The emergence of Beijing family genotypes of Mycobacterium tuberculosis and low-level protection by bacille Calmette–Guérin (BCG) vaccines: is there a link?

Abstract

The world is confronted with major tuberculosis (TB) outbreaks at a time when the protection of bacillus Calmette–Guérin (BCG) vaccine has become inconsistent and controversial. Major TB outbreaks are caused by a group of genetically similar strains of Mycobacterium tuberculosis (Mtb) strains, including the Beijing family genotypes. The Beijing family genotypes exhibit important pathogenic features such high virulence, multi-drug resistance and exogenous reinfection. These family strains have developed mechanisms that modulate/suppress immune responses by the host, such as inhibition of apoptosis of infected macrophages, diminished production of interleukin (IL)-2, interferon (IFN)-γ, tumour necrosis factor (TNF)-α and elevated levels of IL-10 and IL-18. They demonstrate distinct expression of proteins, such as several species of α-crystallin (a known Mtb virulence factor), but decreased expression of some antigens such as heat shock protein of 65 kDa, phosphate transport subunit S and a 47-kDa protein. In addition, the Beijing family strains specifically produce a highly bioactive lipid (a polyketide synthase)-derived phenolic glycolipid. This altered expression of proteins/glycolipids may be important factors underlying the success of the Beijing family strains. The Beijing family strains are speculated to have originated from South-east Asia, where BCG vaccination has been used for more than 60 years. The hypothesis that mass BCG vaccination may have been a selective factor that favoured genotypic and phenotypic characteristic acquired by the Beijing family strains is discussed.

Introduction

The bacille Calmette–Guérin (BCG) vaccine, developed a century ago, remains one of the most widely used vaccines globally. Vaccination at birth with BCG is widely applied as part of the Expanded Programme on Immunization of the WHO and billions of people have been vaccinated since 1921 [1]. Except for tuberculous meningitis in children, the capacity of BCG to protect individuals against tuberculosis (TB) is debated, because randomized clinical trials have provided estimates ranging from 80% to no protection [2,3].

Several explanations have been suggested for these variations in the protective efficacy of BCG: differences among vaccines, interaction of BCG with environmental mycobacteria, nutritional or genetic differences in trial populations, or differences in trial methodologies [2,4–6]. However, in general there is a lack of compelling evidence in favour of any of the proposed mechanisms.

In the past, one dogma that dictated all TB research was the assumption that TB is caused by a single strain of Mycobacterium tuberculosis (Mtb). However, this view has changed since the introduction of DNA fingerprinting in 1990s. To date, we know that TB is caused by different Mtb strains with varying degrees of virulence. The Beijing family is one of the strains associated most often with major TB outbreaks globally.

It is speculated that the Beijing family strain originated in China and spread to neighbouring countries [7]. Molecular epidemiological studies have shown that these strains are distributed worldwide and are able to spread in large clonal clusters [8]. The nature of the force(s) that contributed to the selection and dissemination of strains of the Beijing family is not known. However, the fact that BCG mass vaccination was a factor common to all countries in South-east Asia has led to the hypothesis that BCG might be a selective force that favoured genetic and phenotypic changes which contributed to the spread of these strains. To date, there is a wealth of information pertaining to genetic epidemiological and immunological aspects of the Beijing strains. In this paper, we will discuss current observations that reinforce directly or indirectly the hypothesis that BCG vaccination has been a selective force that favoured the emergence of the Beijing family strains. Particular emphasis will be given to observations that looked into phenotypic factors involved in modulating or suppressing immune responses of the host.

BCG mass vaccination

After the reports of successful BCG vaccination in 1921, substrains of BCG were distributed to different countries for mass vaccination trials. Although formal evaluation of BCG vaccination started in the 1930s, variations in efficacies were noted only in the 1950s. Striking differences had emerged between a major trial in the United Kingdom by the British Medical Research Council (MRC), which showed more than 75% protection, and in trials by the US Public Health Services in Georgia, Alabama and Puerto Rico, which recorded less than 30% protection. Subsequent trials and studies have resulted in a broad range of protective efficacies of the BCG [2]. Moreover, the maximum level of protective efficacy registered in clinical trials of BCG vaccines was 80% [9]. These results imply that a subset of the study population cannot mount a protective immune response against Mtb via immunization or natural immunity. Alternatively, BCG confers protection against some strains of Mtb but not against others.

There was general agreement that BCG vaccination with a potent strain, when given to previously uninfected subjects, is highly effective in preventing the development of TB. This direct effect may be measured in practice in terms of the proportion of cases prevented in the age groups in which the vaccine has been administered. It is also assumed that mass BCG vaccination, especially at school-leaving age, can be expected to yield benefits not only directly, but also indirectly, by breaking the chain of transmission and so preventing the development of TB in unvaccinated subjects. However, studies that compared Norway, Denmark and the Netherlands show that countries that used BCG vaccination (Norway and Denmark) as part of their TB control strategy did not experience any greater reductions in the incidence or mortality rates from TB than countries such as the Netherlands that did not use BCG [10].

As indicated earlier, several factors have been enumerated as factors responsible for variations in the efficacy of BCG [2,5,6,11,12]. The hypothesis that has gained the most widespread acceptance attributes this variation to higher levels of exposure to environmental mycobacteria in areas where the vaccines confer no protection. However, the precise nature of this interference is not known. Based on animal experiments, it has been suggested that the protection provided by environmental mycobacteria partly masks the effect of subsequent BCG vaccination [13], or that environmental mycobacteria have a direct antagonist influence and skew the immune response towards a Th2 direction [14]. However, arguments that heterogeneous protection imparted by environmental mycobacteria masks protection by BCG are inconsistent with the observation that protection afforded by BCG against TB was independent of prior tuberculin status according to some trials, such as that of Puerto Rico [15,16].

Another prominent factor believed to be the cause of variations in the efficacy of BCG is differences in BCG strains. However, the fact that similar protective efficacies with different vaccines, and vice versa, have been observed in trials makes it an unlikely explanation. Prominent examples are provided by Glaxo freeze-dried BCG, which gave good protection against TB in England but not in Malawi, and the Danish BCG, which performed well in the original British trials but provided very little in south India [5]. Host-related differences, such as nutritional, genetic and host immune status, obviously can, to some extent, result in variations of BCG efficacies between populations; however, it is difficult to conceive such huge differences as 80% protection and total lack of protection due to these differences.

To date, an area in which the use of BCG is highly acknowledged is in the protection of childhood TB, particularly in the protection of children against tuberculous meningitis [3]. However, in a study in China, discontinuation of BCG vaccination had no apparent harmful effect on the population in terms of an increase in cases of tuberculous meningitis in children, even though vaccination was discontinued at a higher level of TB notification rate [17]. On the other hand, several studies have indicated high protective efficacy of BCG against tuberculous meningitis in children, notably in Argentina, Brazil and India, and against miliary tuberculosis in Argentina, Burma, Indonesia and Papua New Guinea, reviewed in Trunz et al. [18]. In addition, there are studies that showed high protective efficacy of BCG against pulmonary disease, and even against Mtb infection [19–22]. A study from Jordan showed that BCG had a protective efficacy of 85% against pulmonary TB and 95% protection against extra-pulmonary TB [19]. A study conducted in subtropical Australia indicated that BCG confers higher protection against newly acquired disease than against disease due to endogenous reactivation [20]. A study from Brazil indicated that BCG conferred 69% protection against transmission of TB to close contacts of patients with multi-drug-resistant TB [21]. More importantly, a study from Turkey indicated that BCG vaccination protects not only against disease but also against Mtb infection [22].

BCG vaccine has been used extensively in some South American countries, for instance in Brazil, since at least 1929. One would normally assume a similar trend in the distribution of the Beijing strain as found in South-east Asia, where mass BCG is implicated in the emergence of the Beijing strain. However, the distribution of the Beijing strain is low or absent in South American countries. Two possible explanations for the observed trend may be: (1) the available data on the distribution of strains in these countries are incomplete, and we do not know, for certain, whether or not other new successful strains are already circulating in some communities; and (2) transmission of TB in communities in South American countries may be low compared to South-east Asia, in which case clonal expansion of new strains may take a longer time compared to communities of high transmission.

