In this work, we have studied the role of the MHC class Ib molecules in the selection and maintenance of CD8+ T splenocytes. We have compared the CD8+ T cell repertoires of wild-type, H-2K-deficient, H-2D-deficient, or double knockout C57BL/6 mice. We show that the different CD8+ repertoires, selected either by class Ia and class Ib or by class Ib molecules only, use the various Vα (AV) and Vβ (BV) rearrangements in the same proportion and without biases in the CDR3 size distribution. Furthermore, we have estimated the size of the BV repertoire in the four different strains of mice. Interestingly, we have found that the BV repertoire size is proportional to the overall number of CD8+ splenocytes. This observation implies that BV diversity is positively correlated with the number of CD8+ cells, even when the number of CD8+ splenocytes is dramatically reduced (90% in the double knockout mice).

Immune responses to intracellular pathogens involve the recognition of MHC class I molecules loaded with peptides derived from infectious agents. These peptides/MHC class I complexes activate CD8+ T lymphocytes bearing specific αβ TCRs. In mice, the MHC class I-related molecules include the classical (class Ia) proteins encoded by highly polymorphic genes (H-2K, H-2D, and H-2L), nonclassical (class Ib) ones encoded in Q, T, and M regions, and, finally, the β2-microglobulin-associated molecules such as CD1 encoded by genes not linked to the MHC region. The MHC class Ib molecules H2-M3, HLA-E, and the mouse CD1d have an overall structure comparable to class Ia molecules (1, 2, 3, 4). The major differences with the class Ia molecules are found in the peptide binding groove (1, 3, 4), which can accommodate various ligands such as peptides from class Ia hydrophobic leader sequences for Qa1 and HLA-E (5, 6), N-formylated peptides derived from the N termini of certain bacterial and mitochondrial encoded proteins for H2-M3 (7) and ceramides, as well as some protozoan glycosylphosphatidylinositol for CD1d (8, 9, 10, 11).

The three-dimensional structure of the αβ TCR has shown that each Vα (AV)4 or Vβ (BV) chain forms three loops that interact with the peptide/MHC class Ia molecule (12, 13). These loops correspond to the complementarity-determining regions (CDR). The CDR1 and 2 are encoded within the AV and BV genes. The third CDRs are generated by the somatic rearrangement of V and J segments for AV and V, D, and J segments for BV. The diversity of the TCR repertoire results from rearrangements of various gene segments, their imprecise joining, the addition of template-independent N nucleotides during this process, and the pairing of different α- and β-chains (14, 15).

During development, the TCR repertoire of thymocytes is shaped by two different mechanisms of selection. Positive selection rescues thymocytes from cell death and warrants that mature T lymphocytes are capable of recognizing peptides/MHC complexes. Negative selection ensures that high avidity self-reactive thymocytes are eliminated through clonal deletion (16, 17). Both selective processes involve specific interactions with self-MHC molecules (18, 19, 20, 21) but lead to two different cell fates. The role of naturally processed self-peptides, eluted from MHC class Ia molecules, has recently been addressed by several groups (22, 23). Some of these self-peptides, even though they had few homologies with the antigenic peptide, were able to positively select some but not all CD8+ thymocytes bearing transgenic TCRs (23). These findings imply that weakly specific interactions between TCRs and self-peptides/MHC class Ia complexes promote positive selection of CD8+ thymocytes. Recent results of Berg et al. (24) suggest that positive selection by self-peptides/MHC class Ib molecules may be more specific than for peptides/MHC class Ia. Thus, a transgenic TCR specific for a Listeria-derived peptide presented by the MHC class Ib molecule H2-M3 is selected by a single self-peptide, i.e., the formyl-methionine peptide from an NADH dehydrogenase (24).

Selection of TCRs by self-peptides/MHC class Ia or Ib complexes is rather similar. However, the size and diversity of the CD8+ T lymphocytes selected by these complexes has not been determined yet. To study the size and diversity of the repertoire of CD8+ cells selected on class Ib molecules only, we took advantage of C57BL/6 mice deficient for H-2K and H-2D genes (H-2K°/°, H-2D°/°) (25). In the double knockout (KO) mice, only class Ib molecules persist because the MHC class Ia H-2Lb is missing in the C57BL/6 strain.

Using a method recently described by Casrouge et al. (26), we have estimated the size of the TCRβ repertoire from naive T splenocytes of wild-type, H-2K°/°, H-2D°/°, or double KO C57BL/6 mice.

All mice used in this study were 6- to 8-wk-old C57BL/6, H-2K°/°, H-2D°/°, or double KO C57BL/6 mice raised at the Pasteur Institute animal facility.

FITC-labeled anti-CD8β Abs were purchased from PharMingen (San Diego, CA). Splenocytes from 6- to 8-wk-old mice were depleted of B220+ and CD4+ cells using biotinylated mAbs and streptavidin beads (Dynals, Oslo, Norway). Depleted splenocytes were then incubated with rat anti-mouse CD8β mAb (PharMingen) and anti-rat IgG magnetic-activated cell sorting microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany), passed through MS+/RS+ columns (Miltenyi Biotec), and separated into CD8+ and CD8 lymphocytes. The CD8+ cell populations were counted and their purity was assessed by flow cytometry analyses and found to be ∼98%.

