News & Views | Published:


Fixing the odds against tuberculosis

Nature volume 511, pages 3940 (03 July 2014) | Download Citation

A study that defines the interactions between three immunity-regulating molecules — type 1 interferon, interleukin-1 and prostaglandin E2 — will help to explain the variable outcomes of tuberculosis infections. See Letter p.99

Billions of people worldwide are infected with Mycobacterium tuberculosis. However, only 5–10% of these individuals will develop clinically evident disease — a likelihood similar to winning a game of roulette by betting on a single number. Many genetic and environmental variables are known to increase the risk of progression to clinical tuberculosis (TB). Understanding how these factors interact to determine the outcome of infection could facilitate the design of better therapies and the targeting of their use to those at highest risk. On page 99 of this issue, Mayer-Barber et al.1 describe an immune network that may have a central role in determining TB outcomes and could be manipulated to tip the odds in favour of host resistance.

Numerous mediators of immunity have been implicated in host resistance to TB. These regulators of the immune response come in many forms, including secreted proteins, such as cytokines, and lipid-derived signalling molecules, such as eicosanoids. Most of these were discovered when their complete removal or marked overproduction altered the outcome of infection. Such studies2,3,4 identified roles for the cytokines interleukin-1 (IL-1) and type I interferon (IFN) and the eicosanoid prostaglandin E2 (PGE2) in the pathogenesis of TB.

Mayer-Barber et al. take the step of placing these crucial players into their regulatory context, and discover that they form a cross-regulatory network. In M. tuberculosis-infected mice, the authors find that IL-1 induces PGE2, which enhances the antimicrobial activity of immune cells called macrophages. IFN inhibits the production and activity of IL-1 and is itself inhibited by PGE2 (Fig. 1). This pathway provides a mechanistic framework to explain the previous observations that IFN exacerbates TB5, whereas IL-1 and PGE2 are generally protective2,4. The mechanisms by which IL-1 and PGE2 inhibit bacterial growth remain to be defined, but are probably related, at least in part, to the inhibition of macrophage necrotic cell death6.

Figure 1: Factors influencing the outcome of tuberculosis infection.
Figure 1

Mayer-Barber et al.1 provide evidence from animal models and human clinical data that the depicted network of immune-regulatory molecules influences the gamble that is infection with Mycobacterium tuberculosis. a, Type I interferon (IFN) promotes disease by inducing interleukin-10 (IL-10) and IL-1 receptor antagonist (IL-1RA). These mediators inhibit IL-1-dependent prostaglandin E2 (PGE2), which would otherwise restrict the intracellular growth of the bacteria. b, Although factors that increase the risk of developing tuberculosis probably act by diverse mechanisms, many could influence the activity or balance of this immune circuit, particularly by accentuating IFN production. c, By contrast, Mayer-Barber et al. show that PGE2 supplementation can reduce disease severity in infected mice that have high IFN levels.

The authors also observed that alterations in this IL-1–PGE2–IFN network were associated with more severe disease in humans, providing validation for their animal studies. These clinical data significantly further our ability to delineate distinct disease states using immunological markers. Previous studies demonstrated an association between IFN responses and active TB in humans7. Adding other members of this immune network to the analysis has refined the resolution of this approach, allowing Mayer-Barber and colleagues to differentiate patients with mild disease from those classified as asymptomatic. This type of rationally designed biomarker panel may prove useful in the design of therapeutic trials.

The proposed IL-1–PGE2–IFN pathway could be interpreted as a feedback loop that maintains homeostasis in the body, because excess activity of each mediator triggers an inhibitory response, returning the system to balance. However, the situation is probably more complicated during chronic TB. For example, the kinetics of each mediator's expression may be crucial. IFN production tends to be rapid and transient, whereas IL-1 is typically associated with chronic inflammatory states. Indeed, although IL-1 has a protective role during early infection, its persistent production causes disease-associated symptoms8.

In addition, IFN expression is remarkably stochastic at the single-cell level9, suggesting that IFN-expressing cells could be unevenly distributed in the lung tissue. The observation that different TB lesions in the same lung evolve independently10 supports the idea that the spatial distribution of these mediators is functionally variable. Perhaps most significantly, these innate immune responses influence and are influenced by the T-cell-based adaptive immune response, which can vary throughout a prolonged chronic infection. Thus, although IL-1, PGE2 and IFN can be depicted together as a simple, self-regulating pathway, any number of factors could tip the balance away from protective PGE2 and towards pathological IFN production.

Several genetic and environmental variables are known to increase the odds of developing TB, and might act by modulating the IL-1–PGE2–IFN network. Pre-existing genetic variation in both the host and the pathogen probably have significant roles. For example, highly pathogenic strains of M. tuberculosis are known to express an IFN-inducing phenolic glycolipid, and infection with these strains causes increased IFN production11. Similarly, human genetic variations in eicosanoid biosynthesis are associated with TB12, and variations in IL-1 and IFN production have been found to influence a variety of other infectious diseases. Even non-genetic factors such as viral co-infection could promote TB pathogenesis through IFN induction13.

Are we destined to remain at the mercy of this game of chance? An exciting aspect of the proposed IL-1–PGE2–IFN network is the implied strategy for intervention. In a mouse model of severe TB that is driven by high IFN levels, Mayer-Barber et al. found that PGE2 augmentation suppressed IFN, ameliorated disease and promoted bacterial killing. This is an outstanding example of fundamental biological insight leading to a new therapeutic opportunity. Of course, the potential for this treatment strategy depends on the demonstration that IFN is a key driver of human TB, which is currently unclear. In addition, it is difficult to imagine that a host-directed therapy such as this would be administered without simultaneous antimicrobial chemotherapy, and the benefit of PGE2 in this context would need to be assessed. Nevertheless, when we consider that more than a million people die of TB each year and the continual emergence of antibiotic-resistant M. tuberculosis strains that are insensitive to standard therapy, any chance to fix the game in favour of the host should be considered.


  1. 1.

    et al. Nature 511, 99–103 (2014).

  2. 2.

    et al. J. Immunol. 184, 3326–3330 (2010).

  3. 3.

    et al. J. Immunol. 179, 522–531 (2007).

  4. 4.

    , , , & Nature Immunol. 11, 751–758 (2010).

  5. 5.

    et al. J. Clin. Invest. 120, 1674–1682 (2010).

  6. 6.

    et al. Nature Immunol. 10, 899–906 (2009).

  7. 7.

    et al. Nature 466, 973–977 (2010).

  8. 8.

    et al. Nature Immunol. 14, 52–60 (2013).

  9. 9.

    , , , & PLoS Biol. 10, e1001249 (2012).

  10. 10.

    et al. Nature Med. 20, 75–79 (2014).

  11. 11.

    et al. Proc. Natl Acad. Sci. USA 98, 5752–5757 (2001).

  12. 12.

    et al. Cell 140, 717–730 (2010).

  13. 13.

    et al. J. Infect. Dis. 209, 270–274 (2014).

Download references

Author information


  1. Samuel M. Behar and Christopher M. Sassetti are in the Department of Microbiology and Physiological Systems, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA.

    • Samuel M. Behar
    •  & Christopher M. Sassetti
  2. C.M.S. is also at the Howard Hughes Medical Institute, Chevy Chase, Maryland.

    • Christopher M. Sassetti


  1. Search for Samuel M. Behar in:

  2. Search for Christopher M. Sassetti in:

Corresponding authors

Correspondence to Samuel M. Behar or Christopher M. Sassetti.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing