Anti-TNF immunotherapy and tuberculosis reactivation: another mechanism revealed

Abstract

Anti-TNF immunotherapy has revolutionized the treatment of some inflammatory diseases, such as RA. However, a major concern is that patients receiving this therapy have an increased risk of fungal and bacterial infection, particularly of reactivating latent tuberculosis (TB). In this issue of the JCI, in an effort to understand how anti-TNF immunotherapy affects host mechanisms required to control TB, Bruns and colleagues examined the effects of the anti-TNF therapeutic infliximab on Mycobacterium tuberculosis–specific human lymphocytes (see the related article beginning on page 1167). The authors report that a granulysin-expressing CD45RA⁺ subset of effector memory CD8⁺ T cells that contributes to the killing of intracellular M. tuberculosis is depleted in vivo by infliximab in patients with RA, and that these cells are susceptible to complement-mediated lysis in the presence of infliximab in vitro. The study provides insight into host defense mechanisms that act to control TB infection and how they are affected during anti-TNF immunotherapy for autoimmune disease.

TNF is a pleiotropic cytokine that plays a critical but incompletely understood role in immunity to Mycobacterium tuberculosis and other intracellular bacterial and fungal pathogens. TNF is not only essential for immune control of tuberculosis (TB), it has also been implicated in the immunopathology of the disease (1). While considerable knowledge of the actions of TNF in immunity to M. tuberculosis has been gained from studies in animal models, we have also gained a deeper understanding of TNF’s contributions to the control of TB in humans through the use of TNF-neutralizing drugs for certain chronic inflammatory diseases. While these agents are highly efficacious for the treatment of RA, ankylosing spondylitis, psoriatic arthritis, and Crohn disease, they also promote reactivation (and possibly acquisition) of intracellular pathogens, including M. tuberculosis, resulting in potentially life-threatening infections. Since the early reports of TB in patients treated with infliximab, an anti-TNF monoclonal antibody, described abnormalities in granulomas (2), much of the work to elucidate the mechanisms of TB reactivation has focused on the roles of TNF in the formation and maintenance of these structures. Granulomas are thought to contribute to control of intramacrophage pathogens by providing a local environment in which APCs and lymphocytes interact to suppress progressive growth of the pathogen. In the current issue of the JCI, Bruns et al. describe an additional mechanism for susceptibility to progressive TB in individuals treated with infliximab (3). Using PBMCs from tuberculin skin test–positive subjects, Bruns et al. identified granulysin-rich CD8⁺CD45RA⁺CCR7^– effector memory T cells (CD8⁺ T_EMRA cells) as a major subset of antimycobacterial effector cells and found that T_EMRA cells are selectively depleted by infliximab therapy in patients with RA. Here we discuss the impact of these findings as they relate to the existing knowledge of CD8⁺ T cells and TNF in immunity to M. tuberculosis.

CD8+ T cells and immunity to M. tuberculosis

Cell-mediated immunity is critical for control of M. tuberculosis infection, and it has long been acknowledged that CD4⁺ T cells are important mediators of immunity to M. tuberculosis. More recently, M. tuberculosis–specific CD8⁺ T cells have been increasingly recognized, but their significance remains incompletely understood, especially in humans. CD8⁺ T cells are thought to limit mycobacterial growth directly through the killing of infected cells (4, 5), as well as indirectly through secretion of cytokines that promote activation of macrophages and chemokines that coordinate cell recruitment. Mice deficient in MHC class I molecules, and thus deficient in CD8⁺ T cells due to the absence of positive selection, have higher burdens of M. tuberculosis in the lungs and a modest decrease in survival (6). This is in contrast to CD4⁺ T cell–deficient mice, which display more rapidly progressive bacterial growth and a sharp decline in survival (6). These studies and others imply that CD8⁺ T cells are less crucial during the acute phase of M. tuberculosis infection, but are indispensable during the chronic phase of infection, and therefore may help prevent reactivation of TB (7, 8). When evaluating the contribution of CD8⁺ T cells to M. tuberculosis immunity, it is important to note that mice lack granulysin, a cytolytic granule protein that contributes to killing of M. tuberculosis by human CD8⁺ T cells (4). Therefore, murine studies may undervalue the importance of the antimycobacterial properties of CD8⁺ T cell subsets compared with their roles in human immunity. Human studies are more limited in number, but in vitro experiments provide evidence that CD8⁺ T cells can control M. tuberculosis through killing of infected macrophages and subsequent death of the bacteria, as well as by direct killing of mycobacteria by secreted granulysin (4, 5, 9). Bruns et al. (3) found in their assays that T_EMRA cells stain for granulysin with high frequency and display the highest levels of both cytotoxicity and antimycobacterial activity compared with other T cell subsets.

