NKT cells are a subset of T lymphocytes that is mainly restricted by the nonclassical MHC class I molecule, CD1d, and that includes several subpopulations, in particular CD4+ and CD4CD8 (DN) cells. In the mouse, differential distribution of these subpopulations as well as heterogeneity in the expression of various markers as a function of tissue localization have been reported. We have thus undertaken a detailed study of the DN NKT cell subpopulation. With a highly sensitive semiquantitative RT-PCR technique, its TCR repertoire was characterized in various tissues. We found that mouse DN NKT cells are a variable mixture of two subgroups, one bearing the invariant Vα14 chain paired to rearranged Vβ2, Vβ7, Vβ8.1, Vβ8.2, or Vβ8.3 β-chains and the other exhibiting unskewed α- and β-chains. The proportion of these subgroups varies from about 100:0 in thymus, 80:20 in liver, and 50:50 in spleen to 20:80% in bone marrow, respectively. Finally, further heterogeneity in the tissue-derived DN NKT cells was discovered by sequencing extensively Vβ8.2-Jβ2.5 rearrangements in individual mice. Despite a few recurrences in TCR sequences, we found that each population exhibits its own and broad TCRβ diversity.

Natural killer T cells are a subset of T cells that can be distinguished from conventional T cells by several features. Most of these cells express NK-associated markers, like the NK1.1 and Ly-49 molecules, an intermediate level of TCR and are CD4+ or CD8CD4 double negative (DN)3 (1, 2, 3, 4, 5). In the thymus, most of these cells express an invariant Vα14Jα15 TCR α-chain with TCRβ-chains predominantly skewed toward Vβ8.2 chains (1, 2, 3, 6, 7, 8). In contrast to conventional T cells, which recognize peptides presented by polymorphic MHC molecules, most NKT cells are restricted by a monomorphic MHC class I molecule, CD1d (4, 9, 10), and a sugar moiety of glycolipids bound to CD1d molecules triggers their activation (11, 12, 13, 14, 15, 16, 17).

A recent study has pointed out the different tissular segregation of two NKT cell types (18). One cell type is CD1d restricted, is present in all lymphoid organs, and has a CD4+ or DN phenotype. Its development is mainly thymus dependent. The second one, enriched in spleen and bone marrow, has a CD8+ or DN phenotype, is not CD1d restricted, and develops independently of the thymus. It was further shown that thymus and liver, which are enriched in absolute numbers of CD1d-dependent CD4+ and DN NKT cells, are also enriched in NKT cells with a low expression of NK receptors (Ly-49A and DX5), a memory/activated phenotype (CD62LCD69+), and a TCR repertoire characterized by an overrepresentation of Vβ8.2+ chains and an under-representation of four α-chains other than the invariant Vα14 one (18, 19, 20). Conversely, spleen and bone marrow, which are enriched in absolute numbers of CD1d-independent CD4+, DN, and CD8+ NKT cells, are also enriched in NKT cells with a higher expression of NK receptors, a “naive” phenotype (CD62L+CD69), and a less skewed TCR repertoire (18, 19, 20). Previous work have suggested that tissue-specific and/or cell type-specific ligands could be responsible for the distinct TCR reactivities of CD1d-restricted DN NKT cell hybridomas, derived from different tissues (9, 10, 21, 22). Because the Vα14Jα15 invariant rearrangement is generated at random (8), its expression by an important fraction of NKT cells suggests that some selective pressure is exerted on the TCR of these cells.

This tissue heterogeneity of NKT cells led us to examine in more detail, in individual mice, the TCR chain components expressed by the DN NKT cell subpopulation of various tissue localizations. Due to the small number of NKT cells recovered from an individual organ and the lack of many anti-TCR-Vβ- and anti-TCR-Vα-specific Abs, a powerful RT-PCR technique was used to quantify the Vβ usage and the Vα14 invariant chain expression by thymic, hepatic, splenic, and bone marrow DN NKT cells of individual mice. We found that DN NKT cells localized in different tissues display polyclonal and different TCR repertoires and further demonstrated that these differences reflect the coexistence of two subgroups of DN NKT cells, the proportion of which varies depending on the tissue. DN NKT cells of subgroup I express exclusively the invariant Vα14 chain paired with Vβ2, Vβ7, Vβ8.1, Vβ8.2, and Vβ8.3 chains and represent ∼100, 80, 50, and 20% of DN NKT cells in thymus, liver, spleen, and bone marrow, respectively. In contrast, DN NKT cells of subgroup II express other TCR α-chains paired with unbiased Vβ chains. To further characterize the various DN NKT cell populations, we extensively sequenced the Vβ8.2-Jβ2.5 rearrangements and found that each tissue contains DN NKT cells with a unique TCRβ diversity.

Female 8- to 12-wk-old C57BL/6 mice used for this study were purchased from IFFA-Credo (l’Arbresle, France). They were maintained under specific pathogen-free conditions.

Single-cell suspensions were prepared from thymus, liver, spleen, and bone marrow (femur, tibia). Hepatic leukocytes were recovered from PBS-perfused liver, as described elsewhere (23). After a preincubation step with the 2.4G2 mAb, thymic and hepatic cells were incubated with biotinylated anti-CD8, whereas splenic and bone marrow cells were incubated with biotinylated anti-CD8, anti-CD19, anti-MacI, and anti-GR1 mAbs. After four washes, streptavidin-conjugated beads (Dynal, Oslo, Norway) were added, and the depletion was performed according the procedure described in the Dynal manual. Depleted cell suspensions were recovered after 10 washes.

Anti-Fcγ III/II receptors (2.4G2), PE-anti-NK1.1 (PK136), FITC-anti-TCRβ (H57-597), biotinylated and allophycocyanin-anti-CD4 (RM4-5), biotinylated anti-Ly-6G (RB6-8C5), biotinylated anti-CD19 (1D3), biotinylated anti-CD11b (M1/70), biotinylated anti-CD62L (MEL-14) mAbs, and streptavidin-allophycocyanin were purchased from PharMingen (San Diego, CA). Biotinylated anti-CD8 (CT-CD8a) and streptavidin-tricolor were from Caltag (South San Francisco, CA), and streptavidin red 613 was from Life Technologies (Gaithersburg, MD).

After incubation with the 2.4G2 mAb, depleted cell suspensions were stained with anti-NK1.1, anti TCRβ, anti-CD4, and anti-CD62L (for the four-color cell sorting) mAbs for 30 min on ice. After four washes, cells were incubated with streptavidin conjugate for 15 min, washed, and resuspended in PBS, 1% FCS. The three-color and four-color cell sortings were performed on a FACStarPlus (Becton Dickinson, Mountain View, CA) and an Epics-Elite ESP (Beckman Coulter, Fullerton, CA), respectively.

Total RNA from sorted DN NKT cells was extracted and reverse-transcribed using the avian myeloblastosis virus reverse transcriptase (Boehringer Mannheim, Mannheim, Germany), as previously described (24).

Vβ-specific primers, with the exception of Vβ8.2, have been described elsewhere (25). The Vβ8.2-specific primer was TTCATATGGTGCTGGCAGCACT, and the labeled Cβ-specific primer was 6-carboxyfluorescein (FAM)-CTTGGGTGGAGTCACATTTCTC. Vα-specific primers, with the exception of Vα9, Vα15, and Vα14, have been described previously (26). Vα9-, Vα15-, and Vα14-specific primers were ACACCGTTGTTAAAGGCACC, GAGCCAAAGACTTATAGTTTT, and CTAAGCACAGCACGCTGCACA, respectively. The labeled Cα-specific primer was FAM-ACACAGCAGGTTCTGGGTTC. Primers for CD3ε were 5′-primer GCCTCAGAAGCATGATAAGC and 3′-primer CCTTGGCCTTCCTATTCTTG, the one used for run-off reactions being 3′-FAM-CCCAGAGTGATACAGATGTC. The Jβ2.5-specific primer was GAGCCGAGGAGCACAATCTCC.

The number of CD3ε copies contained in each cDNA sample was quantified as follows. We used a plasmid bearing a 4-bp deleted CD3ε-specific sequence which can be amplified by the CD3ε-specific primers. From 106 to 102 copies of the plasmid were mixed to 1 μl of cDNA sample, and a 40-cycle competitive PCR was performed using the unlabeled CD3ε primers in a final volume of 25 μl. One microliter of the latter PCR was submitted to two cycles of run-off reaction, using the FAM-labeled CD3ε primer. Resulting fluorescent PCR products were loaded on a 6% polyacrylamide gel and run on a automated sequencer (Applied Biosystems, Foster City, CA). The exact number of CD3ε copies contained in each cDNA sample was determined using the immunoscope software (24, 27). Semiquantitative analyses of the TCRαβ repertoires were conducted as follows. cDNA containing 104 copies of CD3ε-specific cDNAs were subjected to 31 PCR cycles using either a Vβ-specific primer and the Cβ-fluorescent primer, or a Vα-specific primer and the Cα-fluorescent primer. Under such conditions, exponential PCRs were generated for all V-C combinations. The latter were run on an automated sequencer. The CDR3 length of PCR products and their fluorescence intensity were determined with the immunoscope software. When working on unprimed and polyclonal T cell populations, the CDR3 size distribution of each V-C-specific PCR product adopts a Gaussian-like distribution, composed of six to eight peaks separated by three nucleotides, because they derive from in-frame mRNAs. The area of peaks is proportional to the intensity of fluorescent band and thus to the initial amount of TCR transcripts. The percentage of fluorescence intensity of a given V-C combination was determined as the ratio of the area of all peaks generated in a single V-C PCR product to areas of all V-C-generated peaks in the T cell population.

