The role of CD8+ T cells in human tuberculosis (TB) remains elusive. We analyzed the T cell repertoire and phenotype in 1) children with active TB (≤4 years), 2) healthy latently Mycobacterium tuberculosis-infected children, and 3) noninfected age-matched (tuberculin skin test-negative) controls. Ex vivo phenotyping of T cell subpopulations by flow cytometry revealed a significant increase in the proportion of CD8+CD45ROCD62LCD28CD27 effector T cells (TEF) in the peripheral blood of children with active TB (22.1 vs 9.5% in latently M. tuberculosis-infected children, vs 8.5% in tuberculin skin test-negative controls). Analyses of TCR variable β-chains revealed markedly skewed repertoires in CD8+ TEF and effector memory T cells. Expansions were restricted to single TCR variable β-chains in individual donors indicating clonal growth. CDR3 spectratyping and DNA sequencing verified clonal expansion as the cause for CD8+ effector T cell enrichment in individual TB patients. The most prominent enrichment of highly similar TEF clones (>70% of CD8+ TEF) was found in two children with active severe TB. Therefore, clonal expansion of CD8+ TEF occurs in childhood TB with potential impact on course and severity of disease.

Childhood tuberculosis (TB)3 accounts for ∼10% of all TB cases worldwide (1). Infants and young children are more susceptible to Mycobacterium tuberculosis than adults, resulting in a higher incidence of disease (2), increased severity of pathogenesis (3), and frequent involvement of extrapulmonary organs (4). T cells and T cell-derived cytokines (e.g., IFN-γ, TNF-α) are crucial for protection against TB (5, 6). Participation of CD8+ T cells in control of TB has been demonstrated in mice (7) and humans (8). CD8+ T cells contribute to control of M. tuberculosis infection by mediating specific effector functions including IFN-γ production (9), lysis of infected host cells (9, 10), and direct killing of mycobacteria (11). A limited number of studies focused on the T cell repertoire in M. tuberculosis infection, demonstrating clonal T cell expansion in granulomas from the latently M. tuberculosis infected (LTBI) (12) and changes in the peripheral blood and pleural fluid T cell repertoire from TB patients (13). So far, changes in the T cell repertoire of children with active TB or LTBI have not been described.

The basic assumption underlying the present study implies that differences in the T cell repertoire in children with TB can be detected at the level of the T cell subpopulation. This is based on the rationale that the proportion of M. tuberculosis-specific T cells in infected children is possibly under the detection limit when compared with the vast majority of naive T cells (Tnaive) reflecting rare encounter with exogenous Ags (14). Comparison of the T cell repertoire on the level of relatively rare differentiated memory and effector T cells can circumvent this quandary and might lead to the identification of repertoire changes. Therefore, we compared the peripheral blood T cell repertoire of infants and young children.

Insights into functionally different subsets of T cells have increased during the last few years (15). Tnaive can be distinguished from memory T cells, which can be further subdivided according to their capacity to enter secondary lymphoid organs. Central memory T cells (TCM) express lymph node homing receptors (e.g., CD62L, CCR7), whereas effector memory T cells (TEM) do not (15). CD8+ TEM down-regulate surface expression of T cell coreceptor molecules, CD45RO, CD28, and CD27, during further differentiation toward an effector T cell (TEF) phenotype (16). This maturation step is likely driven by cytokines in the absence of antigenic stimulation (17). Each of these subpopulations comprises distinct functions. TCM, for example, produce mainly IL-2, whereas TEM are characterized by rapid expression of effector functions and production of IFN-γ or IL-4 (18). TEF express vast amounts of TNF-α, perforin, and granulysin accompanied by diminished IL-2 secretion and proliferation (15).

Analysis of the TCR β (B) chain distribution is widely used to characterize alterations in the T cell repertoire, which can range from extensive diversity (19) to expansion of single T cell clones (20, 21). T cell clones are characterized by the clonotypic TCR (B) chains generated by somatic recombination of V (D only in the B chain), J, and C chain. Each TCR B chain comprises three hypervariable CDRs. CDR1 and CDR2 are encoded by germline parts of the V segment, whereas CDR3, located at the seam of the V(D)J segment, contains nondefined nucleotides and germline-derived residues. The CDR3 recognizes the antigenic peptide bound to the MHC-encoded molecule, whereas CDR1 and CDR2 contact residues of the MHC α helices (22). CDR3 consensus sequences provide evidence for identical epitopes as the cause for clonal expansion (23).

In this study, we combined characterization of T cell subpopulations with TCR B chain analyses and describe for the first time clonally expanded CD8+ TEM and TEF in a subgroup of children with active and active severe TB. Of note, clonal expansion of CD8+ TEF frequently occurred in children suffering from severe forms of TB.

Peripheral blood (3 ml) was obtained from eight children (≤4 years) with TB, eight LTBI and nine tuberculin skin test-negative (TST) age-matched individuals, and two children with active severe TB recruited at Department of Pediatric Pneumology and Immunology (Charité, Berlin). Donor characteristics are summarized in Table I. TB diagnosis was based on patient history, chest x-ray, TST, and mycobacterial culture. M. tuberculosis was the causative agent in all TB cases. HIV infection was excluded. TB patients were treated with isoniazid, rifampin, and pyrazinamide (a subgroup with ethambutol) for 8 wk, followed by isoniazid/rifampin treatment for 16 (a subgroup for 44) weeks. Children with TB had a diverse ethnic background: three were Caucasians, two were from Vietnam, two were from Lebanon, and one was from Turkey. LTBI were close contacts of TB patients, had a TST >10 mm, and were not bacillus Calmette-Guérin vaccinated. Controls were TST children, who were not bacillus Calmette-Guérin vaccinated. Six of these control individuals had atopic eczema. Six TB patients were bled before chemotherapy and two to three times during therapy. The two children with active severe TB were a 3-year-old girl suffering from multidrug-resistant TB affecting the lung and the inner ear (TB9) and a 14-year-old girl with pulmonary TB and TB meningitis (TB10). Both children with active severe TB were Caucasians. Two children with active TB (TB5 and TB6) were siblings. The others were unrelated. The local ethics committee approved this study (EA 2/028/04).

