key: cord-0922307-ignsjxls
authors: Silva, Marta L.; Cá, Baltazar; Osório, Nuno S.; Rodrigues, Pedro N. S.; Maceiras, Ana Raquel; Saraiva, Margarida
title: Tuberculosis caused by Mycobacterium africanum: Knowns and unknowns
date: 2022-05-26
journal: PLoS Pathog
DOI: 10.1371/journal.ppat.1010490
sha: 756e3312a0ca961ca077cd065ede7c7d58aafec5
doc_id: 922307
cord_uid: ignsjxls

Tuberculosis (TB), one of the deadliest threats to human health, is mainly caused by 2 highly related and human-adapted bacteria broadly known as Mycobacterium tuberculosis and Mycobacterium africanum. Whereas M. tuberculosis is widely spread, M. africanum is restricted to West Africa, where it remains a significant cause of tuberculosis. Although several differences have been identified between these 2 pathogens, M. africanum remains a lot less studied than M. tuberculosis. Here, we discuss the genetic, phenotypic, and clinical similarities and differences between strains of M. tuberculosis and M. africanum. We also discuss our current knowledge on the immune response to M. africanum and how it possibly articulates with distinct disease progression and with the geographical restriction attributed to this pathogen. Understanding the functional impact of the diversity existing in TB-causing bacteria, as well as incorporating this diversity in TB research, will contribute to the development of better, more specific approaches to tackle TB.

.

https://doi.org/10.1371/journal.ppat.1010490.g002

countries as Benin and Ghana [15, 16] , while L6 strains are more prevalent in the west part of West Africa, affecting countries as Guinea Bissau [17] , Sierra Leone [18] , and The Gambia [19] . L9 strains have so far only been isolated in Somalia [9] . M. africanum is an important cause of TB in West Africa, where it is estimated to cause up to half of all TB cases [20] ( Fig  2B) . A decline of the incidence of M. africanum has been reported over time for some countries [21-23], although not for others [17, 23] . It will be interesting to monitor the dynamics of M. africanum by continuous assessment of its prevalence in West Africa. A recent study has reinforced the importance of M. africanum strains as a cause of TB in the Volta region of Ghana, particularly in less cosmopolitan areas [24] . TB caused by M. africanum strains outside West Africa, although very rare, has been described in several countries [25] [26] [27] [28] . Importantly, these cases were mostly found in migrants from endemic areas in Africa, reinforcing the idea that M. africanum may be restricted to certain host ancestries [29] . However, exceptions to this have been reported. In a study in California, 2 of 5 patients diagnosed with M. africanum did not present epidemiologic relation to M. africanum endemic areas [30] . Similarly, in Canada, of the 4 M. africanum TB cases reported between 2004 to 2015 [31], 2 of them were of Canadian-born patients with no travel history to West Africa [31] . Additionally, in 2020, a case of M. africanum in a Brazilian woman without association with Africa was reported in Brazil [32] . Although not all these reports provided whole-genome sequence analysis to assess the existence of transmission clusters, they do evoke the possibility of local transmission of M. africanum outside endemic areas. It would be very interesting to dissect the host genetic and immune characteristics across these different cases, in search for potential susceptibility factors associated with M. africanum infections.

M. africanum strains were initially divided into 2 main subtypes based on their geographic origin: M. africanum subtype I, for strains originating from West Africa and M. africanum subtype II for strains from East Africa [33] . Subsequent genetic studies placed strains of M africanum subtype II, also known as the "Uganda genotype," as belonging to a sublineage of L4, thus leaving M. africanum strains restricted to West Africa [33, 34] . More recently, comparative genomics approaches established the presence or absence of specific genomic regions from the common ancestor in M. tuberculosis and M. africanum, allowing M. africanum strains to be subdivided into 2 genetically distinct lineages: L5 and L6 (previously known as Maf1 and Maf2, respectively) [12, 35] (Fig 3) . M. tuberculosis and M. africanum strains are readily distinguishable by the deletion of the region of difference (RD) 9 in the genome of M. africanum strains. The genome of strains within L5 has a further deletion in RD711, while genomes of strains within L6 are characterized by a deletion of RD702 [35] . Through spoligotyping, L5 and L6 can be respectively identified by the loss of spacers 8 to 12 and 37 to 39 (L5), and 7 to 9 and 39 (L6) in their genomes [19, 35] . Together, these genomic deletions can be used as phylogenetic markers distinguishing M. africanum from M. tuberculosis strains and allowing the genomic identification of L5 and L6 strains (Fig 3) . More recently, with the increasing availability of multilocus and whole-genome sequencing and analyses, several specific SNPs in the genome of strains belonging to L5 and L6 have been identified, which are the basis of SNP assays for rapid identification of strains belonging to L5/L6 M. africanum lineages [36,37]. Whole-genome analyses also revealed the existence of genetic diversity within M. africanum L5 and L6 strains [9, [38] [39] [40] and allowed the construction of detailed MTBC phylogeny trees. It is interesting to note that strains of the M. africanum L6 are separated from the other lineages of human-adapted members of the MTBC and instead placed among the animal lineages of the MTBC (Fig 1) [41]. Furthermore, several studies support the idea that the dassie bacillus shared a common ancestor with M. africanum L6 [41, 42] . In all, data are compatible with an evolutionary scenario in which the ancestor of L6 strains was a generalist pathogen that subsequently adapted to different host species, and with the hypothesis that L6 strains may originate from an animal reservoir, as further discussed below.

