Immunology: How the Microbiota Digests Bile to Protect against Viral Infection

Abstract

In a recent issue of Cell, Winkler et al. provide a comprehensive analysis of how secondary bile acids produced by unique members of the microbiota regulate plasmacytoid dendritic cells and monocytes via TLR7 and MyD88 signaling to protect from alphavirus dissemination.

Chikungunya virus (CHIKV) derives its name from an African language meaning ‘to become contorted,’ referring to the severe joint pain caused by viral infection. CHIKV is an alphavirus transmitted by mosquitoes that causes acute febrile illness and joint inflammation, which often progresses to chronic incapacitating arthritis. CHIKV infects millions every year and continues to spread across the world, causing urban epidemics. With no vaccine or targeted treatments available, mosquito avoidance remains the mainstay of infection prevention [1].

In the fight against CHIKV, type I interferon (IFN) has been identified as a critical regulator of disease severity. In mice, type I IFN is essential for protection against CHIKV. Expression of type I IFN by plasmacytoid dendritic cells (pDCs) alone can control CHIKV infection in mice [2,3]. pDCs are a specialized subset of dendritic cells that are known to produce high levels of IFN in response to bacterial and viral challenge. Whereas pDCs help defend against infection, monocytes are thought to contribute to CHIKV-induced arthritis as they infiltrate joints and become infected [4,5]. The microbiota has been shown to be important for priming type I IFN responses including induction of antiviral interferon-stimulated genes. For example, mice that are germ-free or treated with antibiotics are more susceptible to several different viral infections, produce less IFN, and have decreased expression of interferon-stimulated genes [6,7]. A mechanistic understanding of how members of the microbiota — and which specific bacteria — act on the host to increase type I IFN remain unknown. Now, in a new study, Winkler and colleagues [8] take a significant step forward by identifying specific members of the microbiota that govern the immune system’s production of type I IFN and aid in viral clearance.

Given that the microbiota has been shown to influence type I IFN and protect from other types of viral infection [6], Winkler and colleagues [8] sought to determine the impact of the microbiota on alphavirus infection. Conventionally raised, germ-free, or antibiotic-treated animals were infected with CHIKV. When they examined acute infection at days one and three post-infection, both germ-free and antibiotic-treated mice had significantly higher viral titers in the blood and spleen. Seeking to understand what cell types accounted for this increased viral load, whole blood was analyzed. Classical inflammatory monocytes had significantly higher viral titers in antibiotic-treated and germ-free mice than conventional mice. Depleting only monocytes abolished the differences in viral load between antibiotic-treated and control mice, demonstrating that microbiota-induced differences in monocyte infection burden accounted for the difference in infection rates.

To investigate why monocytes from antibiotic-treated mice were more susceptible to infection, single-cell RNA seq analysis and targeted qPCR of monocytes revealed extensive downregulation of interferon-stimulated genes. Additionally, at day one post-infection — but not at baseline — antibiotic-treated and germ-free mice had lower type I IFN. Taken together, this suggests a pivotal role for type I IFN in protecting monocytes from infection. Indeed, infection of type I IFN-receptor-knockout mice (Ifnar1^−/−), treated with or without antibiotics, showed a narrowing of the differences in viremia between antibiotic-treated and control. However, only when Stat1, the signaling molecule downstream of both type I and type II IFN, was specifically knocked out in myeloid cells were the differences in viremia between antibiotic-treated and control mice abolished. This finding indicates that although type I IFN plays a crucial role in the antiviral response, there remains an undetermined but not insignificant antiviral role for type II IFN as well.

Looking to define the cellular source of type I IFN that modulates monocyte infectability, the authors then depleted pDC cells [8], known producers of type I IFN [2]. pDC deficiency reduced both type I IFN levels and the microbiota-mediated antiviral effect. Furthermore, pDC production of type I interferon required the expression of TLR7 and MyD88, which are important for the recognition of pathogens and members of the microbiota, to mediate the antiviral effects of the microbiota.

To identify protective members of the microbiota, the authors compared the ability of different complex bacterial communities to restore host protection from viremia after antibiotic treatment. By doing this they identified 28 Clostridiales of interest. Clostridia are a highly diverse class of bacteria with a broad range of specialized functions that extend far beyond the spore formation and toxin production of the infamous Clostridium difficile [9]. One Clostridium species they identified in the protective community bore great similarity to C. scindens, a commensal species that expresses the enzyme 7-α dehydroxylase, which is only found in 0.0001% of total colonic microbes [10]. After the liver secretes bile acids into the small intestine, most are reabsorbed and return to the liver. Still, some primary bile acids continue to the colon, where they are deconjugated by another microbial enzyme, bile salt hydrolase, which is broadly expressed across the microbiota. Once the primary bile acids are deconjugated, 7-α dehydroxylase from C. scindens converts them into the secondary bile acid deoxycholic acid (Figure 1) [11]. Winkler et al. [8] found that supplementation with deoxycholic acid or colonization of antibiotic-treated mice with C. scindens, but not other bacteria that lack 7-α dehydroxylase activity, restores antiviral protection from CHIKV at day one post-infection in a MyD88-dependent manner. Furthermore, one week of pretreatment with deoxycholic acid alone was capable of lowering acute viremia even in mice that did not receive antibiotics.

Figure 1. How bacterial-derived bile acids protect from viral infection.

Bile acids are produced from cholesterol in the liver and stored in the gallbladder before being released into the duodenum of the small intestine. Bile acids assist in the digestion and absorption of fats, but also have many functions as signaling molecules. Most bile acids are absorbed in the small intestine and return to the liver, but some continue to the colon where they are metabolized by the gut microbiota. Primary bile acids can first be deconjugated by the broadly expressed bacterial enzyme bile salt hydrolase (BSH) and then in the presence of 7-α-dehydroxylase expressed by C. scindens, primary bile acids are converted into the secondary bile acid, deoxycholic acid (DCA). In a TLR7- and MyD88-dependent manner, deoxycholic acid influences pDCs to produce type I IFN that acts on Ly6C^hi monocytes to prevent their infection by alphaviruses. Illustration created with BioRender.com.

This study adds new depth to the understanding of how the microbiota influences the host’s antiviral response. The authors propose that the bacterial derived secondary bile acid, deoxycholic acid, signals to pDCs in a TLR7- and MyD88-dependent manner to stimulate type I IFN production, which can in turn act on monocytes to protect them from alphavirus infection (Figure 1). This proposed mechanism reveals many more testable questions that launch the field of microbiota and host antiviral response forward. In the future, it will be essential to explain how secondary bile acids are signaling to immune cells: many are known to express the G-protein coupled receptors that bind bile acids [12]. Determining which receptors are involved and on what cell types their expression is essential for mediating an antiviral response will be valuable. This study also highlights the rapid speed at which microbiota changes can impact susceptibility to viral infections. Just three days of, albeit powerful, antibiotics is enough to disrupt host antiviral immunity. Discovering how initial viremia and treatments with deoxycholic acid or C. scindens impact long-term recovery from CHIKV will provide critical clues for potential clinical treatments in the future. Additionally, considering how frequently antibiotics are used clinically, investigating the potential therapeutic benefits of C. scindens and deoxycholic acid in protecting from other types of viruses is warranted. This new role for deoxycholic acid in modulating type I IFN levels adds to a growing body of evidence about ways in which the microbiota can influence health and disease.