TEXT

Top Abstract Text References

The tuberculin purified protein derivative (PPD) skin test is a widely used diagnostic and epidemiological tool for tuberculous infection. However, PPD contains antigens that are shared among pathogenic mycobacteria, environmental mycobacteria, and the vaccine strain Mycobacterium bovis BCG (7, 14). The relative response to PPDs prepared from different mycobacterial strains can give some indication of exposure (19), but ideally, a species-specific reagent is required. Such a reagent should distinguish vaccinated individuals or individuals exposed to environmental mycobacteria from patients infected with Mycobacterium tuberculosis.

Recently, two extracellular antigens from M. tuberculosis with potential relevance to the diagnosis of tuberculosis (TB), ESAT-6 and MPT64, have been cloned and characterized (13, 18). They are recognized by T cells in animal models of TB and by human TB patients (3, 4, 9, 16). Both antigens have been found primarily in M. tuberculosis but not in most environmental mycobacteria or in BCG (1a, 8, 18). MPT64 has been evaluated as a skin test reagent in the guinea pig model (5, 17) and in humans (20). In this study, we have further investigated the diagnostic potential of these two antigens using the guinea pig model.

Female outbred guinea pigs of strain Ssc:AL (Statens Serum Institut, Copenhagen, Denmark) were infected intravenously (i.v.) with 10⁴ CFU of M. tuberculosis H37Rv or sensitized intradermally (i.d.) with 2 × 10⁶ CFU of BCG (BCG Danish 1331; Statens Serum Institut) or 2 × 10⁶ CFU of a clinical isolate of M. avium (Atyp.1443; Statens Serum Institut). Skin tests were performed 28 days later with 10 tuberculin units (TU) of PPD (1 TU = 0.02 µg; RT23; Statens Serum Institut) and optimal concentrations (1 µg for each protein [data not shown]) of recombinant ESAT-6 (rESAT-6) and rMPT64 (Fig. 1). rMPT64 and rESAT-6 were prepared as described previously (6, 15). Animals infected with M. tuberculosis H37Rv responded to antigen preparations with positive skin test responses (reaction diameter of 5 mm) at 24 h. Although animals sensitized with BCG or M. avium responded to PPD RT23, no reactions to rESAT-6 or rMPT64 were observed at any time up to 72 h postchallenge, reflecting the lack of expression of ESAT-6 and MPT64 in BCG Danish 1331 or M. avium.

View larger version (21K): [in this window] [in a new window]   FIG. 1.   ESAT-6 and MPT64 discriminate TB infection from BCG vaccination or exposure to environmental mycobacteria. Guinea pigs (n = 5 per group) were infected i.v. with M. tuberculosis H37Rv (M. tub.) or immunized with either BCG or M. avium for 4 weeks. They were then tested with 10 TU (0.2 µg) of PPD (), 1 µg of rESAT-6 (), or 1 µg of rMPT64 ().

Outbred Ssc:AL guinea pigs respond to an epitope in the C-terminal region of the MPT64 molecule (15). We examined regions of the ESAT-6 molecule which could elicit delayed type hypersensitivity (DTH) responses in two different outbred strains of guinea pigs, Ssc:AL and Dunkin Hartley (Møllegaard Breeding and Research Center A/S, Lille Skensved, Denmark). Overlapping peptides from ESAT-6 were synthesized, and 1 µg of each peptide was used for skin testing animals 4 weeks after infection with M. tuberculosis H37Rv (Fig. 2). All animals responded to PPD as expected, and skin reactions in the Dunkin Hartley strain were generally more indurate than those in the Ssc:AL strain. Three of eight Ssc:AL and two of six Dunkin Hartley guinea pigs responded to rESAT-6. All but one of the guinea pigs of both strains responded to peptide P8 from ESAT-6 irrespective of their responses to ESAT-6. In addition, one animal of the Dunkin Hartley strain responded to P7. Only one Ssc:AL guinea pig did not respond to P8 or ESAT-6 (DTH reaction diameter of 5 mm). The surprising finding that animals which were nonresponsive to the intact protein recognized a derived peptide has a corollary in humans, in which peripheral blood lymphocytes (PBL) from several TB patients have responded in gamma interferon assays to peptides at the C terminus of ESAT-6, including P8, but not to intact ESAT-6 itself (15a). This result indicates that genetic variability within these two strains of guinea pigs (and perhaps humans) does not exclusively affect major histocompatibility complex specificity and may also operate at the level of antigen processing. It is possible that the P8 peptide does not need the same extensive processing as the intact molecule or, due to its small size, is more amenable to proteolytic cleavage. In addition, it was confirmed that the Dunkin Hartley guinea pigs, like the Ssc:AL strain, responded exclusively to peptide P12 covering residues 168 to 192 of the MPT64 molecule (data not shown). Neither strain of guinea pig responded to any of the ESAT-6 or MPT64 peptides when they were sensitized with BCG or M. avium (data not shown).

