DX5/CD49b-Positive T Cells Are Not Synonymous with CD1d-Dependent NKT Cells

Mice

C57BL/6, C57BL/6 CD1d^−/−, and C57BL/6 Jα18^−/− mice were housed in the Department of Microbiology and Immunology, University of Melbourne, Animal House. CD1d^−/− mice were originally provided by L. van Kaer (Vanderbilt University School of Medicine, Nashville, TN), and Jα18^−/− mice were originally provided by M. Taniguchi (Chiba University, Tokyo, Japan), and each mutation has been backcrossed to C57BL/6 for at least 10 generations. Mice used for ontogeny experiments ranged from 5 to 90 days, with the day of birth referred to as day 1. All other mice used were 5–7 wk of age and were sex matched within each experiment. The University of Melbourne Animal Ethics Committee approved all mouse experiments.

Cell suspensions

Cell suspensions of thymus, spleen, and peripheral lymph nodes (PLNs) were prepared by gently grinding the organs between the frosted ends of glass microscope slides in PBS containing 2% FCS and 0.02% azide. Bone marrow cells were harvested from femurs by flushing with PBS-FCS-azide. Spleen cells were subsequently treated with red cell lysis buffer (Sigma-Aldrich). Hepatic leukocytes were isolated by cutting individual livers into small pieces and gently pressing through 200-gauge wire mesh. The cells were washed twice in ice-cold PBS with 2% FCS and 0.02% azide and spun through 33.8% Percoll (Amersham Biosciences) for 12 min at 693 × g. Recovered leukocytes were washed and treated with red cell lysis buffer (Sigma-Aldrich).

Culture medium

Culture medium contained RPMI 1640 (Invitrogen Life Technologies) supplemented with 10% FCS (JRH Biosciences), 100 U/ml penicillin (Invitrogen Life Technologies), 100 μg/ml streptomycin (Invitrogen Life Technologies), 2 mM glutamax (Invitrogen Life Technologies), 1 mM sodium pyruvate (Invitrogen Life Technologies), 5 × 10⁻⁵ M 2-ME (Sigma-Aldrich), 0.1 mM nonessential amino acids (Invitrogen Life Technologies), and 15 mM HEPES buffer (Invitrogen Life Technologies).

Abs and flow cytometry

The following reagents were used in multiparameter flow cytometric analysis: anti-: CD4 FITC or biotin (clone RM4-5), CD8 PerCP (clone 53-6.7), CD49b FITC (clone Ha1/29), CD49b PE (clone HMα2), CD49b/Pan-NK cells biotin, FITC or PE (clone DX5), CD62L FITC (clone MEL-14), IFN-γ PE (clone XMG1.2, rat IgG1), IL-4 PE (clone 11B11, rat IgG1), NK1.1 biotin (clone PK136), rat IgG1 isotype control PE (clone R3-34), and βTCR allophycocyanin (clone H57-597). All mAb were purchased from BD Pharmingen. Biotinylated mAb were detected with streptavidin Alexa Fluor 488 (Molecular Probes). Mouse CD1d tetramer loaded with α-galactosylceramide was produced in house, as previously described (31), using recombinant baculovirus encoding his-tagged mouse CD1d and mouse β₂ microglobulin, provided by Prof. M. Kronenberg’s laboratory (La Jolla Institute for Allergy and Immunology, San Diego, CA). FcR block (anti-CD16/CD32, clone 2.4G2) was added to all surface-staining mixtures. One fluorescence detection channel was usually left available for the exclusion of autofluorescent cells. Flow cytometry was performed using a FACSCalibur or LSR-II (BD Biosciences) and analyzed using CellQuest software (BD Biosciences).

Intracellular cytokine staining

Following isolation from the thymus and spleen, lymphocytes (3 × 10⁶ cells/well) were incubated at 37°C in the presence or absence of 10 ng/ml PMA (Sigma-Aldrich) and 4 μM ionomycin (Sigma-Aldrich) for 2 h. All cultures contained 10 μg/ml brefeldin A (Sigma-Aldrich) to prevent secretion of intracellular cytokines. Cultured lymphocytes were labeled with cell surface mAb and CD1d tetramer and washed with PBS before being fixed with 0.5% paraformaldehyde (BDH Chemicals) in PBS in the dark for 45 min at room temperature. Cells were washed in PBS and then incubated with intracellular cytokines in the presence of PBS containing 5% FCS and 0.05% saponin (Sigma-Aldrich) in the dark for 1 h at room temperature. Fixed and permeabilized cells were stained with PE-conjugated anti-IFN-γ (clone XMG1.2), IL-4 (clone 11B11), or rat IgG1 as an isotype control (clone R3-34).

Cytokine analysis by sandwich ELISA

Cell subsets were sorted using either a MoFlo sorter (DakoCytomation) or a FACSaria (BD Biosciences) and stimulated in vitro with plate-bound anti-CD3 with or without anti-CD28. Culture supernatants were harvested at 24–48 h and screened by sandwich ELISA. IL-2 was detected using a commercially available mouse IL-2 sandwich ELISA kit (BD Pharmingen). IL-4 and IFN-γ were detected by a sandwich ELISA, as described previously (24). The substrate used for the detection of all cytokines was tetramethylbenzidine dihydrochloride (Sigma-Aldrich), and the reaction stopped using 2M H₂SO₄. ELISA readings were determined by OD scanning at 450 nm using a Labsystems Multiskan (Multisoft). Generally, the limit of detection for IL-2, IL-4, and IFN-γ was 7 pg/ml, 0.7 U/ml, and 30 pg/ml, respectively. ELISA readings below the limits of detection were designated not detected.

Single cell RT-PCR analysis for cytokine production

Lymphocytes were isolated with a MoFlo sorter (DakoCytomation) fitted with a Cyclone single cell deposition unit. Cells were positively gated for CD49b and αβTCR expression, and either CD1d tetramer⁺ or tetramer⁻ cells were sorted directly into a 96-well PCR plate (Brinkman Instruments) containing 5 ml of cDNA reaction mix. The cDNA mix and reverse-transcription steps have been previously described (32). For the first round of RT-PCR, 2 ml of the cDNA was used to amplify larger (external) regions of either the Vα14Jα18 segment of NKT cell TCR (positive control for tetramer⁺ cells), or β-actin (positive control for tetramer⁻ cells) in a 25-μl amplification reaction. Another 2 μl was used to amplify cytokine (IFN-γ, TNF-α, TGF-β, IL-4, IL-13, IL-10) cDNA. For the second round of PCR, 2 μl from the first PCR was used as a template to amplify a smaller (internal) region. The first and second rounds of PCR were performed with 0.33 U of Taq polymerase/1.5 mM MgCl₂/0.2 mM dNTP (all from Invitrogen Life Technologies)/10 pmol each of sense and antisense primers. The primers used to amplify each region were: Vα14Jα18, Vα14 EXT(S) TGGGAGATACTCAGCAACTCTGG, Caa(AS) TGGCGTTGGTCTCTT TGAAG, Jα18 INT(S) ATAGAGGTTCAGCCTTAGG, Cαβ INT(AS) ACACAGCAGGTTCTGGGTTC. β-actin: EXT(S) GACCCTCAAGTAC CCCATTG, EXT(AS) GTAATCTCCTTCTGCATCCTG, INT(S) GG GACGACATGGAGAAGATC, INT(AS) AGGTCTTTACGGATGTCAA CG. IFN-γ: EXT(S) GGCTGTTTCTGGCTGTTACTG, EXT(AS) TTGTT ACTATAAATACTTCTTTGG, INT(S) TCTGGAGGAACTGGCAAAAG GATGG, INT(AS) CCCCACCCCGAATCAGCAGCGACTC. TNF-α: EX T(S) CGTGGAACTGGCAGAAGAGG, EXT(AS) GGGGCAGGGGCTCTTGACGGC, INT(S) CTTCCAGAACTCCAGGCGGTG, INT(AS) GCTGACGGTGTGGGTGAGGAG. TGF-β: EXT(S) GTGGCTTCTAGTGCTGACGC, EXT(AS) TCATGGATGGTGCCCAGGTC, INT(S) GGGACT CTCCACCTGCAAGA, INT(AS) CTTTGCTGTCACAAGAGCAG. IL-4: EXT(S) TCGTCACTGACGGCACAGA, EXT(AS) GTTCTTCTTCAAGCATGGAG, INT(S) CAGCTAGTTGTCATCCTGC, IN T(AS) GTGAGGACGTTTGGCACATC. IL-13: EXT(S) TCACTGGCTCTGGGCTTCAT, EXT(AS) GCAATATCCTCTGGGTCCTG, INT(S) GACTGCAGTCCTGGCTCTTG, INT(AS) GTCCTGTAGATGGCATTGCA. IL-10: EXT(S) GAGAGCTCCATCATGCCTGG, EXT(AS) CATCATG TATGCTTCTATGC, INT(S) CTATGCTGCCTGCTCTTACT, INT(AS) GCAGTTGATGAAGATGTCA.