Table I.

Characteristics of children with TB, LTBI, and TST

Children with TB
ActiveActive severeLTBITST
Total no. 
Female 
Male  
Age range (years (mean)) 1–4 (2) 3–14 (9) 1–4 (3.3) 1–3 (1.8) 
Disease characteristics     
 Pulmonary TB    
 Miliary TB    
 Active multidrug-resistant TB    
 TB meningitis    
 Atopic eczema    
Children with TB
ActiveActive severeLTBITST
Total no. 
Female 
Male  
Age range (years (mean)) 1–4 (2) 3–14 (9) 1–4 (3.3) 1–3 (1.8) 
Disease characteristics     
 Pulmonary TB    
 Miliary TB    
 Active multidrug-resistant TB    
 TB meningitis    
 Atopic eczema    

T cell subpopulations and expression of TCR BV chains were analyzed by six-color flow cytometry. T cell subpopulation analyses have been performed for all donors recruited. TCR BV repertoire analyses have been performed in seven children with active TB (two of them under chemotherapy), two with active severe TB, six TST, and two LTBI. Three TST have been selected for analyses excluding those with very small TEF proportions to avoid analyses bias due to small cell numbers. Whole blood phenotyping was performed as described previously (24). In brief, 50 μl of peripheral blood was diluted 1/1 in ice-cold PBS and centrifuged (300 × g; 5 min), and the supernatant was discarded. Then, cells were stained for 30 min on ice with allophycocyanin-Cy7 anti-CD4 mAb and PerCP anti-CD8 mAb in combination with PE-Cy7 anti-CD45RO mAb (BD Biosciences), FITC anti-CD62L mAb (German Rheuma Research Center, Berlin, Germany), and allophycocyanin anti-CD45RA (BD Biosciences). For TCR BV analyses, FITC anti-CD62L mAb was replaced by allophycocyanin anti-CD62L mAb (Miltenyi Biotec) and used together with the TCR BV mAb set (BD Biosciences), which allows staining of 24 TCR BV chains. For further characterization of CD8+ T cell subpopulations, allophycocyanin-Cy7 anti-CD3 mAb and PerCP anti-CD8 mAb were combined with PE-Cy7-labeled anti-CD45RO mAb, allophycocyanin-labeled anti-CD62L mAb, FITC-labeled anti-CD28 mAb (eBioscience), and PE-labeled anti-CD27 mAb (German Rheuma Research Center). Four children with active and two children with active severe TB have been analyzed for CD27/CD28 expression on T cell subpopulations. In addition, CD8+ T cell subpopulation markers were combined with FITC-labeled anti-Perforin mAb, or FITC-labeled anti-Granulysin mAb. Red blood lysis buffer (Roche) was used following manufacturer’s instructions. Then, samples were washed twice using ice-cold PBS including 5% FCS and measured with a FACSCanto (BD Biosciences). Analyses were performed using FCS Express 3 software (De Novo). Values are given as medians (±SD only for total numbers). Differences in T cell proportions were assessed using the two-tailed Mann-Whitney U test for nonparametric data and the Wilcoxon signed ranks test (two-tailed) for the comparison of proportions at different time points. The Spearman rank-order correlation test has been determined for analyses of association between T cell subpopulation proportions from individual donors. Nominal p values are given.

Two children with active and one with active severe TB have been analyzed. PBMC were isolated from peripheral blood using Ficoll density gradient centrifugation (Biochrom). Afterward, we cultured PBMC for 0, 1, and 2 h with PMA and ionomycin and without stimulation (37°C, 5% CO2) in 96-well round-bottom plates (Nunc). Cells were then permeabilized and stained as described for immune cell phenotyping.

Two children with active and two with active severe TB have been analyzed. RNA was isolated from 400 μl of peripheral blood with a QIAamp RNA blood mini kit (Qiagen), quantified on a nanodrop analyzer (Agilent Technologies), and reverse transcribed using Superscript III (Invitrogen Life Technologies) following manufacturer’s instructions. Previously described TCR BV-specific primers (25) were used to amplify the expanded TCR BV chain transcript including the CDR3. For TCR BV spectratyping, the reverse primer specific for the TCR constant β-chain was 6-FAM-labeled at its 5′ end. CDR3 products were purified using QIAquick PCR purification kit (Qiagen). CDR3 spectratyping and DNA sequencing was done commercially (Meixner). The TCR nomenclature used in this study is derived from Arden et al. (26).

