Though vital to immunological protection and vaccination against a broad range of microbial pathogens, B cells remain an enigma in the tuberculosis (TB) field. Classically, B cells and antibodies are thought to offer no significant contribution toward protection against Mycobacterium tuberculosis1. Though providing early evidence that humoral immunity may affect mycobacterial infection, the inconsistent results of passive immune therapies (likely due to variability in the preparation of antisera) conducted in the late nineteenth century 2 had led to its replacement by safer, and more consistent antibiotic treatment for TB by the 1950s 3. Within time, the efforts of founding immunologists such as Albert Calmette and Elie Metchnikoff elevated cellular immunity to the forefront of host defense in the twentieth century 4–6, where it has dominated research efforts in the TB field ever since 7. The more recent rise of molecular genetics has only further relegated B cells toward irrelevance in the TB field, as studies of B-cell immunodeficiency in both humans 8, 9 and mice 10, 11 have questioned whether these lymphocytes impart a protective effect against M. tuberculosis.
Immunological advances in the twenty-first century have led researchers to not only consider natural immunity to infection, which itself can be variable in terms of protection 12, but also novel ways to rationally improve vaccination strategies against infectious agents 13–15. Moreover, Ig research has advanced to produce safer and more effective antibody therapies, allowing passive immunization to be reconsidered for infectious disease treatment 16. Seeing immunology within these ever unraveling complexities of host defense has brought us once again back to B cells, in the hopes of unlocking previously underappreciated potentials of these lymphocytes to improve immunity against M. tuberculosis.
B and T cells collaborate to repel infectious challenge
The first components of host defense to encounter pathogens that have breeched anatomical barriers constitute innate immunity, which includes mononuclear phagocytes, natural killer cells and other innate cells of lymphoid origin, neutrophils, and serum factors including complement and natural antibodies 17–19. Encounter with foreign microbes through conserved pattern-recognition receptors activates innate sentinels known as dendritic cells to stimulate T lymphocytes, which in turn provide help to B cells and orchestrate adaptive immune responses 20, 21. Adaptive immunity has evolved in vertebrates to include both cellular and humoral components, with T cells and B cells mediating these effects, respectively. T cells target and promote apoptotic killing of pathogen-infected cells either directly or through cytokine activation of neighboring immune cells, whereas B cells make antibodies that neutralize invasion and target infectious agents for destruction 22. The lungs are particularly vulnerable to infection due to limitations in anatomical barriers allowing for airflow, making effective, but non-pathologic, pulmonary immunity vital for successful host defense 23.
Paramount among the properties of immunity is memory, which, through the differentiation of long-lived and highly specialized lymphocytes, results in decidedly effective secondary responses upon subsequent pathogen exposure. Edward Jenner is credited with first realizing this phenomenon when devising a smallpox vaccine over 200 years ago 24. Vaccination harnesses immunological memory by priming the immune system with an attenuated version of the targeted pathogen, bypassing the risk of primary infection while maintaining the immunological benefits against future exposure with the virulent pathogen 25. As an effective means of preventing pathogen entry and neutralizing microbial toxins, antibodies are of particular importance to nearly all vaccines currently in use 26. BCG is the only vaccine administered to prevent TB, and although it is thought to be protective against pediatric TB meningitis, its efficacy against adult pulmonary TB is limited 27. Efficacy of BCG immunization against TB correlated with specific antibodies in one study 28: however it is conventionally accepted that cellular immunity plays a predominant role. With immune correlates of protection against M. tuberculosis remaining incompletely defined 29, much remains to be understood regarding ways to enhance host defense against M. tuberculosis. We remain hopeful that greater understanding of both B and T cells will contribute to improved strategies of harnessing the complementary functions of these lymphocytes.
The role of B cells in immunity against non-viral intracellular pathogens
Protection against intracellular pathogens is often generalized as exclusively T-cell mediated, with B cells and antibodies playing a much more limited role than they are perceived to play during extracellular infections 30. This follows the concept that antibodies, flowing through the serum, are specialized for targeting extracellular microbes, whereas only T cells, which use antigen-presentation as a means to look within cells, can specifically target intracellular microbes for killing. Exceptions to these rules exist, however, as antibodies confer long-lived protection against many obligate intracellular viruses 31–34, presumably by targeting the extracellular phase of the virion, and T cells are activated by antigen-presentation to stimulate defense against extracellular helminthes 35. Directly relevant to the present review, emerging experimental evidence suggests that B cells play a role in defense against a wide variety of intracellular bacterial, fungal, and parasitic pathogens (see Limitation of single-gene knockout mouse studies in infectious diseases). Immunity likely did not evolve as two disconnected arms of defense that separately deal with intracellular and extracellular pathogens, rather a far more resourceful means of host defense is one developed to achieve maximally adaptive, cooperative, and versatile infection containment.
