T cells that survive thymic selection express a diverse array of unique heterodimeric αβ TCRs that mediate peptide-MHC Ag recognition. The proportion of the total T cell repertoire that expresses a particular Vβ protein may be determined by a variety of factors: 1) germline preference for use of particular Vβ genes; 2) allelic effects on the expression of different Vβ genes; and 3) HLA effects on the expression of different Vβ genes (acting via thymic selection and/or peripheral mechanisms). In this study, we show that Vβ usage by human CD4+ and CD8+ T cells in neonatal and adult donors is highly correlated between unrelated individuals, suggesting that a large proportion of the observed pattern of Vβ expression is determined by factors intrinsic to the TCR-β locus. The presence of identical TCR alleles (within an individual) leads to a significantly better correlation between CD4+ and CD8+ T cells with respect to Vβ expression; these effects are, however, relatively minor. The sharing of HLA alleles between individuals also leads to an increased correlation between their Vβ expression patterns, although this did not reach statistical significance. We therefore conclude that the correlation in Vβ expression patterns between CD4+ and CD8+ T cells can be explained predominantly by germline TCR-β locus factors and not TCR-β allelic or HLA effects.

During T cell development, thymocytes proceed through a series of selection processes leading to the establishment of a dominant pool of αβ TCR-expressing CD4+ and CD8+ T cells that recognize peptide Ags presented in the context of MHC class II and I, respectively. The TCR β-chain protein is generated by the somatic recombination of germline-encoded V, D, and J gene segments (1). This process does not seamlessly join one segment to the other but rather incorporates non-germline-encoded contact sequences between V-D and D-J regions, resulting in a hypervariable region designated the third CDR (CDR3) (2). The CDR3 loops of the TCR α- and β-chains are the main contact domains with the MHC-presented peptide, whereas the germline-encoded CDR1 and 2 predominantly contact the conserved helical residues of the MHC (3). In the αβ T cell lineage, the β-chain is expressed first (4), and thymocytes expressing a functional TCR-β protein are selected before the TCR-α locus rearranges. In this selection process, the TCR β-chain dimerizes with an invariant 33-kDa glycoprotein, the pTα chain, to form the pre-TCR (5, 6). This combination is a critical selection event, which allows only cells with a functional Vβ protein to survive, differentiate into CD4/CD8+ double-positive thymocytes, and proliferate (7, 8), a process often referred to as β-selection (9, 10, 11). A ligand for the pre-TCR has not been identified, and it is generally held that the pre-TCR signals in a cell-autonomous or constitutive manner (12, 13, 14).

It is currently unknown how the human Vβ repertoire is established. The main TCR protein-based filtering process in the thymus selects for T cells that recognize a self-peptide Ag presented by self-MHC with low affinity (15). It would, therefore, follow that the Vβ repertoire is selected by MHC. However, there is some evidence from mouse models that TCR-β locus factors affect the peripheral use of Vβ proteins (16, 17). To date, there has been no systematic attempt to quantify the relative contribution of both factors in either mice or humans. In this study, we assessed the role of both TCR-β locus and HLA-related factors in a heterogeneous population of neonatal and adult donors and find that the TCR-β locus factors play a dominant role in the establishment of the mature peripheral Vβ repertoire in both CD4+ and CD8+ T cells.

PBMC were obtained from 23 healthy adult donors, ranging in age from 20 to 74 years, and 31 umbilical cord blood donors, provided by Dr. P. Rubinstein of the New York Blood Center (New York, NY). The adult donor PBMC were obtained with written informed consent under institutional review board-approved National Heart, Lung, and Blood Institute transplant protocols. High-resolution HLA typing was performed on DNA extracted from 16 of 23 adult donor samples using sequence-based genotyping (18); the data are shown in Table I. Note that allele designations marked with an “a” were ambiguous (18), which precluded high-resolution designation.

Table I.