The PCR conditions were 95°C for 5 min, followed by 40 cycles (external reaction) or 33 cycles (internal reaction) of 95°C for 30 s, 55°C for 45 s, and 72°C for 90 s, followed by 1 cycle of 95°C for 1 min, 55°C for 1 min, and 72°C for 7 min. PCR products were resolved on a 2% agarose gel. Samples that were not Vα14Jα18 positive or β-actin positive were excluded from further analysis. Bulk PMA/ionomycin-stimulated spleen cDNA was used as a positive control for all primer products.

A comparison of different DX5/CD49b reagents

One reason that has been proposed to explain the controversy of whether CD1d-dependent NKT cells express CD49b is that a range of Abs conjugated to different fluorochromes is available that vary in their sensitivity and ability to detect this Ag on NKT cells (22, 29). Therefore, we optimally titrated a range of anti-CD49b reagents and tested their ability to label NKT cells, defined by CD1d tetramer binding, which is the most specific NKT cell marker available (Fig. 1⇓). The only Abs to clearly stain NKT cells were the PE-conjugated reagents, with anti-CD49b PE giving the brightest and most complete labeling, while the FITC- and biotin-conjugated reagents showed much weaker staining of these cells, yet clearly labeled NK cells. Bimodal expression of CD49b can be seen on TCR-negative cells, presumably NK cells and particularly in liver, as has been previously reported (33). For each Ab tested, the absence of nonspecific background staining can be seen by the fact that 99% of thymocytes are negative for this reagent (Fig. 1⇓). This supports previous published data (22, 29) and explains why some reports, including our own, suggested that NKT cells are DX5^−/low, because these reports typically used the FITC-conjugated reagent (23, 24, 25, 26, 27). PE-conjugated anti-CD49b demonstrated that virtually all NKT cells are CD49b⁺ (Fig. 1⇓B). This was apparently due to increased sensitivity of the PE-conjugated reagents, because TCR⁻ cells, presumably NK cells, labeled more intensely with these reagents as well (Fig. 1⇓A). However, these data do not mean that DX5/CD49b⁺ T cells are synonymous with NKT cells. Considering the high percentage of T cells that express CD49b (Fig. 1⇓A), compared with the percentage of cells that typically label with CD1d tetramer (Fig. 1⇓B), these data supported earlier studies suggesting that DX5/CD49b⁺ T cells include at least some non-NKT cells (23, 24, 25, 26, 28).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Detection of CD49b expression varies with different mAb fluorochrome conjugates. Thymus, spleen, and liver lymphocytes were harvested from 5-wk-old C57BL/6 mice and stained for flow cytometric analysis. A, Each column shows the percentage of DX5⁺ or CD49b⁺ cells that are either positive or negative for αβTCR on whole lymphocytes. B, First column, Shows CD1d tetramer staining vs αβTCR on whole lymphocytes with the percentage indicated. The remaining columns show the percentage of DX5⁺ or CD49b⁺ NKT cells gated as shown in the first column. Each Ab was optimally titrated to give maximum staining intensity compared with background. Density and dot plots are representative of at least four mice.

