OUP user menu

Mycobacterium tuberculosis reactive T cell clones from naturally converted PPD-positive healthy subjects: recognition of the M. tuberculosis 16-kDa antigen

Fredrik Oftung, Eva Borka, Abu Salim Mustafa
DOI: http://dx.doi.org/10.1111/j.1574-695X.1998.tb01142.x 319-325 First published online: 1 April 1998


Mycobacterium tuberculosis reactive T cell clones were established from naturally converted PPD-positive healthy subjects and screened for proliferative reactivity against defined M. tuberculosis protein antigens of 16, 19, 65 (HSP65), and 71 (HSP70) kDa recombinantly expressed in Escherichia coli. Among the recombinant antigens tested, the M. tuberculosis 16-kDa protein antigen, as expressed from the λgt11 phage Y3155, was found to induce T cell proliferation. Crossreactivity studies showed that the epitope recognized was present in M. tuberculosis, M. africanum as well as the vaccine strain M. bovis BCG. The M. tuberculosis 16-kDa reactive T cell clone identified showed the CD4+, CD8 phenotype, secreted interferon-γ upon antigen stimulation, and displayed major histocompatibility complex class II restricted cytotoxicity against M. tuberculosis pulsed macrophages. The results obtained suggest that the recombinant M. tuberculosis 16-kDa antigen can be recognized by human Th1 cells with potential relevance to protection.

  • Mycobacterium tuberculosis
  • 16-kDa antigen
  • T cell response

1 Introduction

Based on incidence (15–25 million) and annual mortality (3–4 million), tuberculosis is now ranked as one of the most important infectious diseases worldwide [1]. The currently used BCG vaccine, which has been shown to be effective against tuberculosis in some industrialized countries, does not show any significant protection in most endemic regions [2,3]. This situation, combined with the accelerating effect of the HIV epidemics and the emergence of multidrug-resistant Mycobacteriium tuberculosis strains [4], calls for new and effective control strategies with respect to both prophylaxis and diagnosis. Since protective immunity against M. tuberculosis is mediated by the cellular immune system [5], a subunit vaccine approach has to rely on identification and application of protein antigens recognized by protective T cells within an HLA heterogeneous population. In this context the growing number of defined and available mycobacterial antigens [6] now provides a rational basis for subunit vaccine design and development of specific diagnostic reagents.

In order to survey the human T cell repertoire against M. tuberculosis antigens with relevance to protective immunity, we have established M. tuberculosis reactive T cell clones from naturally converted PPD-positive healthy subjects. To identify antigenic targets of the T cell clones, they were screened for proliferative reactivity against a panel of defined recombinant M. tuberculosis protein antigens expressed in Escherichia coli. In addition, CD4+ T cell clones with any defined antigen specificity were subjected to crossreactivity studies and effector function analyses.

2 Materials and methods

2.1 Donors for T cell cloning

Donors in this study were naturally converted PPD-positive healthy subjects without any known clinical manifestations of tuberculosis. The donors were not BCG vaccinated.

2.2 Establishment of M. tuberculosis reactive T cell clones

Human T cell clones against M. tuberculosis were established as previously described [7]. Briefly, peripheral blood mononuclear cells (PBMC), isolated from venous blood by density gradient centrifugation (Lymphoprep; NYCOMED, Oslo, Norway), were primarily stimulated with an optimal concentration of irradiated whole cell M. tuberculosis (70 µg ml−1) in complete medium (RPMI-1640+glutamine)+15% pooled human AB serum+1% penicillin-streptomycin in 50-ml flasks for 6 days. The cultures were restimulated with the same antigen in the presence of autologous PMBC, and responding T cells cloned by limiting dilution, expanded by antigen-specific stimulation and cryopreserved as earlier described [7].

2.3 Surface phenotype analyses

OKT4 and OKT8 monoclonal antibodies (Ortho Pharmaceutical Corp., Raritan, NJ) were used to determine the T cell subsets of clones. In the second step, we used fluorescein-conjugated sheep anti-mouse Ig, and positive cells were scored by fluorescence microscopy [8].

