We examined the expression of the H4 T cell activation marker in thymic T cell subpopulations and found that TCR-αβ+ CD4+ thymic T cells are segregated into three subpopulations based upon H4 levels. Thymic T cells with either no or low H4 expression differentiate via the mainstream differentiation pathway in the thymus. H4int thymic T cells, which express a skewed Vβ repertoire of Vβ2, -7, and -8 in their TCRs, show the phenotype of NKT cells: CD44high, Ly6Chigh, NK1.1+, and TCR-αβlow. H4high thymic T cells also show a skewed Vβ repertoire, Vβ2, -7, and -8, and predominantly express an invariant Vα14-Jα281+ α-chain in their TCRs but constitute a distinct population in that they are CD44int, Ly6C, NK1.1, and TCR-αβhigh. Thus, invariant Vα14+ thymic T cells consist of ordinary NKT cells and a new type of T cell population. Vβ7+ and Vβ8.1+ invariant Vα14+ thymic T cells are present in DBA/2 mice, which carry mammary tumor virus-7-encoded superantigens, in comparable levels to those in BALB/c mice. Furthermore, Vβ7+ invariant Vα14+ thymic T cells in DBA/2 mice are in the immunologically responsive state, and Yersinia pseudotuberculosis-derived mitogen-induced Vβ7+ invariant Vα14+ thymic T cell blasts from DBA/2 and BALB/c mice exhibited equally enhanced responses upon restimulation with Y. pseudotuberculosis-derived mitogen. Thus, invariant Vα14+ thymic T cells that escape negative selection in DBA/2 mice contain T cells as functionally mature as those in BALB/c mice.

CD4 single-positive (SP)3 and CD8 SP T cells differentiate from CD4+ CD8+ double-positive (DP) thymocytes via the CD4/CD8 pathway in the thymus (1, 2). Two selective events, elimination of self-reactive T cells (negative selection) and survival of T cells that can interact with foreign peptide/MHC complex (positive selection), occur at the stage of DP thymocytes (3, 4, 5). Negative selection is well documented in mice expressing endogenous superantigens that are encoded by mouse mammary tumor viruses (Mtvs) and that eliminate developing T cells in a Vβ-specific fashion (6, 7). Mtv-7-encoded superantigens, for instance, delete Vβ6+, Vβ7+, Vβ8.1+, and Vβ9+ T cells (8). Endogenous superantigens do not always delete all Vβ-specific thymocytes. Blackman and coworkers have previously reported that Mtv-7+ TCR β-chain transgenic mice show only partial deletion of thymocytes expressing the relevant TCR Vβ (9, 10), and that TCR Vα usage in T cells surviving the negative selection is biased in those mice (11). Thus, although TCR Vβ element plays the dominant role in T cell response to superantigens, accumulating evidence has shown that non-Vβ TCR element such as α-chain contributes to superantigen responsiveness (12, 13, 14) and influences the susceptibility to negative selection (10, 11, 12, 15). In addition to conventional T cells, a specialized population of TCR-αβ+ T cells that presumably do not differentiate via the CD4/CD8 pathway resides in the thymus as a small population (∼1%). They express NK surface receptors, and are hence called NKT cells, and a highly restricted TCR composed of an invariant α-chain (Vα14-Jα281 in mice) and a β-chain with restricted Vβ (Vβ2, -7, and -8 in mice) and heterogeneous Dβ and Jβ elements (16, 17).

Several studies have shown that Vβ8.1+ NKT cells are present in the thymus of Mtv-7+ mice at an equal level to Mtv-7 mice (18, 19). Thus, Vβ8.1+ NKT cells escape the negative selection by “self-Ags,” Mtv-7-encoded superantigens. However, it is not currently elucidated whether invariant TCR α-chain in NKT cells contributes to the escape of Vβ8.1+ NKT cells from the negative selection. Further questions arise as to whether Vβ7+ NKT cells also survive in Mtv-7+ thymus and whether Vβ7+ and Vβ8.1+ NKT cells in Mtv-7+ mouse are functionally mature, and how they are regulated to not respond harmfully to Mtv-7-encoded superantigens in vivo. Other contributors reported a T cell activation marker, H4, which comodulates and cocaps with TCR, indicating that the molecule physically associates with TCR (20). We became interested in whether the H4 molecule is expressed on NKT cells.

In the present study, we examined the expression of H4 molecule in thymic T cell subpopulations and found that thymic T cells with an intermediate amount of H4 expression are ordinary NKT cells, whereas thymic T cells with a high amount of H4 expression constitute a new type of invariant Vα14+ T cell. Vβ7+ and Vβ8.1+ invariant Vα14+ thymic T cells were present in Mtv-7+ DBA/2 mice at similar levels to Mtv-7 BALB/c mice. Furthermore, Vβ7+ invariant Vα14+ thymic T cells in DBA/2 mice showed substantial and comparable responses with those in BALB/c mice upon stimulation with a superantigen, Yersinia pseudotuberculosis-derived mitogen (YPM) (21, 22, 23). Based upon these results, we discuss the immune regulation of the potentially self-reactive T cells.

BALB/c and C57BL/6 mice were bred in our own colony at the Department of Microbiology and Immunology, Tokyo Women’s Medical University. DBA/2 mice were purchased from Japan S.L.C. (Hamamatsu, Japan). Female mice, 6–7 wk old, were used in this study.

The YPM superantigen was purified from extract of Escherichia coli XL1-Blue carrying pQE30-6xH·ypm (22) using Ni-NTA Agarose (Qiagen, Chatsworth, CA) followed by Sepharose Fast Flow (Pharmacia LKB Biotechnology, Tokyo, Japan). YPM stimulates murine Vβ7- and Vβ8-bearing T cells (23). The C398.4A mAb specific for H4 was produced in Armenian hamsters by immunizing with the murine T cell clone D10.G4.1 as described previously (20). mAbs to CD8 (83.12.5 and 53-6.7), Vβ2 (B20.6), Vβ4 (KT4-10), Vβ6 (RR4-7), Vβ7 (TR310), Vβ8.1+8.2+8.3 (F23.1), Vβ8.1+8.2 (KJ16), Vβ8.2 (F23.2), Vβ9 (MR10-2), Vβ10 (KT10b-2), Vβ14 (14.2), and Thy1.2 (HO13) were described previously (23, 24, 25). mAbs to TCR-αβ (H57-597) and CD44 (IM7), PE-anti-CD4 (RM4-4), cy-chrom-streptavidin and biotinylated anti-Ly6C (AL-21), and anti-NK1.1 (PK136) were purchased from PharMingen (San Diego, CA). FITC-anti-CD8 (53-6.7) and PE-streptavidin were obtained from Becton Dickinson (Mountain View, CA). The following anti-IgG Abs were used: FITC-goat anti-rat IgG from Zymed Laboratories (South San Francisco, CA), FITC-goat anti-mouse IgG from Biosource (Camarillo, CA), FITC-goat anti-hamster IgG from Southern Biotechnology Associates (Birmingham, AL), and goat anti-rat IgG from Organon Teknika (Westchester, PA). Purified hamster IgG and anti-H4 mAb were conjugated with biotin in our laboratory.

RPMI 1640 supplemented with 100 μg of streptomycin per ml, 100 U of penicillin per ml, 10% fetal bovine serum, and 5 × 10−5 M 2-ME was used for the culture.

To detect expression of H4 in thymic T cell subpopulations, thymocytes were stained by incubation with biotinylated anti-H4 mAb followed by a cocktail of FITC-anti-CD8, PE-anti-CD4, and cy-chrom-streptavidin. The relationship between the expression of H4 and various surface molecules was analyzed as follows. Thymocytes were stained with a combination of unconjugated Abs or anti-H4 mAb and appropriate FITC-anti-IgG and were incubated with an excess amount of normal hamster IgGs to block unbound binding sites on FITC-anti-IgG. After washing, FITC-stained cells were incubated with a combination of biotinylated anti-H4 mAb or biotinylated Abs and PE-streptavidin. To detect expression of H4 in CD4CD8 double-negative (DN) thymocytes, FITC-stained cells with anti-Vβ Abs and appropriate FITC-anti-IgG were incubated with biotinylated anti-H4 mAb followed by a combination of PE-anti-CD4 and cy-chrom-streptavidin. The purity of cells after the preparation was checked by incubation of cells with unconjugated anti-Vβ Abs or anti-H4 mAb followed by appropriate FITC-anti-IgG, or a combination of FITC-anti-CD8 and PE-anti-CD4. Samples of number of viable cells indicated were analyzed by an Epics CS flow cytometer (Coulter Immunology, Hialeah, FL), Epics XL flow cytometer (Coulter Immunology), or FACScan (Becton Dickinson, Cockeysville, MD).