For each sample, an amount of cDNA containing 2 × 104 copies of CD3ε was amplified using Vβ8.2- and Jβ2.5-specific primers with 5 U Pfu DNA polymerase (Stratagene, La Jolla, CA) in the supplier’s buffer. PCR products were ethanol precipitated and cloned in the PCR-Blunt II-Topo vector from the Zero Blunt TOPO PCR Cloning Kit (Invitrogen, Carlsbad, CA), following the manufacturer’s instructions. Transformation of TOP10 One Shot-competent cells by the previous ligation products and sequencing of cloned Vβ8.2-Jβ2.5 rearrangements were performed as described (28). The error rate of sequences, consistent with the experimental procedure, as calculated on Vβ8.2 and the Jβ2.5 germline sequences, was 10−4 mutations/nucleotide.

To address the question of whether DN NKT express distinct repertoires depending on the tissue they colonize, NK1.1+ TCRβintCD4CD8 cells were sorted (>99% pure) from thymus, liver, spleen, and bone marrow of individual mice (Fig. 1). RNA was extracted from a defined number of cells in each population, and the TCR-Vβ usage was measured by a semiquantitative RT-PCR procedure combined to the immunoscope technology, as described in Materials and Methods. Compared with conventional T cells, DN NKT cells from thymus and liver preferentially, although not exclusively, use Vβ2, Vβ7, Vβ8.1, Vβ8.2, and Vβ8.3 segments (Fig. 2). The latter are found in similar proportion in the two organs (Table I). In spleen, the Vβ usage of DN NKT cells is less skewed than in thymus and liver (Fig. 2), Vβ7, Vβ8.2, and Vβ8.3 still being over-represented (Table I). Bone marrow-derived DN NKT cells display a Vβ usage similar to that of conventional T cells (Fig. 2), the only noticeable difference being an over-representation of Vβ7 and Vβ8.2 chains (Table I).

FIGURE 1.

Sorted tissue-derived DN NKT cells. CD8 thymic and hepatic cells, as well as CD8CD19MacIGR1 splenic and bone marrow cells which were purified from individual mice were sorted accordingly to the NK1.1+TCRintCD4 phenotype.

FIGURE 1.

Sorted tissue-derived DN NKT cells. CD8 thymic and hepatic cells, as well as CD8CD19MacIGR1 splenic and bone marrow cells which were purified from individual mice were sorted accordingly to the NK1.1+TCRintCD4 phenotype.

Close modal
FIGURE 2.

TCR-Vβ usage of tissue-derived DN NKT cells. Semiquantitative RT-PCR immunoscope analysis was performed on DN NKT cells (▪) purified from thymus (A), liver (B), spleen (C), and bone marrow (D) of individual mice and on conventional T cells of lymph node origin (□). Results are representative of two independently tested mice. The x-axis depicts TCR-Vβ segments, and the y-axis shows the percentage of their respective fluorescence intensity.

FIGURE 2.

TCR-Vβ usage of tissue-derived DN NKT cells. Semiquantitative RT-PCR immunoscope analysis was performed on DN NKT cells (▪) purified from thymus (A), liver (B), spleen (C), and bone marrow (D) of individual mice and on conventional T cells of lymph node origin (□). Results are representative of two independently tested mice. The x-axis depicts TCR-Vβ segments, and the y-axis shows the percentage of their respective fluorescence intensity.

Close modal
Table I.

Ratios of Vβ2, Vβ7, Vβ8.1, Vβ8.2, and Vβ8.3 segment usages between DN NKT cells from various tissues and conventional T cells

Tissue Localization of DN NKT Cells% of Usage in DN NKT Cells/% of Usage in Conventional T Cells
Vβ2Vβ7Vβ8.1Vβ8.2Vβ8.3
Thymus 2.10 4.25 1.81 3.51 1.78 
Liver 1.85 3.61 1.61 3.26 1.86 
Spleen 1.24 2.06 1.35 1.95 2.34 
Bone marrow 0.93 1.66 1.19 1.90 1.14 
Tissue Localization of DN NKT Cells% of Usage in DN NKT Cells/% of Usage in Conventional T Cells
Vβ2Vβ7Vβ8.1Vβ8.2Vβ8.3
Thymus 2.10 4.25 1.81 3.51 1.78 
Liver 1.85 3.61 1.61 3.26 1.86 
Spleen 1.24 2.06 1.35 1.95 2.34 
Bone marrow 0.93 1.66 1.19 1.90 1.14 

All NKT cells do not express the invariant Vα14 chain (7, 8, 14, 21). It was thus necessary to determine the proportion of Vα14 invariant chain-bearing cells in the sorted DN subpopulations. Because the anti-Vα14 Ab does not recognize all Vα14+ NKT cells (Ref. 7 and data not shown), it precludes this investigation by flow cytometry; we thus used the same semiquantitative RT-PCR approach as above to analyze Vα14 transcripts in DN NKT cells.

In thymus, liver, spleen, and bone marrow, the Vα14Cα immunoscope profiles of DN NKT cells display a single 10-aa CDR3 length peak (Fig. 3,A) which was further identified to the Vα14Jα15 invariant rearrangement (data not shown). This, incidentally, confirms the purity of sorted DN NKT cells because conventional T cells show a polyclonal Vα14Cα profile with several CDR3 size peaks (Fig. 3,A). Quantification of the invariant rearrangement by RT-PCR revealed, however, significant differences between tissues (Fig. 3 B). Relative to the same number of TCR transcripts, expression was highest in thymus and liver, intermediate in spleen, and low in bone marrow. This observation implies that either all DN NKT cells express the Vα14 invariant rearrangement, albeit at different levels, or that Vα14+ cells are mixed with variable amounts of Vα14 cells. Because β-chain expression, as measured by flow cytometry analysis, was similar in all cases, we favored the latter hypothesis. Because thymic DN NKT cells were reported to express almost exclusively the invariant Vα14 chain (2), a reasonable assumption is that thymus, liver, spleen, and bone marrow might contain ∼0, 20, 50, and 80% of Vα14 DN cells, respectively.

FIGURE 3.

Vα14-Cα rearrangements in tissue-derived DN NKT cells and in conventional T cells of individual mice. A, Immunoscope profiles of Vα14-Cα rearrangements in thymic (a), hepatic (b), splenic (c), and bone marrow (d) DN NKT cells and in conventional T cells (e). B, Semiquantitative analysis of Vα14-Cα rearrangements in the same populations. Data are representative of two independently tested mice. Fluorescence intensity is represented on the x-axis, and a value of 1 was arbitrarily attributed to the fluorescence intensity obtained with thymic DN NKT cells. That of conventional T cells corresponds to the value obtained from the peak with a 10-aa CDR3 length, because it is not possible to discriminate between the Vα14 invariant rearrangement and irrelevant Vα14 rearrangements of the same CDR3 size.

FIGURE 3.

Vα14-Cα rearrangements in tissue-derived DN NKT cells and in conventional T cells of individual mice. A, Immunoscope profiles of Vα14-Cα rearrangements in thymic (a), hepatic (b), splenic (c), and bone marrow (d) DN NKT cells and in conventional T cells (e). B, Semiquantitative analysis of Vα14-Cα rearrangements in the same populations. Data are representative of two independently tested mice. Fluorescence intensity is represented on the x-axis, and a value of 1 was arbitrarily attributed to the fluorescence intensity obtained with thymic DN NKT cells. That of conventional T cells corresponds to the value obtained from the peak with a 10-aa CDR3 length, because it is not possible to discriminate between the Vα14 invariant rearrangement and irrelevant Vα14 rearrangements of the same CDR3 size.

Close modal

The fraction of cells overexpressing Vβ2, Vβ7, Vβ8.1, Vβ8.2, and Vβ8.3 segments in the various tissues closely correlates with the proportion of putative Vα14+ cells, suggesting the existence of two subgroups of DN NKT cells with distinct TCR repertoires. Thus, and since, consistent with a recent study (18), we found that the percentage of CD62L DN NKT cells varies in different tissues (99% in thymus, 96% in liver, 50% in spleen, and 40% in bone marrow; data not shown), in a way reminiscent of the Vα14+ cell proportions, we purposefully compared the Vβ and Vα usages of CD62L and CD62L+ DN NKT cells.