View larger version (15K): [in this window] [in a new window]   FIG. 2.   Responses of two different strains of outbred guinea pigs infected with M. tuberculosis to overlapping peptides derived from ESAT-6 (P1, amino acids [aa] 1 to 20; P2, aa 12 to 35; P3, aa 22 to 45; P4, aa 32 to 55; P5, aa 42 to 65; P6, aa 52 to 75; P7, aa 62 to 85; P8, aa 72 to 95). Eight outbred Ssc:AL () and six Dunkin Hartley () guinea pigs were infected by i.v. inoculation with M. tuberculosis H37Rv. Twenty-eight days later the animals were skin tested with 10 TU (0.2 µg) of PPD, 1 µg of rESAT-6, or 1 µg each of the ESAT-6 peptides P1 to P8.

The observation that not all animals infected with M. tuberculosis responded to the intact ESAT-6 molecule is in agreement with previous observations of responses to MPT64 (9). Therefore, we decided to examine the use of a combined skin test reagent containing rESAT-6 and rMPT64. Ssc:AL guinea pigs were infected i.v. with M. tuberculosis H37Rv for 28 days and were then injected with RT23, rESAT-6, rMPT64, and a combination of rESAT-6 and rMPT64 in equal quantities. In this experiment, 10 of 18 animals did not respond to rESAT-6, 4 of 18 animals did not respond to rMPT64, but all animals responded to the combination of rESAT-6 and rMPT64 (Fig. 3). Six Dunkin Hartley guinea pigs were then infected via aerosol with M. tuberculosis Erdman as described previously (9). At 5 weeks postinfection these animals responded to a combination of rESAT-6 and rMPT64 in a manner similar to that of animals infected i.v. (Fig. 4). All animals responded to the P8 peptide of ESAT-6 as well as the P12 peptide of MPT64, and the equal mixture of the two peptides resulted in reactions with diameters slightly larger than those to the MPT64 peptide alone. In addition, the response to the two peptides combined was less than that to the parent proteins in combination, a result which may reflect the increased sensitivity of small peptides to proteolytic degradation and epitope destruction.

View larger version (19K): [in this window] [in a new window]   FIG. 3.   A combination of rESAT-6 and rMPT64 identifies all animals infected with M. tuberculosis. Twenty-eight days following i.v. infection, Ssc:AL guinea pigs (n = 18) were tested with 10 TU (0.2 µg) of PPD, 1 µg of rESAT-6, 1 µg of rMPT64, or the same quantities of rESAT-6 and rMPT64 in combination.

View larger version (19K): [in this window] [in a new window]   FIG. 4.   Guinea pigs infected by aerosol recognize ESAT-6-MPT64 and the reactive peptides. Dunkin Hartley guinea pigs (n = 6) were infected by the respiratory route and tested for DTH responses to rESAT-6-rMPT64 and peptides from ESAT-6 and MPT64. The guinea pigs were skin tested 35 days following exposure to M. tuberculosis. At skin testing guinea pigs received 0.2 µg of PPD, 2 µg of an equal mixture (wt/wt) of ESAT-6 and MPT64, and 1 µg of the peptides ESAT P8, MPT64 P12, or the two peptides in combination. Reactions were read at 24 h after intradermal injection with the skin test reagents.

The potential of ESAT-6 in combination with MPT64 as a skin test reagent for humans as indicated by the studies described above is supported by recent clinical studies. Although two of these studies have provided markedly different estimates of the frequency of human responders to MPT64 (16, 20), data on the recognition of ESAT-6 by patients and their contacts are very encouraging. At least 80% of recently diagnosed TB patients in a study in Kuwait responded to ESAT-6 and produced high amounts of gamma interferon to this antigen (1). The ultimate diagnostic reagent will possibly require other M. tuberculosis-specific antigens to cover all members of a genetically diverse population. Antigens which have been suggested to have some potential include MPT63 (12) and the 38-kDa PhoS protein (2, 10), but both these proteins are expressed by BCG and so would be unable to discriminate patients from vaccinated individuals. In this regard, a recent analysis of the regions deleted from the BCG genome during its generation and subsequent propagation has identified open reading frames that may encode antigens with diagnostic potential (11). One of the deleted regions includes the putative bovine esat-6 gene, and a second region codes for mpt64. Characterization of these regions and cloning of antigens may yield reagents that could supplement the ESAT6-MPT64 combination in the future. Taken together, our results indicate that ESAT-6-MPT64 has potential as a highly specific reagent that should not interfere with screening and vaccination programs.