The fact that BCG is a potent stimulant of Th1-type immunity has extended its use in combating many diseases, such as bladder cancer, skin melanoma and leprosy. More importantly, it is used as a vaccine against human immunodeficiency virus (HIV) infections, and as an adjuvant for the development of vaccines against microbial and parasitic infections, including leishmaniasis and schistosomiasis. Again, it is hard to comprehend why a vaccine used as an adjuvant for developing vaccines for unrelated diseases fails to prevent TB. Although it is difficult to attribute the failures of BCG to the emergence of the Beijing family genotype alone, it is also difficult to rule out the possibility that differences in the strains of Mtb might have contributed to differences in the efficacies of BCG vaccines.

Beijing family strains

Genetic studies

Organisms grouped in the Mycobacterium complex include M. tuberculosis, M. africanum and M. canetti, which are exclusively human pathogens; M. microti, which is a rodent pathogen, and M. bovis, which has a wide host spectrum. However, they are characterized by 99·9% similarity at the nucleotide level and identical IS6110 sequences [23,24]. Studies conducted to determine genetic factors underlying host specificity, pathogenesis and global epidemiology of TB show that Mtb strains are classified into ‘modern’ and ‘ancestral’ strains, based on Mtb specific deletions (TbD1). The ‘modern’ Mtb strains include representatives of major epidemics such as the Beijing, Haarlem and East African Mtb genetic families, whereas the rest are classified under the ‘ancestral’ strain [25].

The Beijing family strains share genetic markers such as similar IS6110 RFLP patterns and spoligotyping [26,27]. Currently, the Beijing strains are identified principally by the number of spacers in the direct repeat (DR) region of Mtb genome, namely spoligotype S00034 or STI, which is characterized by the deletions of spacers 1–34 and the presence of most of the spacers 35–43 [7]. Comparative whole-genome hybridization of Beijing/W strains have shown that there are large sequence polymorphisms (LSPs) which subdivide the Beijing/W family into at least four subgroups, raising the possibility that there are phenotypic differences within the Beijing/W family [28].

It is believed that the Beijing family strains might have diverged from a common ancestor in the recent past [7]. Although the exact date of their divergence from a common ancestor is not known, it has been estimated to be less than a century since the clonal dissemination of the Beijing in some communities. The Beijing family strains were first identified from the People's Republic of China and Mongolia, and are highly prevalent in other Asian countries neighbouring China. The apparent high clonality of strains of the Beijing genotype expressed by the high mutual similarity in their IS6110 restriction fragment length polymorphism (RFLP) patterns and identical spoligotype patterns led to the hypothesis that these strains may have a selective advantage that led to clonal expansion [7]. Reasons for the predominance of a narrow range of genotypes may include limited contact with other populations or a selective advantage of certain strains due to vaccine-induced immunity. For instance, wide-scale application of vaccines against whooping cough (Bordetella pertussis) has led to shifts in the population of circulating pathogens [29]. A factor which is common to all countries in South-east Asia is BCG vaccination, which has been used for the past 2–6 decades.

Recently, the fourth International Spoligotyping Database (SpolBD4), which classified 39 295 strains of Mtb from 141 countries into 62 clades/lineages, indicated that the Beijing and the Beijing-like strains represent about 50% of strains in Far East Asia and only 13% of the isolates globally [30]. It has been hypothesized that co-evolution between human beings and bacilli, and ‘vertical transmission’ (in the household), must have been the main mode of TB transmission throughout centuries and even millennia [31]. In view of this, some studies have shown a stable association between bacillary populations with their human hosts in various environments, and suggested that the hosts geographical origin is predictive of clinical isolates of TB being carried [32].