Because the small numbers of thymic and hepatic CD62L+ DN NKT cells hinder analyses of their repertoire, we focused on spleen that contains comparable numbers of CD62L and CD62L+ as well as Vα14+ and Vα14 DN NKT cells. To obtain a sufficient number of cells, 95% pure splenic CD62L and CD62L+ DN NKT cells were isolated from three spleens, and the V segment usage of their TCR chains was analyzed semiquantitatively. CD62L cells displayed a TCR-Vβ usage strongly biased toward Vβ2, Vβ7, Vβ8.1, Vβ8.2, and Vβ8.3 segments (Fig. 4,A, left) and expressed almost exclusively the invariant Vα14 chain (Fig. 4,B, left). In contrast, CD62L+ cells had an unbiased Vβ usage (Fig. 4,A, right), and TCR α-chains distinct from the invariant Vα14 chain (Fig. 4,B, right). Remarkably, the Vα14 invariant chain with a 10-aa-long CDR3 region is the unique Vα14-Cα rearrangement of CD62L DN NKT cells, as disclosed by immunoscope profile (Fig. 4,C, left). The rare Vα14-Cα rearrangements detected in the CD62L+ population are polyclonal because several CDR3 lengths were observed (Fig. 4 C, right). Therefore, splenic DN NKT cells contain two subgroups: subgroup I cells of CD62L phenotype use the invariant Vα14 chain associated with Vβ2, Vβ7, Vβ8.1, Vβ8.2, and Vβ8.3 chains; and subgroup II cells of CD62L+ phenotype express unbiased TCR α- and β-chains.

FIGURE 4.

Semiquantitative analyses of the TCR-Vβ (A) and TCR-Vα (B) usages of splenic CD62L (left) and CD62L+ (right) DN NKT cells. x- and y-axes are as in Fig. 2. C, Immunoscope Vα14Cα profiles obtained from the latter populations.

FIGURE 4.

Semiquantitative analyses of the TCR-Vβ (A) and TCR-Vα (B) usages of splenic CD62L (left) and CD62L+ (right) DN NKT cells. x- and y-axes are as in Fig. 2. C, Immunoscope Vα14Cα profiles obtained from the latter populations.

Close modal

The respective proportions of Vα14+ and Vα14 cells in thymus, liver, spleen, and bone marrow, are 100:0, 80:20, 50:50, and 20:80, respectively. Thus, we took into account the aforementioned proportions to determine whether the coexistence in different proportions of the two subgroups of DN NKT cells, as defined on the basis of their TCR repertoire, underlies the distinct Vβ usages found in the various tissues.

In view of the Vβ usage obtained in Fig. 4 and the assumed relative tissue distribution of the two subgroups, we calculated a TCR-Vβ usage for each tissue and compared it with the one experimentally obtained and depicted in Fig. 2. Fig. 5 shows that for each tissue, the calculated TCR-Vβ usage of DN NKT cells closely matches the experimental one. It is highly likely, therefore, that thymus contains only subgroup I DN NKT cells whereas liver, spleen, and bone marrow harbor 80:20, 50:50, and 20:80 mixtures of subgroups I and II, respectively. Although all splenic CD62L DN NKT cells express the invariant Vα14 chain, in liver and bone marrow the proportion of CD62L DN NKT cells is higher than the Vα14+ cell proportion. Hence, subgroup I cells, as defined on the basis of their TCR components, do not represent all of the CD62L DN NKT cells in the latter tissues.

FIGURE 5.

Comparison between the DN NKT cell Vβ usage observed in Fig. 2 (□) and the one calculated by assuming that the proportion of subgroup I and II cells, with a Vβ usage as defined in Fig. 4, correspond to those of Vα14+ and Vα14 DN NKT cells (i.e., 100:0, 80:20, 50:50, and 20:80 in thymus, liver, spleen, and bone marrow, respectively) (▪).

FIGURE 5.

Comparison between the DN NKT cell Vβ usage observed in Fig. 2 (□) and the one calculated by assuming that the proportion of subgroup I and II cells, with a Vβ usage as defined in Fig. 4, correspond to those of Vα14+ and Vα14 DN NKT cells (i.e., 100:0, 80:20, 50:50, and 20:80 in thymus, liver, spleen, and bone marrow, respectively) (▪).

Close modal

We next investigated the polyclonality of DN NKT cell TCRβ-chains in individual mice. The standard polyclonality of conventional T cells is reflected into Gaussian-like CDR3 size distributions for every analyzed VβCβ or VβJβ immunoscope profile. Representative profiles obtained for Vβ segments either predominantly used by subgroup I (Vβ2 and Vβ8.2) or used by subgroup II (Vβ5.1 and Vβ12) DN NKT cells are shown in Fig. 6. In all mice, VβCβ combinations displayed a Gaussian-like distribution (Fig. 6 and data not shown). The analysis of Vβ8.2-bearing cells was further refined by measuring the Jβ usage (which showed a bias in favor of Jβ2.1, Jβ2.5, and Jβ2.7 in all cases; data not shown). The 12 Vβ8.2Jβ immunoscope profiles were Gaussian-like as well. Thus, in both subgroups and in all tissues, DN NKT cells were polyclonal. We never detected a clonal expansion that would show a higher CDR3 size peak over the Gaussian polyclonal background.

FIGURE 6.

DN NKT cells exhibit a polyclonal TCRβ repertoire. Immunoscope profiles of representative Vβ-Cβ rearrangements obtained from thymic (a), hepatic (b), splenic (c), and bone marrow-derived (d) DN NKT cells are depicted.

FIGURE 6.

DN NKT cells exhibit a polyclonal TCRβ repertoire. Immunoscope profiles of representative Vβ-Cβ rearrangements obtained from thymic (a), hepatic (b), splenic (c), and bone marrow-derived (d) DN NKT cells are depicted.

Close modal

It was of interest to determine whether various tissues share identical DN NKT cells. We sequenced a large number of β-chain rearrangements among the Vβ8.2-Jβ2.5 combination, most expressed within the Vα14+ subgroup. cDNAs were prepared from thymic, hepatic, splenic, and bone marrow-derived DN NKT cells purified from an individual mouse, and Vβ8.2-Jβ2.5 PCRs were performed on identical amounts of cDNA, containing 2 × 104 copies of CD3ε transcripts (about one-tenth of the total cDNA). The resulting PCR products were cloned in Escherichia coli and sequenced until approaching saturation.

From the number of sequences found once, twice, or more, it is possible to extrapolate and calculate the likely number of distinct sequences present in a given sample, known as maximum likelihood estimate (MLE) (29, 30). As shown in Table II, for each sample, >250 sequences were needed to approach a plateau. MLE values in the same range as in Table II were found in DN NKT cells from identical tissues of other mice (data not shown). Because thymic DN NKT cells are a homogeneous population of subgroup I cells, it is possible to provide an estimate of their Vβ repertoire size. About 50% of thymic DN NKT cells use the Vβ8.2 segment (Ref. 18 and data not shown), and the frequency of Jβ2.5 usage among Vβ8.2 rearrangements in the latter population, as determined by immunoscope analysis (data not shown), is ∼13%. Their Vβ repertoire size can thus be calculated as the number of distinct Vβ8.2-Jβ2.5 sequences (MLE value in Table II) divided by (frequency of Vβ segment × frequency of Jβ segment). The result of this calculation gives an estimate of 3.2 × 103 distinct TCRβ rearrangements for 2 × 105 purified thymic DN NKT cells, each rearrangement being thus shared by 60 thymic DN NKT cells. In comparison, 26–45 naive T splenocytes share a given TCRβ rearrangement that can associate with at least two distinct Vα chains (31). Hence, the average size of DN NKT cell clones is close to that of conventional T cells.

Table II.

Diversity of the Vβ8.2Jβ2.5 rearrangements found in DN NKT cells purified from various tissues of an individual mouse

TissueCDR3 Nucleotide SequencesCDR3 Amino Acid SequencesTotal Sequences% of Recurrent Vβ8.2Jβ2.5 Amino Acid Sequencesa
Redundant DistinctbDistinctcMLEdRedundant DistinctbDistinctcMLEd
Thymus 58 151 210 59 148 202 265 8.78 with liver 
 (38.41%) (56.98%)  (39.86%) (55.84%)   10.13 with spleen 
        1.35 with bone marrow 
Liver 76 182 253 78 179 245 320 7.26 with thymus 
 (41.75%) (56.87%)  (43.57%) (55.93%)   7.26 with spleen 
        1.67 with bone marrow 
Spleen 100 248 403 100 242 374 416 6.19 with thymus 
 (40.32%) (59.61%)  (41.32%) (58.17%)   5.37 with liver 
        2.89 with bone marrow 
Bone marrow 45 84 89 45 84 89 249 2.38 with thymus 
 (53.57%) (33.6%)  (53.57%) (33.6%)   3.57 with liver 
        8.33 with spleen 
TissueCDR3 Nucleotide SequencesCDR3 Amino Acid SequencesTotal Sequences% of Recurrent Vβ8.2Jβ2.5 Amino Acid Sequencesa
Redundant DistinctbDistinctcMLEdRedundant DistinctbDistinctcMLEd
Thymus 58 151 210 59 148 202 265 8.78 with liver 
 (38.41%) (56.98%)  (39.86%) (55.84%)   10.13 with spleen 
        1.35 with bone marrow 
Liver 76 182 253 78 179 245 320 7.26 with thymus 
 (41.75%) (56.87%)  (43.57%) (55.93%)   7.26 with spleen 
        1.67 with bone marrow 
Spleen 100 248 403 100 242 374 416 6.19 with thymus 
 (40.32%) (59.61%)  (41.32%) (58.17%)   5.37 with liver 
        2.89 with bone marrow 
Bone marrow 45 84 89 45 84 89 249 2.38 with thymus 
 (53.57%) (33.6%)  (53.57%) (33.6%)   3.57 with liver 
        8.33 with spleen 
a

Percentages are in number of recurrent amino acid sequences between two organs/number of total distinct amino acid sequences in the considered organ.

b

Percentages are in number of distinct redundant sequences/number of total distinct sequences.

c

Percentages are in number of distinct sequences/number of total sequences.

d

MLE of the number of distinct sequences, calculated with a 95% confidence interval (27, 28).

