Activation of macrophages and interference with CD4⁺ T-cell stimulation by Mycobacterium avium subspecies paratuberculosis and Mycobacterium avium subspecies avium

Abstract

Mycobacterium avium subspecies paratuberculosis (M. ptb) and M. avium subspecies avium (M. avium) are closely related but exhibit significant differences in their interaction with the host immune system. The macrophage line, J774, was infected with M. ptb and M. avium and analysed for cytokine production and stimulatory capacity towards antigen-specific CD4⁺ T cells. Under all conditions J774 cells were activated to produce proinflammatory cytokines. No influence on the expression of major histocompatibility complex (MHC) class II, intracellular adhesion molecule-1 (ICAM-1), B7.1, B7.2 or CD40 was found. However, the antigen-specific stimulatory capacity of J774 cells for a CD4⁺ T-cell line was significantly inhibited after infection with M. ptb, but not with M. avium. When a T-cell hybridoma expressing a T-cell receptor identical to that of the T-cell line was used, this inhibition was not observed, suggesting that costimulation which is essential for the CD4⁺ T-cell line is influenced by the pathogenic bacterium M. ptb.

Introduction

Mycobacterial infections in humans and animals are among the most serious infectious diseases worldwide. Besides Mycobacterium tuberculosis, two members of the M. avium complex (MAC), namely M. avium subspecies paratuberculosis (M. ptb) and M. avium subspecies avium (M. avium), have re-emerged as menacing pathogens. M. ptb causes Johne's disease, a chronic, non-treatable enteritis in ruminants and one of the most widespread bacterial infections of domestic animals, which has enormous economic impact.^1–3 M. avium is the causative agent of avian tuberculosis and a major cause of opportunistic infections in immunocompromised humans, such as patients infected with human immunodeficiency virus (HIV).⁴ Despite close genetic and antigenic similarities⁵ the two subspecies show remarkable differences with respect to pathogenicity. This might be explained, in part, by differential capacities to modify macrophages (Mφ).⁶ However, specific efforts to prevent or control the diseases caused by MAC are still hindered by the lack of detailed knowledge on the mechanisms used by either of these pathogens to persist in such cells.

Strong evidence suggests that pathogenic mycobacteria share common mechanisms for entry and survival in Mφ. Schorey et al.⁷ showed that pathogenic mycobacteria, in contrast to apathogenic mycobacteria, were able to recruit the complement fragment C2a to form a C3 convertase. The mycobacterial-associated C2a cleavage resulted in opsonization of the mycobacteria via C3b and, in turn, in binding and invasion of Mφ. In addition, pathogenic mycobacteria have been shown to parasitize Mφ, at least in part, as a result of their ability to block the maturation of mycobacterial phagosomes into phagolysosomes.^8–12 Therefore, phagosomes containing M. tuberculosis or M. avium are not acidified. This may be caused by lack of the vacuolar proton ATPase required for acidification of such vesicles.^13,14 Therefore, it seems that maturation of the mycobacterial phagosome into a highly degradative compartment is retarded in such a way that it resembles a sorting endosome rather than a mature phagolysosome.^15–17

In tuberculosis, additional immunosuppressive effects are known which are attributed to a decreased ability of infected Mφ to activate T cells. This is a result of defective antigen presentation, down-regulation of major histocompatibility complex (MHC) class II and B7 surface molecules, as well as induction of a sustained secretion of interleukin (IL)-6.^18–21

In M. ptb infections, induction of cytokines such as IL-1β, IL-6, tumour necrosis factor-α (TNF-α) and granulocyte–macrophage colony-stimulating factor (GM–CSF) has also been documented²² (reviewed in refs ^{1, 2} and ³). As it is believed that the successful elimination of intracellular mycobacteria by the infected Mφ requires activated, interferon-γ (IFN-γ)-producing CD4⁺ T cells, as well as Mφ-derived IL-1, IL-6, TNF-α and GM–CSF,^23–26 it is not clear whether these cytokines facilitate elimination of M. ptb in a way similar to their roles in the elimination of other mycobacteria. In fact, one could speculate that M. ptb has developed specific immune-escape mechanisms, as indicated by a significant decrease in the IFN-γ mRNA level in lymphoid tissues of experimentally infected lambs²⁷ and by the fact that IFN-γ only moderately affected intracellular survival of M. ptb in bovine monocytes.²⁸ In addition, it has recently been demonstrated that M. ptb and M. avium had differential effects on bovine Mφ. Only M. ptb was able to cause a decrease in the expression of MHC class I and II molecules, while M. avium showed no effects.⁶ This supports the idea of specific modifications of Mφ induced by M. ptb.