High-resolution HLA-typing data of 18 donors used in the current study

DonorHLA-AHLA-BHLA-CHLA-DR
2601, 6801 3503, 3901 0401, 1203 1401, 1601 
240201, 240301 380101, 5001 0602, 1203 1101, 1502 
010101, 6801 1401, 380101 0602, 1203 0701, 1401 
020101, 110101 3501, 4001 0304, 0401 0101, 1302 
020101, 260101 130201, 570101 0602, 0701 0701 
020101, 310102 510101 150201 0101, 0402 
030101, 110101 3503, 510101 040101 0407, 1104 
2301, 3301 140201, 4901 0701, 0802 0102, 1104 
02,a31a 18,a391301 05,a07a 0301, 0411 
10 240201, 290201 3901, 4403 120301, 160101 0701, 1501 
11 010101, 0301 080101, 3503 040101, 0701 0301, 0401 
12 010101, 6802 080101, 140201 0701, 0802 0301, 1303 
13 020101, 6802 150101, 570101 030301, 0602 0701 
14 010101, 300101 080101, 4402 0501, 0701 0301, 1104 
15 2301, 300201 130201, 4101 0602, 17a 0701, 0803 
16 0205, 6601 4402, 5001 0501, 0602 1401, 1501 
17 3,b26b 7,b35b 4,b7b 0101, 1501 
18 1,b0201 62,b63b 3b 1301, 1302 
DonorHLA-AHLA-BHLA-CHLA-DR
2601, 6801 3503, 3901 0401, 1203 1401, 1601 
240201, 240301 380101, 5001 0602, 1203 1101, 1502 
010101, 6801 1401, 380101 0602, 1203 0701, 1401 
020101, 110101 3501, 4001 0304, 0401 0101, 1302 
020101, 260101 130201, 570101 0602, 0701 0701 
020101, 310102 510101 150201 0101, 0402 
030101, 110101 3503, 510101 040101 0407, 1104 
2301, 3301 140201, 4901 0701, 0802 0102, 1104 
02,a31a 18,a391301 05,a07a 0301, 0411 
10 240201, 290201 3901, 4403 120301, 160101 0701, 1501 
11 010101, 0301 080101, 3503 040101, 0701 0301, 0401 
12 010101, 6802 080101, 140201 0701, 0802 0301, 1303 
13 020101, 6802 150101, 570101 030301, 0602 0701 
14 010101, 300101 080101, 4402 0501, 0701 0301, 1104 
15 2301, 300201 130201, 4101 0602, 17a 0701, 0803 
16 0205, 6601 4402, 5001 0501, 0602 1401, 1501 
17 3,b26b 7,b35b 4,b7b 0101, 1501 
18 1,b0201 62,b63b 3b 1301, 1302 
a

Ambiguous alleles.

b

Serological typing

mAbs to CD3, CD4, and CD8 were obtained from BD Biosciences. The 22 Abs to the Vβ proteins were obtained from Biodesign and from Immunotech/Beckman Coulter. This set of mAb detects 27 of 46 (59%) functional Vβ proteins. In our set of samples, we found that, using this panel of anti-Vβ mAb, a median of 55% of all expressed Vβ proteins was detected. The IMGT guidelines were followed for the Vβ nomenclature (http://imgt.cines.fr/).

PBMC were stained for CD3, CD4, and CD8, and individual Vβ proteins using standard procedures (19, 20). Data acquisition was performed on a BD Biosciences FACSCalibur and LSRII, and analyzed using CellQuest and DIVA software (both obtained from BD Biosciences). Quadrants were set using CD3, then CD8 and CD4; Vβ expression was determined within CD4high and CD8high expressing CD3+ cells in the lymphocyte gate.

The correlation coefficient between the Vβ frequencies in CD4+ and CD8+ T cells between and within individuals was calculated using Matlab’s “corr” function (Matlab). A nonparametric (ranked; Spearman) correlation coefficient was calculated because Vβ frequencies for most donors did not follow a normal distribution. Vβ28 (IMGT nomenclature; Vβ3.1 in the nomenclature of Arden et al. (21), which was used by Posnett et al. (22)) was excluded from the analysis due to the known effect of a single polymorphism in the 23-bp spacer region on its usage by T cells.

To test whether the distribution of CD4/CD8 Vβ correlations within individuals equaled the distribution of CD4/CD8 correlations between individuals, a modification of the Wilcoxon rank-sum test was used. The correlations were arranged in an n by n matrix with the same ordering of the n individuals for rows and columns. Thus, the diagonal entries were the within individual correlations and the off-diagonal entries were the between individual correlations. A permutation procedure was used where the rows were shuffled while leaving the columns fixed. Ten thousand random shuffles were generated. The test statistic used in this permutation procedure was the two-sample Wilcoxon rank-sum statistic for the within-individual (diagonal) and between-individual (off-diagonal) groups.

The effect of HLA on Vβ frequencies was calculated in the adult donor data set by determining the Spearman correlation coefficient between Vβ frequencies (with the exclusion of Vβ28) in CD8+ (n = 14) or CD4+ (n = 18) T cells sharing no HLA alleles or ≥1 HLA class I or HLA-DR allele. To test whether an HLA match increased the correlation between two individuals, a different modification of the Wilcoxon rank-sum test was used. The correlations for all pairs of individuals for a population of cells (e.g., CD8+) using a specific matching criterion (e.g., HLA class I matched) were separated into two groups: pairs with 0 matches and pairs with ≥1 match. If there was no effect of matching, then the ranks of these correlations should be similar for the 0-match and ≥1-match groups. A permutation procedure was used where the n by n matrix of correlations was fixed while the n by n matrix of matches was shuffled; a permuted list of names was used to define both the row and column labels of the (permuted) matrix of matches. Ten thousand random shuffles were generated. The test statistic used in this permutation procedure was the two sample Wilcoxon rank-sum statistic for the 0 match and ≥1 match groups.