As mentioned previously, B- and T-cell responses function complementarily to repel natural infection, and can likewise both contribute to long-lived protection as a result of vaccination 36, 37. For example, collaboration between cellular and humoral immunity protects young children against Haemophilus influenzae type b, as a conjugation of the T-cell-independent carbohydrate antigen polyribosylribitol phosphate to a T-cell-dependent protein immunogen stimulates long-lived antibody protection 38, 39. Th cells greatly impact the development and specialization of B-cell responses 40, and B cells conversely impact T-cell activation by acting as or upon antigen-presenting cells 41. Within this intricate relationship between both arms of immunity, cellular and humoral responses together determine the outcome of intracellular infection.
Viruses often rely on limited means of cellular entry, and antibodies that neutralize the molecular interactions mediating infection allow for long-term immunity against viral pathogens 42. Intracellular bacteria, such as mycobacteria, and parasites, like malaria, present a much more complicated picture by utilizing multiple potential modes of cell entry 43, expanding the epitope requirements for neutralizing antibodies. Given this complexity, it is likely that successful antibody responses against intracellular bacteria tend to prevent disease rather than infection 44, through activation of complement pathways and cellular immunity 45. Further research efforts are necessary to understand how protective antibodies can be incorporated into successful vaccines against intracellular bacterial pathogens, such as M. tuberculosis, that continue to cause significant global morbidity and mortality.
Limitation of single-gene knockout mouse studies in infectious diseases
The small size, limited expense, available genetics and laboratory reagents, and an immune system that significantly mirrors that of humans maintain the mouse as the preferred choice for infectious disease study. Mice with targeted deletions that “knockout” specific genes are useful tools for immunologists, and can be utilized to garner much information regarding the roles of individual immune components within the host response to specific pathogens. Knockout mice have been useful for examining the effects of B-cell deficiency upon disease progression and survival for a wide range of intracellular bacteria and parasites. Enhanced susceptibility of B-cell-deficient mice to infections with Chlamydia trachomatis, Francisella tularensis, Leishmania major, Plasmodium chabaudi chabaudi, Pneumocystis carinii, and Salmonella enterica serovar Typhimurium have all been reported 46–52, demonstrating that B lymphocytes mediate optimal immunity against many intracellular pathogens in addition to viruses.
Studies of M. tuberculosis infection in B-cell-deficient mice have been variable, with reports of immunity being diminished, pathologic progression being delayed, or no apparent effects from the genetic ablation of B cells 10, 11, 53–55. This broad range of results emphasizes the limitations and variances inherent to mouse models of infectious disease, particularly TB. Single knockout studies in M. tuberculosis-infected mice may yield results, whose interpretations are not straightforward because of functionally overlapping immune components 56, 57, may not account for potential host response differences between mice and humans 58, and do not provide any information regarding the potential effects of augmenting the particular immune component in question. Consequently, knockout mouse studies can lead to premature conclusions regarding the role of a particular component of immunity, if not interpreted thoroughly. Additionally, experimental conditions can have marked effects on results. For example, increasing the infection inoculum exposed a requirement for B cells in the optimal TB host response in mice 53, 54. Supporting a role for B cells in host immune response to M. tuberculosis, genetic ablation of the polymeric Ig receptor heightened susceptibility of mice, implicating a role for secretory IgA within optimal TB immunity 59. Finally, protective effects of intravenous Ig (IVIG) suggest further impact of humoral components upon host defense in TB 60. Though the results of certain knockout mouse studies and the IVIG experiment indicate that B cells and their products mediate protection against M. tuberculosis, the important question that remains is whether B-cell responses can be augmented to improve immunity against M. tuberculosis through immunotherapy or vaccination.
