Identification and HLA Restriction of Naturally Derived Th1-Cell Epitopes from the Secreted Mycobacterium tuberculosis Antigen 85B Recognized by Antigen-Specific Human CD4⁺ T-Cell Lines

Abstract

Antigen 85B (Ag85B/MPT59) is a major secreted protein from Mycobacterium tuberculosis which is a promising candidate antigen for inclusion in novel subunit vaccines against tuberculosis (TB). The present study was undertaken to map naturally derived T-cell epitopes from M. tuberculosis Ag85B in relation to major histocompatibility complex (MHC) class II restriction. Antigen-specific CD4⁺ T-cell lines were established from HLA-typed TB patients and Mycobacterium bovis BCG vaccinees by stimulation of peripheral blood mononuclear cells with purified Ag85B in vitro. The established T-cell lines were then tested for proliferation and gamma interferon (IFN-γ) secretion in response to 31 overlapping synthetic peptides (18-mers) covering the entire sequence of the mature protein. The results showed that the epitopes recognized by T-cell lines from TB patients were scattered throughout the Ag85B sequence whereas the epitopes recognized by T-cell lines from BCG vaccinees were located toward the N-terminal part of the antigen. The T-cell epitopes represented by peptides p2 (amino acids [aa] 10 to 27), p3 (aa 19 to 36), and p11 (aa 91 to 108) were frequently recognized by antigen-specific T-cell lines from BCG vaccinees in both proliferation and IFN-γ assays. MHC restriction analysis demonstrated that individual T-cell lines specifically recognized the complete Ag85B either in association with one of the self HLA-DRB1, DRB3, or DRB4 gene products or nonspecifically in a promiscuous manner. At the epitope level, panel studies showed that peptides p2, p3, and p11 were presented to T cells by HLA-DR-matched as well as mismatched allogeneic antigen-presenting cells, thus representing promiscuous epitopes. The identification of naturally derived peptide epitopes from the M. tuberculosis Ag85B presented to Th1 cells in the context of multiple HLA-DR molecules strongly supports the relevance of this antigen to subunit vaccine design.

Tuberculosis (TB) is one of the most important infectious diseases worldwide based on incidence (8 to 10 million cases) and annual mortality (3 to 4 million cases) (World Health Organization fact sheet 93, 1995). The rapid spread of TB in Africa and Asia is being accelerated by the AIDS epidemic, and the emergence of multidrug-resistant TB underlines the need for new and efficient control measures. Vaccination with Mycobacterium bovis BCG has been used for more than 70 years, but its efficacy varies tremendously in different populations (9). Identification and characterization of candidate antigens to be used in novel TB vaccines with protective effect in all parts of the world is therefore required.

Since protection against TB is mediated by cellular immune responses, a primary criterion for selection of any antigen as a subunit vaccine candidate is its ability to induce protective T-cell responses. Mycobacterium tuberculosis is rich in antigens that induce cell-mediated immunity, and the presence of such antigens in purified cell walls, the cytosolic fraction, and culture filtrates (CF) has been reported (1, 23, 61). However, several recent studies have demonstrated that antigens present in CF are among the primary inducers of protective immunity against challenge with live M. tuberculosis in mice and guinea pigs (reviewed in references 1 and 8). Furthermore, the use of memory immune mice has demonstrated that CF antigens with molecular masses of 6 to 10 kDa (ESAT-6) and 26 to 34 kDa (Ag85 complex) are strongly recognized by the T helper 1 (Th1) type of CD4⁺ T cells during infection with M. tuberculosis (2). In addition, DNA vaccination with ESAT-6, Ag85B, and Ag85A induces CD8⁺ cytotoxic T cells and protection against challenge with live M. tuberculosis or M. bovis BCG in mice (13, 16, 20).

By screening peripheral blood mononuclear cells (PBMC) for proliferation and gamma interferon (IFN-γ) secretion in response to a panel of well-defined secreted and cytosolic antigens, we have previously shown that Ag85B is frequently recognized by human Th1 cells after natural infection with M. tuberculosis (24). However, to induce protection in an HLA-heterogeneous human population, subunit vaccine antigens should contain epitopes recognized by T cells in the context of multiple HLA class II molecules. Synthetic peptides covering the Ag85B sequence have previously been used to map epitopes recognized by nonselected human PBMC (54, 55). Although these studies suggested that multiple HLA class II molecules were able to present Ag85B peptides to T cells, this approach did not allow major histocompatibility complex restriction analysis of individual T-cell epitopes relevant to natural processing of the antigen.

