SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Mannose-capped lipoarabinomannan (ManLAM) is considered an important virulence factor of Mycobacterium tuberculosis. However, while mannose caps have been reported to be responsible for various immunosuppressive activities of ManLAMobserved in vitro, there is conflicting evidence about their contribution to mycobacterial virulence in vivo. Therefore, we used Mycobacterium bovis BCG and M. tuberculosis mutants that lack the mannose cap of LAM to assess the role of ManLAM in the interaction of mycobacteria with the host cells, to evaluate vaccine-induced protection and to determine its importance in M. tuberculosis virulence. Deletion of the mannose cap did not affect BCG survival and replication in macrophages, although the capless mutant induced a somewhat higher production of TNF. In dendritic cells, the capless mutant was able to induce the upregulation of co-stimulatory molecules and the only difference we detected was the secretion of slightly higher amounts of IL-10 as compared to the wild type strain. In mice, capless BCG survived equally well and induced an immune response similar to the parental strain. Furthermore, the efficacy of vaccination against a M. tuberculosis challenge in low-dose aerosol infection models in mice and guinea pigs was not affected by the absence of the mannose caps in the BCG. Finally, the lack of the mannose cap in M. tuberculosis did not affect its virulence in mice nor its interaction with macrophages in vitro. Thus, these results do not support a major role for the mannose caps of LAM in determining mycobacterial virulence and immunogenicity in vivo in experimental animal models of infection, possibly because of redundancy of function.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The cell envelope of Mycobacterium tuberculosis is considered a major determinant of virulence in this global pathogen. A major component of the cell envelope of all mycobacteria is lipoarabinomannan (LAM) (Chatterjee and Khoo, 1998; Nigou et al., 2003; Briken et al., 2004; Gilleron et al., 2008), which appears to play a key role in the interaction with the host, and modulation by the bacterium of the host response (Nigou et al., 2002; 2003; Briken et al., 2004; Gilleron et al., 2008; Mishra et al., 2011). This complex molecule of approximately 17 kDa is the largest member of a series of lipoglycans of varying size based on a conserved mannosyl-phosphatidyl-myo-inositol anchor. In slow-growing mycobacteria, such as M. tuberculosis, the presence of one to three mannopyranosyl residues linked to the non-reducing ends of the arabinan domain constitutes the mannose cap of LAM (ManLAM), while in fast-growing mycobacteria, such as Mycobacterium smegmatis, this latter domain is capped with phospho-inositol residues (PILAM) (Nigou et al., 2003) and in some species like Mycobacterium chelonae, no such capping motif is present (Guerardel et al., 2002). The proportion of LAM non-reducing termini that are capped with mannose varies among different species of slow-growing mycobacteria and among strains of M. tuberculosis, with fully virulent M. tuberculosis laboratory strains having up to 70% capping (Chatterjee et al., 1992; Khoo et al., 1995). The number of mannose residues per cap is also variable even within LAM from the same species or strain (Nigou et al., 2003).

A large number of studies have assigned a role in virulence to LAM. Initial studies aiming at determining the role of LAM in mycobacterial virulence addressed the in vitro effects of the purified molecules. These studies showed that LAM from M. tuberculosis is able to alter macrophage functions associated with protective immunity (Sibley et al., 1988; Chan et al., 1991; Fratti et al., 2003; Kang et al., 2005; Pathak et al., 2005). It was additionally shown that LAM from mycobacteria of different virulences have different immunomodulatory activities (Chatterjee et al., 1992; Adams et al., 1993; Bradbury and Moreno, 1993; Roach et al., 1993; Yoshida and Koide, 1997; Knutson et al., 1998; Vergne et al., 2003). Numerous studies have so far suggested a role of ManLAM in binding to and in modulating the function of macrophages and dendritic cells (DCs) (Sibley et al., 1988; Chan et al., 1991; Schlesinger, 1993; Schlesinger et al., 1994; Nigou et al., 2001; Fratti et al., 2003; Geijtenbeek et al., 2003; Maeda et al., 2003; Tailleux et al., 2003; Vergne et al., 2003; Kang et al., 2005; Torrelles et al., 2006; Welin et al., 2008).

The studies above have led to the conclusion that ManLAM is an important virulence factor in tuberculosis, and that mannose capping plays an essential role. However, interpretation of these experiments is complex, with the experiments comparing ManLAM from fully virulent M. tuberculosis either with ManLAM from strains with a lower proportion of mannose capping, or with PILAM (Khoo et al., 1995). It is therefore important to carry out experiments directly comparing LAM produced by isogenic strains differing only in their terminal mannosyl decoration.

Enzymatic removal of mannose residues with α-mannosidase revealed that the inhibition of IL-12 secretion by human DCs caused by BCG ManLAM is strictly dependent on an intact molecule (Nigou et al., 2001). However, to unambiguously study the role of mannose capping of LAM in an infectious setting, we have been using genetically engineered mutants in which the cap is not added. A mannosyltransferase encoded by the M. tuberculosis capA gene (rv1635c) is responsible for the addition of the first mannose residue of the mannose cap in an α(1[RIGHTWARDS ARROW]5) linkage (Dinadayala et al., 2006). Previously, we identified homologous enzymes in Mycobacterium marinum and Mycobacterium bovis BCG (Appelmelk et al., 2008). Disruption of the capA gene resulted in bacteria deficient in the biosynthesis of the mannose cap of LAM (Dinadayala et al., 2006; Appelmelk et al., 2008).

When we tested capless mutants of M. marinum and M. bovis BCG in vitro and in vivo (Appelmelk et al., 2008), we obtained the surprising result that there was no evidence for an altered host–pathogen interaction, and that capless M. marinum and BCG mutants were not less virulent than their respective parent strains. Thus, the data obtained with live isogenic strains were discrepant with the mass of data obtained earlier with purified ManLAM.

Here we follow up on our previous study of the role of the mannose cap further using isogenic strains differing only at the capA locus: we (i) look at the role of the mannose cap in the protective efficacy of BCG in vaccination studies, and (ii) extend our studies on virulence to look at M. tuberculosis itself.

Mycobacterium bovis BCG, the vaccine for tuberculosis (TB) protects well against disseminated TB, including meningitis in childhood, but protection against adult pulmonary TB is highly variable, ranging from 80% to no protection at all (reviewed by Colditz et al., 1994), and a more effective vaccine is therefore required. As ManLAM has been reported as blocking phagolysosome fusion, and/or inducing IL-10, and LAM lacking mannose caps does not exhibit these activities, we hypothesized both that a capless mycobacterium would induce a different type of immune response, being less immunosuppressive and thus more protective. We have therefore tested a capless BCG as a vaccine both in a low-dose challenge model in mice, in contrast to the high-dose challenge model used by Festjens et al. (2011) and in the guinea pig model.

To date, in vivo virulence studies have been carried out with capless mutants of BCG and M. marinum, but not M. tuberculosis itself. Here we test the virulence of capless M. tuberculosis in a low-dose aerosol model.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We first studied a capless mutant of BCG. This strain has already been used in an earlier publication where it was fully characterized and shown to synthesize LAM devoid of mannose caps (see Fig. S4 from Appelmelk et al., 2008). Here we extended the analysis of its interaction with the host cells initially in in vitro assays using cultured macrophages and DCs and subsequently in in vivo experiments.

Response of mouse macrophages to wild type BCG and capless BCG mutant

Macrophages are in the first line of cellular defence against mycobacterial infection. However, as intracellular pathogens, mycobacteria have evolved strategies to manipulate the host cell mechanisms responsible for their killing. ManLAM inhibits the maturation of phagosomes (Fratti et al., 2003; Vergne et al., 2003) and prevents macrophage activation by IFNγ (Sibley et al., 1988; Chan et al., 1991). To assess the importance of the mannose caps in BCG survival and macrophage activation, we infected bone marrow-derived macrophages (BMDM) from Balb/c mice with either the parental BCG strain or the capless mutant strain and measured mycobacterial growth and the production of TNF and reactive nitrogen species by the macrophages. No differences in phagocytosis were observed between the mutant and parental BCG, with both BCG strains being internalized to the same extent following exposure of the macrophage monolayers to similar multiplicity of infection (moi = 2) leading to similar colony-forming unit (CFU) counts at time 0 of infection, i.e. after 4 h of contact of the macrophages with the inocula (Fig. 1A). Both strains replicated in macrophages with similar growth rates (Fig. 1A). Activation of the macrophages with IFNγ or IFNγ plus TNF caused macrophages to kill BCG, but again both BCG strains behaved in the same manner (Fig. 1B). Despite the same degree of replication in resting macrophages and the similar bacterial loads found at the end of the experiment, the mutant induced higher amounts of TNF secretion at day 7 of infection (Fig. 1C). The production of reactive nitrogen species was also assessed, with both BCG strains inducing the same amount (Fig. 1D). These data show that the lack of mannose caps in LAM leads to different signalling in infected macrophages but has no consequences on the ability of BCG to proliferate or survive intracellularly.

figure

Figure 1. In vitro interaction of BCG with macrophages.

