PeRSpeCTIveS
clusters that express proliferation-associated OPINION markers and secrete M. tuberculosis-specific antibodies14. However, depletion of B cells Antibodies and tuberculosis: using the therapeutic monoclonal antibody rituximab resulted in highly heterogeneous effects in local granuloma immune finally coming of age? modulation, perhaps owing to the variety of antibody and non-antibody Hao Li and Babak Javid functions of B cells, but had no effect on disease outcome15. Abstract | Are antibodies important for protection against tuberculosis? The jury While these studies did not by has been out for more than 100 years. B cell depletion in experimental themselves prove a lack of B cell or Mycobacterium tuberculosis infection failed to identify a major role for these cells antibody involvement in immunity to in immunity to tuberculosis. However, recent identification of naturally occurring M. tuberculosis, they did cast doubt on antibodies in humans that are protective during M. tuberculosis infection has the potential importance of B cells and reignited the debate. Here, we discuss the evidence for a protective role for antibodies in animal models of TB infection. In parallel, however, studies using antibodies antibodies in tuberculosis and consider the feasibility of designing novel in passive immunization suggested that tuberculosis vaccines targeting humoral immunity. certain antibodies are protective against TB. Administration of monoclonal antibodies The role of humoral immunity in vaccine strategy in TB and discuss the key raised against M. tuberculosis antigens or tuberculosis (TB) is controversial. Almost all questions that still need to be addressed in pooled mouse or human immunoglobulins clinical and preclinical vaccine candidates this field. could ameliorate disease in animal models for TB that are in development target in most16–23, although not all24, cases. Perhaps cell-mediated immunity1. As an intracellular Antibodies in tuberculosis owing to caution in interpretation of some pathogen, it has been assumed that Why is antibody-mediated immunity of the data — antibodies mixed with Mycobacterium tuberculosis, which is the controversial? Koch’s 1882 discovery of M. tuberculosis in vitro before infection, causative agent of TB, would be inaccessible M. tuberculosis as the causative agent of TB as in some of the experiments, might have to circulating antibodies; however, cells stimulated an interest in specific therapies, induced clumping and exaggerated the infected with M. tuberculosis would be including serum therapy, which had shown degree of offered protection — there was vulnerable to killing by the effector arm considerable success in the treatment of greater enthusiasm for the development of the cell-mediated immune response. other infectious diseases (reviewed in ref.8). of recombinant vaccines that elicited Several publications have stimulated However, dozens of studies using serum potent cell-mediated immune responses a re-evaluation of the contribution of therapy in TB showed highly variable in animal models. antibody-mediated immunity to protection results, partly owing to differences in against M. tuberculosis2–5, but a number antigens, antibody titres in serum and host Antibodies back in vogue. In 2013, the first of important questions remain. Why has animals (reviewed in ref.9). As such, late-stage clinical trial in the modern era antibody-mediated immunity been largely interest in serum therapy waned rapidly for a new vaccine against TB was reported; overlooked for vaccine development? and came to a halt following the more disappointingly, it showed no efficacy over Which approaches led to the recent series promising discoveries of the BCG and above routine BCG immunization25. of high-profile papers re-examining the vaccine strain and streptomycin, the first The vaccine, known as MVA85A, delivered role of antibodies in human immunity to anti-tubercular antibiotic. a recombinant immunodominant epitope, TB? How do antibodies offer protection There was also increasing evidence for antigen 85A, via modified vaccinia Ankara against M. tuberculosis? As the only licensed the importance of cell-mediated responses virus. MVA85A performed as engineered: vaccine for TB, bacillus Calmette–Guerin in protection against TB. Mice lacking key it elicited a potent response of CD4+ T cells (Mycobacterium bovis-BCG or BCG) elements of cell-mediated immunity, such secreting IFNγ25,26. Therefore, its failure offers highly variable protection against as IFNγ, tumour necrosis factor (TNF) or (though unsurprising27 to some) led to TB6 and almost no protection against the T cells, and patients with HIV are highly a re-evaluation of what may constitute a transmissible form of the disease, there is an susceptible to TB10,11. This contrasted with protective immune response against urgent need for a new TB vaccine if we are the relatively modest disease phenotypes M. tuberculosis. Some questioned whether to interrupt the current TB emergency; in seen when mice lacking B cells were infected IFNγ-secreting CD4+ T cells were the key 2016 M. tuberculosis claimed more lives than with M. tuberculosis12,13. It was shown to cell-mediated immunity28, and there was any other single pathogen7. In this Opinion in experimentally infected cynomolgus also renewed interest in the potential role, piece, we consider whether antibody-based macaques that M. tuberculosis-containing if any, of antibody-mediated immunity in vaccines could show promise as a new granulomata are surrounded by B cell M. tuberculosis infection.