To further understand the molecular basis for the permissive T-cell recognition of Ag85B, we have established and screened antigen-specific CD4⁺ T-cell lines from HLA-DR-typed TB patients and healthy BCG vaccinees for proliferation and IFN-γ secretion in response to synthetic peptides covering the mature Ag85B sequence. Importantly, this approach allowed us to map naturally derived T-cell epitopes in relation to major histocompatibility complex restriction. The results showed that T-cell lines from TB patients responded to peptides scattered throughout the Ag85B sequence whereas the response of T-cell lines from BCG vaccinees was restricted to the N-terminal part of the antigen. In addition, we have identified peptide epitopes relevant to natural processing of this antigen which are presented to the responding Th1 cells in association with multiple HLA-DR molecules encoded by the HLA-DRB1, DRB3, and DRB4 genes.

MATERIALS AND METHODS

Complex and purified mycobacterial antigens.

Irradicated M. tuberculosis was kindly provided by J. Eng (National Institute of Public Health, Oslo, Norway). CF highly enriched for secreted antigens of M. tuberculosis with only trace amounts of intracellular soluble antigens such as GroES, GroEL, and DnaK was prepared as previously described (41, 59). In brief, M. tuberculosis H37Rv (ATCC 27294) was cultured as surface pellicles on the wholly synthetic Sauton medium. The culture fluid was concentrated by 80% ammonium sulfate precipitation, extensively dialyzed against phosphate-buffered saline (pH 7.4), and further washed twice by ultrafiltration in Centricon 3 tubes (Amicon, Inc., Beverly, Mass.) and centrifugation at 7,500 × g for 2 h. CF antigen was finally sterile filtered and reconstituted to 1.1 mg/ml. Ag85B (MPT59) (lot 12455A2) was purified from M. tuberculosis as previously described (40). Lyophilized purified antigen was solubilized in phosphate-buffered saline washed twice in Centricon 3 tubes, sterile filtered, and reconstituted to 1.2 mg/ml. Aliquots were kept frozen at −20°C until use.

Synthetic peptides.

Thirty-one peptides (18-mers overlapping by 9 amino acids [aa]) spanning the mature Ag85B sequence from M. bovis BCG (Fig. 1) were synthesized on Tenta-Gel-S-RAM and purified by high-pressure liquid chromatography as described before (14, 18). This sequence is identical to that of M. tuberculosis except for an L instead of an F at position 100. Peptides p13 and p14 were synthesized with P and S at positions 121 and 122, respectively, and peptides p18 and p19 were synthesized without P and S between residues 162 and 163 (Fig. 1).

FIG. 1.

Thirty-one synthetic overlapping 18-mer peptides covering the amino acid sequence of mature Ag85B. The protein sequence is given in the one-letter code for amino acids. The regions covered by peptides p1 to p31 are indicated by horizontal bars.

Isolation of PBMC.

PBMC were isolated from heparinized blood of smear- and culture-positive pulmonary TB patients treated for 1 month (attending the Chest Diseases Hospital, Kuwait, Kuwait) and from buffy coats of BCG-vaccinated healthy subjects (blood donors at the Central Blood Bank, Kuwait, Kuwait, and the Blood Bank at Ullevål Hospital, Oslo, Norway) by flotation on Lymphoprep gradients using standard procedures (28). The cells were finally suspended in complete tissue culture medium (RPMI 1640, 10% human AB serum, 100 U of penicillin per ml, 100 μg of streptomycin per ml, 40 μg of gentamicin per ml, 2.5 μg of amphotericin B per ml) and counted in a Coulter Counter (Coulter Electronics Ltd., Luton, England).

HLA typing of PBMC.

PBMC were HLA typed genomically by using sequence-specific primers in a PCR as described by Olerup and Zetterquist (49). HLA-DR “low-resolution” kits containing the primers to type for DRB1, DRB3, DRB4, and DRB5 alleles were purchased from Dynal AS (Oslo, Norway) and used in a PCR as specified by the manufacturer. DNA amplifications were carried out in a Gene Amp PCR system 2400 (Perkin-Elmer Cetus), and the amplified products were analyzed by gel electrophoresis, using standard procedures (25). Serologically defined HLA-DR specificities were determined from the genotypes by following the guidelines provided by Dynal AS.

Antigen-induced proliferation of PBMC.

Antigen-induced proliferation of PBMC was performed by standard procedures (12, 29, 37). In brief, PBMC (2 × 10⁵ cells/well) suspended in 50 μl of complete tissue culture medium were seeded into 96-well tissue culture plates (Nunc, Roskilde, Denmark). Antigen in 50 μl of complete medium was added to the wells in duplicate or triplicate at an optimal concentration of 5 μg/ml (24). The final volume of the culture in the wells was adjusted to 200 μl. The plates were incubated at 37°C in a humidified atmosphere of 5% CO₂–95% air. The cultures were pulsed (4 h) on day 6 with 1 μCi of [³H]thymidine (Amersham Life Science, Little Chalfont, England) and harvested on filter mats with a Skatron harvester (Skatron Instruments AS, Oslo, Norway), and the radioactivity incorporated was measured by liquid scintillation counting as described previously (27, 31, 35).