TNF blockade and progression of M. tuberculosis

The authors go beyond defining T_EMRA cells as potentially important antimycobacterial effector cells, to address the consequences of TNF blockade on these cells and on mycobacterial growth in cultured cells (3). In doing so, they reveal what we believe to be a novel potential mechanism for TNF neutralization to promote progression of TB. Previous studies in mice and in zebrafish embryos have revealed variable increases in bacillary load in the absence of TNF, ranging from small changes to multiple orders of magnitude (10–13). These observations indicated that TNF contributes directly to control of mycobacteria in granuloma macrophages (10) and may also function indirectly to suppress mycobacterial growth by modulating formation and/or maintenance of granulomas. Additional evidence suggests that TNF blockade may impair mycobacterial control as a result of effects on T lymphocytes. TNF blockade has previously been shown to induce production of Tregs via TGF-β in patients with RA (14, 15), and in mice infected with M. tuberculosis, depletion of Tregs leads to improved control of mycobacterial growth (16). Bruns et al. provide evidence that TNF blockade may also promote reactivation of TB through selective reduction of T_EMRA cells (3). As a mechanism for their observation that T_EMRA cells in peripheral blood are depleted following infliximab therapy, the authors demonstrate that T_EMRA cells from RA patients possess high levels of membrane-bound (presumably transmembrane) TNF compared with healthy controls and are susceptible to in vitro complement-mediated lysis in the presence of infliximab. Additionally, PBMCs from patients receiving infliximab therapy displayed diminished capacity to limit mycobacterial growth in culture, an impairment that was rescued by the addition of autologous CD8⁺ T_EMRA cells, to a slightly lesser extent by CD8⁺CCR7^–CD45RA^– effector memory T cells, but not by circulating central memory CD8⁺ T cells.

Implications for future study

As with any provocative results, the findings reported by Bruns et al. (3) raise several questions, with implications for understanding immunity to TB and with practical application to clinical medicine. First, do infliximab and etanercept differ in their ability to deplete T_EMRA cells? Infliximab, a monoclonal antibody to TNF, and etanercept, a soluble TNF receptor, are both TNF-neutralizing agents used in the treatment of RA and other inflammatory conditions. For unclear reasons, infliximab causes significantly more infectious complications, including reactivation of M. tuberculosis, than does etanercept; this has been attributed by some to the differential ability of these agents to bind membrane TNF. Several studies have found that apoptosis ensues upon binding of infliximab to membrane TNF on T cells (17), monocytes (18), and cells transfected to express membrane TNF (19, 20), which may lead to a reduction in the number of antimycobacterial effector cells and/or to dissolution of granulomas. In contrast, etanercept does not cause apoptosis of cells that express membrane TNF (17, 19, 20). Additionally, cell lines transfected with membrane TNF have been used to evaluate complement-dependent cytotoxicity (CDC) and antibody-dependent, cell-mediated cytotoxicity (ADCC); while both etanercept and infliximab lead to ADCC, only infliximab induces CDC (20, 21). Consistent with these results, Bruns et al. (3) revealed that T_EMRA cells are susceptible to CDC in the presence of infliximab (Figure 1). If etanercept does not cause CDC of T_EMRA cells, this outcome would suggest that the differential risk of reactivation of TB with these agents may be caused, at least in part, by relative differences in depletion of T_EMRA cells. Likewise, if etanercept does not deplete T_EMRA cells to the same extent as does infliximab, this could at least partially explain the therapeutic differences observed with these 2 drugs. While both drugs are equally efficacious for the treatment of RA, infliximab is superior to etanercept in the treatment of other chronic inflammatory diseases, especially Crohn disease (22). If selective reduction of T_EMRA cells does not occur following etanercept therapy, this finding could provide insight into differences in the immunopathology of these inflammatory conditions. In contrast, if etanercept does deplete T_EMRA cells to the same extent as does infliximab, it would weaken the case that T_EMRA cells are an essential element of protective immunity against TB.