Mice were treated with dexamethasone (Sigma, St. Louis, MO) as reported by Screpanti et al. (26). Various doses indicated in 40% ethanol/saline were injected s.c. Thymuses were taken 2 days later for the preparation of cells.

Single thymocyte suspensions were prepared in Hanks’ solution (Nissui Pharmaceutical, Tokyo, Japan) with 2% FCS. CD4 SP+DN thymocytes were obtained from BALB/c and C57BL/6 mice by two rounds of treatment of thymocytes with 83.12.5 and guinea pig complement (C′) (CD8+ cells, <0.2%). DBA/2 CD4 SP+DN thymocytes were obtained by two rounds of indirect immunomagnetic depletion according to the manufacturer’s instructions. Briefly, after thymocytes were incubated with 53-6.7, 53-6.7-coated cells were rosetted with sheep anti-rat IgG bound magnetic beads (Dynabeads, Dynal, Great Neck, NY) and separated with a magnet (CD8+ cells, <0.1%).

To obtain Vβ7+ invariant Vα14+ thymic T-enriched population, BALB/c and DBA/2 CD4 SP+DN thymocytes were obtained by one round of treatment with 83.12.5 and C′ and by taking nonadherent 53-6.7-treated cells from goat anti-rat IgG (30 μg/ml)-coated dishes (no. 25020; Corning, Corning, NY), respectively. Vβ7+ thymic T cells were then enriched by treatment with a cocktail of anti-Vβ mAbs (anti-Vβ4, -8, -10, and -14) and sheep anti-rat IgG-beads and sheep anti-mouse IgG-beads (Dynabeads, Dynal). These cells were further incubated with anti-H4 mAb followed by anti-mouse IgG-beads, and H4+ cells were positively selected with a magnet. After overnight culture, cells free from beads were separated with a magnet.

To obtain CD4+ Vβ7+ mainstream thymic T-enriched population, CD4 SP+DN thymocytes obtained from 500 μg of dexamethasone-treated BALB/c mice were incubated with a cocktail of anti-Vβ mAbs and anti-H4 mAb, and Ab-coated cells were depleted using anti-IgG-beads as described above. T-depleted spleen cells were obtained by treatment of spleen cells with anti-Thy1.2 and C′ followed by inactivation with mitomycin C (Kyowa Hakko Kogyo, Tokyo, Japan) and used as APCs.

Total mRNA was extracted by oligo(dT)-latex (Nippon Roche, Tokyo, Japan) from thymic T cell blasts and reverse transcribed into cDNA at 42°C for 2 h using RAV-2 reverse transcriptase (Takara Biomedicals, Osaka, Japan) and random hexamer primers (Takara Biomedicals). The cDNA was amplified on a thermocycler (programmable temp-control system PC-700, Sci-media, Tokyo, Japan) using Taq DNA polymerase (Takara Biomedicals) and oligonucleotides pairs specific for Vβ7 and Cβ or those for a panel of Vα and Cα primers (Vβ7, 5′-TACAGGGTCTCACGGAAGAAGCG-3′; Cβ, 5′-CTGCTCGGCCCCAGGCCTCT-3′; Vα2, 5′-AGCACTTTTAACTACTTCCCA-3′; Vα8, 5′-AATATCTCAACGAAGCCCCT-3′; Vα10, 5′-CGCAGCTCTTTGCACATTTC-3′; Vα11, 5′-GTTCTGCTCTGAGATGCAAT-3′; Vα14, 5′-AGTGTGACCCCCGACAAC-3′; Cα, 5′-TTGCTCTTGGAATCCATAGCT-3′). PCR amplification using Vβ7 and Cβ primers was performed for 28 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 1 min, followed by a final extension of 72°C for 10 min. The PCR condition using primers of a panel of Vα and Cα primers was 30 cycles of 94°C for 30 s, 62°C for 30 s, and 72°C for 1 min. PCR products were separated by electrophoresis on a 2.0% agarose gel and visualized by UV light after ethidium bromide staining.

DBA/2 and BALB/c CD4 SP+DN thymocytes were sorted into Vβ7+ H4high cells and Vβ7+ H4low cells, respectively, by an Epics CS flow cytometer. The sorted cells were stimulated with 3 μg/ml of YPM and syngenic APCs. After 3 days of culture, YPM-induced thymic T cell blasts were collected by applying the cell suspension to a Percoll gradient (density, 1.075) and expanded with 100 U/ml of human rIL-2 (a gift from Shionogi, Osaka, Japan) for an additional 2 days. First-strand cDNAs synthesized from mRNA from YPM-induced T cell blasts from Vβ7+ H4high thymocytes were amplified by PCR using Vα14 and Cα primers. The products were then ligated into pCR vector (Invitrogen, Carlsbad, CA) and were transfected into E. coli BL21 competent cells (Pharmacia Biotech, Uppsala, Sweden). After random selection of transformants, cloned plasmic DNAs were purified using Qiagen Plasmid Mini Kit (Qiagen). The products from the subsequent 25 cycles of PCR performed using cloned plasmic DNAs, Vα14 primers, and Dye Terminator Cycle Sequencing FC Ready Reaction Kit (PE Applied Biosystems, Foster City, CA) were sequenced by ABI PRISM 310 Genetic Analyzer (PE Applied Biosystems). According to the procedure described above, junctional sequences of Vβ7+ β-chain in Vβ7+ H4high and Vβ7+ H4low cells were obtained using Vβ7 and Cβ primers.

An indicated number of thymic T cell subpopulations were cultured with titrated amounts of YPM in the presence of 2 × 105 syngenic APCs in flat-bottom microtiter plate (3072 Falcon, Becton Dickinson Labware, Oxnard, CA) in culture medium. After 2.5 days of culture, culture medium was removed for the determination of lymphokine concentration. In parallel with these assays, an indicated number of thymic T cells was cultured with 3 μg/ml of YPM in the presence of 5 × 105 syngenic APCs in 48-well culture plate (3078 Falcon, Becton Dickinson Labware). After 3 days of culture, YPM-induced thymic T cell blasts were collected as described above and expanded with 100 U/ml of human rIL-2 for 4 more days. A total of 1 × 105 YPM-induced thymic T cell blasts were restimulated with YPM as described above. After 24 h of culture, culture supernatants were obtained for the determination of lymphokine concentration.

Concentration of IL-2 in culture supernatants was determined in a bioassay as proliferation of IL-2-dependent CTLL-2 cells as described previously (21). IL-4 in culture supernatants was quantitated by sandwich ELISA according to the manufacturer’s instructions (PharMingen). Anti-mouse IL-4 (11B11) used as coating mAbs and biotinylated anti-mouse IL-4 (BVD6-24G2) used as detecting mAbs were purchased from PharMingen. Standard curves were generated using mouse rIL-4 (PharMingen).

Statistical significance between any two groups was analyzed by Student’s t test.

BALB/c thymic T cell subpopulations defined by CD4/CD8 phenotypes were analyzed for the expression of H4 molecule (Fig. 1, A and B). H4 is expressed in ∼25% of CD4 SP, ∼55% of CD8 SP, and ∼10% of DN thymocytes, but not at all in DP thymocytes. Based upon the amount of H4 expressed, CD4 SP thymocytes could be segregated into three groups; none to low, intermediate, and high expression, suggesting the heterogeneity of CD4 SP thymocytes. In CD8 SP thymocytes, only low H4 expression was observed. DN thymocyte expression of H4 ranged from low to high.

FIGURE 1.