Nevertheless, little is known about the molecular mechanisms of immune escape of MAC species. It seems clear that they are able to alter the Mφ in such a way that induction of specific T cells might be obscured. Therefore, in the present study we attempted to analyse and dissect the effects of M. ptb and M. avium on the Mφ using the well-established murine Mφ-like cell line, J774, as a model. Both types of bacteria induced expression of cytokines in these cells but did not alter the expression of several surface antigens involved in T-cell stimulation. Despite this, infection with M. ptb, but not by M. avium, resulted in inhibition of the antigen-specific stimulation of CD4⁺ T cells, probably by interference with the expression of an essential costimulatory molecule on the host cell surface.

Materials and methods

Bacterial strains and growth conditions

The field isolate M. ptb 6783 has been described previously.²⁹ M. avium strains DSM 44156 and DSM 44158 were obtained from the DSMZ (German Collection of Micro-organisms and Cell Cultures, Braunschweig, Germany). Mycobacteria were grown in Watson Reid liquid medium, which was supplemented with Mycobactin J (1 mg/l) for the growth of M. ptb. Bacteria were grown at 37° for 12 days (M. ptb) or 2 weeks (M. avium). For infection of Mφ, mycobacteria were harvested by centrifugation (10 min at 1000 g) and resuspended in 5 ml of antibiotic-free cell-culture medium. Homogenization of the mycobacteria was performed as described by Silver et al.¹⁰ Briefly, bacterial suspensions were vortexed in the presence of glass beads (3 mm diameter) for 5 min and centrifuged for 5 min at 50 g. Microscopy confirmed that the resulting supernatant contained mainly single bacteria. Viability of the bacteria was determined using the BacLight system, as recommended by the manufacturer (Molecular Probes, Eugene, OR). Bacterial suspensions from M. ptb usually contained between 70% and 80% viable bacteria, those from M. avium ≈ 95%. Heat inactivation of bacteria was achieved by incubating bacterial suspensions at an optical density (OD)₆₆₀ of 1·0 for 15 min at 85°, according to Zurbrick & Czubrynski.³⁰ Loss of viability was confirmed by serial platings on Middlebrooke 7H9 medium with supplements (Difco, Detroit, MI).

Cell lines

The murine Mφ line J774 (H-2^d) was used for infection experiments. This cell line expresses MHC class II molecules constitutively at the cell surface under our culture conditions. The T-cell hybridoma 16.2.11^31,32 was kindly provided by Dr K. Karjalainen (Institute for Immunology, Basel, Switzerland). This hybridoma is specific for an epitope comprising amino acids 110–120 of the haemagglutinin (HA) of influenza virus, PR8, presented by the MHC class II molecule I-E^d. The T-cell line, HA-TCL, was established from a mouse that was transgenic for the T-cell receptor of the T-cell hybridoma, 16.2.11.³³ Spleen cells from these mice were restimulated repeatedly with irradiated syngeneic spleen cells as antigen-presenting cells (APC) and with a synthetic peptide comprising amino acids 110–120 of the HA. All eukaryotic cells were grown in Iscove's modified Dulbeco's Medium (IMDM) supplemented with 10% fetal calf serum (FCS).

Infection of Mφ

J774 cells were plated at a density of 1·5 × 10⁶/well in a six-well plate 24 hr before infection. Cells were washed with phosphate-buffered saline (PBS), 1 ml of a bacterial suspension (OD₆₆₀ = 0·1) in IMDM was added and, after incubation for 2 hr, cells were washed again with PBS. Then, fresh IMDM containing 10% FCS was added, and infected J774 cells were incubated for 48 or 90 hr before testing the T-cell stimulatory capacity on cytokines in the cell supernatants. Infection efficiency was monitored by the use of microscopy. To achieve this, infected Mφ were fixed with 0·37% formaldehyde at the respective times and examined for intracellular acid-fast bacilli by the Ziehl–Neelsen procedure. Under these conditions, almost all cells were infected and the cell viability of such cultures was close to 100% at the time of testing.