Total RNA was extracted using Trizol reagent (Life Technologies, Gaithersburg, MD) as recommended by the manufacturer, with the addition of 20 mg/ml of glycogen (Boehringer Mannheim, Mannheim, Germany) to the samples. Total RNAs were reverse transcribed into cDNA using oligo(dT)17 primers, 1 mM dNTP, 40 U Rnasin (Promega, Madison, WI), and 200 U of reverse transcriptase from Moloney murine leukemia virus (Life Technologies) at 43°C for 1 h, followed by an incubation at 53°C for 10 min.

The number of CD3ε copies in each cDNA preparation was quantified as described by Apostolou et al. (27). A plasmid containing a four-nucleotides-deleted form of a mouse CD3ε cDNA fragment was used as competitor. Copies (106–102) of the plasmid were mixed with 1 μl of cDNA preparation, and a 40-cycle competitive PCR was performed. The amplified products were subjected to run-off reaction with fluorescent anti-sense CD3ε-specific primer designed to hybridize equally well to wild-type and deleted CD3ε. These products were loaded on a sequencing gel and analyzed with Immunoscope software, then the number of CD3ε copies was determined.

Semiquantitative analyses of the TCR repertoires were conducted in 25 μl containing 105–106 copies of CD3ε transcript equivalents, with 2 U of Taq polymerase (Goldstar; Eurogentec, Brussels, Belgium) in the buffer provided by the supplier. cDNA was amplified (25–28 cycles) using TCR AV- or BV-specific sense primers and fluorescent antisense primer designed to hybridize in the AC or BC segments (28, 29).

The fluorescent products, corresponding to the elongation of individual AV or BV PCR products with various CDR3 sizes, were loaded on polyacrylamide gels and subjected to electrophoresis in an automated DNA sequencer. CDR3 size distribution and signal intensities were then analyzed with Immunoscope software (29, 30). The patterns observed from unprimed splenocytes usually contain six to eight size peaks spaced each by three nucleotides, corresponding to the lengths of in-frame transcripts. The area of each size peak is proportional to the quantity of the TCR transcripts of the corresponding CDR3 length in the sample. Each peak corresponding to a given CDR3 length usually contains multiple distinct sequences. An increase in the height and area of a size peak signals clonal expansion occurring against polyclonal background.

Each product was then diluted v/v in 25 mM EDTA-formamide solution. This mix was heat denatured for 10 min at 80°C, and a 2-μl aliquot was loaded on a 6% (or 4.25%) polyacrylamide 8 M urea gel. Gel electrophoresis was performed on a 373A or 377 DNA sequencer (Applied Biosystems, Foster City, CA). Peak size and fluorescence intensity were determined with Immunoscope software (30).

PCRs were performed using specific BV-BJ primers in 50 μl containing 105 copies of CD3ε transcript equivalents of the cDNA with 2 U of PFU polymerase (Stratagene, La Jolla, CA) in the supplier’s buffer. The elongation starts with 1 min at 94°C, followed by 40 cycles each consisting of 45 s at 94°C, 45 s at 60°C, 45 s at 72°C, and ending with a step at 72°C for 10 min. The PCR product was ethanol precipitated and resuspended in 10 μl formamide containing 0.05% bromophenol blue and 0.05% xylene cyanol. This mixture was heat denatured for 10 min at 80°C and then loaded on an 8% polyacrylamide 7 M urea gel. After migration, PCR products were visualized by silver staining (31) (DNA Silver Staining System; Promega) following manufacturer’s instructions. We usually obtained six to eight bands spaced each by three nucleotides, corresponding to in-frame transcripts of the V-J combinations. Bands of interest, corresponding to a given CDR3 length, were cut out from the gel and disrupted in 40 μl water. A second PCR was conducted using the same primers on 2 μl of the isolated PCR product with 2 U of PFU polymerase (Stratagene) in the supplier buffer for 20 cycles. Further purification was performed on a 15% nondenaturing acrylamide gel in Tris-borate EDTA electrophoresis buffer one time. Staining of this gel was obtained with a 30-min bath in a 0.5 mg/ml ethidium bromide solution. PCR product was electroeluted in Tris-borate EDTA electrophoresis buffer one time, and purity of the sample was estimated by a runoff reaction with the fluorescent BJ primer. We usually obtained 90–95% purity of the final product with the expected size. PCR products were then cloned in pCR Blunt II-TOPO vector using the Zero Blunt TOPO PCR cloning kit (InVitrogen, Carlsbad, CA) as recommended by the manufacturer.

For sequencing purpose, PCR was conducted directly on TOP 10 Escherichia coli colonies (InVitrogen) in a final volume of 30 μl using universal primers RP and M13(-40) designed to hybridize on each side of the polylinker where the BV-BJ PCR product was cloned. Five microliters of this PCR product was treated for 40 min at 37°C with 0.3 U of shrimp alkaline phosphatase (Amersham, Little Chalfont, U.K.) and 3 U of exonuclease I (Amersham) in a final volume of 7 μl. Both enzymes were heat denatured at 80°C for 20 min. Sequencing reactions were conducted directly on these products using M13(-20) primer and with the ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction kit or the ABI PRISM Big Dye Terminator Cycle Sequencing Ready Reaction Kit following manufacturer’s instructions (Applied Biosystems). Reaction products were loaded on a 5% Long Ranger polyacrylamide (FMC Bioproducts, Rockland, ME) 8 M urea gel. Gel electrophoresis was performed on a 373 or 377 DNA sequencer (Applied Biosystems), and CDR3 region corresponding sequences were extracted and analyzed using a software designed for this purpose. The BV segment sequences were taken from Arden et al. (32) and from the sequence of BV locus submitted by Rowen et al. (DNA Data Base in Japan/European Molecular Biology Laboratory/GenBank databases under accession numbers AE000663 and AE000522).