CD45RO and CD62L as well as TCR BV surface expression were determined in CD4+ and CD8+ T cells to determine the distribution of subpopulations (Tnaive, TCM,TEM, TEF) in PBMC from infants and young children (≤4 years) with active TB, LTBI, and TST. Fig. 1,a graphically depicts flow cytometry analysis. Differences in CD4+ T cell proportions were detected for TCM (CD62L+, CD45RO+) with a lower percentage in TB (18.4%) compared with TST (30.6%) (p = 0.05) (Fig. 1,b, upper graph). A similar tendency was found in comparison to LTBI (28.2%) without reaching significant levels (p = 0.2). Marked differences were found for CD8+ T cells. In children with TB, TEF proportions (CD62L, CD45RO) (22.1%) were significantly higher compared with LTBI (9.4%; p = 0.01), and TST children (8.5%; p < 0.01) (Fig. 1 b, lower graph) accompanied by decreased CD8+ Tnaive proportions in individual donors (Spearman rank-order correlation test, rs = 0.95, p < 0.01). TEM proportions were slightly higher in children with TB (15.4 vs 7.6% in LTBI (p = 0.38), vs 5.0% in TST (p = 0.1)). In addition, we detected lower CD8+ TCM proportions in children with TB compared with LTBI (5.9 vs 11.3% (p = 0.02)) but not compared with TST (6.3%). Since differences in total lymphocyte numbers in M. tuberculosis infection have been described (13, 27), we compared also total CD4+ and CD8+ T cell numbers between study groups. No significant differences for CD4+ or CD8+ T cells were detected, although the numbers were slightly lower in LTBI (CD4+, 18,571 ± 8,641; CD8+, 7,835 ± 4,332) compared with children with TB (CD4+, 25,867 ± 11,782; CD8+, 11,243 ± 4,304) and TST (CD4+, 28,465 ± 26,566; CD8+, 12,644 ± 7,585). Children with TB with marked TEF expansions did not differ from the study group mean (data not shown). Therefore, changes in total cell numbers did not affect our analyses. We focused on enriched CD8+ TEF in children with TB to characterize the phenotype and functions of these cells.

FIGURE 1.

Ex vivo analysis of T cell subpopulations in blood from eight children with TB, eight LTBI, and nine TST. a, Representative example of flow cytometry in active TB. CD8+ and CD4+ T cells were analyzed for subpopulations according to expression of CD45RO and CD62L. In these subpopulations (Tnaive, TCM, TEM, and TEF), the expression of TCR BV chains (here TCR BV8, BV13-1, and BV-13-6) was further determined. Arrows describe the sequence of analytical steps. Numbers indicate the percentages of cells in each quadrant. b, Subpopulation distribution for CD4+ (upper graph) and CD8+ (lower graph) T cells. From left to right, panels (separated by dotted lines) show Tnaive, TCM, TEM, and TEF for children with active TB (▪), LTBI (▨), and TST (□). Bars, mean percentages and SDs. Asterisks indicate significant differences between the indicated study groups, as determined by the Mann-Whitney U test.

FIGURE 1.

Ex vivo analysis of T cell subpopulations in blood from eight children with TB, eight LTBI, and nine TST. a, Representative example of flow cytometry in active TB. CD8+ and CD4+ T cells were analyzed for subpopulations according to expression of CD45RO and CD62L. In these subpopulations (Tnaive, TCM, TEM, and TEF), the expression of TCR BV chains (here TCR BV8, BV13-1, and BV-13-6) was further determined. Arrows describe the sequence of analytical steps. Numbers indicate the percentages of cells in each quadrant. b, Subpopulation distribution for CD4+ (upper graph) and CD8+ (lower graph) T cells. From left to right, panels (separated by dotted lines) show Tnaive, TCM, TEM, and TEF for children with active TB (▪), LTBI (▨), and TST (□). Bars, mean percentages and SDs. Asterisks indicate significant differences between the indicated study groups, as determined by the Mann-Whitney U test.

Close modal

A characteristic feature of the differentiation stage of CD8+ TEF is the expression pattern of the T cell costimulatory molecules, CD27 and CD28 (16). Therefore, we analyzed expanded CD8+ TEF from individual TB diseased children for the expression of these markers. Fig. 2,a shows dot plots from a representative experiment, and Fig. 2,b summarizes analyses. Although Tnaive and TCM were mostly CD28+ and CD27+, the majority of TEM were CD28 and some TEM down-regulated CD27, too. The vast majority of TEF were CD28 and CD27 (Fig. 2, a, lower left graph, and b), a characteristic feature of well-differentiated CD8+ T cells. In addition, TEF expressed CD45RA (data not shown), which defines the so-called TEMRA phenotype of CD8+ T cells (16). A major function of TEF is the lysis of infected host cells to eradicate invading pathogens (9, 10). One crucial mechanism in this process involves granulysin and perforin (9, 10). We determined perforin (Fig. 2,c, left graphs) and granulysin (Fig. 2,c, right graphs) expression in T cell subpopulations from selected TB patients with expanded TEF. The majority of Tnaive expressed low amounts of granulysin and perforin (Fig. 2,c, upper graphs). Higher proportions of granulysin and perforin expressing T cells were detected in TCM and TEM subpopulations, but the expression varied markedly (Fig. 2,c, lower graphs). In contrast, the vast majority of TEF expressed a huge amount of granulysin and perforin (Fig. 2,c, upper graphs). We analyzed the capacity of TEF to secrete granulysin and perforin upon in vitro stimulation in two children with active disease and one child with active severe TB. PMA/ionomycin induced a gradual decrease of perforin- and granulysin-positive cells within TEF compared with the nonstimulated control (Fig. 2,d). In addition, we detected a reduction in the mean expression of perforin and granulysin in TEF by comparison of the mean fluorescence intensities (Fig. 2 d). We conclude that expanded TEF in children with active and active severe TB possess the phenotype and the capacity to secrete effector molecules necessary to exert lytic antimycobacterial functions.

FIGURE 2.