H4 expression in thymic T cell subpopulations. A, BALB/c thymocytes were stained with PE-anti-CD4, FITC-anti-CD8, and biotinylated anti-H4 mAb followed by cy-chrom-streptavidin. A sample of 300,000 viable cells was analyzed by FACScan. The histograms in B represent H4 expression (solid line) on CD4/CD8-defined subpopulations gated as indicated in A with background staining with biotinylated hamster IgG (dashed line). C, BALB/c CD4 SP+DN thymocytes were obtained and stained with a combination of anti-TCR-αβ antibody/FITC-anti-hamster IgG and biotinylated anti-H4 mAb/avidin-PE. Samples of 50,000 viable cells were analyzed by an Epics CS flow cytometer. Circles in the figure show H4low, H4int, and H4high thymic T cells defined with differential expression of H4 on TCR-αβ+ thymic T cells. Numbers indicate the percentages of these thymic T cell subpopulations in TCR-αβ+ CD4 SP+DN thymocytes that were calculated by subtracting background staining. Data show typical results from one of four experiments.

FIGURE 1.

H4 expression in thymic T cell subpopulations. A, BALB/c thymocytes were stained with PE-anti-CD4, FITC-anti-CD8, and biotinylated anti-H4 mAb followed by cy-chrom-streptavidin. A sample of 300,000 viable cells was analyzed by FACScan. The histograms in B represent H4 expression (solid line) on CD4/CD8-defined subpopulations gated as indicated in A with background staining with biotinylated hamster IgG (dashed line). C, BALB/c CD4 SP+DN thymocytes were obtained and stained with a combination of anti-TCR-αβ antibody/FITC-anti-hamster IgG and biotinylated anti-H4 mAb/avidin-PE. Samples of 50,000 viable cells were analyzed by an Epics CS flow cytometer. Circles in the figure show H4low, H4int, and H4high thymic T cells defined with differential expression of H4 on TCR-αβ+ thymic T cells. Numbers indicate the percentages of these thymic T cell subpopulations in TCR-αβ+ CD4 SP+DN thymocytes that were calculated by subtracting background staining. Data show typical results from one of four experiments.

Close modal

Thymic T cell preparations, which were obtained by removing cells expressing CD8 and contain ∼85% CD4 SP thymocytes and 15% DN thymocytes, were analyzed for expression of TCR-αβ and H4. Three distinct subpopulations were defined in TCR-αβ+ thymic T cells (Fig. 1 C); TCR-αβhigh thymic T cells with none to a low amount of H4 (hereafter, designated as H4low thymic T cells) which contained a majority of CD4 SP thymocytes, TCR-αβlow thymic T cells with an intermediate amount of H4 (H4int thymic T cells), and TCR-αβhigh thymic T cells with a high amount of H4 (H4high thymic T cells).

We next performed TCR analysis of these thymic T cell subpopulations because H4int thymic T cells express a low level of TCRs, similar to NKT cells (16, 19). The NKT cell population expresses a skewed Vβ repertoire of Vβ2, -7, and -8 in their TCRs (16, 17, 18, 19). Therefore, CD4 SP+DN thymocytes obtained from BALB/c thymocytes were stained with combinations of available anti-Vβs and anti-H4 mAb. The profiles in Fig. 2,A indicate that H4int and H4high thymic T cells exclusively express Vβ2+, Vβ7+, and Vβ8+ TCR, whereas H4low thymic T cells express all of the TCR Vβs examined. The proportions of Vβ7+, Vβ8.2+, and Vβ8.3+ cells in H4high thymic T cells are 2- to 3-fold higher than those in whole TCR-αβ+ thymocytes as controls, and those of Vβ2+ and Vβ8.1+ cells in H4high thymic T cells are almost same as controls (Table I). The proportions of Vβ2+, Vβ7+, Vβ8.1+, Vβ8.2+, and Vβ8.3+ cells in H4int thymic T cells were equivalent to those in H4high thymic T cells (data not shown). Thus, a biased usage of Vβ2+, -7+, and -8+ TCR in H4int and H4high thymic T cells strongly suggests that NKT cells are enriched in these subpopulations. Furthermore, H4int and H4high thymic T cells in DN thymocytes expressed a skewed Vβ repertoire of Vβ7 and Vβ8 TCR in their TCRs (Fig. 3), suggesting that these cells are also NKT cells. In C57BL/6 mice, H4high thymic T cells are seen at much lower levels (∼0.7% in TCR-αβ+ CD4 SP+DN thymocytes) (see the profiles in Fig. 4), and they show a biased usage of TCR Vβ as in BALB/c mice (Table I).

FIGURE 2.

Preferential expression of Vβ2+, Vβ7+, and Vβ8+ TCR in H4int and H4high thymic T cells. BALB/c CD4 SP+DN thymocytes (A) or DBA/2 CD4 SP+DN thymocytes (B) were obtained and stained with combinations of anti-Vβ or anti-TCR-αβ/appropriate FITC-anti-IgG and biotinylated anti-H4 mAb/avidin-PE. Samples of 100,000 viable cells were analyzed by an Epics CS flow cytometer. In B, profiles of staining for representative Vβ and H4 are shown. Data show typical results from one of three experiments.

FIGURE 2.

Preferential expression of Vβ2+, Vβ7+, and Vβ8+ TCR in H4int and H4high thymic T cells. BALB/c CD4 SP+DN thymocytes (A) or DBA/2 CD4 SP+DN thymocytes (B) were obtained and stained with combinations of anti-Vβ or anti-TCR-αβ/appropriate FITC-anti-IgG and biotinylated anti-H4 mAb/avidin-PE. Samples of 100,000 viable cells were analyzed by an Epics CS flow cytometer. In B, profiles of staining for representative Vβ and H4 are shown. Data show typical results from one of three experiments.

Close modal
Table I.

TCR Vβ expression on H4high and H4low thymic T cellsa

BALB/c (%)C57BL/6 (%)DBA/2 (%)
H4high thymic T cellsH4low thymic T cellsTCR-αβ+ thymic T cellsH4high thymic T cellsH4low thymic T cellsTCR-αβ+ thymic T cellsH4high thymic T cellsH4low thymic T cellsTCR-αβ+ thymic T cells
6.8 6.4 6.8 6.3 5.8 5.9 6.0 6.5 7.1 
0.2 7.6 7.2 0.6 7.6 7.4 ND ND ND 
1.1 9.6 9.1 1.2 8.4 8.1 1.0 0.1 0.4 
13.4 4.6 6.2 16.7 2.4 3.4 17.3 0.8 4.5 
8.1 5.6 7.5 6.0 8.1 6.5 7.2 5.3 0.1 5.1 
8.2 45.1 10.9 15.7 33.1 10.2 11.6 43.0 13.6 20.8 
8.3 17.6 9.5 10.2 11.2 4.4 5.3 11.7 12.0 10.0 
0.4 1.1 1.1 0.4 1.3 1.3 ND ND ND 
10 0.5 6.6 6.3 0.8 4.3 4.2 0.8 9.3 7.5 
14 1.0 7.9 7.2 0.9 7.9 7.5 ND ND ND 
BALB/c (%)C57BL/6 (%)DBA/2 (%)
H4high thymic T cellsH4low thymic T cellsTCR-αβ+ thymic T cellsH4high thymic T cellsH4low thymic T cellsTCR-αβ+ thymic T cellsH4high thymic T cellsH4low thymic T cellsTCR-αβ+ thymic T cells
6.8 6.4 6.8 6.3 5.8 5.9 6.0 6.5 7.1 
0.2 7.6 7.2 0.6 7.6 7.4 ND ND ND 
1.1 9.6 9.1 1.2 8.4 8.1 1.0 0.1 0.4 
13.4 4.6 6.2 16.7 2.4 3.4 17.3 0.8 4.5 
8.1 5.6 7.5 6.0 8.1 6.5 7.2 5.3 0.1 5.1 
8.2 45.1 10.9 15.7 33.1 10.2 11.6 43.0 13.6 20.8 
8.3 17.6 9.5 10.2 11.2 4.4 5.3 11.7 12.0 10.0 
0.4 1.1 1.1 0.4 1.3 1.3 ND ND ND 
10 0.5 6.6 6.3 0.8 4.3 4.2 0.8 9.3 7.5 
14 1.0 7.9 7.2 0.9 7.9 7.5 ND ND ND 
a

BALB/c, C57BL/6, and DBA/2 CD4 SP+DN thymocytes were stained and analyzed as in Fig. 2. Numbers indicate mean percentages of TCR Vβ+ cells among whole H4high, H4low, and TCR-αβ+ thymic T cells from three separate experiments. ND, not detected. Underlining indicates preferential expression.

