Antigen Specificity of Type I NKT Cells Is Governed by TCR β-Chain Diversity

Garth Cameron, Daniel G. Pellicci, Adam P. Uldrich, Gurdyal S. Besra, Petr Illarionov, Spencer J. Williams, Nicole L. La Gruta, Jamie Rossjohn and Dale I. Godfrey
J Immunol November 15, 2015, 195 (10) 4604-4614; DOI: https://doi.org/10.4049/jimmunol.1501222

Abstract

NKT cells recognize lipid-based Ags presented by CD1d. Type I NKT cells are often referred to as invariant owing to their mostly invariant TCR α-chain usage (Vα14-Jα18 in mice, Vα24-Jα18 in humans). However, these cells have diverse TCR β-chains, including Vβ8, Vβ7, and Vβ2 in mice and Vβ11 in humans, joined to a range of TCR Dβ and Jβ genes. In this study, we demonstrate that TCR β-chain composition can dramatically influence lipid Ag recognition in an Ag-dependent manner. Namely, the glycolipids α-glucosylceramide and isoglobotrihexosylceramide were preferentially recognized by Vβ7+ NKT cells from mice, whereas the α-galactosylceramide analog OCH, with a truncated sphingosine chain, was preferentially recognized by Vβ8+ NKT cells from mice. We show that the influence of the TCR β-chain is due to a combination of Vβ-, Jβ-, and CDR3β-encoded residues and that these TCRs can recapitulate the selective Ag reactivity in TCR-transduced cell lines. Similar observations were made with human NKT cells where different CDR3β-encoded residues determined Ag preference. These findings indicate that NKT TCR β-chain diversity results in differential and nonhierarchical Ag recognition by these cells, which implies that some Ags can preferentially activate type I NKT cell subsets.

Introduction

Natural killer T cells are CD1d-restricted, lipid Ag–reactive, αβ T cells that are present in mice and humans. There are two broad classes of NKT cells: the most extensively studied are those classified as type I NKT cells, which mostly express an invariant TCR α-chain paired with a limited, but not invariant, array of TCR β-chains. In contrast, type II NKT cells, although also CD1d-restricted and lipid Ag reactive, express more diverse TCR α- and β-chains and appear to recognize different lipid Ags from those detected by type I NKT cells. This study is focused on type I NKT cells, which we simply refer to as NKT cells. In mice, NKT cells typically express an invariant Vα14-Jα18 TCR α-chain paired with either Vβ8, Vβ7, or Vβ2, whereas in humans, most of these cells use the orthologous Vα24-Jα18 TCR α-chain paired with Vβ11 (1). However, in both mice and humans, TCR β-chain diversity is generated through the use of variable TCR Dβ and Jβ genes and non–germline-encoded nucleotide additions during TCR β-chain gene rearrangement, thus resulting in a high level of CDR3β loop diversity and a polyclonal repertoire in both species (2–5). Furthermore, despite their limited TCR diversity, NKT cells recognize a diverse array of exogenous and endogenous CD1d-restricted ligands, which coincides with the participation of these cells in a range of different diseases (6, 7).

Whether the TCR β-chain can influence lipid Ag specificity of NKT cells has been a controversial and long-standing question. Earlier studies provided evidence that in mice, Vβ7+ NKT cells were selectively expanded or selected in response to self-lipid Ags, such as isoglobotrihexosylceramide (iGb3) and α-glucosylceramide (α-GlcCer) presented by CD1d (8–11), whereas others had suggested that the TCR β-chain primarily modulated the affinity of TCR binding to CD1d, independently of the lipid Ag that was presented by CD1d (12–14). Florence et al. (15) reported that different CDR3β loops modified through alanine substitution at different residues in the context of a Vβ8+ NKT TCR differentially influenced glycolipid Ag recognition in the context of CD1d. Structural studies examining NKT TCR interactions have indicated that the TCR β-chain binds to CD1d but does not come into direct contact with the Ag (6, 7). However, the TCR β-chain can indirectly impact on the binding affinity through differential interactions with CD1d (16, 17) and, furthermore, via modulating the TCR α-chain–Ag interactions (18). This raises the possibility that the NKT cell TCR β-chain can contribute to Ag specificity of these cells.

In this study, data are provided that support an important role for the TCR β-chain involving both germline-encoded (Vβ and Jβ) and non–germline-encoded residues working in a complementary manner in determining lipid Ag specificity of NKT cells. This suggests that the TCR β-chain can influence the recognition of particular glycolipid Ags presented by CD1d.

Materials and Methods

Glycolipids

The glycolipid Ags used in this study were either commercially purchased [α-galactosylceramide (α-GalCer; C26) (19), OCH (20), and iGb3 and Gb3 (21) from Enzo Life Sciences; β-GalCer (C12) (22) from Avanti Polar Lipids] or synthesized as previously described [α-GalCer (C20:2) (23), α-GlcCer (C20:2) (24), α-GlcCer (C26) (19), α-l-fucosylceramide (α-l-FucCer) (C26) (25), diglucosyldiacylglycerol (DGlcDAG; β-d-Glc-(1→6)-β-d-Glc-(1→3)-diacylglycerol) (26), α-GalCer (C24:1)/PBS-44 (27), α-glucuronosylceramide (GSL-1) and α-galacturonosylceramide (GSL-1′) (28), and α-glucuronosyl diacylglycerol (α-GlcA-DAG) (29)]. Glycolipid stocks were prepared using Tween 20–based (0.5% [v/v] Tween 20, sucrose [56 mg/ml], and l-histidine [7.5 mg/ml] in PBS) or tyloxapol-based (0.5% [v/v] tyloxapol in TBS) vehicle reagents.

Mice

C57BL/6 mice were bred in-house. C57BL/6 Tcra-Jtm1Tg (referred to as Jα18−/−) mice were originally obtained from M. Taniguchi (RIKEN Institute, Saitama, Japan) and backcrossed at least 10 times. All mice used in this study were derived from the Department of Microbiology and Immunology Animal House, University of Melbourne, and all procedures were approved by the University of Melbourne, Biochemistry and Molecular Biology, Dental Science, Medicine (Royal Melbourne Hospital), Microbiology and Immunology, and Surgery (Royal Melbourne Hospital) Animal Ethics Committee.

CD1d tetramers

Mouse and human CD1d tetramers were produced using a baculovirus expression system, similar to that previously described (30). Glycolipids were incubated with purified CD1d-biotin at 6:1 or 30:1 (lipid/protein) molar ratios at 37°C overnight. When required, CD1d-biotin stock was diluted with 10 mM Tris (pH 8.0) prior to the addition of glycolipid to ensure the concentration of vehicle reagent did not exceed 33% (v/v). Streptavidin-PE (BD Biosciences), streptavidin-allophycocyanin (BD Biosciences), or streptavidin–Brilliant Violet 421 (BioLegend) was used to generate tetramers. iGb3-loaded CD1d tetramer staining required iGb3 stock to be prepared in tyloxapol-based vehicle reagent.

Single-cell TCR sequencing

RNA was extracted from sort-purified mouse and human NKT cells, and cDNA was synthesized using a Sensiscript reverse transcription kit (Qiagen) (31) or SuperScript VILO cDNA synthesis kit (Invitrogen), respectively. Transcribed cDNA was amplified by two rounds of nested PCR amplification, as previously described (31, 32). PCR product was purified using a QIAquick PCR purification kit (Qiagen). Dye terminator sequencing reaction was performed using BigDye Terminator v3.1 cycle sequencing kit (Applied Biosystems) prior to sequencing electrophoresis (Molecular Diagnostics).