Reverse transcription–polymerase chain reaction (RT–PCR)

Poly (A)+ RNA was prepared as described previously.³⁴ First-strand synthesis of cDNA was performed from 200 ng of poly (A)+ RNA with 200 U MMLV reverse transcriptase (Gibco BRL, Eggenstein, Germany), used according to the manufacturer's instructions. PCR was performed using Taq polymerase (Gibco, BRL), according to the manufacturer's instructions. After an initial denaturation at 91° for 3 min, 35 cycles were performed as follows: denaturation for 1 min at 95°, annealing for 1 min at 60°, and extension for 2 min at 72°. PCR for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and GM–CSF were performed with 25 and 40 cycles, respectively. Oligonucleotide primers for IL-6, GM–CSF and GAPDH were purchased from Clontech (Heidelberg, Germany) and those for TNF-α and IL-1β were from Stratagene (Heidelberg, Germany). The oligonucleotide primers for IL-10 (5′-AATTTGACATCTTCATCAACT-3′; 5′-TTTCTTCACAACTCTCTTAGG-3′) and IL-12p40 (5′-CATCACTGTCAAAGAGTTTCT-3′; 5′-CAGGACACTGAATACTTCTCA-3′) were synthesized by MWG-BioTech (Ebersberg, Germany). PCR products were fractionated on 1·5% agarose gels at 75 V for 2·5 hr, stained with ethidium bromide and documented using a gel-documentation system (Bio-Rad, Munich, Germany).

Stimulation of T cells

J774 cells that had been treated with either live or heat-killed mycobacteria, or incubated with IMDM alone (controls), were harvested using a rubber ‘policeman’. Cells were washed and irradiated with 10 000 rads and subsequently plated at a density of 2·5 × 10⁴ in combination with 10⁴ HA-TCL and either 10 µg/ml HA^110−120 peptide or 10 µg/ml HA-Ig protein.³⁵ Supernatants were harvested after 24 hr and tested for IFN-γ by enzyme-linked immunosorbent assay (ELISA) or for IL-2 and IL-3 by a bioassay using cell lines CTLL and FDC-1, respectively, as indicators. Alternatively, 5 × 10⁵ spleen cells, irradiated with 4000 rads, were used as APC. Similarly, 5 × 10⁴ 16.2.11 T-cell hybridomas were incubated with 2·5 × 10⁴ infected or control J774 in combination with 10 µg/ml HA^110−120 peptide. After 24 hr, production of IL-2 was estimated in the bioassay.

ELISA analysis of cytokines

IL-6, IL-12, IFN-γ, TNF-α and GM–CSF were estimated in standard ELISAs using the following antibodies: IL-6: MP5–20F3/MP5–32C11-bio; IL-12p40/p70: C15.6/C17.8-bio; GM–CSF: MP1–22E9/MP1–31G6-bio; TNF-α: G281-2626/MP6-XT3-bio; IFN-γ: AN-18.74.24/R4–6A2-bio. Assays were developed using streptavidin-conjugated horseradish peroxidase. All reagents, except those used for determination of IFN-γ, were obtained from PharMingen (San Diego, CA).

Flow cytometry

The following antibodies were used: I-E^d, 14.4-4S–bio (ATCC HB32); intracellular adhesion molecule-1 (ICAM-1), YN1/1-7–fluorescein isothiocyanate (FITC); B7.1, 16-10A1–FITC; B7.2, GL1–phycoerythrin (PE); and CD40, FGK45.5–bio (kindly supplied by Dr J. Andersson, Basel). Antibodies for B7.1, B7.2 and CD40, and streptavidin–PE conjugate were obtained from PharMingen. Analysis was performed on a FACSCalibur using Cellquest software (Becton-Dickinson, Heidelberg, Germany).