NaTuRe ReviewS | Immunology volume 18 | SEPTEMBER 2018 | 591 Perspectives
Important lessons had been learnt in were infected with M. tuberculosis, HEBUI) with no evidence of prior infection the interim. The focus shifted to studying antibodies from the individuals with LTBI by M. tuberculosis may represent human naturally elicited human antibody responses. were better at restricting intracellular ‘restrictors’ that can control M. tuberculosis It was also clear from the renaissance in growth of the pathogen than antibodies infection — although an alternative broadly neutralizing antibodies in the HIV from patients with active TB4. explanation might be that current tests for field that not all HIV-infected individuals Our own group and the groups of LTBI merely fail to identify the mediators made these protective antibody responses29. Kaufmann and Wardemann chose to of immune exposure in these individuals33. As such, pooling antibodies from multiple compare antibody responses between By contrast, no patients with active TB made donors might drown out signals of patients with active TB and health-care protective antibody responses3. Intriguingly, protection within the noise of the majority workers (HCWs) who had potential M. tuberculosis growth restriction owing of donors that did not make protective occupational exposure to M. tuberculosis3,5. to antibodies from protective donors was responses. Below, we focus on four Zimmermann et al. identified M. tuberculosis- completely abrogated both in vivo and key papers published since 2016 that specific antibody responses in both patients in vitro in the absence of CD4+ T cells and applied these principles to advance with active TB and HCWs. Approximately required formation of immune complexes the field2–5. 40% of monoclonal antibodies isolated for full efficacy, suggesting that protection A low titre of antibodies against from the plasmablasts of these individuals was not due simply to neutralization of lipoarabinomannan (LAM), which is a recognized mycobacterial antigens, with M. tuberculosis by antibodies3. key glycolipid found on the surface of the majority of M. tuberculosis-specific Together, these studies revived interest Mycobacterium spp., had been associated antibodies targeting surface-exposed in humoral immunity to TB and suggested with disseminated TB in children30, antigens: either binding intact bacteria that subjects with high exposure to and passive immunization of mice with or surface antigens5. It was found that M. tuberculosis but without active TB monoclonal antibodies against the sugar 7 of 12 isolated immunoglobulin A (IgA) represent the ideal population in which to component of LAM, arabinomannan, had monoclonal antibodies but 0 of 16 isolated study and isolate protective antibodies. been demonstrated to be protective in IgG monoclonal antibodies were able to One further study deserves highlighting. mouse infection models19,21. Achkar restrict M. tuberculosis growth within the Unlike the preceding four studies, it focused and colleagues obtained serum from lung epithelial cell line A549. However, on BCG-elicited immunity in murine TB 30 healthy volunteers immunized with BCG, although HCWs had slightly higher titres of and did not involve human antibodies. which stimulates both cell-mediated and M. tuberculosis-specific IgA, both restrictive Casadevall and colleagues showed that when humoral immunity2. They demonstrated and ineffective M. tuberculosis-specific mice were immunized with BCG grown with that BCG elicited antibody responses monoclonal antibodies were isolated from an intact capsule, they were better able to against both arabinomannan and LAM and patients with active TB and the healthy resist subsequent infection with pathogenic that post-vaccination serum was able to HCWs. Therefore, it could not be inferred M. tuberculosis34. Importantly, encapsulated opsonize M. tuberculosis for phagocytosis from these studies how important antibodies BCG was not only better at eliciting higher by macrophages, enhance phagolysosome were for protection from active TB. titres of class-switched antibody responses, fusion and inhibit intracellular growth of It is difficult to control for relative it was also more effective at generating IFNγ M. tuberculosis. Importantly, the titre exposure to M. tuberculosis when studying and poly-functional T cell responses, which of arabinomannan-specific responses in protective immunity, which can confound together led to an order of magnitude lower individuals correlated with the magnitude true immunity from lack of exposure. Our bacterial burden34. of these protective responses2. subjects were from the Beijing TB hospital, Alter and Fortune examined polyclonal which treats thousands of patients with Antibody-mediated protection serum from 20 patients with active TB and TB annually. Unlike in more developed How do antibodies protect against TB? As 22 individuals with latent TB infection economies, HCWs in China rarely use already mentioned above, what is clear is (LTBI) — these are individuals with adequate personal protective equipment that protection appears to