Generation of TCR-transduced NKT cell lines

Individual NKT cell TCR constructs containing TCRα and TCRβ genes separated by a 2A linker were synthesized (GenScript) and inserted into pMIG2 vectors essentially as previously described (33). Human embryonic kidney (293T) cells were transfected with 4 μg TCR expression vector, 2 μg pMIG expression vector containing CD3 sequence (a gift from S. Turner, University of Melbourne), 4 μg packaging vector pEQ-Pam3(-E), and 2 μg packaging vector pVSV-G using FuGENE 6 transfection reagent (Promega). Retrovirus containing supernatant was collected twice daily and used to transduce a TCR-deficient mouse thymoma cell line (BW58) in the presence of Polybrene (Sigma-Aldrich) for a total of 4 d.

In vitro proliferation assay

Splenocytes from Jα18−/− mice were pulsed overnight with iGb3 or α-GalCer in 48-well flat-bottom plates (3 × 106 cells/well), harvested, and washed with fresh media to remove excess glycolipid. Wild-type thymocytes were enriched for NKT cells by complement depletion of anti–CD24- and anti–CD8-labeled cells. Dead cells were removed via Histopaque-1083 (Sigma-Aldrich) density gradient centrifugation. Thymocytes were then labeled with either 2 μM CFSE (Molecular Probes) or 1 μM CellTrace Violet (Invitrogen) and incubated for 10–15 min at 37°C. Splenocytes (3 × 105/well) and thymocytes (5 × 104/well) were cocultured for 72 h in 96-well round-bottom plates.

Cell surface staining

In vitro–stimulated cells were stained with 7-aminoactinomycin D (Sigma-Aldrich) for exclusion of nonviable cells, anti–CD45R/B220-allophycocyanin-Cy7 (BD Biosciences, RA3-6B2), α-GalCer (PBS-44)–loaded CD1d tetramer (PE), and anti–TCRβ-allophycocyanin (BD Biosciences, H57-597). For Vβ analysis, anti-TCRβ was excluded from the Ab mixture and replaced with anti-CD3 (145-2C11). Subsequently, cells were stained using a panel of anti–Vβ-FITC (BD Biosciences). Proliferative responses were assessed using a BD LSR II, LSRFortessa, or BD FACSCanto II flow cytometer and FlowJo (Tree Star) software. TCR-transduced NKT cell lines were stained with anti–CD3-Pacific Blue (eBioscience, 17A2) or anti–CD3-Brilliant Violet 421 (BioLegend, 17A2) and anti–TCRβ-allophycocyanin (BD Biosciences, H57-597) at 4°C for 30 min, followed by glycolipid-loaded CD1d tetramer (PE) at room temperature for 30 min. CD1d tetramer staining experiments of retrovirally transduced BW58 cell lines were performed using a BD LSRFortessa. Mouse CD1d tetramer costaining experiments were performed on NKT cell–enriched thymocytes simultaneously incubated with saturating concentrations of α-GalCer (analogs)–loaded CD1d tetramer (allophycocyanin or Brilliant Violet 421) and α-GlcCer (analogs)–loaded CD1d tetramer (PE) at 37°C for 30 min. For Vβ analysis, cells were then chilled on ice for 5 min and stained with anti–Vβ-FITC at 4°C for 30 min. Human CD1d tetramer costaining experiments were performed on in vitro–expanded NKT cells simultaneously incubated with saturating concentrations of α-GalCer (analogs)–loaded CD1d tetramer (Brilliant Violet 421) and α-GlcCer (analogs)–loaded CD1d tetramer (PE) at 37°C for 30 min. CD1d tetramer costaining experiments were performed using a BD LSRFortessa.

Human NKT cell enrichment and expansion

Blood samples from healthy volunteers were obtained from the Australian Red Cross Blood Service, with ethics approval from both Red Cross (approval no. 11-05VIC-12) and University of Melbourne Human Research and Ethics Committee (approval no. 1035100). PBMCs were isolated and NKT enriched/expanded as previously described (34). Briefly, CD1d–α-GalCer tetramer+ NKT cells were enriched using anti-PE magnetic beads (Miltenyi Biotec) and cell sorting, followed by stimulation in vitro with anti-CD3/anti–CD28/PHA/IL-2/IL-7 for 48 h, followed by maintenance with IL-2 and IL-7 for 14–21 d.

Results

iGb3 recognition is restricted to a subset of NKT cells

iGb3 is a mammalian glycolipid and when presented by CD1d acts as a weak agonist for NKT cells in mice (9, 21). Some studies have suggested that Vβ7+ NKT cells are more responsive to iGb3 than are Vβ7− NKT cells (8, 9), although this is not always observed (12). We investigated this further by labeling mouse NKT cell–enriched thymocytes with a fluorescent dye (CellTrace Violet) to track cell division, stimulating these cells in vitro in the presence of spleen cells from NKT cell–deficient Jα18−/− mice with either iGb3 (10 μg/ml) or α-GalCer (100 ng/ml) (Fig. 1A), and comparing the proportions of Vβ2, Vβ7, Vβ8.1/8.2, and Vβ8.3+ NKT cells among the proliferating fractions with those from unstimulated (preculture, day 0) NKT cells (Fig. 1B, 1C). Most notably, the percentage of Vβ7+ NKT cells was consistently increased in the iGb3-responsive fraction and decreased in the α-GalCer–responsive fraction, compared with preculture NKT cells (Fig. 1B, 1C). Corresponding to the increase in Vβ7+ NKT cells in the iGb3-responsive fraction, there was also a significant reduction in Vβ2+ cells.

FIGURE 1.
FIGURE 1.

Vβ usage biases Ag recognition. (A) Thymocytes (C57BL/6) enriched for NKT cells were labeled with CellTrace Violet and cocultured with Jα18−/− (C57BL/6) splenocytes that had been pulsed overnight with iGb3 (10 μg/ml) or α-GalCer (100 ng/ml) and washed with fresh media to remove excess glycolipid. Plots indicate the percentage of NKT cells among 7-aminoactinomycin D, B220− lymphocytes following 3 d in vitro culture (top). Histograms represent the dilution of CellTrace Violet of gated NKT cells. Numbers on plots indicate the percentages of proliferating NKT cells (middle) and the estimated precursor frequency that has divided as measured by FlowJo proliferation analysis software (bottom). Data are representative of 12 similar experiments. (B) Representative plots indicating the percentages of Vβ+ CD1d–α-GalCer tetramer+ NKT cells prior to culture (top) and among proliferating NKT cells following exposure to either iGb3- (middle) or α-GalCer–pulsed (bottom) splenocytes. (C) Graph depicts the percentage of the indicated Vβ subset among proliferating NKT cells relative to the naive population (mean ± SEM). Data are pooled from three independent experiments. Statistical significance was assessed via a two-way ANOVA, followed by Bonferroni posttests comparing naive to iGb3 and α-GalCer. *p < 0.05, ***p < 0.001. (D) Plots represent the dilution of CellTrace Violet among the indicated Vβ+ NKT cell subsets. Data are representative of at least three independent experiments. (E) Graph depicts the percentage of Vβ7+ NKT cells within the nondivided or most divided fractions (mean ± SEM) as gauged by CellTrace Violet or CFSE dilution. Data are pooled from seven independent experiments. Statistical significance was assessed by a Friedman test followed by a Dunn posttest comparison. *p < 0.05, ***p < 0.0005.

