Evolutionary genetic and experimental analyses suggest that mutations causing familial Mediterranean fever have been positively selected in the Middle East, probably because they confer heightened resistance against Yersinia pestis infection.
Familial Mediterranean fever (FMF) is a common, hereditary autoinflammatory disease caused by recessive mutations in the gene encoding the inflammatory response protein pyrin. FMF mainly affects populations from the Eastern Mediterranean basin, where ~8% of individuals carry one of the disease-causing pyrin variants1. Upon bacterial infection, pyrin forms the pyrin inflammasome complex, which elicits the release of proinflammatory cytokines and induces pyroptosis2. To counteract the host response, pathogenic Yersinia species secrete a virulence effector, YopM, that specifically inhibits the pyrin inflammasome3. In this issue of Nature Immunology, Park et al.4 show that FMF-causing mutations are associated with increased release of proinflammatory cytokines from human leukocytes and increased survival of mice following Y. pestis infection. They show that FMF variants arose >1,800 years ago, prior to the first deadly plague pandemic, and have conferred a selective advantage to people from the Middle East, probably because they increase resistance to the plague.
Plague pandemics were among the most devastating epidemics in the history of the northern hemisphere, causing the death of 30–60% of its total population. Because Y. pestis outbreaks have exerted strong selection pressure on humans, they have regularly been invoked to explain the heterogeneous geographic distribution of genetic variants associated with immune-related disease. In 1962, Pettenkofer et al.5 argued that the frequencies of the ABO histo-blood group antigens have been shaped by natural selection because they confer resistance to Y. pestis infection. Likewise, the high prevalence and estimated age of the CCR5-Δ32 variant, which today protects against human immunodeficiency virus infection, prompted researchers to test its effects on plague susceptibility6. However, the evolutionary genetic and experimental analyses were debatable, and subsequent studies showed that the evidence from these experiments was inconclusive7,8.
The role of pyrin in plague susceptibility was first proposed when a specific Yersinia virulence factor, YopM, was shown to inhibit the pyrin inflammasome and to be involved in the evasion of host antimicrobial responses3. YopM is secreted by pathogenic Yersinia species to activate the host kinases that negatively regulate pyrin through phosphorylation. The 14-3-3 regulatory proteins bind phosphorylated pyrin and block inflammasome activation, resulting in the inhibition of proinflammatory cytokine release and pyroptosis (Fig. 1). Because pyrin variants causing FMF lead to excessive inflammasome activation2 and are relatively common in regions where the plague was endemic1, researchers have hypothesized that these mutations have been selected to counteract the virulence strategy of Y. pestis. Nevertheless, no studies to date had formally investigated whether FMF-causing mutations affect the host response to Yersinia infection or whether their high frequency in present-day Mediterranean populations is due to natural selection.
Using state-of-the-art approaches in evolutionary genetics, Park et al.4 inspected the genomic diversity of more than 2,000 individuals from Turkey to test whether common amino-acid altering mutations in MEFV, the gene encoding pyrin, have evolved under positive selection. They found that genetic variation surrounding FMF mutations is more conserved than that surrounding random mutations that occur at similar frequency in the Turkish population. This suggests that MEFV variants have rapidly increased in frequency in the last few thousand years, consistent with the action of recent positive selection9. By modelling the genetic diversity of the MEFV locus, they also estimated that the two main FMF mutations occurred ~1,800–2,600 years ago and conferred a selective advantage that is comparable to that of the lactose tolerance allele in Europeans. Based on these results, the authors conclude that, despite their deleterious effects, FMF mutations have been under positive selection in Eastern Mediterranean populations, possibly because they confer resistance to the plague. In support of this, the MEFV variants emerged prior to the first recorded plague pandemic, the Plague of Justinian, which began ~1,500 years ago and afflicted the entire Mediterranean Basin.
