University, Ghent, Belgium Search for more papers by this author Dirk M Zajonc Corresponding Author Dirk M Zajonc Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA Search for more papers by this author Dirk Elewaut Corresponding Author Dirk Elewaut Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium Search for more papers by this author Author Information Sandrine Aspeslagh1,‡, Yali Li2,‡, Esther Dawen Yu2, Nora Pauwels3, Matthias Trappeniers3, Enrico Girardi2, Tine Decruy1, Katrien Van Beneden1, Koen Venken1, Michael Drennan1, Luc Leybaert4, Jing Wang2, Richard W Franck5, Serge Van Calenbergh3, Dirk M Zajonc 2 and Dirk Elewaut 1 1Department of Rheumatology, Faculty of Medicine and Health Sciences, Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium 2Division of Cell Biology, La Jolla Institute for Allergy and Immunology, La Jolla, CA, USA 3Faculty of Pharmaceutical Sciences, Laboratory for Medicinal Chemistry, Ghent University, Ghent, Belgium 4Department of Basic Medical Sciences-Physiology Group, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium 5Department of Chemistry, Hunter College of CUNY, New York City, NY, USA ‡These authors contributed equally to this work *Corresponding authors: Division of Cell Biology, La Jolla Institute for Allergy and Immunology, 9420 Athena Circle, La Jolla, CA 92037, USA. Tel.: +1 858752 6605; Fax: +1 858752 6985; E-mail: [email protected] of Rheumatology, Laboratory for Molecular Immunology and Inflammation, Ghent University, De Pintelaan 185, 9000 Ghent, Belgium. Tel.: +32 (9)3322240; Fax: +32 (9)3323803; E-mail: [email protected] The EMBO Journal (2011)30:2294-2305https://doi.org/10.1038/emboj.2011.145 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Invariant natural killer T (iNKT) cells are known to have marked immunomodulatory capacity due to their ability to produce copious amounts of effector cytokines. Here, we report the structure and function of a novel class of aromatic α-galactosylceramide structurally related glycolipids with marked Th1 bias in both mice and men, leading to superior tumour protection in vivo. The strength of the Th1 response correlates well with enhanced lipid binding to CD1d as a result of an induced fit mechanism that binds the aromatic substitution as a third anchor, in addition to the two lipid chains. This induced fit is in contrast to another Th1-biasing glycolipid, α-C-GalCer, whose CD1d binding follows a conventional key-lock principle. These findings highlight the previously unexploited flexibility of CD1d in accommodating galactose-modified glycolipids and broaden the range of glycolipids that can stimulate iNKT cells. We speculate that glycolipids can be designed that induce a similar fit, thereby leading to superior and more sustained iNKT cell responses in vivo. Introduction Invariant natural killer T (iNKT) cells form a subset of regulatory T cells with features of both innate and adaptive immunity. In contrast to conventional T cells that are activated by a peptide presented by an MHC class I or II molecule, iNKT cells recognize lipid derivatives present in the context of CD1d, a non-classical MHC I molecule expressed on antigen presenting cells (APCs) (Godfrey et al, 2004). The list of CD1d binding glycolipids that can activate iNKT cells is consistently growing and contains different endogenous and exogenous glycolipids. The latter includes very diverse microbial glycolipids and synthetically derived compounds, of which α-galactosylceramide (α-GalCer) is a prototypical glycolipid antigen. α-GalCer consists of a galactose head group that is α-anomerically linked to a phytoceramide, which is composed of a phytosphingosine chain coupled to an acyl chain. The crystal structures of both mouse and human CD1d bound to α-GalCer show that both alkyl chains fit into two pockets of CD1d, the F′ and A′ pocket, while the galactose is exposed at the CD1d surface for