Ectopic B-cell aggregates as a consequence of chronic inflammation and infection
Our renewed interest in B lymphocytes emerged from observations of follicle-like B-cell aggregates in the lungs of TB patients 61, 62. These B-cell aggregates are the predominant foci of cellular proliferation in human TB lungs 61, and are characteristics of TB granulomatous progression in mice 11, 62, 63. B-cell aggregates are noted pathological findings of many chronic inflammatory diseases, including multiple sclerosis and rheumatoid arthritis 64, 65. Similar B-cell clusters have been observed in other infections, such as with influenza virus and Helicobacter66, 67. Certain pathogens have evolved means of promoting expansion and inhibiting apoptosis of B cells 68, 69, which may contribute to the accumulation of these lymphocytes during infection.
What are the effects of these ectopic B-cell aggregates upon immunity to M. tuberculosis? It has been hypothesized that these follicle-like structures function to perpetuate local host responses, with the majority of proliferative activity occurring in the proximity of B-cell aggregates 61. T cells can be scattered throughout these B-cell clusters 62, making these potential sites of both antigen-presentation and B-cell maturation. It is also notable that these B-cell aggregates are components of tertiary lymphoid tissue within TB lungs 70, containing markers of germinal centers 54. Further suggesting that B cells foster localized architecture of the immune response, we noted that B-cell-deficient mice had exacerbated pulmonary pathology and disruption of granulomatous organization in the lungs 54. Much remains to be understood regarding the significance of lymphoid neogenesis in TB, but it has been demonstrated that these inducible lymphoid structures can prime protective immunity in the lungs and memory responses against pulmonary influenza virus challenge in the absence of secondary lymphoid organs 66, 71.
T-cell activation is strongly influenced by B cells
B cells are professional antigen-presenting cells with notable impact upon T-cell responses. By capturing antigens via cell-surface receptors, B lymphocyte activation is initiated. Subsequent progression through cellular interactions with CD4+ helper T cells includes antigen-presentation events, which stimulate T cells to produce the cytokines that reciprocally regulate the antibody responses of B cells 72. Though the significance of B-cell-mediated antigen-presentation varies with antigen and immunological conditions 73, these lymphocytes can be targeted to present antigen to T cells by specific vaccination strategies 74, 75. In fact, one such B-cell-targeting vaccine vector has been effective in boosting BCG primed immunity against M. tuberculosis76. In a recently published example, the presence of B cells, but not specific antibody, protected against reactivation of chronic virus infection, presumably through antigen-presentation to T cells 77. An antibody-independent function of B cells in priming optimal primary and secondary immune responses against the intracellular bacterial pathogen F. tularensis has also been reported 78. Furthermore, activated B cells serving as antigen-presenting cells have been utilized to enhance anti-tumor immunity 79. It has also been noted that antigen-presentation by B cells can help perpetuate autoimmunity 80, whereas resting B cells are thought to mediate immune tolerance 81, 82. By studying B cells as targets for immune augmentation as well as in the suppression of autoimmunity 83, future research will likely expose creative means of targeting this often over-looked pathway of antigen-presentation. Finally, accumulating evidence suggests that B cells are required to prime memory T-cell responses 52, 84, placing the B- and T-cell relationship in particularly relevant context to vaccine biology.
B cells can also act upon other antigen-presenting cells, thus influencing the evolution of T-cell responses in a more indirect manner. T cells are stimulated to clonally expand by antigen-presenting cells that concurrently express co-stimulatory molecules on their surface while presenting antigen 85. This illustrates the two-signal hypothesis for T-cell activation in which engagement of both the antigen receptor and CD28 on the surface of T cells is required to initiate adaptive immune responses, with lymphocyte tolerance a consequence of an absent second signal 86. Antigen-presenting cells up-regulate surface expression of co-stimulatory proteins, such as those of the B7 family, during a maturation process that follows the activation of innate immunity 17. These co-stimulatory molecules then interact with CD28 or other co-receptors to promote activation of T cells 87. B cells can modulate this maturation process through the production of antibodies and cytokines, which can either enhance or suppress immune responses 88–93. B cells have been reported to polarize T-cell responses through the production of cytokines 94, and the provision of IL-10 is thought to be a means by which B cells can promote Th2 differentiation in mice 95. Additionally, natural antibodies can modulate antigen-presentation by binding to and altering the activity of the co-stimulatory molecules B7 and CD40 41, 96, 97.