FIGURE 3.

Biased usage of Vβ7+ and Vβ8+ TCR in H4int and H4high thymic T cells in DN thymocytes. BALB/c CD4 SP+DN thymocytes were obtained and stained with anti-Vβ/appropriate FITC-anti-IgG followed by biotinylated anti-H4 mAb/cy-chrom-streptavidin and PE-anti-CD4. Samples of 380,000 viable cells were analyzed by an Epics XL flow cytometer. The profiles in B represent Vβ and H4 expression on DN thymocytes gated as indicated in A. In C, percentages of TCR Vβ+ cells among H4high thymic T cells are shown. Data show typical results from one of three experiments.

FIGURE 3.

Biased usage of Vβ7+ and Vβ8+ TCR in H4int and H4high thymic T cells in DN thymocytes. BALB/c CD4 SP+DN thymocytes were obtained and stained with anti-Vβ/appropriate FITC-anti-IgG followed by biotinylated anti-H4 mAb/cy-chrom-streptavidin and PE-anti-CD4. Samples of 380,000 viable cells were analyzed by an Epics XL flow cytometer. The profiles in B represent Vβ and H4 expression on DN thymocytes gated as indicated in A. In C, percentages of TCR Vβ+ cells among H4high thymic T cells are shown. Data show typical results from one of three experiments.

Close modal
FIGURE 4.

Relationship between H4 and CD44, Ly6C, and NK1.1 expression. BALB/c (A) and C57BL/6 (B and C) CD4 SP+DN thymocytes were obtained and stained with combinations of anti-CD44/FITC-anti-rat IgG and biotinylated anti-H4 mAb/avidin-PE (A), and anti-H4 mAb/FITC-anti-hamster IgG and biotinylated anti-Ly6C/avidin-PE (B) or biotinylated anti-NK1.1/avidin-PE (C). Samples of 50,000 viable cells were analyzed by an Epics CS flow cytometer. Data show typical results from one of three experiments.

FIGURE 4.

Relationship between H4 and CD44, Ly6C, and NK1.1 expression. BALB/c (A) and C57BL/6 (B and C) CD4 SP+DN thymocytes were obtained and stained with combinations of anti-CD44/FITC-anti-rat IgG and biotinylated anti-H4 mAb/avidin-PE (A), and anti-H4 mAb/FITC-anti-hamster IgG and biotinylated anti-Ly6C/avidin-PE (B) or biotinylated anti-NK1.1/avidin-PE (C). Samples of 50,000 viable cells were analyzed by an Epics CS flow cytometer. Data show typical results from one of three experiments.

Close modal

In many strains of inbred mouse, endogeneous superantigens induce the negative selection of developing T cells in the thymus in a Vβ-specific fashion (6, 7). DBA/2 mice carry Mtv-7-encoded superantigens that delete Vβ6+, -7+, -8.1+, and -9+ T cells (8). To examine the influence of negative selection on the subpopulations defined by H4 expression, CD4 SP+DN thymocytes from DBA/2 mice were stained in the same way as above. As shown in Fig. 2,B and Table I, the proportion of Vβ7+ H4high and Vβ8.1+ H4high thymic T cells was comparable to that in BALB/c, consistent with the notion that NKT cells escape negative selection in the thymus (18, 19). Vβ6+, Vβ7+, and Vβ8.1+ H4low thymic T cells are markedly reduced, whereas other Vβ+ (such as Vβ2, Vβ8.2, Vβ8.3, Vβ10) H4low thymic T cells are observed at levels comparable to those in BALB/c. Thus, H4low thymic T cells differentiate via the mainstream differentiation pathway in the thymus.

NKT cells have been defined with CD44high, Ly6Chigh, and NK1.1+ (16, 18, 19). To determine the expression of these molecules on H4int and H4high thymic T cells, CD4 SP+DN thymocytes from BALB/c and C57BL/6 mice were stained with combinations of anti-H4 mAb and anti-CD44, or anti-H4 mAb and anti-Ly6C or anti-NK1.1, respectively. H4int thymic T cells express high levels of CD44, whereas H4high thymic T cells express an intermediate amount of CD44 (Fig. 4,A). Moreover, H4high thymic T cells do not coexpress Ly6C or NK1.1 molecules at all (Fig. 4, B and C). Ly6Chigh and NK1.1+ thymocytes correspond to H4int thymic T cells (Fig. 4, B and C). Thus, H4int thymic T cells are ordinary NKT cells, whereas H4high thymic T cells constitute a specialized NKT-like T cell subpopulation that is not defined by the NKT phenotype of CD44high, Ly6Chigh, and NK1.1+.

Pairing with polyclonal Vβ2+, -7+, and -8+ β-chains, an invariant α-chain composed of Vα14 and Jα281 is expressed in TCR of murine NKT cells (16, 17, 27). We examined whether the invariant α-chain is expressed in NKT-like H4high thymic T cells. In addition, it might be possible that H4int and H4high thymic T cells that escape negative selection in DBA/2 mice are oligoclones. Therefore, the repertoire of TCRα- and β-chains in Vβ7+ H4high thymic T cells in DBA/2 mice was analyzed and compared with that in Vβ7+ H4low thymic T cells.

T cell blasts selectively expressing Vβ7 were induced from DBA/2 Vβ7+ H4high thymic T cells (Fig. 5,A) and BALB/c Vβ7+ H4low thymic T cells (Fig. 5,B) by in vitro stimulation with YPM. PCR of cDNA obtained from these T cell blasts using Vα and Cα primers indicated that Vα14 transcripts are predominantly expressed over the other Vα examined in H4high thymic T cells (Fig. 5,Ca), whereas the amount of Vα14 transcripts is decreased in H4low thymic T cells (Fig. 5,Cb). Furthermore, only one nucleotide insertion (data not shown) and Jα281 usage in all junctional regions resulted in identical amino acid sequences derived from Vα14+ α-chains in Vβ7+ H4high thymic T cells (Fig. 5,D). Both H4low and H4high thymic T cells revealed heterogeneous junctional sequences in the Vβ7+ β-chain, and the length of CDR3 region (the distance from the J region-encoded GXG triplet to the nearest preceding V region-encoded cysteine according to Rock et al. (28)) was diverse (Fig. 5, E and F). Thus, it is clear that invariant Vα14-Jα281+ thymic T cells (hereafter designated as Vα14+ thymic T cells) are enriched in H4high thymic T cells, and those in DBA/2 mice do not consist of oligoclones. These results together with those in Fig. 4 indicate that Vα14+ thymic T cells consist of ordinary NKT cells and a new type of Vα14+ T cell.

FIGURE 5.

Invariant Vα14-Jα281+ T cells are enriched in H4high thymic T cells. DBA/2 CD4 SP+DN thymocytes and BALB/c CD4 SP+DN thymocytes were obtained and sorted into Vβ7+ H4high cells (A) and Vβ7+ H4low cells (B), respectively. The sorted cells were stimulated with YPM in the presence of APCs followed by the expansion of blast cells with rIL-2. C, First-strand cDNAs obtained from YPM-induced DBA/2 (a) and BALB/c (b) thymic T cell blasts were amplified by PCR using various Vαs and Cα primers. D, First-strand cDNAs derived from YPM-induced Vβ7+ H4high thymic T cell blasts were amplified by PCR using Vα14 and Cα primers. Junctional sequences of Vα14+ α-chain in cDNA clones were obtained as described in Materials and Methods. E and F, First-strand cDNAs derived from YPM-induced Vβ7+ H4high or H4low thymic T cell blasts were amplified by PCR using Vβ7 and Cβ primers. Junctional sequences of Vβ7+ β-chain in cDNA clones were obtained as described in Materials and Methods.

FIGURE 5.