be more complex immunological evidence of exposure to when encountering patients with TB. than simple neutralization of bacteria and M. tuberculosis but with no symptoms Therefore, HCWs have a substantial requires cell-mediated immune functions. or signs of active infection4. They used a M. tuberculosis exposure burden, and many For example, macrophages, immune ‘systems serology’ approach to interrogate show evidence of LTBI or develop active complex formation (Fig. 1) and T cell approximately 70 features of antibody TB31,32, but many others remain healthy, with responses have all been proposed to be function that are not associated with the no evidence of LTBI. We isolated polyclonal involved (Fig. 2) — although evidence variable domain of the antibody molecule. immunoglobulin from patients with TB for many of these mechanisms is still Of these features, 19 differed significantly and HCWs — some of whom had LTBI but highly preliminary. between antibodies derived from South also some who were highly exposed but African patients with LTBI and those uninfected (HEBUI). We tested antibodies What are the protective antigens? derived from South African patients from single donors in two models: one The M. tuberculosis genome encodes with active TB, and 9 of these features examined relative protection from approximately 4,000 genes35, and the function were sufficient to distinguish between M. tuberculosis in a mouse model of aerosol and subcellular localization of the products the two groups. Some of the features, infection, and one examined protection of many of these genes remain unknown. in particular, glycosylation status of the in an in vitro human whole blood assay. Furthermore, M. tuberculosis has a complex crystallizable fragment (Fc) portion of the Antibodies from a subset of both donors glycolipid and glycoprotein capsule36, which antibody molecule, were also validated in with LTBI and HCWs who were HEBUI would represent the surface of the pathogen samples from Texas, USA, and Mexico. (7 of 48) restricted M. tuberculosis growth in natural infection. To complicate matters, In experiments where human macrophages in both assays. Those donors (who were for pragmatic purposes, M. tuberculosis is
592 | SEPTEMBER 2018 | volume 18 www.nature.com/nri Perspectives
CD/VWDGTEWNQUKU of antibodies targeting soluble antigens #PVKDQF[ 2JCIQUQOGs N[UQUQOGHWUKQP from M. tuberculosis failed to remove the protective fraction, whereas the depletion of %QORNGOGPV antibodies targeting intact M. tuberculosis TGEGRVQT (E4 abrogated protection. Together, these data strongly suggested that protective antibodies were directed against surface-displayed capsule antigens3. However, what was the nature of these antigens? Panning an M. tuberculosis proteome array37 failed to identify potential protective antigens3. Possibly, the majority of protective responses 2JCIQUQOG were directed against non-protein antigens, /CETQRJCIG and indeed carbohydrate–protein conjugate +PETGCUGFRJCIQN[UQUQOG 38 'PJCPEGFRJCIQE[VQUKU HQTOCVKQP vaccines against arabinomannan and a peptide mimotope vaccine targeting LAM39 E F +PȱCOOCUQOG have shown efficacy in mouse models. /VWDGTEWNQUKU But while there is a clear correlation between arabinomannan antibody titres #PVKDQF[ +PȱCOOCUQOG CEVKXCVKQP and protection2,30, only one study to date has isolated naturally occurring human monoclonal antibodies against M. tuberculosis5. The antigenic targets of 0GWVTCNK\CVKQP most of the human antibodies that recognize M. tuberculosis have not been identified, and although some antibodies were shown to bind the cell surface epitopes LAM and +.β heparin-binding haemagluttinin20, antibody- RTQFWEVKQP 'PJCPEGFKPȱCOOCUQOG mediated protection in this study appeared to CEVKXCVKQP be independent of these antigens5. G Roles of invariant antibody functions. )TCP\[OG Although antibody-mediated immunity CPFRGTHQTKP to viruses involves simple neutralization, broadly neutralizing antibodies to HIV also require antibody Fc-mediated functions for 0-EGNN in vivo efficacy40. Systems serology identified a number of Fc-mediated antibody functions: principally, glycosylation of the Fc molecule, which had previously been demonstrated to be important for protection of passive antibody transfer in mice41, but 'PJCPEGFE[VQVQZKE0-EGNNCEVKXKV[ immune complex and inflammasome activation functions were also identified as Fig. 1 | Potential mechanisms of antibody-mediated protection against Mycobacterium tuberculosis. The figure indicates the different ways in which antibodies may enhance immunity to being strongly associated with protective 4 Mycobacterium tuberculosis. a | Complexes of M. tuberculosis bound to antibody can be more readily responses derived from LTBI donors . phagocytosed by macrophages via crystallizable fragment (Fc) receptors (FcRs) and complement rece M. tuberculosis has adapted to an ptors. b | Antibodies specific for M. tuberculosis can increase phagosome–lysosome fusion — which intracellular lifestyle — therefore, the first M. tuberculosis usually interferes with — and thereby restrict the growth of M. tuberculosis. c | Antibodies