To assess the generic effects of the TCR-β locus (independent of allelic and HLA effects), we determined the correlation between the hierarchy of Vβ expression in CD4+ and CD8+ T cells across unrelated individuals in the adult donor set by calculating the Spearman correlation coefficient between Vβ expression in CD4+ T cells of donor x and Vβ expression in CD8+ T cells of donor y. An example of this analysis is shown in Fig. 1,A. The median correlation coefficient was 0.651 (Table II), indicating that between any two randomly chosen individuals, variation in the Vβ hierarchy of CD4+ T cells in one individual would explain ∼42% of the variation in the Vβ hierarchy of CD8+ T cells in another. This analysis was repeated in the neonatal blood samples, where the correlation between CD4+ T cells of donor x and CD8+ T cells of donor y was 0.686, indicating that ∼47% of the Vβ hierarchy can be explained by generic TCR-β locus factors (Fig. 2).

FIGURE 1.

TCR-β locus factors strongly affect the use of Vβ proteins by mature CD4+ and CD8+ T cells. Shown are examples of the analysis of the interindividual (A) and intraindividual (B) correlation between Vβ usage by CD4+ and CD8+ T cells. C, Vβ usage is not affected by the position of the Vβ gene at the TCR-β locus. D and E, Correlation between Vβ28 (D) and 5-1 (E) usage by CD4+ and CD8+ T cells within individuals.

FIGURE 1.

TCR-β locus factors strongly affect the use of Vβ proteins by mature CD4+ and CD8+ T cells. Shown are examples of the analysis of the interindividual (A) and intraindividual (B) correlation between Vβ usage by CD4+ and CD8+ T cells. C, Vβ usage is not affected by the position of the Vβ gene at the TCR-β locus. D and E, Correlation between Vβ28 (D) and 5-1 (E) usage by CD4+ and CD8+ T cells within individuals.

Close modal
Table II.

Spearman correlation analysis of CD4++ and CD8++ T cells between and within individuals

GroupNo. PairsMedian rp (rank sum)
Adults, within donor 23 0.7209 <0.0001 
 Between donors 506 0.6510  
UCB, within donor 31 0.7293 <0.0001 
 Between donors 930 0.6857  
Within individuals 23 0.7209 0.5007 
 31 0.7293  
GroupNo. PairsMedian rp (rank sum)
Adults, within donor 23 0.7209 <0.0001 
 Between donors 506 0.6510  
UCB, within donor 31 0.7293 <0.0001 
 Between donors 930 0.6857  
Within individuals 23 0.7209 0.5007 
 31 0.7293  
FIGURE 2.

Spearman correlation coefficient distribution between Vβ expression patterns in CD4+ and CD8+ T cells. Shown are the frequency distributions of the correlation coefficient calculated between Vβ expression patterns in CD4+ and CD8+ T cells between (□) and within (▪) adult (A) and neonatal (B, UCB) donors. C, Comparison of intraindividual Vβ expression patterns between CD4+ and CD8+ T cells in neonatal and adult donor samples. D, Effect of age on correlation of intraindividual Vβ usage between CD4+ and CD8+ T cells (right) as determined in neonatal (age set at 0) and adult donors.

FIGURE 2.

Spearman correlation coefficient distribution between Vβ expression patterns in CD4+ and CD8+ T cells. Shown are the frequency distributions of the correlation coefficient calculated between Vβ expression patterns in CD4+ and CD8+ T cells between (□) and within (▪) adult (A) and neonatal (B, UCB) donors. C, Comparison of intraindividual Vβ expression patterns between CD4+ and CD8+ T cells in neonatal and adult donor samples. D, Effect of age on correlation of intraindividual Vβ usage between CD4+ and CD8+ T cells (right) as determined in neonatal (age set at 0) and adult donors.

Close modal

To assess whether proximity to the DJβ cluster at the 3′ end of this locus affected the Vβ usage by mature T cells, all Vβs analyzed in CD4+ T cells from the neonatal donors were arranged according to their position along the TCR-β locus. As can be seen in Fig. 1 C, the proximity to the DJβ cluster did not correlate with Vβ usage by T cells, suggesting that Vβ usage is not simply determined by geographics but by other TCR-β-associated factors.