To further strengthen their hypothesis, Park et al. evaluated the protective role of FMF mutations against Y. pestis infection using comprehensive ex vivo and in vivo experiments. They first confirmed that the Yersinia virulence factor YopM inhibits IL-1β release by the pyrin inflammasome-dependent pathway in primed macrophages. Several host protein kinases (from the RSK family) phosphorylate pyrin only in the presence of YopM, indicating that Yersinia bacteria can hijack host proteins to inhibit pyrin inflammasome activation. Using pulldown assays and mutant YopM, the authors showed that YopM-induced pyrin phosphorylation requires direct interactions between YopM, pyrin and RSK proteins. Then they explored the effect of FMF-causing pyrin mutations on Yersinia–host interactions. In monocytes of patients with FMF infected by Y. pestis, both the binding of YopM to pyrin and pyrin phosphorylation were reduced relative to healthy controls. Furthermore, Y. pestis–induced IL-1β release was higher in human cells expressing pyrin with FMF mutations than in cells expressing wild-type pyrin (Fig. 1). Intriguingly, FMF mutations alter the C-terminal domain of pyrin, while YopM interacts directly with the N-terminal domain; however, the C-terminal and N-terminal pyrin domains interact, and FMF mutations reduce these interactions. Finally, knockin mice expressing human pyrin with FMF mutations exhibit increased survival to Y. pestis as compared to mice expressing the wild-type human pyrin, and this is mediated by the IL-1β-signaling pathway. In light of the beneficial effects of FMF mutations on the human immune response to Y. pestis and their recent history of positive selection, Park et al. conclude that plague epidemics have played a key role in selecting for mutations that cause FMF in Eastern Mediterranean peoples today.
Examples of beneficial mutations that confer a well-understood selective advantage are extremely rare in humans because it is notoriously difficult to infer their evolution and effects on survival in time and space. In this study, Park et al. elegantly combined evolutionary genetic and experimental analyses to support the view that FMF-causing mutations have been positively selected to confer resistance to the plague. The hypothesis is very reasonable in light of the increasing evidence supporting a role for past epidemics in selecting for mutations associated with common present-day autoimmune and autoinflammatory diseases10. Nevertheless, direct evidence is still required: the pyrin inflammasome may be repressed by other pathogens, and FMF mutations may have thus evolved under positive selection in response to other microbial pressures. Fortunately, while the task was impossible in 19627, recent advances in ancient DNA sequencing now offer the unique opportunity to demonstrate the direct role of past infections in shaping present-day disease risk. Indeed, it will soon be possible to compare the genomes of victims and survivors of past pandemics and to detect genetic variation that conferred resistance to lethal infections without bias. Likewise, ancient DNA research now allows the reconstruction of how allele frequencies have evolved over time, which greatly facilitates the inference of natural selection11. Finally, experiments involving infection with pathogens engineered to resemble those isolated from victims of pandemics12 will offer promising new avenues for research on the evolution of microbial virulence. Overall, there is no doubt that the study by Park et al. will stimulate new research to increase our understanding of the causes and consequences of historical epidemics, such as the plague, whose re-emergence poses a significant health risk today13.
Koshy, R., Sivadas, A. & Scaria, V. Clin. Genet. 93, 92–102 (2018).
Schnappauf, O., Chae, J. J., Kastner, D. L. & Aksentijevich, I. Front. Immunol. 10, 1745 (2019).
Chung, L. K. et al. Cell Host Microbe 20, 296–306 (2016).
Park, Y. H. et al. Nat. Immunol. https://doi.org/10.1038/s41590-020-0705-6 (2020).
Pettenkofer, H. J., Stoess, B., Helmbold, W. & Vogel, F. Nature 193, 445–446 (1962).
Elvin, S. J. et al. Nature 430, 418 (2004).
Springer, G. F. & Wiener, A. S. Nature 193, 444–445 (1962).
Mecsas, J. et al. Nature 427, 606 (2004).
Voight, B. F., Kudaravalli, S., Wen, X. & Pritchard, J. K. PLoS Biol. 4, e72 (2006).
Barreiro, L. B. & Quintana-Murci, L. Nat. Rev. Genet. 11, 17–30 (2010).
Dehasque, M. et al. Evol. Lett. 4, 94–108 (2020).
Spyrou, M. A., Bos, K. I., Herbig, A. & Krause, J. Nat. Rev. Genet. 20, 323–340 (2019).
Randremanana, R. et al. Lancet Infect. Dis. 19, 537–545 (2019).
My laboratory is supported by the Institut Pasteur, the CNRS, the French Government’s Investissement d’Avenir program, Laboratoires d’Excellence “Integrative Biology of Emerging Infectious Diseases” (ANR-10-LABX-62-IBEID) and “Milieu Intérieur” (ANR-10-LABX-69-01) projects, the Fondation pour la Recherche Médicale (Equipe FRM DEQ20180339214) and the Agence Nationale de la Recherche grant ‘MORTUI’ (ANR-19-CE35-0005-01).
The author declares no competing interests.
About this article
Cite this article
Patin, E. Plague as a cause for familial Mediterranean fever. Nat Immunol 21, 833–834 (2020). https://doi.org/10.1038/s41590-020-0724-3
The plague that never left: restoring the Second Pandemic to Ottoman and Turkish history in the time of COVID-19
New Perspectives on Turkey (2020)
The American Journal of Medicine (2020)