step of productive infection is phagocytosis against M. tuberculosis may have direct microbicidal or neutralizing activity , or they may prevent the by a host cell. The potential protective uptake of the bacteria by cells not expressing the appropriate FcRs to promote M. tuberculosis killing. role for opsonophagocytosis by antibody d | Antibodies against M. tuberculosis have been shown to promote inflammasome activation in macro is therefore less straightforward42 than | phages, which is associated with ASC speck formation and IL-1β secretion. e Antibodies specific for that for extracellular pathogens such as M. tuberculosis may stimulate killing of M. tuberculosis-infected cells via natural killer (NK) cell-mediated pneumococcal species, where it plays a antibody-dependent cell cytotoxicity. It should be noted that there is only limited evidence for some 43 of these mechanisms. clearly protective role . The intriguing finding that the IgA1 isoform of a particular monoclonal antibody was protective, routinely grown in the laboratory in media offered more robust protection against whereas the IgG1 isoform of the same containing mild detergent, which strips off M. tuberculosis34. Protective antibody monoclonal antibody was permissive for much of the capsule36. responses from HCWs were significantly growth, in lung epithelial cells suggested As already mentioned, immunization of enhanced when assessed against capsule- that opsonizing antibodies in TB act as mice with BCG grown with an intact capsule intact M. tuberculosis. Furthermore, depletion ‘trojan horses’ (ref.5). In support of this idea,
NaTuRe ReviewS | Immunology volume 18 | SEPTEMBER 2018 | 593 Perspectives
C D -KNNKPHGEVGFEGNNU S. Typhimurium — specifically after /VWDGTEWNQUKU vaccination47. While we do not know precisely how CD4+ T cells and antibodies )TCP\[OG $%4 CPFRGTHQTKP cooperate to protect against M. tuberculosis infection, there are several possibilities, $EGNN 2NCUOC 0- including enhanced antigen presentation EGNN EGNN /CETQRJCIG to T cells following phagocytosis of M. tuberculosis–antibody immune /*% complexes or CD4+ T cell help for natural ENCUU++ %[VQMKPGU #PVKDQF[ killer cell-mediated antibody-dependent #EVKXCVKQP cell cytotoxicity (Fig. 2). 6%4
+. Antibody-based vaccines? %& #PVKDQF[ It is now clear that naturally occurring RTQFWEVKQP #EVKXCVKQP 6EGNN human antibodies that are likely to be protective against M. tuberculosis infection exist in at least a subset of individuals. But can this observation lead to the development E -KNNKPHGEVGFEGNNU -KNNKPHGEVGFEGNNU of a preventive vaccine? We first need to prioritize the characterization of the )TCP\[OG CPFRGTHQTKP antigens that elicit protection, followed by )TCP\[OG the characterization of the molecular and CPFRGTHQTKP cellular mechanisms mediating protection. %& 6EGNN This will be impossible without further #PVKDQF[ RTQNKHGTCVKQP studies of isolated protective monoclonal 0- %& EGNN 6EGNN antibodies. However, even then, there are at least three further hurdles that need to be addressed. First, what are the /*% in vitro correlates of in vivo protection? ENCUU++ ↑#PVKIGP Second, are protective antibody responses RTGUGPVCVKQP 6%4 of sufficient magnitude and efficacy to +. +. prevent disease? Third, can immunization #EVKXCVKQP #EVKXCVKQP with either recombinant antigens or a killed or attenuated mycobacterial strain elicit sufficiently protective antibodies in susceptible individuals?
+ Fig. 2 | Potential CD4 T cell-dependent mechanisms of antibody-mediated protection. There are In vitro correlates of protection. For many several ways in which CD4+ T cells and antibodies may cooperate to promote immunity to viral and bacterial infections, simple in Mycobacterium tuberculosis. a | CD4+ T cells can provide help to B cells to stimulate the maturation of | + vitro tests can predict in vivo protection by antibody responses. b CD4 T cells may stimulate antibody-mediated killing of M. tuberculosis- 48,49 infected cells indirectly by producing cytokines that can activate natural killer (NK) cells and enhance antibodies . However, for M. tuberculosis, antibody-dependent cell cytotoxicity responses. c | Complexes of M. tuberculosis and antibody may a major bottleneck in vaccine development result in increased processing and presentation of M. tuberculosis antigens to CD4+ T cells by profes has been the total absence of any correlates sional phagocytes (for example, macrophages or dendritic cells). This can result in increased activation for M. tuberculosis protection. Therefore, of helper CD4+ T cells that in turn promote cytotoxic CD8+ T cell and NK cell responses to enhance efficacy of vaccine candidates, even in M. tuberculosis killing. BCR, B cell receptor; TCR , T cell receptor. non-human primates, cannot reliably predict protection in humans. As such, only late-stage clinical trials can properly the relevant lung epithelial cells expressed cells that were required for protection. test protection. This extremely costly neonatal Fc receptors, which bound only IgG Importantly, we identified a critical role for option limits the potential number