A. Growth of wild type (WT) or capless (CapA) mutant BCG in bone marrow-derived macrophages (BMDM) from Balb/c mice.

B. Survival of WT and capless (CapA) mutant BCG in untreated BMDM or BMDM treated with 50 U TNF, 100 U IFNγ or both (results expressed as ‘log CFU increase’, that corresponds to the difference of growth, in log CFU, between day 7 and day 0).

C. Secretion of TNF into the culture medium by BMDM infected with the BCG strains.

D. Release of nitrite into the culture supernatants by BMDM infected with BCG and treated with IFNγ. Data represent the mean ± 1 SD of a representative experiment out of a total of three experiments. Statistically significant differences are labelled with an asterisk.

Download figure to PowerPoint

Response of mouse dendritic cells to wild type BCG and capless BCG mutant

Dendritic cells are key players in the induction of cellular immune responses against mycobacteria (Murray, 1999). It has been shown that ManLAM (but not non-capped LAM) prevents human DC maturation, inhibits the production of the pro-inflammatory cytokines IL-12 and TNF (Nigou et al., 2001; 2002) and triggers the production of the immunosuppressive cytokine IL-10 in lipopolysaccharide (LPS)-activated DCs (Geijtenbeek et al., 2003). To evaluate the effect of the mannose capping of LAM in the activation and modulation of DC function, we infected bone marrow-derived dendritic cells (BMDC) from Balb/c mice with the parental BCG or the mutant strain and measured the induction of co-stimulatory molecules and the production of cytokines released into the culture supernatants. Both wild type and mutant BCG activated DCs and induced the same extent of upregulation of co-stimulatory molecules (MHC-II, CD40 and CD86) (Fig. 2A). They also induced the secretion of the same amount of IL12p70 and TNF but, interestingly, the mutant strain induced higher amounts of IL-10 (Fig. 2B). Thus, both in macrophages and DCs, mannose capping of LAM affects cytokine induction.

figure

Figure 2. In vitro interaction of BCG with dendritic cells.

A. Expression of co-stimulatory molecules in bone marrow-derived dendritic cells infected for 24 h with wild type (WT) or capless (CapA) mutant BCG as evaluated by flow cytometry.

B. Secretion of IL-12p70, IL-10 and TNF by dendritic cells infected with WT or capless (CapA) mutant BCG. Data represent the mean ± 1 SD of a representative experiment out of a total of three experiments. Statistically significant differences are labelled with an asterisk.

Download figure to PowerPoint

Capless BCG mutant replicates and induces identical immune responses in vivo as compared to the wild type strain

As described above, ManLAM is proposed to be a key molecule in mycobacterial virulence (reviewed by Nigou et al., 2003; Briken et al., 2004) and the mannose capping of LAM may be related to an immunosuppressive activity of this lipoglycan impacting on vaccine efficacy of BCG. We therefore wished to determine the protective efficacy of the capless BCG strain.

Before testing the protective abilities of the parent BCG and its capless mutant, we compared their replication in vivo, and the immune response they induced. In order to compare the growth of wild type and capless BCG, Balb/c mice were intravenously infected with 5 × 104 CFU of either the parental strain or the capless mutant and groups of five mice were sacrificed at days 1, 10, 20, 30 and 60 post infection. No differences in mycobacterial loads were observed for the two strains in either the spleen or the liver (Fig. 3A). In the spleen, both BCG strains proliferated until around day 20, with mice reducing the bacterial load afterwards, while in the liver, the growth of the parent and mutant strains was slowly controlled after day 10, with bacterial numbers decreasing over time, as has been classically described (Gheorghiu et al., 1985).

figure

Figure 3. Capless M. bovis BCG replicates similarly and induces identical immune responses as compared to the wild type (WT) strain.

A. Proliferation of WT or capless (CapA) mutant BCG in the liver and spleen of BALB/c mice intravenously infected with 5 × 104 CFUs.

B. Secretion of IFNγ, TNF and IL-10 into the culture supernatants by splenocytes from the infected animals following in vitro re-stimulation with 4 μg ml−1 M. bovis BCG extract for 72 h. A total of five to seven mice per time point were used, and all results are representative of at least two independent experiments.

Download figure to PowerPoint

We next took spleen cells from infected mice, and stimulated them in vitro with BCG antigens to assess cytokine responses. Cells from capless mutant-infected animals released the same amount of IFNγ, TNF and IL-10 as cells from mice infected with the wild type strain (Fig. 3B). No differences were found in the number of several different immune cell populations (Table 1).

Table 1. M. bovis BCG CapA mutant induces identical immune responses as compared to the wild type (WT) strain
DaysAverage cells × 107 ± SD
CD4+CD8+CD4+Foxp3+CD19+CD3+DX5+DX5+CD11b+
WTCapAWTCapAWTCapAWTCapAWTCapAWTCapAWTCapA
  1. Balb/c mice were intravenously infected with 5 × 104 CFUs of M. bovis BCG WT or CapA mutant. The animals were sacrificed at various time points and spleen cells were labelled with specific antibodies for flow cytometric analysis of the splenic cell population. A total of five to seven mice per time point were used, and all results are representative of at least two independent experiments.

101.2 ± 0.21.2 ± 0.20.6 ± 0.10.6 ± 0.10.2 ± 0.020.2 ± 0.022.4 ± 0.62.1 ± 0.90.1 ± 0.010.1 ± 0.020.2 ± 0.080.2 ± 0.070.3 ± 0.10.3 ± 0.1
201.5 ± 0.31.5 ± 0.40.7 ± 0.10.6 ± 0.10.2 ± 0.030.2 ± 0.053.6 ± 0.73.6 ± 0.90.1 ± 0.020.1 ± 0.030.3 ± 0.080.3 ± 0.090.6 ± 0.20.6 ± 0.2
301.8 ± 0.41.5 ± 0.10.6 ± 0.20.5 ± 0.10.3 ± 0.060.2 ± 0.033.9 ± 0.63.3 ± 0.70.1 ± 0.030.1 ± 0.020.4 ± 0.040.4 ± 0.070.7 ± 0.20.6 ± 0.1
601.6 ± 0.21.4 ± 0.30.6 ± 0.10.5 ± 0.10.2 ± 0.030.2 ± 0.034.1 ± 0.63.4 ± 0.50.1 ± 0.040.1 ± 0.030.3 ± 0.070.3 ± 0.030.8 ± 0.10.6 ± 0.1

These data do not substantiate our hypothesis that a capless BCG would induce a different type of immune response.

Wild type and capless BCG induce the same level of protection against a M. tuberculosis challenge in a murine low-dose aerosol infection model

We proceeded to carry out a protection study comparing BCG and its capless mutant. Festjens et al. (2011) used a high-dose (5 × 104 CFU intravenous or 2 × 105 CFU intratracheal) virulent M. tuberculosis challenge to assess the protection afforded by wild type or capless BCG. We reasoned that this high dose might overwhelm host immunity and hence mask protective efficacy, and therefore used a low-dose exposure aerosol challenge with virulent M. tuberculosis to compare the protection afforded by either wild type BCG or its capless mutant.

Balb/c mice were intradermally inoculated with 5 × 104 CFU of the parental or the mutant strain and the dissemination of BCG to the lung and spleen was determined 70 days post vaccination. Very small numbers of bacteria were detected in the spleen and no CFUs could be detected in the lung of both vaccinated groups (Fig. 4A), indicating a poor dissemination of BCG after intradermal inoculation. We then determined if a specific immune response could be observed in the lung of these vaccinated mice. Seventy days post vaccination, leucocytes isolated from the lungs of both immunized groups produced the same amounts of the protective cytokines IFNγ, TNF and IL-17 after in vitro re-stimulation with BCG antigens (Fig. 4B). Furthermore, both BCG strains induced the same number of CD4+IFNγ+-producing cells (Fig. 4C), showing that both the parental strain and the capless mutant induce a similar type of immune response in the lung. No differences in cytokine production by spleen cells from either group were detected (data not shown).

figure

Figure 4. Capless and wild type (WT) M. bovis BCG induce the same level of protection to a M. tuberculosis challenge in the low-dose aerosol mouse model.