The most recently proposed M. africanum lineage, L9, resulted from the analysis of 675 M. africanum genomes, of which 5 could not be classified into any of the known MTBC lineages [9] . L9 appears as a sister clade of L6, being placed between L6 and the animal-adapted lineages. Genomes of strains belonging to L9 share some genomic deletions with those of strains belonging to L6, as RD702, but not others also present in genomes of strains of animal-associated lineages, such as RD1 and RD5. Owing to its distinct geographic location and the still significative genetic separation from L5/L6 strains genomes, a new M. africanum lineage was thus proposed [9] . Of note, phylogenetic markers to identify L9 based on genomic deletions have yet not been described (Fig 3) . Instead, a list of SNPs common to all L9 genomes and absent from other lineages was proposed as phylogenetic markers for L9 [9] . This can now be used as a framework for the discovery of other strains within L9, as well as for further functional studies of this lineage.

Given the close phylogenomic proximity of M. tuberculosis and M. africanum strains these 2 pathogens may be considered as belonging to the same species [43, 44] , which is still a matter of debate. Of note, the phylogeographic distribution and the host preference of the different members of the MTBC suggest that this complex could represent host-adapted ecotypes [45] . Also, there are important molecular and phenotypical differences between M. tuberculosis sensu stricto and M. africanum (Table 1) . Initial studies showed that unlike M. tuberculosis, M. africanum strains are unable to use glycerol as a sole carbon source, a characteristic similar to M. bovis, which implies the need to supplement the culture medium with pyruvate in . Therefore, globally, L5 and L6 strains present several characteristics that are distinctive of M. tuberculosis. No phenotypic data pertaining L9 strains exist thus far, and so it is as yet not possible to include L9 strains in these comparisons. Hence, the classification of TB-causing isolates into the different lineages of M. tuberculosis or M. africanum based in genotypic differences (Fig 3) is more accurate than one based on phenotypic assays, also because one cannot exclude phenotypic diversity among isolates of the same lineage or even sublineage.

At the genetic level, M. africanum strains present important variations in some "classical" virulence factors of M. tuberculosis. An example are the DosR regulon genes, which have been linked to virulence in the MTBC strains, and shown to be down-regulated in the sputum of M. africanum L6-infected patients as compared to infections by M. tuberculosis L4 [53]. M. africanum strains accumulate mutations in major DosR regulon genes, and the lower expression of the DosR regulon may explain why M. africanum strains present microaerobic growth and associates with extrapulmonary disease [53] . Strains of the M. africanum lineages were also described to accumulate mutations in genes encoding enzymes of the electron transport chain 59] . Because this system is involved in several pathogenic processes, as the secretion of the virulence factor ESAT-6, biosynthesis of acyltrehalose-based lipids, and the modulation of antigen export, such mutations were expected to render the pathogen avirulent. However, M. africanum L6 strains evolved to compensate the deleterious effects of the phoP/R mutations and so maintain their pathogenic ability. Indeed, the RD8-specific deletion in both animal-adapted and M. africanum L6 strains restores ESAT-6 secretion by a phoP/R-independent mechanism, by ensuring high levels of expression of the operon espACD, which is required for ESAT-6 secretion [58, 59] . How this occurs is not fully understood, but it is possible that polymorphisms upstream the espA gene might increase the affinity of PhoP or EspR for this promoter region, which results in espA expression in the absence of a fully functional phoP/R system [58] . In a different study, M. africanum L5 isolates were also shown to efficiently secrete and induce immune responses against ESX-1 substrates [38].