Much has been uncovered regarding antibody regulation of antigen-presentation through Fcγ receptors 98, and this immunological pathway has garnered interest as a potential means of improving vaccination against intracellular pathogens 99. Fcγ receptors are divided into stimulatory and inhibitory types based on the presence of intracellular ITAM or ITIM motifs, respectively 100, 101. FcγRIIB, the lone inhibitory Fcγ receptor, has an intricate involvement in T-cell activation by limiting dendritic cell maturation and subsequent antigen-presentation, whereas the stimulatory Fcγ receptors appear to promote both processes 89, 90. FcγRIIB also plays a significant role in mediating peripheral tolerance of T-cell responses in murine autoimmunity models 102. Furthermore, selective blockade of FcγRIIB leads to enhanced T-cell activity in experimental tumor models 89–91. Stimulatory Fcγ receptors appear to promote T-cell responses, the polarization of which is shaped by the prevailing inflammatory context and can be either Th1 or Th2 dominated 103. Fcγ receptor engagement can be strongly influenced by antibody isotype 104, and immunization methods that preferentially target stimulatory Fcγ receptors have the potential effects of enhancing cellular immunity against intracellular pathogens 99, 105. Interestingly, it has been reported that FcγRIII mediates immune suppressive effects in an IVIG model, illustrating that association with an ITAM motif may not restrict function to pro-inflammatory activity 106.
Genetic disruption of Fcγ receptor function in mice has implicated a role for these receptors in microbial infection 107. In particular, specific ablation of the shared stimulatory Fcγ-chain compromises optimal immunity against a variety of intracellular pathogens, including influenza virus, Leishmania species, Plasmodium berghei, and S. enterica49, 108–112. Passive immunization against Cryptococcus neoformans with IgG1 mAb was also dependent upon functional stimulatory Fcγ receptors 113, further implicating the stimulatory Fcγ receptors in cellular host defense against intracellular pathogens. We have found that disruption of stimulatory Fcγ receptor activity heightens susceptibility to M. tuberculosis Erdman, corresponding with compromised bacterial containment and worsened immunopathology 114. Conversely, genetic deletion of inhibitory FcγRIIB improves mycobacterial containment, with increases in IFN-γ production and Th1 polarization detected in the lungs 114. Efforts are underway to study how these apparent effects of Fcγ receptors can be harnessed to improve immunity against M. tuberculosis.
These immune-enhancing effects from specific targeting of Fcγ receptors may provide mechanisms of improving vaccine-induced immunity against intracellular pathogens 99, 105. For example, viral vectors can be targeted to Fcγ receptor-bearing cells through antibody-dependent infection enhancement 115, providing a means by which antigen-presenting cells can potentially be manipulated for activation and immunization. Interestingly, a P. falciparum merozoite surface antigen has been identified that preferentially induces isotype class-switching to IgG2b in mice 116, a cytophilic Ig subtype with preferential affinity for stimulatory Fcγ receptors 104. Future research should further expose how Fcγ receptors can be targeted and utilized to enhance immunity against intracellular pathogens.
Are antibodies protective against M. tuberculosis, an intracellular pathogen?
Contrary to the generally accepted notion that humoral immunity is insignificant in protection against the tubercle bacillus, passive administration of antibodies had reported efficacy against M. tuberculosis since the late nineteenth century 2. However, serum therapy for TB fell out of favor when efficacy of treatment and reagent preparations were found to be inconsistent 117, and the advent of pharmacologic anti-mycobacterials, such as streptomycin, by the middle of the twentieth century, offered a far more reliable option 118. Excellent reviews exist regarding the history and recent developments of antibody-mediated immunity to M. tuberculosis2, 119. Recently, mAb against mycobacterial arabinomannan, heparin-binding hemagglutinin and 16 kDa α-crystallin have all demonstrated efficacy in mouse models of TB 120–124. These antibodies mediate protection in different manners, some by diminishing tissue mycobacterial burden while others enhance animal survival through apparent decreases in inflammatory progression 119. Notable, however, is the fact that an M. tuberculosis arabinomannan–protein conjugate vaccine has been reported to induce more robust antibody responses than BCG, but without an apparent survival improvement in mice 125. The question remains as to whether the mouse is an optimal modality to measure enhanced protection, given the chronically inflammatory and persistent infectious burden of the model and limited correlations of the pathology with that of human disease. Thus, future studies are required to unravel antibody correlates of protection in humans and animal models if effectively protective immunization-induced B-cell responses against M. tuberculosis are to be generated. Finally, it is noteworthy that virtually nothing is known of the impact of innate or natural antibody responses 126–128 upon mycobacterial infection. Given that complex lipids are major constituents of the hydrophobic cell wall of mycobacteria, it would be interesting to examine the significance of T-independent antibody responses, most notably of B1 and marginal zone B cells, in the defense against M. tuberculosis.