Results

Interaction of bacteria with APC should result in various signals dependent on the particular micro-organism and the type of APC it interacts with. Therefore, we first wanted to examine whether M. ptb, or two strains of M. avium, were able to induce cytokines that are typical for activated Mφ. Viable or heat-killed bacteria were added to J774 cells and tested for phagocytosis and survival. All bacteria, viable or heat killed, were rapidly taken up and persisted for several days without harming the J774 cells (Fig. 1a). Almost all cells contained similar amounts of intracellular mycobacteria, irrespective of treatment with viable or heat-killed bacteria.

Figure 1.

Induction of cytokines in J774 cells after incubation with live or heat-killed Mycobacterium avium subspecies paratuberculosis (M. ptb) and Mycobacterium avium subspecies avium (M. avium). (a) Detection of intracellular bacteria by light microscopy 2 days (2d) or 5 days (5d) after addition of the micro-organisms to cultures. Cell smears were stained using the Ziehl–Neelsen procedure. (b) Induction of proinflammatory cytokine mRNA by M. ptb and M. avium. Poly (A)+ RNA was isolated from uninfected J774 macrophages (Mφ)¹ and from J774 Mφ following infection with M. ptb strain 6783,² M. avium strain 44156,³ or 44158,⁴ for 24 hr. Reverse transcription–polymerase chain reaction (RT–PCR) for the indicated genes was performed with 200 ng of poly (A)+ RNA. PCR products were resolved on a 1·5% agarose gel and stained with ethidium bromide. The specific PCR product size is given on the left. DNA marker fragments (M) were a 100-bp ladder. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; IL, interleukin; TNF-α, tumour necrosis factor-α; (c) Production of cytokines induced by M. ptb and M. avium. Supernatants derived from cultures 2 days after initiation were tested for the indicated cytokines by using enzyme-linked immunosorbent assay (ELISA). Identical results were obtained on day 5. To allow direct comparison, samples of one experiment were measured on the same ELISA plate. Data represent the mean of duplicates. M. ptb, M. paratuberculosis; M. ptb hk, heat-killed M. ptb; M.a. 44156, M. avium strain 44156; M.a. 44156 hk, heat-killed M. avium strain 44156.

To investigate whether cytokines were induced, J774a cells were tested (by RT–PCR), 24 hr after the addition of bacteria to the cells, for the presence of cytokine mRNA. Induction of IL-1β, IL-6, TNF-α and low levels of IL-12 and GM–CSF were observed, while IL-10 mRNA seemed to be down-regulated (Fig. 1b). This was corroborated by adding viable or heat-killed bacteria to the J774 cultures and testing the culture supernatants 2 and 5 days later (by ELISA) for the presence of IL-1β, IL-6, TNF-α, GM–CSF and IL-12. Only the induction of IL-6, TNF-α and, to some extent, IL-12 could be confirmed (Fig. 1c). The lack of detection of the other cytokines might be a result of the lower sensitivity of the assay used.

Obviously, these bacteria interacted strongly with J774 Mφ. Therefore, we also wanted to investigate whether cell-surface molecules involved in T-cell stimulation were altered by the mycobacterial infection. Often, MHC class II or costimulatory molecules are up-regulated under such circumstances. However, under our culture conditions, J774 cells already expressed some of these molecules, i.e. MHC class II, B7.1, B7.2, ICAM-1 (Fig. 2) and CD40 (data not shown). No obvious change in the surface expression of these molecules was observed, regardless of whether they were tested 2 or 5 days after addition of the micro-organism (Fig. 2 and data not shown).

Figure 2.

Flow cytometric analysis of cell-surface markers on J774 cells after incubation with live Mycobacterium avium subspecies paratuberculosis (M. ptb), Mycobacterium avium subspecies avium (M. avium) or heat-killed M. ptb (M. ptb hk). Displayed are live gated cells derived from cultures 2 days after infection. Identical results were obtained with cultures from day 5 of infection. ICAM-1, intracellular adhesion molecule-1.