There are other studies that support the BCG selection hypothesis [7]. For instance, one study that compared genetic heterogeneity among Mtb isolates from 501 patients in Ethiopia, Tunisia and the Netherlands by analysis of DNA polymorphism driven by insertion element IS6110 indicated that the percentage of isolates displaying two or more identical patterns differed greatly in the three countries: it was highest among Tunisian isolates and lowest in Dutch isolates. In contrast to isolates from Dutch subjects infected with Mtb, the majority of strains from Ethiopia and Tunisia were from a few families of genetically highly related strains. Furthermore, little overlap was observed among isolates from the three countries, indicating strict isolation of the bacterial reservoirs in these countries. The study shows more DNA polymorphism in Ethiopia than in Tunisia, although the incidence of tuberculosis in Ethiopia is about five times that in Tunisia. Although a number of socio-demographic factors make interpretation of these results difficult, it has been suggested that vaccination in Tunisia may have favoured the selection of Mtb strain that resist BCG-induced immunity because of mass vaccination practice in Tunisia, compared to Ethiopia, where BCG vaccination is not a common practice [33].

A study in Vietnam indicated that the Beijing genotype was less associated with BCG vaccination but was frequently associated with younger age, suggestive of active transmission [34]. For the first time, a recent study described differences between ‘typical and atypical’ strains of the Beijing clade in the Netherlands, Vietnam, and Hong Kong (the term ‘Beijing clade’ is used to designate all strains with a Beijing characteristic spoligotype, in accordance with the definition of the Beijing lineage, and ‘typical’ refers to strains with a typical IS6110 RFLP pattern, containing a high number of IS6110 copies) [35]. It was shown that typical family strains among the Beijing clade isolate ranged from 74·6% in Vietnam to 86·4% in Hong Kong. Secondly, typical Beijing family isolates were found significantly less frequently among patients older than 75 years. The typical Beijing isolates were significantly more often isolated from BCG-vaccinated individuals. A reduced sensitivity to vaccine-induced immunity of the typical Beijing family strains also explains the significantly higher proportion of Beijing strains in Hong Kong, where mass BCG vaccination was started in 1952, compared to Vietnam, where BCG vaccination began only in 1980 [27].

Virulence

Studies in both humans and animal models show that among the Beijing family strains there are hyper-virulent phenotypes that are associated with relapse and treatment failure in humans and distinct immune responses in animal models [36–40]. Several studies show that there are factors underlying susceptibility of the host, as well as virulence of the mycobacteria. In general, there is evidence of multi-factorial genetic predisposition in humans that influences susceptibility to TB [41,42]. Regarding Mtb, there is association between the presence of polymorphisms in single genes, e.g. the human NRAMP gene, and TB [43], although this is not the case with some other geographical regions (e.g. Morocco) [44].

Until recently, differences in virulence between Mtb strains or genotypes had not been studied, except in animal models. In fact, until the introduction of DNA fingerprinting in the early 1990s, very limited possibilities were available to distinguish between different strains of the genetically highly conserved Mtb complex. Therefore, most of the immunological research has been based on a limited number of laboratory strains, such as H37Rv and Erdman, with H37Ra as an avirulent laboratory variant. However, tens of thousands of different Mtb complex strains have been distinguished by DNA fingerprinting techniques over the last decade [8,38,45,46]. Based on fingerprinting results, it was found that the population structure of Mtb in high prevalence areas is much more conserved than in areas with a low incidence of TB [7]. This suggests that selective advantages of predominant Mtb genotypes may play a role in the spread of TB in high incidence settings.

Animal models

The Mtb Beijing genotype family is one of the major contributors to the current TB epidemic [7,8,26,46]. The high prevalence of the Beijing genotype worldwide is indicative of the success of this Mtb strain type as a human pathogen [8]. The success may be related to selective advantages, such as advanced mechanisms to circumvent BCG-induced immunity over Mtb strains belonging to other genotypes [38]. This is supported by observations that Beijing strains are more virulent and elicit non-protective immune response compared to other genotypes, during experimental disease in animal models [38,47,48].