interaction with the TCR (Koch et al, 2005; Zajonc et al, 2005a). The invariant α chain of the TCR recognizes the sugar moiety, whereas the TCR-β chain interacts with CD1d residues. However, in contrast to conventional T cells, the complementarity determining regions of the iNKT TCR do not alter their conformation upon antigen recognition (Borg et al, 2007) and appear to be functionally conserved in both mouse and man (Pellicci et al, 2009). In this context, TCR-dependent recognition of α-GalCer results in the robust production of Th1 (e.g., IFN-γ and IL-12), Th2 (e.g., IL-4) and also Th17 cytokines both by iNKT cells itself and by activated bystander cells (Coquet et al, 2008). As several pathological processes are characterized by Th1- or Th2-polarized immune responses, much attention has been focused on attempting to skew iNKT cell cytokine responses towards a more Th1- or Th2-biased cytokine profile. Such strategies have been shown to be effective for Th2 analogues such as OCH, which has been shown to improve the outcome of several autoimmune models including experimental autoimmune encephalitis and arthritis (Miyamoto et al, 2001; Chiba et al, 2004). Conversely, an α-GalCer analogue, in which the oxygen of the glycosidic linkage has been altered to methylene (α-C-GalCer), is known to skew iNKT cell responses towards a Th1 profile and has been applied to the B16 model murine model for melanoma metastasis (Schmieg et al, 2003). Although the underlying mechanisms are currently unclear, several groups have attempted to provide models that explain iNKT cell cytokine polarization. For example, for Th2 analogues such as OCH, the analysis of the crystal structure when bound to CD1d shows the absence of an induced fit above the F′ pocket of CD1d (F′ roof). This has been proposed to result in a shorter half-life for CD1d–OCH complexes, which may modulate the interaction with the iNKT TCR (Sullivan et al, 2010). As a consequence, the ability of iNKT cells to activate accessory cell types such as NK cells is also diminished. For other Th2-polarizing analogues, the ability to directly load CD1d on the surface of cellular membranes was proposed (Bai et al, 2009; Im et al, 2009). Although α-C-GalCer displayed a reduced TCR avidity similar to OCH in vitro, it was shown that the complexes with CD1d were much more stable in vivo, which allowed a more sustained activation of iNKT cells (Sullivan et al, 2010). This in turn is correlated with its uniquely Th1-biasing properties, such as inducing IL-12 in a CD40L-dependent manner (Fujii et al, 2006). Additionally, the role of APCs has been emphasized by the observation that loading of bone marrow dendritic cells (BMDCs) with α-GalCer and α-C-GalCer enhances its Th1-polarizing potencies in contrast to its soluble injection (Fujii et al, 2002). Generally, the currently described Th1- and Th2-biasing α-GalCer analogues, such as α-C-GalCer or OCH, tend to exhibit reduced overall antigenic potencies compared with α-GalCer in vivo. Affirmatively, both α-C-GalCer and OCH have been shown to display a relatively weak interaction with CD1d and TCR. Here, we describe that introduction of selected aromatic groups at position 6′′ of the galactopyranosyl ring may afford analogues with unique functional features, that is, a strong Th1 polarization and tumour suppression. These interesting properties are believed to result from a reinforced interaction with CD1d, caused by the induction of significant conformational changes in CD1d that allows accommodation of the aromatic ring. These findings highlight an hitherto unknown flexibility of CD1d in accommodating glycolipids by inducing the formation of an additional small pocket on top of the A′ roof, independently of the F′ and A′ pockets, leading to more sustained iNKT cell responses in vivo and also broadening the range of glycolipids that can activate iNKT cells. Results 6′′-derivatized α-GalCer