CD1d-dependent NKT cells are a minor subset of CD49b⁺ T cells

To directly investigate the extent to which DX5/CD49b⁺ T cells are CD1d dependent, we examined these cells using the PE-conjugated DX5 and CD49b reagents, shown to clearly label NKT cells in Fig. 1⇑, in a range of different tissues from WT mice compared with CD1d^−/− mice (known to be NKT deficient) (Fig. 2⇓A). These data showed that DX5/CD49b⁺ T cells were clearly present in CD1d^−/− mice, and in fact, when compared with WT mice, they were present at similar frequencies in spleen and bone marrow, were higher in lymph node, but lower in thymus and liver in the absence of CD1d. TCR Jα18^−/− mice, which also specifically lack NKT cells, were also examined for DX5/CD49b⁺ T cells, with similar results to the CD1d^−/− mice (data not shown). As shown in Fig. 2⇓B, when CD49b⁺ T cells in WT mice were stained with CD1d tetramer vs NK1.1, most CD49b⁺ T cells in spleen and bone marrow, and many in thymus and liver, were not labeled with CD1d tetramer. This suggests that many, and in some tissues most, DX5/CD49b⁺ T cells are not CD1d-dependent NKT cells. It was also noteworthy that of the CD49b⁺ T cells in WT mice, most did not label with NK1.1, while conversely, of the CD1d tetramer⁻ NK1.1⁺ T cells in WT mice, most (but not all) were CD49b⁺. Collectively, these results show that CD1d-dependent, Vα14Jα18 invariant NKT cells make up only a small subset of DX5/CD49b⁺ T cells in mice, and that furthermore, CD1d-independent T cells expressing NK cell markers such as NK1.1 and CD49b represent a heterogeneous population.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

DX5/CD49b⁺ T cells are present in CD1d^−/− mice. Thymus, spleen, liver, bone marrow, and PLN lymphocytes were harvested from 6-wk-old WT and CD1d^−/− mice and stained for flow cytometric analysis. A, Each tissue was examined for αβTCR staining vs either DX5 or CD49b. The bar graphs show accumulated data for WT and CD1d^−/− mice; the error bars represent SEM. B, CD49b⁺ αβTCR⁺ cells (left panels) or NK1.1⁺ αβTCR⁺ cells (right panels) from various tissues were examined for NK1.1 (left panels) or CD49b (right panels) expression vs CD1d tetramer binding. FACS plots are representative of between four and nine mice per group.

CD1d-independent CD49b⁺ T cells include subsets defined by CD4/CD8 expression

Analysis of CD49b⁺ T cells in WT and CD1d^−/− mice revealed that CD1d-independent CD49b⁺ T cells included CD4⁺, CD4⁻CD8⁻, and CD8⁺ subpopulations (Fig. 3⇓), while, in contrast, CD1d tetramer⁺ cells included CD4⁺ and CD4⁻CD8⁻ subsets, but not CD8⁺ cells, as expected from previous studies (25). The CD1d tetramer⁻ population of CD49b⁺ T cells in WT mice had a similar CD4/CD8 profile to CD49b⁺ T cells in CD1d^−/− mice, suggesting that these cells were similar between the two types of mice, regardless of whether CD1d-dependent NKT cells are present or not. This also reiterates the point made above, that the CD1d-independent CD49b⁺ T cells are a heterogeneous population.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Phenotype of CD49b⁺ T cells in WT and CD1d^−/− mice. Lymphocytes were harvested from thymus, spleen, liver, bone marrow, and PLN, and stained for flow cytometric analysis to examine CD4 vs CD8 expression on CD49b⁺ αβTCR⁺ cells, or subsets thereof, from tissues, as indicated. Top row, Shows results from WT mice; second row, shows results from CD1d^−/− mice. Third and fourth rows, Show CD4 vs CD8 on the CD1d tetramer⁺ and CD1d tetramer⁻ fractions, respectively, of CD49b⁺ T cells from WT mice. Dot plots are representatives of between four and nine mice per group.

CD1d-independent CD49b⁺ T cells have a distinct profile of activation/memory markers