In a study carried out by Lopez et al. [38], it was found that the Beijing family strains were characterized by extensive pneumonia, early but ephemeral tumour necrosis factor (TNF)-α, an inducible isoform of nitric oxide synthetase (iNOS) expression and significantly higher earlier mortality, compared with H37Rv. In contrast, Canetti strains induced limited pneumonia, sustained TNF-α and iNOS expression in lungs, and almost 100% survival. Previous BCG vaccination protected less effectively against infection with the Beijing strain than against H37RV strain. The Beijing family strains consistently induced accelerated bacterial multiplication, early and massive pneumonia and death compared to the Canetti, Haarlem and Somali clades. The more severe pathology induced by the Beijing bacteria co-existed with a host immune response probably driven towards non-protective mechanisms. Although macrophages infected with the Beijing strains were initially activated and produced high expression of TNF-α and iNOS, they rapidly failed to stimulate Th1 cells efficiently enough to arrest bacillary multiplication, resulting in massive tissue damage and early mortality [38]. Similarly, Dorman et al. [48] studied survival, lung pathology, bacterial load and delayed-type hypersensitivity (DTH) responses of BALB/c mice after intratracheal infection with 19 different Mtb strains of 11 major genotype families. The results indicated that these strains varied with regard to histopathology, bacillary burden at different time-points after infection and DTH responses. A good correlation was found between the pathology, virulence and bacillary load. All five strains with the highest score for histopathology also induced the highest mortality, with death of all mice within 92 days after infection, and showed the highest bacillary burden. On the other hand, all six strains with the lowest score for histopathology caused no or only minimal mortality after 112 days and showed the lowest bacillary burden. Strains with a high histopathology score, including the Beijing, East African and Haarlem, were highly virulent as indicated by the short survival time.

A recent study that compared four clinical isolates of H37Rv (CSU93, CSU22, CSU15, CSU21), representing a range of virulence in a mouse macrophage model, indicated that those of higher virulence grew faster in macrophages; induced higher production of TNF-α and cell death by necrosis but induced minimal apoptosis [49]. Apoptosis (programmed cell death) is a distinct form of cell death that is essential for the regulation of the immune system. Fas (APO-1/CD95) [50], a member of the TNF receptor family, and its ligand (FasL) [51] play an important role in the induction of apoptosis. Apoptosis of cells infected with intracellular pathogens may benefit the host by eliminating a supportive environment for bacterial growth. Because apoptosis is involved in the killing of intracellular mycobacteria, inhibition of apoptosis may represent a strategy for mycobacterial survival [52]. Several studies support this view [53–56]. In a study that compared apoptosis of alveolar macrophages following infection with Mtb complex strains of differing virulence and by M. kansasii indicate that avirulent or attenuated bacilli (Mtb H37Rva, M. bovis BCG and M. kansasii) induced significantly more macrophage apoptosis than virulent strains (Mtb H37Rv, Erdman, Mtb clinical isolate BMC 96·1, and M. bovis wild-type). Increased apoptosis was not due to greater intracellular bacterial multiplication because virulent strains grew more rapidly in alveolar macrophages than attenuated strains. These findings suggest the existence of mycobacterial virulence determinants that modulate the apoptosis response of alveolar macrophages to intracellular infection [53].

A recent study that compared H37Rv with high virulent strain of the Beijing strain (code 9501000) in a mouse model of pulmonary TB demonstrated that the Beijing strain induced twofold lower apoptosis of activated macrophages at days 1 and 3 post-infection. A high percentage of vacuolated macrophages (8·7% for H37Rv and only 1·4% for the Beijing strain) in pneumonic areas were found at day 60 days post-infection. On the other hand, a high percentage of vacuolated macrophages expressed the anti-apoptotic molecule (Bcl-2), 83% for H37Rv and 95% for the Beijing strain [54].

Spira et al. [55] have shown that a group of pro-apoptotic genes are down-regulated after infection by virulent Mtb strain H37Rv, whereas infection with avirulent Mtb H37Rva led to a gene expression profile that would favour macrophage apoptosis. These data reveal that apoptosis-related genes are regulated differently by virulent and avirulent/attenuated Mtb strains, and are consistent with the hypothesis that virulent Mtb strains interfere with TNF death signalling. Interferon (IFN)-γ and TNF-α are known to be potent inducers of nitric oxide (NO) production, which has been proposed as an important effector molecule in apoptosis. Given the importance of TNF-α in host defence against TB, the ability to repress the expression of genes activated by TNF-α may constitute a bacillary virulence mechanism.