Ex vivo analysis of T cell subpopulations for CD27, and CD28 expression, as well as perforin and granulysin expression and release. a, Representative example of flow cytometry analysis for expression of CD27 and CD28 on CD8+ Tnaive (upper left graph), TCM (upper right graph), TEM (lower right graph), and TEF (lower left graph). A representative experiment is shown. b, CD27 and CD28 expression on CD8+ T cell subpopulations (Tnaive, TCM, TEM, and TEF) determined in four children with active TB and two children with active severe TB. Bar charts represent mean percentages and SDs. c, Expression of perforin (left graphs) and granulysin (right graphs) as determined by flow cytometry for CD8+ Tnaive (gray curve) and TEF (black hatched curve) (upper graphs), as well as TCM (gray curve) and TEM (black curve) (lower graphs). An isotype control is shown (bright gray curve). Fluorescence intensity is shown on the x-axis and the number of events on the y-axis. Granulysin- and perforin-positive cells are indicated by a marker. A representative experiment of four is shown. d, Perforin (left graph; ○, •) and granulysin (right graph; ▵, ▴) expression after stimulation with PMA/ionomycin (•, ▴) or without stimulus (○, ▵) for 0, 1, and 2 h (x-axis). Percentages of positive TEF are shown on the y-axes. Numbers in brackets indicate mean fluorescence intensity. A representative experiment of three is shown.

FIGURE 2.

Ex vivo analysis of T cell subpopulations for CD27, and CD28 expression, as well as perforin and granulysin expression and release. a, Representative example of flow cytometry analysis for expression of CD27 and CD28 on CD8+ Tnaive (upper left graph), TCM (upper right graph), TEM (lower right graph), and TEF (lower left graph). A representative experiment is shown. b, CD27 and CD28 expression on CD8+ T cell subpopulations (Tnaive, TCM, TEM, and TEF) determined in four children with active TB and two children with active severe TB. Bar charts represent mean percentages and SDs. c, Expression of perforin (left graphs) and granulysin (right graphs) as determined by flow cytometry for CD8+ Tnaive (gray curve) and TEF (black hatched curve) (upper graphs), as well as TCM (gray curve) and TEM (black curve) (lower graphs). An isotype control is shown (bright gray curve). Fluorescence intensity is shown on the x-axis and the number of events on the y-axis. Granulysin- and perforin-positive cells are indicated by a marker. A representative experiment of four is shown. d, Perforin (left graph; ○, •) and granulysin (right graph; ▵, ▴) expression after stimulation with PMA/ionomycin (•, ▴) or without stimulus (○, ▵) for 0, 1, and 2 h (x-axis). Percentages of positive TEF are shown on the y-axes. Numbers in brackets indicate mean fluorescence intensity. A representative experiment of three is shown.

Close modal

We serially analyzed the CD8+ TEF proportions in six children with TB to determine whether expansions were modulated during chemotherapy. In five of six children with TB, proportions of TEF were increased before chemotherapy compared with the mean TEF percentage of TST and LTBI (indicated by the dotted line; Fig. 3 a). Interestingly, these donors showed a strong initial decrease in the proportion of TEF. In three of six TB children, the CD8+ TEF percentage fell short compared with the mean proportion of controls and remained stable at this level. In the three TB patients with the highest TEF percentage at disease onset, the initial decrease was followed by an increase in TEF proportions. TB patients TB2 and TB3 showed rapid and strong increase, whereas patient TB4 showed a mild increase until day 200. However, in this patient, only one measurement during the early, apparently sensitive phase could be performed. To determine the statistical significance of these findings, we compared the first time point (before chemotherapy) with the second time point (19–72 days) and the last time point of bleeding (146–240 days) and detected a significant decrease during initial therapy (p = 0.03) and tendency to decrease in comparison between the first and the last time point (p = 0.06).

FIGURE 3.

Longitudinal analysis of CD8+ T subpopulation distribution in individual TB patients. a, Percentage of TEF is shown on the y-axis and length of chemotherapy on the x-axis. Data are shown for six TB patients (TB1–TB6). Each patient is represented by a symbol, and symbol positions indicate the respective day of blood sampling. b, Analyses of T cell subpopulations from six children with active TB. Percentages of CD8+ T cell subpopulations (y-axis) under chemotherapy course (x-axis) are shown. Colors indicate Tnaive (bright gray), TCM (medium gray), TEM (dark gray), and TEF (black).

FIGURE 3.

Longitudinal analysis of CD8+ T subpopulation distribution in individual TB patients. a, Percentage of TEF is shown on the y-axis and length of chemotherapy on the x-axis. Data are shown for six TB patients (TB1–TB6). Each patient is represented by a symbol, and symbol positions indicate the respective day of blood sampling. b, Analyses of T cell subpopulations from six children with active TB. Percentages of CD8+ T cell subpopulations (y-axis) under chemotherapy course (x-axis) are shown. Colors indicate Tnaive (bright gray), TCM (medium gray), TEM (dark gray), and TEF (black).

Close modal

How changes in the proportions of TEF affect the other CD8+ T cell subsets was addressed by comparing CD8+ T cell subpopulation proportions in six children with active disease under chemotherapy (Fig. 3 b). In five of six children (TB1 to -5), the initial decrease of TEF was accompanied by increased Tnaive proportions. Although TCM proportions were relatively stable in most children with active TB, TEM showed similar fluctuations compared with TEF in three (TB1, TB2, TB4) of six patients. In conclusion, the decrease in CD8+ TEF during initial chemotherapy was accompanied by an increase of Tnaive proportions. Thereafter, unique trends were detected in individual children with active TB.