Figure 1

Effect of TNF neutralization with infliximab on the antimycobacterial action of CD8⁺ T_EMRA cells. In the absence of the TNF-neutralizing drug infliximab (i), cytoxic T_EMRA cells are present and release their granules containing perforin and granulysin, resulting in the death of M. tuberculosis–infected macrophages and intracellular and extracellular mycobacteria. In this issue of the JCI, Bruns et al. (3) report that in the presence of infliximab (ii), membrane TNF on T_EMRA cells is bound by the antibody, and CDC ensues. The depletion of T_EMRA cells results in suboptimal control of mycobacterial growth, leading to the potential spread of M. tuberculosis infection.

Second, how long does infliximab-mediated depletion of T_EMRA cells persist? The frequency of progressive TB in people treated with infliximab is highest in the first 90 days after initiating the therapy (23), but the studies by Bruns et al. were limited to a single time point 2 weeks after initiating therapy (3). Additional studies of the duration of depletion of T_EMRA cells after anti-TNF therapy should shed further light on the roles of these cells in protection against TB.

Third, what are the antigens recognized by T_EMRA cells, and what determines differentiation of CD8⁺ effector cells into T_EMRA cells rather than CD45RA^– effector memory T cells? Answers to these questions could provide guidance in the design of improved TB vaccines by delineating optimal antigens and adjuvants and might also provide insight into the mechanisms of differential susceptibility and resistance to TB in populations and in individuals. In addition, they may identify Ag-specific T_EMRA cells as potential surrogate markers of vaccine efficacy and thereby provide improved predictive information in the design and evaluation of clinical trials of novel TB vaccines.

Finally, while the findings reported by Bruns et al. clearly demonstrate that T_EMRA cells are targets of the anti-TNF action of infliximab (3), they do not provide proof that TNF has an essential function in the development, maintenance, or effector functions of this interesting subset of CD8⁺ T cells. If membrane TNF is simply a bystander, then future development of agents that block TNF activities without depleting T_EMRA cells may provide for safer therapy of chronic inflammatory diseases.

Footnotes

Conflict of interest: The authors have declared that no conflict of interest exists.

Nonstandard abbreviations used: CDC, complement-dependent cytotoxicity; TB, tuberculosis; T_EMRA, CD45RA⁺ effector memory T (cell).

Reference information: J. Clin. Invest.119:1079–1082 (2009). doi:10.1172/JCI39143

See the related article at Anti-TNF immunotherapy reduces CD8⁺ T cell–mediated antimicrobial activity against Mycobacterium tuberculosis in humans.