Several studies have been performed with the aim of unveiling associations between clinical and epidemiologic data and the infecting bacteria, i.e., M. tuberculosis or M. africanum. It is important to note that these studies have been performed in different countries, at different times, and using different methodologies. Therefore, multicentric studies are still in need, and it is somehow not surprising that some discrepant results are seen across different reports, as further discussed below. No marked differences were found in the chest X-ray presentation of TB caused by M. africanum or M. tuberculosis [60] . Furthermore, both pathogens were shown to respond similarly to the standard 4 first-line drugs in TB treatment, although patients diseased with M. tuberculosis L4 strains responded faster to TB treatment than those with M. africanum L6 strains [61] . The slow clinical recovery of M. africanum-infected patients as compared to M. tuberculosisinfected ones may result from a higher content of persister-like M. africanum bacilli in sputum at diagnosis [62] . Despite the similar clinical presentation of TB, several studies associated infections with M. africanum strains with more vulnerable hosts. Studies conducted in The Gambia found M. africanum infections to be more common in HIV-coinfected patients, as well as in older individuals and individuals presenting severe malnutrition [60] . The association between M. africanum infection and elder patients was also reported in Ghana [63] and with lower body mass index individuals in Mali [64] . However, a clear association between M. africanum and HIV coinfection is still controversial. A recent study in Ghana showed no significant differences between the prevalence of M. tuberculosis or M. africanum infections in individuals with diabetes, another important comorbidity in TB [65] . Furthermore, it is possible that M. africanum infections correlate with slower progression to active TB. A study from The Gambia showed that despite similar rates of transmission, individuals exposed to M. tuberculosis strains were more likely to progress to active TB disease than those infected with M. africanum ones [66] . This was supported by another study in Mali associating a longer time between symptom onset and TB diagnosis in M. africanum infections [64] . In line with these findings, infection of mouse models with M. africanum strains showed a slower progression of the disease [51, 67, 68] with mild lung pathology even in mice lacking IFN-γ, which are highly susceptible to M. tuberculosis infection [51] . Of note, whether transmission rates are equivalent between M. tuberculosis and M. africanum is not fully set, as a study in Ghana associated M. africanum strains with reduced recent transmission rates [69] . Importantly, whereas in The Gambia, the prevalent M. africanum lineage is L6 [66] , in Ghana, it is L5 [69] , and so differences in transmission rates may reflect the specific characteristics of L5 or L6 strains. Thus, all in all, M. africanum strains present several differences when compared to M. tuberculosis ones (Table 1 ) and is generally viewed as a less virulent pathogen than M. tuberculosis. As mentioned before, it is as yet not possible to establish comparisons between the recently identified L9 strains and those of L5/L6 or M. tuberculosis.