Expansions in CD8+ T cells can be caused by oligoclonal or clonal proliferation (28). To address this question, we characterized the TCR BV composition of CD8+ T cell subpopulations. Because the TCR BV composition in CD8+ Tnaive reflects the endogenous repertoire, we subtracted the individual Tnaive percentage of each TCR BV chain from the respective TCM, TEM, and TEF proportions. Fig. 4,a graphically depicts this transformation introduced to improve the comprehensibility of the analysis. Four children with active TB before chemotherapy with TEF expansions (24.1–52.6% TEF within CD8+ T cells) (Fig. 4,b) and three children with active TB without TEF expansions (TB5 and TB6 at day 240 of chemotherapy, TB8 before chemotherapy) (Fig. 4,c) were analyzed. In two children with TEF expansions (TB1, TB2), we detected markedly increased proportions of single TCR BV chains (BV1, BV22) leading to skewed TCR repertoires in TEM and TEF cells (Fig. 4,b). TEM comprised an even higher proportion of the respective TCR BV chain-expressing cells compared with TEF cells. In TB3, we detected enrichment of three TCR BV, whereas in TB4 only moderate changes in the TCR BV repertoire were detected. Notably, children with active TB without TEF expansions showed no skewing of the TCR BV repertoire (Fig. 4 c).

FIGURE 4.

Skewed TCR BV expression by CD8+ T cells. a, Representative example for data transformation of the proportions for CD8+ Tnaive (▥), TCM (□), TEM (▤), and TEF (▪) in T cell subpopulations (left graph, x-axis) for 24 TCR BV chains (left graph, y-axes) to the expression difference for TCM, TEM, and TEF vs Tnaive (right graph). b, Analyses of four children with active TB before therapy (TB1, TB2, TB3, TB4). c, Analyses of three children with active TB who had normal TEF proportions (TB8, TB6) or after the initial decrease of TEF proportion at the end of chemotherapy (TB5) are shown. d, Analyses of two children with active severe TB (TB9, TB10). e, Analyses of three TST and two LTBI are shown. Percentages of TEF are indicated in brackets for each donor.

FIGURE 4.

Skewed TCR BV expression by CD8+ T cells. a, Representative example for data transformation of the proportions for CD8+ Tnaive (▥), TCM (□), TEM (▤), and TEF (▪) in T cell subpopulations (left graph, x-axis) for 24 TCR BV chains (left graph, y-axes) to the expression difference for TCM, TEM, and TEF vs Tnaive (right graph). b, Analyses of four children with active TB before therapy (TB1, TB2, TB3, TB4). c, Analyses of three children with active TB who had normal TEF proportions (TB8, TB6) or after the initial decrease of TEF proportion at the end of chemotherapy (TB5) are shown. d, Analyses of two children with active severe TB (TB9, TB10). e, Analyses of three TST and two LTBI are shown. Percentages of TEF are indicated in brackets for each donor.

Close modal

We included two additional children with active severe TB in the CD8+ TCR BV repertoire analysis to determine a possible role of expanded TEF in severity of disease. Both patients had marked TEF expansions (62.9 and 41.0%) and a highly skewed TCR BV repertoire (Fig. 4 d). Notably, in these cases, the same TCR chain (BV16) dominated the CD8+ TCR BV repertoire and expanded T cells exclusively expressed the TEF phenotype suggesting potential association with disease severity.

The TCR BV repertoire of CD8+ T cells from three TST and two LTBI children (Fig. 4 e) showed minor differences in comparison with Tnaive. Only two donors (TB5, LTBI2) revealed slight enrichment of a single TCR BV chain (TB5: BV7-2, 7% for TEF; LTBI2: BV5-2, 5% for TEM and 7% for TEF). We conclude that TEF expansion found in a subgroup of children with active TB is partially caused by enrichment of single or few TCR BV chain-expressing T cells. In two children with severe active TB, expanded TEF predominantly expressed a single TCR BV chain. Although skewed TCR BV chains in children with active TB were expressed by T cells of a TEM or TEF phenotype, skewed TCR BV-expressing T cells from children with severe TB were exclusively TEF.

To precisely define the type of skewed TCR BV we analyzed the CDR3 by PCR spectratyping. Fig. 5, upper graphs, depicts the normal distribution of a polyclonal T cell population (here TCR BV2) comprising various different CDR3 length fragments for two active TB patients (Fig. 5,a) and two TB patients with severe disease (Fig. 5,b). However, the expanded TCR BV (Fig. 5, lower graphs) revealed more or less one prominent peak. Because this is the typical feature of a monoclonal T cell response, we performed ex vivo DNA sequencing of the expanded TCR BV in the blood. The CDR3 from the expanded TCR BV from the four TB patients shown could be sequenced (Table II), whereas nonexpanded TCR BV2 CDR3 DNA sequencing terminated after the BV2 chain (data not shown). Notably, both children with active TB as well as those with active severe TB had highly similar CDR3 amino acid sequences. We conclude that clonal expansion was the cause for TCR BV skewing of TEM and TEF in two children with active TB and exclusively TEF in two children with active severe TB.

FIGURE 5.

TCR BV CDR3 spectratyping. Amplified DNA fragment length analyses of expanded TCR BV chains (lower graphs) and nonexpanded TCR BV chain 2 (upper graphs) from two children with active TB (TB1, TB2) (a) and two TB patient with active severe disease (TB9, TB10) (b). Curve peaks indicate fragments of a certain length (x-axis).

FIGURE 5.

TCR BV CDR3 spectratyping. Amplified DNA fragment length analyses of expanded TCR BV chains (lower graphs) and nonexpanded TCR BV chain 2 (upper graphs) from two children with active TB (TB1, TB2) (a) and two TB patient with active severe disease (TB9, TB10) (b). Curve peaks indicate fragments of a certain length (x-axis).

Close modal
Table II.