Inspection of individual amino acid sequences revealed that thymic, hepatic, and splenic DN NKT cell populations exhibited the greatest sequence recurrence (i.e., shared sequences), about 9% for thymus with liver or spleen, 7% for liver with thymus or spleen, and 6% for spleen with thymus or liver, whereas their sequence recurrence with bone marrow was 3- to 7-fold less (Table II). This is not unexpected, given the observation that these organs contain the highest proportion of subgroup I cells. Reciprocally, bone marrow, mostly composed of subgroup II cells, showed the greatest recurrence with spleen (Table II) which, in contrast to thymus and liver, also contains a high proportion of subgroup II cells. In all cases, each tissue-derived DN NKT cell population shares only a few of its Vβ8.2-Jβ2.5 rearrangements with the other populations. Hence, in every tissue, DN NKT cells display a mostly unique Vβ8.2-Jβ2.5 rearrangement diversity.

Table III shows that 6 distinct nucleotide sequences were shared between thymus and liver, 9 between thymus and spleen, and 8 between liver and spleen of 151, 182, and 248 distinct sequences in thymus, liver, and spleen, respectively (Table II). None was found between bone marrow (84 distinct sequences, Table II) and thymus, and only one was found between bone marrow and liver or spleen. Translation of CDR3 nucleotide sequences revealed a greater recurrence at the amino acid level (Table III). Two amino acid sequences, 8- and 9-aa-long CDR3, were common to thymus, liver, and spleen, and several others found in two of the three organs. As for bone marrow, only two and three sequences were common to, respectively, thymus and liver, and seven to the spleen. The amino acid CDR3 sequences that originate from distinct nucleotide sequences are indicated in Table III. Their relatively high number among recurrent sequences suggests that a selection takes place at the protein level.

Table III.

Shared nucleotide and amino acid Vβ8.2Jβ2.5 sequences between tissue-derived DN NKT cells of an individual mouse

CDR3 SizeRecurrent CDR3 Nucleotide SequencesRecurrent CDR3 Amino Acid Sequences
OccurrenceaSequenceOccurrenceaVβ8.2CDR3Jβ2.5
 Thymus Liver  Thymus Liver    
0 /1 0 /1  0 /1 0 /1    
2 /3 3 /5 TGTGCCAGCGGTGATAAAGAC 2 /3 3 /5 CAS GDKDTQ YFG 
 1 /3 1 /5 TGTGCCAGCGGTGAGGGAAAC 1 /3 1 /5 CAS GEGNTQ YFG 
0 /9 0 /11  0 /9 0 /11    
0 /19 0 /24  1 /19 5 /24 CAS GDAGGDTQ YFGb,c 
    2 /19 2 /24 CAS GGTGEDTQ YFGb 
1 /46 5 /69 TGTGCCAGCGGTGGGACTGGGGGCCAAGAC 4 /44d 5 /67d CAS GGTGGQDTQ YFGc 
 1 /46 1 /69 TGTGCCAGCGGTGATGCGCCAGGGAGAGAC 1 /44d 1 /67d CAS GDAPGRDTQ YFG 
    1 /44d 4 /67d CAS GEGLGQDTQ YFGb 
10 1 /40 1 /37 TGTGCCAGCGGTGGGATGGGGGGGCGCGAAGAC 1 /40 1 /37 CAS GGMGGREDTQ YFG 
    1 /40 2 /37 CAS GEGTGGQDTQ YFGb 
    1 /40 1 /37 CAS GDAEVNQDTQ YFGb 
    1 /40 2 /37 CAS GETGGSQDTQ YFGb 
11 0 /21 0 /22  1 /20d 3 /21d CAS GDAGGGDQDTQ YFGb 
12 1 /12 1 /9 TGTGCCAGCGGTGGACCGGGACTGGGGGGTAACCAAGAC 1 /12 1 /9 CAS GGPGLGGNQDTQ YFG 
13 0 /0 0 /4  0 /0 0 /4    
14 0 /0 0 /0  0 /0 0 /0    
CDR3 SizeRecurrent CDR3 Nucleotide SequencesRecurrent CDR3 Amino Acid Sequences
OccurrenceaSequenceOccurrenceaVβ8.2CDR3Jβ2.5
 Thymus Liver  Thymus Liver    
0 /1 0 /1  0 /1 0 /1    
2 /3 3 /5 TGTGCCAGCGGTGATAAAGAC 2 /3 3 /5 CAS GDKDTQ YFG 
 1 /3 1 /5 TGTGCCAGCGGTGAGGGAAAC 1 /3 1 /5 CAS GEGNTQ YFG 
0 /9 0 /11  0 /9 0 /11    
0 /19 0 /24  1 /19 5 /24 CAS GDAGGDTQ YFGb,c 
    2 /19 2 /24 CAS GGTGEDTQ YFGb 
1 /46 5 /69 TGTGCCAGCGGTGGGACTGGGGGCCAAGAC 4 /44d 5 /67d CAS GGTGGQDTQ YFGc 
 1 /46 1 /69 TGTGCCAGCGGTGATGCGCCAGGGAGAGAC 1 /44d 1 /67d CAS GDAPGRDTQ YFG 
    1 /44d 4 /67d CAS GEGLGQDTQ YFGb 
10 1 /40 1 /37 TGTGCCAGCGGTGGGATGGGGGGGCGCGAAGAC 1 /40 1 /37 CAS GGMGGREDTQ YFG 
    1 /40 2 /37 CAS GEGTGGQDTQ YFGb 
    1 /40 1 /37 CAS GDAEVNQDTQ YFGb 
    1 /40 2 /37 CAS GETGGSQDTQ YFGb 
11 0 /21 0 /22  1 /20d 3 /21d CAS GDAGGGDQDTQ YFGb 
12 1 /12 1 /9 TGTGCCAGCGGTGGACCGGGACTGGGGGGTAACCAAGAC 1 /12 1 /9 CAS GGPGLGGNQDTQ YFG 
13 0 /0 0 /4  0 /0 0 /4    
14 0 /0 0 /0  0 /0 0 /0    
ThymusSpleenThymusSpleen
1 /1 3 /1 TGTGCCAGCGGAGGGGAC 1 /1 3 /1 CAS GGDTQ YFG 
0 /3 0 /4  0 /3 0 /2d    
2 /9 2 /10 TGTGCCAGCGGTGGACAAAAAGAC 2 /9 2 /10 CAS GGQKDTQ YFG 
1 /19 1 /33 TGTGCCAGCGGTGATGCGGGGGGAGAC 1 /19 2 /32d CAS GDAQQDTQ YFG 
    2 /19 1 /32d CAS GGGSQDTQ YFGb 
    1 /19 1 /32d CAS GDAGGDTQ YFGb,c 
2 /46 1 /79 TGTGCCAGCGGTGATGACTGGGGGCAAGAC 2 /44d 1 /77d CAS GDDWGQDTQ YFG 
 2 /46 2 /79 TGTGCCAGCGGTGATGCAGGTTTCCAAGAC 2 /44d 2 /77d CAS GDAGFQDTQ YFG 
    4 /44d 2 /77d CAS GGTGGQDTQ YFGb,c 
    1 /44d 1 /77d CAS GETGGQDTQ YFGb 
10 6 /40 2 /64 TGTGCCAGCGGGGGACTGGGGGTTAACCAAGAC 6 /40 2 /63d CAS GGLGVNQDTQ YFG 
 1 /40 1 /64 TGTGCCAGCGGTGATGAGAGACAGGGCCAAGAC 1 /40 1 /63d CAS GDERQGQDTQ YFG 
 3 /40 1 /64 TGTGCCAGCGGTGGACTGGGGGGGCGCCAAGAC 3 /40 1 /63d CAS GGLGGRQDTQ YFG 
    2 /40 1 /63d CAS GRLGGRQDTQ YFGb 
11 0 /21 0 /36  2 /20d 1 /36 CAS GDAGLGGQDTQ YFGb 
12 1 /12 3 /16 TGTGCCAGCGGTGATGCGGGACTGGGGATTAACCAAGAC 1 /12 3 /16 CAS GDAGLGINQDTQ YFG 
13 0 /0 0 /4  0 /0 0 /4    
14 0 /0 0 /1  0 /0 0 /1    
ThymusSpleenThymusSpleen
1 /1 3 /1 TGTGCCAGCGGAGGGGAC 1 /1 3 /1 CAS GGDTQ YFG 
0 /3 0 /4  0 /3 0 /2d    
2 /9 2 /10 TGTGCCAGCGGTGGACAAAAAGAC 2 /9 2 /10 CAS GGQKDTQ YFG 
1 /19 1 /33 TGTGCCAGCGGTGATGCGGGGGGAGAC 1 /19 2 /32d CAS GDAQQDTQ YFG 
    2 /19 1 /32d CAS GGGSQDTQ YFGb 
    1 /19 1 /32d CAS GDAGGDTQ YFGb,c 
2 /46 1 /79 TGTGCCAGCGGTGATGACTGGGGGCAAGAC 2 /44d 1 /77d CAS GDDWGQDTQ YFG 
 2 /46 2 /79 TGTGCCAGCGGTGATGCAGGTTTCCAAGAC 2 /44d 2 /77d CAS GDAGFQDTQ YFG 
    4 /44d 2 /77d CAS GGTGGQDTQ YFGb,c 
    1 /44d 1 /77d CAS GETGGQDTQ YFGb 
10 6 /40 2 /64 TGTGCCAGCGGGGGACTGGGGGTTAACCAAGAC 6 /40 2 /63d CAS GGLGVNQDTQ YFG 
 1 /40 1 /64 TGTGCCAGCGGTGATGAGAGACAGGGCCAAGAC 1 /40 1 /63d CAS GDERQGQDTQ YFG 
 3 /40 1 /64 TGTGCCAGCGGTGGACTGGGGGGGCGCCAAGAC 3 /40 1 /63d CAS GGLGGRQDTQ YFG 
    2 /40 1 /63d CAS GRLGGRQDTQ YFGb 
11 0 /21 0 /36  2 /20d 1 /36 CAS GDAGLGGQDTQ YFGb 
12 1 /12 3 /16 TGTGCCAGCGGTGATGCGGGACTGGGGATTAACCAAGAC 1 /12 3 /16 CAS GDAGLGINQDTQ YFG 
13 0 /0 0 /4  0 /0 0 /4    
14 0 /0 0 /1  0 /0 0 /1    
LiverSpleenLiverSpleen
0 /1 0 /1  0 /1 0 /1    
0 /5 0 /4  0 /5 0 /2d    
0 /11 0 /10  1 /11 3 /10 CAS GDATDTQ YFGb 
1 /24 1 /33 TGTGCCAGCGGTGATGGGGGGGCAGAC 1 /24 1 /32d CAS GDGGADTQ YFG 
    5 /24 1 /32d CAS GDAGGDTQ YFGb,c 
3 /69 2 /79 TGTGCCAGCGGTGAAGAGACTGGGGGGGTC 3 /67d 2 /77d CAS GEETGGVTQ YFG 
 3 /69 3 /79 TGTGCCAGCGGTGCCGGGACTGGGGAAGAC 3 /67d 3 /77d CAS GAGTGEDTQ YFG 
 2 /69 3 /79 TGTGCCAGCGGGGCGACAATTAACCAAGAC 2 /67d 3 /77d CAS GATINQDTQ YFG 
 2 /69 1 /79 TGTGCCAGCGGTGATGGGACTGGGGAAGAC 3d /67d 1 /77d CAS GDGTGEDTQ YFG 
 1 /69 4 /79 TGTGCCAGCGGTGATGTGGGGATCCAAGAC 1 /67d 4 /77d CAS GDVGIQDTQ YFG 
    1 /67d 1 /77d CAS GDAGGQDTQ YFGb 
    1 /67d 3 /77d CAS GEDWGQDTQ YFGb 
    5 /67d 2 /77d CAS GGTGGQDTQ YFGb,c 
10 1 /37 3 /64 TGTGCCAGCGGTGATGCCAGGGGTAACCAAGAC 1 /37 3 /63d CAS GDARGNQDTQ YFG 
11 0 /22 0 /36  0 /21d 0 /36    
12 1 /9 1 /16 TGTGCCAGCGGTGATGAATCCCTACTGGGGAACCAAGAC 1 /9 1 /16 CAS GDESLLGNQDTQ YFG 
13 0 /4 0 /4  0 /4 0 /4    
14 0 /0 0 /1  0 /0 0 /1    
LiverSpleenLiverSpleen
0 /1 0 /1  0 /1 0 /1    
0 /5 0 /4  0 /5 0 /2d    
0 /11 0 /10  1 /11 3 /10 CAS GDATDTQ YFGb 
1 /24 1 /33 TGTGCCAGCGGTGATGGGGGGGCAGAC 1 /24 1 /32d CAS GDGGADTQ YFG 
    5 /24 1 /32d CAS GDAGGDTQ YFGb,c 
3 /69 2 /79 TGTGCCAGCGGTGAAGAGACTGGGGGGGTC 3 /67d 2 /77d CAS GEETGGVTQ YFG 
 3 /69 3 /79 TGTGCCAGCGGTGCCGGGACTGGGGAAGAC 3 /67d 3 /77d CAS GAGTGEDTQ YFG 
 2 /69 3 /79 TGTGCCAGCGGGGCGACAATTAACCAAGAC 2 /67d 3 /77d CAS GATINQDTQ YFG 
 2 /69 1 /79 TGTGCCAGCGGTGATGGGACTGGGGAAGAC 3d /67d 1 /77d CAS GDGTGEDTQ YFG 
 1 /69 4 /79 TGTGCCAGCGGTGATGTGGGGATCCAAGAC 1 /67d 4 /77d CAS GDVGIQDTQ YFG 
    1 /67d 1 /77d CAS GDAGGQDTQ YFGb 
    1 /67d 3 /77d CAS GEDWGQDTQ YFGb 
    5 /67d 2 /77d CAS GGTGGQDTQ YFGb,c 
10 1 /37 3 /64 TGTGCCAGCGGTGATGCCAGGGGTAACCAAGAC 1 /37 3 /63d CAS GDARGNQDTQ YFG 
11 0 /22 0 /36  0 /21d 0 /36    
12 1 /9 1 /16 TGTGCCAGCGGTGATGAATCCCTACTGGGGAACCAAGAC 1 /9 1 /16 CAS GDESLLGNQDTQ YFG 
13 0 /4 0 /4  0 /4 0 /4    
14 0 /0 0 /1  0 /0 0 /1    