The TCR-β locus contains 284 single nucleotide polymorphisms (SNPs)3 which may affect Vβ usage by CD4+ and CD8+ T cells as has been established for Vβ28 (22). The allelic effect was determined in adult and neonatal T cells by calculating the Spearman correlation coefficient between the Vβ repertoire in CD4+ and CD8+ T cells of the same individual. This comparison assumes that Vβ expression is not HLA-dependent because CD4+ and CD8+ T cells are selected in the thymus by HLA class II and I, respectively. However, within the same individual, the same set of SNPs affects the same set of Vβ proteins in both CD4+ and CD8+ T cells whereas between individuals these alleles will not be matched in most cases due to random sampling. We found that the median correlation within adult individuals (r = 0.7209) was significantly higher than between individuals (p < 0.0001) (Table II; an example is shown in Fig. 1,B), indicating that matching of TCR-β alleles led to a significantly better correlation between CD4+ and CD8+ T cell Vβ hierarchy. However, the allelic effects of different Vβ alleles were relatively small, explaining only an additional ∼10% of the expression hierarchy (Fig. 2). This same analysis was repeated in the neonatal blood samples where it was found that the correlation coefficient within individuals (r = 0.7293) was again significantly higher than between individuals (r = 0.6857; p < 0.0001), suggesting that allelic effects explained an additional ∼6% of the hierarchy.

Surprisingly, the within-individual correlation for Vβ expression between CD4+ and CD8+ T cells was not significantly different within the adult donors than within the umbilical cord blood (UCB) (p = 0.5007; Fig. 2,C). Similarly, the within-individual correlation between CD4+ and CD8+ T cell Vβ expression was not significantly correlated with age in the adult donors (Fig. 2 D), suggesting an overall maintenance of the expression hierarchy over time. Thus, these data suggest that TCR-β locus factors contribute ∼50% to the shaping of the mature Vβ repertoire in both the naive (UCB) and adult T cell repertoires.

As discussed above, Vβ28 was excluded from the analysis due to its known allelic effects on Vβ expression. To confirm the allelic effects of Vβ28 and to investigate which other Vβ contribute to the allelic effects seen in the absence of Vβ28, we analyzed for each specific Vβ the correlation between the expression in CD4+ and CD8+ T cells within each donor (using the Spearman correlation). This analysis showed (Fig. 1,D) that Vβ28 expression was very significantly correlated between CD4+ and CD8+ T cells (r = 0.90415; p = 3.28 × 10−6), indicating a strong allelic effect. Several other Vβ proteins also exhibited strong correlations between CD4+ and CD8+ T cells, including Vβ5-1 (Fig. 1,E), Vβ12-3/12-4, and Vβ4-1/4-2/4-3 (Table III). However, these correlations were not significant after correction for the number of comparisons.

Table III.

Correlation between individual Vβ frequencies in CD4+ and CD8+ T cells across the adult donors

r (Spearman)p
0.415 0.050 
3-1 0.133 0.544 
4-1, 4-2, 4-3 0.541 0.008 
5-1 0.560 0.006 
5-5 0.421 0.051 
5-6 0.290 0.191 
6-5, 6-6, 6-9 0.461 0.027 
6-6 0.384 0.071 
0.454 0.030 
10-3 0.448 0.032 
11-2 0.176 0.512 
12-3, 12-4 0.519 0.011 
13 0.170 0.439 
14 −0.005 0.980 
18 0.459 0.064 
19 0.357 0.095 
20-1 −0.184 0.401 
25-1 −0.084 0.702 
27 −0.064 0.773 
28 0.904 3.28 × 10−6 
30 0.350 0.102 
r (Spearman)p
0.415 0.050 
3-1 0.133 0.544 
4-1, 4-2, 4-3 0.541 0.008 
5-1 0.560 0.006 
5-5 0.421 0.051 
5-6 0.290 0.191 
6-5, 6-6, 6-9 0.461 0.027 
6-6 0.384 0.071 
0.454 0.030 
10-3 0.448 0.032 
11-2 0.176 0.512 
12-3, 12-4 0.519 0.011 
13 0.170 0.439 
14 −0.005 0.980 
18 0.459 0.064 
19 0.357 0.095 
20-1 −0.184 0.401 
25-1 −0.084 0.702 
27 −0.064 0.773 
28 0.904 3.28 × 10−6 
30 0.350 0.102 

We next assessed whether HLA matching significantly affected the hierarchy of Vβ expression. To this end, the correlation between Vβ hierarchies in the CD8+ T cell compartment between individuals was calculated, separating the donors into subjects sharing no HLA allele and individuals who shared one or more HLA allele. Only the data from adult donors with high-resolution HLA-typing data were used. The median correlation in the CD8 compartment between pairs of individuals who shared one or more HLA class I alleles had a trend toward being higher than for individuals expressing unique HLA alleles (r = 0.740 vs 0.648, respectively), although this was not significant (p = 0.16) (Table IV; Fig. 3). Thus, a median of ∼42% of the Vβ hierarchy of one donor is explained by the Vβ hierarchy of another donor in the CD8+ T cell compartment, whereas in donors with one or more shared HLA allele this increases to 54.8% (i.e., an increase of 12.8%). Thus, sharing of a single HLA class I allele could contribute at most ∼13% to the observed sharing, and in our sample this was not significantly different.

Table IV.