The equation used by Barth et al. (33) and Behlke et al. (34) was used to estimate the maximum probable number of distinct CDR3 sequences found in the cDNA preparation. Namely, the maximum likelihood estimate (MLE) of the number of distinct sequences is a value that maximizes the equation:

\[P(L){=}\ \frac{1}{{\sum_{K{=}M}^{{\infty}}}\ \frac{K!}{(K{-}M)!K^{N}}}\ \frac{L!}{(L{-}M)!L^{N}}\]

in which L is the number of distinct sequences in the cDNA preparation and M is the number of observed distinct sequences among the N sequences performed. A 95% confidence interval was computed as the narrowest interval in which the cumulated values of P(L) is >0.95.

Size of the BV repertoire = Number of distinct sequences found in a CDR3 peak (MLE value) divided by (frequency of BV × frequency of BJ segment × frequency of the CDR3 peak in the Immunoscope profile of the BV-BJ rearrangements under study).

Number of cells present in a CDR3 peak of given length = Total number of T cells in the analyzed sample × frequency of BV × frequency of BJ segment × frequency of the CDR3 peak in the Immunoscope profile of the BV-BJ rearrangements under study.

Number of cells bearing the same CDR3 sequence = Calculated number of cells present in a CDR3 peak of given length divided by the number of distinct nucleotide sequences (MLE value) found in the CDR3 peak.

CD8-positive splenocytes were purified from different animals, and their mRNA was extracted and reverse transcribed into cDNA. Twenty-three BV- and 22 AV-specific PCRs were then conducted in semiquantitative conditions as described in Materials and Methods. BC- or AC-specific fluorescent primers were used for the PCR. These fluorescent products were visualized on a 373A sequencer and analyzed by Immunoscope software. It was shown previously in our laboratory that the CDR3 size profiles in functional AV and BV rearrangements from naive splenocytes display six to eight peaks separated by three nucleotides, corresponding to in-frame transcripts. These peaks give a typical bell-shaped distribution of the CDR3 lengths. However, for BV17 and BV19, which are pseudogenes in C57BL/6 mice (35, 36), in- and out-of-frame transcripts are mixed (data not shown).

In Fig. 1 are shown four representative profiles of AV (A) and BV (B) rearrangements obtained from the CD8+ splenocytes of wild-type, H-2K°/°, H-2D°/°, or double KO C57BL/6 animals; they all show bell-shaped patterns characteristic of polyclonal T lymphocytes. All other rearrangements have been tested in the different strains of mice, and the results clearly show that 21 BV and 22 AV are used and that the CDR3 size profiles of all combinations have a Gaussian-like pattern.

FIGURE 1.

Profiles of the fluorescent AV-AC (A) and BV-BC (B) semiquantitative PCR products obtained from splenocytes of wild-type, H-2K°/°, H-2D°/°, or double KO C57BL/6 mice. The CDR3 patterns shown are representative of the 21 BV and 22 AV combinations tested. The fluorescence intensity is represented in arbitrary units as a function of the CDR3 size in amino acids.

FIGURE 1.

Profiles of the fluorescent AV-AC (A) and BV-BC (B) semiquantitative PCR products obtained from splenocytes of wild-type, H-2K°/°, H-2D°/°, or double KO C57BL/6 mice. The CDR3 patterns shown are representative of the 21 BV and 22 AV combinations tested. The fluorescence intensity is represented in arbitrary units as a function of the CDR3 size in amino acids.

Close modal

We and others have shown previously that the expansion of specific T cell clones or the presence of oligoclonal populations lead to a modification of the CDR3 size profile with the appearance of prominent peaks detected by the Immunoscope technique (37, 38, 39, 40, 41). To demonstrate that, in our working conditions, we were able to detect such oligoclonal expansions, we looked for two invariant chains, i.e., the AV14 AJ15 and the AV19 AJ33 in different subpopulations of T lymphocytes from each strain of mouse. The AV14 segment is preferentially used by two subsets of CD1-restricted NKT cells, which are either CD4CD8 double negative or bear the CD4+ marker (42). The AV19 segment was recently shown to be overexpressed in a double negative subset of T cells (43). As shown in Fig. 2, the characteristic AV14 peak with a CDR3 of 10 aa in size is observed in the CD4+ T cell and more strongly in the double negative CD4/CD8 population from all different strains of mice used in this study. In contrast, a Gaussian-like profile is found in the CD8+ T cells from all groups of mice. Similarly, the Immunoscope spectra observed for the AV19 segment showed a prominent 7-aa CDR3 peak in the double negative T cell populations of the four different strains of mice while it yielded a Gaussian-like profile in the purified CD8+ T cell populations (data not shown).

FIGURE 2.