analogues induce a marked Th1 polarization The crystal structure of the hCD1d–α-GalCer–TCR trimolecular complex highlighted the importance of the 2′′-OH, 3′′-OH and 4′′-OH groups of the galactose head group, as all these hydroxyl groups form strong hydrogen bonds with Gly96a, Ser30a and Phe29a located on the human invariant TCR-α chain (Borg et al, 2007). The 6′′-OH group is the only hydroxyl group not involved in hydrogen bond formation, suggesting the possibility of introducing modifications without the loss of important interactions. Therefore, we generated a series of analogues with aromatic groups connected via different linkages to the C6′′ of the galactose group aimed at generating extra hydrophobic interactions. For this study, we selected five novel synthetic glycolipids, structurally related to α-GalCer, comprising the 3-epimer of α-GalCer (xylo-α-GalCer) and four 6′′-derivatives (NU-α-GalCer, PhU-α-GalCer, BzNH-α-GalCer and 4ClBzHN-α-GalCer) (Figure 1). Their Th1–Th2 profile was analysed by determining the IFN-γ and IL-4 levels in the serum. NU-α-GalCer, featuring a naphthylurea (NU) substituent, induced serum levels of IFN-γ comparable to α-GalCer combined with markedly reduced IL-4 levels, thereby inducing the most significant Th1 bias from all the studied 6′′-derivatives. By contrast, injection with xylo-α-GalCer induces a weak Th2 profile as compared with the conventionally used Th2 polarizer OCH (Miyamoto et al, 2001; Chiba et al, 2004). Some iNKT cell subsets are known to secrete IL-17 (Coquet et al, 2008). However, IL-17 secretion was undetectable in the serum at different time points by any of the studied analogues (data not shown). Furthermore, 6′′-derivatives induced also a Th1 cytokine bias in mice of a BALB/c background (data not shown) and this was CD1d dependent (Supplementary Figure S2). Thus, addition of an aromatic moiety at the 6′′-position of the galactose moiety leads to a marked functional Th1 polarization in vivo. Figure 1.Th1–Th2 profile of novel glycolipids. (A) Serum cytokine levels at different time points after injection of 5 μg glycolipid (i.p.). IL-4 levels peak at 4 h whereas IFN-γ levels peak at 16 h after injection (upper panel). Graphs indicate the mean with s.e.m. for at least eight mice. Data are of two independent experiments. Therefore, this analysis was extended to all 6′OH analogues at these time points (lower panel). IFN-γ levels of OCH are significantly lower compared with NU-α-GalCer and α-GalCer (Kruskal–Wallis test, Dunn's multiple comparison post-analysis test). Bars indicate the mean with s.e.m. for 8–16 mice/group. Data are of five independent experiments. (B) Serum cytokine levels after injection of glycolipid-pulsed BMDCs. IFN-γ and IL-12 concentration is significantly higher for NU-α-GalCer compared with α-GalCer (two-tailed Mann–Whitney U-test). Bars indicate the mean with s.e.m. of 6 mice/group. Data are representative for two independent experiments. (C) Cytokine levels in supernatant of co-culture human iNKT cells with glycolipid-loaded irradiated PBMCs. Data represent mean with s.e.m. from four experiments using cells from four different donors (two-tailed Mann–Whitney U-test). Download figure Download PowerPoint It has previously been shown that α-GalCer-induced IFN-γ production in vivo can be prolonged by loading the glycolipids onto BMDCs (Fujii et al, 2002). We therefore investigated whether the Th1 profile induced by NU-α-GalCer could be extended when using NU-α-GalCer-loaded BMDCs. As shown in Figure 1B and Supplementary Figure S1, the IFN-γ response to NU-α-GalCer-pulsed BMDCs is markedly higher compared with α-GalCer, leading to a sustained Th1 bias in vivo. In this context, most iNKT cell-dependent IFN-γ production is derived from NK cell activation in an IL-12-dependent manner (Kitamura et al, 1999). In line with this, IL-12 production was found to be markedly elevated for NU-α-GalCer