Intestinal infection with rotavirus is a major cause of diarrhoea in infants, and can be fatal. The identification of immune sensor proteins that detect and restrict this viral infection now illuminates the body's defence system. See Letter p.667
Intestinal cells can be infected by a type of virus known as a rotavirus, which is one of the leading causes of severe diarrhoea in infants and young children worldwide. Although effective rotavirus vaccines have been available since 2006, routine vaccination has been adopted in only a few developing countries, and it is estimated that the virus causes more than 200,000 childhood deaths annually1. There is only limited understanding of how intestinal cells sense rotavirus infection and mount an antiviral response. Now, on page 667, Zhu et al.2 identify host proteins that are key components of this response.
Rotavirus is mainly spread by direct oral ingestion through contact with contaminated objects, or from water or food. Once ingested, the virus tends to infect epithelial cells that line the intestine, and this is where the virus, which contains double-stranded RNA, replicates. Zhu and colleagues investigated the host-defence response to the infection.
“A key component of the response is the formation of a multiprotein complex called an inflammasome.”
A key component of the response is the formation of a multiprotein complex called an inflammasome3. This complex contains core proteins, which are present in most inflammasomes, and sensor proteins that respond to specific types of pathogen and are present in only a subset of inflammasome complexes. Inflammasome activation usually promotes protective host defence and repair mechanisms. However, if activated in diseases such as chronic inflammatory disorders, these complexes can contribute to tissue damage and disease development4.
Inflammasome complexes mainly serve to recruit and engage the enzyme caspase-1. When activated in response to specific pathogen- or host-derived cues, caspase-1 acts as 'molecular scissors', cleaving proteins containing certain sequences of amino acids that include aspartate. This cleavage activates key immune regulators, including signalling proteins known as cytokines, which convert biologically inert precursor proteins such as pro-interleukin-1β (pro-IL-1β) and pro-IL-18 into the pro-inflammatory proteins IL-1β and IL-18, respectively3. Caspase-1 can also cleave the gasdermin D protein, releasing its amino-terminal fragment. This fragment can generate pores in cellular membranes and cause a type of cell death called pyroptosis, which occurs through cellular rupture5,6,7,8,9.
Zhu and colleagues investigated whether inflammasomes might be involved in host defence against the virus. They observed that, compared with wild-type mice, animals lacking a functional copy of the core inflammasome proteins Asc or caspase-1 were more susceptible to rotavirus infection. By contrast, animals lacking known inflammasome sensor proteins were not more susceptible. This prompted the authors to search for other sensor proteins that might detect rotaviral infection and engage a protective inflammasome response.
The authors focused on the evolutionarily conserved family of nucleotide binding domain and leucine-rich (NLR) proteins, which have diverse roles in immunity10. They investigated the protein Nlrp9b, a previously uncharacterized member of this family, because it is expressed mainly in intestinal epithelial cells. The authors used genetic engineering to produce mice that either lacked Nlrp9b throughout their bodies or lacked it only in intestinal epithelial cells. They found that both groups of mice were more susceptible than wild-type mice to rotaviral infection, indicating that Nlrp9b has a crucial role in protection against the infection (Fig. 1).
The intestine is home to a complex ecosystem of bacterial species that affect various metabolic and immune responses. When Zhu and colleagues compared the resident intestinal bacteria in wild-type and Nlrp9-deficient mice, they found no significant differences, indicating that this protein acts through a mechanism that is independent of the gut bacterial composition.
Using state-of-the-art stem-cell technology, the authors generated in vitro 'mini-gut' structures to investigate inflammasome activation. Using mini-guts that expressed or lacked Nlrp9b, and assessing IL-18 levels, the authors revealed a role for Nlrp9b in activation. However, they found that inflammasome activation was only partially attenuated in Nlrp9b-deficient mini-guts compared with Asc-deficient mini-guts. This implies that another inflammasome component might account for the residual inflammasome activity still present in the Nlrp9b-deficient mini-guts. The authors observed that mice deficient in the NLR protein Nlrp6 are more susceptible than wild-type mice to rotaviral infection. Perhaps Nlrp6 or an as yet unknown inflammasome sensor in intestinal epithelial cells might be responsible for this residual activity.