Profiles of the fluorescent AV14-AC PCR products from CD4+ (left), CD8+ (middle), or CD4CD8 double negative splenocytes from the four categories of animals. The fluorescence intensity is represented in arbitrary units as a function of the CDR3 size in amino acids.

FIGURE 2.

Profiles of the fluorescent AV14-AC PCR products from CD4+ (left), CD8+ (middle), or CD4CD8 double negative splenocytes from the four categories of animals. The fluorescence intensity is represented in arbitrary units as a function of the CDR3 size in amino acids.

Close modal

Overall, these results show that the BV and AV T cell repertoire of CD8+ splenocytes from double KO mice, i.e., selected on MHC class Ib molecules only, is polyclonal without any indication of oligoclonal expansions or invariant chain usage.

The relative BV and AV usage in the CD8+ T cell population from wild type and the three mutant mice is shown in Fig. 3. We used a semiquantitative PCR to precisely compare the different AV and BV. The relative frequency of each BV or AV was calculated by adding the areas of the six or eight peaks corresponding to a single AV or BV combination and dividing this value by the sum of the areas of all peaks from all AV or BV rearrangements. This analysis revealed that the BV and AV usage is similar between the different mice except for the higher usage of the BV5.1 segment in the three mutant mice comparatively to wild-type animals (4–7% in the mutants vs 1% in the wild type).

FIGURE 3.

AV (A) and BV (B) usage of CD8+ splenocytes from C57BL/6 □, H-2K°/° ▪, H-2D°/° ▨, and double KO ▩ mice. In the abcissa are shown the AV and BV chains analyzed. In the ordinate are represented the relative frequency of AV and BV usage calculated as follows: % = area of the six or eight peaks corresponding to a single AV or BV combination/area of all peaks from all AV or BV rearrangements.

FIGURE 3.

AV (A) and BV (B) usage of CD8+ splenocytes from C57BL/6 □, H-2K°/° ▪, H-2D°/° ▨, and double KO ▩ mice. In the abcissa are shown the AV and BV chains analyzed. In the ordinate are represented the relative frequency of AV and BV usage calculated as follows: % = area of the six or eight peaks corresponding to a single AV or BV combination/area of all peaks from all AV or BV rearrangements.

Close modal

Altogether, these results strongly suggest that the CD8+ T cell repertoires selected either by class Ia and Ib (wild-type, H-2K°/°, or H-2D°/° mice) or by class Ib molecules only (double KO mouse) use the various AV and BV rearrangements in the same proportion and without biases in the CDR3 size distribution.

To test the effect of different class I molecules on the BV diversity of CD8+ T cells, we determined the size of the BV repertoire in the four categories of mice used in this study. For this purpose, we used a method that we have recently described (26). Briefly, this method involves the isolation of a CDR3 band of a given length from a defined BV-BJ rearrangement, the cloning of these PCR fragments into a plasmid vector, and an extensive sequencing of the plasmid present in individual bacterial clones. This allowed us to determine the number of distinct nucleotide sequences found in a single CDR3 peak and to evaluate the size of the BV repertoire of naive T splenocytes as indicated in Table I. The equation used by Barth et al. (33) and Behlke et al. (34) was used to estimate the maximum probable number of distinct CDR3 sequences (MLE of the number of distinct sequences) present in a cDNA preparation. Importantly, the values calculated with the MLE were very close to or identical with the number of different sequences obtained experimentally, indicating that we had reached a plateau by exhaustive sequencing (Table I).

Table I.

Estimation of the size of the TCRBV repertoire of CD8+ splenocytes from wild-type and the three MHC class I mutant micea

Mouse No. of CD8+ T splenocytesC57BL/6H-2K°/°H-2D°/°H-2K°/°D°/°H-2K°/°D°/°
10 × 1064 × 1068 × 1067 × 1060.7 × 1061 × 106
Studied BV-BJ rearrangements BV7-BJ1.2 BV10-BJ1.2 BV7-BJ1.2 BV10-BJ1.2 BV7-BJ1.2 BV10-BJ1.2 BV7-BJ1.2 BV10-BJ1.2 BV7-BJ1.2 
% of CD8+ T splenocytes bearing the BV-BJ1.2 rearrangementb 0.38 0.52 0.37 0.46 0.3 0.37 0.35 0.3 0.35 
% of BV-BJ1.2 rearrangements with 6 aa CDR3β lengthc 8.5 11 6.8 10.7 9.8 
Total no. of sequences performed 494 240 91 81 132 172 159 190 92 
MLE of distinct sequencesd 169 (168–170) 90 (89–91) 48 (47–49) 27 (26–28) 34 (33–35) 56 (55–57) 10 (9–11) 28 (27–29) 15 (14–16) 
Deduced size of CD8+ T cells BV repertoire 5.2× 105 2.5× 105 1.2 × 105 8.6× 104 2.8× 105 1.7× 105 2.7× 104 1.2× 105 4.4× 104 
Mouse No. of CD8+ T splenocytesC57BL/6H-2K°/°H-2D°/°H-2K°/°D°/°H-2K°/°D°/°
10 × 1064 × 1068 × 1067 × 1060.7 × 1061 × 106
Studied BV-BJ rearrangements BV7-BJ1.2 BV10-BJ1.2 BV7-BJ1.2 BV10-BJ1.2 BV7-BJ1.2 BV10-BJ1.2 BV7-BJ1.2 BV10-BJ1.2 BV7-BJ1.2 
% of CD8+ T splenocytes bearing the BV-BJ1.2 rearrangementb 0.38 0.52 0.37 0.46 0.3 0.37 0.35 0.3 0.35 
% of BV-BJ1.2 rearrangements with 6 aa CDR3β lengthc 8.5 11 6.8 10.7 9.8 
Total no. of sequences performed 494 240 91 81 132 172 159 190 92 
MLE of distinct sequencesd 169 (168–170) 90 (89–91) 48 (47–49) 27 (26–28) 34 (33–35) 56 (55–57) 10 (9–11) 28 (27–29) 15 (14–16) 
Deduced size of CD8+ T cells BV repertoire 5.2× 105 2.5× 105 1.2 × 105 8.6× 104 2.8× 105 1.7× 105 2.7× 104 1.2× 105 4.4× 104 

a The percentage of a given CDR3 length peak in a defined BV-BJ rearrangement was calculated by dividing the peak area studied by the total area of the different peaks found in this BV-BJ1.2 combination. The percentages of BV7 or BV10 chain usage in T splenocytes were determined from Fig. 3 for all categories of mice. The BJ1.2 usage was derived from Kato et al. (44 ) and Candeias et al. (45 ). The size of the CD8+ T cell repertoire was calculated as follows: Number of distinct CDR3 sequences found in given peak × 1/percentage of a given CDR3 length peak in a defined BV-BJ1.2 (to correct for all CDR3 peaks present in a single BV-BJ1.2 rearrangement) × 1/percentage of the BV studied (to correct for all Bvs) × 100/5 (to correct for all BJs since BJ1.2 was used in 5% of the rearrangements).

b The percentages of BV7 and BV10 usage among CD8+ T splenocytes were determined as described in Fig. 3. The BJ1.2 usage was derived from Kato et al. (44 ).

c The percentage of BV-BJ rearrangements bearing the indicated CDR3β length was calculated from the Immunoscope profile.

d MLE was calculated as described in Materials and Methods.

The results show that the TCR BV repertoire in wild-type mice is ∼2.5–5.2 × 105, in the H-2D°/° 1.7–2.8 × 105, in H-2K°/° 0.86–1.2 × 105, and in the double KO 0.3–1.2 × 105 (Table I). There is a linear relation between the size of the BV repertoire and the number of CD8+ cells (Fig. 4). The average number of T cells bearing the same BV chain was calculated from the equation of the linear regression curve and found to be equal to 29. Thus, the size reduction of the analyzed repertoires found in the different strains of mice correlates well with the reduction in CD8+ T cell numbers, which is limited in the H-2D°/° mouse (30% of wild type), substantial in the H-2K°/° mouse (60%), and profound in the double KO mouse (90%) (Perarneau et al. (25) and Table I).

FIGURE 4.

Size estimates of the BV repertoire of CD8+ splenocytes. In the abcissa are plotted the number of CD8+ splenocytes found in wild-type, H-2D°/°, H-2K°/°, and double KO C57BL/6 animals. The size estimates of the BV repertoires are shown in the ordinate. The estimates are calculated from the BV10-BJ1.2 ▪ and from the BV7-BJ1.2 □ rearrangements. The linear regression is drawn from all values. The average number of T cells sharing the same BV chain was calculated from the equation of the linear regression curve and found to be equal to 29.

FIGURE 4.

Size estimates of the BV repertoire of CD8+ splenocytes. In the abcissa are plotted the number of CD8+ splenocytes found in wild-type, H-2D°/°, H-2K°/°, and double KO C57BL/6 animals. The size estimates of the BV repertoires are shown in the ordinate. The estimates are calculated from the BV10-BJ1.2 ▪ and from the BV7-BJ1.2 □ rearrangements. The linear regression is drawn from all values. The average number of T cells sharing the same BV chain was calculated from the equation of the linear regression curve and found to be equal to 29.

Close modal

In this work, we have studied the role of the MHC class Ib molecules in the selection and maintenance of CD8+ T splenocytes of H-2K°/°, H-2D°/°, and double KO C57BL/6 mice. More specifically, we have addressed the following questions: 1) are MHC class Ib molecules able to select oligoclonal or polyclonal AV and BV repertoires; and 2) what is the size of the BV repertoire of CD8+ T splenocytes in double KO animals?

Perarnau et al. (25) have observed that the CD8αβ T cell number is reduced by 30% in H-2D°/°, 60% in H-2K°/°, and 90% in double KO animals when compared with C57BL/6 mice. Because class Ia-negative mice, like wild-type animals, develop potent Qa1-restricted CTL responses against Listeria monocytogenes, it is most likely that observations made in double KO mice also apply to wild-type animals. It is striking that, although the pattern and level of expression of class I molecules detected on thymus cryosections with anti-mouse β2-microglobulin mAb (25) are similar in class Ia negative and wild-type mice, very few CD8 αβ lymphocytes are found in class Ia-negative animals (25). Many nonexclusive possibilities may account for this paucity: 1) expression of class Ib molecules often is restricted to certain tissues, and may be missing or expressed at low levels on cells responsible for positive selection; 2) a weaker CD8 interaction than with class Ia molecules has been documented with Q7 and Q9 molecules; 3) class Ib molecules may present a limited set of self-peptides as shown for H-2M3 and Qa1; 4) class Ib molecules may not be expressed in the periphery in sufficient amounts and/or on the right cells, resulting in a reduced T cell survival; and 5) finally, the class Ib-mediated positive and negative selection processes may be unbalanced as compared with class Ia in favor of the latter. Considering the large number of class Ib genes and molecules and the limited number of specific reagents at hand, none of these possibilities should be excluded.