compared with α-GalCer (Figure 1B; Supplementary Figure S2). Furthermore, NU-α-GalCer was also able to induce a Th1-biased cytokine secretion in cultures of human peripheral blood mononuclear cells (PBMCs) (data not shown) and purified human iNKT cells (Figure 1C), highlighting the conserved nature of the Th polarization. Superior tumour protection of 6′′-modified analogues Encouraged by the fact that IL-12 and IFN-γ are the primary mediators of prevention of lung metastasis in the B16 melanoma model (Kim et al, 2000; Kakuta et al, 2002; Airoldi et al, 2007), we decided to directly test the in vivo impact of the Th1 bias elicited by NU-α-GalCer. We therefore evaluated whether the Th1 bias elicited by NU-α-GalCer provided an enhanced protection in the B16 melanoma model when compared with α-GalCer or xylo-α-GalCer, a weak Th2-biasing analogue. In this context, the glycolipids were either directly injected or delivered via BMDCs. When administered directly at high doses, all of the tested glycolipids prevented tumour growth (Figure 2A), whereas at lower doses only NU-α-GalCer, and to a lesser extent α-GalCer, could provide protection (Figure 2A). When glycolipids were loaded onto BMDCs and adoptively transferred, NU-α-GalCer was significantly more potent in preventing tumour growth compared with α-GalCer and xylo-α-GalCer, the latter eliciting very little if any tumour protection under these conditions (Figure 2B and C). When the tumour load was doubled the observed differential effect of NU-α-GalCer versus α-GalCer and xylo-α-GalCer was even more striking (Figure 2D). Figure 2.Tumour suppression by iNKT cell stimulation in a B16 melanoma lung metastasis model. (A) At 5 μg, there was almost no growth of lung nodules. Both with NU-α-GalCer and α-GalCer, at 1 ng, there is still a significant reduction of the amount of lung nodules. (8 mice/group for 5 μg and 16 mice/group for 1 ng) Data are representative of two independent experiments. (B, C) When 10 000 BMDCs loaded with glycolipid were injected, NU-α-GalCer was able to reduce the quantity of lung nodules significantly more than α-GalCer. (NU-α-GalCer versus DMSO, P=0.0002). Each dot represents an individual mouse with at least 6 mice/group. Data are representative of two independent experiments. (D) The protective effect of BMDCs pulsed with NU-α-GalCer is even more pronounced when the tumour load is increased to 400 000 B16 cells. Error bars express median and interquartile range and a Kruskal–Wallis test (Dunn's multiple comparison post-analysis test) was used for statistical analysis. Download figure Download PowerPoint Increased calcium flux and IL-2-dependent NKT cell expansion To further investigate the anti-tumour potency of NU-α-GalCer, we assayed iNKT cell frequency and expansion at several time points after administration of the glycolipid-pulsed BMDCs to mice. Consistent with earlier studies, iNKT expansion could be observed 2 days after transfer of α-GalCer-pulsed BMDCs. Strikingly, NU-α-GalCer induced a more prolonged iNKT cell expansion in both spleen and liver (Figure 3; Supplementary Figure S3). Five days after injection of BMDCs pulsed with NU-α-GalCer, iNKT cell numbers were increased up to 2% of all lymphocytes in spleen and 35% in the liver, representing a nearly six- and two-fold expansion, respectively. To directly evaluate the proliferative capacity of iNKT cells, DNA labelling studies (with BrdU) were performed after adoptive transfer of glycolipid-pulsed BMDCs. Recipients of NU-α-GalCer-pulsed BMDCs had an up to two-fold increase in the BrdU incorporation as opposed to mice injected with α-GalCer-loaded cells, indicating a marked increase in iNKT cell expansion (Figure 3A). We also assessed the impact on human iNKT cell expansion. Here, purified human iNKT cells were labelled with CFSE and co-cultured with PBMCs that had been loaded with NU-α-GalCer or α-GalCer. As was observed in mice in vivo, NU-α-GalCer induced an increased CFSE dilution and thus proliferation of human iNKT cells when compared with α-GalCer (Figure 3B) and this NU-α-GalCer-mediated expansion was dependent upon IL-2 (data not shown). Accordingly upon recognition of the NU-α-GalCer–CD1d complex, iNKT cells produce IL-2 in much higher and sustained quantities than the control glycolipids (Figure 4A). Thus, iNKT cell recognition by NU-α-GalCer induces an increased IL-2 secretion as compared with α-GalCer and ultimately results in a stronger, iNKT cell proliferation. Figure 3.Glycolipid-pulsed bone marrow dendritic cells (BMDCs) induce iNKT cell expansion. (A) Spleen iNKT cells (TCRbintermediate, CD1d–α-GalCer tetramer-positive cells) were stained 1, 2, 3, 4 and 5 days after adoptive transfer of 600 000 glycolipid-loaded BMDCs. Animals were injected intraperitoneally with BrdU 16 h before they were sacrificed. Dot plots and histograms of iNKT cells show a marked increase in expansion and proliferation (as measured by BrdU uptake) 5 days after injection of NU-α-GalCer-loaded BMDCs compared with α-GalCer. At each time point, 3 mice/group were analysed. Data are representative of two independent experiments. (B) Proliferation of human iNKT cells after co-culture with glycolipid-loaded PBMCs. Human iNKT cells (104) from five healthy individuals were labelled with CFSE (1 μM) and stimulated with α-GalCer, NU-α-GalCer or DMSO-loaded irradiated autologous PBMC (105), in the presence of IL-2 (0.1 U/ml). At day 7, cells were harvested and labelled with 6B11 (Ab recognizing the human invariant Vα24Jα18-CDR3 loop) and 7-AAD to gate out dead cells. CFSE dilution of iNKT cells (6B11+) was measured by flow cytometry. Similarly as for mice, NU-α-GalCer induces an increased expansion of human iNKT cells as observed by the stronger CFSE dilution compared with α-GalCer. Data are shown as mean percentage of CFSE diluted cells (with s.e.m.). Graphs shown are representative for one out of five independent experiments. Download figure Download PowerPoint Figure 4.NU-α-GalCer induces a stronger iNKT cell response when bound to bone marrow-derived dendritic cells (BMDCs). BMDCs were grown with GM-CSF for 10 days and subsequently loaded with a glycolipid (100 ng/ml) for 2 h. (A) Co-culture with an iNKT cell line (2C12) was set up for 16 h and IL-2 production was measured by ELISA. Xylo-α-GalCer, NU-α-GalCer and DMSO significantly differ from α-GalCer (two-tailed Mann–Whitney U-test). Data are shown as mean with s.e.m. (n=5). One representative experiment of two independent experiments. (B) The iNKT cell line (2C12) was loaded with FURA-2 and added to glycolipid-pulsed BMDCs. Five minutes later, calcium signals in iNKT cells were measured using epifluorescence microscopy. On average, about 50 cells were monitored for 10 min. An increase of fluorescence of about 10% was determined as a cell displaying an increase in intracellular calcium concentration. The percentage of increase was normalized to 100% for α-GalCer. NU-α-GalCer induced more cells to increase their calcium concentration. Representation of pooled data from four independent experiments. (C) Peak value of transient calcium changes expressed as ratiometric FURA-2 measurements. NU-α-GalCer displays higher increases in calcium concentration compared with α-GalCer. Every dot represents a cell displaying a calcium transient. Data are shown as mean with s.e.m. Four similar experiments were performed. Download figure Download PowerPoint Because IL-2 synthesis is triggered by TCR mediated recognition of antigens and subsequent increase of intracellular calcium levels, we focused on calcium flux elicited by iNKT cells after recognition of the glycolipid/CD1d complex. The role of calcium-dependent signals has not been extensively studied in the context of iNKT cells, although it was suggested to be correlated with the formation of a more stable immunological