A. Dissemination of WT and capless (CapA) mutant BCG, after subcutaneous immunization with 5 × 104 CFU. The numbers of CFU in the spleen and lung were assessed 70 days after immunization.

B. Quantification of cytokine responses in the lungs of vaccinated mice. Seventy days post immunization, lung cell suspensions were re-stimulated in vitro with 4 μg ml−1 BCG extract for 72 h and cytokines measured in the supernatants.

C. Number of CD4+IFNγ+ cells in the lungs of vaccinated mice at day 70 post immunization, determined by intracellular cytokine staining of lung cell suspensions re-stimulated in vitro with PMA and ionomycin.

D. Protective efficacy of WT versus capless (CapA) mutant BCG in a M. tuberculosis challenge. Balb/c mice were immunized with 5 × 104 CFU of parental or mutant CapA BCG. Two months later, mice were aerogenically infected with approximately 100 CFU of M. tuberculosis H37Rv. Mice were sacrificed at days 1, 30 and 90 post infection and the number of bacteria in lungs and spleens determined. In all experiments a total of five to seven mice per time point were used and all results are representative of at least two independent experiments. Significant values are labelled by an asterisk.

Download figure to PowerPoint

Seventy days post vaccination, the two groups of immunized mice and a control group of non-immune animals were subjected to a low-dose aerosol challenge with a virulent strain M. tuberculosis (strain H37Rv) leading to the implantation of < 100 CFU in the lung. The lungs and spleens were harvested at different time points and the numbers of CFU were determined. As expected, 30 days after the challenge with M. tuberculosis, a significantly reduced bacterial growth was observed in the lungs and spleens of vaccinated mice when compared to unvaccinated controls (Fig. 4D, P < 0.001). The capless mutant induced the same level of protection as the wild type strain.

Wild type and capless BCG induce the same level of protection against a M. tuberculosis challenge in a low-dose guinea pig aerosol model

Guinea pigs are a key TB vaccine model, as they present pathological features resembling the human disease and, like humans, express CD1b an antigen presenting molecule which is not expressed by mice and that is known to present LAM to antigen-specific T cells (Sieling et al., 1995; Prigozy et al., 1997). Thus, potential LAM-mediated immunity may not be evident in mice, making tests in the guinea pig model prone to show differences in protection not seen in mice. We thus tested capless versus wild type BCG in a low-dose aerosol challenge (10–50 CFU in the lung upon infection) in the guinea pig model. Figure 5 shows that the capless BCG is not more protective as a capped BCG.

figure

Figure 5. Capless and wild type (WT) M. bovis BCG induce the same level of protection against a M. tuberculosis challenge in the low-dose aerosol guinea pig model. Bacterial load of viable M. tuberculosis in spleen and lung of guinea pigs was determined. Adult guinea pigs were infected, sacrificed (eight per group) after 4 weeks and lung and spleen homogenates were plated for enumeration of bacilli (total CFU). Horizontal bars indicate medians after log transformation; error bars indicate range and P-values represent statistical comparisons (Mann–Whitney test) between BCG WT (closed circles) and PBS (closed squares) control groups and capless BCG (open circles).

Download figure to PowerPoint

Isolation of a capless M. tuberculosis mutant

The results obtained here as well as the already published information (Appelmelk et al., 2008; Festjens et al., 2011) point to a redundancy of the mannose cap in BCG and M. marinum virulence. However, BCG and M. tuberculosis differ drastically in terms of virulence and one could argue that the mannose cap in M. tuberculosis may have a more essential role for its virulence than in BCG. We thus studied a M. tuberculosis mutant lacking the ability to synthesize the mannose cap to evaluate the role of ManLAM in M. tuberculosis virulence. A capless mutant of M. tuberculosis H37Rv was constructed using the pGOAL-pNIL procedure (Parish and Stoker, 2000) exactly as described for M. bovis BCG, including isolation of the complementant (Appelmelk et al., 2008). Figure S1 provides genetic evidence that indeed Rv1635c was disrupted in the mutant and that in the complementant we succeeded in the reintroduction of an intact copy of this gene.

Mild acid hydrolysis of purified LAM followed by analysis of liberated oligosaccharides by capillary electrophoresis (Nigou et al., 2000) provided evidence that all cap motifs were missing (see Fig. S2). We investigated if differences in glycosylation, other than the absence of the cap, were present in the knockout by two independent approaches, i.e. SDS-PAGE followed by immunoblotting with glycosylation-specific probes (Fig. S3) and, second, mass spectrometric analysis of isolated phosphatidyl-myo-inositol mannosides (PIMs) (Fig. S4). No evidence for differences in glycosylation other than in the cap was found.

Wild type and capless M. tuberculosis show similar survival in murine macrophages and induce similar amounts of TNF and NO in vitro

We tested the interaction of wild type and capless M. tuberculosis with Balb/c macrophages. As shown in Fig. 6A, the phagocytosis and growth rate of the two strains in resting BMDM was similar. The restriction of their growth in macrophages following activation by cytokines was also of the same extent (Fig. 6B). The induction of TNF secretion and nitrite production did not differ significantly between the two strains (Fig. 6C and D). Hence, the difference in TNF production observed in capless BCG (see Fig. 1C) was not found for capless M. tuberculosis.

figure

Figure 6. In vitro interaction of M. tuberculosis with macrophages.

A. Growth of wild type (WT) or capless (CapA) mutant M. tuberculosis in bone marrow-derived macrophages (BMDM) from Balb/c mice.

B. Survival of WT and capless (CapA) mutant M. tuberculosis in untreated BMDM or BMDM treated with 50 U TNF, 100 U IFNγ or both (results expressed as ‘log CFU increase’, that corresponds to the difference of growth, in log CFU, between day 7 and day 0).

C. Secretion of TNF into the culture medium by BMDM infected with the M. tuberculosis strains.

D. Release of nitrite into the culture supernatants by BMDM infected with M. tuberculosis and treated with IFNγ. Data represent the mean ± 1 SD of a representative experiment out of a total of three experiments. No statistically significant differences were found.

Download figure to PowerPoint

Wild type and capless M. tuberculosis behave similarly in vivo with regard to replication and cytokine induction

The M. tuberculosis wild type strain, capless mutant, as well as a complemented strain were then used for in vivo infection studies. Balb/c mice were infected by the low-dose aerosol route. Infection resulted in the implantation of 10 to 25 bacilli in the lungs of each animal. Mice were sacrificed at days 1, 60 and 120 post infection and the bacterial loads in the lung and spleen determined. The course of the infection is presented in Fig. 7A and shows that all strains replicated to similar extents in both organs, with mice stabilizing the infection after day 60 post infection in both the lung and the spleen at similar bacterial loads. In addition, an identical immune response was observed for all the strains, with the production of similar amounts of IFNγ, TNF and IL-17 by spleen cells of mice with 60 days of infection (Fig. 7B). No IL-10 was detected in culture supernatants (data not shown). We extended this study to the widely used C57/BL6 mouse strain. Mice were infected as previously described and sacrificed at days 1 and 120 post infection. Results are expressed as ‘log CFU increase’ corresponding to the difference of growth, in log CFU between day 1 and day 120. Like in the Balb/c mouse model, there were no significant differences in growth between the parental and the capless strains (Fig. 7C).

figure

Figure 7. Capless and wild type (WT) M. tuberculosis have the same virulence.

A. Proliferation of WT, capless (CapA) mutant and complementant strains of M. tuberculosis in the lung and spleen of Balb/c mice following an aerogenic infection leading to the implantation of 10 to 25 bacilli in the lung of each animal.

B. Secretion of IFNγ, TNF and IL-17 into the culture supernatants by lung leucocytes from Balb/c mice infected for 60 days and following in vitro re-stimulation with live M. tuberculosis bacilli for 72 h.

C. Proliferation of M. tuberculosis WT or capless (CapA) mutant in C57/BL6 mice following an aerogenic infection. Mice were sacrificed at days 1 and 120 post infection and the bacterial loads in the lung and spleen determined. Results are expressed as ‘log CFU increase’, that corresponds to the difference of growth, in log CFU, between day 1 and day 120. A total of five to seven mice per time point were used.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The cell envelope of M. tuberculosis plays an important role in the pathogenesis of tuberculosis. For years, the ManLAM has been considered a key factor in host cell recognition and immunomodulation. The present work questions the importance of mannose capping of LAM as a requirement for virulence by showing that the in vivo growth in mice of mycobacteria (M. bovis BCG and M. tuberculosis) that are deficient in the biosynthesis of the mannose cap is not affected, and that a BCG mutant lacking mannose caps is as effective as a vaccine as its capped parent strain in low-dose (< 100 CFU) aerosol models in mice and guinea pigs. These data are surprising as mannose capping of LAM is found in slow-growing mycobacteria, among which are many pathogenic species, but is (almost completely) lacking in fast-growing, environmental non-pathogenic species.

It has been shown in vitro that, at the level of the macrophage, purified ManLAM interferes with phagosome maturation (Fratti et al., 2003; Vergne et al., 2003) and IFNγ-dependent activation (Sibley et al., 1988; Chan et al., 1991). ManLAM inhibits IL-12 secretion by LPS-stimulated human DCs (Nigou et al., 2001). In addition, ManLAM also prevents DC activation as measured by CD80, CD83 and CD86 expression and triggers secretion of IL-10 by LPS-primed human DCs through DC-SIGN binding (Geijtenbeek et al., 2003). Together, these data demonstrate an immunosuppressive activity of this lipoglycan and suggest that a capless Mycobacterium would be attenuated and would replicate less well in macrophages. Importantly, these studies were all conducted with purified ManLAM, not live bacteria.

Our in vitro findings comparing BCG with isogenic capless BCG showed that mannose capping of LAM is not crucial for mycobacteria survival in macrophages (Fig. 1A and B), with the capless BCG mutant growing exactly as the wild type strain both in non-activated and in activated bone marrow-derived primary murine macrophages (Fig. 1B) Our data are in agreement with two other studies where capless BCG replicated as well as its parent either in human THP-1 macrophages (Appelmelk et al., 2008) or in the murine macrophage cell line Mf4/4 (Festjens et al., 2011). Hence, the prediction that the mannose cap of LAM would influence mycobacterial survival does not come true in mice. With regard to cytokine induction in vitro, no difference was seen for IL-12p70 (Fig. 2B); for TNF, a difference was seen after 7 days only (Figs 1C and 2B); in contrast to data obtained with purified ManLAM and non-capped LAM (Geijtenbeek et al., 2003), BCG with capless LAM induced more IL-10 than parent BCG (Fig. 2B). Also in vivo, after intravenous injection in C57/Bl6 mice (Fig. 3), capless BCG was not attenuated as compared to parent strains, confirming earlier data following intranasal challenge in C57/Bl6 mice (Appelmelk et al., 2008) or intravenous injection in Balb/c mice (Festjens et al., 2011). Altogether, these data show that studies with live capless LAM mutant bacteria yield data conflicting with those obtained with purified LAM. This could be due to several reasons. First, to show enhanced IL-10 (Geijtenbeek et al., 2003) or decreased IL-12 (Nigou et al., 2001; Pathak et al., 2005) production by purified ManLAM as compared to LAM without mannose caps, DCs and macrophages were primed by the TLR4 ligand LPS; however, live mycobacteria are poor triggers of TLR4 signalling, (Reiling et al., 2008) and hence, not surprisingly, live mycobacteria do not recapitulate the behaviour of purified ManLAM. Second, we assume there is redundancy for the role of ManLAM. For example, binding of BCG to DCs is dominated by DC-SIGN ligand interactions (Geijtenbeek et al., 2003; Appelmelk et al., 2008; Geurtsen et al., 2009) and when the cap, i.e. one of the ligands, is removed, binding to DCs stays fully intact as a sufficient number of back-up ligands remain available, e.g. PIMs (Driessen et al., 2009), lipomannan and glycoproteins (Pitarque et al., 2005). We conclude that ligands other than the mannose cap determine binding to DCs, and assume that a similar redundancy exists for other effects of ManLAM, such as the inhibition of phagolysosome fusion.

A major goal of our studies was to evaluate the potency of capless BCG as a vaccine in a low-dose M. tuberculosis aerosol challenge model. Based on the ability of ManLAM to induce immunosuppressive IL-10 (Geijtenbeek et al., 2003), one could expect that a capless BCG might be a more effective vaccine than parent BCG. Using a high-dose murine M. tuberculosis challenge model (5 × 104 CFU intravenously or 2 × 105 CFU intratracheally), Festjens et al. (2011) showed that prior immunization with a capless BCG (105 CFU subcutaneously) indeed appeared to be more protective than parent BCG: mean survival time increased from 26.5 to 27.5 weeks (intravenous challenge) and from 48 to 52 weeks (intratracheal challenge); also after capless BCG immunization and intravenous M. tuberculosis challenge, the weight loss was delayed as compared to immunization with parent BCG. However, the differences in protection between capless BCG and parent strain were small and not statistically significant (N. Festjens, pers. comm.). We reasoned, based on early experience, that differences in immune protection could become more evident in a low-dose challenge model (Appelmelk et al., 1986). The low-dose aerosol M. tuberculosis infection model is currently considered to be the golden standard to evaluate protective efficacy of tuberculosis vaccines. In the aerogenic model, infectious doses are as low as 10–25 CFU. The outcome of our studies (Fig. 4A) is that BCG and capless BCG hardly disseminate to the spleen or lung after intradermal injection, that (Fig. 4B and C) immune parameters following immunization were similar for both vaccines, and most importantly, the capless BCG had a protective efficacy identical to its parent strain (Fig. 4D).

A second goal was to evaluate capless BCG in a particularly susceptible host, the guinea pig, an animal species which is also able to present the glycolipid LAM to T cells. Specialized CD1b lipid antigen presenting molecules are present in humans, but are lacking in mice. ManLAM has long been known to be presented by CD1b to antigen-specific T cells (Sieling et al., 1995), and hence in mice, protection mediated by LAM-specific CD1b-restricted T cells will not be evident. Guinea pigs, in contrast, express CD1b. In addition, unlike in mice, experimental tuberculosis in the guinea pigs causes caseating granulomas. Hence, as compared to mice, the guinea pig model is seen as more representative of human disease and is an essential step in human tuberculosis vaccine development. To mimic natural disease, the aerogenic challenge route was again chosen. Figure 5 shows that no differences in protection to M. tuberculosis infection were seen in this model following vaccination with the capless BCG or the parent strain.

A final goal was to test the role in virulence of the mannose cap of M. tuberculosis, a species not tested hitherto. In vitro, we found no differences between wild type and capless M. tuberculosis with regard to replication rate in mouse macrophages, susceptibility to cytokine-activated macrophages, and induction of TNF and nitrite secretion. To maximize the possibility to observe potential LAM-mediated immunomodulatory effects, we challenged via the aerogenic route two immunologically contrasting mouse strains, i.e. Balb/c (a Th2-skewed strain) and C57/Bl6 (a Th1-skewed strain) (see Fig. 7A and B respectively). However, in both mouse strains capless M. tuberculosis and parent strain proliferated equally well.

Altogether, our data suggest that the dominant role attributed to the mannose cap of LAM, which was predominantly based on in vitro studies with purified ManLAM (Mishra et al., 2011), cannot be confirmed by in vivo studies in mice with isogenic pairs of mutants. The cap does not affect immunoprotection by BCG, nor does it affect virulence of pathogenic mycobacteria in mice. One explanation for this lack of effect is that the role of the cap might be redundant as discussed above. Still, the outcome of our studies is puzzling: we investigated, within the taxonomic tree of the genus Mycobacterium, which species expressed a ManLAM and we found it almost exclusively in slow-growing species, many of which are pathogens; in fast-growing, often environmental species, the cap was mostly lacking (N.N. Driessen and B.J. Appelmelk, unpublished). This suggests evolutionary pressure to preserve the cap. It cannot be excluded that the cap, albeit not relevant in the biological assay systems tested so far with live mycobacteria, may still be important for example for transmission of M. tuberculosis from one person to another. As the DC-SIGN system in humans differs strongly from that of mice (Park et al., 2001), this aspect of tuberculosis is not accessible to animal experimentation; possibly, non-human primates mimic the human DC-SIGN system more accurately.