Infection of human monocyte-derived macrophages with distinct strains of the MTBC showed variation of the induced cytokine responses including between the 2 isolates of M. africanum tested, with 1 inducing strong cytokine release and another inducing a weak response [70] . Interestingly, in the same study, both M. africanum isolates seemed to grow less inside resting macrophages than their M. tuberculosis counterparts [70] . Another study, focusing on the pathogen transcriptional adaptation upon macrophage infection, reported distinct MTBC lineage signatures, including the failure of M. africanum strains to induce the phthiocerol dimycocerosate (PDIM) locus, a complex cell wall lipid unique to mycobacteria associated with its virulence [71] . More recently, an isolate of M. africanum L6 was shown to induce considerably less IFN-β by infected bone marrow-derived macrophages than M. tuberculosis strains from L2 or L4 [72] . Although the M. africanum isolate also triggered cGAS and STING, infections of macrophages by this pathogen induced less mitochondrial stress, thus decreased production of mitochondrial reactive oxygen species that contributed to less type I IFN being produced [72] . The in vivo effect of IFN-αβ signalling during infection by M. africanum strains was subsequently studied in mouse models. In agreement with the detrimental role of type I IFN in TB [73] , lack of type I IFN receptor signaling led to reduced lung bacterial burdens and less severe histopathological findings upon M. africanum infection [67] . These results highlight that even the lowest levels of IFN-αβ induced during chronic M. africanum infection are potentially pathogenic [67] . Collectively, these studies and others [51] demonstrate that M. africanum strains infect macrophages, inducing a cytokine response, while adapting to the host cell. The molecular mechanisms underlying these responses, such as the recognition of M. africanum strains by pattern recognition receptors, remain however elusive. M. africanum strains were shown to bind recombinant human mannose-binding lectin (MBL), a plasma opsonin, more efficiently than M. tuberculosis strains and a protective association between TB and the human MBL2 G57E variant, associated with lower MBL levels, was described, only in TB caused by M. africanum [74] . It is possible that the stronger binding of M. africanum strains to MBL may favour the bacteria uptake by macrophages, promoting the establishment of infection in vivo, and thus the protective MBL deficiency may have been selected in the human population in regions endemic for M. africanum. Another study has identified increased levels of TLR9 expression in unstimulated blood of patients infected with M. africanum isolates as compared to other MTBC strains infections [75] , which may suggest a role for TLR9 in innate immune responses to M. africanum strains. Of note, the levels of IL-12p70 and IL12A were also significantly higher in M. africanum-infected patients, while those of IL-15, IL8, and MIP-1α were higher in M. tuberculosis-infected patients [75] . A broader study comparing peripheral blood gene expression profiles between M. africanum-and M. tuberculosisinfected patients showed no differences at diagnosis, although there were distinct signatures associated with each infection posttreatment, predominantly associated with immune responses and metabolic diseases [76] .

Comparison of T cell responses from M. tuberculosis-or M. africanum-infected TB patients before chemotherapy and following overnight stimulation of whole blood with ESAT-6/CFP-10 or with purified protein derivative (PPD) showed higher single-TNF-α-producing CD4 and CD8 T cells and lower single-IL-2-producing T cells in the case of M. africanum infections [77] . Additionally, a persistently high proportion of activated T cells was reported in M. africanum-infected individuals posttreatment [77] . However, the frequencies of PPD-specific polyfunctional CD4 T cells did not differ between the 2 infections [77] , both before and after treatment, suggesting an overall uniform immune response triggered by either pathogen. This is in line with the abovementioned studies on peripheral blood transcriptomic and metabolic profiles obtained at diagnosis [76] . Interestingly, stimulation of whole blood with ESAT-6/ CFP-10 stimulation after treatment induced significantly higher production of pro-inflammatory markers, such as IFN-γ, in the case of M. tuberculosis-infected TB patients [75, 76] . In the mouse model of infection, a modest immune response has been reported upon infection with a M. africanum isolate, also associated with restricted lung pathology [51] . Taken all this together, it is possible that a lower immune response takes place upon M. africanum infection, which although precluding the clearance of the pathogen, may protect the host from tissue immune pathology. This hypothesis is compatible with a slower progressing infection and may be explained by pathogen-associated factors. Pathogens belonging to the MTBC are known to have remarkably hyperconserved T cell epitopes, suggesting that ensuring T cell responses is more important to these agents than evading them [78] . Interestingly, a study showed that the L6 strains of M. africanum were significantly more genetically diverse than the L5 ones, including in predicted T cell epitopes [79] . Additionally, even though the majority of the T cell epitopes were conserved between the 2 lineages, a higher ratio of nonsynonymous to synonymous single nucleotide variation was detected in the epitopes from L6 strains relatively to L5 ones [39] . Thus, it is tempting to speculate that the evolutive pressure to hyperconserve T cell epitopes may be weaker in the case of L6 strains, leading to lower T cell responses and favouring the persistence of the pathogen in its host population. Further studies are required to address these hypotheses and investigate the contribution of T cell responses to TB caused by M. africanum strains.