The limited number of CD8+ T cells educated and maintained by class Ib molecules only led us to analyze more precisely their AV and BV repertoire. Some class Ib molecules have been shown to promote the education of T cells with a limited TCR diversity. NK1 T lymphocytes, which are restricted by CD1, a class Ib-like molecule not encoded in the MHC, have a limited repertoire diversity. These NK1 T lymphocytes are CD4+ or CD4CD8 and display an invariant α-chain with a 10-amino-acid long CDR3 associated with a limited set of BV segments (42). More recently, another subset of T lymphocytes expressing an invariant α-chain was described in human, mouse, and cattle (43). This invariant chain, AV19AJ33 in mouse, is expressed by a subpopulation of CD4CD8 T lymphocytes selected on a yet unidentified MHC class Ib molecule different from CD1 (43). This AV19AJ33 chain pairs predominantly with BV8 and BV6 segments (43). The existence of MHC class Ib-restricted T cell subpopulations, which display a limited T cell repertoire, led us to evaluate the diversity of the CD8αβ T lymphocytes in mice lacking one or all classical MHC class I molecules.

We first analyzed RNA from purified CD8αβ splenocytes from H-2K°/°, H-2D°/°, double KO, and C57BL/6 animals by the Immunoscope method. In the three different strains of mice tested, CD8αβ profiles were bell-shaped and similar to those found in wild-type animals. No clonal or oligoclonal expansion was detected. Moreover, using a semiquantitative PCR, we observed that the AV and BV usage in the three deficient strains closely matched that seen in C57BL/6 (Fig. 3). The only significant difference between mutant mice and wild-type animals concerns the BV5.1 segment. The decrease of BV5.1 in wild-type mice is probably not due to a superantigenic effect because 1) no endogenous superantigen restricted by MHC class I molecules has been described yet; and 2) in single KO mice, the expression of only one of the class Ia molecules (H-2Kb or H-2Db) does not result in size reduction of the BV5.1 CD8+ T cell subset.

These analyses showed clearly that the CD8αβ T cell repertoires of the three mutant mice match closely the wild-type one and use similarly the different AV and BV rearrangements without biases in the CDR3 size distribution. However, the methods used provided no information on the number of distinct CDR3 sequences present in a given BV rearrangement. To evaluate the diversity of the BV repertoires in the four strains of mice, we used a method recently described by Casrouge et al. (26). In the four strains of mice, the BV repertoire sizes were calculated from two different rearrangements, the BV7-BJ1.2 and the BV10-BJ1.2. When we plotted the BV repertoire sizes vs the number of CD8αβ lymphocytes, a linear regression curve was obtained. From it we calculated the average number of T cells sharing the same BV chain: 29 (Fig. 4). This value is in good agreement with those previously found for unfractionated T splenocytes from C57BL/6 and DBA/2 mice (26).

In summary, this work shows that the diversity (usage of AV and BV segments, CDR3 sizes) of the CD8+ T cell repertoire selected by class Ib molecules is similar to the diversity of the CD8+ repertoire in normal mice, taking into account the 90% reduction of CD8+ splenocytes. Furthermore, we have estimated the size of the BV repertoire in the four different strains of mice and we have found that the BV repertoire size is proportional to the overall number of CD8+ splenocytes. This observation implies that BV diversity is positively correlated with the number of CD8+ cells. Considering that certain class Ib molecules have evolved to present peculiar sets of peptides, i.e., N-formylated peptides for H-2 M3, endoplasmic reticulum signal peptides for Qa1 and complement functionally class Ia molecules, significant diversification of their corresponding CD8+ T cell repertoire is evidently advantageous.

We thank Dr. Irina Apostolou, Dr. Iris Motta, and Dr. Jean-Pierre Cabaniols for discussion and critical review of the manuscript. The help of Annick Lim for technical advice with the semiquantitative PCR is greatly appreciated.

1

This work was supported by Association de Recherches contre le Cancer, Ligue de Recherches contre le Cancer, and by a grant from the European Economic Community (no. 96-0077). D.L. was supported by fellowships from Association des Universités Partiellement ou Entièrement de Langue Française-Université des Réseaux d’Expression, from the International Network of the Pasteur Institute, and from Collège de France.

4

Abbreviations used in this paper: AV, Vα; BV, Vβ; KO, knockout; CDR, complementarity-determining regions; °/°, deficient; MLE, maximum likelihood estimate.