2.4 Whole cell mycobacterial antigens

M. leprae was kindly supplied by Dr. R.J.W. Rees, Mill Hill, London, from the IMMLEP (WHO) bank. All other mycobacterial species were kindly donated by Dr. Otto Closs, National Institute of Health, Oslo, Norway. The bacilli were killed by irradiation (2.5 Mrad).

2.5 M. tuberculosis antigens expressed in E. coli

Crude E. coli lysates containing M. tuberculosis antigens were prepared by inducing λgt11 recombinant lysogens of the E. coli strain Y1089 [9]. Single colonies of recombinant lysogens were inoculated into LB medium and cultured at 32°C. When the cultures had grown to OD600=0.5, the temperature was raised to 45°C and incubation was continued for 20 min. Isopropyl β-d-thiogalactopyranoside was added to 10 mM and the cells were additionally cultured at 38°C for 60 min. The cells were harvested by centrifugation, resuspended in 1/20 of the original culture volume in saline, and frozen immediately in liquid nitrogen. Thawing of frozen cells resulted in lysis of the induced lysogens. Viscosity of the crude lysates was reduced by brief sonication (3×10 s) and protein content was determined by measurement of OD280. After sterilization by Millipore filtration, lysates were stored frozen at −70°C. Expression of the individual antigens was confirmed by probing crude lysates on nitrocellulose filters with the appropriate monoclonal antibodies [10]. All handling of lysates was performed on ice to reduce proteolysis.

2.6 Antigen-induced proliferation of T cell clones

T cell clones were tested for reactivity against mycobacterial antigens and E. coli lysates in an in vitro proliferation assay. Cloned cells (1×104) and 1×105 autologous irradiated PBMC were distributed to each well of 96-well, flat-bottomed plates. Mycobacterial antigens as whole bacilli and E. coli lysates were added in triplicate or duplicate at 25 µg and 50 µg protein ml−1 respectively. The total culture volume was kept at 200 µl. After 72 h of incubation at 37°C in a humidified atmosphere of 5% CO2 and 95% air, cultures were given a 4-h pulse of 0.045 mBq [3H]thymidine (specific activity 185×103 mBq mM−1). The cells were harvested and the radioactivity incorporated was determined by liquid scintillation counting. The results are expressed as stimulation index (SI) values which are defined as cpm with Ag/cpm without Ag. A T cell clone was considered to be proliferating in response to a given antigen when SI> 5.

2.7 Interferon-γ (IFN-γ) assay

T cell clones (2×105 cells ml−1) were distributed to wells of 24-well Costar plates with adherent cells from 1×106 irradiated autologous PBMC and M. tuberculosis at optimal concentrations. Cell-free supernatants, collected after 48 h of incubation, were assayed for IFN-γ activity by the method of Dahl and Degre [11], by using human embryonic fibroblasts and vesicular stomatitis virus as the challenge virus.

2.8 Cytotoxicity assay

Adherent cells from 1×106 autologous irradiated PBMC in 24-well Costar plates were pulsed with Ag at optimal concentrations and the density of cloned T cells were adjusted to 1×105 cells per well. After 7 days of incubation at 37°C, the T cells were washed off, 0.5 ml 0.03% neutral red (in saline+10% FBS) was added to each well and the plates were incubated for 30 min. Neutral red was then removed from the wells by washing, and the dye taken up by macrophages was released by adding 0.5 ml 0.05 M acetic acid in 50% ethanol [12]. Percentage cytotoxicity was calculated from spectrophotometric measurement at OD540 according to the formula: Embedded Image where OD540 control=OD540 of cultures with adherent cells+T cell clone and OD540 experimental=OD540 of cultures with adherent cells+T cell clone+Ag.

3 Results

3.1 Screening of M. tuberculosis reactive T cell clones against defined recombinant antigens expressed in E. coli

M. tuberculosis reactive T cell clones were established from PBMC of naturally converted PPD-positive healthy donors (not BCG vaccinated) by using whole M. tuberculosis as the primary stimulation antigen in vitro. Expanded T cell clones were retested against M. tuberculosis in a proliferative assay before screening against recombinant antigens.