Establishment of antigen-specific T-cell lines.

Antigen-specific T-cell lines were established from the donors by stimulating PBMC with purified Ag85B by procedures described previously (21, 22, 34). In brief, PBMC (2 × 10⁵ cells/well) were stimulated with the antigen (5 μg/ml) in triplicate in 96-well plates and incubated at 37°C in an atmosphere of 5% CO₂–95% air. After 6 days, interleukin-2 (100 U/well) (Amersham Life Sciences) was added twice a week until the cell culture density allowed transfer to 24-well tissue culture plates (Nunc). The growing T-cell lines were expanded in 24-well plates, with addition of interleukin-2 twice a week, until tested for antigen reactivity.

Antigen- and peptide-induced proliferation of T-cell lines.

The T-cell lines were tested for antigen- and peptide-induced proliferation in the presence of autologous and allogeneic HLA-typed antigen-presenting cells (APC), using procedures described previously (30, 38, 42). In brief, irradiated (2,400 rads) PBMC were seeded into the wells of 96-well plates at a concentration of 10⁵ cells/well. The plates were incubated at 37°C in a humidified atmosphere of 5% CO₂–95% air. Nonadherent cells were removed, and adherent cells were washed three times with tissue culture medium (RPMI 1640) and used as APC. Antigen-specific T-cell lines were harvested, washed three times, and added to the APC-containing wells at a concentration of 5 × 10⁴ cells/well. Antigen and peptides were added in triplicate at a final concentration of 5 μg/ml, and the culture volume in the wells was made up to 200 μl with complete tissue culture medium. The plates were incubated at 37°C in an atmosphere of 5% CO₂–95% air. On day 3, the cultures were pulsed (4 h) with 1 μCi of [³H]thymidine and harvested on filter mats, and the radioactivity incorporated was determined by liquid scintillation counting as described previously (26, 43, 46).

Interpretation of proliferation results.

The radioactivity incorporated was obtained as counts per minute (cpm). Average cpm were calculated from triplicate or duplicate cultures stimulated with each antigen or peptide. Cellular proliferation results are presented as a stimulation index (SI), which is defined as follows: SI = cpm in antigen-stimulated cultures/cpm in cultures without antigen. An SI of ≥5 was considered a positive proliferative response (24), and such values are given in bold in the tables.

IFN-γ assay.

Supernatants (100 μl) were collected from antigen-stimulated cultures of PBMC and T-cell lines (96-well plates) before being pulsed with [³H]thymidine. The supernatants were kept frozen at −70°C until assayed for IFN-γ activity. The amount of IFN-γ in the supernatants was quantitated by using PREDICTA immunoassay kits (Genzyme Co., Cambridge, Mass.) as specified by the manufacturer. The detection limit of the IFN-γ assay kit was 8 pg/ml. Secretion of IFN-γ in response to a given antigen or peptide was considered positive when delta IFN-γ (the IFN-γ concentration in cultures stimulated with antigen minus the IFN-γ concentration in cultures without antigen) was ≥500 pg/ml (24). Such values are given in bold in the tables.

Inhibition assays with monoclonal anti-HLA antibodies.

Inhibition of antigen-induced T-cell proliferation was studied as described previously (26, 32, 43, 45, 46) with the monoclonal antibodies W6/32 (anti-HLA class I) and L243 (anti-HLA-DR), purchased from American Type Culture Collection, Rockville, Md., as well as FN81 (anti-HLA-DQ), a gift from S. Funderud, Oslo, Norway. In brief, adherent APC in the wells of 96-well flat-bottom plates were preincubated with the antibodies for 30 min at 37°C in an atmosphere of 5% CO₂–95% air. After preincubation, antigen-induced proliferation of T-cell lines was assayed as described above. The results were expressed as percent inhibition [1 − (cpm in antigen-stimulated cultures in the presence of antibodies/cpm in antigen-stimulated cultures in the absence of antibodies)] × 100.

RESULTS

Identification of Ag85B-responding donors.

To identify suitable donors for establishing Ag85B-reactive T-cell lines, PBMC from 22 TB patients and 19 BCG-vaccinated healthy subjects were screened for proliferation and IFN-γ secretion in response to M. tuberculosis, CF antigen, and purified Ag85B (Table 1). The results showed that PBMC from a majority of the donors responded to M. tuberculosis and CF antigen whereas PBMC from approximately half of the donors in both groups responded to Ag85B in proliferation and IFN-γ assays (Table 1). Ag85B-responding patients and BCG vaccinees were then selected as donors for establishing antigen-specific T-cell lines.

TABLE 1.