Most NKT cells in the cultures were proliferating following 3 d in vitro culture with either α-GalCer or iGb3 Ags; however, a larger portion of the population remained undivided, even when exposed to a 100-fold higher concentration of iGb3 (Fig. 1A). Proliferation analysis software, which accounts for the exponential expansion of dividing cells, suggested that only a minor subset of the starting population of NKT cells was responding to iGb3 in these cultures, with division estimates ranging from 15 to 30% of the population. Analysis of the proliferative responses of individual Vβ subsets indicated that a larger percentage of Vβ7+ population responded to iGb3 compared with Vβ2+, Vβ8.1/8.2+, and Vβ8.3+ cells (Fig. 1D, 1E). This is consistent with the relative expansion of Vβ7+ NKT cells when cultured with iGb3 in the present study and in other studies (8, 9). Collectively, these data suggest that iGb3 is capable of activating a subset of NKT cells and that a higher proportion of Vβ7+ NKT cells are iGb3-responsive compared with NKT cells expressing other Vβ-chains.

To directly test whether some NKT cells are simply unable to respond to iGb3, CFSE-labeled thymocytes enriched for NKT cells were stimulated in vitro for 3 d with 10 μg/ml iGb3 (Fig. 2A), and undivided thymocytes (including iGb3-unresponsive NKT cells) were purified by flow cytometry, based on a lack of CFSE dilution to exclude divided cells, as well as the exclusion of large cells to remove potentially activated cells that were yet to proliferate (Fig. 2B). CD1d tetramer was not used to purify NKT cells in this experiment, as this may have activated the cells. The FACS-sorted undivided cells were then restimulated with iGb3 (10 μg/ml), α-GalCer (100 ng/ml), or no glycolipid. After 2–3 more days, the vast majority of NKT cells in these cultures had divided following exposure to α-GalCer, indicating that iGb3-unresponsive cells from the initial culture were competent to proliferate to an appropriate antigenic stimulus (Fig. 2C); however, following reculture with iGb3, very little Ag-induced NKT cell proliferation was evident (Fig. 2C). These data suggest that the recognition of the mammalian Ag iGb3 is restricted to a subset of the NKT cell population, whereas iGb3-unresponsive cells are still capable of responding to more broadly activating Ags such as α-GalCer.

FIGURE 2.
FIGURE 2.

iGb3 recognition is restricted to a subset of NKT cells. (A) CFSE-labeled thymocytes enriched for NKT cells were cocultured with Jα18−/− splenocytes pulsed with iGb3 (10 μg/ml). (B) CFSEhi and forward scatter area (FSC-A)lo thymocytes were sort purified. (C) Sort-purified cells were then cocultured with Jα18−/− splenocytes pulsed with iGb3 (10 μg/ml), α-GalCer (100 ng/ml), or no glycolipid. Data are representative of two similar experiments.

TCR Jβ usage influences Ag recognition

Despite the evidence that only a subset of NKT cells appeared to respond to iGb3, TCR Vβ usage alone was unlikely to account for the restricted recognition because NKT cells from each of the major Vβ subsets showed differential responses to iGb3 (Vβ7 > Vβ8 > Vβ2) (Fig. 1D). Another possibility was that iGb3 responsiveness was partly mediated by the hypervariable CDR3β loop. To examine this, CFSE-labeled thymocytes enriched for NKT cells were stimulated for 3 d with 100 ng/ml iGb3, which is typically a suboptimal dose (Supplemental Fig. 1), or the maximum dose of 10 μg/ml. NKT cells were then single cell sorted based on CFSE staining intensity to separate divided and nondivided cells and the expression of either Vβ7 or Vβ8.1/8.2. Isolated cells were then examined using PCR amplification with nested primers specific for either Vβ7 or Vβ8.1/8.2 and the TCRβ C region (31, 32, 35). Using this approach, >650 single NKT cell TCR β-chains were examined. The iGb3-responsive NKT cells were predominantly polyclonal, with 247 of the 336 TCR β-chain sequences being unique at the nucleotide level. In the nondivided fraction, 312 of the 338 sequences were unique, highlighting the diversity of the NKT cell TCRβ repertoire (3). Although analysis of CDR3β sequences failed to reveal any obvious trends that could explain the differential reactivity to iGb3 (Supplemental Fig. 2), the distribution of Jβ sequence usage was heavily biased between divided and nondivided fractions (Fig. 3). Furthermore, trends in Jβ distribution were distinct between Vβ7- and Vβ8.1 and Vβ8.2-expressing cells. Thus, Jβ1.2+ cells were overrepresented among divided Vβ7+ cells but not among the Vβ8.1/8.2+ divided cells. In contrast, Jβ2.7+ cells were overrepresented in the Vβ8.1/8.2+ divided cells but not the Vβ7+ divided cells. Other Jβ gene segments appeared to be inhibitory for iGb3 recognition. For example, no Jβ2.3 or Jβ2.4 gene segments were identified in the Vβ8.1/8.2+ divided fractions, yet they were clearly represented among the Vβ8.1/8.2+ nondivided cells. In contrast, no clear difference in Jβ2.3 and Jβ2.4 expression was observed in the Vβ7+ divided cells compared with Vβ7+ nondivided cells. Taken together, these data suggested that recognition of iGb3 by NKT cells is influenced by a combination of TCR Vβ, Jβ, and CDR3β regions.

FIGURE 3.
FIGURE 3.

TCR Jβ usage influences Ag recognition. Thymocytes enriched for NKT cells were cocultured for 3 d with Jα18−/− splenocytes pulsed with either 100 ng/ml or 10 μg/ml iGb3. Cells were then single-cell sorted based on α-GalCer–loaded CD1d tetramer reactivity, CFSE intensity to differentiate the most divided and nondivided NKT cells, and the expression of either Vβ7 or Vβ8.1/8.2. cDNA from sorted cells was then PCR amplified and sequenced to determine TCR β-chain gene usage. Graphs depict the percentages of Jβ gene segment usage among unique nucleotide sequences derived from within either the divided or nondivided fractions. The 100 ng/ml data are from a single experiment, and 10 μg/ml data are pooled from two independent experiments.

NKT cell TCR β-chain composition confers differential Ag recognition

To establish whether NKT TCR β-chain composition is directly responsible for iGb3 recognition, a selection of TCRs derived from NKT cells that were either responsive of nonresponsive to iGb3 were used to generate a panel of TCR-transduced cell lines. To examine the role of divergent CDR3β sequence usage in the context of the same Vβ-Jβ paring, two Vβ8.2-Jβ2.7–responsive (Vβ8.2-Jβ2.7r1 and Vβ8.2-Jβ2.7r2) and two Vβ8.2-Jβ2.7–nonresponsive sequences (Vβ8.2-Jβ2.7n1 and Vβ8.2-Jβ2.7n2) were selected, alongside one responsive Vβ7-Jβ1.2 (Vβ7-Jβ1.2r) and one nonresponsive Vβ7-Jβ1.2 sequence (Vβ7-Jβ1.2n) (Fig. 4A). Given that Vβ8+ cells that expressed Jβ2.4 were unable to respond to iGb3 (Fig. 3), two of these sequences were also selected from the iGb3 nonresponsive cells (Vβ8.2-Jβ2.4n1 and Vβ8.2-Jβ2.4n2). Each of these TCR β-chains was coupled with the invariant Vα14-Jα18 TCR α-chain prior to retroviral transduction. With the intention of isolating TCR sequences with highest responsiveness to iGb3, all responsive sequences were selected from NKT cells that proliferated to the lower dose of iGb3 (100 ng/ml) with the exception of one iGb3-responsive clone that used a Vβ8.2-Jβ2.5 pairing (Vβ8.2-Jβ2.5r), chosen from a 10 μg/ml iGb3-stimulated culture because this was the largest clonal population identified among iGb3-responsive cells (data not shown). Conversely, to investigate TCRs that were nonresponsive to iGb3, sequences were taken from NKT cells that failed to proliferate following high-dose iGb3 exposure (10 μg/ml) (Fig. 4A).

FIGURE 4.
FIGURE 4.

The NKT cell TCR β-chain contributes to CD1d-iGb3 reactivity. (A) TCR β-chain sequences used for the retroviral transduction of BW58 cell lines (n, TCR sequences derived from iGb3 nonresponsive cells; r, TCR sequences derived from iGb3 responsive cells). (B) Representative plots of TCR-transduced cell lines stained with CD1d tetramers loaded with either iGb3, Gb3, α-GalCer (PBS-44), or vehicle alone. Plots represent the intensity of CD1d tetramer staining (black) relative to a non-agonist control CD1d tetramer (Gb3, gray). Cells have been gated on an equivalent level of TCR expression during analysis; numbers indicate the geometric mean fluorescence intensity (GMFI) of each CD1d tetramer. (C) Representative plots demonstrating the relationship between TCR expression levels and iGb3- or Gb3-loaded CD1d tetramer staining intensity of cell lines transduced with the invariant NKT TCR α-chain and TCR β-chain sequences derived from an iGb3-responsive and three iGb3-nonresponsive NKT cells.

Cell lines were subsequently tested with CD1d tetramers loaded with either iGb3, Gb3, α-GalCer, or vehicle alone. CD1d tetramer staining intensity was measured on cells gated for a comparable level of TCR expression by each cell line. As predicted, iGb3-responsive TCRs (Vβ7-Jβ1.2r, Vβ8.2-Jβ2.5r, Vβ8.2-Jβ2.7r1, and Vβ8.2-Jβ2.7r2) conferred superior binding of iGb3-loaded CD1d tetramer (Fig. 4B). Interestingly no such pattern was observed when these cell lines were stained with α-GalCer–loaded CD1d tetramer, and Gb3-loaded tetramer failed to stain any of these cell lines. These data confirmed that both Vβ-Jβ pairing and non–germline-encoded variation within the CDR3β region can facilitate iGb3 recognition by NKT cells, demonstrating a context-dependent role for the hypervariable CDR3β region. Interestingly, some cell lines expressing TCRs from nonresponsive NKT cell clones, including Vβ8.2-Jβ2.7n2 and Vβ7-Jβ1.2n, exhibited weak reactivity to iGb3-loaded CD1d tetramer compared with the control Gb3-loaded tetramer, but this required a higher level of TCR expression relative to iGb3-responsive cell lines. This is demonstrated in Fig. 4C comparing Vβ8.2-Jβ2.7r2 staining with iGb3-loaded or Gb3-loaded CD1d tetramer versus TCRβ (without the narrow TCRβ gate that was used in Fig. 4B), relative to the nonresponsive cell lines Vβ8.2-Jβ2.7n2, Vβ7-Jβ1.2n, and Vβ8.2-Jβ2.4n1. These data suggested that, in some instances, reactivity to CD1d-iGb3 by NKT TCRs could be influenced by both TCR β-chain composition and affinity, as well as the level of TCR expression, and thus avidity.

Given that these NKT cell lines displayed differential recognition of iGb3 and α-GalCer, the reactivity of other CD1d-restricted glycolipid Ags was also investigated. These included mammalian β-GalCer analog C12 [β-GalCer (C12)] (22, 36), tumor-associated Ag α-l-FucCer (C26) (25, 37), bacterial glycolipids α-GlcA-DAG from Mycobacterium smegmatis (29), GSL-1 and GSL-1′ from Sphingomonas spp. (28, 38), and DGlcDAG from Staphylococcus aureus. Although iGb3 and GSL-1–loaded tetramers shared similar trends among the nine NKT cell lines, ligands such as α-l-FucCer, DGlcDAG, α-GlcA-DAG, GSL-1′, and β-GalCer (C12) displayed CD1d tetramer staining patterns that were distinct from both iGb3 and α-GalCer (Fig. 5). The influence of different CDR3β sequence usage was highlighted by comparing cell lines with the same Vβ-Jβ pairings. For example, the Vβ8.2-Jβ2.7n2 line displayed the highest intensity staining for α-GalCer, whereas the Vβ8.2-Jβ2.7r2 line stained brightest for α-GlcA-DAG, DGlcDAG, and GSL-1′, and Vβ8.2-Jβ2.7r1 conferred the brightest staining with α-l-FucCer (Fig. 5). The fact that Ag-dependent differences were observed among such a limited selection of TCR sequences suggests that differential antigenic recognition among the NKT cell TCR repertoire may extend to a wider array of glycolipid Ags than previously appreciated. Consistent with an important role for TCRβ diversity in glycolipid Ag recognition, CD1d tetramer staining on thymic NKT cells with these Ags revealed differential staining (Supplemental Fig. 3), suggesting that only those expressing particular TCR β-chains had sufficiently high avidity to be clearly labeled. Importantly, CD1d tetramers loaded with lower affinity NKT cell agonists, including iGb3, demonstrated limited staining on thymic NKT cells. This suggested that low-affinity interactions with CD1d-Ag, combined with the relatively low levels of TCR expressed by NKT cells ex vivo, may not provide sufficient avidity to achieve clear tetramer staining, even when CD1d is loaded with glycolipids that are demonstrably antigenic.

FIGURE 5.
FIGURE 5.

The NKT cell TCR β-chain diversity confers differential Ag recognition. Graphs depict the ratio of glycolipid-loaded CD1d tetramer staining intensity (geometric mean fluorescence intensity) relative to a non-agonist control (Gb3-loaded CD1d tetramer) for each cell line and a TCR negative control (mean ± SEM). Data are pooled from 3 to 12 independent experiments.

Distinct TCR Vβ+ NKT cell subsets display divergent antigenic preferences

To further investigate the correlation between β-chain diversity and differential Ag recognition, NKT cell–enriched mouse thymocytes were simultaneously costained with saturating concentrations of CD1d tetramers loaded with different glycolipid Ags, a technique we have previously demonstrated to segregate NKT cell subsets with divergent antigenic preferences (29). The competitive nature of CD1d tetramer binding in this context resulted in some NKT cells being preferentially stained with one or the other CD1d tetramer. In particular, when costaining NKT cells with the α-GalCer analog OCH (20) and α-GlcCer (including either C26 or C20:2 variants), NKT cells were segregated into at least three major subsets (OCHhi/α-GlcCerlo-hi, OCHlo-hi/α-GlcCerint-hi, and OCHlo/α-GlcCerint-hi) (Fig. 6A) that were strikingly correlated with Vβ usage (Fig. 6B). Vβ8+ NKT cells displayed a bias toward CD1d-OCH tetramer, compared with Vβ7+ and Vβ2+ NKT cells, which were biased toward CD1d–α-GlcCer tetramer (Fig. 6B). The antigenic preferences of the minor Vβ6+, Vβ9+, Vβ10b+, and Vβ14+ NKT cell subsets (3, 39) were less obvious, but there did appear to be some variation; for example, Vβ14+ NKT cells were preferentially bound by CD1d–α-GlcCer tetramer, whereas Vβ9+ NKT cells displayed a bias toward CD1d-OCH tetramer staining (Fig. 6B). The observation that for each TCR Vβ, tetramer staining ranged from single-positive to double-positive and for some cells showed the opposite bias to the majority, supports the concept that although Vβ usage influences Ag recognition by NKT cells, the additional diversity contributed by non–germline- and Jβ-encoded residues are also important factors.

FIGURE 6.
FIGURE 6.

TCR β-chain diversity influences Ag recognition. (A) NKT cell–enriched thymocytes were simultaneously costained with saturating concentrations of the indicated α-GalCer–loaded CD1d tetramers (allophycocyanin) and α-GlcCer–loaded CD1d tetramers (PE) (left and middle panels), α-GalCer (C26)–loaded CD1d tetramer alone, or vehicle-loaded CD1d tetramer controls for each fluorochrome (right panel). Data are representative of at least three independent experiments. (B) CD1d-OCH and CD1d–α-GlcCer tetramer colabeling of NKT cell–enriched thymocytes. Black dots depict thymocytes expressing the indicated Vβ region, whereas the gray dots represent remaining cells. Data are representative of two to six independent experiments.