In short, together with two earlier studies, our novel data provide overwhelming evidence that the mannose cap of LAM does not dominate the interaction of mycobacteria with the three experimental animal hosts (zebrafish, mice and guinea pigs) tested.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Mycobacterial strains and growth conditions

The BCG capA mutant, lacking the mannose cap of LAM, has been described and characterized before (Appelmelk et al., 2008). The capA mutant in M. tuberculosis H37Rv has not been described before and was constructed in a similar way via a two-step p1NIL-pGOAL19 approach developed by Parish and Stoker (2000), leading to a markerless deletion in the gene of interest (rv1635c). In fact, exactly the same plasmids were used to genetically modify BCG and M. tuberculosis: the DNA sequences of Rv1635c and its BCG homologue are 100% identical. Briefly, after the second step (sucrose selection) of the p1NIL-pGOAL procedure, 14 colonies were picked. The presence of this deletion was investigated by PCR. Phenotypically, the lack of mannose caps was investigated in immunoblot. In this colony dot blot, 11 of the colonies were non-reactive with the cap-specific Mab 55.921A (Appelmelk et al., 2008) and three were reactive. In PCR those three yielded a rv1635c product, which was absent in the other 11 colonies. We concluded that of the 14 colonies picked, three were revertants and 11 were (markerless) mutants lacking Rv1635c. These three mutants were investigated in SDS-PAGE-Immunoblot with monoclonal antibodies F30-5, 183-24 and 55.92.1A1, concanavalin A and DC-SIGN-Fc and further evidence was obtained that they missed the cap (Fig. S3). Finally, one of the capless colonies was further investigated in capillary electrophoresis again providing evidence that the cap is missing (Fig. S2). A complementant was also made: rv1635c was cloned into the shuttle vector pSMT3 and the construct electroporated in the capA mutant of M. tuberculosis H37Rv. M. bovis BCG Copenhagen and mutant CapA were grown in liquid Middlebrook 7H9 broth (Difco) supplemented with 0.05% Tween 80, until the log phase. The H37Rv M. tuberculosis wild type, capless mutant and complemented strain were grown in Proskauer-Beck medium containing 0.05% Tween 80 to mid-log phase. Cultures were aliquoted and frozen at −80°C until the day of use.

Chemical analysis of the mannose caps

Lipoarabinomannans from M. tuberculosis wild type and M. tuberculosis capless mutant were analysed for presence of the mannose cap by the capillary electrophoresis (CE) technique described earlier (Nigou et al., 2000). Briefly, purified LAM was partially degraded by controlled acid hydrolysis (0.1 M HCl for 20 min at 110°C), and the oligosaccharides liberated tagged with the fluorescent label 8-aminopyrene-1,3,6-trisulfonate (APTS). During CE, the labelled oligosaccharides are separated and peaks are detected by laser-induced fluorescence and migration times compared with the appropriate controls.

Analysis of purified PIMs by mass spectrometry

Lipids were obtained by chloroform/methanol extraction of bacteria and subjected to MALDI-TOF MS analysis in the negative ion mode as previously described (Gilleron et al., 2003).

Laboratory mice

Eight-week-old female BALB/c and C57/B6 mice were purchased from Charles River (L'Arbresle, France) and housed under specific pathogen-free conditions in our facilities. Sterile chow and tap water ad libitum was given. All experiments were approved by and performed according to the guidelines of the animal ethical committee of the Instituto de Biologia Molecular e Celular (IBMC, Porto, Portugal). Five to seven animals were used per experimental group for each time point.

BCG antigens for cell stimulation

Mycobacteria antigens were prepared as described elsewhere (Pais et al., 2000). BCG was grown until log phase, at 37°C, in Middlebrook 7H9 medium (Difco) supplemented with 10% albumin/dextrose/catalase (ADC) and 0.05% Tween 80. The culture was centrifuged (10 000 g, 40 min, 4°C) and the remaining pellet washed and resuspended with phosphate-buffered saline (PBS), containing 0.1% Tween 80 (Sigma), 1 mM MgCl2 (Merck, Darmstadt, Germany) and 1 mM benzamidine (Sigma). The bacteria in suspension were disrupted through sonication with pulses of 1 min at maximum power, with the sample kept in ice during the whole procedure. The sonicate was centrifuged, to discard intact mycobacteria (30 min at 2700 g), and the supernatant was dialysed against PBS (molecular weight cut-off of 12 000), followed by ultra-centrifugation for 2 h at 150 000 g. The remaining pellet, containing the envelope proteins, was resuspended in PBS, and the supernatant, enriched in cytosolic proteins, was precipitated with 80% ammonium sulfate and dialysed against PBS. Aliquots were quantified and stored at −80°C until the day of use.

Generation of bone marrow-derived macrophages (BMMØ) and dendritic cells (BMDC)

Bone marrow cells were flushed from the femurs of mice with 5 ml of cold Hanks' balanced salt solution (HBSS; Gibco, Paisley, UK) using a 26-gauge needle. For macrophage generation, the resulting cell suspension was centrifuged for 10 min at 1200 r.p.m., 4°C, resuspended in Dulbecco's modified Eagle's medium (DMEM; Gibco) supplemented with 10 mM Hepes (Sigma, St Louis, MO, USA), 1 mM sodium pyruvate (Gibco), 10 mM glutamine (Gibco), 10% heat-inactivated fetal bovine serum (Sigma), 10% L929 cell conditioned medium (LCCM, as a source of M-CSF) and cultured for a period of 4 h on cell culture dishes (Nunc, Naperville, IL, USA) in order to remove already differentiated cells. Non-adherent cells were then collected with warm HBSS, counted, distributed in 24-well plates at a density of 5 × 105 cells/well and incubated in 1 ml of similar media at 37°C in a 5% CO2 atmosphere. On day 4 after seeding, 100 μl of LCCM was added and on day 7 the medium was renewed. Macrophages were infected at day 10. For DC differentiation, the cell suspension was cultured at a density of 106 cells ml−1 in RPMI 1640 containing GlutaMAX-I, supplemented with 5% v/v fetal calf serum, 50 mM β-mercaptoethanol and 10% GM-CSF-containing culture supernatant from transformed J558 cells. Every 2 days, one half of the media was removed and supplemented with complete medium with GM-CSF. On day 9, the non-adherent cells were cultured at 2 × 105 cells ml−1 in 96-well plates (100 μl/well), and infected at day 10. The purity of the population was determined by FACS analysis of specific surface markers and ranged from 85% to 95%.

Macrophage infection

A bacterial suspension containing 5 × 106 CFU ml−1 was prepared and 200 μl was added to each well to obtain a multiplicity of infection of two bacteria per macrophage. After 4 h of incubation at 37°C in a 5% CO2 atmosphere, cells were washed with warm HBSS to remove the non-internalized bacteria, and re-incubated in DMEM with 10% LCCM. All treatments (100 U IFNγ; 50 U TNF) were applied from day 0 until day 4. At different time points 100 μl of supernatant was collected for subsequent cytokine measurements. For the CFU assay, the infected cells were lysed with a 10% saponin solution, in sterilized water with 0.05% Tween 80, and serial dilutions of triplicate wells were performed. The number of viable bacteria was assessed by counting the colonies 3–4 weeks after plating on 7H11 Agar medium (Difco) supplemented with 10% oleic acid/albumin/dextrose/catalase (OADC) and incubated at 37°C.

Dendritic cell infection

A bacterial suspension containing 5 × 106 CFU ml−1 was prepared and 100 μl was added to each well in order to obtain a multiplicity of infection of two bacteria per DC. Supernatants were collected at different time points for IL-12p70, TNF and IL-10 cytokine measurements and the levels of expression of co-stimulatory molecules in cells were assessed by flow cytometry.

Cytokine measurement

Cytokine detection in the supernatants was performed by ELISA. For IFN-γ quantification, affinity-purified monoclonal antibodies (R4-6A2 as capture and biotinylated AN-18 as detecting antibody) were used, while commercial kits were used according to the manufacturer's instructions for the detection of TNF, IL-10 (R&D Systems), IL-12p70 and IL-17A (Biolegend).