Looking beyond antigen-specific neutralization, antibodies also have notable general effects upon inflammation, including complement activation, Fcγ receptor cross-linking, and release of microbial products due to direct anti-microbial activity 45. With regards to pulmonary host responses, antibodies can modulate architectural changes in airway epithelium and vessels, as in response to mycoplasma infection 129. In our own studies, we found that adoptive transfer of B cells resolved the inflammatory exacerbation in B-cell-deficient mice upon airborne challenge with M. tuberculosis Erdman 54. Suggesting a role of Ig-mediated “endocrine” immune regulation, this reduction in pathology occurred in conjunction with detectable levels of antibodies in the serum but without the presence of B cells locally within lungs 54. It is quite apparent that antibodies can have a variety of protective effects during infection with intracellular pathogens, which includes limiting inflammatory pathology 130 in addition to well-established roles in neutralizing and opsonizing microbes.
How B cells shape the immune response against M. tuberculosis
The correlates of vaccine-mediated protection against M. tuberculosis are incompletely defined, but most evidence suggests that T cells are of paramount importance 7. Consequently, the aim of all the novel TB vaccines currently in development is to enhance cellular immune responses against the pathogen 131. As reminded by recent failure in a clinical trial of a T-cell vaccine against HIV 132, it is prudent not to limit vaccine research too narrowly. It is quite likely that protective antibody responses will be required for optimally successful immunization against M. tuberculosis in the future 44, and further research designed to uncover how humoral immunity can best be harnessed to mediate this protection seems warranted. In this direction, efforts to understand B-cell biology and its relationship with TB have demonstrated that these often over-looked lymphocytes can significantly influence cytokine production, bacillary containment, and immunopathologic progression during M. tuberculosis infection. As one possible mechanism by which B cells shape the immune response, we hypothesize that Ig, acting upon Fcγ receptors, influences antigen-presenting cell maturation to produce the noted phenotypes of gene-deficient mice challenged with M. tuberculosis (Fig. 1). However, this is but one pathway by which B lymphocytes modulate the host response in TB infection. B cells can conceivably shape anti-tuberculous immunity through a variety of means, including direct effects of antibody upon the pathogen, antigen-presentation, cytokine production, as well as influencing the intracellular killing mechanisms of leukocytes.
Figure 1. Schematic of how B cells shape the immune response against M. tuberculosis. B cells modulate the murine host response against M. tuberculosis in a variety of ways. Upon acute infection, selective engagement of stimulatory Fcγ receptors by antibody complexes heightens the Th1 response and promotes mycobacterial containment with minimal inflammation. Conversely, engagement of inhibitory FcγRIIB increases IL-10 production and compromises immunity against M. tuberculosis. Interestingly, an immunosuppressive phenotype in the absence of B cells subverts optimal containment of acute M. tuberculosis challenge initially, but delays inflammatory progression during chronic TB.
Download figure to PowerPoint
As addressed earlier, T-cell activation is achieved through antigen-presentation by specialized cells that must undergo a maturation process in order to serve this function 17. We previously described how B cells can have a significant impact upon antigen-presenting cell maturation through the engagement of Fcγ receptors by antibodies. Using mice deficient in inhibitory FcγRIIB receptor function, we found that the absence of FcγRIIB enhanced containment of M. tuberculosis Erdman in mice 114. Pulmonary IFN-γ production and Th1 cell frequency were correspondingly increased in FcγRIIB-deficient mice. Conversely, genetic ablation of the common γ-chain shared among stimulatory Fcγ receptors impaired M. tuberculosis containment and worsened overall survival 110. Thus, B cells can significantly impact host immunity and disease outcome by engagement of Fcγ receptors during TB, influencing both Th1 activation and mycobacterial containment.