Vα24+Vβ11+ human NKT cells display divergent antigenic preferences

The degree of Vβ-chain diversity among Vα24+Jα18+ human NKT cells is less extensive than their mouse counterparts, being largely restricted to Vβ11, the human ortholog of mouse Vβ8.2 (4, 40, 41). However, similar to their mouse counterparts, human NKT cells also exhibit extensive Jβ and non–germline-encoded diversity at the CDR3β junction. Given that Ag-dependent differences were observed among mouse NKT cell TCRs expressing the same Vβ gene segment (Figs. 4, 5), this raised the possibility that differential Ag recognition may also be present among human NKT cells. To investigate this, in vitro–expanded human NKT cells (sort purified from PBMCs, expanded in vitro with anti-CD3/anti–CD28/PHA/IL-2/IL-7) from two donors were costained with human CD1d tetramers loaded with several α-GalCer and α-GlcCer variants. Although cells from both donors were >98% Vα24+Vβ11+ (Fig. 7A), CD1d tetramer costaining segregated these cells into subsets, revealing differences in their antigenic preference (Fig. 7B). These data suggest that, as in mice, TCR β-chain diversity has the potential to differentially influence CD1d-Ag recognition in humans. Furthermore, the observation that CD1d tetramer costaining profiles were distinct between donors suggests that the random nature of NKT cell TCR gene rearrangement could potentially generate Vα24+Jα18+ repertoires with divergent recognition characteristics in different individuals.

FIGURE 7.
FIGURE 7.

TCR β-chain diversity influences Ag recognition. (A) CD1d–α-GalCer tetramer+ cells were sort purified from healthy donor PBMCs and expanded in vitro. Plots represent the percentages of donor cells coexpressing Vα24 and Vβ11, and the percentages of CD1d–α-GalCer tetramer+ cells. (B) Sort-purified, in vitro–expanded human NKT cells were then simultaneously costained with saturating concentrations of CD1d tetramers loaded with the indicated α-GalCer and α-GlcCer analogs. (C) Plots represent the segregation of NKT cells based on CD1d tetramer costaining used for single-cell sorting, and tables show unique amino acid sequences derived from each single cell–sorted NKT cell population.

To explore potential correlations between TCR β-chain usage and preferential Ag recognition, the three most prominent subpopulations from each donor were single-cell sorted (Fig. 7C) for TCR sequence analysis. The majority of donor 1’s NKT cells (population 1A) stained with similar intensity for α-GalCer– and α-GlcCer–loaded CD1d tetramers. From 22 derived sequences, five clones were identified expressing unique TCR β-chain sequences. Interestingly, two of these five clones had atypical CDR3α usages, where the Vα24-encoded Ser residue was substituted with either a non–germline-encoded Thr or Phe at the Vα24 and Jα18 junction. Population 1B exhibited a reduction in CD1d–α-GlcCer, but not CD1d–α-GalCer staining intensity compared with most cells from this donor. Only one sequence was derived from population 1B, which contained a Vβ11-Jβ1.1 gene pairing not observed among other NKT cells from this donor. In contrast, population 1C displayed similar staining intensity for CD1d–α-GlcCer as 1A; however, the intensity of α-GalCer–loaded CD1d tetramer was reduced. NKT cells from 1C appeared to be clonal, with 22 of 22 TCR α- and β-chain sequences being identical. This TCR contained a unique CDR3α region, due to the deletion of the Vα24-encoded Ser and the addition of a non–germline-encoded Asn residue at the Vα24-Jα18 junction. This TCR was composed of a Vβ11-Jβ1.6 gene combination, as did one clone from 1A, with the only differences being deletions of germline and additions of non–germline-encoded residues in the CDR3α and CDR3β loops. Therefore, differences in CD1d tetramer costaining could not be specifically attributed to either TCR chain, but this does demonstrate the potential for random sequence variation to have Ag-dependent effects. The three subpopulations from donor 2 appeared more polyclonal, with 16 of the 48 cells sequenced carrying unique TCRs. The TCR α-chain sequences were almost entirely invariant, with only one clone containing a non-germline CDR3α residue, namely a Thr (Fig. 7C), indicating that differences in CD1d tetramer costaining were likely a result of TCR β-chain diversity. No particular patterns stood out for TCR Jβ-chain usage or CDR3β composition in either donor 1 or donor 2, other than that the TCR Jβ distribution was largely distinct between different subpopulations (Fig. 7C). This is in line with the findings from mouse NKT cells that Jβ usage contributes to Ag-specific biases (Figs. 3–5). These data suggest that despite limited diversity among human Vα24+Vβ11+ NKT cells, there remains scope for TCR β-chain composition to impact CD1d recognition in an Ag-dependent manner.

Discussion

TCR interactions with peptides presented by the MHC often involve the hypervariable CDR3 loops (reviewed in Ref. 7). In contrast, the NKT TCR interacts with lipid Ags solely via the invariant TCR α-chain (6, 18, 42), suggesting an innate-like characteristic of this receptor, a concept supported by mutant studies of the NKT TCR (14, 18, 43). However the NKT TCR β-chain is diverse with our sequence analysis reiterating the hypervariable nature of the CDR3β repertoire (2–5), which encompasses considerable non–germline-encoded variation. Divergent CDR3β sequence usage has been reported to modulate affinity for CD1d in an Ag-independent manner (12–14), consistent with previous structural reports that when the CDR3β loop is involved in these interactions, it solely contacts CD1d (reviewed in Ref. 6). Surprisingly, our study has demonstrated that the composition of the CDR3β region modulates affinity for CD1d in an Ag-dependent manner, suggesting that there may be an immunological consequence of non–germline-encoded sequence variation within the NKT cell TCR β-chain beyond simply modulating the extent of autoreactivity (13).

The results in this study support the correlation between Vβ usage and the preferential recognition of CD1d-restricted ligands such as iGb3, α-GlcCer, and OCH (8–10), despite reports of the CDR1β and CDR2β loops not contacting Ag (reviewed in Ref. 6). Importantly, the observed Vβ biases were dependent on the Jβ/CDR3β composition, indicating that both germline and non-germline components of the TCR are important, working in a complementary manner to modulate recognition of different glycolipid Ags. These findings provide a rationale for the differential responses to α-GalCer and iGb3 observed among a panel of Vβ8.2+ NKT cell hybridomas expressing diverse Jβ-CDR3β usages (15). However, the presence of differential Ag recognition in the context of the same Vβ-Jβ pairings observed in the present study highlights the potential for non–germline-encoded residues within the CDR3β loop to have an Ag-dependent influence. This implies that differential antigenic recognition may not always manifest in clear Vβ- or Jβ-specific biases. A reason why previous studies did not detect preferential recognition of iGb3 by Vβ7+ NKT cells may have been due to the use of a single Jβ/CDR3β loop, derived from a non–NKT cell line that in combination with Vβ7 did not favor preferential iGb3/CD1d binding (12). In contrast, other studies that have demonstrated Ag-specific TCR Vβ biases have used naturally occurring, polyclonal NKT TCR repertoires (8–10, 15). Our data suggests that the extensive natural variation that exists within the NKT cell TCR β-chain repertoire needs to be considered when investigating the recognition of different CD1d-restricted ligands.