Invariant Vα14-Jα281+ T cells are enriched in H4high thymic T cells. DBA/2 CD4 SP+DN thymocytes and BALB/c CD4 SP+DN thymocytes were obtained and sorted into Vβ7+ H4high cells (A) and Vβ7+ H4low cells (B), respectively. The sorted cells were stimulated with YPM in the presence of APCs followed by the expansion of blast cells with rIL-2. C, First-strand cDNAs obtained from YPM-induced DBA/2 (a) and BALB/c (b) thymic T cell blasts were amplified by PCR using various Vαs and Cα primers. D, First-strand cDNAs derived from YPM-induced Vβ7+ H4high thymic T cell blasts were amplified by PCR using Vα14 and Cα primers. Junctional sequences of Vα14+ α-chain in cDNA clones were obtained as described in Materials and Methods. E and F, First-strand cDNAs derived from YPM-induced Vβ7+ H4high or H4low thymic T cell blasts were amplified by PCR using Vβ7 and Cβ primers. Junctional sequences of Vβ7+ β-chain in cDNA clones were obtained as described in Materials and Methods.

Close modal

Because Vβ7+ T cells are potentially self-reactive in DBA/2 mice, it is of interest whether DBA/2 Vβ7+ Vα14+ thymic T cells are responsive or anergic. In the experiment shown in Fig. 5, a substantial number of Vβ7+ Vα14+ thymic T cell blasts was generated upon in vitro stimulation of DBA/2 H4high thymic T cells with YPM, indicating that Vβ7+ Vα14+ thymic T cells in DBA/2 mice are in an immunologically responsive state. We have previously found that CD1a CD4+ human thymic and cord blood T cells generated massive Vβ2+ T cell blasts upon primary stimulation with TSST-1, and that those T cell blasts are in the anergic state upon restimulation with TSST-1. By contrast, Vβ2+ CD4+ T cell blasts generated by stimulating human peripheral blood CD4+ T cells with TSST-1 exhibited enhanced IL-2 production upon restimulation with TSST-1 (29, 30). Thus, we consider that the responsiveness in this culture system can reflect the functional maturity of T cells. To further analyze the function of DBA/2 Vβ7+ Vα14+ thymocytes, the primary and secondary responses upon stimulation with YPM were examined.

Vβ7+ Vα14+ thymic T-enriched cells were obtained from BALB/c and DBA/2 CD4 SP+DN thymocytes by sequential immunomagnetic cell depletion and positive selection of H4+ cells containing mainly H4int and H4high thymic T cells (Fig. 6, Aa and Ab), and stimulated with YPM in the presence of APCs. BALB/c Vβ7+ H4int and H4high thymic T cells produced a substantial amount of IL-2 (Fig. 6,Ac) and IL-4 (Fig. 6,Ad) upon stimulation with YPM. DBA/2 H4int and H4high thymic T cells had a much lower response than BALB/c H4int and H4high thymic T cells (Fig. 6, Ac and Ad). The lower proportion of TCR-αβhigh H4high thymic T cells in DBA/2 H4+ thymic T cells than BALB/c H4+ thymic T cells (Fig. 6, Aa and Ab) may be the cause of lower responsiveness. Alternatively, cells not triggered by the stimulation with YPM may be present in DBA/2 mice.

FIGURE 6.

Responsiveness of Vβ7+ Vα14+ thymic T cells. A, Vβ7+ Vα14+ thymic T-enriched population was obtained by serial treatments as described in Materials and Methods. The purity of cells from BALB/c (a) and DBA/2 (b) after preparation was analyzed by flow cytometry. Numbers indicate the percentage of H4+ cells calculated by subtracting background staining. A total of 5 × 104 cells were cultured with titrated amounts of YPM in the presence of 2 × 105 syngenic APCs. After 2.5 days of culture, IL-2 production (c) and IL-4 production (d) were evaluated. Results are expressed as mean ± SD of triplicate cultures. B, In parallel with the above culture, 1 × 105 cells from BALB/c mice and 5 × 105 cells from DBA/2 mice were cultured with 3 μg/ml YPM in the presence of 5 × 105 syngenic APCs. Blast cells were purified after 3 days of culture, expanded with rIL-2 for 4 days, and used as YPM-induced thymic T cell blasts. The purity of YPM-induced thymic T cell blasts from BALB/c (a) and DBA/2 (b) was analyzed by flow cytometry and PCR using Vαs and Cα primers. A total of 1 × 105 YPM-induced thymic T cell blasts were restimulated with titrated amounts of YPM in the presence of 2 × 105 syngenic APCs. After 24 h of culture, IL-2 (c) and IL-4 (d) production were evaluated as in A. Data shown are from one of three experiments which gave similar results.

FIGURE 6.

Responsiveness of Vβ7+ Vα14+ thymic T cells. A, Vβ7+ Vα14+ thymic T-enriched population was obtained by serial treatments as described in Materials and Methods. The purity of cells from BALB/c (a) and DBA/2 (b) after preparation was analyzed by flow cytometry. Numbers indicate the percentage of H4+ cells calculated by subtracting background staining. A total of 5 × 104 cells were cultured with titrated amounts of YPM in the presence of 2 × 105 syngenic APCs. After 2.5 days of culture, IL-2 production (c) and IL-4 production (d) were evaluated. Results are expressed as mean ± SD of triplicate cultures. B, In parallel with the above culture, 1 × 105 cells from BALB/c mice and 5 × 105 cells from DBA/2 mice were cultured with 3 μg/ml YPM in the presence of 5 × 105 syngenic APCs. Blast cells were purified after 3 days of culture, expanded with rIL-2 for 4 days, and used as YPM-induced thymic T cell blasts. The purity of YPM-induced thymic T cell blasts from BALB/c (a) and DBA/2 (b) was analyzed by flow cytometry and PCR using Vαs and Cα primers. A total of 1 × 105 YPM-induced thymic T cell blasts were restimulated with titrated amounts of YPM in the presence of 2 × 105 syngenic APCs. After 24 h of culture, IL-2 (c) and IL-4 (d) production were evaluated as in A. Data shown are from one of three experiments which gave similar results.

Close modal

In parallel, blast cells were generated during the primary stimulation of DBA/2 and BALB/c H4+ thymic T cells with YPM, expanded with rIL-2, and used as YPM-induced T cell blasts. As can be seen in Fig. 6, Ba and Bb, although the proportions of TCR+ cells in the preparations of YPM-induced blast cells were low, almost all of those thymic T cell blasts expressed Vβ7+ TCR. In addition, PCR using Vα and Cα primers showed the preferential usage of Vα14 in their TCR, thus indicating enrichment for Vβ7+ Vα14+ thymic T cells. BALB/c- and DBA/2-derived Vβ7+ Vα14+ thymic T cell blasts exhibited comparable levels of IL-2 (Fig. 6,Bc) and IL-4 (Fig. 6 Bd) production upon restimulation with YPM. These results indicate that Vβ7+ Vα14+ thymic T cells that survive without being deleted in DBA/2 mice contain mature T cells that are as functional as those in BALB/c mice.

Next, we assessed the biological difference between Vα14+ thymic T cells and mainstream thymic T cells. Thymic T cell subpopulations have quite different sensitivities to steroid hormones (26, 31); SP thymocytes are relatively resistant, whereas DP thymocytes are sensitive. We examined the effect of treatment of mice with dexamethasone on H4int and H4high thymic T cells. The number of BALB/c thymic T cell subpopulations was plotted against dexamethasone dose (Fig. 7). Dexamethasone induced the deletion of H4int and H4high thymic T cells that was around 3- and 4-fold larger than that of CD4 SP thymocytes at 300 μg and 900 μg of dexamethasone, respectively. Thus, Vα14+ thymic T cells are relatively sensitive to dexamethasone compared to mainstream CD4 SP thymocytes.

FIGURE 7.

Vα14+ thymic T cells are relatively sensitive to dexamethasone in vivo. Various doses of dexamethasone in ethanol/saline were injected into BALB/c mice s.c. Two days later, thymocytes were stained with combinations of FITC-anti-CD8 and PE-anti-CD4, or anti-TCR-αβ/FITC-anti-hamster IgG and biotin-anti-H4 mAb/avidin-PE. Relative cell number of thymic T cell subpopulations were calculated. Results are expressed as the mean and SD of 4 mice/group. (∗, p < 0.001 when compared with the relative cell number of CD4 SP thymocytes at each dose.) Data shown are typical results from one of three experiments.