synapse (McCarthy et al, 2007) and activation of additional cytokine production (Wang et al, 2008). To directly evaluate the impact of the C6′′-modification, we measured intracellular calcium concentration by video microscopy. As indicated in Figure 4, the higher IL-2 production by NU-α-GalCer versus α-GalCer was associated with a marked increase in both the frequency of calcium signalling cells and the calcium signal peak (see also Supplementary videos 1–4). Differential calcium fluxes elicited by glycolipid ligands may be directly related to a different binding stability of the CD1d–glycolipid–T-cell receptor (McCarthy et al, 2007). We therefore focused on the molecular interaction of NU-α-GalCer with mCD1d and the iNKT TCR by surface plasmon resonance (SPR) and crystallography. For comparative purposes, we included the prototypical Th1 analogue α-C-GalCer, as well as the synthetic analogue BnNH-GSL-1′. The latter also possesses an aromatic moiety, which, however, is linked to the sugar via an amide (5′′-CO-NH-) linkage. In that respect it may be considered as an amide derivative of GSL-1′, a galacturonic acid congener of α-GalCer isolated from Sphingomonas species (Kinjo et al, 2005). BnNH-GSL-1′ is also a Th1 polarizer, however, it has a weaker profile as it induces both less IL-4 and IFN-γ compared with NU-α-GalCer (Supplementary Figure S4). Biochemical characterizations We measured the equilibrium binding constants of the refolded Vα14Vβ8.2 TCR towards the α-GalCer analogues NU-α-GalCer, α-C-GalCer and BnNH-GSL-1′ bound to mouse CD1d by SPR (Supplementary Figure S1). With an equilibrium binding constant (KD) of 39.6 nM, the TCR-binding affinity towards NU-α-GalCer is slightly weaker than to α-GalCer (11.2 nM) (Wang et al, 2010), while both BnNH-GSL-1′ and α-C-GalCer are much weaker than α-GalCer (KD of 187 and 247 nM, respectively) but higher in affinity than the bacterial antigen GalA-GSL aka GSL-1′ (KD=0.7 μM) (Wang et al, 2010). The kinetic parameters reveal that the TCR binds all four glycolipids with a similar association rate (ka between 0.98 × 105 and 1.18 × 105/M per second). However, the dissociation of NU-α-GalCer is slightly faster than α-GalCer (2–3 times) (kd=3.8 × 10−3 per second), while both BnNH-GSL-1′ and α-C-GalCer dissociate 15–20 times faster (kd=2.2 × 10−2 and 2.83 × 10−2 per second, respectively) than α-GalCer (ka=1.3 × 105/M per second and kd=1.45 × 10−3 per second) (Wang et al, 2010). In summary, the kinetic data indicate that the initial binding of the TCR to all four mCD1d–glycolipids complexes is very similar. This suggests that their binding to CD1d before TCR engagement is similar, in contrast to microbial antigens that bind differently to mCD1d before TCR engagement and, therefore, have a different TCR association rate (Li et al, 2010). As a result, the chemical differences between the analysed glycolipids are mostly affecting the dissociation rate, and thus the stability of the CD1d–glycolipid–TCR complexes. Crystal structures of the TCR complexes To elucidate the structural details of the mCD1d–NU-α-GalCer–TCR interaction and to understand how the relatively large 6′′-naphthylurea substituent can be accommodated between the mCD1d–TCR interface without a significant reduction in TCR-binding affinity, we crystallized the TCR–NU-α-GalCer–mCD1d complex and determined the structure to 2.3 Å resolution (Figure 5; Supplementary Table S1). For comparison, we have also determined the crystal structures of the mCD1d–α-C-GalCer–TCR complex, as well as the mCD1d–BnNH-GSL-1′–TCR complex to 2.9 and 2.25 Å, respectively. Each structure contains one ternary complex in the asymmetric unit of the crystal and exhibited very clear electron density for each of the glycolipid ligands after molecular replacement, which allowed for unambiguous fitting of the glycolipid ligand (Figure 5). Figure 5.Stereoview of the electron densit

Referência(s)