Analysis of the contribution of HLA to the shape of the adult peripheral blood Vβ repertoire in CD4+ and CD8+ T cells

PopulationHLA ClassNo. PairsaMedian Correlationbp Value
0 Matches≥1 Matches
CD8+ 16 0.680 0.728 0.88 
 11 0.658 0.719 0.43 
 32 0.651 0.754 0.14 
 A, B, C 43 0.648 0.740 0.16 
CD4+ Class II 29 0.812 0.860 0.12 
PopulationHLA ClassNo. PairsaMedian Correlationbp Value
0 Matches≥1 Matches
CD8+ 16 0.680 0.728 0.88 
 11 0.658 0.719 0.43 
 32 0.651 0.754 0.14 
 A, B, C 43 0.648 0.740 0.16 
CD4+ Class II 29 0.812 0.860 0.12 
a

Number of pairs of donors with ≥1 match for the respective HLA class I/II alleles.

b

Wilcoxon correlation coefficient.

FIGURE 3.

Minor contribution of HLA to the shape of the Vβ hierarchy in adult donors. Correlation coefficients were calculated between the Vβ repertoire of each individual that shared one or more HLA-DR (A) or HLA class I (B) allele (▪) and plotted as a frequency distribution of r. Parallel analyses were conducted for individuals that did not share HLA alleles (□).

FIGURE 3.

Minor contribution of HLA to the shape of the Vβ hierarchy in adult donors. Correlation coefficients were calculated between the Vβ repertoire of each individual that shared one or more HLA-DR (A) or HLA class I (B) allele (▪) and plotted as a frequency distribution of r. Parallel analyses were conducted for individuals that did not share HLA alleles (□).

Close modal

There was also a trend toward an HLA allele-specific effect in the CD4+ T cell compartment, because a higher correlation between donors was found when donors were matched for one or more HLA-DR allele(s) (Fig. 3) although this was not statistically significant (r = 0.812 for unmatched individuals vs r = 0.860 for individuals matched at ≥1 HLA-DR loci; p = 0.12). Thus, the sharing of one HLA-DR allele again had at most a relatively minor effect on the Vβ distribution, increasing the extent to which variation in Vβ expression in one individual predicted that in another by only ∼8%. Inclusion of HLA-DQB1 and -DPB1 lead to a smaller difference between matched and unmatched individuals (data not shown), suggesting that these MHC class II molecules have an even smaller effect on shaping the Vβ repertoire.

In the present study, we determined the Vβ repertoire in neonatal and adult blood, and quantified the contribution of TCR-β locus factors and HLA to the shape of the mature Vβ repertoire as determined at the protein level. We found that the Vβ repertoire in both CD4+ and CD8+ T cells was very similar, even in unrelated individuals who would be expected to have different TCR loci and different HLA molecules. The contribution of generic TCR-β locus factors was quantified by comparing the Vβ repertoire in CD4+ T cells and CD8+ T cells between all pairs of individuals. This comparison eliminates both HLA and Vβ allelic effects as variables in the analysis of the Vβ repertoire of CD4+ and CD8+ T cells. A 42–47% concordance of the CD4+ Vβ repertoire with the Vβ repertoire in CD8+ T cells of two unrelated individuals was observed (Table II); i.e., 42–47% of the mature Vβ repertoire in adult and neonatal donors is determined by generic TCR-β locus factors. Furthermore, by comparing the repertoire in CD4+ and CD8+ T cells within individuals, the effects of polymorphisms in the TCR-β locus could be calculated. We found that an additional 6–10% can be attributed to allelic differences at the TCR-β locus as determined by comparing the Vβ repertoire between CD4+ and CD8+ T cells within donors (Table II). Donor age, and, indirectly, exposure to environmental Ags, may skew the Vβ repertoire in adult donors. However, comparison of the correlation in Vβ expression between CD4+ and CD8+ T cells within an individual in neonatal vs adult donor samples failed to identify a significant difference (Fig. 2,C). It is, however, of interest to point out that the oldest individual showed the lowest CD4 vs CD8 correlation (Fig. 2 D), suggesting that, although there was no effect from neonatal to adult samples, there may be effects at the extremes of age, although we have insufficient samples from elderly individuals to confirm this.