DNA and translated protein sequences of the CDR3 fragments of expanded TCRa

PatientBVBV regionBD NDNBJ RegionBJ
TB1 22 c a s s a g p g v l t q f f g a g 2–6 
  tgtgccagcagt gccgggccgggg gtcctgacccagttcttcggggccgga  
TB2 c a s s g p n s p l h f g n g 
  tgtgccagcagc ggacct aattcacccctccactttgggaatggg  
TB9 16 c a s s p d r r e q f f g p g 2–1 
  tgtgccagcagc ccagacaggcgg gagcagttcttcgggccaggg  
TB10 16 c a s s r d r n y e q y f g p g 2–7 
  tgtgccagcagc cgggacaggaac taccgagcagtacttcgggccggg  
PatientBVBV regionBD NDNBJ RegionBJ
TB1 22 c a s s a g p g v l t q f f g a g 2–6 
  tgtgccagcagt gccgggccgggg gtcctgacccagttcttcggggccgga  
TB2 c a s s g p n s p l h f g n g 
  tgtgccagcagc ggacct aattcacccctccactttgggaatggg  
TB9 16 c a s s p d r r e q f f g p g 2–1 
  tgtgccagcagc ccagacaggcgg gagcagttcttcgggccaggg  
TB10 16 c a s s r d r n y e q y f g p g 2–7 
  tgtgccagcagc cgggacaggaac taccgagcagtacttcgggccggg  
a

BV, Variable β-chain identifier; BV region, variable β-chain cDNA and amino acid sequences; BD NDN, diversity β-chain (nondefined nucleotides) cDNA and amino acid sequences; BJ region, TCR joining β-chain cDNA and amino acid sequences; BJ, TCR joining β-chain identified.

CD8+ T cells have gained increasing attention during recent years as they have been shown to play a critical role in chronic viral infections (29). In the present study, we identified clonal and oligoclonal expansions of Ag-specific CD8+ T cells in children with active TB and active severe TB. Clonal expansions in children with active TB involved TEM as well as TEF, whereas clones in severe active TB exclusively had the TEF phenotype. The latter showed markedly skewed TCR repertoires with >60% of TEF expressing TCR BV16 including consensus CDR3 sequences. Variant expression of CD45RO on expanded CD8+ T cell clones from children with active TB suggested possible phenotypic changes leading to down-regulation of the CD45RO molecule. Because low levels of CD45RO were expressed on CD8+ T cell clones from children with active severe TB, it is tempting to speculate that the persistence of TEF in some children with active severe TB is involved in disease pathogenesis.

We did not find significant differences of TEF proportions in adolescents with TB, LTBI, or TST (data not shown). This could explain contradictory results in children with TB and LTBI (age, >6 years) described very recently (30). This study described decreased proportions of CD8+CD45RA+CCR7 T cells (TEF) in older children with active TB, which increased during the course of disease. We could exclude that analysis of different markers caused this discrepancy by comparing proportions of CD8+CD45RA+CD62L with CD8+CD45ROCD62L T cells in our study groups (data not shown). Therefore, we assume that TEF expansion occurs when the inexperienced immune system of a young child or infant (age <4 years) is confronted with M. tuberculosis.

The consequences of clonal CD8+ T cell expansions have to be addressed by further studies. Nevertheless, it is tempting to speculate that clonotypic TEF disturb the subtle balance between immune response and pathogen. In accordance, although the TEF clones expanded specifically in response to M. tuberculosis Ags, these cells failed to control infection. This is in agreement with clonal evasion postulated to occur in viral infections (31). In these viral infections, changes in the Ag-specific CD8+ T cell repertoire between active and chronic stages are likely due to distinct Ags expressed by the virus during these stages (32). These variations in Ag expression represent an escape mechanism of the virus leading to expansion of CD8+ T cells unable to fulfill required antiviral effector functions (31, 32). A similar scenario could hold true for TB where the metabolism of M. tuberculosis differs markedly between active and dormant stages (33). Thus, secreted Ags dominate during metabolically active stages, whereas Ags controlled by the DosR regulon prevail during dormant stages (38). Accordingly, the Ag profile at one stage could induce clonal proliferation against an epitope, which is not expressed at the other stage. Clonotypic expansion of T cells with irrelevant specificity and inappropriate function could thus cause competition with CD8+ T cells with relevant specificity and appropriate effector functions. We detected only weak IFN-γ and TNF-α secretion by TEF in individual children with active TB, which could not be attributed to T cell clones unequivocally (data not shown). The question whether this unresponsiveness was due to functional impairment as described for CD8+ T cells in clonal evasion (31) will be addressed in ongoing studies.

Differences in mycobacterial burden provide another explanation for fluctuating proportions of T cell subpopulations and expansion of distinct T cell clones. Because most children with active TB are paucibacillary and culture for mycobacteria from gastric fluid is not pursued during chemotherapy, we cannot verify whether changes in TEF proportions reflected mycobacterial burden, although the expansion before chemotherapy and the initial decrease during the first weeks support this assumption.

Expanded TEF had a CD28 and CD27 double-negative phenotype, a typical feature of end-differentiated TEF. The regulation of the expression of CD27 and CD28 seems to be highly conserved in host immunity against chronic viral infections (34). Chronic EBV and hepatitis C virus infection were characterized by CD28+ and CD27+ effector T cells, whereas HIV-specific T cells were CD28CD27+, and the vast majority of CMV-specific T cells were CD28CD27 as well. To determine the possible confounding influence of TEF expansion due to CMV infection in the present study we assessed the stimulatory capacity of three CMV specific proteins, IE-1, IE-2, and pp65 (provided by Dr. F. Kern, University Hospital Charité, Berlin, Germany) in individual donors. CMV proteins induced cytokine production in all tested CMV-positive donors (including TB9) but expanded TEF did not respond (data not shown). This renders CMV infection as the cause for CD8+ TEF expansion in children with active and active severe TB unlikely.