Table III (continued)

To determine whether identical sequences would be found in other animals, we analyzed thymic and hepatic DN NKT cells from a second mouse, splenic and bone marrow DN NKT cells from a third mouse (data not shown). Eight, 10, 13, and 2 amino acid sequences, mostly encoded by different nucleotide sequences, were recurrent to thymi, livers, spleens, and bone marrows, respectively. Some of them were encoded in a given organ by two distinct nucleotide sequences. These observations show that the Vβ8.2-Jβ2.5 repertoire of DN NKT cells irrespective of the tissue localization displays a large individual variability, as has already been described for conventional T cells (32).

Mouse NKT cells comprise two major subpopulations with thus far not well-defined functions, the CD4+ and the DN NKT cells. The present study investigates the cellular basis of the tissue heterogeneity of the latter subpopulation. To that purpose, we analyzed in detail the TCR repertoire of these cells, using the immunoscope technique, combined with a semiquantitative RT-PCR procedure.

In thymus, liver, spleen, and bone marrow, ∼100, 80, 50, and 20% of DN NKT cells expressing the invariant Vα14 chain were, respectively, observed. The skewing in the TCR-Vβ usage toward Vβ2, Vβ7, Vβ8.1, Vβ8.2, and Vβ8.3 segments of each tissue-derived population decreased in proportion to Vα14 expression. The latter correlation was suggestive of the existence of two distinct subgroups, one expressing a TCR biased for both chains and the other expressing an unbiased TCR. Because CD62L and CD62L+ DN NKT cells segregate in these tissues in proportions reminiscent of the Vα14+ and Vα14 populations, particularly in spleen, the Vα and Vβ usages of splenic CD62L and CD62L+ DN NKT cells were analyzed. CD62L DN NKT cells were found to express exclusively the invariant Vα14 chain associated with Vβ2, Vβ7, Vβ8.1, Vβ8.2, or Vβ8.3 chains, whereas CD62L+ cells express TCR α-chains other than Vα14, associated with unbiased TCR β-chains. We further demonstrated that subgroup I DN NKT cells, as defined by the expression of a biased TCR repertoire, and subgroup II DN NKT cells, as defined by the expression of an unbiased TCR repertoire, coexist in thymus, liver, spleen, and bone marrow in proportions close to 100:0, 80:20, 50:50, and 20:80, respectively.

The above findings are consistent with several observations reported by other groups : DN NKT cells were recently found to be diminished in Jα281−/− mice by ∼90, 75, 40, and 60–80% in thymus, liver, spleen, and bone marrow, respectively (19, 20). In sharp contrast to CD1−/− mice in which DN NKT cells express an unbiased TCRαβ repertoire, DN NKT cells from Jα281−/− mice express, exclusively in thymus and liver, a TCRαβ repertoire which is biased toward the single TCR β-chain (18). These thymic and hepatic cells are therefore likely to be selected by CD1d molecules in Jα281−/− mice, being otherwise not selected in C57BL/6 animals in which they display a TCR repertoire highly skewed for both TCR chains (subgroup I DN NKT cells), as shown in the present work. Hence, thymus and liver of C57BL/6 mice might comprise a percentage of Vα14+ DN NKT cells that is slightly higher than that deduced from residual DN NKT cells of Jα281−/− mice. The overexpression of Vβ8.2+ chains and the low expression of non Vα14 TCR α-chains which were observed on thymic and hepatic DN NKT cells by flow cytometry suggested that the majority of these cells were harboring a TCR repertoire biased for both chains (18), which was confirmed by our data. However, the bias toward Vβ8.2+ chain expression simultaneously with an important expression of TCR α-chains other than Vα14 by splenic and bone marrow DN NKT cells was interpreted as the existence, in these tissues, of cells bearing a TCR repertoire in which only the β-chain is biased (18). In view of our results, the flow cytometry stainings of Eberl et al. can be reinterpreted as reflecting the tissue-variable mixture of subgroups I and II of DN NKT cells that we identified, rather than the existence of DN NKT cells biased either for both TCR chains, in thymus and liver, or for the single TCR β-chain, in spleen and bone marrow.