FIGURE 7.

Vα14+ thymic T cells are relatively sensitive to dexamethasone in vivo. Various doses of dexamethasone in ethanol/saline were injected into BALB/c mice s.c. Two days later, thymocytes were stained with combinations of FITC-anti-CD8 and PE-anti-CD4, or anti-TCR-αβ/FITC-anti-hamster IgG and biotin-anti-H4 mAb/avidin-PE. Relative cell number of thymic T cell subpopulations were calculated. Results are expressed as the mean and SD of 4 mice/group. (∗, p < 0.001 when compared with the relative cell number of CD4 SP thymocytes at each dose.) Data shown are typical results from one of three experiments.

Close modal

In parallel, Vβ7+ mainstream thymic T cells were cultured in the same way as in Fig. 6, and their responses were compared with those of Vβ7+ Vα14+ thymic T cells. Because 500 μg of dexamethasone did not affect the lymphokine production of YPM-induced thymic T cell blasts in preliminary experiments, we pretreated the mice with dexamethasone to substantially deplete H4int and H4high thymic T cells and DP thymocytes beforehand. Vβ7+ mainstream thymic T-enriched cells were obtained from BALB/c mice pretreated with dexamethasone by sequential immunomagnetic cell depletion and stimulated with YPM in the presence of APCs. After expansion with rIL-2, YPM-induced T cell blasts were obtained. The exclusive proportion (91%) of Vβ7+ TCR in TCR+ thymocytes, and a lower amount of Vα14 transcript indicate the enrichment for Vβ7+ mainstream thymic T cell blasts (Fig. 8,A). They produced ∼10-fold higher amounts of IL-2 (Fig. 8,B) and ∼100-fold lower amounts of IL-4 (Fig. 8 C) than Vα14+ thymic T cell blasts upon restimulation with YPM. Taken together, these results indicate that Vα14+ thymic T cells are biologically quite different from mainstream thymic T cells.

FIGURE 8.

Comparison of responsiveness between Vβ7+ mainstream thymic T cells and Vβ7+ Vα14+ thymic T cells. A total of 5 × 105 BALB/c Vβ7+ mainstream thymic T-enriched cells obtained as described in Materials and Methods were cultured with 3 μg/ml YPM in the presence of 5 × 105 syngenic APCs. After expansion of blast cells with rIL-2, YPM-induced Vβ7+ mainstream thymic T cell blasts were obtained. The purity of these cells (A) was analyzed as in Fig. 6. In parallel, BALB/c-derived YPM-induced Vβ7+ Vα14+ thymic T cell blasts were obtained as in Fig. 6. The proportion of TCR-αβ+ T cells and Vβ7+ T cells in the preparation of YPM-induced blast cells was 55.3% and 55.5%, respectively, and PCR showed the preferential usage of Vα14 in their TCRs (data not shown). IL-2 (B) and IL-4 (C) productions of YPM-induced Vβ7+ mainstream and Vβ7+ Vα14+ thymic T cell blasts upon restimulation with YPM were analyzed as in Fig. 6. Data shown are from one of three experiments which gave similar results.

FIGURE 8.

Comparison of responsiveness between Vβ7+ mainstream thymic T cells and Vβ7+ Vα14+ thymic T cells. A total of 5 × 105 BALB/c Vβ7+ mainstream thymic T-enriched cells obtained as described in Materials and Methods were cultured with 3 μg/ml YPM in the presence of 5 × 105 syngenic APCs. After expansion of blast cells with rIL-2, YPM-induced Vβ7+ mainstream thymic T cell blasts were obtained. The purity of these cells (A) was analyzed as in Fig. 6. In parallel, BALB/c-derived YPM-induced Vβ7+ Vα14+ thymic T cell blasts were obtained as in Fig. 6. The proportion of TCR-αβ+ T cells and Vβ7+ T cells in the preparation of YPM-induced blast cells was 55.3% and 55.5%, respectively, and PCR showed the preferential usage of Vα14 in their TCRs (data not shown). IL-2 (B) and IL-4 (C) productions of YPM-induced Vβ7+ mainstream and Vβ7+ Vα14+ thymic T cell blasts upon restimulation with YPM were analyzed as in Fig. 6. Data shown are from one of three experiments which gave similar results.

Close modal

The present study has demonstrated that TCR-αβ+ CD4+ thymic T cells are composed of three subpopulations based on H4 expression: H4low, H4int, and H4high thymic T cells. H4low thymic T cells branch off the mainstream differentiation pathway in the thymus. H4int thymic T cells are ordinary NKT cells, whereas H4high thymic T cells constitute a specialized population of Vα14+ thymic T cells. Vβ7+ and Vβ8.1+ Vα14+ thymic T cells are present in DBA/2 mice at comparable levels to BALB/c mice, whereas Vβ7+ and Vβ8.1+ mainstream thymic T cells were deleted, indicating that Vα14+ thymic T cells escape negative selection in the thymus. Furthermore, Vβ7+ Vα14+ thymic T cells in DBA/2 mice were shown to be in an immunologically responsive state and contain T cells as functionally mature as BALB/c mice.

It was shown that the NKT phenotype of CD44high, Ly6Chigh, and NK1.1+ corresponds to that of H4int thymic T cells, whereas H4high thymic T cells do not express a high amount of CD44, Ly6C, or NK1.1. Because H4high thymic T cells express a high amount of TCR, whereas H4int thymic T cells express a low amount of TCR (Figs. 1,C and 2), these results are consistent with the notion that thymic NKT cells express a lower amount of TCR than mainstream thymic T cells (16, 19). However, more importantly, data presented here indicate that there is a specialized population in the thymus expressing a high amount of an invariant Vα14+ TCR that is associated with H4, but not with CD44, Ly6C, and NK1.1. Thus, Vα14+ thymic T cells consist of ordinary NKT cells and a new type of Vα14+ T cells. The reason that this type of Vα14+ T cell population has not been demonstrated to date is probably because Vα14+ NKT cells have been characterized mainly by using NK1.1+ C57BL/6 mice, and H4high thymic T cells are present at much lower levels in C57BL/6 mice than BALB/c mice (Fig. 4). This new type of Vα14+ T cell raises several questions. Do they originate from a common differentiation pathway as NKT cells, indicating the different maturation steps of Vα14+ T cells, or from separate pathways? Are their activation requirements different from those of ordinary NKT cells? In peripheral lymphoid organs, how much do they occupy and how do they regulate immune responses? In BALB/c splenic CD4+ T cells, some cells (∼8%) express a substantial amount of H4 irrespective of Vβ in their TCRs, suggesting that H4 is the surface marker on activated/memory CD4+ T cells. This fact makes it difficult to demonstrate Vα14+ T cells in the periphery with relation to H4. However, we are currently examining how much a new type of Vα14+ T cell is present in the periphery. Obviously, further experiments will be required to resolve these questions.

Vα14+ thymic T cells were shown to be biologically different from mainstream thymic T cells in several ways. First, Vα14+ thymic T cells escape the negative selection in the thymus of Mtv-7+ DBA/2 mice, whereas mainstream thymic T cells are deleted. Second, the pattern of lymphokine production is distinct. YPM-induced Vα14+ thymic T cell blasts produce markedly higher levels of IL-4 and lower levels of IL-2 than YPM-induced mainstream thymic T cell blasts upon in vitro restimulation with YPM, consistent with the findings of primary stimulation of NKT cells with anti-TCR-αβ Abs (32, 33). Third, Vα14+ thymic T cells are relatively sensitive to dexamethasone compared with CD4 SP mainstream thymocytes. Tamada et al. (34) have recently reported the resistance of NKT cells in both spleen and liver to dexamethasone-induced apoptosis. There are some possible explanations for the distinct reactivity of Vα14+ thymic T cells to dexamethasone. Although NKT cell precursors are suggested to reside in the thymus (35, 36, 37), the sites for the generation of Vα14+ thymic T cells and peripheral NKT cells might be separate as suggested by a couple of reports (38, 39), leading to the different susceptibility to dexamethasone-induced apoptosis. Because the activation of cells induces the translocation of glucocorticoid receptor to the nucleus (40, 41), it is likely that the status of activation of cells is a determinant of susceptibility to apoptosis. From this viewpoint, Vα14+ thymic T cells may be more activated than peripheral NKT cells.

