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

Genetic Perspectives

Comas et al. have recently proposed that M. tuberculosis is capable of bimodal disease characteristics: the ancient strains having evolved to persist in an era of low human population density, and modern strains behaving like a crowd disease that are more virulent and spread more rapidly.[75] The exploitation of what may broadly be considered 'fungal PRR' may be one mechanism by which the emerging strains promote a Th2-type response that favors dissemination in regions of high human population density. In essence, these strains may be 'devolving' the ancient characteristics that enabled the Th1-type host response to either resolve or contain the infection at a manageable chronic persistence.

The acquisition of fungal-like characteristics by M. tuberculosis must by necessity be preserved by the forces of natural selection. While some traits may be the vestigial adaptations of phylogenetic ancestors, for example to environmental stress, others may reflect the co-evolution with viruses; however, the most extensively studied are factors pertaining to the human immune system. Modern technology has enabled the transcriptional response from whole-blood samples of the host in response to M. tuberculosis infection to be profiled, and unique signatures have been observed that distinguish it from other bacterial infections.[47] It would be interesting to conduct similar experiments to directly compare the response between virulent strains of mycobacteria and fungi. From the pathogen perspective, the exposure of a given strain of M. tuberculosis to greater genetic variation among hosts as a consequence of modern travel and the integration of mixed populations is a force that drives what may be seen as a change in the historically stable nature of host–pathogen interaction, creating a state of flux in the evolutionary history of this disease.

The genetic history of M. tuberculosis suggests very little antigenic variation with regard to predicted T-cell epitopes, leading to speculation that these epitopes may elicit a response that is beneficial to the bacteria.[1,76] This may reflect an equilibrium state whereby both host and pathogen had evolved an almost commensal relationship, namely the above-described induction of a Th1 inflammatory response that controls the infection but allows persistence in the granuloma. However, induction of a Th2 response by virulent strains promotes survival of bacilli inside macrophages, resulting in less antigen presentation to T cells among these strains. Therefore, and so long as the Th2 phenotype is preserved, antigenic variation may occur in these strains that is not registered by the host immune system; this is only exacerbated in the immunocompromised host. Freed of the selective pressure to maintain long-preserved T-cell receptor epitopes, more virulent, drug-resistant, antigenically and morpholigally diverse strains may arise. On a related note, the ability of virulent strains to survive intracellularly and thereby limit the direct presentation of antigen to T cells may partially explain the variable efficacy of BCG M. bovis vaccine that generates immune memory only against antigens that are readily expressed by the major histocompatibility complex (MHC). Virulent strains may also reduce cross-presentation of antigen to T cells due to their ability to inhibit apoptosis of infected cells,[77] a feature shared with some fungal species, such as Histoplasma sp..[78]

An exception to the observation of low genetic volatility in M. tuberculosis can be found among the PE/PPE genes clusters,[79] which have been proposed as mechanisms of immunogenic variation and capable of redirecting an immune response away from a protective cell-mediated to a more permissive humoral response.[80] Interestingly, analogous regions have been discovered in fungal genomes. Termed 'megasatellites', they are comprised of large DNA tandem repeats in protein-coding regions with a similarly poorly defined function as their mycobacterial counterparts.[81] Furthermore, there are predicted fungal proteins that resemble PPE family members, such as EMBL CCF43578.1 from Colletotrichum higginsianum and EMBL EMR87799.1 from Botryotinia fuckeliana.

Sulfolipid-1 is an abundant cell wall glycolipid that has been linked with mycobacterial virulence. One of the key enzymes required for its synthesis shares conserved domains with PE/PPE family proteins and mycobacterial cutinase-like proteins.[82] The presence of such a domain is curious. Cutinases are secreted enzymes found in some soil-dwelling bacteria, but the mycobacterial cutinase-like domains more closely resemble the crystal structure of those from fungi such as Fusarium solani and Cryptococcus sp., despite low sequence homology.[83,84] Saprophytic fungi use cutinases to break down the surface of plants; however, they have also been proposed to degrade fatty acids as well as generate diverse immunological effects.[85]M. tuberculosis does not encounter cutin polymers in the host, and instead mycobacterial cutinases such as Rv1984c and Rv3452 may help metabolize an alternative carbon source for the bacteria, as well as provide acyl carbon chains for the synthesis of mycolic acids. While a comparable tertiary structure derived from variable genetic code may suggest convergent evolution, the enzyme in question, Rv3822, is actually predicted to have been acquired by M. tuberculosis through horizontal gene transfer (reviewed by Stinear et al.[86]). Fungi are themselves proposed to have gained cutinases by gene transfer from actinomycetes[87] and, in general, gene transfer from bacteria to fungi is now largely accepted.[88] Thus, what we think of as fungal characteristics in M. tuberculosis may have originated from genes that are paralogues rather than orthologues.

Whole-genome comparisons of mycobacteria reveal adaptations from environmental to pathogenic lifestyles.[1,89] Fungi and soil-dwelling actinomycetes must retain genes required for environmental survival, and fungi are typically considered to be only opportunistic human pathogens, unlike M. tuberculosis, which is a specialist. Even among the Mycobacterium spp., M. tuberculosis is unique in that is an obligate pathogen: there is no environmental reservoir, unlike Mycobacteriummarinum,[86]Mycobacterium bovis,[90]Mycobacteriumavium[91] and probably Mycobacterium leprae.[92] Transmission involves only a very brief stage outside the host. Therefore, M. tuberculosis has been able to dispense with ancestral genes that are not required for this challenge but retains a core genome,[1,86] and retains and evolves traits that may be found in some actinomycetes and fungi that initially appeared for environmental survival. For example, cell wall components such as trehalose, discussed above, which protects against dessication and ultraviolet light, is similarly beneficial at resisting oxidative stress from an activated macrophage. Therefore, the common features between fungi and M. tuberculosis described presently are not simply restricted to examples of convergent evolution, but also gene loss, mutation, duplication and gene transfer.

Future research may shed light on more subtle comparisons, where perhaps there is not an immediately apparent relationship between what are clearly two distinct branches of the evolutionary tree. The discovery of shared genetic heritage between mycobacteria and proteobacteria is a case in point.[93] What is known for certain, is that M. tuberculosis is truly distinct from other bacteria in its evolutionary origins, as evidenced by its possession of a number of genes that are either of eukaryotic origin or prokaryotic–eukaryotic gene fusion events, including the cutinases, described above, and proteins involved in sterol interaction.[94,95]