Antigen-induced proliferation and secretion of IFN-γ from PBMC of TB patients and BCG-vaccinated healthy subjects in response to M. tuberculosis, CF, and Ag85B

Antigen	Response of PBMC [no. positivea/no. tested (%)] from:
	TB patients		BCG vaccinees
	Proliferation	IFN-γ	Proliferation	IFN-γ
M. tuberculosis	20/22 (91)	19/19 (100)	18/19 (95)	18/19 (95)
CF	19/22 (86)	18/19 (95)	17/18 (94)	15/18 (83)
Ag85B	10/22 (45)	9/19 (47)	12/18 (67)	9/19 (47)

^aAntigen-induced proliferation with an SI of ≥5 and IFN-γ secretion with a delta IFN-γ of ≥500 pg/ml were considered positive responses. 

Proliferation and IFN-γ secretion of antigen-specific T-cell lines in response to synthetic Ag85B peptides.

Primary stimulation of PBMC with purified Ag85B resulted in establishment of antigen-specific T-cell lines from 9 TB patients (Table 2) and 11 BCG vaccinees (Table 3) as demonstrated by Ag85B-induced proliferation and IFN-γ secretion. Surface marker analysis revealed that all of the established T-cell lines showed the CD4+ CD8− phenotype.

TABLE 2.

Proliferation of Ag85B-induced T-cell lines from TB patients in response to Ag85B and synthetic peptides

T-cell line		Assay	Proliferation (SI) and IFN-γ secretion (102 pg/ml) of T-cell lines in response to:
No.	HLA-DR		Ag85B	Peptidesa
				p3	p4	p5	p6	p9	p11	p13	p16	p17	p21	p23	p26	p30
1	6,7,52,53	Proliferation	18	1.0	1.0	1.0	1.0	20	1.0	2.0	6.0	2.0	1.0	1.0	4.0	1.0
		IFN-γ	>250	<5	<5	<5	<5	80	<5	<5	50	<5	<5	<5	6.0	<5
2	1,4,53	Proliferation	30	3.0	5.8	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0
		IFN-γ	15	<5	<5	<5	<5	<5	<5	<5	<5	<5	<5	<5	<5	<5
3	2,51	Proliferation	5.0	1.0	1.0	1.0	2.0	1.0	10	1.0	2.0	1.0	8.0	1.0	1.0	1.0
		IFN-γ	11	<5	<5	<5	<5	<5	250	<5	<5	<5	5.0	<5	<5	<5
4	2,7,51,53	Proliferation	239	1.0	1.0	1.0	2.0	1.0	1.0	41	115	91	1.0	2.0	31	1.0
		IFN-γ	>250	<5	<5	<5	<5	<5	<5	8.0	17	8.0	<5	<5	<5	<5
5	2,51	Proliferation	16	1.0	3.0	3.0	28	2.0	3.0	1.0	2.0	2.0	1.0	2.0	2.0	5.6
		IFN-γ	8.0	<5	<5	<5	25	<5	<5	<5	<5	<5	<5	<5	<5	<5
6	5,52	Proliferation	47	1.0	1.0	1.0	1.3	2.0	53	1.0	1.0	1.0	1.0	9.0	1.0	1.0
		IFN-γ	>250	<5	<5	<5	<5	<5	200	<5	<5	<5	<5	5.0	<5	<5
7	6,7,52,53	Proliferation	37	4.0	2.1	1.8	1.2	1.1	1.6	7.8	1.3	1.2	1.0	2.2	0.7	0.7
		IFN-γ	37	<5	<5	<5	<5	<5	<5	11	<5	<5	<5	<5	<5	<5
8	7,53	Proliferation	13	40	14	5.3	2.8	0.9	1.5	1.7	13	3.4	1.5	1.6	1.4	0.6
		IFN-γ	158	208	36	7.0	25	<5	<5	<5	168	15	<5	<5	<5	<5
9	6,52	Proliferation	25	1.7	1.3	11	2.4	1.5	1.3	1.9	28	15	2.7	1.3	1.9	1.2
		IFN-γ	25	<5	<5	6.0	<5	<5	<5	<5	16	14	<5	<5	<5	<5
No. positive/no. tested		Proliferation	9/9	1/9	2/9	2/9	1/9	1/9	2/9	2/9	4/9	2/9	1/9	1/9	1/9	1/9
		IFN-γ	9/9	1/9	1/9	2/9	2/9	1/9	2/9	2/9	4/9	3/9	1/9	1/9	1/9	1/9

^aPeptides of Ag85B not shown in the table did not induce positive responses. Positive responses (SI ≥ 5 and delta IFN-γ ≥ 500 pg/ml) are given in bold type. 

TABLE 3.