Previous studies have highlighted the importance of the CDR3β loop in modulating CD1d–glycolipid Ag binding, where different CDR3β loops engineered into NKT TCRs can either abrogate or enhance binding (12, 14, 44). Moreover, prior investigations have suggested that trends within CDR3β sequence composition could be directly correlated with the recognition of self-ligands (16); however, our findings suggest a more collaborative role for the individual TCR β-chain components (Vβ, Jβ, and CDR3β) in determining specificity for individual glycolipid Ags. Although the observed Ag-specific influence of the TCR β-chain may seem somewhat counterintuitive considering the absence of direct CDRβ interactions with glycolipid Ag (reviewed in Ref. 6), altered amino acid composition within the NKT cell TCR β-chain not only has the potential to affect TCR contacts with CD1d-Ag, it also has the potential to impact the structure of the αβTCR heterodimer due to the molecular interplay between the α- and the β-chains (11, 18, 45). Indeed, structural reports have previously demonstrated that CDR3α loop-mediated contacts with glycolipid Ag may be altered when NKT cell TCRs with different β-chains are used (11, 18). Collectively, these data suggest that both germline and non–germline-encoded amino acid composition of the NKT cell TCR that do not directly contact glycolipid Ag may still have the potential to modulate Ag specificity.

The inability of most α-GalCer–reactive NKT cells to respond to the mammalian ligand iGb3 indicated that TCR β-chain diversity has the potential to limit activation to subsets of the NKT cell population. Analysis of TCR-transfected cell lines where individual NKT TCRs are expressed from low to very high levels suggested that this limited recognition was not solely the result of Ag specificity, rather, it may reflect a spectrum of avidity for CD1d-iGb3 throughout the population, mediated by both β-chain composition and TCR expression levels. Given that thymic NKT cells can vary considerably in their TCR expression levels (46), this may also be a relevant factor in intrathymic NKT cell selection. In contrast, regardless of TCR expression levels, Gb3-loaded CD1d tetramers failed to stain these TCR transfected cells, regardless of their TCR levels. The potential for the NKT cell population to maintain spectrums of avidity for CD1d-Ag has been demonstrated previously using CD1d tetramers loaded with various bacterial ligands (28, 47, 48), with similar observations being made in this investigation with regard to CD1d tetramers loaded with Ags other than α-GalCer. Therefore, the uniform reactivity of high-affinity synthetic Ags such as α-GalCer may obscure the variability of antigenic recognition that exists for lower affinity, physiologically relevant, CD1d-restricted ligands. These data provide rationale for the presence of context-dependent, immunodominant NKT cell subsets to exist as a function of TCR β-chain diversity.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

α-GalCer (C24:1)/PBS-44, GSL-1, and GSL-1′ were provided by Prof. P. Savage (Brigham Young University). The retroviral expression and packaging vectors were supplied by Dr. D. Vignali (St. Jude Children's Research Hospital). The CD3 expression vector was supplied by Prof. S. Turner (University of Melbourne). S. aureus pellets for production of DGlcDAG were supplied by Prof. S. Foster (University of Sheffield). We thank Dr. Sophie Doak for helpful advice with single-cell TCR sequencing.

Footnotes

  • This work was supported by National Health and Medical Research Council of Australia Grants 1013667, 1021972, 5671222, and 1046333 and by program and project grants from the Cancer Council Victoria. G.C. was supported by a Cancer Research Institute predoctoral fellowship. D.G.P. is supported by National Health and Medical Research Council of Australia Biomedical Fellowship 1054431; A.P.U. is supported by Australian Research Council Future Fellowship FT140100278; N.L.L.G. is supported by a Sylvia and Charles Viertel Foundation senior medical research fellowship; D.I.G. is supported by National Health and Medical Research Council of Australia Senior Principal Research Fellowship 1020770; and J.R. is supported by National Health and Medical Research Council of Australia Fellowship AF50.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    DGlcDAG
    diglucosyldiacylglycerol
    α-GalCer
    α-galactosylceramide
    α-GlcA-DAG
    α-glucuronosyl diacylglycerol
    α-GlcCer
    α-glucosylceramide
    GSL-1
    α-glucuronosylceramide
    GSL-1′
    α-galacturonosylceramide
    iGb3
    isoglobotrihexosylceramide
    α-l-FucCer
    α-l-fucosylceramide.

  • Received June 1, 2015.
  • Accepted September 3, 2015.