1
Wang, C. R., A. R. Castano, P. A. Peterson, C. Slaughter, K. F. Lindahl, J. Deisenhofer.
1995
. Nonclassical binding of formylated peptide in crystal structure of the MHC class Ib molecule H2–M3.
Cell
82
:
655
2
O’Callaghan, C. A., J. Tormo, B. E. Willcox, C. D. Blundell, B. K. Jakobsen, D. I. Stuart, A. J. McMichael, J. I. Bell, E. Y. Jones.
1998
. Production, crystallization, and preliminary x-ray analysis of the human MHC class Ib molecule HLA-E.
Protein Sci.
7
:
1264
3
O’Callaghan, C. A., J. I. Bell.
1998
. Structure and function of the human MHC class Ib molecules HLA-E, HLA-F and HLA-G.
Immunol. Rev.
163
:
129
4
Zeng, Z., A. R. Castano, B. W. Segelke, E. A. Stura, P. A. Peterson, I. A. Wilson.
1997
. Crystal structure of mouse CD1: an MHC-like fold with a large hydrophobic binding groove.
Science
277
:
339
5
Aldrich, C. J., A. DeCloux, A. S. Woods, R. J. Cotter, M. J. Soloski, J. Forman.
1994
. Identification of a Tap-dependent leader peptide recognized by alloreactive T cells specific for a class Ib antigen.
Cell
79
:
649
6
Braud, V., E. Y. Jones, A. McMichael.
1997
. The human major histocompatibility complex class Ib molecule HLA-E binds signal sequence-derived peptides with primary anchor residues at positions 2 and 9.
Eur. J. Immunol.
27
:
1164
7
Wang, C. R., B. E. Loveland, K. F. Lindahl.
1991
. H-2M3 encodes the MHC class I molecule presenting the maternally transmitted antigen of the mouse.
Cell
66
:
335
8
Kawano, T., J. Cui, Y. Koezuka, I. Toura, Y. Kaneko, K. Motoki, H. Ueno, R. Nakagawa, H. Sato, E. Kondo, et al
1997
. CD1d-restricted and TCR-mediated activation of vα14 NKT cells by glycosylceramides.
Science
278
:
1626
9
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
10
Joyce, S., A. S. Woods, J. W. Yewdell, J. R. Bennink, A. D. De Silva, A. Boesteanu, S. P. Balk, R. J. Cotter, R. R. Brutkiewicz.
1998
. Natural ligand of mouse CD1d1: cellular glycosylphosphatidylinositol.
Science
279
:
1541
11
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
12
Garboczi, D. N., P. Ghosh, Q. R. Fan, W. E. Biddison, D. C. Wiley.
1996
. Structure of the complex between human T-cell receptor, viral peptide and HLA-A2.
Nature
384
:
134
13
Garcia, K. C., M. Degano, R. L. Stanfield, A. Branmark, M. R. Jackson, P. A. Peterson, L. Teytou, I. A. Wilson.
1996
. An αβ T cell receptor structure at 2.5 Å and its orientation in the TCR-MHC complex.
Science
274
:
209
14
Davis, M. M., P. A. Bjorkman.
1988
. T-cell antigen receptor gene and T-cell recognition.
Nature
334
:
395
15
Wilson, R. K., E. Lai, P. Concannon, R. K. Barth, L. E. Hood.
1988
. Structure, organization and polymorphism of murine and human T-cell receptor α and β chain gene families.
Immunol. Rev.
101
:
149
16
Kappler, J. W., N. Roehm, P. Marrack.
1987
. T cell tolerance by clonal elimination in the thymus.
Cell
49
:
273
17
Kisielow, P., H. Blüthmann, U. D. Staerz, M. Steinmetz, H. von Boehmer.
1988
. Tolerance in T-cell-receptor transgenic mice involves deletion of nonmature CD4+8+ thymocytes.
Nature
333
:
742
18
Kisielow, P., H. S. Teh, H. Blüthmann, H. von Boehmer.
1988
. Positive selection of antigen-specific T cells in thymus by restricting MHC molecules.
Nature
335
:
730
19
Scott, B., H. Blüthmann, H. S. Teh, H. von Boehmer.
1989
. The generation of mature T cells require interaction of the αβ T-cell receptor with major histocompatibility antigens.
Nature
338
:
591
20
Sha, W. C., C. A. Nelson, R. D. Newberry, D. M. Kranz, J. H. Russell, D. Y. Loh.
1988
. Positive and negative selection of an antigen receptor on T cells in transgenic mice.
Nature
336
:
73
21
Pircher, H., K. Burki, R. Lang, H. Hengartner, R. M. Zinkernagel.
1989
. Tolerance induction in double specific T-cell receptor transgenic mice varies with antigen.
Nature
342
:
559
22
Hogquist, K. A., A. J. Tomlinson, W. C. Kieper, M. A. McGargill, M. C. Hart, S. Naylor, S. C. Jameson.
1997
. Identification of a naturally occurring ligand for thymic positive selection.
Immunity
6
:
389
23
Hu, Q., C. R. Bazemore Walker, C. Girao, J. T. Opferman, J. Shabanowitz, D. F. Hunt, P. G. Ashton-Rickardt.
1997
. Specific recognition of thymic self-peptides induces the positive selection of cytotoxic lymphocytes.
Immunity
7
:
221
24
Berg, R. E., M. F. Princiotta, S. Irion, J. A. Moticka, K. R. Dahl, U. D. Staerz.
1999