of and not IgA. Therefore, the protective role both immune complex function and CD4+ candidates that can be tested1,50, and of IgA may have been due to binding and T cells in antibody-mediated protection3. late-stage failures can influence funding inhibition of M. tuberculosis phagocytosis Antibodies play an important role for of other candidates. by epithelial cells5. D’Arcy Hart’s classic protection against another intracellular As already discussed, there are many observation that opsonizing antibodies pathogen, Salmonella enterica subsp. potential mechanisms by which antibodies may also promote phagolysosomal fusion enterica serovar Typhimurium16,46, for might mediate protection against and killing by reactive oxygen species in which a vaccine with 50–80% efficacy is M. tuberculosis — both antigen-dependent macrophages44 may explain why some commercially available. Although cell- and antigen-independent (Figs. 1,2). But studies showed that opsonization was mediated immunity was sufficient for which should be used to test for protective protective2,45 but others did not3,4. clearance of attenuated S. Typhimurium antibody production? It is far from clear A whole blood assay allowed us to in a murine infection model, antibody which of these mechanisms might be interrogate mediators beyond single was required for clearance of virulent physiologically relevant, as in the example of
594 | SEPTEMBER 2018 | volume 18 www.nature.com/nri Perspectives
opsonophagocytosis. Most groups have used correlates of protection and potent in vitro Hao Li and Babak Javid * M. tuberculosis cell-infection models to test and in vivo efficacy of broadly neutralizing Centre for Global Health and Infectious Diseases, antibody-mediated protection in vitro2,4,5, antibodies, eliciting such protective Collaborative Innovation Centre for the Diagnosis and but this approach may miss more complex antibodies by immunization has been Treatment of Infectious Diseases, Tsinghua University School of Medicine, Beijing, China. interactions that involve other components extremely problematic. The TB field has yet *e-mail: [email protected] of the immune system3. Relatively unbiased to pass the first two hurdles, but the urgency approaches such as systems serology may be of the global TB problem has meant that https://doi.org/10.1038/s41577-018-0028-0 Published online 5 June 2018 able to identify which antibody functions more than a dozen clinical and hundreds of are most critical for protection — but preclinical vaccine candidates have been 1. Kaufmann, S. H. et al. Progress in tuberculosis vaccine development and host-directed therapies—a state of lack of a preclinical gold standard will proposed. Although the vast majority have the art review. Lancet Respir. Med. 2, 301–320 require validation in multiple models focused on cell-mediated immunity, (2014). 2. Chen, T. et al. Association of human antibodies before the field can rely on in vitro a number of candidates also elicit humoral to arabinomannan with enhanced mycobacterial correlates. responses1,57. Other considerations are opsonophagocytosis and intracellular growth reduction. J. Infecti. Diseases 214, 300–310 (2016). which models to test for protective antibody 3. Li, H. et al. Latently and uninfected healthcare workers Are antibodies sufficient for protection? generation and efficacy? Even non-human exposed to TB make protective antibodies against 58 Mycobacterium tuberculosis. Proc. Natl Acad. Sci. The relative magnitude of M. tuberculosis primate models can be highly variable . USA 114, 5023–5028 (2017). growth restriction of naturally occurring Increasingly, human challenge models are 4. Lu, L. L. et al. A functional role for antibodies in 3 59–62 tuberculosis. Cell 167, 433–443.e414 (2016). human protective antibodies both in vivo being used for testing vaccine efficacy 5. Zimmermann, N. et al. Human isotype-dependent and in vitro2–5 was less than one order of and, although still some way off, are now inhibitory antibody responses against Mycobacterium tuberculosis. EMBO Mol. Med. 8, 1325–1339 magnitude, raising questions about the starting to be developed for TB. Adjuvant (2016). physiological importance of antibody- choice also has a major impact on the 6. Mangtani, P. et al. Protection by BCG vaccine against tuberculosis: a systematic review of randomized mediated protection. Furthermore, quality and not just on the magnitude of controlled trials. Clin. Infect. Dis. 58, 470–480 while one study used a clinical isolate the provoked immune response63–65, further (2014). 3,51 7. World Health Organization. Global Tuberculosis of M. tuberculosis , most studies have complicating the number of parameters that Report 2017. WHO http://www.who.int/tb/ employed the laboratory-adapted strain need to be tested. publications/global_report/en/ (2017). 8. Casadevall, A. & Scharff, M. D. Return to the past: H37Rv, and no studies have tested multiple the case for antibody-based therapies in infectious clinical isolates. It is thought that natural Concluding remarks diseases. Clin. Infect. Dis. 21, 150–161 (1995). 