As some species of mycobacteria are known to interfere with antigen presentation, we incubated J774 cells with bacteria, as described above, and measured the T-cell stimulatory capacity on day 2 or day 5 after treatment by adding a CD4⁺ T-cell line specific for the epitope 110–120 of the HA of influenza virus P8. Either protein or synthetic peptide was added to these cultures as antigen. Figures 3 and 4 show that M. avium and heat-killed M. ptb had no influence on the induction of IL-2, IFN-γ (Fig. 3), or IL-3 (Fig. 4) by the antigen-specific CD4⁺ T-cell line, regardless of whether tests were performed 2 days (Figs 3 and 4) or 5 days (data not shown) after treatment was started. In contrast, J774 cells infected with M. ptb were unable to stimulate production of IL-2 and IFN-γ in the CD4⁺ T-cell line, regardless of whether protein or peptide was used as the antigen. Results testing the antigen-specific induction of IL-3 in these T cells were not as clear as for IL-2 or IFN-γ. As shown in Fig. 4, in one experiment, induction of IL-3 was completely eliminated, while in the second experiment IL-3 production was only reduced. IL-2 and IFN-γ production had been completely inhibited in this second experiment (data not shown). Therefore, induction of IL-3 in these T-cell lines might be less prone to the inhibitory mechanism of M. ptb. This is consistent with the finding that induction of IL-3 in CD4⁺ T cells requires weaker signals then induction of IL-2.³⁶

Figure 3.

Infection of J774 cells with Mycobacterium avium subspecies paratuberculosis (M. ptb) inhibits stimulation of antigen-specific CD4⁺ T cells. J774 cells were treated as described in the legend to Fig. 1. After 2 days, T cells specific for the epitope comprising amino acids 110–120 of the haemagglutinin (HA) of influenza virus, PR8, (HA^110−120) were added in combination with either 10 µg of protein antigen or 10 µg of synthetic peptide. Supernatants were harvested after 24 hr and assessed for interleukin-2 (IL-2) in a CTLL bioassay or for interferon-γ (IFN-γ) by enzyme-linked immunosorbent assay (ELISA). Similar experiments were performed on day 5, with identical results obtained. Data represent the mean of duplicate samples. The experiments were carried out three times, with identical results obtained on each occasion. A, absorbance; c.p.m., counts per minute.

Figure 4.

Infection of J774 cells with live Mycobacterium avium subspecies paratuberculosis (M. ptb) inhibits stimulation of interleukin-3 (IL-3) production by antigen-specific CD4⁺ T cells. Control indicates stimulation with uninfected J774. Cultures were set up as described in the legend to Fig. 3. The results of two different experiments are shown, one in which IL-3 production was completely inhibited (Exp. 1) and one in which IL-3 production was only reduced (Exp. 2). IL-3 production was assessed using proliferation of the cell line FDC-1 as indicator. In both experiments, interleukin-2 (IL-2) and interferon-γ (IFN-γ) production by the T cells were completely inhibited when J774 cells were infected with live M. ptb (results not shown). c.p.m., counts per minute.

There are several possibilities for explaining this inhibitory effect on J774 cells by M. ptb; for example, the Mφ line could produce a soluble inhibitory factor after infection with live M. ptb. Therefore, we isolated supernatants of J774 cells that were infected with M. ptb and which had been shown to be unable to stimulate the antigen-specific CD4⁺ T cells. When such supernatants were added to the T-cell assays, no inhibition of IL-2 or IL-3 production was observed, even when up to 25% conditioned medium was used (Fig. 5). As expected, no inhibition was observed when supernatants from J774 cells infected with M. avium or treated with heat-killed M. ptb were added to the T-cell cultures.

Figure 5.

Supernatants from J774 cells incubated with live Mycobacterium avium subspecies paratuberculosis (M. ptb), Mycobacterium avium subspecies avium (M. avium) or heat-killed M. ptb (M. ptb hk) do not affect the antigen-specific induction of CD4⁺ T cells. J774 cells were incubated as described in the legend to Fig. 1 and supernatants were harvested on day 2 or day 5. Parallel cultures were tested for the stimulatory capacity of these cells, as described in the legend to Fig. 3. Graded volumes of supernatants were added to cultures of the CD4⁺ T-cell line specific for the epitope comprising amino acids 110–120 of the haemagglutinin (HA) of influenza virus, PR8, (HA^110−120), with synthetic peptide as the antigen and normal irradiated spleen cells as antigen-presenting cells (APC). Interleukin (IL)-2 and IL-3 were assessed by measuring the proliferation of indicator lines CTLL and FDC-1, respectively. c.p.m., counts per minute.

