Shared Characteristics Between Mycobacterium tuberculosis and Fungi Contribute to Virulence

Sam Willcocks; Brendan W Wren


Future Microbiol. 2014;9(5):657-668. 

In This Article

The Immune Response

Th1 vs Th2

In terms of the host response to infection, several mycobacterial cell wall components, such as LAM, LM, PIM2 and PIM6, have been described to activate Toll-like receptor (TLR)2 and drive a protective Th1-type proinflammatory response, including the upregulation of nitric oxide, IL-12, IL-1β and TNF-α production.[36] TNF-α is crucial in controlling tuberculosis infection due to its ability to maintain the granuloma, activate macrophages, mature dendritic cells (DC), and induce apoptosis in infected cells. M. tuberculosis therefore clearly expresses pathogen-associated molecular patterns (PAMP), that may result in its destruction by virtue of the innate immune response, providing selective pressure to evade this fate.

A moderate but not excessive Th2 response is required by the host to protect from hyper-immune pathology. However, induction of a Th2 inflammatory response and its associated cytokines is normally associated with defense against extracellular pathogens such as helminths, as well as the allergic response. A relative shift towards Th2 is counter to a robust Th1 response that is effective against intracellular pathogens, such as M. tuberculosis, which is able to survive and divide inside macrophages. Th1-associated cytokines such as IFN-γ and TNF-α activate macrophages, enabling them to overcome mycobacterium-induced delay in phagosome maturation and reduce their intracellular survival.[23] Reduced IFN-γ may also exacerbate symptoms of disease, since one function of this cytokine is to antagonize fibrosis of lung tissue, a potential consequence of cavitation that contributes to respiratory dysfunction in pulmonary tuberculosis. The Th2 milieu is required for efficient development of a B-cell response, but the contribution of B cells to control of tuberculosis is questionable, since patients with high antibody titers against M. tuberculosis are not protected from the disease.[37]

Challenge with M. tuberculosis does not induce a purely Th1 nor Th2 pathway however, but rather parallel arms of the immune response.[38] Concomitant induction of IL-10, as with recognition of mycobacterial PAMPs by dendritic cells, can promote disease due to its potent anti-inflammatory effects, including inhibition of phagosome maturation.[39] Strains of M. tuberculosis that preferentially induce a Th2-like response, featuring induction of IL-4 and IL-13 are more virulent.[40] This tipping of the balance from Th1 towards Th2 is perhaps a hallmark of the more virulent strains of M. tuberculosis. Indeed, the absence of the IL-4 gene in mice significantly diminishes bacterial growth.[41] The relative abundance of IL-4 in infections with virulent strains is also highly significant, since it alters the effect of TNF-α from being one of protection to one of toxicity.[38] IL-4 also counteracts Th1-dependent mycobacterial killing, including IFN-γ-induced autophagy in macrophages.[42]

There are interesting parallels to be drawn here, since the advantageous bias towards Th2 – a response geared towards resolving allergic and parasitic insult – is a strategy also demonstrated by successful fungal infections.[43] For example, a Th2-dominant allergic inflammatory response in the airway eliciting eosinophil maturation and mucus production is induced by Candida albicans and Aspergillus fumigatus. This pathogenesis strategy fails in the presence of bacteria that redirect the response towards Th1 and Th17 bias,[44] the latter of which has recently been described to be important in resolving both fungal and mycobacterial infection, and is driven in a TLR-dependent manner upon challenge with M. tuberculosis.[43,45,46]M. tuberculosis, like these fungi, is more virulent when bacterial immunity is bypassed. Hijacking a pathway that has evolved to cope with a different class of pathogen has been noted previously with mycobacteria in the context of the host antiviral response.[47,48] One consequence of this is the perversion of the traditional host–pathogen co-evolution scenario of reciprocal 'tit-for-tat' changes, since the host is under independent selective pressure from the other class of pathogen and risks vulnerability to these if it changed its receptors to prevent their use by mycobacteria.

Engagement of Pattern Recognition Receptors

The innate immune response is orchestrated by an array of pattern recognition receptors (PRRs) at the cell surface that recognize specific PAMPs. The lung, the primary site of tuberculosis transmission, is considered to be a relatively immuno-permissive environment.[49] This has evolved to tolerate constant challenge by foreign particles including fungal spores encountered during inhalation. This is regulated with a remarkable degree of elegance; for example, the ability of resident dendritic cells to discriminate between PAMPs of potentially colonizing fungal conidia versus hyphae and appropriately eliciting either a Th1 or Th2 inflammatory response, respectively.[50] Differential PRR ligands are expressed by dividing yeast that are masked in hyphae; similar heterogeneity of molecular expression at the cell surface has now been described for M. tuberculosis using labelled synthetic sugars to track the preferential incorporation and accumulation of these components at the poles of dividing bacilli.[51]