9. Glatman-Freedman, A. & Casadevall, A. Serum transmission of M. tuberculosis is usually in Re-examining humoral immunity to TB therapy for tuberculosis revisited: reappraisal of the microdroplets containing one bacillus52,53, in the context of naturally arising human role of antibody-mediated immunity against Mycobacterium tuberculosis. Clin. Microbiol. Rev. 11, while the typical murine challenge model antibodies from individual donors has led 514–532 (1998). infects mice with 100–200 bacilli per to renewed interest in antibody-mediated 10. Cooper, A. M. Mouse model of tuberculosis. Cold Spring Harb. Perspect. Med. 5, a018556 (2015). animal. In the HIV field, shifting to more immunity to M. tuberculosis infection, but 11. Havlir, D. V. & Barnes, P. F. Tuberculosis in patients physiological ‘low-dose’ challenge models this enthusiasm needs to be tempered with with human immunodeficiency virus infection. N. Engl. J. Med. 340, 367–373 (1999). resulted in more reliable assessment of a sober realization of the many hurdles 12. Maglione, P. J., Xu, J. & Chan, J. B cells moderate 50,54 vaccine candidates . Such low-dose that need to be overcome before we can inflammatory progression and enhance bacterial containment upon pulmonary challenge with aerosol challenges have been developed for determine whether an antibody-based Mycobacterium tuberculosis. J. Immunol. 178, 53 murine M. tuberculosis infection but are vaccine can be developed to prevent 7222–7234 (2007). not widely used. Our observation that CD4+ active TB. Development of vaccines that 13. Vordermeier, H. M., Venkataprasad, N., Harris, D. P. & Ivanyi, J. Increase of tuberculous infection in the T cells are required for antibody efficacy target both cell-mediated and humoral organs of B cell-deficient mice. Clin. Exp. Immunol. may also mean that humanized mice55,56, immunity57 may prove the most effective 106, 312–316 (1996). 14. Phuah, J. Y., Mattila, J. T., Lin, P. L. & Flynn, J. L. or non-human primates whose immune approach, with some already in late Activated B cells in the granulomas of nonhuman systems would be better able to crosstalk clinical trials66. primates infected with Mycobacterium tuberculosis. Am. J. Pathol. 181, 508–514 (2012). with human antibodies, may be superior Are there interim milestones that can be 15. Phuah, J. et al. Effects of B cell depletion on early to inbred rodents for testing antibody tested more readily? Therapeutic antibodies Mycobacterium tuberculosis infection in cynomolgus macaques. Infect. Immun. 84, 1301–1311 (2016). 67 efficacy. Coupling these approaches are effective in HIV and other infections , 16. Achkar, J. M. & Casadevall, A. Antibody-mediated with culturing multiple clinical isolates and killed whole-cell vaccines are being immunity against tuberculosis: implications for vaccine development. Cell Host Microbe 13, 250–262 (2013). of M. tuberculosis with intact capsule, developed as adjuncts to antibiotic therapy 17. Balu, S. et al. A novel human IgA monoclonal antibody which has been associated with increased in TB68. Such approaches are unlikely to protects against tuberculosis. J. Immunol. 186, 3,34 3113–3119 (2011). protective responses , may allow be cost-effective for the treatment of the 18. Buccheri, S. et al. Prevention of the post- researchers to determine whether vaccine majority of TB cases — after all, TB is a chemotherapy relapse of tuberculous infection by combined immunotherapy. Tuberculosis 89, 91–94 candidates can truly elicit sterilizing disease of poverty and prevalent in resource- (2009). protective immunity. limited settings. However, adjunctive therapy 19. Hamasur, B. et al. A mycobacterial lipoarabinomannan specific monoclonal antibody and its F(ab’) fragment with therapeutic monoclonal antibodies may prolong survival of mice infected with Mycobacterium Can vaccination elicit protective antibody have a role in the treatment of extensively tuberculosis. Clin. Exp. Immunol. 138, 30–38 (2004). 20. Pethe, K. et al. The heparin-binding haemagglutinin of responses? Generation of antibodies from drug-resistant TB, which is far more difficult M. tuberculosis is required for extrapulmonary natural infection and exposure and those and expensive to treat than standard TB dissemination. Nature 412, 190–194 (2001). 21. Teitelbaum, R. et al. A mAb recognizing a surface elicited by immunization can be vastly cases. It would perhaps be fitting if the first antigen of Mycobacterium tuberculosis enhances host different in terms of both function and ease clinical application of our recent advances survival. Proc. Natl Acad. Sci. USA 95, 15688–15693 (1998). of generation. Although the HIV experience in understanding M. tuberculosis antibody 22. Guirado, E. et al. Passive serum therapy with may be useful in determining priorities for responses was the use of the 21st century polyclonal antibodies against Mycobacterium tuberculosis protects against post-chemotherapy antibody research in TB, it also sounds a equivalent of serum therapy to fight the relapse of tuberculosis infection in SCID mice. note of caution. Even with excellent in vitro old enemy TB. Microbes Infect. 8, 1252–1259 (2006).