Forty-five M. tuberculosis reactive T cell clones were tested for their proliferative response against four well defined recombinant M. tuberculosis antigens. These proteins were initially expressed in E. coli by the λgt11 clones Y3155, Y3147, Y3143, and Y3272 that were previously isolated from an M. tuberculosis genomic DNA library with the monoclonal antibodies TB68, TB23, TB78, and IT-11 directed against well defined protein antigens of 16, 19, 65 (HSP65), and 71 (HSP70) kDa, respectively [13,14]. Table 1 provides basic information of the recombinant M. tuberculosis antigens used. Lysogens of recombinant λgt11 phages carrying each of the four antigen-encoding genes were established in E. coli Y1089, and antigen expression of fusion protein from the LacZ operon was induced by IPTG [9]. Before using the lysates as antigen in T cell proliferation assays, appropriate expression of each M. tuberculosis protein antigen was confirmed by probing lysates with the corresponding monoclonal antibodies in dot blot analysis (results not shown).

View this table:
Table 1

Recombinant M. tuberculosis protein antigens used in this study

Vector DNA recombinantProtein antigen expressedDistributionmAb
Y315516 kDaM. tuberculosis complex-specificTB68
Y314719 kDaLimited crossreactiveTB23
Y314365 kDa, HSP65CrossreactiveTB78
Y327271 kDa, HSP70CrossreactiveIT-11

All of the M. tuberculosis reactive T cell clones were then screened for reactivity against the M. tuberculosis 16-, 19-, 65-, and 71-kDa recombinant antigens by adding the respective crude E. coli lysate to a final concentration of 50 µg ml−1 in the in vitro proliferation assay. E. coli Y1089 lysate made from bacteria infected with the wild-type λgt11 vector, lacking mycobacterial DNA insert, was used as a negative control in the proliferation assays. This screening resulted in the identification of one T cell clone (IMK3) recognizing the M. tuberculosis 16-kDa protein antigen as expressed by the recombinant phage Y3155. In addition, by titration of the recombinant antigen added to the proliferation assay, it was possible to obtain a dose-response curve of the 16-kDa antigen-specific T cell response with a maximum response in the range of 50–200 µg ml−1E. coli lysate added (Fig. 1). None of the M. tuberculosis reactive T cell clones tested here responded to the recombinant 19-, 65-, or 71-kDa protein antigens as expressed by the phages Y3147, Y3143, and Y3272, respectively.

Figure 1

Proliferation (cpm) of the CD4+ T cell clone IMK3 in response to E. coli lysate containing the M. tuberculosis 16-kDa antigen expressed from phage Y3155 (back row) and lysate from E. coli infected with vector without DNA insert used as negative control (front row).

3.2 Species specificity and functional properties of 16-kDa antigen reactive T cells

The specificity of the 16-kDa antigen reactive T cell clone (IMK3) was investigated by testing this clone for proliferative reactivity against a panel of pathogenic as well as environmental mycobacterial species. The results showed that the proliferative response of this T cell clone was restricted to M. tuberculosis, M. africanum, and M. bovis BCG (Fig. 2).

Figure 2

Proliferation (SI) of the M. tuberculosis 16-kDa antigen reactive T cell clone IMK3 to a panel of pathogenic and environmental mycobacterial species.

With respect to cytokine analysis, the T cell clone IMK3 was able to secrete a significant level of INF-γ (20 U ml−1) when stimulated with M. tuberculosis in the presence of autologous PBMC. Surface marker analyses revealed that this clone showed the CD4+, CD8 phenotype. In addition, the ability of the 16-kDa antigen reactive T cell clone to exert major histocompatibility complex (MHC) class II restricted cytotoxicity was demonstrated by lysis of M. tuberculosis pulsed autologous macrophages in vitro (74% cytotoxicity).

4 Discussion

The lack of BCG-induced protection against tuberculosis in endemic regions [2,3] has motivated the search for new strategies to control the disease. In this context, identification of individual key antigens with relevance to T cell-mediated protection is important to develop novel subunit vaccines. In addition, the need for specific diagnostic reagents, which make it possible to discriminate between exposure to M. tuberculosis and environmental mycobacteria, also makes it important to identify such antigens.