A specific adaptation of M. africanum to the host population, particularly to the host immune response, is a conceivable hypothesis to explain the geographic restriction of this pathogen. Previous studies provide compelling evidence towards this hypothesis in the case of M. africanum L5 strains. In a study in Ghana, M. africanum was significantly more common in TB patients belonging to the Ewe ethnic group an association that was mainly driven by L5 strains [80] . Possible interactions between M. africanum infection and human genetic diversity were also described in other studies. A polymorphism in the exonic allele (g.760A) of the ALOX5 gene (which encodes for 5-lipoxygenase, an important regulator of the immune response in TB [81, 82] ) was associated with higher risk of TB in Ghana, an association that was stronger in infections caused by M. africanum L6 strains [83] . Furthermore, another study identified a highly frequent variant of the human immunity-related GTPase M (IGRM), a regulator of the autophagic process, in the Ghanaian population and associated it with protection against M. tuberculosis L4 strains, but not against M. africanum isolates [84] . Thus, higher frequencies of genetic variants conferring increased susceptibility to M. africanum strains in West African individuals may at least partially explain the geographical restriction of this pathogen to this region. Still, more studies linking human and pathogen genetic diversities are needed to validate this hypothesis. In this line, specific HLA genetic associations may be of potential interest to explain the geographic distribution of M. africanum versus M. tuberculosis infections. A study conducted in Mali identified various class I HLA-A and HLA-B alleles associated with active TB disease caused by either pathogen. However, several class II HLA-DR variants were found to be associated with M. tuberculosis but not M. africanum strains, with only the variant DRB1 � 03:01 being associated with both groups [85] . It is tempting to speculate that specific associations between HLA variants and M. africanum strains may reflect variations affecting T cell epitopes in M. africanum, which as described above are not as hyperconserved as in M. tuberculosis.

More recently, the hypothesis that differences in the intestinal microbiota of patients infected with M. africanum isolates could contribute to the high susceptibility of West African individuals to infections with this pathogen has been proposed [86] . Patients infected with M. africanum strains presented less microbiome diversity than individuals infected with M. tuberculosis isolates or healthy controls and were enriched in bacteria from the Enterobacteriaceae phylum Proteobacteria as compared to healthy controls [86] . Since a positive correlation between the abundance of Enterobacteriaceae and an inflammatory gene expression profile was reported, differences in the intestinal microbiome may contribute as host-associated factors predisposing to infections by M. africanum.

Other 2 hypotheses may explain the geographic restriction of M. africanum infections, which are less related to the host immune response. It is possible that M. africanum is an attenuated member of the human-adapted TB-causing bacteria, being therefore outcompeted by M. tuberculosis. This hypothesis is supported by the reduction of TB cases caused by M. africanum strains over time for some countries [21-23], although it is not observed in other countries [17,23] as discussed above. Finally, M. africanum L6 strains share a common ancestor with animal (nonhuman) adapted strains (Fig 1) . Animal-adapted lineages are composed of Mycobacteria that infect different species of animals as preferential hosts, including nonhuman primates and other mammals. The ancestry of M. africanum L6 allows raising the hypothesis that, despite having become a human pathogen, strains of this lineage may still be adapted to an animal that could function as a reservoir in West Africa. Although M. africanum strains have been isolated from several animal species, including pigs and cows [87] , an animal reservoir has never been identified. Of note, the animal reservoir hypothesis is less likely to prove valid in the case of L5 strains since this lineage is phylogenetically less related to animal adapted members of MTBC [41] (Fig 1) .

Up to 50% of the TB cases in West Africa have been attributed to M. africanum strains. A striking feature characterizing these TB-causing bacteria is its geographical restriction, which contrasts with the widespread distribution of M. tuberculosis strains and remains largely unexplained. Adaptation of M. africanum strains to the West African population, perhaps mediated through differential modulation of the immune response, is a likely hypothesis. Importantly, infections of humans and experimental models with strains of M. africanum are generally more attenuated than those with M. tuberculosis strains. This offers an opportunity to learn from M. africanum and its interactions with the host, with the aim of better controlling M. tuberculosis. There are several outstanding questions that would advance our knowledge in this field towards better strategies to tackle TB: 5. Are there differences in latency establishment, duration, or reactivation between M. tuberculosis and M. africanum strains? Elucidating this question is hampered by our inability to categorize the pathogen species in latent cases. However, by revealing immune signatures of M. africanum infections, it may be possible to then look at latent and progressing cohorts in an attempt to establish latent signatures specific of the different MTBC members. This will further our knowledge on the natural history of TB, again potentially offering novel targets to control TB.

Answering these questions to understand the differences between M. africanum and M. tuberculosis strains will provide valuable knowledge towards identifying the cellular and molecular determinants allowing the widespread of the M. tuberculosis strain lineages, which are a global threat. This knowledge will also advance our understanding on the biology of M. africanum and its interactions with the human host, which is highly relevant considering the TB burden in West Africa. Furthermore, consistently stratifying for the type of infecting bacteria in human-based studies will contribute to a better interpretation of novel TB intervention tools, including diagnosis and vaccines. All this will in turn inform the development of better, more specific approaches to tackle TB. 