References

  1. ↵
    1. Godfrey D. I.,
    2. H. R. MacDonald,
    3. M. Kronenberg,
    4. M. J. Smyth,
    5. L. Van Kaer
    . 2004. NKT cells: what’s in a name? Nat. Rev. Immunol. 4: 231–237.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Lantz O.,
    2. A. Bendelac
    . 1994. An invariant T cell receptor alpha chain is used by a unique subset of major histocompatibility complex class I-specific CD4+ and CD4−8− T cells in mice and humans. J. Exp. Med. 180: 1097–1106.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Matsuda J. L.,
    2. L. Gapin,
    3. N. Fazilleau,
    4. K. Warren,
    5. O. V. Naidenko,
    6. M. Kronenberg
    . 2001. Natural killer T cells reactive to a single glycolipid exhibit a highly diverse T cell receptor β repertoire and small clone size. Proc. Natl. Acad. Sci. USA 98: 12636–12641.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Dellabona P.,
    2. E. Padovan,
    3. G. Casorati,
    4. M. Brockhaus,
    5. A. Lanzavecchia
    . 1994. An invariant Vα24-JαQ/Vβ11 T cell receptor is expressed in all individuals by clonally expanded CD4−8− T cells. J. Exp. Med. 180: 1171–1176.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Porcelli S.,
    2. C. E. Yockey,
    3. M. B. Brenner,
    4. S. P. Balk
    . 1993. Analysis of T cell antigen receptor (TCR) expression by human peripheral blood CD4−8− alpha/beta T cells demonstrates preferential use of several V beta genes and an invariant TCR alpha chain. J. Exp. Med. 178: 1–16.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Rossjohn J.,
    2. D. G. Pellicci,
    3. O. Patel,
    4. L. Gapin,
    5. D. I. Godfrey
    . 2012. Recognition of CD1d-restricted antigens by natural killer T cells. Nat. Rev. Immunol. 12: 845–857.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Rossjohn J.,
    2. S. Gras,
    3. J. J. Miles,
    4. S. J. Turner,
    5. D. I. Godfrey,
    6. J. McCluskey
    . 2015. T cell antigen receptor recognition of antigen-presenting molecules. Annu. Rev. Immunol. 33: 169–200.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Schümann J.,
    2. M. P. Mycko,
    3. P. Dellabona,
    4. G. Casorati,
    5. H. R. MacDonald
    . 2006. Cutting edge: influence of the TCR Vβ domain on the selection of semi-invariant NKT cells by endogenous ligands. J. Immunol. 176: 2064–2068.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Wei D. G.,
    2. S. A. Curran,
    3. P. B. Savage,
    4. L. Teyton,
    5. A. Bendelac
    . 2006. Mechanisms imposing the Vβ bias of Vα14 natural killer T cells and consequences for microbial glycolipid recognition. J. Exp. Med. 203: 1197–1207.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Wun K. S.,
    2. G. Cameron,
    3. O. Patel,
    4. S. S. Pang,
    5. D. G. Pellicci,
    6. L. C. Sullivan,
    7. S. Keshipeddy,
    8. M. H. Young,
    9. A. P. Uldrich,
    10. M. S. Thakur,
    11. et al
    . 2011. A molecular basis for the exquisite CD1d-restricted antigen specificity and functional responses of natural killer T cells. Immunity 34: 327–339.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Patel O.,
    2. D. G. Pellicci,
    3. A. P. Uldrich,
    4. L. C. Sullivan,
    5. M. Bhati,
    6. M. McKnight,
    7. S. K. Richardson,
    8. A. R. Howell,
    9. T. Mallevaey,
    10. J. Zhang,
    11. et al
    . 2011. Vβ2 natural killer T cell antigen receptor-mediated recognition of CD1d-glycolipid antigen. Proc. Natl. Acad. Sci. USA 108: 19007–19012.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Mallevaey T.,
    2. J. P. Scott-Browne,
    3. J. L. Matsuda,
    4. M. H. Young,
    5. D. G. Pellicci,
    6. O. Patel,
    7. M. Thakur,
    8. L. Kjer-Nielsen,
    9. S. K. Richardson,
    10. V. Cerundolo,
    11. et al
    . 2009. T cell receptor CDR2β and CDR3β loops collaborate functionally to shape the iNKT cell repertoire. Immunity 31: 60–71.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Matulis G.,
    2. J. P. Sanderson,
    3. N. M. Lissin,
    4. M. B. Asparuhova,
    5. G. R. Bommineni,
    6. D. Schümperli,
    7. R. R. Schmidt,
    8. P. M. Villiger,
    9. B. K. Jakobsen,
    10. S. D. Gadola
    . 2010. Innate-like control of human iNKT cell autoreactivity via the hypervariable CDR3β loop. PLoS Biol. 8: e1000402.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Scott-Browne J. P.,
    2. J. L. Matsuda,
    3. T. Mallevaey,
    4. J. White,
    5. N. A. Borg,
    6. J. McCluskey,
    7. J. Rossjohn,
    8. J. Kappler,
    9. P. Marrack,
    10. L. Gapin
    . 2007. Germline-encoded recognition of diverse glycolipids by natural killer T cells. Nat. Immunol. 8: 1105–1113.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Florence W. C.,
    2. C. Xia,
    3. L. E. Gordy,
    4. W. Chen,
    5. Y. Zhang,
    6. J. Scott-Browne,
    7. Y. Kinjo,
    8. K. O. Yu,
    9. S. Keshipeddy,
    10. D. G. Pellicci,
    11. et al
    . 2009. Adaptability of the semi-invariant natural killer T-cell receptor towards structurally diverse CD1d-restricted ligands. EMBO J. 28: 3579–3590.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Mallevaey T.,
    2. A. J. Clarke,
    3. J. P. Scott-Browne,
    4. M. H. Young,
    5. L. C. Roisman,
    6. D. G. Pellicci,
    7. O. Patel,
    8. J. P. Vivian,
    9. J. L. Matsuda,
    10. J. McCluskey,
    11. et al
    . 2011. A molecular basis for NKT cell recognition of CD1d-self-antigen. Immunity 34: 315–326.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Pellicci D. G.,
    2. A. J. Clarke,
    3. O. Patel,
    4. T. Mallevaey,
    5. T. Beddoe,
    6. J. Le Nours,
    7. A. P. Uldrich,
    8. J. McCluskey,
    9. G. S. Besra,
    10. S. A. Porcelli,
    11. et al
    . 2011. Recognition of β-linked self glycolipids mediated by natural killer T cell antigen receptors. Nat. Immunol. 12: 827–833.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Pellicci D. G.,
    2. O. Patel,
    3. L. Kjer-Nielsen,
    4. S. S. Pang,
    5. L. C. Sullivan,
    6. K. Kyparissoudis,
    7. A. G. Brooks,
    8. H. H. Reid,
    9. S. Gras,
    10. I. S. Lucet,
    11. et al
    . 2009. Differential recognition of CD1d-α-galactosyl ceramide by the Vβ8.2 and Vβ7 semi-invariant NKT T cell receptors. Immunity 31: 47–59.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Kawano T.,
    2. J. Cui,
    3. Y. Koezuka,
    4. I. Toura,
    5. Y. Kaneko,
    6. K. Motoki,
    7. H. Ueno,
    8. R. Nakagawa,
    9. H. Sato,
    10. E. Kondo,
    11. et al
    . 1997. CD1d-restricted and TCR-mediated activation of vα14 NKT cells by glycosylceramides. Science 278: 1626–1629.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Miyamoto K.,
    2. S. Miyake,
    3. T. Yamamura
    . 2001. A synthetic glycolipid prevents autoimmune encephalomyelitis by inducing TH2 bias of natural killer T cells. Nature 413: 531–534.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Zhou D.,
    2. J. Mattner,
    3. C. Cantu III.,
    4. N. Schrantz,
    5. N. Yin,
    6. Y. Gao,
    7. Y. Sagiv,
    8. K. Hudspeth,
    9. Y. P. Wu,
    10. T. Yamashita,
    11. et al
    . 2004. Lysosomal glycosphingolipid recognition by NKT cells. Science 306: 1786–1789.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Parekh V. V.,
    2. A. K. Singh,
    3. M. T. Wilson,
    4. D. Olivares-Villagómez,
    5. J. S. Bezbradica,
    6. H. Inazawa,
    7. H. Ehara,
    8. T. Sakai,
    9. I. Serizawa,
    10. L. Wu,
    11. et al
    . 2004. Quantitative and qualitative differences in the in vivo response of NKT cells to distinct α- and β-anomeric glycolipids. J. Immunol. 173: 3693–3706.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Yu K. O.,
    2. J. S. Im,
    3. A. Molano,
    4. Y. Dutronc,
    5. P. A. Illarionov,
    6. C. Forestier,
    7. N. Fujiwara,
    8. I. Arias,
    9. S. Miyake,
    10. T. Yamamura,
    11. et al
    . 2005. Modulation of CD1d-restricted NKT cell responses by using N-acyl variants of α-galactosylceramides. Proc. Natl. Acad. Sci. USA 102: 3383–3388.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Jervis P. J.,
    2. N. Veerapen,
    3. G. Bricard,
    4. L. R. Cox,
    5. S. A. Porcelli,
    6. G. S. Besra
    . 2010. Synthesis and biological activity of α-glucosyl C24:0 and C20:2 ceramides. Bioorg. Med. Chem. Lett. 20: 3475–3478.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Watanabe K.,
