The finding that innate lymphoid cells can control the activity of CD4+ T cells reveals another potential form of immune-system regulation, and may help to explain how the body distinguishes resident from pathogenic bacteria. See Letter p.113
The lining of our intestines is a border zone at which our own cells peacefully coexist with resident bacteria. Cells of the immune system patrol this area to prevent infiltration by invasive pathogenic bacteria. However, in some individuals, the immune system mistakenly targets the benign commensal bacteria, triggering an inflammatory process that damages the intestinal mucosa and leads to inflammatory bowel disease1,2. In a report on page 113 of this issue, Hepworth et al.3 identify a mechanism that could be crucial for preventing an overly exuberant immune response to commensal bacteria. Intriguingly, this mechanism relies on the ability of a rare type of immune cell — innate lymphoid cells — to control T cells of the adaptive immune system.
Innate lymphoid cells (ILCs) are classified into three groups, depending on their expression of and developmental dependence on certain transcription factors and secreted molecules. Hepworth and colleagues studied group 3 ILCs, which reinforce the intestinal barrier against pathogenic bacteria by producing the soluble molecules IL-22 and IL-17A. These cytokines augment the capacity of epithelial cells to produce antimicrobial peptides and also recruit other immune cells, such as granulocytes4,5.
The development of group 3 ILCs is driven by the transcription factor RORγt (ref. 6), and so these cells are absent from RORγt-deficient mice. Hepworth et al. noted that RORγt-deficient mice had symptoms that were characteristic of immune activation: their spleens were enlarged and contained activated T cells expressing the CD4 receptor (CD4+ T cells). Moreover, the serum of the mice contained antibodies that bind to commensal bacteria, suggesting specific immune activity against these bacteria. Consistent with this, the authors could ameliorate the CD4+ T-cell activation by using antibiotic treatment to eliminate the commensal bacteria. The researchers saw a similar effect when they depleted the group 3 ILCs in normal mice, confirming that these cells are essential for controlling CD4+ T-cell activation.
But how does this regulation occur? Although group 3 ILCs produce IL-22 and IL-17A, mice that lacked these cytokines did not have symptoms of immune activation. To gain more insight, Hepworth and colleagues turned to analysis of the transcriptome — a cell's complement of RNA molecules. This revealed that the genes that encode MHC class II proteins are highly expressed by a subset of group 3 ILCs called lymphoid tissue inducer (LTi)-like cells. LTi-like cells appear after birth in the intestinal mucosa, in which they promote the generation of post-natal lymphoid tissue by recruiting B cells to form isolated lymphoid follicles7. MHC class II molecules capture protein fragments called peptides and present these antigens to CD4+ T cells, which then scrutinize the peptide–MHC complexes for the presence of 'self' or foreign (typically microbial) peptides. Thus, it seems possible that group 3 ILCs influence CD4+ T-cell activation through this antigen-presentation process.
Capture, processing and presentation of antigens is mainly a function of dendritic cells — 'sentinel' immune cells that instruct T cells to tolerate MHC molecules in complex with self peptides, but to react against foreign-peptide–MHC complexes8. Whether dendritic cells induce T-cell activation or tolerance depends on whether the dendritic cells are concurrently activated by microbial stimuli that induce the accessory expression of co-stimulatory molecules. Hepworth et al. demonstrate that group 3 ILCs can capture model proteins (chicken ovalbumin or Eα protein), degrade them, and present the peptides on MHC class II, just like dendritic cells. The group 3 ILCs could also present a peptide derived from commensal bacteria. However, the cells did not induce proliferation of CD4+ T cells that expressed receptors specific for these antigens, probably because presentation by MHC class II on ILCs is not coupled with co-stimulatory molecules and hence induces T-cell tolerance rather than activation.
To conclusively demonstrate the relevance of MHC class II expression by group 3 ILCs to the control of CD4+ T cells, Hepworth and colleagues generated mice that lacked MHC class II proteins in group 3 ILCs only. They observed signs of CD4+ T-cell activity against commensal bacteria, just as in RORγt-deficient mice. In addition, over time these mice spontaneously developed rectal prolapse, which is characteristic of inflammatory bowel disease.
Overall, Hepworth and colleagues' study reveals that group 3 ILCs can capture and present antigens much like dendritic cells, but that they regulate CD4+T-cell function in a manner somewhat resembling that of another class of T cell, regulatory T cells. These findings are provocative and raise important questions. Where and how do group 3 ILCs capture microbial peptides or proteins? And where do the ILCs and CD4+T cells interact? Both processes could take place in the mesenteric lymph node, in which bacterial products are collected from lymph and CD4+ T cells are initially activated. Alternatively, group 3 ILCs might act in the intestinal mucosa, where they are in close contact with commensal bacteria and where CD4+T cells perform most of their effector functions.
Details on the regulatory function of ILCs also remain uncertain. Do these cells induce CD4+T-cell death or functional paralysis, or do they actively suppress CD4+T cells, like regulatory T cells do? Hepworth et al. propose that group 3 ILCs interact directly with CD4+T cells (Fig. 1a). However, it is possible that the ILCs compete with dendritic cells for interaction with CD4+ T cells, and thereby prevent their activation (Fig. 1b). Group 3 ILCs might also enable regulatory interactions of CD4+T cells with other cells by ensuring an appropriate architecture of the lymphoid tissue in the intestinal mucosa.
Finally, how do the group 3 ILCs inhibit CD4+T-cell responses to commensal bacteria but not to pathogenic bacteria? ILCs have been shown to promote the memory function of CD4+T cells through the co-stimulatory molecules OX40L and CD30L9. Perhaps pathogenic infection and the consequent release of inflammatory cytokines (such as IL-23) enhance expression of these co-stimulatory molecules on ILCs, thereby changing their interactions with CD4+T cells (Fig. 1c). In this regard, it would be interesting to know the status of ILCs during inflammation. Alternatively, recruitment and activation of dendritic cells might overwhelm the regulatory ILCs during infection (Fig. 1d). Further delineation of the regulatory and immunogenic functions of innate lymphoid cells, including group 3 ILCs, will not only help us to understand immune regulatory processes but could also provide the basis for new, refined therapeutic intervention in conditions such as inflammatory bowel disease.
About this article
Journal of Molecular Medicine (2016)