All DN NKT cells display a memory phenotype because they bear high levels of the CD44 molecule. However, Oehen et al. (33) reported that only conventional memory T cells having down-regulated the CD62L molecule display an immediate effector function and suggested that chronic Ag activation is required to maintain CD62L at low levels. It is possible that this correlation also applies to NKT cells because they participate in the very early stages of immune responses (34, 35, 36) and produce high levels of cytokines within few hours after TCR engagement (31, 34, 35, 37, 38). Thus, subgroup I DN NKT cells (Vα14+CD62L) may be subject to a chronic activation in all tissues. In spleen, all subgroup II cells are CD62L+, hence, likely not chronically stimulated. In liver and bone marrow, DN NKT cells comprise higher numbers of CD62L than Vα14+ cells, meaning that their CD62L population has not only subgroup I cells but also subgroup II cells. Therefore, liver and bone marrow environments may deliver stimulatory signals both to subgroup I and subgroup II cells. Furthermore, there is a good relevance between the differential ability of DN NKT cells from various tissues to secrete rapidly cytokines and their phenotype of effector and/or resting memory cells in these tissues (19). Overall, CD62L is differentially expressed by subgroup II cells depending on the tissue, which strongly suggest that the two subgroups may be regulated by distinct sets of ligands. Our work also reveals that CD62L DN NKT cells are more heterogeneous than their CD4+ counterpart because the latter has been recently demonstrated to exhibit a TCR biased for both chains whatever their tissue localization (18).

CD4+ and DN NKT cells which develop in CD1d−/− mice have been shown to display an unbiased TCR repertoire (18). This observation has led to the hypothesis that CD1d-dependent and -independent NKT cell populations could be distinguished from each other by the expression of a biased and an unbiased TCR repertoire, respectively. However, all tissues comprise about 70% of CD1d-restricted DN NKT cells (18) whereas the two subgroups segregate differently. No strict correlation can be thus established between subgroup I and II DN NKT cells, and CD1d restriction in wild-type mice. In particular, subgroup II DN NKT cells are likely a mixture of CD1d-dependent and -independent cells, in spleen and bone marrow of wild-type mice. Moreover, on immunoscope analysis, the TCR β-chains expressed by both DN NKT cell subgroups displayed Gaussian-like distributions with respect to the CDR3 length, which is indicative of a considerable DN NKT cell TCRβ diversity and of the absence of clonally expanded cells. These observations indicate that a wide range of TCR α- and β-chains allow the recognition of CD1d molecules and suggest that homeostasis is maintained, despite the activated phenotype of DN NKT cells but consistent with their being nonaggressive in the mouse, by inhibitory mechanisms of their effector functions (probably involving Ly-49 molecules).

The tissue heterogeneity of the DN NKT cells was further shown by the extensive sequencing of Vβ8.2-Jβ2.5 rearrangements of thymic, hepatic, splenic, and bone marrow populations purified from an individual mouse, because we found that each tissue-derived population shared only a few nucleotide and amino acid sequences with the others. This observation is consistent with two non-mutually exclusive hypotheses: either a tissue-specific selection of DN NKT cell TCR occurs; or DN NKT cells are subject to continuous fluxes between tissues. Previous work reported that when challenged with distinct CD1d+ cell types, CD1d-restricted NKT cell hybridomas with different TCRs display their own autoreactivity patterns independently of the level of CD1d expression, even when they share same α-chain and Vβ segment (21, 22), and that the reactivity of one of two NKT cell hybrids bearing the Vα14 invariant chain but different Vβ-chains requires CD1d trafficking to endosomal compartments (22). Moreover, the composition of hybridoma TCR chains appeared to be critical in the recognition of α-galactosylceramide (14), a synthetic glycolipid capable of eliciting in vivo and in vitro NKT cell responses (11, 14, 34, 35, 36) and a recent study suggested that oligoclonal expanded NKT cells can discriminate between distinct glycosylphosphatidylinositol Ags (15). These studies, which emphasize on one hand the NKT cell reactivity to endogenous and cell type specific-ligands and on the other hand the role of NKT cell TCR in the recognition of glycolipid Ags, argue in favor of a TCR tissue-specific selection.

Vβ8.2Jβ2.5 rearrangements are predominantly used by subgroup I cells and consistent with the proportion of this subgroup in the various tissues, the highest occurrence of common sequences was observed among thymic, hepatic, and splenic DN NKT cells. Moreover, it is a very unlikely event to find T cells with identical nucleotide TCRβ rearrangement and which developed from different precursors. Therefore, because thymus comprises almost exclusively subgroup I cells, the nucleotide recurrence of the latter populations show that at least some subgroup I DN NKT cells share a common precursor.

The present report thus demonstrate: 1) that DN NKT cells comprise two subgroups of cells: subgroup I, which expresses the invariant Vα14 chain in association with Vβ2, Vβ7, Vβ8.1, Vβ8.2, and Vβ8.3 chains, and subgroup II, which express other α-chains paired with unskewed β-chains; 2) that the proportion of subgroup I and subgroup II DN NKT cells is about 100:0 in thymus, 80:20 in liver, 50:50 in spleen, and 20:80 in bone marrow; 3) that in all tissues, both subgroups are polyclonal; and 4) that DN NKT cells from each tissue express a large individual TCR Vβ8.2Jβ2.5 sequence diversity.

During the past years, many studies have focused on NKT cell involvement, mainly through their cytokine production, in rejection of tumors (39, 40), autoimmune diseases (41, 42, 43), and infectious processes (15, 44, 45). Bone marrow DN NKT cells have been recently implicated in the IL4-mediated suppression of acute graft vs host disease (46). Another study reports that insulin-dependent diabetes mellitus in nonobese diabetic mice is in part related to a dysfunction of IL12-mediated IFN-γ secretion by TCR-stimulated NKT cells and that cytokine production by TCR-stimulated NKT cells of distinct tissue localization is heterogeneous (42). Because CD4 vs DN NKT cell proportions and subgroup I vs subgroup II DN NKT cell proportions vary depending on the tissue, it will be of interest to determine the cytokines produced by these respective NKT cell subpopulations on TCR engagement and to determine whether all these NKT cell subpopulations are identically implicated in the various immune responses. Lastly, almost complete disappearance of CD4+ NKT cells was observed in liver of LFA-1-deficient mice (47), the residual NKT cells being thus likely of DN phenotype (48). It is intriguing that in the latter, residual NKT cells do not display the Ly-49 phenotype of liver but instead that of splenic wild-type NKT cells. Whether these cells belong to subgroup I and/or subgroup II DN NKT cells is an interesting question to be addressed. Overall, this report and those of others contribute to a better definition of NKT cell subpopulations and thus should help to unravel their respective functions in homeostasis and immunopathology.

Table 3AA.

(continues).

CDR3 SizeRecurrent CDR3 Nucleotide SequencesRecurrent CDR3 Amino Acid Sequences
OccurrenceaSequenceOccurrenceaVβ8.2CDR3Jβ2.5
 Thymus Bone marrow  Thymus Bone marrow    
         
0/1 0/0  0/1 0/0    
0/3 0/1  0/3 0/1    
0/3 0/1  0/3 0/1    
0/19 0/16  2/19 2/16 CAS GDWGQDTQ YFGb 
0/46 0/28  0/44d 0/28    
10 0/40 0/19  1/40 3/19 CAS GETGGAQDTQ YFGb 
11 0/21 0/8  0/20d 0/8    
12 0/12 0/11  0/12 0/11    
13 0/0 0/0  0/0 0/0    
14 0/0 0/0  0/0 0/0    
CDR3 SizeRecurrent CDR3 Nucleotide SequencesRecurrent CDR3 Amino Acid Sequences
OccurrenceaSequenceOccurrenceaVβ8.2CDR3Jβ2.5
 Thymus Bone marrow  Thymus Bone marrow    
         