There were some findings concerning the function of Vα14+ thymic T cells. BALB/c Vβ7+ Vα14+ thymic T cells responded well to YPM in both primary and secondary culture, suggesting that they are capable of responding upon rechallenge with Ags. DBA/2 Vβ7+ Vα14+ thymic T cells that escaped the negative selection were shown to be in an immunologically responsive state. The lower response of Vβ7+ Vα14+ thymic T cells from DBA/2 mice upon primary stimulation with YPM than that from BALB/c mice may be due to the presence of some anergized T cells under the influence of Mtv-7-encoded superantigens. Further experiments are required to clarify this. However, the same degree of response of YPM-induced Vβ7+ Vα14+ thymic T cell blasts from DBA/2 and BALB/c mice was observed upon restimulation with YPM. This finding clearly indicates that Vβ7+ Vα14+ thymic T cells in DBA/2 mice contain T cells as functionally mature as those in BALB/c mice. Therefore, the mechanism of escape of Vα14+ thymic T cells from the negative selection is important. Some possible mechanisms can be proposed. Vα14+ thymic T cells that develop in contact with CD1+ cortical thymocytes (16, 42, 43) may not encounter their deleting ligands on stromal cells. However, previous reports have shown that DN NKT cells are selected in the thymus of anti-HY TCR/Rag2−/− mice (44) or in the presence of specific class I MHC Ag (45). Thus, we infer that Vα14+ thymic T cells develop by interaction with Mtv-7+ stromal cells. Alternatively, it is possible that H4 molecules affect TCR-mediated signaling in thymic selection. The interaction of H4 with unknown ligands may reduce the overall avidity of TCRs on Vβ7+ and Vβ8.1+ Vα14+ thymic T cells for Mtv-7-encoded superantigens, rescuing them from negative selection. Finally, the TCR structure may be involved. Because hypervariable region 4 of the Vβ domain is crucial for the recognition of superantigens (46, 47), sequences of hypervariable region 4 in Vβ7+ and Vβ8.1+ β-chain of Vα14+ thymic T cells should be analyzed. In addition, it is known that non-Vβ TCR elements such as Vα repertoire and the combination of Vα and Jα usage influence T cell recognition of Mls-1a (12, 13, 14). Furthermore, Mtv-7+ mice show that most, but not all, thymocytes with the relevant TCR Vβ are deleted in the thymus (9, 10, 11, 12), and that TCR Vα repertoire in T cells surviving the negative selection is skewed (10, 11, 12, 15). Thus, skewed Vα repertoire or invariant α-chain may affect the affinity of TCR for Mtv-7-encoded superantigens, decreasing it to below the range for negative selection. In the present study, besides the predominance of Vα14, the Vα repertoire in Vβ7+ H4high thymic T cells is different from that in Vβ7+ mainstream thymic T cells. The use of Vα8 in the former and that of Vα11 in the latter was higher than other Vαs (Figs. 5, 6, and 8), thus supporting this hypothesis.

An important question arises as to whether these mature Vβ7+ and Vβ8.1+ Vα14+ T cells readily react to Mtv-7-encoded super-antigens in the periphery, leading to the harmful response to the host? Because the DBA/2 mouse is not an autoimmune model mouse, it is apparent that immunological self-tolerance is maintained in the mouse. Thus, the above possibility is less likely. Some mechanisms can be suggested to explain self-tolerance of Vβ7+ and Vβ8.1+ Vα14+ T cells in DBA/2 mice. DBA/2 Vβ7+ and Vβ8.1+ Vα14+ T cells may be induced into a state of anergy after interaction with Mtv-7+ cells in the periphery. However, the responsiveness of DBA/2 Vβ7+ Vα14+ thymic T cells to YPM in vitro suggests that the induction of anergy in these cells is less likely. An alternative mechanism, which we favor, is based upon the above-mentioned hypothesis that a distinct TCR structure causes the decreased affinity for Mtv-7-encoded superantigens. Vβ7+ and Vβ8.1+ Vα14+ T cells interact with Mtv-7-encoded superantigens in the peripheral lymphoid organs, but their signaling will not result in full activation, but survival as in the thymus. Thus, Vβ7+ and Vβ8.1+ Vα14+ T cells in DBA/2 mice seem to be regulated to not exceed triggering thresholds at the interaction with Mtv-7-encoded superantigens, while maintaining the responsiveness to natural ligands. The data presented here thus provide the basis for future work exploring the novel mechanisms of self-tolerance.

We thank Drs. K. Tomonari, G. Suzuki, H. Nariuchi, and T. Nakayama for provision of mAbs. We also thank Hisako Yagi for technical help.

1

This work was supported in part by grants from the Ministry of Education, Science, Sports, and Culture of Japan, the Ministry of Public Welfare of Japan, and the Associazione Italiana Ricerca sul Cancro (Milan).

3

Abbreviations used in this paper: SP, single positive; DP, double positive; Mtv, mouse mammary tumor virus; TSST-1, toxic shock syndrome toxin-1; YPM, Yersinia pseudotuberculosis-derived mitogen; DN, double negative.