MHC class II expression on J774 infected with M. ptb was unaltered, as shown by flow cytometry (Fig. 2). On the other hand, even T-cell stimulation using synthetic peptides that should be loaded on the outside of an APC was strongly reduced. Thus, it was possible that a costimulatory molecule of J774 cells that we did not test for could be influenced by the M. ptb infection. In general, T-cell hybridomas are less dependent on costimulation than T-cell lines or clones. In fact, T-cell hybrids can react to isolated MHC class II molecules on planary membranes.³⁷ We therefore treated J774 cells with mycobacteria as described above. A T-cell hybridoma with a T-cell receptor identical to that of the above T-cell line was added in combination with either the protein antigen or the synthetic peptide, and tested for activation. As shown in Fig. 6, J774 cells infected with M. ptb were able to stimulate the antigen-specific secretion of IL-2 by the T-cell hybridoma, similar to uninfected J774 Mφ. The antigen-specific T-cell line was used as a control in parallel cultures to assure that infection with live M. ptb had indeed resulted in complete inhibition of stimulation of the T-cell line (data not shown). Thus, the infection did not alter the presentation of MHC class II-dependent antigens, but rather affected the expression of an accessory molecule at the surface of the APC that is not essential for activation of the T-cell hybridoma, but which is required for the activation of T-cell lines.

Figure 6.

Stimulation of an antigen-specific CD4⁺ T-cell hybridoma by J774 infected with Mycobacterium avium subspecies paratuberculosis (M. ptb) and Mycobacterium avium subspecies avium (M. avium). J774 was treated as described in the legend to Fig. 3, and on day 2 or 5, the T-cell hybridoma, 16.2.11, expressing an identical receptor to the T-cell line used in the experiments described in Figs 3–5, was added to the culture with either protein antigen or synthetic peptide. After overnight incubation, supernatants were tested for interleukin-2 (IL-2) in the CTLL bioassay. Only results obtained on day 2 are displayed. Identical results were obtained with cultures from day 5 of infection. These data also indicate that no inhibitory factor for the IL-2 bioassay is produced by J774 cells infected with M. ptb. Data represent means of duplicate experiments. Experiments were carried out three times, with identical results obtained on each occasion. c.p.m., counts per minute.

Discussion

Our data on the induction of cytokines by M. ptb and M. avium after infection of J774 are in agreement with other findings where proinflammatory cytokines have been induced after infection of phagocytic cells with bacteria.³⁸ Interestingly, however, there seemed to be no difference between infection of J774 with viable M. ptb and avium or treatment with heat-killed bacteria. Thus, the bacterial components that suffice to induce the required signals were not denaturable by the heat treatment and therefore were probably not proteins.

No change in expression of the surface molecules tested on J774 cells was found in our experiments. Often an increase of surface expression of molecules such as MHC class II and ICAM-1 is observed when Mφ or dendritic cells are infected with bacteria.³⁹ However, under our growth conditions such molecules might already be optimally expressed and, hence, not permit any further enhancement. In agreement with this is the finding that J774 can also be negative for MHC class II molecules and might require the presence of IFN-γ to be able to present antigen via MHC class II (A. Darji, unpublished).

Despite the unchanged expression of the surface molecules tested to date, stimulation of CD4⁺ T cells was inhibited after infection of J774 with viable M. ptb. Inhibition of stimulation of MHC II-restricted CD4⁺ T cells by Mφ infected with mycobacteria has been observed previously^18,23 and some effects have been characterized in detail.^19,20,40 Differences between various pathogenic mycobacterial species have been reported.⁶ Here, we demonstrated a novel specific modification of the Mφ–T cell interactions by M. ptb, which differ from those caused by other mycobacteria of the MAC, i.e. M. avium.

The lack of antigen-specific stimulation of CD4⁺ T cells by M. ptb-infected J774 cells could be explained most straightforwardly by a reduction in either antigen uptake or antigen processing by these infected cells. Alternatively, the cellular traffic of MHC class II molecules could have been impeded, resulting in a down-regulation of cell-surface MHC class II molecules, or at least in blockage of the export of newly formed MHC class II–peptide complexes. Such phenomena have been observed for other micro-organisms, including mycobacteria.^19,40 This would be consistent with the recently described modulation of phagosomal maturation and acidification by M. ptb and M. avium,⁴¹ which could interfere with the antigen-presentation mechanism. However, none of these possibilities is applicable to J774 cells infected with M. ptb, as the inhibition is seen only when using a T-cell line, but not when a T-cell hybridoma expressing an identical T-cell receptor is used. Thus, MHC class II–peptide complexes appear to be unaltered in quantity and quality on the surface of J774 infected with M. ptb. This is observed regardless of whether they are derived from loading synthetic epitopes on the outside of the APC or from processing and loading via the intracellular machinery after uptake of the native protein antigen.

A change in the adhesion properties of the presenting cells, as has been described for T. cruzi-infected APC,⁴² can be excluded for the same reason, namely that the T-cell line – but not the T-cell hybridoma – is affected. Therefore, the most probable explanation for the lack of stimulation of CD4⁺ T cells by M. ptb-infected J774 is a change in a costimulatory molecule. T-cell hybridomas are known to be less dependent on costimulation compared with T-cell lines or clones. One of those costimulatory molecules, B7.1, has been shown to be down-regulated in M. tuberculosis-infected Mφ, in addition to down-regulation of MHC class II molecules.²⁰ However, cell-surface expression of B7.1 and other costimulatory molecules were found to be unaltered on J774 cells infected with M. ptb, as shown by flow cytometry analyses. Nevertheless, a large number of novel costimulatory molecules have been recently described⁴³ and it is highly probable that one of these, or another yet-unidentified costimulatory molecule, is down-regulated by the M. ptb infection. Alternatively, an alteration in the post-translational modification of such molecules could cause the block of antigen-specific CD4⁺ T-cell stimulation. Hypoglycosylated B7.2 has been shown to lack costimulatory properties for CD4⁺ T-cell clones.⁴⁴

Inhibitory effects of M. ptb on T-cell activation were clearly connected with bacterial viability. This is in agreement with published observations that indicate a role of viable mycobacteria in intracellular trafficking and immunosuppression.^14,45 It is also supported by our recent finding that inhibition of phagosomal acidification by M. ptb depends on bacterial viability.⁴¹ However, there are conflicting results for M. tuberculosis, where inhibition of the MHC II processing pathway seemed not to depend on the viability of the bacteria.¹⁹

How do our results using the established Mφ-like cell line, J774, reflect the situation in primary murine Mφ? Preliminary data suggest that in the latter cells the effect is even more pronounced. MHC class II molecules appeared to be completely down-regulated (A. Darji et al. unpublished). Similarly, a decrease of MHC class II has been described for primary bovine Mφ infected with M. ptb.⁶ Therefore, the use of the J774 cell line in which only a costimulatory molecule appeared to be affected was advantageous in that it allowed us to dissect a complex interference of a bacterial pathogen with the antigen-specific stimulation of CD4⁺ T cells. This adds to the suitability of J774 cells as a model system for using to analyse molecular events during the interaction of Mφ with intracellular bacteria, including mycobacteria, which has been documented in numerous studies.^46–48

Of particular interest is the difference between the two closely related species of MAC. Both species induce activation of J774, as judged by the induction of cytokine production. However, only M. ptb (not M. avium) is also able to interfere with some of the other immune functions of this cell line, i.e. stimulation of CD4⁺ T cells. This is consistent with recent results on differential immune modulatory effects of these species in bovine Mφ.⁶ Thus, further characterization of these functions and the underlying genomic differences between the two species of MAC will surely contribute to our understanding of the evolution of pathogenicity.

Abbreviations

APC

antigen-presenting cell

GM–CSF

granulocyte–macrophage colony-stimulating factor

HA

haemagglutinin

IFN-γ

interferon-γ

IL

interleukin

Mφ

macrophage

MAC

Mycobacterium avium complex

M. ptb

Mycobacterium paratuberculosis

RT–PCR

reverse transcription–polymerase chain reaction

TNF-α

tumour necrosis factor-α