NaTuRe ReviewS | Immunology volume 18 | SEPTEMBER 2018 | 595 Perspectives
23. Roy, E. et al. Therapeutic efficacy of high-dose lipoarabinomannan. PLOS One 12, e0185945 57. Gengenbacher, M., Nieuwenhuizen, N. E. & intravenous immunoglobulin in Mycobacterium (2017). Kaufmann, S. BCG — old workhorse, new skills. tuberculosis infection in mice. Infect. Immun. 73, 40. Bournazos, S. et al. Broadly neutralizing anti-HIV-1 Curr. Opin. Immunol. 47, 8–16 (2017). 6101–6109 (2005). antibodies require Fc effector functions for in vivo 58. Maiello, P. et al. Rhesus macaques are more 24. Forget, A., Benoit, J. C., Turcotte, R. & activity. Cell 158, 1243–1253 (2014). susceptible to progressive tuberculosis than Gusew-Chartrand, N. Enhancement activity of 41. Olivares, N. et al. The protective effect of cynomolgus macaques: a quantitative comparison. anti-mycobacterial sera in experimental immunoglobulin in murine tuberculosis is dependent Infect. Immun. 86, e00505-17 (2017). Mycobacterium bovis (BCG) infection in mice. on IgG glycosylation. Pathog. Dis. 69, 176–183 59. Sauerwein, R. W., Roestenberg, M. & Moorthy, V. S. Infect. Immun. 13, 1301–1306 (1976). (2013). Experimental human challenge infections can 25. Tameris, M. D. et al. Safety and efficacy of MVA85A, 42. Li, H., Wu, M., Shi, Y. & Javid, B. Over-expression of accelerate clinical malaria vaccine development. a new tuberculosis vaccine, in infants previously the mycobacterial trehalose-phosphate phosphatase Nat. Rev. Immunol. 11, 57–64 (2011). vaccinated with BCG: a randomised, OtsB2 results in a defect in macrophage phagocytosis 60. Collins, A. M. et al. First human challenge testing placebo-controlled phase 2b trial. Lancet 381, associated with increased mycobacterial-macrophage of a pneumococcal vaccine. Double-blind randomized 1021–1028 (2013). adhesion. Front. Microbiol. 7, 1754 (2016). controlled trial. Am. J. Respiratory Crit. Care Med. 26. Beveridge, N. E. et al. Immunisation with BCG and 43. Bogaert, D., Sluijter, M., De Groot, R. & Hermans, P. W. 192, 853–858 (2015). recombinant MVA85A induces long-lasting, Multiplex opsonophagocytosis assay (MOPA): 61. Dobinson, H. C. et al. Evaluation of the clinical polyfunctional Mycobacterium tuberculosis-specific a useful tool for the monitoring of the 7-valent and microbiological response to Salmonella CD4+ memory T lymphocyte populations. Eur. pneumococcal conjugate vaccine. Vaccine 22, Paratyphi A infection in the first paratyphoid human J. Immunol. 37, 3089–3100 (2007). 4014–4020 (2004). challenge model. Clin. Infect. Dis. 64, 1066–1073 27. Beverley, P. Selective presentation of MVA85A 44. Armstrong, J. A. & Hart, P. D. Phagosome-lysosome (2017). tuberculosis booster vaccine preclinical animal data. interactions in cultured macrophages infected with 62. Kirkpatrick, B. D. et al. The live attenuated dengue Int. J. Epidemiol. 45, 581–582 (2016). virulent tubercle bacilli. Reversal of the usual vaccine TV003 elicits complete protection against 28. Sakai, S. et al. CD4 T cell-derived IFN-gamma plays a nonfusion pattern and observations on bacterial dengue in a human challenge model. Sci. Transl Med. minimal role in control of pulmonary Mycobacterium survival. J. Exp. Med. 142, 1–16 (1975). 8, 330ra336 (2016). tuberculosis infection and must be actively repressed 45. Kumar, S. K., Singh, P. & Sinha, S. Naturally produced 63. Garcon, N., Heppner, D. G. & Cohen, J. Development by PD-1 to prevent lethal disease. PLoS Pathog. 12, opsonizing antibodies restrict the survival of of RTS,S/AS02: a purified subunit-based malaria e1005667 (2016). Mycobacterium tuberculosis in human macrophages vaccine candidate formulated with a novel adjuvant. 29. Burton, D. R. & Hangartner, L. Broadly neutralizing by augmenting phagosome maturation. Open Biol. 5, Expert Rev. Vaccines 2, 231–238 (2003). antibodies to HIV and Their role in vaccine design. 150171 (2015). 64. Ackerman, M. E., Barouch, D. H. & Alter, G. Systems Annu. Rev. Immunol. 34, 635–659 (2016). 46. Wahid, R. et al. Live oral Salmonella enterica serovar serology for evaluation of HIV vaccine trials. Immunol. 30. Costello, A. M. et al. Does antibody to mycobacterial Typhi vaccines Ty21a and CVD 909 induce Rev. 275, 262–270 (2017). antigens, including lipoarabinomannan, limit opsonophagocytic functional antibodies in humans 65. Vaccari, M. et al. Adjuvant-dependent innate dissemination in childhood tuberculosis? Trans. R. Soc. that cross-react with S. Paratyphi A and S. Paratyphi B. and adaptive immune signatures of risk of Trop. Med. Hyg. 86, 686–692 (1992). Clin. Vaccine Immunol. 