0/1 0/0  0/1 0/0    
0/3 0/1  0/3 0/1    
0/3 0/1  0/3 0/1    
0/19 0/16  2/19 2/16 CAS GDWGQDTQ YFGb 
0/46 0/28  0/44d 0/28    
10 0/40 0/19  1/40 3/19 CAS GETGGAQDTQ YFGb 
11 0/21 0/8  0/20d 0/8    
12 0/12 0/11  0/12 0/11    
13 0/0 0/0  0/0 0/0    
14 0/0 0/0  0/0 0/0    
LiverBone marrowLiverBone marrow
0 /1 0 /0  0 /1 0 /0    
0 /5 0 /1  0 /5 0 /1    
0 /11 0 /1  0 /11 0 /1    
0 /24 0 /16  2 /24 4 /16 CAS GDGTGDTQ YFGb 
0 /69 0 /28  1 /67d 1 /28 CAS GDSGDQDTQ YFGb 
10 0 /37 0 /19  0 /37 0 /19    
11 0 /22 0 /8  0 /21d 0 /8    
12 2 /9 8 /11 TGTGCCAGCGGTGATGGACTGGGGGGGCGCGACCAAGAC 2 /9 8 /11 CAS GDGLGGRDQDTQ YFG 
13 0 /4 0 /0  0 /4 0 /0    
14 0 /0 0 /0  0 /0 0 /0    
LiverBone marrowLiverBone marrow
0 /1 0 /0  0 /1 0 /0    
0 /5 0 /1  0 /5 0 /1    
0 /11 0 /1  0 /11 0 /1    
0 /24 0 /16  2 /24 4 /16 CAS GDGTGDTQ YFGb 
0 /69 0 /28  1 /67d 1 /28 CAS GDSGDQDTQ YFGb 
10 0 /37 0 /19  0 /37 0 /19    
11 0 /22 0 /8  0 /21d 0 /8    
12 2 /9 8 /11 TGTGCCAGCGGTGATGGACTGGGGGGGCGCGACCAAGAC 2 /9 8 /11 CAS GDGLGGRDQDTQ YFG 
13 0 /4 0 /0  0 /4 0 /0    
14 0 /0 0 /0  0 /0 0 /0    
SpleenBone marrowSpleenBone marrow
0 /1 0 /0  0 /1 0 /0    
0 /4 0 /1  0 /2d 0 /1    
0 /10 0 /1  0 /10 0 /1    
0 /33 0 /16  2 /32d 3 /16 CAS GDWGEDTQ YFGb 
0 /79 0 /28  1 /77d 1 /28 CAS GDWGDQDTQ YFGb 
    1 /77d 1 /28 CAS GDWGNQDTQ YFGb 
    1 /77d 1 /28 CAS GDAWGQDTQ YFGb 
    1 /77d 1 /28 CAS GELGAQDTQ YFGb 
10 4 /64 5 /19 TGTGCCAGCGGTGATGCCTGGGGGGACCAAGAC 4 /63d 5 /19 CAS GDAWGDQDTQ YFG 
    1 /63d 2 /19 CAS GDAPGVQDTQ YFGb 
11 0 /36 0 /8  0 /36 0 /8    
12 0 /16 0 /11  0 /16 0 /11    
13 0 /4 0 /0  0 /4 0 /0    
14 0 /1 0 /0  0 /1 0 /0    
SpleenBone marrowSpleenBone marrow
0 /1 0 /0  0 /1 0 /0    
0 /4 0 /1  0 /2d 0 /1    
0 /10 0 /1  0 /10 0 /1    
0 /33 0 /16  2 /32d 3 /16 CAS GDWGEDTQ YFGb 
0 /79 0 /28  1 /77d 1 /28 CAS GDWGDQDTQ YFGb 
    1 /77d 1 /28 CAS GDWGNQDTQ YFGb 
    1 /77d 1 /28 CAS GDAWGQDTQ YFGb 
    1 /77d 1 /28 CAS GELGAQDTQ YFGb 
10 4 /64 5 /19 TGTGCCAGCGGTGATGCCTGGGGGGACCAAGAC 4 /63d 5 /19 CAS GDAWGDQDTQ YFG 
    1 /63d 2 /19 CAS GDAPGVQDTQ YFGb 
11 0 /36 0 /8  0 /36 0 /8    
12 0 /16 0 /11  0 /16 0 /11    
13 0 /4 0 /0  0 /4 0 /0    
14 0 /1 0 /0  0 /1 0 /0    
a

Sequence occurrence/total distinct sequences.

b

Recurrent amino acid sequence, encoded in both organs by distinct nucleotide sequences.

c

Amino acid sequence, common to thymic, liver, and splenic DN NKT cells.

d

Presence of amino acid sequences, encoded by distinct nucleotide sequences.

We thank D. Laouini for intensive discussions; I. Motta, J. Kanellopoulos, and M. Weksler for critical reading of the manuscript; A. Lim for technical advice; S. Dalle for constructing the CD3ε plasmid; and M.-C. Gendron for the four-color cell sorting.

1

This work was supported by the Institut National pour la Santé et la Recherche Médicale, the “Axe Immunologie des Tumeurs” of La Ligue Nationale Contre le Cancer, and the European Community. I.A. is a recipient of a fellowship from l’Association pour la Recherche sur le Cancer.

3

Abbreviations used in this paper: DN, double negative; CDR3, complementarity-determining region 3; MLE, maximum likelihood estimate; FAM, 6-carboxyfluorescein; int, intermediate.

1
Bendelac, A., N. Killeen, D. R. Littman, R. H. Schwartz.
1994
. A subset of CD4+ thymocytes selected by MHC class I molecules.
Science
263
:
1774
2
Lantz, O., A. Bendelac.
1994
. An invariant T cell receptor α chain is used by a unique subset of major histocompatibility complex class I-specific CD4+ and CD48 T cells in mice and humans.
J. Exp. Med.
180
:
1097
3
Ohteki, T., H. R. MacDonald.
1994
. Major histocompatibility complex class I related molecules control the development of CD4+CD8 and CD4CD8 subsets of natural killer 1.1+ T cell receptor-αβ+ cells in the liver of mice.
J. Exp. Med.
180
:
699
4
Bendelac, A., M. N. Rivera, S. H. Park, J. H. Roark.
1997
. Mouse CD1-specific NK1 T cells: development, specificity, and function.
Annu. Rev. Immunol.
15
:
535
5
MacDonald, H. R., R. K. Lees, W. Held.
1998
. Developmentally regulated extension of Ly-49 receptor expression permits maturation and selection of NK1.1+ T cells.
J. Exp. Med.
187
:
2109
6
Arase, H., N. Arase, K. Ogasawara, R. A. Good, K. Onoe.
1992
. An NK1.1+ CD4+8 single-positive thymocyte subpopulation that expresses a highly skewed T-cell antigen receptor Vβ family.
Proc. Natl. Acad. Sci. USA
89
:
6506
7
Makino, Y., R. Kanno, T. Ito, K. Higashino, M. Taniguchi.
1995
. Predominant expression of invariant Vα14+ TCR α chain in NK1.1+ T cell populations.
Int. Immunol.
7
:
1157
8
Shimamura, M., T. Ohteki, U. Beutner, H. R. MacDonald.
1997
. Lack of directed Vα14-Jα281 rearrangements in NK1+ T cells.
Eur. J. Immunol.
27
:
1576
9
Bendelac, A., O. Lantz, M. E. Quimby, J. W. Yewdell, J. R. Bennink, R. R. Brutkiewicz.
1995
. CD1 recognition by mouse NK1+ T lymphocytes.
Science
268
:
863
10
Bendelac, A..
1995
. Positive selection of mouse NK1+ T cells by CD1-expressing cortical thymocytes.
J. Exp. Med.
182
:
2091
11
Kawano, T., J. Cui, Y. Koezuka, I. Toura, Y. Kaneko, K. Motoki, H. Ueno, R. Nakagawa, H. Sato, E. Kondo, H. Koseki, M. Taniguchi.
1997
. CD1d-restricted and TCR-mediated activation of Vα14 NKT cells by glycosylceramides.
Science
278
:
1626
12
Brossay, L., M. Chioda, N. Burdin, Y. Koezuka, G. Casorati, P. Dellabona, 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
13
Brossay, L., O. Naidenko, N. Burdin, J. Matsuda, T. Sakai, M. Kronenberg.
1998
. Structural requirements for galactosylceramide recognition by CD1-restricted NK T cells.
J. Immunol.
161
:
5124
14
Burdin, N., L. Brossay, Y. Koezuka, S. T. Smiley, M. J. Grusby, M. Gui, M. Taniguchi, K. Hayakawa, M. Kronenberg.
1998
. Selective ability of mouse CD1 to present glycolipids: α-galactosylceramide specifically stimulates Vα14+ NK T lymphocytes.
J. Immunol.
161
:
3271
15
Schofield, L., M. J. McConville, D. Hansen, A. S. Campbell, B. Fraser-Reid, M. J. Grusby, S. D. Tachado.
1999
. CD1d-restricted immunoglobulin G formation to GPI-anchored antigens mediated by NKT cells.
Science
283
:
225
16
Sakai, T., O. V. Naidenko, H. Iijima, M. Kronenberg, Y. Koezuka.
1999
. Syntheses of biotinylated α-galactosylceramides and their effects on the immune system and CD1 molecules.
J. Med. Chem.
42
:
1836
17
Naidenko, O. V., J. K. Maher, W. A. Ernst, T. Sakai, R. L. Modlin, M. Kronenberg.
1999
. Binding and antigen presentation of ceramide-containing glycolipids by soluble mouse and human CD1d molecules.
J. Exp. Med.
190
:
1069
18
Eberl, G., R. Lees, S. T. Smiley, M. Taniguchi, M. J. Grusby, H. R. MacDonald.
1999
. Tissue-specific segregation of CD1d-dependent and CD1d-independent NK T cells.
J. Immunol.
162
:
6410
19
Hammond, K. J. L., S. B. Pelikan, N. Y. Crowe, E. Randle-Barrett, T. Nakayama, M. Taniguchi, M. J. Smyth, I. R. van Driel, R. Scollay, A. G. Baxter, et al
1999
. NKT cells are phenotypically and functionally diverse.
Eur. J. Immunol.
29
:
3768
20
Zeng, D. F., G. Gazit, S. Dejbakhsh-Jones, S. P. Balk, S. Snapper, M. Taniguchi, S. Strober.
1999
. Heterogeneity of NK1.1(+) T cells in the bone marrow: divergence from the thymus.
J. Immunol.
163
:
5338
21
Park, S. H., J. H. Roark, A. Bendelac.
1998
. Tissue-specific recognition of mouse CD1 molecules.
J. Immunol.
160
:
3128
22
Brossay, L., S. Tangri, M. Bix, S. Cardell, R. Locksley, M. Kronenberg.
1998
. Mouse CD1-autoreactive T cells have diverse patterns of reactivity to CD1+ targets.
J. Immunol.
160
:
3681
23
Eberl, G., H. R. MacDonald.
1998
. Rapid death and regeneration of NKT cells in anti-CD3ε- or IL-12-treated mice: a major role for bone marrow in NKT cell homeostasis.
Immunity
9
:
345
24
Pannetier, C., M. Cochet, S. Darche, S. Casrouge, M. Zšller, P. Kourilsky.
1993
. The sizes of the CDR3 hypervariable regions of the murine T-cell receptor β chain vary as a function of the recombined germ-line segments.
Proc. Natl. Acad. Sci. USA
90
:
4319
25
Pannetier, C., J. Even, P. Kourilsky.
1995
. T-cell repertoire diversity and clonal expansions in normal and clinical samples.
Immunol. Today
16
:
176
26
Casanova, J. L., P. Romero, C. Widmann, P. Kourilsky, J. L. Maryanski.
1991
. T cell receptor genes in a series of class I major histocompatibility complex-restricted cytotoxic T lymphocyte clones specific for a Plasmodiumberghei nonapeptide: implications for T cell allelic exclusion and antigen-specific repertoire.
J. Exp. Med.
174
:
1371
27
Pannetier, C., J.-P. Levraud, A. Lim, J. Even, and P. Kourilsky. 1997. The immunoscope approach for the analysis of T cell repertoires. iN The T-Cell Receptor: Selected Protocols and Applications. J. R. Okseznberg, L. Wang, and J.-Y. Y. Jeffery, eds. Chapman and Hall, New York, p. 287.
28
Gapin, L., Y. Fukui, J. Kanellopoulos, T. Sano, A. Casrouge, V. Malier, E. Beaudoing, D. Gautheret, J. M. Claverie, T. Sasazuki, et al
1998
. Quantitative analysis of the T cell repertoire selected by a single peptide-major histocompatibility complex.
J. Exp. Med.
187
:
1871
29
Barth, R. K., B. S. Kim, N. C. Lan, T. Hunkapiller, N. Sobieck, A. Winoto, H. Gershenfeld, C. Okada, D. Hansburg, I. L. Weissman, et al
1985
. The murine T-cell receptor uses a limited repertoire of expressed Vβ gene segments.
Nature
316
:
517
30
Behlke, M. A., D. G. Spinella, H. S. Chou, W. Sha, D. L. Hartl, D. Y. Loh.
1985
. T-cell receptor β-chain expression: dependence on relatively few variable region genes.
Science
229
:
566
31
Casrouge, A., E. Beaudoing, S. Dalle, C. Pannetier, J. Kanellopoulos, and P. Kourilsky. Size estimate of the αβ TCR repertoire of naive mouse splenocytes. J. Immunol. 165:5782.
32
Bousso, P., A. Casrouge, J. D. Altman, M. Haury, J. Kanellopoulos, J. P. Abastado, P. Kourilsky.
1998
. Individual variations in the murine T cell response to a specific peptide reflect variability in naive repertoires.
Immunity
9
:
169
33
Oehen, S., K. Brduscha-Riem.
1998
. Differentiation of naive CTL to effector and memory CTL: correlation of effector function with phenotype and cell division.
J. Immunol.
161
:
5338
34
Singh, N., S. Hong, D. C. Scherer, I. Serizawa, N. Burdin, M. Kronenberg, Y. Koezuka, L. Van Kaer.
1999
. Cutting edge: activation of NK T cells by CD1d and α-galactosylceramide directs conventional T cells to the acquisition of a Th2 phenotype.
J. Immunol.
163
:
2373
35
Burdin, N., L. Brossay, M. Kronenberg.
1999
. Immunization with α-galactosylceramide polarizes CD1-reactive NK T cells towards Th2 cytokine synthesis.
Eur. J. Immunol.
29
:
2014
36
Toura, I., T. Kawano, Y. Akutsu, T. Nakayama, T. Ochiai, M. Taniguchi.
1999
. Inhibition of experimental tumor metastasis by dendritic cells pulsed with α-galactosylceramide.
J. Immunol.
163
:
2387
37
Kitamura, H., K. Iwakabe, T. Yahata, S. Nishimura, A. Ohta, Y. Ohmi, M. Sato, K. Takeda, K. Okumura, L. van Kaer, et al
1999
. The natural killer T (NKT) cell ligand α-galactosylceramide demonstrates its immunopotentiating effect by inducing interleukin IL-12 production by dendritic cells and IL-12 receptor expression on NKT cells.
J. Exp. Med.
189
:
1121
38
Kawamura, T., K. Takeda, S. K. Mendiratta, H. Kawamura, L. Van Kaer, H. Yagita, T. Abo, K. Okumura.
1998
. Critical role of NK1+ T cells in IL-12-induced immune responses in vivo.
J. Immunol.
160
:
16
39
Cui, J., T. Shin, T. Kawano, H. Sato, E. Kondo, I. Toura, Y. Kaneko, H. Koseki, M. Kanno, M. Taniguchi.
1997
. Requirement for Vα14 NKT cells in IL-12-mediated rejection of tumors.
Science
278
:
1623
40
Kawano, T., J. Cui, Y. Koezuka, I. Toura, Y. Kaneko, H. Sato, E. Kondo, M. Harada, H. Koseki, T. Nakayama, et al
1998
. Natural killer like non specific tumor cell lysis mediated by specific ligand activated Vα14 NKT cells.
Proc. Natl. Acad. Sci. USA
95
:
5690
41
Hammond, K. J. L., L. D. Poulton, L. J. Palmisano, P. A. Silveira, D. I. Godfrey, A. G. Baxter.
1998
. αβ-T cell receptor TCR+ CD4CD8 (NKT) thymocytes prevent insulin-dependent diabetes mellitus in nonobese diabetic NOD/Lt mice by the influence of interleukin IL-4 and/or IL-10.
J. Exp. Med.
187
:
1047
42
Falcone, M., B. Yeung, L. Tucker, E. Rodriguez, N. Sarvetnick.
1999
. A defect in interleukin 12-induced activation and interferon γ secretion of peripheral natural killer T cells in nonobese diabetic mice suggest new pathogenic mechanisms for insulin-dependent diabetes mellitus.
J. Exp. Med.
190
:
963
43
Mieza, M. A., T. Itoh, J.Q. Cui, Y. Makino, T. Kawano, K. Tsuchida, T. Koike, T. Shirai, H. Yagita, A. Matsuzawa, et al
1996
. Selective reduction of Vα14+ NK T cells associated with disease development in autoimmune-prone mice.
J. Immunol.
156
:
4035
44
Flesch, I. E. A., A. Wandersee, S. H. E. Kaufmann.
1997
. IL-4 secretion by CD4+NK1+ T cells induces monocyte chemoattractant protein-1 in early listeriosis.
J. Immunol.
159
:
7
45
Emoto, M., Y. Emoto, I. B. Buchwalow, S. H. E. Kaufmann.
1999
. Induction of IFN-γ-producing CD4+ natural killer T cells by Mycobacterium bovis bacillus Calmette-Guérin.
Eur. J. Immunol.
29
:
650
46
Zeng, D. F., D. Lewis, S. Dejbakhsh-Jones, F. S. Lan, M. Garcia-Ojeda, R. Sibley, S. Strober.
1999
. Bone marrow NK1.1 and NK1.1+ T cells reciprocally regulate acute graft versus host disease.
J. Exp. Med.
189
:
1073
47
Emoto, M., H. W. Mittrucker, R. Schmits, T. W. Mak, S. H. E. Kaufmann.
1999
. Critical role of leukocyte function-associated antigen-1 in liver accumulation of CD4+NKT cells.
J. Immunol.
162
:
5094
48
Ohteki, T., C. Maki, S. Koyasu, T. W. Mak, P. S. Ohashi.
1999
. LFA-1 is required for liver NK1.1+TCRαβ+ cell development: evidence that liver NK1.1+TCRαβ+ cells originate from multiple pathways.
J. Immunol.
162
:
3753