References

1. Mohan, V.P., et al. 2001. Effects of tumor necrosis factor alpha on host immune response in chronic persistent tuberculosis: possible role for limiting pathology. Infect. Immun. 69:1847-1855.
   View this article via: CrossRef PubMed Google Scholar
2. Keane, J., et al. 2001. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N. Engl. J. Med. 345:1098-1104.
   View this article via: CrossRef PubMed Google Scholar
3. Bruns, H., et al. 2009. Anti-TNF immunotherapy reduces CD8⁺ T cell–mediated antimicrobial activity against Mycobacterium tuberculosis in humans. J. Clin. Invest. 119:1167-1177.
   View this article via: JCI Google Scholar
4. Stenger, S., et al. 1998. An antimicrobial activity of cytolytic T cells mediated by granulysin. Science. 282:121-125.
   View this article via: CrossRef PubMed Google Scholar
5. Stenger, S., et al. 1997. Differential effects of cytolytic T cell subsets on intracellular infection. Science. 276:1684-1687.
   View this article via: CrossRef PubMed Google Scholar
6. Mogues, T., Goodrich, M.E., Ryan, L., LaCourse, R., North, R.J. 2001. The relative importance of T cell subsets in immunity and immunopathology of airborne Mycobacterium tuberculosis infection in mice. J. Exp. Med. 193:271-280.
   View this article via: CrossRef PubMed Google Scholar
7. Ordway, D., et al. 2007. XCL1 (lymphotactin) chemokine produced by activated CD8 T cells during the chronic stage of infection with Mycobacterium tuberculosis negatively affects production of IFN-gamma by CD4 T cells and participates in granuloma stability. J. Leukoc. Biol. 82:1221-1229.
   View this article via: CrossRef PubMed Google Scholar
8. van Pinxteren, L.A., Cassidy, J.P., Smedegaard, B.H., Agger, E.M., Andersen, P. 2000. Control of latent Mycobacterium tuberculosis infection is dependent on CD8 T cells. Eur. J. Immunol. 30:3689-3698.
   View this article via: CrossRef PubMed Google Scholar
9. Carranza, C., et al. 2006. Mycobacterium tuberculosis growth control by lung macrophages and CD8 cells from patient contacts. Am. J. Respir. Crit. Care Med. 173:238-245.
   View this article via: PubMed Google Scholar
10. Clay, H., Volkman, H.E., Ramakrishnan, L. 2008. Tumor necrosis factor signaling mediates resistance to mycobacteria by inhibiting bacterial growth and macrophage death. Immunity. 29:283-294.
    View this article via: CrossRef PubMed Google Scholar
11. Flynn, J.L., et al. 1995. Tumor necrosis factor-alpha is required in the protective immune response against Mycobacterium tuberculosis in mice. Immunity. 2:561-572.
    View this article via: CrossRef PubMed Google Scholar
12. Kindler, V., Sappino, A.P., Grau, G.E., Piguet, P.F., Vassalli, P. 1989. The inducing role of tumor necrosis factor in the development of bactericidal granulomas during BCG infection. Cell. 56:731-740.
    View this article via: CrossRef PubMed Google Scholar
13. Zganiacz, A., et al. 2004. TNF-alpha is a critical negative regulator of type 1 immune activation during intracellular bacterial infection. J. Clin. Invest. 113:401-413.
    View this article via: JCI PubMed Google Scholar
14. Ehrenstein, M.R., et al. 2004. Compromised function of regulatory T cells in rheumatoid arthritis and reversal by anti-TNFalpha therapy. J. Exp. Med. 200:277-285.
    View this article via: CrossRef PubMed Google Scholar
15. Nadkarni, S., Mauri, C., Ehrenstein, M.R. 2007. Anti-TNF-alpha therapy induces a distinct regulatory T cell population in patients with rheumatoid arthritis via TGF-beta. J. Exp. Med. 204:33-39.
    View this article via: CrossRef PubMed Google Scholar
16. Scott-Browne, J.P., et al. 2007. Expansion and function of Foxp3-expressing T regulatory cells during tuberculosis. J. Exp. Med. 204:2159-2169.
    View this article via: CrossRef PubMed Google Scholar
17. Van den Brande, J.M., et al. 2003. Infliximab but not etanercept induces apoptosis in lamina propria T-lymphocytes from patients with Crohn’s disease. Gastroenterology. 124:1774-1785.
    View this article via: CrossRef PubMed Google Scholar
18. Lugering, A., et al. 2001. Infliximab induces apoptosis in monocytes from patients with chronic active Crohn’s disease by using a caspase-dependent pathway. Gastroenterology. 121:1145-1157.
    View this article via: CrossRef PubMed Google Scholar
19. Mitoma, H., et al. 2005. Infliximab induces potent anti-inflammatory responses by outside-to-inside signals through transmembrane TNF-alpha. Gastroenterology. 128:376-392.
    View this article via: CrossRef PubMed Google Scholar
20. Mitoma, H., et al. 2008. Mechanisms for cytotoxic effects of anti-tumor necrosis factor agents on transmembrane tumor necrosis factor alpha-expressing cells: comparison among infliximab, etanercept, and adalimumab. Arthritis Rheum. 58:1248-1257.
    View this article via: CrossRef PubMed Google Scholar
21. Scallon, B.J., Moore, M.A., Trinh, H., Knight, D.M., Ghrayeb, J. 1995. Chimeric anti-TNF-alpha monoclonal antibody cA2 binds recombinant transmembrane TNF-alpha and activates immune effector functions. Cytokine. 7:251-259.
    View this article via: CrossRef PubMed Google Scholar
22. Sandborn, W.J., et al. 2001. Etanercept for active Crohn’s disease: a randomized, double-blind, placebo-controlled trial. Gastroenterology. 121:1088-1094.
    View this article via: CrossRef PubMed Google Scholar
23. Dinarello, C.A. 2005. Differences between anti-tumor necrosis factor-alpha monoclonal antibodies and soluble TNF receptors in host defense impairment. J. Rheumatol. Suppl. 74:40-47.
    View this article via: PubMed Google Scholar