Replication of M. bovis BCG in mice

Mice were infected intravenously through the lateral tail vein, with 5 × 104 CFU M. bovis BCG, wild type or mutant, in 200 μl of PBS. Bacterial loads in the organs of infected mice were evaluated at different time points post infection. Organs were homogenized in sterile water with 0.05% Tween 80 and 10-fold serial dilutions of organ homogenates were plated in duplicate onto Middlebrook 7H11 agar plates containing OADC. Plates where incubated at 37°C and colonies were counted 21 days later. Results are expressed as log CFU per organ.

Immunization and TB challenge

Balb/c mice were immunized by a single intradermal injection with 5 × 104 CFUs of M. bovis BCG, wild type or mutant. Two months later, mice were aerogenically infected with M. tuberculosis H37Rv using a Glass-col aerosol generation device chamber (Terre Haute, IN, USA). Briefly, mice were exposed for 30 min to an aerosol produced by nebulizing 10 ml of PBS-Tween 80 containing 106 CFU ml−1 that resulted in the implantation of 10–25 bacilli in the lung of each animal. Bacterial loads in the organs of infected mice were assessed by plating organ homogenates onto Middlebrook 7H11 agar plates, and counting the colonies formed 14 to 21 days after incubation at 37°C. Results are expressed as log CFU per organ.

Guinea pig immunization and infection

Groups of eight Dunkin–Hartley guinea pigs, weighing between 250 and 300 g (and free of infection), obtained from a commercial supplier (Harlan, UK), were used to evaluate the efficacy of capless BCG compared with BCG Danish 1331 (Statens Serum Institute, Copenhagen, Denmark), both delivered subcutaneously in a single dose at a concentration of 5 × 104 CFU and a negative control (PBS-vaccinated) group. Individual animals were identified using subcutaneously implanted microchips (PLEXX BV, the Netherlands). Guinea pig experimental work was conducted according to UK Home Office legislation for animal experimentation and was approved by the local ethics committee.

Animals were infected with a low aerosol dose (10–50 CFU retained dose in the lung) of M. tuberculosis H37Rv (Williams et al., 2000) 12 weeks after vaccination. Aerosol challenge was performed using a fully contained Henderson apparatus as previously described (Chambers et al., 2000; Lever et al., 2000; Clark et al., 2011) in conjunction with the AeroMP (Biaera) control unit (Hartings and Roy, 2004). Fine particle aerosols of M. tuberculosis H37Rv, with a mean diameter of 2 μm (diameter range, 0.5–7 μm) (Hartings and Roy, 2004), were generated using a Collison nebulizer and delivered directly to the animal snout. The aerosol was generated from a water suspension containing 5 × 106 CFU ml−1 in order to obtain an estimated retained, inhaled dose of approximately 10–50 CFU/lung. The Henderson apparatus allows controlled delivery of aerosols to the animals and the reproducibility of the system and relationship between inhaled CFU and the concentration of organisms in the nebulizer has been described previously (Chambers et al., 2000; Clark et al., 2011). The challenge system is controlled by an AeroMP: the aerosol management platform controls, monitors and records all relevant parameters during an aerosol procedure including air flow rate, temperature and relative humidity (Hartings and Roy, 2004). At 4 weeks post challenge, guinea pigs were killed humanely by intraperitoneal injection of pentabarbitone (Euthatal). Post-mortem, tissues were aseptically removed for bacteriology analysis. Tissues were homogenized in 5 ml of sterile distilled water using a rotating blade macerator system (Ystral, UK). Viable counts were performed on the macerate by preparing serial dilutions in sterile deionized water and plating 100 μl aliquots onto Middlebrook 7H11 + OADC agar (BioMerieux, UK). Plates were incubated at 37°C for 3 weeks before counting the number of M. tuberculosis colonies (CFU).

Cell preparation and in vitro stimulation

Spleens were gently disrupted with the help of a cell glass homogenizer. The resulting cell suspension was passed through a 70 μm nylon cell strainer, in order to remove large pieces and debris. Lung cell suspensions were obtained as follows: thoracic cavities were opened, and sterile PBS was gently injected into the right heart ventricles to perfuse lungs. Lungs were excised, sectioned and incubated with digestion media [DMEM supplemented with Collagenase IX (0.7 mg ml−1; Sigma)] for 30 min at 37°C. The final cell suspension was obtained by passing the digested lung tissue through a 70 μm nylon cell strainer. In all cell suspensions the red blood cells were lysed with a haemolytic solution [155 mM NH4Cl, 10 mM KHCO3 (pH 7.2)] during 5 min at room temperature. Cells were then distributed into 96-well plates (2.5 × 105 cells/well) and incubated in triplicate with different stimuli: DMEM culture medium and 4 μg ml−1 ConA (Sigma-Aldrich) as negative and positive controls, respectively, and depending on the experiment, 4 μg ml−1 BCG extract or 1.25 × 105 of live M. tuberculosis H37Rv bacilli. After 72 h of incubation at 37°C in a 5% CO2 atmosphere, the supernatants were collected for cytokine measurement. For intracellular staining, 1 × 106 cells ml−1 were incubated for 4 h at 37°C in the presence of PMA (Sigma-Aldrich) plus ionomycin (Calbiochem) at a final concentration of 25 μg ml−1 each, followed by an incubation of 2 h in the presence of 0.01 mg ml−1 brefeldin A (Sigma-Aldrich). Then, cells were fixed, permeabilized and stained with IFNγ-specific antibodies.

Flow cytometry

Cells were labelled with specific antibodies for CD3 (clone 145-2C11), CD4 (clone RM4-5), CD8 (clone 53-6.7), CD11b (clone M1/70), CD19 (clone 6D5), DX5 (clone HMα2), CD25 (clone PC61), CD86 (clone GL-1), CD40 (clone 5C3), MHC-II (clone M5/114.15.2) from BioLegend (San Diego, CA, USA) and FOXP3 (e-bioscience). Cell populations were acquired in a FACS Calibur instrument equipped with CellQuest software. Data were analysed using FlowJo software (Tree Star, Ashland, OR, USA).

Statistical analysis

Data were analysed by using Student's t-test. Statistical analyses of guinea pig data were performed using Minitab (version 13.32). The CFU data were analysed by non-parametric Mann–Whitney test comparisons to compare median values of the vaccine group with either the saline or BCG control groups.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