Proliferation and IFN-γ secretion by Ag85B-induced T-cell lines from BCG-vaccinated healthy subjects in response to Ag85B and synthetic peptides

T-cell line		Proliferation (SI) of T-cell lines in response to:						IFN-γ secretion (102 pg/ml) by T-cell lines in response to:
No.	HLA-DR	Ag85B	Peptidesa					Ag85B	Peptidesa
			p1	p2	p3	p4	p11		p1	p2	p3	p4	p11
1	2,6,51,52	25	6.0	39	15	1.0	6.0	50	<5	<5	<5	<5	23
2	7,53	6.0	61	1.0	6.0	1.0	18	25	250	<5	25	<5	25
3	4,6,52,53	11	1.0	1.0	1.0	1.0	20	NDb	ND	ND	ND	ND	ND
4	6,8,52	45	1.0	1.0	1.0	1.0	5.0	1,500	<5	<5	<5	<5	150
5	5,7,52,53	216	2.0	67	125	1.0	34	250	<5	240	500	<5	230
6	3,4,52,53	294	4.0	502	67	1.0	8.0	600	<5	22	300	<5	22
7	3,4,52,53	19	1.0	38	13	4.0	1.0	18	<5	50	15	<5	<5
8	3,5,52	23	1.0	6.0	1.0	1.0	6.0	17	<5	8.0	<5	<5	20
9	1,6,52	6.6	1.0	2.0	1.0	22	1.0	25	<5	<5	<5	10	<5
10	6,7,52,53	49	1.0	2.0	2.0	2.0	12	60	<5	<5	<5	<5	250
11	3,3,52	48	1.0	41	9.0	2.0	14	ND	ND	ND	ND	ND	ND
No. positive/no. tested		11/11	1/11	6/11	6/11	1/11	9/11	9/9	1/9	4/9	4/9	1/9	7/9

^aPeptides of Ag85B not shown in the table did not induce positive responses. Positive responses (SI ≥ 5 and delta IFN-γ ≥ 500 pg/ml) are given in bold. 

^bND, not determined. 

All the Ag85B-induced T-cell lines were then screened for both proliferation and IFN-γ secretion in response to 31 overlapping synthetic peptides covering the mature Ag85B protein sequence (Fig. 1). The results obtained with T-cell lines from TB patients showed that they responded to a total of 13 and 12 peptides from the tested series of 31 peptides in proliferation and IFN-γ assays, respectively (Table 2). The following Ag85B peptides were shown to be stimulatory: p3 (aa 19 to 36), p4 (aa 28 to 45), p5 (aa 37 to 54), p6 (46 to 63), p9 (aa 73 to 90), p11 (aa 91 to 108), p13 (aa 109 to 126), p16 (aa 136 to 153), p17 (aa 145 to 162), p21 (aa 181 to 198), p23 (aa 199 to 216), p26 (aa 226 to 243), and p30 (aa 262 to 279). All peptides which induced the proliferation of T-cell lines in this group (except peptide p30) also stimulated the secretion of IFN-γ. However, on an individual basis, most of the peptides were stimulatory for T-cell lines from one or two donors, except for peptide p16, which stimulated T-cell lines from four of the nine donors in both proliferation and IFN-γ assays (Table 2).

In contrast to the T-cell lines from TB patients, the corresponding lines from BCG vaccinees responded to only five peptides from the N terminus of the Ag85B in proliferation and IFN-γ assays (Table 3). These stimulatory peptides were p1 (aa 1 to 18), p2 (aa 10 to 27), p3 (aa 19 to 36), p4 (aa 28 to 45), and p11 (aa 91 to 108). Among these peptides, p1 and p4 stimulated only one T-cell line each whereas p2, p3, and p11 stimulated 6, 6, and 9 T-cell lines (out of a total of 11) in proliferation assays and 4, 4, and 7 T-cell lines (out of a total of 9) in IFN-γ assays, respectively (Table 3).

HLA restriction analysis of Ag85B and peptide presentation to antigen-specific T-cell lines.

The HLA restriction of the Ag85B T-cell responses observed here was analyzed by combining results from HLA typing, antibody-blocking assays, and panel studies with HLA-typed APC. HLA typing of the TB patients and BCG vaccinees used to establish T-cell lines showed that they constituted a heterogeneous group with respect to HLA-DR, expressing DR1, DR2, DR3, DR4, DR5, DR6, DR7, DR8, DR51, DR52, and DR53 molecules (Tables 2 and 3). Inhibition assays with monoclonal antibodies against HLA class I and II molecules (anti-HLA-DR and anti-HLA-DQ) showed that Ag85B and the selected positive peptides were presented to T cells in the context of HLA-DR molecules. The results of representative experiments demonstrating anti-HLA-DR blocking of antigen- and peptide-induced T-cell proliferation are shown in Fig. 2.

FIG. 2.

Inhibition of the proliferative response of the T-cell line S-45 in the presence of anti-HLA class I and class II monoclonal antibodies. The T-cell line S-45 was established from a TB patient after stimulation of PBMC with Ag85B as described in Materials and Methods. The T-cell line was stimulated with Ag85B (A), peptide p13 (B), peptide p16 (C), or peptide p17 (D) in the presence of defined monoclonal antibodies to HLA class I and class II molecules at the concentrations indicated. The percent inhibition (defined in Materials and Methods) of the proliferative response is given for anti-HLA class I (□), anti-HLA-DQ (○), and anti-HLA-DR (●) antibodies.