Various factors may contribute to the final shape of the TCRVβ repertoire, such as TCR-β locus accessibility (23, 24, 25, 26), accessibility of the recombination signal sequences to V(D)J recombinases (27, 28), differential individual Vβ promoter activities (29, 30), recombination efficiency which depends on variations in the recombination signal sequence (RSS) (31), proximity to the DJβ cluster (16), and pairing efficiency with the pre-Tα chain (16). Wilson et al. (16) showed that DJβ proximity and efficiency of pairing with pre-Tα did not affect murine Vβ usage. Instead, it was found that Vβ usage in mice was already skewed at the earliest stages of protein expression (even before cell surface expression), suggesting that genetic rather than thymic selection pressures were the driving force in determining the Vβ repertoire. We also confirm that proximity to the DJβ cluster/TCR-β enhancer does not determine the Vβ usage by mature T cells in humans (Fig. 1,C). The Vβ genes are already transcribed before rearrangement (30), and the transcriptional activity of different Vβs is different. However, Chen et al. (30) showed that many of the differences were leveled off by rearrangement and/or β-selection, suggesting that other factors change the Vβ usage at different stages during thymocyte development, the most likely candidate being differences in RSS heptamer, nonamer, and 23-bp spacer region (32). Posnett et al. (22) have shown that a SNP in the 23-bp spacer region of Vβ28 strongly correlated with the usage of this Vβ by T cells. We have confirmed this in our data set. The analysis of the correlation in Vβ28 usage by CD4+ and CD8+ across all individuals showed a (Spearman) correlation coefficient of 0.90415 (p < 0.00001; Fig. 1 D). Several other Vβs analyzed in the present study also showed strong correlations, which suggests that other SNPs may similarly affect the usage of these Vβ proteins. However, these correlations were not significant after correction for the number of comparisons. Larger studies, or studies focused specifically on the highly correlated Vβ alleles, may be necessary to determine the extent of allelic effects in other Vβ genes. Others similarly have shown that the RSS has a strong effect on the usage of TCR-β segments in the mature T cell repertoire. The TCR-β locus is rich in SNPs, which are dispersed over coding and noncoding regions (the promoters, introns, and the RSS); 284 polymorphisms have been reported, which averages to ∼1 SNP per 200 bp of the TCR-β locus (33). It has long been known that slight dissimilarities in the RSS can have profound effects on targeting RAG1 and RAG2; i.e., there is a correlation between how well the RSS targets recombinase activity and the frequency of lymphocytes expressing that gene segment (34). This has been demonstrated for murine Igs (reviewed in Ref. 34) and also for the murine TCR-β locus (17). Importantly, the biases established during recombination are largely maintained in mature T cells (35, 36). Together, these data suggest that the generic TCR-β locus factors and allelic effects of Vβ genes determine ∼50% of the Vβ usage in mature T cells.

We subsequently quantified the contribution of HLA to the shape of the Vβ repertoire by comparing Vβ frequencies in adult donors that shared one or more HLA class I or II alleles with donors that shared no HLA alleles for CD8+ and CD4+ T cells, respectively. The median extent to which the hierarchy of one individual’s Vβ expression pattern predicted the distribution in another unrelated individual increased by ∼8–13% when ≥1 HLA allele was shared, but this was not significant in our studies. This indicates that HLA makes at most only a minor contribution to the shape of the mature Vβ repertoire in these circumstances.

Others have suggested that HLA is the dominant molding force for Vβ frequencies in CD4+ and CD8+ T cells (37, 38). In these studies, the expression of a limited set of Vβ proteins was determined in HLA identical siblings, HLA haploidentical siblings, and HLA disparate individuals. However, HLA identical siblings share not only HLA identity but also genomic identity and thus such an analysis is confounded by HLA-independent factors. Alternatively, the analysis of a limited set of Vβ proteins (4, 5, 6, 7, 8, 9) may have confounded these studies as well because others analyzing 13 Vβ proteins in individuals from five multisibling families could not confirm previous findings (39).

In quantifying the contribution of different factors to the hierarchy of Vβ expression in humans, we observed that there is a significant correlation even when TCR-β allelic effects and HLA have been excluded, suggesting that factors generic to the TCR locus play a major role in determining Vβ expression patterns. By contrast, TCR-β allelic effects (with the exception of Vβ28) and sharing of HLA alleles play a significant but relatively minor role.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

D.A.P. is a Medical Research Council (U.K.) Senior Clinical Fellow. M.P.D. is a Sylvia and Charles Viertel Charitable Foundation Senior Medical Research Fellow. This work was supported in part by the James S. McDonnell Foundation 21st Century Research Award/Studying Complex Systems, the Australian Research Council, the National Health and Medical Research Council, and the Intramural Research Programs of the National Heart, Lung, and Blood Institute and the National Institute of Allergy and Infectious Diseases, U.S. National Institutes of Health.

3

Abbreviations used in this paper: SNP, single nucleotide polymorphism; UCB, umbilical cord blood; RSS, recombination signal sequence.