Murine studies of M. tuberculosis infection also indicate that B cells have a significant impact upon the production of IL-10 in the lungs. IL-10 is produced by a broad range of leukocytes, including B cells, dendritic cells, macrophages, and T cells, imparting mostly anti-inflammatory functions upon host responses 133. Heightened production of IL-10 in the lungs during M. tuberculosis infection has been reported in two different murine models of B-cell deficiency 54, 134. We also observed an IL-10 increase in the lungs of stimulatory-Fcγ-chain-deficient mice infected with M. tuberculosis Erdman 114.
The cellular source of this IL-10 increase is unclear as intracellular cytokine-staining techniques to detect IL-10 are not optimally sensitive for detecting significant quantities of these cells in mice. The recent development of IL-10 reporter mice may facilitate such studies in the future 135, 136. IL-10 production is a characteristic of incompletely activated dendritic cells that have not undergone complete maturation 137. Thus, by influencing cellular activation via immune complex engagement of Fcγ receptors, B cells may influence the production of IL-10 by these antigen-presenting cells. It has also been reported that B cells can activate or inhibit regulatory T cells, a significant cellular source of IL-10 138, 139. Finally, the increase in IL-10 may be in reaction to the exacerbated immunopathology induced during M. tuberculosis infection of B cell−/− and Fcγ-chain−/− mice 54, as the production of anti-inflammatory cytokines, such as IL-10, may be a compensatory attempt to limit excessive inflammation. More work is needed to assess the relationship of B cells and cytokines during TB, particularly given that B cells themselves can be a source of the vital anti-mycobacterial cytokine IFN-γ 94.
As also described previously in this article, B cells are prominent components of TB pulmonary granulomatous inflammation, with large aggregates of these lymphocytes as prominent histological characteristics of the infection 61–63. The absence of B cells, while compromising optimal immunity leading to exacerbated pulmonary pathology upon acute M. tuberculosis infection 54, leads paradoxically to a delay in inflammatory progression during chronic infection in mice 55. One possible explanation for this paradox is that B-cell function during the course of TB is infection phase-specific: During acute infection, B cells are required for an optimal granulomatous response and effective immunity against pulmonary challenge with M. tuberculosis, and in their absence, dysregulation of granuloma formation results and increased pulmonary inflammation is required to contain infection. In contrast, during chronic phase of infection when persistent bacilli are contained, the immunologically active B-cell clusters 54, 61, 62 likely promote the perpetuation of local host immunity against M. tuberculosis and may aid in prevention of reactivation disease. This perpetuation of inflammation may occur in part through B cells acting as antigen-presenting cells. Indeed, T cells have been observed to be associated with B-cell nodules in both human and mouse tuberculous lung tissues (62, P. Maglione and J. Chan, unpublished). In this regard, antigen uptake may occur in a specific manner via the BCR, though it has been shown that B cells can load MHC with peptide not in line with their receptor specificity 140. Such “non-specific” antigens could be the result of Toll-like receptor uptake 141. Thus, the inflammatory paradox of B-cell-deficient mice seems to reflect the role of B cells shifting from optimizing host defense during acute challenge to perpetuating the chronic inflammatory response during persistent infection.
TB morbidity and mortality results from an aberrant and damaging host response 142, 143: one that is both excessive in pathologic consequence yet ineffective in containing the pathogen. This apparent paradox may be viewed as a product of ineffective immune containment of the tubercle bacillus leading to excessive compensatory recruitment of leukocytes within the lungs. This is best evidenced in patients with reactivation TB. Rather than being devoid of immune activity, which could explain the re-emergence of a dormant pathogen, the lungs of patients with reactivation diseases contain areas of intense pulmonary infiltrate 143. Prior to treatment, active TB pulmonary infiltrate is dominated by neutrophils 61, an innate immune cell that aids in early protection but promotes inflammatory damage in a variety of acute pulmonary diseases 144. Similar to humans, susceptibility to TB is associated with increased pulmonary neutrophil influx in mice 145. In contrast, incidental findings of Ghon complex pathology in humans emphasize how successful containment of M. tuberculosis need not have significant immunopathology as a by-product. Existing data strongly suggest that B cells play a role in modulating the inflammatory response during TB infection 54, 55.