To identify the HLA-DR molecules responsible for presentation of Ag85B and individual peptides, selected T-cell lines were then tested for proliferation in the presence of HLA-DR-typed autologous and allogeneic APC. This analysis demonstrated that individual T-cell lines specifically recognized the complete Ag85B either in association with one of the self HLA-DRB1, DRB3, or DRB4 gene products or nonspecifically in a promiscuous manner. Ag85B was exclusively presented to the T-cell lines S-45, S-47, SN-24, SN-23, and SN-39 by HLA-DR7, HLA-DR2, HLA-DR3, HLA-DR52, and HLA-DR53, respectively (Tables 4 and 5), whereas the T-cell lines SN-21 and S-50 were able to recognize Ag85 in the presence of multiple HLA-DR molecules (Table 5). With respect to HLA restriction for peptide presentation, the peptides recognized by T-cell lines from single donors were specifically restricted by one of the self HLA-DRB1 gene products expressed by the donor. Peptides p13, p16, p17, and p26, recognized by the T-cell line S-45 (HLA-DR2,7,51,53), were presented by HLA-DR7, and peptide p6, recognized by the T-cell line S-47 (HLA-DR2,51), was presented by HLA-DR2 (Table 4). In contrast, Ag85B peptides recognized by T-cell lines from several donors showed the ability to be presented by multiple HLA-DR molecules. A representative example is peptide p2, which was presented by HLA-DR53, HLA-DR52, and HLA-DR3 to the T-cell lines SN-21, SN-23, and SN-24, respectively (Table 5). In addition, we found that the peptides p3 and p11 were presented to the responding T-cell lines in an even more promiscuous fashion, involving multiple HLA-DR molecules (Table 5).

TABLE 4.

Proliferation of T-cell lines in response to Ag85B and its peptides in the presence of HLA-DR-typed APC^a

APC HLA-DR	Proliferation (SI) of T-cell line S-45 (HLA-DR2,7,51,53)					Proliferation (SI) of T-cell line S-47 (HLA-DR2,51)		Proliferation (SI) of T-cell line S-39 (HLA-DR1,4,53), Ag85B
	Ag85B	Peptide p13	Peptide p16	Peptide p17	Peptide p26	Ag85B	Peptide p6
2,7,51,53	171	41	41	97	12	3.5	3.4	NDb
7,53	142	60	76	62	17	1.2	2.0	ND
5,7,52	78	3.8	13	7.7	7.0	ND	ND	ND
4,7,53	44	2.2	14	6.1	1.7	ND	ND	ND
4,5,52,53	1.4	2.1	2.7	1.3	1.4	4.2	1.9	ND
2,51	2.2	1.7	1.4	1.0	0.9	7.2	7.2	0.8
2,6,51,52	1.3	1.1	0.8	1.1	1.0	13	7.6	ND
2,4,51,53	1.7	1.1	1.0	1.0	1.1	ND	ND	ND
1,6,52	ND	ND	ND	ND	ND	1.1	1.1	1.1
1,4,53	ND	ND	ND	ND	ND	ND	ND	65
3,10,52	ND	ND	ND	ND	ND	ND	ND	1.2
1,7,53	ND	ND	ND	ND	ND	ND	ND	83
5,6,52	ND	ND	ND	ND	ND	ND	ND	0.8
3,5,52	ND	ND	ND	ND	ND	ND	ND	0.4
HLA-DR restriction	DR7	DR7	DR7	DR7	DR7	DR2	DR2	DR53

^aPositive responses (SI ≥ 5) are given in bold. 

^bND, not determined. 

TABLE 5.

Proliferation of T-cell lines in response to Ag85B and its peptides in the presence of HLA-DR-typed APC^a