1
Bassing, C. H., W. Swat, F. W. Alt.
2002
. The mechanism and regulation of chromosomal V(D)J recombination.
Cell
109
: (Suppl.):
S45
-S55.
2
Tonegawa, S..
1983
. Somatic generation of antibody diversity.
Nature
302
:
575
-581.
3
Garcia, K. C., E. J. Adams.
2005
. How the T cell receptor sees antigen–a structural view.
Cell
122
:
333
-336.
4
Dik, W. A., K. Pike-Overzet, F. Weerkamp, D. de Ridder, E. F. de Haas, M. R. Baert, P. van der Spek, E. E. Koster, M. J. Reinders, J. J. van Dongen, et al
2005
. New insights on human T cell development by quantitative T cell receptor gene rearrangement studies and gene expression profiling.
J. Exp. Med.
201
:
1715
-1723.
5
Ramiro, A. R., C. Trigueros, C. Marquez, J. L. San Millan, M. L. Toribio.
1996
. Regulation of pre-T cell receptor (pTα-TCRβ) gene expression during human thymic development.
J. Exp. Med.
184
:
519
-530.
6
Carrasco, Y. R., C. Trigueros, A. R. Ramiro, V. G. de Yebenes, M. L. Toribio.
1999
. β-selection is associated with the onset of CD8β chain expression on CD4+CD8αα+ pre-T cells during human intrathymic development.
Blood
94
:
3491
-3498.
7
Fehling, H. J., A. Krotkova, C. Saint-Ruf, H. von Boehmer.
1995
. Crucial role of the pre-T-cell receptor α gene in development of αβ but not γδ T cells.
Nature
375
:
795
-798.
8
Hoffman, E. S., L. Passoni, T. Crompton, T. M. Leu, D. G. Schatz, A. Koff, M. J. Owen, A. C. Hayday.
1996
. Productive T-cell receptor β-chain gene rearrangement: coincident regulation of cell cycle and clonality during development in vivo.
Genes Dev.
10
:
948
-962.
9
von Boehmer, H., I. Aifantis, O. Azogui, J. Feinberg, C. Saint-Ruf, C. Zober, C. Garcia, J. Buer.
1998
. Crucial function of the pre-T-cell receptor (TCR) in TCR β selection, TCR β allelic exclusion and αβ versus γδ lineage commitment.
Immunol. Rev.
165
:
111
-119.
10
Kruisbeek, A. M., M. C. Haks, M. Carleton, A. M. Michie, J. C. Zuniga-Pflucker, D. L. Wiest.
2000
. Branching out to gain control: how the pre-TCR is linked to multiple functions.
Immunol. Today
21
:
637
-644.
11
Borowski, C., C. Martin, F. Gounari, L. Haughn, I. Aifantis, F. Grassi, H. von Boehmer.
2002
. On the brink of becoming a T cell.
Curr. Opin. Immunol.
14
:
200
-206.
12
Irving, B. A., F. W. Alt, N. Killeen.
1998
. Thymocyte development in the absence of pre-T cell receptor extracellular immunoglobulin domains.
Science
280
:
905
-908.
13
Saint-Ruf, C., M. Panigada, O. Azogui, P. Debey, H. von Boehmer, F. Grassi.
2000
. Different initiation of pre-TCR and γδTCR signalling.
Nature
406
:
524
-527.
14
Panigada, M., S. Porcellini, E. Barbier, S. Hoeflinger, P. A. Cazenave, H. Gu, H. Band, H. von Boehmer, F. Grassi.
2002
. Constitutive endocytosis and degradation of the pre-T cell receptor.
J. Exp. Med.
195
:
1585
-1597.
15
Starr, T. K., S. C. Jameson, K. A. Hogquist.
2003
. Positive and negative selection of T cells.
Annu. Rev. Immunol.
21
:
139
-176.
16
Wilson, A., C. Marechal, H. R. MacDonald.
2001
. Biased Vβ usage in immature thymocytes is independent of DJβ proximity and pTα pairing.
J. Immunol.
166
:
51
-57.
17
Livak, F., D. B. Burtrum, L. Rowen, D. G. Schatz, H. T. Petrie.
2000
. Genetic modulation of T cell receptor gene segment usage during somatic recombination.
J. Exp. Med.
192
:
1191
-1196.
18
Adams, S. D., K. C. Barracchini, D. Chen, F. Robbins, L. Wang, P. Larsen, R. Luhm, D. F. Stroncek.
2004
. Ambiguous allele combinations in HLA class I and class II sequence-based typing: when precise nucleotide sequencing leads to imprecise allele identification.
J. Transl. Med.
2
:
30
-35.
19
Melenhorst, J. J., R. Eniafe, D. Follmann, R. Nakamura, M. Kirby, A. J. Barrett.
2002
. Molecular and flow cytometric characterization of the CD4 and CD8 T-cell repertoire in patients with myelodysplastic syndrome.
Br. J. Haematol.
119
:
97
-105.
20
Melenhorst, J. J., L. Sorbara, M. Kirby, N. F. Hensel, A. J. Barrett.
2001
. Large granular lymphocyte leukaemia is characterized by a clonal T-cell receptor rearrangement in both memory and effector CD8+ lymphocyte populations.