A previous study (22) reported that NK1.1⁺ or DX5⁺ T cells in spleen showed a distinct activation/memory phenotype from their counterparts in thymus, including a major population in spleen that expressed high levels of CD62L and CD44, yet low levels of CD69, whereas NK1.1⁺ or DX5⁺ T cells in thymus were mostly CD62L^lowCD44^high and CD69⁺. Although this led them to suggest that spleen and thymus NKT cells differ in their activation/memory status, an alternative possibility was that the CD62L^high, CD69⁻ T cells in spleen were not NKT cells. We examined CD49b⁺ T cell populations from thymus, spleen, and liver, for expression of CD62L, CD69, CD25, and CD44, with the intention of comparing CD1d tetramer⁺ (NKT) and CD1d tetramer⁻ (non-NKT) subsets (Fig. 4⇓). The results indicated that CD1d-dependent NKT cells in thymus, spleen, and liver were identical for expression of these activation/memory markers, and that the CD1d tetramer⁻ fraction of DX5⁺ T cells exclusively accounted for the previously reported variability. This is consistent with other studies suggesting that NKT cells are typically CD69⁺ and CD62L^low (23, 28). Interestingly, a subset of the CD1d tetramer⁻ population labeled with anti-CD25. This subset was also CD4⁺, and ∼6% of CD4⁺CD25⁺ T regulatory cells in spleen were found to be CD49b⁺ and NK1.1⁻ (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Activation/Memory marker expression is similar on thymus, spleen, and liver CD49b⁺ αβTCR⁺ CD1d tetramer⁺ cells. Thymus, spleen, and liver lymphocytes were harvested from WT mice and stained for flow cytometric analysis. First column, Shows CD49b staining vs αβTCR on whole lymphocytes for the various tissues. The next columns show activation/memory marker expression vs CD1d tetramer reactivity on CD49b⁺ αβTCR⁺ cells gated, as shown in first column. Density and dot plots are representative of at least three mice per group.

Only the CD1d-dependent population of CD49b⁺ T cells produces high levels of IL-4

A hallmark of NKT cells is the potent and rapid production of both IL-4 and IFN-γ. Other studies have reported that NK1.1⁺ or CD49b⁺ T cells in spleen produced lower levels of these cytokines than their counterparts in thymus (22, 24). Considering the low frequency of CD1d-dependent NKT cells within the splenic NK1.1/CD49b⁺ T cell population, it was possible that the different cytokine production simply reflected the lower frequency of NKT cells within this compartment. We therefore conducted intracellular cytokine staining, following in vitro stimulation with PMA and ionomycin, to compare IFN-γ and IL-4 production by CD1d tetramer⁺ NKT cells with other T cell types, including CD49b⁺ CD1d tetramer⁻ T cells. These experiments were technically difficult due to the fact that, at least in our hands, anti-IL-4 PE is the only Ab to give clear intracellular IL-4 staining, yet anti-CD49b PE is the only Ab to clearly define T cell subsets, including both CD1d-dependent and CD1d-independent populations. Because at least some CD49b⁺ CD1d-independent T cells also express NK1.1, and most NK1.1⁺ CD1d-independent T cells express CD49b (Fig. 2⇑B), we substituted NK1.1 with CD49b. Thus, we used anti-NK1.1 biotin/streptavidin Alexa-488 in combination with anti-IL-4 PE or anti-IFN-γ PE and compared cytokine production by CD1d tetramer⁺ NK1.1⁺ T cells with CD1d tetramer⁻ NK1.1⁺ T cells. The results revealed that only CD1d tetramer⁺ fraction of CD49b⁺ T cells produced detectable levels of IL-4 following stimulation (Fig. 5⇓A), while both populations of NK1.1⁺ cells made IFN-γ, although this cytokine was produced by a higher frequency of NKT cells compared with the CD1d tetramer⁻ fraction under these stimulation conditions. This staining allowed a direct comparison of thymic and splenic CD1d-dependent NKT cells, which suggested that these populations make comparable levels of IL-4 and IFN-γ, at least by intracellular cytokine staining following PMA/ionomycin stimulation. To verify the results from the intracellular cytokine staining, and to extend the analysis of CD49b⁺ CD1d tetramer⁻ cells beyond the NK1.1⁺ fraction, CD49b⁺ CD1d tetramer⁺ and CD49b⁺ CD1d tetramer⁻ cells were FACS sorted from spleen and stimulated in vitro on plate-bound anti-CD3 with or without anti-CD28 costimulation (Fig. 5⇓B). These ELISA results supported the intracellular cytokine staining, showing that only CD1d tetramer⁺ fraction of CD49b⁺ T cells produces high levels of IL-4, while both populations produce IFN-γ; although somewhat unexpectedly, the CD1d tetramer⁻ fraction produced higher levels of this cytokine. Both populations produced IL-2, and CD28 enhanced production of this cytokine.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Intracellular cytokine staining of CD1d tetramer⁺ and CD1d tetramer⁻ NK1.1^+/− T cells. A, Lymphocytes from the thymus and spleen were harvested from WT mice and stimulated in vitro in the presence of PMA and ionomycin for 2 h. Cells were stained with cell surface markers, fixed, permeabilized, and then stained with cytokine-reactive mAb before examination by flow cytometric analysis. First column, Shows NK1.1 staining vs αβTCR on whole lymphocytes. The remaining columns show CD1d tetramer reactivity vs IFN-γ, IL-4, or isotype control on αβTCR⁺ NK1.1⁺ cells gated, as shown in first column. Density and dot plots are representative of at least four mice per group. B, CD49b⁺CD1d tetramer⁺ and CD49b⁺CD1d tetramer⁻ T cells were sorted from spleens of WT mice and cultured in 96-well plates coated with anti-CD3 with or without anti-CD28. Supernatants were collected after 1 and 2 days of culture and tested for cytokines using sandwich ELISA. □, Represents cytokine production by CD49b⁺CD1d tetramer⁺ NKT cells (n = 5–6). ▪, Represents cytokine production by CD49b⁺CDld tetramer⁻ T cells (n = 9). Data are represented as mean ± SE.