21, 427–434 (2014). SIVmac251 acquisition. Nature Med. 22, 762–770 31. Chu, H. et al. Risk of tuberculosis among healthcare 47. McSorley, S. J. & Jenkins, M. K. Antibody is required (2016). workers in an intermediate-burden country: a for protection against virulent but not attenuated 66. Nieuwenhuizen, N. E. et al. The recombinant nationwide population study. J. Infect. 69, 525–532 Salmonella enterica serovar typhimurium. Infect. Bacille Calmette-Guerin vaccine VPM1002: ready for (2014). Immun. 68, 3344–3348 (2000). clinical efficacy testing. Front. Immunol. 8, 1147 32. Zhou, F. et al. Latent tuberculosis infection and 48. van Els, C. et al. Fast vaccine design and development (2017). occupational protection among health care workers in based on correlates of protection (COPs). Hum. Vaccin 67. Salazar, G., Zhang, N., Fu, T. M. & An, Z. Antibody two types of public hospitals in China. PLOS One 9, Immunother. 10, 1935–1948 (2014). therapies for the prevention and treatment of viral e104673 (2014). 49. Corey, L. et al. Immune correlates of vaccine infections. NPJ Vaccines 2, 19 (2017). 33. Pai, M. et al. Tuberculosis. Nat. Rev. Dis. Primers 2, protection against HIV-1 acquisition. Science Transl 68. Cardona, P. J. The progress of therapeutic vaccination 16076 (2016). Med. 7, 310rv7 (2015). with regard to tuberculosis. Front. Microbiol. 7, 1536 34. Prados-Rosales, R. et al. The type of growth medium 50. Karp, C. L., Wilson, C. B. & Stuart, L. M. Tuberculosis (2016). affects the presence of a mycobacterial capsule and is vaccines: barriers and prospects on the quest for a associated with differences in protective efficacy of transformative tool. Immunol. Rev. 264, 363–381 Acknowledgements BCG vaccination against Mycobacterium tuberculosis. (2015). This work was in part funded by a grant from the National J. Infecti. Diseases 214, 426–437 (2016). 51. Su, H. W. et al. The essential mycobacterial Science Foundation of China (81661128044) to B.J. B.J. is a 35. Cole, S. T. et al. Deciphering the biology of amidotransferase GatCAB is a modulator of specific Wellcome Trust Investigator. Mycobacterium tuberculosis from the complete translational fidelity. Nat. Microbiol. 1, 16147 (2016). genome sequence. Nature 393, 537–544 (1998). 52. Riley, R. L. et al. Aerial dissemination of pulmonary Author contributions 36. Sani, M. et al. Direct visualization by cryo-EM of the tuberculosis. A two-year study of contagion in a Both authors contributed to researching data, discussion of mycobacterial capsular layer: a labile structure tuberculosis ward. 1959. Am. J. Epidemiol. 142, content and the writing of the article. B.J. reviewed and containing ESX-1-secreted proteins. PLoS Pathog. 6, 3–14 (1995). edited the manuscript before submission. e1000794 (2010). 53. Saini, D. et al. Ultra-low dose of Mycobacterium 37. Deng, J. et al. Mycobacterium tuberculosis proteome tuberculosis aerosol creates partial infection in mice. Competing interests microarray for global studies of protein function and Tuberculosis 92, 160–165 (2012). The authors declare no competing interests. immunogenicity. Cell Rep. 9, 2317–2329 (2014). 54. Gautam, R. et al. A single injection of anti-HIV-1 38. Prados-Rosales, R. et al. Enhanced control of antibodies protects against repeated SHIV challenges. Publisher’s note Mycobacterium tuberculosis extrapulmonary Nature 533, 105–109 (2016). Springer Nature remains neutral with regard to jurisdictional dissemination in mice by an arabinomannan-protein 55. Calderon, V. E. et al. A humanized mouse model of claims in published maps and institutional affiliations. conjugate vaccine. PLoS Pathog. 13, e1006250 tuberculosis. PLOS One 8, e63331 (2013). (2017). 56. Grover, A. et al. Humanized NOG mice as a model for Reviewer information 39. Shin, H. J., Franco, L. H., Nair, V. R., Collins, A. C. & tuberculosis vaccine-induced immunity: a comparative Nature Reviews Immunology thanks S. Kaufmann and the Shiloh, M. U. A baculovirus-conjugated mimotope analysis with the mouse and guinea pig models of other, anonymous reviewer(s) for their contribution to vaccine targeting Mycobacterium tuberculosis tuberculosis. Immunology 152, 150–162 (2017). the peer review of this work.
596 | SEPTEMBER 2018 | volume 18 www.nature.com/nri