Human studies

The first indication of a peculiar interaction between Beijing strain and the host immune response was observed in a study in Indonesia, where patients affected by Beijing strains developed early febrile responses to treatment twice as often as patients infected by other Mtb strains [56]. In Vietnam, the Beijing strains were associated with young patients but not with BCG vaccination. With increasing age, a decreasing proportion of cases due to primary TB, indicating that the Beijing genotype is associated with transmission of TB in Vietnam. Thus, these observations suggest a recent spread of the Beijing genotype in Vietnam [34]. Moreover, this study indicated that the Beijing genotype occurs more commonly in BCG vaccinated than in non-BCG vaccinated individuals. It has been suggested that this is likely to represent a cohort effect of BCG vaccination rather than reduced sensitivity to vaccine-induced immunity of the Beijing genotype strains. Because of increasing BCG coverage in Vietnam over the last two decades, young people are more likely to be vaccinated than older people. The study of Qian et al. [57], in which spoligotype was performed on paraffin-embedded material, indicated that the Beijing genotype was presumably already prevalent in the Beijing region 30–40 years ago.

The precise mechanisms that are required for containment of Mtb during in vivo infection are poorly known. In mice, this depends on reactive oxygen and nitrogen radicals, whereas these do not seem to be the effector mechanism in humans [58,59]. In humans, granulysin in combination with perforin is believed to be one important mechanism by which the host cells kill the bacilli [60]. However, it has been shown that this protective activity fails if there is a marked release of Th2 type cytokines [61,62]. Thus, the Th1/Th2 balance is thought to determine the outcome of the encounter with the pathogen, although this view remains controversial [63,64]. However, at least in a BALB/c model of pulmonary TB, the initial phase is dominated by high production of Th1 cell cytokines, in co-existence with high levels of TNF-α and iNOS, which temporarily control the infection. Three weeks after infection, a rise in IL-4 production co-exists with a drop in cells expressing IL-2, TNF-α and iNOS [65,66].

Endogenous reactivation versus exogenous reinfection

Previously, it was assumed that post-primary TB is usually caused by reactivation of endogenous infection rather than by a new, exogenous reinfection. More specifically, it has been argued that, in countries with low incidence of TB (i.e. industrialized countries), a combination of host immunity and limited exposure to Mtb made endogenous reactivation a more plausible explanation for recrudescence of disease [67]. Mycobacterial genotyping techniques demonstrate that exogenous reinfection plays an important role in the pathogenesis of post-primary TB in adults, both in areas of high incidence [68,69] and that of low incidence of TB, including subpopulations in industrialized countries [70].

Using DNA fingerprinting, it has been shown that exogenous reinfection has been a major cause of post-primary TB after a primary cure in an area with a high incidence of TB [68]. A study carried out in South Africa shows that 75% of recurrent cases had developed disease as a consequence of exogenous reinfection [71]. Moreover, Warren et al. [9] have shown that 19% of patients studied in Cape Town, South Africa were infected simultaneously with Beijing and non-Beijing strains, and 57% of patients infected with the Beijing strains were also infected with non-Beijing strains. Multiple infections were more frequent in retreatment cases compared with new cases.

In another study in Cape Town, it was shown that cases of reactivation in HIV-negative individuals occurred as little as 7 or 8 months after previous cure. These results suggest that in immunocompetent people living in TB endemic areas, reinfection and progression to active disease may occur at any time after treatment has been discontinued [71].

The above observation is not consistent with the widely held view that in immunocompetent people, reinfection is rare during the first 2–5 years after a first infection because of the immune response mounted against Mtb antigens that develop after primary infection [67]. The observations in Vietnam and South Africa imply that the host mounts immune responses that are inadequate to contain mixed infections, either because the previous infection will render the host susceptible to superinfection or because there are resistant strains that suppress/modulate immune responses of the host. In people infected with HIV, it has been shown that reinfection can occur not only years after a previous infection but even during treatment for active TB [72].

What does this imply?

Active TB reflects an inefficient host immune response against Mtb, as most infected people mount a protective immunity and only 5–10% develop active disease. The mechanisms underlying susceptibility to TB presumably involve immunosuppression (for example, as HIV patients) and/or genetic predisposition of the host [73,74]. Although the exact mechanism is yet to be elucidated, one important feature of multi-drug resistant strain of Mtb, specifically the Beijing family strain, is the capacity to alter immune responses of the host.

