Poised or Confused? DCs, Type I interferons, and the microbiota
In a previous article, we discussed how metabolites produced by the gut microbiota can influence proliferation of regulatory T cells through the action of dendritic cells (DCs). More recently, the Probst group at the University Medical Center Mainz in Germany published a study in the journal Cell in which they explore the role of the microbiota in the development not of T cells, but of the DCs themselves. Past studies have shown that compared to specific pathogen free (SPF) mice, germ-free mice—which lack resident microbes throughout their bodies—have immune system defects, including impairment in the ability of innate immune cells such as DCs to respond to infectious agents such as viruses and effectively produce cytokines, signals that activate or regulate the immune system. However, the signaling pathways by which resident microbiota influence the “poised” state—or the potential of these DCs to respond to future immune challenges—are not fully understood.
The authors of this paper first set out to explore to what extent, in their hands, lack of microbial signaling within the mouse could influence DC function—in particular, a subset called conventional DCs (cDCs). Using a mouse model in which administration of an antibody against the costimulatory molecule CD40 (anti-CD40) induces DC activation and subsequent inflammation, they first confirmed that splenic cDCs from germ-free mice did, as expected, have a reduced capacity to secrete the pro-inflammatory cytokine TNFα compared to the “microbially-intact” SPF mice. They also used a variety of mice lacking adaptor molecules involved in signaling pathways downstream of pathogen sensing mechanisms, as well as germ-free mice colonized with either one bacterial strain or a consortium of strains, to show that TNFα production by cDCs was impacted by the presence of microbial signals. In addition, some of these experiments were done with a challenge from a viral or bacterial component and germ-free DCs were not able to properly mount an immune response to either.
Besides the ability of DCs to produce cytokines, the authors also looked at their ability to induce T cell responses (specifically CD8+ T cells). Previously, their group had developed a mouse model (the DIETER mouse) in which DCs overexpress a specific antigen that can then induce peripheral tolerance in the form of CD4+ T cells. They took advantage of their previous finding that depletion of CD4+ T cells in this model leads to a CD8+ T cell response that is specific to the otherwise “harmless” antigen expressed by the body’s own cells. They found that transgenic mice that were treated with antibiotics following depletion of CD4+ T cells actually had a reduction in the number of antigen-specific CD8+ T cells, further suggesting that microbial signals play a role in the ability of DCs to influence other cell types.
What are the mechanisms by which microbes influence DC function? To find the answer, the authors used now-standard techniques to query the genetic material (RNA) actively being made in a cell. This genetic material will be used to produce proteins that carry out the cell’s functions. Through these techniques (whole RNA sequencing and qRT-PCR), the authors found that genes downstream of Type I interferon signaling were significantly downregulated in germ-free mice. Type I interferons (IFNα and IFNβ) are cytokines important in initiating anti-viral immune responses and the typical, albeit low, basal and constitutive (“tonic”) production of IFNβ throughout the body is absent in the germ-free mice. Unsurprisingly, they found that another subset of DCs called plasmacytoid DCs (pDCs)—which are also important in fighting viruses—were the major producers of Type I interferons in the spleen of SPF mice.
Through analysis of RNA sequencing data, the authors show that major gene signatures of the germ-free status resemble those of mice in which there is a lack of Type I interferon signaling. Specifically, these mice have a genetic deletion of the receptor involved in sensing Type I interferons (Ifnar1r-/-). These two groups’ core transcriptional programs are compared to the control’s—SPF mice. Like what was shown in germ-free mice, Ifnar1r-/- mice also had reduced capacity of cDCs to produce TNFα and induce antigen-specific CD8+ T cells in the DIETER mouse model. Administration of anti-CD40 antibody was not sufficient to push the system in the Ifnar1r-/- mice and revert back to normal levels of cDC function in both experiments. The authors suggest that exposure to Type I interferon signaling is important for DCs early on in the steady state, prior to encountering an inflammatory trigger, that there is a certain “mark” or transcriptional signature imprinted on these cDCs. This was evident in the altered epigenetic signatures and metabolic state of both germ-free and Ifnar1r-/- cDCs compared to SPF cDCs.
In summary, the authors found that tonic Type I interferon signaling, which is normally present in SPF mice and primarily produced by pDCs in the spleen, “educates” splenic cDCs in the steady state and “poises” them to be ready to respond to microbial challenges in the future through cytokine production and activation of CD8+ T cells. In this paper, the authors did not explore directly whether only microbiota-induced Type I interferon signaling was important in cDC function and why or how microbiota-derived signals instruct pDCs to make Type I interferons specifically. However, this study did bring into light the concept that DCs receive epigenetically-imprinted “instructions” from the microbiota that prepare them for potential challenges, while at the same time making them more susceptible to reacting to otherwise “harmless” antigens in the absence of CD4+ T cells and peripheral tolerance. This new piece of evidence could provide some insight on conditions such as Type I diabetes or systemic lupus erythematosus, which are characterized by increased Type I interferon-induced gene signatures early on in disease development.
By Geil Merana
Microbiota educate dendritic cells via Type 1 interferons in Schaupp et al (2020).