1
Fowlkes, B. J., D. M. Pardoll.
1989
. Molecular and cellular events of T cell development.
Adv. Immunol.
44
:
207
2
Nikolic-Zugic, J..
1991
. Phenotypic and functional stages in the intrathymic development of αβ T cells.
Immunol. Today
12
:
65
3
Nossal, G. J. V..
1994
. Negative selection of lymphocytes.
Cell
76
:
229
4
Fink, P. J., M. J. Bevan.
1995
. Positive selection of thymocytes.
Adv. Immunol.
59
:
99
5
Kisielow, P., H. von Boehmer.
1995
. Development and selection of T cells: facts and puzzles.
Adv. Immunol.
58
:
87
6
Kappler, J. W., N. Roehm, P. Marrack.
1987
. T-cell tolerance by clonal elimination in the thymus.
Cell
49
:
273
7
MacDonald, H. R., R. Schneider, R. K. Lees, R. C. Howe, H. Acha-Orbea, H. Festenstein, R. M. Zinkernagel, H. Hengartner.
1988
. T-cell receptor Vβ use predicts reactivity and tolerance to Mlsa-encoded antigens.
Nature
332
:
40
8
Tomonari, K., S. Fairchild, O. A. Rosenwasser.
1993
. Influence of viral superantigens on Vβ- and Vα-specific positive and negative selection.
Immunol. Rev.
131
:
131
9
Blackman, M. A., H. Gerhard-Burgert, D. L. Woodland, E. Palmer, J. W. Kappler, P. Marrack.
1990
. A role for clonal inactivation in T cell tolerance to Mls-1a.
Nature
345
:
540
10
Smith, H. P., P. Le, D. L. Woodland, M. A. Blackman.
1992
. T cell receptor α-chain influences reactivity to Mls-1 in Vβ8.1 transgenic mice.
J. Immunol.
149
:
887
11
Blackman, M. A., H. P. Smith, P. Le, D. L. Woodland.
1993
. Influence of the T cell receptor α-chain on T cell reactivity and tolerance to Mls-1 in T cell receptor β-chain transgenic mice.
J. Immunol.
151
:
556
12
Blackman, M. A., D. L. Woodland.
1996
. Role of the T cell receptor α-chain in superantigen recognition.
Immunol. Res.
15
:
98
13
Vacchio, M. S., O. Kanagawa, K. Tomonari, R. J. Hodes.
1992
. Influence of T cell receptor Vα expression on Mlsa superantigen-specific T cell responses.
J. Exp. Med.
175
:
1405
14
Woodland, D. L., H. P. Smith, S. Surman, P. Le, R. Wen, M. A. Blackman.
1993
. Major histocompatibility complex-specific recognition of Mls-1 is mediated by multiple elements of the T cell receptor.
J. Exp. Med.
177
:
433
15
Chies, J. A. B., G. Marodon, A.-M. Joret, A. Regnault, M.-P. Lembezat, B. Rocha, A. A. Freitas.
1995
. Persistence of Vβ6+ T cells in Mls-1a mice: a role for the third complementarity-determining region (CDR3) of the T cell receptor β-chain in superantigen recognition.
J. Immunol.
155
:
4171
16
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
17
Makino, Y., H. Koseki, Y. Adachi, T. Akasaka, K. Tsuchida, M. Taniguchi.
1994
. Extrathymic differentiation of a T cell bearing invariant Vα14Jα281 TCR.
Int. Rev. Immunol.
11
:
31
18
Takahama, Y., S. O. Sharrow, A. Singer.
1991
. Expression of an unusual T cell receptor (TCR)-Vβ repertoire by Ly-6C+ subpopulations of CD4+ and/or CD8+ thymocytes: evidence for a developmental relationship between Ly-6C+ thymocytes and CD4CD8TCR-αβ+ thymocytes.
J. Immunol.
147
:
2883
19
Arase, H., N. Arase, K. Ogasawara, R. A. Good, K. Onoé.
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
20
Redoglia, V., U. Dianzani, J. M. Rojo, P. Portolés, M. Bragardo, H. Wolff, D. Buonfiglio, S. Bonissoni, C. A. Janeway, Jr.
1996
. Characterization of H4: a murine T lymphocyte activation molecule functionally associated with the CD3/TCR.
Eur. J. Immunol.
26
:
2781
21
Uchiyama, T., T. Miyoshi-Akiyama, H. Kato, W. Fujimaki, K. Imanishi, X.-J. Yan.
1993
. Superantigenic properties of a novel mitogenic substance produced by Yersinia pseudotuberculosis isolated from patients manifesting acute and systemic symptoms.
J. Immunol.
151
:
4407
22
Miyoshi-Akiyama, T., A. Abe, H. Kato, K. Kawahara, H. Narimatsu, T. Uchiyama.
1995
. DNA sequencing of the gene encoding a bacterial superantigen, Yersinia pseudotuberculosis-derived mitogen (YPM), and characterization of the gene product, cloned YPM.
J. Immunol.
154
:
5228
23
Miyoshi-Akiyama, T., W. Fujimaki, X.-J. Yan, J. Yagi, K. Imanishi, H. Kato, K. Tomonari, T. Uchiyama.
1997
. Identification of murine T cells reactive with the bacterial superantigen Yersinea pseudotuberculosis-derived mitogen (YPM) and factors involved in YPM-induced toxicity in mice.
Microbiol. Immunol.
41
:
345
24
Yan, X.-J., X.-Y. Li, K. Imanishi, Y. Kumazawa, T. Uchiyama.
1993
. Study of activation of murine T cells with bacterial superantigens. In vitro induction of enhanced responses in CD4+ T cells and of anergy in CD8+ T cells.
J. Immunol.
150
:
3873
25
Yagi, J., M. Nakata, T. Uchiyama, M. Nishikawa, Y. Mizushima, K. Nishioka, K. Ito, H. Yagita, K. Okumura, C. A. Janeway, Jr, K. Yamamoto.
1994
. Superantigen-like properties of an antibody bispecific for MHC class II molecules and the Vβ domain of the T cell antigen receptor.
J. Immunol.
152
:
3833
26
Screpanti, I., S. Morrone, D. Meco, A. Santoni, A. Gulino, R. Paolini, A. Crisanti, B. J. Mathieson, L. Frati.
1989
. Steroid sensitivity of thymocyte subpopulations during intrathymic differentiation: effects of 17 β-estradiol and dexamethasone on subsets expressing T cell antigen receptor or IL-2 receptor.
J. Immunol.
142
:
3378
27
Oliver, L., 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
28
Rock, E. P., P. R. Sibbald, M. M. Davis, Y. Chien.
1994
. CDR3 length in antigen-specific immune receptors.
J. Exp. Med.
179
:
323
29
Takahashi, N., K. Imanishi, H. Nishida, T. Uchiyama.
1995
. Evidence for immunologic immaturity of cord blood T cells: cord blood T cells are susceptible to tolerance induction to in vitro stimulation with a superantigen.
J. Immunol.
155
:
5213
30
Imanishi, K., K. Seo, H. Kato, T. Miyoshi-Akiyama, R.-H. Zhang, Y. Takanashi, Y. Imai, T. Uchiyama.
1998
. Post-thymic maturation of migrating human thymic single-positive T cells: thymic CD1a CD4+ T cells are more susceptible to anergy induction by toxic shock syndrome toxin-1 than cord blood CD4+ T cells.
J. Immunol.
160
:
112
31
Homo, F., D. Duval, J. Hatzfeld, C. Evrard.
1980
. Glucocorticoid-sensitive and -resistant cell populations in the mouse thymocytes.
J. Steroid Biochem.
13
:
135
32
Bendelac, A., R. H. Schwartz.
1991
. CD4+ and CD8+ T cells acquire specific lymphokine secretion potentials during thymic maturation.
Nature
353
:
68
33
Arase, H., N. Arase, K. Nakagawa, R. A. Good, K. Onoé.
1993
. NK1.1+ CD4+ CD8 thymocytes with specific lymphokine secretion.
Eur. J. Immunol.
23
:
307
34
Tamada, K., M. Harada, K. Abe, T. Li, K. Nomoto.
1998
. IL-4-producing NK1.1+ T cells are resistant to glucocorticoid-induced apoptosis: implications for the Th1/Th2 balance.
J. Immunol.
161
:
1239
35
Sykes, M..
1990
. Unusual T cell populations in adult murine bone marrow. Prevalence of CD3+CD4CD8 and αβTCR+NK1.1+ cells.
J. Immunol.
145
:
3209
36
Levitsky, H. I., P. T. Golumbek, D. M. Pardoll.
1991
. The fate of CD48 T cell receptor-αβ+ thymocytes.
J. Immunol.
146
:
1113
37
Hammond, K., W. Cain, I. van Driel, D. Godfrey.
1998
. Three day neonatal thymectomy selectively depletes NK1.1+ T cells.
Int. Immunol.
10
:
1491
38
Makino, Y., N. Yamagata, T. Sasho, Y. Adachi, R. Kanno, H. Koseki, M. Kanno, M. Taniguchi.
1993
. Extrathymic development of Vα14-positive T cells.
J. Exp. Med.
177
:
1399
39
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
40
Dieken, E. S., R. L. Miesfeld.
1992
. Transcriptional transactivation functions localized to the glucocorticoid receptor N terminus are necessary for steroid induction of lymphocyte apoptosis.
Mol. Cell. Biol.
12
:
589
41
King, L. B., J. D. Ashwell.
1993
. Signaling for death of lymphoid cells.
Curr. Opin. Immunol.
5
:
368
42
Bendelac, A..
1995
. Positive selection of mouse NK1+ T cells by CD1-expressing cortical thymocytes.
J. Exp. Med.
182
:
2091
43
Chen, Y.-H., N. M. Chiu, M. Mandal, N. Wang, C. R. Wang.
1997
. Impaired NK1+ T cell development and early IL-4 production in CD1-deficient mice.
Immunity
6
:
459
44
Schulz, R.-J., A. Parkes, E. Mizoguchi, A. K. Bhan, S. Koyasu.
1996
. Development of CD4CD8 αβTCR+NK1.1+ T lymphocytes: thymic selection by self antigen.
J. Immunol.
157
:
4379
45
Curnow, S. J., C. Boyer, M. Buferne, A.-M. Schmitt-Verhulst.
1995
. TCR-associated ζ-FcεRIγ heterodimers on CD4CD8NK1.1+ T cells selected by specific class I MHC antigen.
Immunity
3
:
427
46
Pullen, A., T. Wade, P. Marrack, J. W. Kappler.
1990
. Identification of the region of T cell receptor β chain that interacts with the self-superantigen Mls-1a.
Cell
61
:
1365
47
Cazenave, P.-A., P. N. Marche, E. Jouvin-Marche, D. Voegtle, F. Bonhomme, A. Bandeira, A. Coutinho.
1990
. Vβ17 gene polymorphism in wild-derived mouse strains: two amino acid substitutions in the Vβ17 region greatly alter T cell receptor specificity.
Cell
63
:
717