Immune evasion mechanisms

The interplay between mycobacteria and macrophages seems to be a crucial factor determining the outcome of infection. M. tuberculosis persists within macrophages through a variety of immune evasion mechanisms, including prevention of the recognition of infected macrophages by T cells, evading the macrophage killing mechanism and prevention of phagosome maturation into an acidic hydrolytic compartment with microbicidal activity [75–78].

Proteins actively secreted (e.g. ESAT-6, CFP-10) by Mtb are partly responsible for evasion of the immune defence and complications ensuing disease in susceptible individuals [76]. Earlier, it has been speculated that strains belonging to the Beijing family have a genetic advantage to cause disease and that the wide dispersion of this family compared to other less prevalent clinical isolates may be related to differential protein expression [79]. Recent studies demonstrated that differential protein expression is one of the factors underlying heterogeneous immune responses to Mtb infection [80–82]. In a study that compared protein expression of two clinical strains belonging to the Beijing family and F23 (family 23) from South Africa with H37Rv strain has revealed a number of differences in antigen expression levels across the three strains, with several changes being particular to the Beijing strain. Several species of α-crystallin, an Mtb virulence factor [83–85], were more highly expressed in the Beijing strain compared to H37Rv and F23. Thus, altered expression of this protein may be a factor contributing to the virulence of the Beijing family strains.

Other mechanisms used by the Beijing family strains for immune evasion include decreased expression of Hsp65 (heat shock protein of 65 kDa) and PstS (phosphate transport subunit S) and the 47 kDa protein. It has been suggested previously that reduced expression of certain major antigens may allow strains to evade the host immune responses [81]. PstS1 is a 38-kDa Mtb complex-specific phosphate-binding lipoprotein, and a known B and T cell stimulant [86]. Earlier studies have shown that PstS1 was promising for serodiagnosis of TB, with sensitivities of 70% and 73% for smear-negative and extra-pulmonary TB patients, respectively [82]. However, in a recent evaluation of commercially available tests the highest sensitivity achieved with PstS1 was 55% [87]. The absence of PstS1 antibodies in some TB patients was attributed to decreased expression of PstS1 in some clinical strains, as observed for the Beijing family strains [80]. Put together, the above results suggest that altered expression of proteins during infections by the Beijing family strains may be an important factor contributing to the virulence and/or suppression of immune responses by the host. Thus, altered gene/protein expression by the Beijing family strain and the associated immune modulation/suppression may have an implication on the efficacy of BCG.

Recently, a recombinant BCG mutant that secretes listeriolysin of Listeria monocytogenes has yielded a profound potential in protecting against the Beijing genotype family of Mtb in mice [88]. Listeriolysin promotes antigen translocation into the cytoplasm (cross-priming) and induces apoptosis of infected macrophages. These results are encouraging, but more work is required to fully realize the full potential of this vaccine.

Conclusion

The Beijing family genotype, believed to have originated in China, is now spreading globally. Recent studies indicate that one-third of global TB is caused by the Beijing family strains. BCG mass vaccination in South-east Asia has been implicated as a selective force for the emergence of the Beijing family genotype. Information presented in this review indicates that BCG vaccine is a risk factor rather than a preventive agent against TB in populations infected by the Beijing strains. However, in some countries, where the Beijing strain is low or absent, high protective efficacy of BCG (at least against tuberculous meningitis in children) has been maintained. This family strain has defied almost all measures to combat TB (BCG vaccine, host natural immunity and drug treatment). Measures designed for TB prevention have selected for more adaptive strains. However, virulence and drug resistance is not limited to the Beijing strain alone. To date, there are different Mtb strains with varying degrees of virulence and drug resistance in different geographical areas. New strains are emerging before we are able to resolve outstanding issues. Why do immunocompetent individuals capable of producing Th1 immunity (believed to be protective against TB) develop disease? What kind of immune response should a vaccine be capable of inducing in an immunocompetent host? Can the global population be rescued from TB with a single potent vaccine, or are different vaccines required for different Mtb lineages? Future vaccine development programmes must take into account the host population, and the strains of Mtb in the different endemic settings.