APC HLA-DR	Proliferation (SI) of T-cell line SN-21 (HLA-DR 5,7,52,53)				Proliferation (SI) of T-cell line SN-23 (HLA-DR 3,4,52,53)			Proliferation (SI) of T-cell line SN-24 (HLA-DR 3,4,52,53)			Proliferation (SI) of T-cell line S-50 (HLA-DR5,52)
	Ag85B	Peptide p2	Peptide p3	Peptide p11	Ag85B	Peptide p2	Peptide p3	Ag85B	Peptide p2	Peptide p3	Ag85B	Peptide p11
5,7,52,53	168	88	120	28	10	7.0	1.0	NDb	ND	ND	ND	ND
3,4,52,53	214	94	128	24	206	276	19	ND	ND	ND	ND	ND
7,53	83	81	67	20	4.0	7.0	2.0	ND	ND	ND	ND	ND
5,52	135	2.0	18	14	19	19	1.0	ND	ND	ND	47	9.0
2,51	21	3.0	12	25	ND	ND	ND	2.0	3.0	2.0	19	23
5,10,52	169	3.0	29	32	22	16	1.0	ND	ND	ND	28	30
3,5,52	138	3.0	11	22	45	44	5.0	54	88	11	63	40
1,7,53	96	27	56	22	1.0	2.0	1.0	3.0	3.0	4.0	ND	ND
3,4,52,53	65	13	46	11	83	87	10	17	31	9.0	ND	ND
3,5,52	152	2.0	9.0	32	33	34	17	ND	ND	ND	ND	ND
4,7,53	123	6.0	89	23	1.0	2.0	1.0	3.0	2.0	1.0	92	95
2,4,51,53	46	21	37	15	1.0	4.0	1.0	ND	ND	ND	6.2	8.1
6,7,52,53	ND	ND	ND	ND	ND	ND	ND	2.0	3.0	1.0	ND	ND
3,5,52	ND	ND	ND	ND	ND	ND	ND	19	46	3.0	ND	ND
6,52	ND	ND	ND	ND	ND	ND	ND	1.0	2.0	1.0	ND	ND
6,7,52	ND	ND	ND	ND	ND	ND	ND	2.0	3.0	1.0	ND	ND
3,5,52	ND	ND	ND	ND	ND	ND	ND	48	64	4.0	ND	ND
5,6,52	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	13	5.2
6,8,52	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND	31	36
HLA-DR restriction	Promiscuous	DR53	Promiscuous	Promiscuous	DR52	DR52	DR3	DR3	DR3	DR3	Promiscuous	Promiscuous

^aPositive responses (SI ≥ 5) are given in bold. 

^bND, not determined. 

DISCUSSION

In this work, we have mapped Ag85B epitopes relevant to natural antigen-processing and presentation pathways by using synthetic peptides and antigen-specific human T-cell lines established after primary stimulation of PBMC with purified Ag85B in vitro. The results obtained showed that Ag85B epitopes recognized by CD4⁺ T cells from TB patients were scattered throughout the antigen sequence (Table 2) whereas the antigen-specific T-cell responses in BCG vaccinees were directed mainly toward the N-terminally located peptides p2 (aa 10 to 27), p3 (aa 19 to 36), and p11 (aa 91 to 108) (Table 3). These results are consistent with an earlier report showing that PBMC from BCG-vaccinated donors responded frequently to the N-terminal peptides of Ag85B (54) whereas there was no such skewing of the T-cell response in TB patients tested with the peptides of either Ag85B (55) or 85A (18). We have previously shown that recognition of N-terminal peptides of Ag85A in mice injected with M. bovis BCG is dependent on the expression of MHC haplotype; i.e., mice expressing the H-2^d haplotype recognized only the N-terminal peptides whereas H-2^b-expressing mice recognized both the N- and C-terminal peptides of Ag85A (14). These results suggest that nonrecognition of C-terminal peptides of Ag85B may be associated with the expression of specific MHC molecules. However, this does not seem to be the case for BCG vaccinees, because they were quite heterogeneous with respect to the expression of MHC (HLA-DR) molecules (Table 3).

In addition to our results, the studies with PBMC from BCG vaccinees (54) as well as M. tuberculosis-exposed and purified protein derivative-positive healthy donors from different geographical regions have shown a predominant recognition of Ag85B (55) and Ag85A (18) sequences covered by peptides p2, p3, and p11. Moreover, the regions overlapping with peptides p2, p3, and p11 were also recognized by T cells from mice immunized with BCG (14) or Ag85A DNA (7) and guinea pigs immunized with Ag85B (19). The recognition of these as well as other peptides by Ag85B-specific T-cell lines described in this study demonstrated that the corresponding epitopes are not cryptic but are relevant to natural processing and presentation of this antigen to CD4⁺ T cells. Such knowledge is a prerequisite for application of synthetic peptides in subunit vaccine design.

Experimental work in animal models suggests that both CD4⁺ and CD8⁺ T cells are required for optimal protection against TB (5, 11, 50, 52). Consistent with the use of purified Ag85B as an exogenously added antigen to establish T-cell lines in vitro, the peptide results obtained here are restricted to reflect CD4⁺ T-cell responses. Although some antigens secreted from M. tuberculosis (ESAT-6 and the 38-kDa antigen) during natural infection probably have access to the HLA class I antigen presentation pathway and can induce cytotoxic CD8⁺ T-cell responses (17, 60), this has not yet been ruled out for Ag85B in humans. In mice, injection of Ag85A DNA but not M. tuberculosis and M. bovis BCG induced CD8⁺ and HLA class I-restricted cytotoxic T lymphocytes (7). Mapping with synthetic peptides identified three cytotoxic T-lymphocyte epitopes in Ag85A. Two of these epitopes, represented by regions aa 64 to 81 and aa 145 to 162, were cross-reactive with Ag85B (7). The region from aa 145 to 162 of Ag85B, represented by p17, of our series was recognized by T-cell lines from TB patients (Table 2).