Br. J. Haematol.
112
:
189
-194.
21
Arden, B., S. P. Clark, D. Kabelitz, T. W. Mak.
1995
. Human T-cell receptor variable gene segment families.
Immunogenetics
42
:
455
-500.
22
Posnett, D. N., C. S. Vissinga, C. Pambuccian, S. Wei, M. A. Robinson, D. Kostyu, P. Concannon.
1994
. Level of human TCRBV3S1 (Vβ3) expression correlates with allelic polymorphism in the spacer region of the recombination signal sequence.
J. Exp. Med.
179
:
1707
-1711.
23
Bories, J. C., J. Demengeot, L. Davidson, F. W. Alt.
1996
. Gene-targeted deletion and replacement mutations of the T-cell receptor β-chain enhancer: the role of enhancer elements in controlling V(D)J recombination accessibility.
Proc. Natl. Acad. Sci. USA
93
:
7871
-7876.
24
Bouvier, G., F. Watrin, M. Naspetti, C. Verthuy, P. Naquet, P. Ferrier.
1996
. Deletion of the mouse T-cell receptor β gene enhancer blocks αβ T-cell development.
Proc. Natl. Acad. Sci. USA
93
:
7877
-7881.
25
McMurry, M. T., C. Hernandez-Munain, P. Lauzurica, M. S. Krangel.
1997
. Enhancer control of local accessibility to V(D)J recombinase.
Mol. Cell. Biol.
17
:
4553
-4561.
26
Mathieu, N., W. M. Hempel, S. Spicuglia, C. Verthuy, P. Ferrier.
2000
. Chromatin remodeling by the T cell receptor (TCR)-β gene enhancer during early T cell development: implications for the control of TCR-β locus recombination.
J. Exp. Med.
192
:
625
-636.
27
Yancopoulos, G. D., T. K. Blackwell, H. Suh, L. Hood, F. W. Alt.
1986
. Introduced T cell receptor variable region gene segments recombine in pre-B cells: evidence that B and T cells use a common recombinase.
Cell
44
:
251
-259.
28
Golding, A., S. Chandler, E. Ballestar, A. P. Wolffe, M. S. Schlissel.
1999
. Nucleosome structure completely inhibits in vitro cleavage by the V(D)J recombinase.
EMBO J.
18
:
3712
-3723.
29
Whitehurst, C. E., S. Chattopadhyay, J. Chen.
1999
. Control of V(D)J recombinational accessibility of the Dβ1 gene segment at the TCR β locus by a germline promoter.
Immunity
10
:
313
-322.
30
Chen, F., L. Rowen, L. Hood, E. V. Rothenberg.
2001
. Differential transcriptional regulation of individual TCR Vβ segments before gene rearrangement.
J. Immunol.
166
:
1771
-1780.
31
Lee, A. I., S. D. Fugmann, L. G. Cowell, L. M. Ptaszek, G. Kelsoe, D. G. Schatz.
2003
. A functional analysis of the spacer of V(D)J recombination signal sequences.
PLoS Biol.
1
:
56
-69.
32
Glusman, G., L. Rowen, I. Lee, C. Boysen, J. C. Roach, A. F. Smit, K. Wang, B. F. Koop, L. Hood.
2001
. Comparative genomics of the human and mouse T cell receptor loci.
Immunity
15
:
337
-349.
33
Mackelprang, R., C. S. Carlson, L. Subrahmanyan, R. J. Livingston, M. A. Eberle, D. A. Nickerson.
2002
. Sequence variation in the human T-cell receptor loci.
Immunol. Rev.
190
:
26
-39.
34
Livak, F., H. T. Petrie.
2001
. Somatic generation of antigen-receptor diversity: a reprise.
Trends Immunol.
22
:
608
-612.
35
Feeney, A. J..
1991
. Junctional sequences of fetal T cell receptor β chains have few N regions.
J. Exp. Med.
174
:
115
-124.
36
Candeias, S., C. Waltzinger, C. Benoist, D. Mathis.
1991
. The Vβ17+ T cell repertoire: skewed Jβ usage after thymic selection; dissimilar CDR3s in CD4+ versus CD8+ cells.
J. Exp. Med.
174
:
989
-1000.
37
Gulwani-Akolkar, B., D. N. Posnett, C. H. Janson, J. Grunewald, H. Wigzell, P. Akolkar, P. K. Gregersen, J. Silver.
1991
. T cell receptor V-segment frequencies in peripheral blood T cells correlate with human leukocyte antigen type.
J. Exp. Med.
174
:
1139
-1146.
38
Akolkar, P. N., B. Gulwani-Akolkar, R. Pergolizzi, R. D. Bigler, J. Silver.
1993
. Influence of HLA genes on T cell receptor V segment frequencies and expression levels in peripheral blood lymphocytes.
J. Immunol.
150
:
2761
-2773.
39
Genevee, C., F. Farace, V. Chung, A. Diu, C. Raffoux, D. Charron, T. Hercend, F. Triebel.
1994
. Influence of human leukocyte antigen genes on TCR V gene segment frequencies.
Int. Immunol.
6
:
1497
-1504.