. Positive selection of an H2–M3 restricted T cell receptor.
Immunity
11
:
33
25
Perarnau, B., M. F. Saron, B. R. San Martin, N. Bervas, H. Ong, M. J. Soloski, A. G. Smith, J. M. Ure, J. E. Gairin, F. A. Lemonnier.
1999
. Single H2Kb, H2Db and double H2KbDb knockout mice: peripheral CD8+ T cell repertoire and anti-lymphocytic choriomeningitis virus cytolytic responses.
Eur. J. Immunol.
29
:
1243
26
Casrouge, A., E. Beaudoing, S. Dalle, C. Pannetier, J. Kanellopoulos, P. Kourilsky.
2000
. Size estimate of the α β TCR repertoire of naive mouse splenocytes.
J. Immunol.
164
:
5782
27
Apostolou, I., A. Cumano, G. Gachelin, P. Kourilsky.
2000
. Evidence for two subgroups of CD4CD8 NKT cells with distinct TCRαβ repertoires and differential distribution in lymphoid tissues.
J. Immunol.
165
:
2481
28
Pannetier, C., S. Delassus, S. Darche, C. Saucier, P. Kourilsky.
1993
. Quantitative titration of nucleic acids by enzymatic amplification reactions run to saturation.
Nucleic Acids Res.
21
:
577
29
Pannetier, C., M. Cochet, S. Darche, A. Casrouge, M. Zoller, P. Kourilsky.
1993
. The sizes of the CDR3 hypervariable regions of the murine T-cell receptor β chains vary as a function of the recombined germ-line segments.
Proc. Natl. Acad. Sci. USA
90
:
4319
30
Pannetier, C., J.-P. Levraud, J. Even, P. Kourilsky.
1997
. The Immunoscope approach for the analysis of T cell repertoires. J. R. Oksenberg, ed.
The Antigen T Cell Receptor: Selected Protocols and Applications
287
Chapman & Hall, Austin, TX.
31
Raaphorst, F. M., J. Tami, I. E. Sanz.
1996
. Cloning of size-selected human immunoglobulin heavy-chain rearrangements from third complementarity-determining region fingerprint profiles.
Biotechniques
20
:
78
32
Arden, B., S. P. Clark, D. Kabelitz, T. W. Mak.
1995
. Mouse T-cell receptor variable gene segments families.
Immunogenetics
42
:
501
33
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 V β gene segments.
Nature
316
:
517
34
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
35
Louie, M. C., C. A. Nelson, D. Y. Loh.
1989
. Identification and characterization of new murine T cell receptor β chain variable region (Vβ) genes.
J. Exp. Med.
170
:
1987
36
Wade, T., J. Bill, P. C. Marrack, E. Palmer, J. W. Kappler.
1988
. Molecular basis for the nonexpression of vβ17 in some strains of mice.
J. Immunol.
141
:
2165
37
Cibotti, R., J. P. Cabaniols, C. Pannetier, C. Delarbre, I. Vergnon, J. M. Kanellopoulos, P. Kourilsky.
1994
. Public and private V β T cell receptor repertoires against hen egg white lysozyme (HEL) in nontransgenic versus HEL transgenic mice.
J. Exp. Med.
180
:
861
38
Gapin, L., Y. Bravo de Alba, A. Casrouge, J.-P. Cabaniols, P. Kourilsky, J. Kanellopoulos.
1998
. Dendritic cells focus T-cell response against an immunodominant peptide: study in the hen egg-white lysozyme (HEL) model.
J. Immunol.
160
:
1555
39
Gapin, L., J. P. Levraud, J. Kanellopoulos, P. Kourilsky.
1996
. Public T cells in response to foreign and self-tumor antigens. J. Banchereau, and B. Dodet, and R. Schwartz, and E. Trannoy, eds.
Immune Tolerance
21
Elsevier, Paris.
40
Douillard, P., C. Pannetier, R. Josien, S. Menoret, P. Kourilsky, J. P. Soulillou, M. C. Cuturi.
1996
. Donor-specific blood transfusion induced tolerance in adult rats with a dominant TcR-V β rearrangement in heart allografts.
J. Immunol.
157
:
1250
41
Dietrich, P. Y., A. Caignard, A. Lim, V. Chung, J. L. Pico, C. Pannetier, P. Kourilsky, T. Hercend, J. Even, F. Triebel.
1994
. In vivo T-cell clonal amplification at time of acute graft-versus-host disease.
Blood
84
:
2815
42
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
43
Tilloy, F., E. Treiner, S. H. Park, C. Garcia, F. Lemonnier, H. de la Salle, A. Bendelac, M. Bonneville, O. Lantz.
1999
. An invariant T cell receptor α chain defines a novel TAP-independent major histocompatibility complex class Ib-restricted α/β T cell subpopulation in mammals.
J. Exp. Med.
189
:
1907
44
Kato, T., S. Suzuki, H. Sasakawa, K. Masuko, Y. Ikeda, K. Nishioka, K. Yamamoto.
1994
. Comparison of the J β gene usage among different T cell receptor V β families in spleens of C57BL/6 mice.
Eur. J. Immunol.
24
:
2410
45
Candeias, S., C. Waltzinger, C. Benoist, D. Mathis.
1991
. The V β 17+ T cell repertoire: skewed J β usage after thymic selection; dissimilar CDR3s in CD4+ versus CD8+ cells.
J. Exp. Med.
174
:
989