Both thymus and spleen NKT cells respond to costimulation

At least two studies, including one from our own laboratory (22, 24), have suggested that spleen NKT cells make less cytokine than thymus NKT cells when stimulated by plate-bound anti-CD3, suggesting an important functional distinction between these two populations. Furthermore, spleen NK1.1⁺ or DX5⁺ T cells appeared to require CD28 costimulation for optimal cytokine production, while those from thymus seemed to function independently of CD28 (22). A problem with both reports (22, 24) was the use of surrogate markers DX5/CD49b or NK1.1 to define and isolate NKT cells from thymus and spleen, which might explain the difference in cytokine production and CD28 responsiveness, because, as shown above, many more DX5/CD49b/NK1.1⁺ T cells were CD1d independent in spleen compared with thymus (Fig. 2⇑B). Therefore, CD1d tetramer⁺ NKT cells were specifically sorted from thymus and spleen, and their cytokine production in response to anti-CD3 with or without CD28 costimulation was tested by ELISA (Fig. 6⇓). Conventional CD1d tetramer⁻ T cells were also sorted and tested as a control population to show that CD28 costimulation was giving the expected results. Our results in part verify earlier findings, showing that thymus NKT cells (in this study defined by CD1d tetramer) were a more potent source of IL-4 and IFN-γ than spleen NKT cells when assayed by ELISA. However, spleen NKT cells were a richer source of IL-2 at 48 h (but not 24 h). The influence of CD28 on these subsets was dependent on the cytokine, and the time point, being tested. In agreement with the previous study (22), CD28 enhanced IL-4 and IFN-γ production from spleen, but not thymus NKT cells at 48 h. However, at the 24-h time point, both thymus and spleen NKT cells showed enhanced production of IFN-γ by anti-CD28 costimulation. Furthermore, CD28 dramatically boosted IL-2 production from both thymus and spleen NKT cells at the 48-h time point. Taken together, this supports and extends earlier studies, showing that thymus NKT cells are a more potent source of IL-4 and IFN-γ than spleen NKT cells, following in vitro stimulation; however, spleen NKT cells can produce higher levels of IL-2 than thymus NKT cells under these conditions. This emphasizes the fact that NKT cells in different tissues may have distinct functional roles in the immune system. Furthermore, and in contrast to the conclusion from an earlier study, both thymus and spleen NKT cells can respond to CD28 costimulation, although the outcome differs between these two populations of CD1d-dependent NKT cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Thymus and spleen NKT cells differ in cytokine production and response to CD28 costimulation. CD1d tetramer⁺ αβTCR⁺ cells, and CD1d tetramer⁻ αβTCR⁺ cells were sorted from thymus and spleen, and cultured in 96-well plates coated with anti-CD3 with or without anti-CD28, for the indicated time. IL-2, IL-4, and IFN-γ levels were determined by sandwich ELISA, and results plotted as mean ± SE for n = 4–5 samples per group. When comparing CD3 with CD3/CD28 stimulation for each group, significance is indicated as shown with the asterisk (∗, p ≤ 0.05, Mann-Whitney rank sum U test).