Among the best described PRRs are the TLR. The majority have evolved to ligate bacterial targets: TLR2 recognizes Gram-positive lipoproteins; TLR4 recognizes LPS from Gram-negative bacteria; TLR5 recognizes flagellin, and TLR9 recognizes bacterial DNA. These receptors are germline encoded and highly conserved. Their engagement is therefore open to manipulation by bacteria that are able to modify the PAMP they display on a vastly more swift evolutionary time scale due to their much shorter generation time. Accordingly, M. tuberculosis displays strategies to minimize TLR engagement in favor of other cell surface receptors, and virulent strains have even been reported to downregulate a key TLR signalling molecule, MyD88.[52] Fungi too are generally considered to be poor agonists of TLR, and rather engage alternative PRRs, including Dectin-1 (also known as CLEC7A), MR, DC-SIGN and Mincle,[53] all of which have also been reported as key receptors for M. tuberculosis. These receptors are members of the C-type lectin (CTL) family, which increasingly have been shown to be capable of antagonizing TLR-driven Th1 responses while permitting phagocytosis and Th2 inflammation. CTL may also interplay with other classes of immune receptors; for example, their triggering in the presence of FcγR ligation enhances IL-10 production.[54] Since MR and Mincle lack their own cytoplasmic signalling capability, docking of either receptor with their ligand may therefore sequester ligands that would otherwise trigger an intracellular signalling cascade via TLR.

Another strategy to avoid TLR-Th1 immunity is to conceal potent TLR agonists. For example, the archetypal ligands for TLR2 and TLR4, peptidoglycan (PGN)[55] and LPS, are respectively buried deep underneath the lipid-rich mycobacterial cell wall or are absent completely. The fact that PGN is located beneath capsular material, mycolic acids and arabinogalactan reduces its exposure at the cell surface.[56] PGN may have a role in the innate immune response to M. tuberculosis through engagement of NOD2 receptors, but since these are cytosolic receptors, any such interaction occurs after uptake of the bacterium and would require possible shedding of cell wall PGN or phagosomal escape, which would all occur downstream of TLR recognition.[57] The role of NOD2 in control of M. tuberculosis infection is still not clear, and neither are fungi historically thought to activate NOD2, although recent research suggests invasive, colonizing Aspergillus conidia can indeed engage this pathway.

Mycobacterium abscessus has been described to express glycopeptidolipids that mask underlying lower-order PIMs that would otherwise trigger TLR2-dependent TNF-α production,[38] and M. tuberculosis employs a similar strategy to enhance virulence, masking TLR-triggering PAMPs by the cell wall lipid, phthiocerol dimycoceroserate (PDIM).[58] Concealing antigens beneath the thick cell wall of mycobacteria is analogous to the capsule of virulent fungal strains of Cryptococcus neoformans[59] that, similar to M. tuberculosis, is known to establish latent infection and disseminate without triggering protective inflammation.[60] For example, C. neoformans is able to impair the production of granulocyte–macrophage colony-stimulating factor (GM-CSF) and TNF-α by natural killer cells, which has likewise been observed among multi-drug-resistant (MDR)-TB strains.[61]

The primary composition of the Cryptococcus sp. capsule are polysaccharides rich in mannans, which compares favorably with the composition of polysaccharides extracted from the capsule of M. tuberculosis.[62,63] The capsule of M. tuberculosis had previously remained elusive for study due to the difficulty in recovering it intact using typical in vitro methods. The potential for fungal and mycobacterial mannan components to mask proinflammatory ligands represents a novel function for this sugar to those previously described. Capping of LAM with mannose also serves this purpose, as LAM is a reported ligand for TLR2.[64] The capsule of both organisms may further contribute to the evasion of humoral defense: as with M. tuberculosis, antibodies specific to the outer capsule of the fungus are found in patient serum, but the avidity is poor and the antibodies are either non- or weakly opsonic.

In summary, the selective engagement of receptors that minimize Th1 and maximize Th2 outcome arguably represents convergent evolution of the most virulent strains of M. tuberculosis to exploit a similar inflammatory response to pathogenic fungi. Specific targeting of CTL receptors over TLR enables efficient uptake into cells while minimizing inflammation. Bypassing the TLR machinery still permits production of the chemokine CXCL8, resulting in an influx of neutrophils to the lung that are responsible for much of the tissue pathology seen with tuberculosis due to release of reactive oxygen species, helping to create a privileged habitat for the proliferation of mycobacteria. Neutrophils are also the main producers of IL-10 during infection with M. tuberculosis, further antagonizing the protective Th1 response.