Increased proportions of TEF in active adult TB patients and fluctuations during disease course have been described earlier (35). In this study, one of the two TB patients analyzed during disease course showed a strong decrease (∼40–10%) until week 3 followed by an increase (∼30%) thereafter. The other patient had a lower proportion of TEF cells before chemotherapy (∼15%), decreasing to <10% until week 24. Marked variances in the proportions of TEF cells from active TB patients observed by others (35) and also in children with TB in the present study likely reflect interindividual variances, e.g., due to differences in the HLA background, which influence the probability to develop TEF expansions in TB. Ongoing studies aim at characterizing the function of TB-induced TEM and TEF in detail and address the question of whether the occurrence of TEF and changes in the TCR BV repertoire are associated with TB severity and susceptibility to recurrent disease in high-incidence countries. The data presented in this study demonstrating clonal expansion of CD8+ TEF in children suffering from severe active TB render this scenario a likely one worthy of further investigation.

We thank M. L. Grossman for carefully revising the manuscript as well as Drs. J. Geginat and J. Schreiber for helpful comments on the manuscript.

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

This work was supported, in part, by financial support from the Bill and Melinda Gates Foundation, Grand Challenge 6 (to M.J. and S.H.E.K.), and from the Fonds Chemie (to S.H.E.K.).

3

Abbreviations used in this paper: TB, tuberculosis; LTBI, latently Mycobacterium tuberculosis infected; Tnaive, naive T cell; TCM, central memory T cell; TEM, effector memory T cell; TEF, effector T cell; TST, tuberculin skin test.