Single cell RT-PCR analysis of cytokine production by CD1d tetramer⁺ and CD1d tetramer⁻ DX5/CD49b⁺ T cells

As an additional assay for the cytokine-producing potential of CD49b⁺ T cells, single cell RT-PCR analysis was used, comparing CD1d tetramer⁺ NKT and CD1d tetramer⁻ subsets. Single cells from each population (30 tetramer⁻ and 24 tetramer⁺) were stimulated with PMA and ionomycin, and then tested for cytokine mRNA, including: IL-4, IFN-γ, IL-10, IL-13, TNF, and TGF-β (Fig. 7⇓). This supported and extended the above results, showing that both populations of CD49b⁺ T cells produced message for IFN-γ, and also TNF, at similar frequency. Only the NKT cells produced IL-4 mRNA, and IL-13 mRNA was also primarily associated with these cells. IL-10 mRNA was not detected in either subset, but was clearly detected in whole unfractionated spleen cells, while TGF-β was produced by a small proportion of cells in both CD49b⁺ T cell populations. Fig. 7⇓C shows the number of individual cells that display a given pattern of cytokine mRNA, as indicated by the table.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

CD49b⁺ αβTCR⁺ CD1d tetramer⁺ cells differ in cytokine mRNA production from their CD1d tetramer⁻ counterparts. Splenocytes were FACS sorted (>98% purity) on the basis of CD49b, αβTCR, and CD1d tetramer staining. CD49b⁺ αβTCR⁺ tetramer⁺ or tetramer⁻ populations were sorted and stimulated in vitro with PMA/ionomcyin for 2 h as a bulk population, and then resorted as single cells (n = 24 tetramer⁺ cells and n = 30 tetramer⁻ cells), and each cell was examined for cytokine mRNA using single cell RT-PCR analysis. A, An example of five single cell RT-PCR products from each group is shown. B, The proportion of single cells in tetramer⁺ and tetramer⁻ groups producing each type of cytokine mRNA. Contingency tables were formulated, and statistical analysis was performed using Fisher’s exact test. ∗∗∗, p < 0.0001; ∗, p < 0.05. C, Cytokine mRNA expression patterns for tetramer⁺ and tetramer⁻ cells are indicated by shaded boxes. First six columns, Each represents individual cytokines, and the numbers in the two right-hand columns represent a percentage of cells producing the cytokines in that row. The total number of cells examined is shown in the bottom row rather than the percentage.

CD49b does not distinguish different NKT cell developmental stages

An earlier study suggested that the most immature thymic NKT cells could be defined as DX5^−/low, became DX5⁺ as they matured, but then reverted to a DX5^−/low phenotype in older mice, such that by 10 wk of age, most NKT cells were DX5^−/low (29). We examined thymuses ranging in age from 5 days neonatal, which represents the earliest time point at which NKT cells have been detected (34), through to 13 wk when mice are fully mature. As we have previously shown, at the earliest time point stage, NKT cells are extremely rare, all are NK1.1⁻, and almost all CD4⁺ (34). In accordance with published data, we found that most (but not all) very immature, NK1.1⁻ NKT cells did not stain with DX5 PE, which was the same Ab as used in the previous study of NKT cell maturation (29). Staining levels increased as NKT cells matured and began to acquire NK1.1, but we were unable to verify the subsequent loss of this marker as mice reach full maturity, apart from a slight decrease in staining intensity (Fig. 8⇓). In contrast with the results using the DX5 Ab, we were unable to detect a clear relationship between NKT cell maturation and anti-CD49b PE staining, such that the majority of NKT cells were CD49b⁺ regardless of age (Fig. 8⇓). These data suggest that CD49b expression does not appear to be a robust marker of NKT cell maturation, and only some Abs might be able to distinguish between immature and mature NKT cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 8.

Phenotype of NKT cells during development. Thymocytes were harvested from C57BL/6 mice at various ages and stained for flow cytometric analysis. First column, Shows CD1d tetramer staining vs αβTCR on whole lymphocytes with the percentage indicated. Second column, Shows DX5 vs NK1.1 expression on NKT cells gated, as shown in first column. Third and fourth columns, Show CD49b vs NK1.1 and CD4 expression (respectively) on NKT cells gated, as shown in first column. FACS plots are representative of at least four mice per group.