Human and tuberculosis co-evolution: An integrative view

Global Tuberculosis Report-World Health Organization

The potential impact of COVID-19-related disruption on tuberculosis burden

Global Epidemiology of Tuberculosis

Co-evolution of Mycobacterium tuberculosis and Homo sapiens

Ecology and evolution of Mycobacterium tuberculosis

A sister lineage of the Mycobacterium tuberculosis complex discovered in the African Great Lakes region

Phylogenomics of Mycobacterium africanum reveals a new lineage and a complex evolutionary history

Mycobacterium tuberculosis lineage 4 comprises globally distributed and geographically restricted sublineages

A new evolutionary scenario for the Mycobacterium tuberculosis complex

Variable host-pathogen compatibility in Mycobacterium tuberculosis

Out-of-Africa migration and Neolithic coexpansion of Mycobacterium tuberculosis with modern humans

Mycobacterial lineages causing pulmonary and extrapulmonary tuberculosis

Spatio-temporal distribution of Mycobacterium tuberculosis complex strains in Ghana

The First Phylogeographic population structure and analysis of transmission dynamics of M. africanum West African 1-Combining molecular data from Benin, Nigeria and Sierra Leone

Evolutionary history of tuberculosis shaped by conserved mutations in the PhoPR virulence regulator

Differences between TB cases infected with M. africanum, West-African type 2, relative to Euro-American M. tuberculosis-an update

Mycobacterium africanum (Lineage 6) shows slower sputum smear conversion on tuberculosis treatment than Mycobacterium tuberculosis

Association of slow recovery of Mycobacterium africanum-infected patients posttreatment with high content of Persister-Like bacilli in pretreatment sputum

Molecular epidemiology and drug susceptibility profiles of Mycobacterium tuberculosis complex isolates from Northern Ghana

Association of Mycobacterium africanum infection with slower disease progression compared with Mycobacterium tuberculosis in Malian patients with tuberculosis

TB-diabetes co-morbidity in Ghana: The importance of Mycobacterium africanum infection

Progression to active tuberculosis, but not transmission, varies by Mycobacterium tuberculosis lineage in The Gambia

Type I interferon is pathogenic during chronic Mycobacterium africanum infection

Impaired fitness of Mycobacterium africanum despite secretion of ESAT-6

Reduced transmission of Mycobacterium africanum compared to Mycobacterium tuberculosis in urban West Africa

Clade-specific virulence patterns of Mycobacterium tuberculosis complex strains in human primary macrophages and aerogenically infected mice

Functional genetic diversity among Mycobacterium tuberculosis complex clinical isolates: delineation of conserved core and lineage-specific transcriptomes during intracellular survival

The mechanism for type I Interferon induction by Mycobacterium tuberculosis is bacterial strain-dependent

Type I interferons in tuberculosis: Foe and occasionally friend

Variant G57E of mannose binding lectin associated with protection against tuberculosis caused by Mycobacterium africanum but not by M. tuberculosis

Host immune responses differ between M. africanum-and M. tuberculosis-infected patients following standard anti-tuberculosis treatment

Differential transcriptomic and metabolic profiles of M. africanum-and M. tuberculosis-infected patients after, but not before, drug treatment

Differences in T-cell responses between Mycobacterium tuberculosis and Mycobacterium africanum-infected patients

Human T cell epitopes of Mycobacterium tuberculosis are evolutionarily hyperconserved

tuberculosis T cell epitope analysis reveals paucity of antigenic aariation and identifies rare variable TB antigens

Mycobacterium africanum is associated with patient ethnicity in Ghana

Host control of Mycobacterium tuberculosis is regulated by 5-lipoxygenase-dependent lipoxin production

Eicosanoid pathways regulate adaptive immunity to Mycobacterium tuberculosis

ALOX5 variants associated with susceptibility to human pulmonary tuberculosis

Autophagy gene variant IRGM -261T contributes to protection from tuberculosis caused by Mycobacterium tuberculosis but not by M. africanum strains

Differential HLA Allele Frequency in M. africanum vs. M. tuberculosis in Mali

Patients infected with Mycobacterium africanum versus Mycobacterium tuberculosis possess distinct intestinal microbiota

An outbreak of tuberculosis in pigs and cattle caused by Mycobacterium africanum