We have used both proliferation and IFN-γ release as readout for antigen-specific CD4⁺ T-cell responses in vitro. The importance of the Th1 cytokine IFN-γ as the main mediator of protective immunity against mycobacterial infections has been well demonstrated in animal models (10, 51, 56). In addition, the observation of strong IFN-γ responses in purified protein derivative-positive healthy subjects and TB patients with localized disease (4, 15, 53, 57, 60) compared to weak responses in TB patients with advanced disease (15, 57, 60) suggested a critical role for IFN-γ in protective immune responses in humans. The observation of peptide-specific Th1-cell responses (secretion of IFN-γ) allowed us to suggest that recognition of the corresponding Ag85B epitopes is relevant to protective immune responses.

Recognition of mycobacterial antigens and epitopes by circulating CD4⁺ T cells is mostly restricted by HLA-DR molecules (26, 32, 33, 39, 43–47). Consistent with this, inhibition assays with anti-HLA class I and II antibodies showed that only anti-HLA-DR antibody was able to block peptide presentation. To identify the HLA-DR molecules involved in presentation of the individual Ag85B peptides, we have screened the relevant T-cell lines for antigen- or peptide-induced proliferation in the presence of allogeneic APC expressing defined DR types. In addition to presentation by the highly polymorphic HLA-DRB1 gene products, the results showed that Ag85B and peptide p2 also were presented to T cells in association with the HLA-DRB3 and DRB4 gene products, HLA-DR52 and HLA-DR53, respectively. It is well known that, due to linkage disequilibrium, HLA-DR3-, HLA-DR5-, and HLA-DR6-positive donors also express HLA-DR52 whereas HLA-DR4-, HLA-DR7-, and HLA-DR9-positive donors also express HLA-DR53. Therefore, compared to the allelic products of HLA-DRB1, which are expressed in a minor proportion of individuals in a given population, HLA-DR52 and HLA-DR53 are expressed in a large proportion of individuals (up to 80%) (3, 36). In addition to presentation of Ag85B in association with defined HLA-DR molecules (DR2, DR3, DR7, DR52, and DR53), our results showed that Ag85B and peptides p3 and p11 were presented to T-cell lines by allogeneic APC in a promiscuous manner. The promiscuous recognition pattern of these T cells could imply that the peptides involved bind mainly the nonpolymorphic HLA-DR chain and do so with less stringency to the polymorphic HLA-DR chains and that the T cells involved neglect the DR chain differences when stimulated by non-self peptide-loaded APC.

The M. tuberculosis Ag85B is a member of the Ag85 complex, which consists of four antigenic moieties known as Ag85A, Ag85B, Ag85C, and MPT51, encoded by four closely related but distinct genes (48, 58). At the mature-protein level, the sequence homology between Ag85B and Ag85A and between Ag85B and Ag85C is 80 and 68%, respectively (6). DNA vaccination with the Ag85 complex, Ag85A, and Ag85B has been shown to be protective whereas vaccination with Ag85C does not protect against mycobacterial challenge in mice (13, 16, 20). These studies suggest that the sequences recognized by protective T cells may be common to Ag85A and Ag85B but distinct from Ag85C. To possibly identify such protective sequences, we compared the peptides of Ag85B identified as predominant Th1-cell epitopes in this study with the corresponding sequences in Ag85A and Ag85C (Fig. 3). This analysis revealed that Ag85B and Ag85A are completely identical in the region covered by peptide p2 (aa 10 to 27) and showed 2 amino acid substitutions in the regions covered by peptides p3 and p11 (Fig. 3). In contrast, Ag85B and 85C have a nonconservative substitution in the peptide p2 region and five and six substitutions in the regions covered by peptides p3 and p11, respectively (Fig. 3). This analysis further suggests that the peptides p2, p3, and p11 may be relevant for protective immunity.

FIG. 3.

Sequence alignment of the Ag85B peptides p2, p3, and p11 with the corresponding sequences in Ag85A and Ag85C of M. tuberculosis. Amino acid residues in the Ag85B sequence which are identical to those in the Ag85A and Ag85C sequences are represented by dashes. The sequence information is based on data in reference 6.

In conclusion, the identification of multiple synthetic peptides from the M. tuberculosis Ag85B sequence, which can be recognized by T-cell lines induced by the complete antigen, shows that these epitopes are relevant to natural processing and presentation pathways. Furthermore, our demonstration that the Ag85B and some of its synthetic peptides are recognized by IFN-γ-secreting Th1 cells in association with multiple HLA-DR molecules strongly supports the notion that this antigen is relevant to the design of subunit vaccines against TB.