The authors found that mice lacking gasdermin D or Nlrp9b had a similar sensitivity to infection, and showed markedly increased levels of rotavirus and diarrhoea compared with infected wild-type mice. However, when Zhu and colleagues conducted similar viral-infection tests in mice deficient in IL-18, the level of diarrhoea in these animals resembled that of infected wild-type mice. The authors therefore concluded that gasdermin D-dependent pyroptosis is more crucial than an IL-18-mediated immune response in restricting rotavirus infection in vivo.
Pyroptosis can promote the release of molecules generated through inflammasome action, including IL-18 and IL-1β, along with other types of molecules known as alarmins, which can help to activate an immune response in target cells5,6. Perhaps deficiencies in the release of these molecules account for the observed inability of gasdermin-D-deficient mice to efficiently curb rotaviral infection. Pyroptosis of infected intestinal cells eliminates the place where the virus replicates, thereby decreasing the level of virus, an effect that may also be part of the host response to rotaviral infection. Determining the relative importance of these mechanisms will require further investigation.
To address potential ways in which host cells might sense rotaviral infection and assemble the Nlrp9b inflammasome, Zhu and colleagues used cultured human cells, which they engineered to express tagged versions of mouse Asc and Nlrp9b and human NLRP9. A weak interaction between NLRP9 and Asc was observed only after rotavirus infection, whereas Asc and Nlrp9b failed to interact. Zhu et al. found that if cultured cells were engineered to express human NLRP9, this protein bound rotaviral RNA and short synthetic RNA molecules more readily than did the other NLR proteins tested. However, mouse Nlrp9b did not bind the synthetic RNA.
These differences between murine Nlrp9b and human NLRP9 were not investigated further. Perhaps human NLRP9, but not mouse Nlrp9b, can interact with an RNA-binding protein present in the cultured human cells that is needed for inflammasome assembly. Consistent with this idea, Zhu et al. showed that human NLRP9 recruits the RNA helicase enzyme DHX9 only on rotaviral infection, and that a decrease in DHX9 expression results in impaired binding of NLRP9 to both double-stranded RNA and ASC. Determining whether there is a physical association between mouse Dhx9 and Nlrp9b should be the subject of future analysis. However, consistent with a role for Dhx9 in an inflammasome response to infection, when the authors analysed rotaviral infection of Dhx9-deficient mini-guts, the results were similar to those found in mini-guts lacking Nlrp9b, and showed elevated rotavirus shedding, impaired pyroptotic cell death and defective IL-18 production compared with wild-type cells.
The finding that Nlrp9b is an inflammasome adaptor specific to intestinal epithelial cells, and potentially cooperates with Dhx9 on rotavirus infection, is a conceptual advance that is likely to prompt investigation of this pathway in many clinically relevant viral gut infections. And the finding raises many questions. Is Dhx9 the sole upstream sensor that engages Nlrp9b, or is there a diverse set of upstream sensors that activate the Nlrp9b inflammasome? Is the Nlrp9b inflammasome activated during non-viral infections, and does it contribute to the pathogenesis of inflammatory bowel diseases? DHX9 cooperates with the host RNA-editing enzyme ADAR (ref. 11), raising the question of whether (and, if so, how) Nlrp9b regulates responses against host-derived double-stranded RNA species. Addressing these and other questions would provide substantial insight into the roles of Nlrp9b in health and disease, and might result in improved ways to define and treat intestinal inflammation. Footnote 1
Tate, J. E. et al. Clin. Infect. Dis. 62 (Suppl. 2), S96–S105 (2016).
Zhu, S. et al. Nature 546, 667–670 (2017).
Lamkanfi, M. & Dixit, V. M. Cell 157, 1013–1022 (2014).
Saavedra, P. H., Demon, D., Van Gorp, H. & Lamkanfi, M. Semin. Immunopathol. 37, 313–322 (2015).
Kayagaki, N. et al. Nature 526, 666–671 (2015).
Shi, J. et al. Nature 526, 660–665 (2015).
Aglietti, R. A. et al. Proc. Natl Acad. Sci. USA 113, 7858–7863 (2016).
Liu, X. et al. Nature 535, 153–158 (2016).
Sborgi, L. et al. EMBO J. 35, 1766–1778 (2016).
Van Gorp, H., Kuchmiy, A., Van Hauwermeiren, F. & Lamkanfi, M. FEBS J. 281, 4568–4582 (2014).
Aktaş, T. et al. Nature 544, 115–119 (2017).
About this article
Cite this article
Saavedra, P., Lamkanfi, M. Gut sensor halts viral attack. Nature 546, 606–608 (2017). https://doi.org/10.1038/nature23090
Science Signaling (2017)