This work was supported by grant ImmunovacTB, ref. 37388 of the FP6 from the European Union, the NEWTBVAC project, ref. 241745 of the FP7 from the EU and by a grant from the Gulbenkian Foundation and TBVI. AAB, GTR, SSG, CN and SVC were supported by fellowships from Fundação para a Ciência e a Tecnologia (FCT) from the Portuguese Government. FM was supported by Wellcome Trust grant 073237. JG is financially supported by the Netherlands Organization for Scientific Research (NWO) through a VENI research grant (016.101.001). AAB is enrolled in the PhD Program in Experimental Biology and Biomedicine (PDBEB), Center for Neuroscience and Cell Biology, University of Coimbra, Portugal. We thank Marion Sparrius, Amsterdam, for technical assistance.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Adams, L.B., Fukutomi, Y., and Krahenbuhl, J.L. (1993) Regulation of murine macrophage effector functions by lipoarabinomannan from mycobacterial strains with different degrees of virulence. Infect Immun 61: 41734181.
  • Appelmelk, B.J., Verwey-van Vught, A.M., Maaskant, J.J., Schouten, W.F., Thijs, L.G., and Maclaren, D.M. (1986) Use of mucin and hemoglobin in experimental murine gram-negative bacteremia enhances the immunoprotective action of antibodies reactive with the lipopolysaccharide core region. Antonie Van Leeuwenhoek 52: 537542.
  • Appelmelk, B.J., den Dunnen, J., Driessen, N.N., Ummels, R., Pak, M., Nigou, J., et al. (2008) The mannose cap of mycobacterial lipoarabinomannan does not dominate the Mycobacterium-host interaction. Cell Microbiol 10: 930944.
  • Bradbury, M.G., and Moreno, C. (1993) Effect of lipoarabinomannan and mycobacteria on tumour necrosis factor production by different populations of murine macrophages. Clin Exp Immunol 94: 5763.
  • Briken, V., Porcelli, S.A., Besra, G.S., and Kremer, L. (2004) Mycobacterial lipoarabinomannan and related lipoglycans: from biogenesis to modulation of the immune response. Mol Microbiol 53: 391403.
  • Chambers, M.A., Williams, A., Gavier-Widen, D., Whelan, A., Hall, G., Marsh, P.D., et al. (2000) Identification of a Mycobacterium bovis BCG auxotrophic mutant that protects guinea pigs against M. bovis and hematogenous spread of Mycobacterium tuberculosis without sensitization to tuberculin. Infect Immun 68: 70947099.
  • Chan, J., Fan, X., Hunter, S.W., Brennan, P.J., and Bloom, B.R. (1991) Lipoarabinomannan, a possible virulence factor involved in persistence of Mycobacterium tuberculosis within macrophages. Infect Immun 59: 17551761.
  • Chatterjee, D., and Khoo, K.H. (1998) Mycobacterial lipoarabinomannan: an extraordinary lipoheteroglycan with profound physiological effects. Glycobiology 8: 113120.
  • Chatterjee, D., Roberts, A.D., Lowell, K., Brennan, P.J., and Orme, I.M. (1992) Structural basis of capacity of lipoarabinomannan to induce secretion of tumor necrosis factor. Infect Immun 60: 12491253.
  • Clark, S.O., Hall, Y., Kelly, D.L., Hatch, G.J., and Williams, A. (2011) Survival of Mycobacterium tuberculosis during experimental aerosolization and implications for aerosol challenge models. J Appl Microbiol 111: 350359.
  • Colditz, G.A., Brewer, T.F., Berkey, C.S., Wilson, M.E., Burdick, E., Fineberg, H.V., and Mosteller, F. (1994) Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the published literature. JAMA 271: 698702.
  • Dinadayala, P., Kaur, D., Berg, S., Amin, A.G., Vissa, V.D., Chatterjee, D., et al. (2006) Genetic basis for the synthesis of the immunomodulatory mannose caps of lipoarabinomannan in Mycobacterium tuberculosis. J Biol Chem 281: 2002720035.
  • Driessen, N.N., Ummels, R., Maaskant, J.J., Gurcha, S.S., Besra, G.S., Ainge, G.D., et al. (2009) Role of phosphatidylinositol mannosides in the interaction between mycobacteria and DC-SIGN. Infect Immun 77: 45384547.
  • Festjens, N., Bogaert, P., Batni, A., Houthuys, E., Plets, E., Vanderschaeghe, D., et al. (2011) Disruption of the SapM locus in Mycobacterium bovis BCG improves its protective efficacy as a vaccine against M. tuberculosis. EMBO Mol Med 3: 222234.
  • Fratti, R.A., Chua, J., Vergne, I., and Deretic, V. (2003) Mycobacterium tuberculosis glycosylated phosphatidylinositol causes phagosome maturation arrest. Proc Natl Acad Sci USA 100: 54375442.
  • Geijtenbeek, T.B., Van Vliet, S.J., Koppel, E.A., Sanchez-Hernandez, M., Vandenbroucke-Grauls, C.M., Appelmelk, B., and Van Kooyk, Y. (2003) Mycobacteria target DC-SIGN to suppress dendritic cell function. J Exp Med 197: 717.
  • Geurtsen, J., Chedammi, S., Mesters, J., Cot, M., Driessen, N.N., Sambou, T., et al. (2009) Identification of mycobacterial alpha-glucan as a novel ligand for DC-SIGN: involvement of mycobacterial capsular polysaccharides in host immune modulation. J Immunol 183: 52215231.
  • Gheorghiu, M., Mouton, D., Lecoeur, H., Lagranderie, M., Mevel, J.C., and Biozzi, G. (1985) Resistance of high and low antibody responder lines of mice to the growth of avirulent (BCG) and virulent (H37Rv) strains of mycobacteria. Clin Exp Immunol 59: 177184.
  • Gilleron, M., Quesniaux, V.F., and Puzo, G. (2003) Acylation state of the phosphatidylinositol hexamannosides from Mycobacterium bovis bacillus Calmette Guerin and Mycobacterium tuberculosis H37Rv and its implication in Toll-like receptor response. J Biol Chem 278: 2988029889.
  • Gilleron, M., Jackson, M., Nigou, J., and Puzo, G. (2008) Structure, biosynthesis, and activities of the phosphatidyl-myo-inositol based lipoglycans. In The Mycobacterial Cell Envelope. Daffé, M. , and Reyrat, J.-M. (eds). Washington, DC: ASM Press, pp. 75105.
  • Guerardel, Y., Maes, E., Elass, E., Leroy, Y., Timmerman, P., Besra, G.S., et al. (2002) Structural study of lipomannan and lipoarabinomannan from Mycobacterium chelonae. Presence of unusual components with alpha 1,3-mannopyranose side chains. J Biol Chem 277: 3063530648.
  • Hartings, J.M., and Roy, C.J. (2004) The automated bioaerosol exposure system: preclinical platform development and a respiratory dosimetry application with nonhuman primates. J Pharmacol Toxicol Methods 49: 3955.
  • Kang, P.B., Azad, A.K., Torrelles, J.B., Kaufman, T.M., Beharka, A., Tibesar, E., et al. (2005) The human macrophage mannose receptor directs Mycobacterium tuberculosis lipoarabinomannan-mediated phagosome biogenesis. J Exp Med 202: 987999.
  • Khoo, K.H., Dell, A., Morris, H.R., Brennan, P.J., and Chatterjee, D. (1995) Inositol phosphate capping of the nonreducing termini of lipoarabinomannan from rapidly growing strains of Mycobacterium. J Biol Chem 270: 1238012389.
  • Knutson, K.L., Hmama, Z., Herrera-Velit, P., Rochford, R., and Reiner, N.E. (1998) Lipoarabinomannan of Mycobacterium tuberculosis promotes protein tyrosine dephosphorylation and inhibition of mitogen-activated protein kinase in human mononuclear phagocytes. J Biol Chem 273: 645652.
  • Lever, M.S., Williams, A., and Bennett, A.M. (2000) Survival of mycobacterial species in aerosols generated from artificial saliva. Lett Appl Microbiol 31: 238241.
  • Maeda, N., Nigou, J., Herrmann, J.L., Jackson, M., Amara, A., Lagrange, P.H., et al. (2003) The cell surface receptor DC-SIGN discriminates between Mycobacterium species through selective recognition of the mannose caps on lipoarabinomannan. J Biol Chem 278: 55135516.
  • Mishra, A.K., Driessen, N.N., Appelmelk, B.J., and Besra, G.S. (2011) Lipoarabinomannan and related glycoconjugates: structure, biogenesis and role in Mycobacterium tuberculosis physiology and host-pathogen interaction. FEMS Microbiol Rev 35: 11261157.
  • Murray, P.J. (1999) Defining the requirements for immunological control of mycobacterial infections. Trends Microbiol 7: 366372.
  • Nigou, J., Vercellone, A., and Puzo, G. (2000) New structural insights into the molecular deciphering of mycobacterial lipoglycan binding to C-type lectins: lipoarabinomannan glycoform characterization and quantification by capillary electrophoresis at the subnanomole level. J Mol Biol 299: 13531362.
  • Nigou, J., Zelle-Rieser, C., Gilleron, M., Thurnher, M., and Puzo, G. (2001) Mannosylated lipoarabinomannans inhibit IL-12 production by human dendritic cells: evidence for a negative signal delivered through the mannose receptor. J Immunol 166: 74777485.
  • Nigou, J., Gilleron, M., Rojas, M., Garcia, L.F., Thurnher, M., and Puzo, G. (2002) Mycobacterial lipoarabinomannans: modulators of the dendritic cell function and the apoptotic response. Microbes Infect 4: 945953.
  • Nigou, J., Gilleron, M., and Puzo, G. (2003) Lipoarabinomannans: from structure to biosynthesis. Biochimie 85: 153166.
  • Pais, T.F., Cunha, J.F., and Appelberg, R. (2000) Antigen specificity of T-cell response to Mycobacterium avium infection in mice. Infect Immun 68: 48054810.
  • Parish, T., and Stoker, N.G. (2000) Use of a flexible cassette method to generate a double unmarked Mycobacterium tuberculosis tlyA plcABC mutant by gene replacement. Microbiology 146: 19691975.
  • Park, C.G., Takahara, K., Umemoto, E., Yashima, Y., Matsubara, K., Matsuda, Y., et al. (2001) Five mouse homologues of the human dendritic cell C-type lectin, DC-SIGN. Int Immunol 13: 12831290.
  • Pathak, S.K., Basu, S., Bhattacharyya, A., Pathak, S., Kundu, M., and Basu, J. (2005) Mycobacterium tuberculosis lipoarabinomannan-mediated IRAK-M induction negatively regulates Toll-like receptor-dependent interleukin-12 p40 production in macrophages. J Biol Chem 280: 4279442800.
  • Pitarque, S., Herrmann, J.L., Duteyrat, J.L., Jackson, M., Stewart, G.R., Lecointe, F., et al. (2005) Deciphering the molecular bases of Mycobacterium tuberculosis binding to the lectin DC-SIGN reveals an underestimated complexity. Biochem J 392: 615624.
  • Prigozy, T.I., Sieling, P.A., Clemens, D., Stewart, P.L., Behar, S.M., Porcelli, S.A., et al. (1997) The mannose receptor delivers lipoglycan antigens to endosomes for presentation to T cells by CD1b molecules. Immunity 6: 187197.
  • Reiling, N., Ehlers, S., and Holscher, C. (2008) MyDths and un-TOLLed truths: sensor, instructive and effector immunity to tuberculosis. Immunol Lett 116: 1523.
  • Roach, T.I.A., Barton, C.H., Chatterjee, D., and Blackwell, J.M. (1993) Macrophage activation: lipoarabinomannan from avirulent and virulent strains of Mycobacterium tuberculosis differentially induces the early genes c-fos, KC, JE, and tumor necrosis factor-α. J Immunol 150: 18861896.
  • Schlesinger, L. (1993) Macrophage phagocytosis of virulent but not attenuated strains of Mycobacterium tuberculosis is mediated by mannose receptors in addition to complement receptors. J Immunol 150: 29202930.
  • Schlesinger, L.S., Hull, S.R., and Kaufman, T.M. (1994) Binding of the terminal mannosyl units of lipoarabinomannan from a virulent strain of Mycobacterium tuberculosis to human macrophages. J Immunol 152: 40704079.
  • Sibley, L.D., Hunter, S.W., Brennan, P.J., and Krahenbuhl, J.L. (1988) Mycobacterial lipoarabinomannan inhibits gamma interferon-mediated activation of macrophages. Infect Immun 56: 12321236.
  • Sieling, P.A., Chatterjee, D., Porcelli, S.A., Prigozy, T.I., Mazzaccaro, R.J., Soriano, T., et al. (1995) CD1-restricted T cell recognition of microbial lipoglycan antigens. Science 269: 227230.
  • Tailleux, L., Schwartz, O., Herrmann, J.L., Pivert, E., Jackson, M., Amara, A., et al. (2003) DC-SIGN is the major Mycobacterium tuberculosis receptor on human dendritic cells. J Exp Med 197: 121127.
  • Torrelles, J.B., Azad, A.K., and Schlesinger, L.S. (2006) Fine discrimination in the recognition of individual species of phosphatidyl-myo-inositol mannosides from Mycobacterium tuberculosis by C-type lectin pattern recognition receptors. J Immunol 177: 18051816.
  • Vergne, I., Chua, J., and Deretic, V. (2003) Tuberculosis toxin blocking phagosome maturation inhibits a novel Ca2+/calmodulin-PI3K hVPS34 cascade. J Exp Med 198: 653659.
  • Welin, A., Winberg, M.E., Abdalla, H., Sarndahl, E., Rasmusson, B., Stendahl, O., and Lerm, M. (2008) Incorporation of Mycobacterium tuberculosis lipoarabinomannan into macrophage membrane rafts is a prerequisite for the phagosomal maturation block. Infect Immun 76: 28822887.
  • Williams, A., Davies, A., Marsh, P.D., Chambers, M.A., and Hewinson, R.G. (2000) Comparison of the protective efficacy of bacille Calmette-Guerin vaccination against aerosol challenge with Mycobacterium tuberculosis and Mycobacterium bovis. Clin Infect Dis 30: 299301.
  • Yoshida, A., and Koide, Y. (1997) Arabinofuranosyl-terminated and mannosylated lipoarabinomannans from Mycobacterium tuberculosis induce different levels of interleukin-12 expression in murine macrophages. Infect Immun 65: 19531955.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
cmi12065-sup-0001-si.tif84K

Fig. S1. Genetic evidence for the deletion and complementation of gene Rv1635c. The left panel shows the PCR with deletion primers GGCGAACGGAACACTGTGAT and CCGTCTGGCTGACCGTATTA in the deleted region of Rv1635c. Lane 1: M. tuberculosis H37Rv WT; Lane 2: M. tuberculosis H37Rv ΔRv1635c; Lane 3: Complemented M. tuberculosis ΔRv1635c; Lane 4: Negative control; Lane 5: Molecular Weight marker. The predicted product size is 347 bp. As expected parent strain and complementant show a product which is of the correct seize, while the knockout does not yield a product. The right panel shows the PCR with flanking primers GTGCCGGTGTGGTCGCTATT and ACCTGGCAACCTGCCGACTT in the regions upstream and downstream of gene Rv1635c. Lane 1: M. tuberculosis H37Rv WT; Lane 2: M. tuberculosis H37Rv ΔRv1635c; Lane 3: Complemented M. tuberculosis ΔRv1635c; Lane 4: Negative control; Lane 5: M. bovis BCG; Lane 6: Molecular Weight markers. The predicted size of the product in M. tuberculosis H37Rv and M. bovis BCG is 3256 bp; predicted size in H37Rv ΔRv1635c is 2492 bp. As expected the complemented strain still yields the knockout product size as the complementation plasmid carries a full-length copy of Rv1635c but no flanking regions. Together the two PCRs prove the identity of parent strain, knockout and complementant.

cmi12065-sup-0002-si.tif149K

Fig. S2. Mannooligosaccharide cap analysis of LAM. Purified LAM was analysed for presence of the mannose caps by capillary electrophoresis monitored by laser-induced fluorescence after mild acid hydrolysis and 8-aminopyrene-1,3,6-trisulfonate tagging. Shown are the profiles of LAM from M. tuberculosis wild type (trace 1) and M. tuberculosis capless mutant (trace 2). A, Ara-APTS; M, Man-APTS; S, internal standard, mannoheptose-APTS; AM, Manp-(α1[RIGHTWARDS ARROW]5)-Ara-APTS (monomannoside cap); AMM, Manp-(α1[RIGHTWARDS ARROW]2)-Manp-(α1[RIGHTWARDS ARROW]5)-Ara-APTS (dimannoside cap); AMMM, Manp-(α1[RIGHTWARDS ARROW]2)-Manp-(α1[RIGHTWARDS ARROW]2)-Manp-(α1[RIGHTWARDS ARROW]5)-Ara-APTS (trimannoside cap).

cmi12065-sup-0003-si.tif3700K

Fig. S3. Evidence for a LAM-specific defect in mannosylation. M. tuberculosis H37Rv, capless mutant and complementant were grown on plates, suspended to 50 mg ml−1 wet-weight, disrupted by beat-beating and 10 μl samples subjected to SDS-PAGE (12% acrylamide), blotted to PVDF membranes and probed with various monoclonal antibodies or lectins at 1–2 mg ml−1 and immunostained. Mab F30-5 recognizes the arabinan domain of LAM; conA and DC-SIGN are lectins specific for mannosyl residues; Mab 183-24 recognizes tri-mannosyl residues in both the mannose cap of LAM and PIMs; Mab 55.92.1A1 recognizes mannose residues in cap only. Arrows indicate the migration of the indicated molecules. Ponceau staining proves adequate transfer to PVDF. The five immunoblots together show that only the mannose cap of LAM is affected in the capless mutant with no other changes in mannosylation visible. Molecular weight markers (expressed in kDa) are indicated along the Ponceau staining.

cmi12065-sup-0004-si.tif284K

Fig. S4. MALDI-TOF Mass Spectrometry analysis of PIMs from M. tuberculosis wild type (A) and capless mutant (B) strains. Lipids were obtained by chloforform/methanol extraction of bacteria and subjected to MALDI-TOF MS analysis in the negative ion mode as previously described (Gilleron et al., 2003).

cmi12065-sup-0005-si.doc30KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.