    2. T. Matsubara,
    3. S. Hakomori
    . 1976. α-l-Fucopyranosylceramide, a novel glycolipid accumulated in some of the human colon tumors. J. Biol. Chem. 251: 2385–2387.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Fedtke I.,
    2. D. Mader,
    3. T. Kohler,
    4. H. Moll,
    5. G. Nicholson,
    6. R. Biswas,
    7. K. Henseler,
    8. F. Götz,
    9. U. Zähringer,
    10. A. Peschel
    . 2007. A Staphylococcus aureus ypfP mutant with strongly reduced lipoteichoic acid (LTA) content: LTA governs bacterial surface properties and autolysin activity. Mol. Microbiol. 65: 1078–1091.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Freigang S.,
    2. E. Landais,
    3. V. Zadorozhny,
    4. L. Kain,
    5. K. Yoshida,
    6. Y. Liu,
    7. S. Deng,
    8. W. Palinski,
    9. P. B. Savage,
    10. A. Bendelac,
    11. L. Teyton
    . 2012. Scavenger receptors target glycolipids for natural killer T cell activation. J. Clin. Invest. 122: 3943–3954.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Kinjo Y.,
    2. D. Wu,
    3. G. Kim,
    4. G. W. Xing,
    5. M. A. Poles,
    6. D. D. Ho,
    7. M. Tsuji,
    8. K. Kawahara,
    9. C. H. Wong,
    10. M. Kronenberg
    . 2005. Recognition of bacterial glycosphingolipids by natural killer T cells. Nature 434: 520–525.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Uldrich A. P.,
    2. O. Patel,
    3. G. Cameron,
    4. D. G. Pellicci,
    5. E. B. Day,
    6. L. C. Sullivan,
    7. K. Kyparissoudis,
    8. L. Kjer-Nielsen,
    9. J. P. Vivian,
    10. B. Cao,
    11. et al
    . 2011. A semi-invariant Vα10+ T cell antigen receptor defines a population of natural killer T cells with distinct glycolipid antigen-recognition properties. Nat. Immunol. 12: 616–623.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Matsuda J. L.,
    2. O. V. Naidenko,
    3. L. Gapin,
    4. T. Nakayama,
    5. M. Taniguchi,
    6. C. R. Wang,
    7. Y. Koezuka,
    8. M. Kronenberg
    . 2000. Tracking the response of natural killer T cells to a glycolipid antigen using CD1d tetramers. J. Exp. Med. 192: 741–754.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Kedzierska K.,
    2. V. Venturi,
    3. K. Field,
    4. M. P. Davenport,
    5. S. J. Turner,
    6. P. C. Doherty
    . 2006. Early establishment of diverse T cell receptor profiles for influenza-specific CD8+CD62Lhi memory T cells. Proc. Natl. Acad. Sci. USA 103: 9184–9189.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Turner S. J.,
    2. G. Diaz,
    3. R. Cross,
    4. P. C. Doherty
    . 2003. Analysis of clonotype distribution and persistence for an influenza virus-specific CD8+ T cell response. Immunity 18: 549–559.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Holst J.,
    2. A. L. Szymczak-Workman,
    3. K. M. Vignali,
    4. A. R. Burton,
    5. C. J. Workman,
    6. D. A. Vignali
    . 2006. Generation of T-cell receptor retrogenic mice. Nat. Protoc. 1: 406–417.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Uldrich A. P.,
    2. J. Le Nours,
    3. D. G. Pellicci,
    4. N. A. Gherardin,
    5. K. G. McPherson,
    6. R. T. Lim,
    7. O. Patel,
    8. T. Beddoe,
    9. S. Gras,
    10. J. Rossjohn,
    11. D. I. Godfrey
    . 2013. CD1d-lipid antigen recognition by the γδ TCR. Nat. Immunol. 14: 1137–1145.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Kedzierska K.,
    2. S. J. Turner,
    3. P. C. Doherty
    . 2004. Conserved T cell receptor usage in primary and recall responses to an immunodominant influenza virus nucleoprotein epitope. Proc. Natl. Acad. Sci. USA 101: 4942–4947.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Ortaldo J. R.,
    2. H. A. Young,
    3. R. T. Winkler-Pickett,
    4. E. W. Bere Jr..,
    5. W. J. Murphy,
    6. R. H. Wiltrout
    . 2004. Dissociation of NKT stimulation, cytokine induction, and NK activation in vivo by the use of distinct TCR-binding ceramides. J. Immunol. 172: 943–953.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Veerapen N.,
    2. F. Reddington,
    3. G. Bricard,
    4. S. A. Porcelli,
    5. G. S. Besra
    . 2010. Synthesis and biological activity of α-l-fucosyl ceramides, analogues of the potent agonist, α-d-galactosyl ceramide KRN7000. Bioorg. Med. Chem. Lett. 20: 3223–3226.
    OpenUrlCrossRefPubMed
  38. ↵
    1. Mattner J.,
    2. K. L. Debord,
    3. N. Ismail,
    4. R. D. Goff,
    5. C. Cantu III.,
    6. D. Zhou,
    7. P. Saint-Mezard,
    8. V. Wang,
    9. Y. Gao,
    10. N. Yin,
    11. et al
    . 2005. Exogenous and endogenous glycolipid antigens activate NKT cells during microbial infections. Nature 434: 525–529.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Gui M.,
    2. J. Li,
    3. L. J. Wen,
    4. R. R. Hardy,
    5. K. Hayakawa
    . 2001. TCRβ chain influences but does not solely control autoreactivity of Vβ14J281T cells. J. Immunol. 167: 6239–6246.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    1. Brossay L.,
    2. M. Chioda,
    3. N. Burdin,
    4. Y. Koezuka,
    5. G. Casorati,
    6. P. Dellabona,
    7. M. Kronenberg
    . 1998. CD1d-mediated recognition of an α-galactosylceramide by natural killer T cells is highly conserved through mammalian evolution. J. Exp. Med. 188: 1521–1528.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    1. Davodeau F.,
    2. M. A. Peyrat,
    3. A. Necker,
    4. R. Dominici,
    5. F. Blanchard,
    6. C. Leget,
    7. J. Gaschet,
    8. P. Costa,
    9. Y. Jacques,
    10. A. Godard,
    11. et al
    . 1997. Close phenotypic and functional similarities between human and murine αβ T cells expressing invariant TCR α-chains. J. Immunol. 158: 5603–5611.
    OpenUrlAbstract
  42. ↵
    1. Borg N. A.,
    2. K. S. Wun,
    3. L. Kjer-Nielsen,
    4. M. C. Wilce,
    5. D. G. Pellicci,
    6. R. Koh,
    7. G. S. Besra,
    8. M. Bharadwaj,
    9. D. I. Godfrey,
    10. J. McCluskey,
    11. J. Rossjohn
    . 2007. CD1d-lipid-antigen recognition by the semi-invariant NKT T-cell receptor. Nature 448: 44–49.
    OpenUrlCrossRefPubMed
  43. ↵
    1. Wun K. S.,
    2. N. A. Borg,
    3. L. Kjer-Nielsen,
    4. T. Beddoe,
    5. R. Koh,
    6. S. K. Richardson,
    7. M. Thakur,
    8. A. R. Howell,
    9. J. P. Scott-Browne,
    10. L. Gapin,
    11. et al
    . 2008. A minimal binding footprint on CD1d-glycolipid is a basis for selection of the unique human NKT TCR. J. Exp. Med. 205: 939–949.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Kjer-Nielsen L.,
    2. N. A. Borg,
    3. D. G. Pellicci,
    4. T. Beddoe,
    5. L. Kostenko,
    6. C. S. Clements,
    7. N. A. Williamson,
    8. M. J. Smyth,
    9. G. S. Besra,
    10. H. H. Reid,
    11. et al
    . 2006. A structural basis for selection and cross-species reactivity of the semi-invariant NKT cell receptor in CD1d/glycolipid recognition. J. Exp. Med. 203: 661–673.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. López-Sagaseta J.,
    2. L. V. Sibener,
    3. J. E. Kung,
    4. J. Gumperz,
    5. E. J. Adams
    . 2012. Lysophospholipid presentation by CD1d and recognition by a human natural killer T-cell receptor. EMBO J. 31: 2047–2059.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Stenström M.,
    2. M. Sköld,
    3. A. Andersson,
    4. S. L. Cardell
    . 2005. Natural killer T-cell populations in C57BL/6 and NK1.1 congenic BALB.NK mice-a novel thymic subset defined in BALB.NK mice. Immunology 114: 336–345.
    OpenUrlCrossRefPubMed
  47. ↵
    1. Fischer K.,
    2. E. Scotet,
    3. M. Niemeyer,
    4. H. Koebernick,
    5. J. Zerrahn,
    6. S. Maillet,
    7. R. Hurwitz,
    8. M. Kursar,
    9. M. Bonneville,
    10. S. H. Kaufmann,
    11. U. E. Schaible
    . 2004. Mycobacterial phosphatidylinositol mannoside is a natural antigen for CD1d-restricted T cells. Proc. Natl. Acad. Sci. USA 101: 10685–10690.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Kinjo Y.,
    2. E. Tupin,
    3. D. Wu,
    4. M. Fujio,
    5. R. Garcia-Navarro,
    6. M. R. Benhnia,
    7. D. M. Zajonc,
    8. G. Ben-Menachem,
    9. G. D. Ainge,
    10. G. F. Painter,
    11. et al
    . 2006. Natural killer T cells recognize diacylglycerol antigens from pathogenic bacteria. Nat. Immunol. 7: 978–986.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section