To map the human T cell repertoire with potential relevance to protection, we have established M. tuberculosis reactive T cell clones from naturally converted PPD-positive healthy subjects without BCG vaccination and known clinical disease. A panel of recombinant phages expressing defined M. tuberculosis protein antigens have previously been identified and isolated from a λgt11 M. tuberculosis genomic DNA expression library by screening with monoclonal antibodies [13,14]. In this work four of these recombinant antigens were expressed in E. coli and screened for T cell reactivity against the M. tuberculosis reactive T cell clones established. The results showed that the recombinant 16-kDa protein antigen was able to induce a significant and dose-dependent proliferative T cell response. These findings confirm and extend our previous observations that human T cell clones can recognize well characterized antigens expressed from recombinant DNA in foreign hosts [1518]. By using the same approach, we and other groups have earlier identified both polyclonal and clonal T cell populations from different donor groups recognizing the recombinantly expressed 19-, 65-, and 71-kDa antigens of M. tuberculosis[7,1921]. In addition, both the 65- and 71-kDa heat shock proteins have been mapped with respect to peptide-defined epitopes and HLA molecules used in antigen presentation [2224].

The crossreactivity data obtained revealed that the M. tuberculosis 16-kDa antigen reactive T cell clone (IMK3) was M. tuberculosis complex specific in the sense that it only recognized M. tuberculosis, M. africanum, as well as the vaccine strain M. bovis BCG. These results are consistent with the earlier report that the monoclonal antibody TB68, which recognizes the M. tuberculosis 16-kDa antigen, showed the same species specificity when tested in Western blots including a variety of mycobacterial species [25,26]. Our finding is relevant to subunit vaccine design, because prophylaxis based on cellular immune responses against M. tuberculosis-specific antigens may be favorable to avoid downregulating by a high level of exposure to crossreactive antigens present in most environmental mycobacterial species [3]. However, due to the M. tuberculosis complex-specific distribution, the 16-kDa antigen is also relevant to diagnosis as previously demonstrated by serological approaches [25,27]. In addition, the M. tuberculosis complex-specific nature of the clonal T cells studied here suggests that at least some parts of the 16-kDa antigen sequence have a diagnostic potential as measured at the cellular immune response level.

Consistent with our results, others have reported that native affinity purified 16-kDa antigen from M. tuberculosis sonic extracts (TB68 affinity) can stimulate human T cells [28], but the present work represents the first report on T cell recognition of the E. coli expressed recombinant protein within subjects who are able to control infection with M. tuberculosis without BCG vaccine-induced protection. This observation suggests that the 16-kDa antigen might be one of several M. tuberculosis antigens used as targets for a protective immune response in humans. In addition, the relevance of the 16-kDa antigen reactive CD4+, CD8 T cells to protective immunity functions was demonstrated by their ability to secrete INF-γ and exhibit MHC class II restricted cytotoxicity against autologous macrophages pulsed with M. tuberculosis. INF-γ producing CD4+ T cells with cytotoxic activity (Th1 subset) have been shown to play an important role in protective immunity against M. tuberculosis[29]. Our results are further supported by data from animal studies where both lymphocyte proliferation in vitro and delayed-type hypersensitivity reactions in vivo were induced by immunization with the recombinantly expressed 16-kDa antigen [30]. With respect to nomenclature, this antigen was originally designated a 14-kDa molecule, but DNA sequencing of its gene revealed a deduced molecular mass of 16.3 kDa [25]. In addition, sequence homology studies showed that the antigen, which is considered to be a major membrane protein [27], belongs to the a-crystallin family of low molecular mass heat shock proteins [25]. Interestingly, this finding is consistent with the notion that heat shock proteins are immunodominant targets of the immune response against mycobacteria [29,31,32].

Similar to earlier findings [7], the majority of the T cell clones established and studied here did not respond to any of the recombinant antigens tested. This may be due to the fact that only a small donor group with a limited HLA background was included in the screening. In addition, the T cell clones were tested against a limited number of protein antigens available by recombinant DNA technology. In designing effective subunit vaccines against tuberculosis, one should consider T cell screening of a large range of cytosolic, cell wall-associated as well as secreted antigens which have recently been claimed to be important to protection [33].


This work was supported financially by Laurine Maarschalks Funding, Oslo, Norway. We want to thank Dr. R.A. Young for providing recombinant phages encoding defined M. tuberculosis antigens.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
View Abstract