1
Dolin, P. J., M. C. Raviglione, A. Kochi.
1994
. Global tuberculosis incidence and mortality during 1990–2000.
Bull. World Health Organ.
72
:
213
-220.
2
Miller, F., R. Seale, M. Taylor.
1963
.
Tuberculosis in Children
Little Brown, Boston.
3
Walls, T., D. Shingadia.
2004
. Global epidemiology of paediatric tuberculosis.
J. Infect.
48
:
13
-22.
4
Carrol, E. D., J. E. Clark, A. J. Cant.
2001
. Non-pulmonary tuberculosis.
Paediatr. Respir. Rev.
2
:
113
-119.
5
Keane, J., S. Gershon, R. P. Wise, E. Mirabile-Levens, J. Kasznica, W. D. Schwieterman, J. N. Siegel, M. M. Braun.
2001
. Tuberculosis associated with infliximab, a tumor necrosis factor α-neutralizing agent.
N. Engl. J. Med.
345
:
1098
-1104.
6
Cooper, A. M., D. K. Dalton, T. A. Stewart, J. P. Griffin, D. G. Russell, I. M. Orme.
1993
. Active tuberculosis in interferon-γ gene-disrupted mice.
J. Exp. Med.
178
:
2243
-2247.
7
Flynn, J. L., M. M. Goldstein, K. J. Triebold, B. Koller, B. R. Bloom.
1992
. Major histocompatibility complex class I-restricted T cells are required for resistance to Mycobacterium tuberculosis infection.
Proc. Natl. Acad. Sci. USA
89
:
12013
-12017.
8
Lewinsohn, D. M., M. R. Alderson, A. L. Briden, S. R. Riddell, S. G. Reed, K. H. Grabstein.
1998
. Characterization of human CD8+ T cells reactive with Mycobacterium tuberculosis-infected antigen-presenting cells.
J. Exp. Med.
187
:
1633
-1640.
9
Cho, S., V. Mehra, S. Thoma-Uszynski, S. Stenger, N. Serbina, R. J. Mazzaccaro, J. L. Flynn, P. F. Barnes, S. Southwood, E. Celis, et al
2000
. Antimicrobial activity of MHC class I-restricted CD8+ T cells in human tuberculosis.
Proc. Natl. Acad. Sci. USA
97
:
12210
-12215.
10
Lalvani, A., R. Brookes, R. J. Wilkinson, A. S. Malin, A. A. Pathan, P. Andersen, H. Dockrell, G. Pasvol, A. V. Hill.
1998
. Human cytolytic and interferon-γ-secreting CD8+ T lymphocytes specific for Mycobacterium tuberculosis.
Proc. Natl. Acad. Sci. USA
95
:
270
-275.
11
Stenger, S., R. J. Mazzaccaro, K. Uyemura, S. Cho, P. F. Barnes, J. P. Rosat, A. Sette, M. B. Brenner, S. A. Porcelli, B. R. Bloom, R. L. Modlin.
1997
. Differential effects of cytolytic T cell subsets on intracellular infection.
Science
276
:
1684
-1687.
12
Tully, G., C. Kortsik, H. Hohn, I. Zehbe, W. E. Hitzler, C. Neukirch, K. Freitag, K. Kayser, M. J. Maeurer.
2005
. Highly focused T cell responses in latent human pulmonary Mycobacterium tuberculosis infection.
J. Immunol.
174
:
2174
-2184.
13
Gambon-Deza, F., M. Pacheco Carracedo, T. Cerda Mota, J. Montes Santiago.
1995
. Lymphocyte populations during tuberculosis infection: Vβ repertoires.
Infect. Immun.
63
:
1235
-1240.
14
Saule, P., J. Trauet, V. Dutriez, V. Lekeux, J. P. Dessaint, M. Labalette.
2006
. Accumulation of memory T cells from childhood to old age: central and effector memory cells in CD4+ versus effector memory and terminally differentiated memory cells in CD8+ compartment.
Mech. Ageing Dev.
127
:
274
-281.
15
Sallusto, F., J. Geginat, A. Lanzavecchia.
2004
. Central memory and effector memory T cell subsets: function, generation, and maintenance.
Annu. Rev. Immunol.
22
:
745
-763.
16
Hamann, D., P. A. Baars, M. H. Rep, B. Hooibrink, S. R. Kerkhof-Garde, M. R. Klein, R. A. van Lier.
1997
. Phenotypic and functional separation of memory and effector human CD8+ T cells.
J. Exp. Med.
186
:
1407
-1418.
17
Geginat, J., A. Lanzavecchia, F. Sallusto.
2003
. Proliferation and differentiation potential of human CD8+ memory T-cell subsets in response to antigen or homeostatic cytokines.
Blood
101
:
4260
-4266.
18
Sallusto, F., D. Lenig, R. Forster, M. Lipp, A. Lanzavecchia.
1999
. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions.
Nature
401
:
708
-712.
19
Silins, S. L., S. M. Cross, S. L. Elliott, S. J. Pye, J. M. Burrows, D. J. Moss, I. S. Misko.
1997
. Selection of a diverse TCR repertoire in response to an Epstein-Barr virus-encoded transactivator protein BZLF1 by CD8+ cytotoxic T lymphocytes during primary and persistent infection.
Int. Immunol.
9
:
1745
-1755.
20
Moss, P. A., R. J. Moots, W. M. Rosenberg, S. J. Rowland-Jones, H. C. Bodmer, A. J. McMichael, J. I. Bell.
1991
. Extensive conservation of α and β chains of the human T-cell antigen receptor recognizing HLA-A2 and influenza A matrix peptide.
Proc. Natl. Acad. Sci. USA
88
:
8987
-8990.
21
Argaet, V. P., C. W. Schmidt, S. R. Burrows, S. L. Silins, M. G. Kurilla, D. L. Doolan, A. Suhrbier, D. J. Moss, E. Kieff, T. B. Sculley, I. S. Misko.
1994
. Dominant selection of an invariant T cell antigen receptor in response to persistent infection by Epstein-Barr virus.
J. Exp. Med.
180
:
2335
-2340.
22
Davis, M. M., P. J. Bjorkman.
1988
. T-cell antigen receptor genes and T-cell recognition.
Nature
334
:
395
-402.
23
Fink, P. J., L. A. Matis, D. L. McElligott, M. Bookman, S. M. Hedrick.
1986
. Correlations between T-cell specificity and the structure of the antigen receptor.
Nature
321
:
219
-226.
24
Jacobsen, M., S. Cepok, E. Quak, M. Happel, R. Gaber, A. Ziegler, S. Schock, W. H. Oertel, N. Sommer, B. Hemmer.
2002
. Oligoclonal expansion of memory CD8+ T cells in cerebrospinal fluid from multiple sclerosis patients.
Brain
125
:
538
-550.
25
Walters, G., S. I. Alexander.
2004
. T cell receptor BV repertoires using real time PCR: a comparison of SYBR green and a dual-labelled HuTrec fluorescent probe.
J. Immunol. Methods
294
:
43
-52.
26
Arden, B., S. P. Clark, D. Kabelitz, T. W. Mak.
1995
. Human T-cell receptor variable gene segment families.
Immunogenetics
42
:
455
-500.
27
Onwubalili, J. K., A. J. Edwards, L. Palmer.
1987
. T4 lymphopenia in human tuberculosis.
Tubercle
68
:
195
-200.
28
Kjer-Nielsen, L., C. S. Clements, A. W. Purcell, A. G. Brooks, J. C. Whisstock, S. R. Burrows, J. McCluskey, J. Rossjohn.
2003
. A structural basis for the selection of dominant αβ T cell receptors in antiviral immunity.
Immunity
18
:
53
-64.
29
van Leeuwen, E. M., G. J. de Bree, I. J. ten Berge, R. A. van Lier.
2006
. Human virus-specific CD8+ T cells: diversity specialists.
Immunol. Rev.
211
:
225
-235.
30
Caccamo, N., S. Meraviglia, C. La Mendola, G. Guggino, F. Dieli, A. Salerno.
2006
. Phenotypical and functional analysis of memory and effector human CD8 T cells specific for mycobacterial antigens.
J. Immunol.
177
:
1780
-1785.
31
Zajac, A. J., J. N. Blattman, K. Murali-Krishna, D. J. Sourdive, M. Suresh, J. D. Altman, R. Ahmed.
1998
. Viral immune evasion due to persistence of activated T cells without effector function.
J. Exp. Med.
188
:
2205
-2213.
32
Price, D. A., S. M. West, M. R. Betts, L. E. Ruff, J. M. Brenchley, D. R. Ambrozak, Y. Edghill-Smith, M. J. Kuroda, D. Bogdan, K. Kunstman, et al
2004
. T cell receptor recognition motifs govern immune escape patterns in acute SIV infection.
Immunity
21
:
793
-803.
33
Boshoff, H. I., C. E. Barry, III.
2005
. Tuberculosis—metabolism and respiration in the absence of growth.
Nat. Rev. Microbiol.
3
:
70
-80.
34
Appay, V., P. R. Dunbar, M. Callan, P. Klenerman, G. M. Gillespie, L. Papagno, G. S. Ogg, A. King, F. Lechner, C. A. Spina, et al
2002
. Memory CD8+ T cells vary in differentiation phenotype in different persistent virus infections.
Nat. Med.
8
:
379
-385.
35
Hohn, H., M. Julch, H. Pilch, C. Kortsik, G. Tully, C. Neukirch, K. Freitag, M. Maeurer.
2003
. Definition of the HLA-A2 restricted peptides recognized by human CD8+ effector T cells by flow-assisted sorting of the CD8+CD45RA+CD28 T cell subpopulation.
Clin. Exp. Immunol.
131
:
102
-110.