Epithelial IKKα licenses ILC3s to defend the intestinal barrier

In this issue, Giacomin et al. report on a novel function of IKKα acting in intestinal epithelial cells (IECs) to control intestinal immunity against enteropathogenic bacteria. Human enteropathogenic and enterohemorrhagic E. coli constitute a serious health concern, as they can cause life-threatening intestinal infections. Infection of mice with Citrobacter rodentium, a natural enteric mouse pathogen, serves as a relevant model to study the mechanisms of host immunity against this family of enteropathogenic bacteria.

Giacomin et al. found that mice with IEC-specific ablation of IKKα (IKKα^ΔIEC), but not IKKβ, were highly susceptible to C. rodentium infection. A series of elegant follow-up experiments revealed that epithelial IKKα deficiency caused impaired production of IL-22, a cytokine that promotes epithelial regeneration and antimicrobial defense, by group 3 innate lymphoid cells (ILC3s) in infected mice. Administration of recombinant IL-22 or transfer of IL-22 producing ILCs restored effective immunity to C. rodentium in IKKα^ΔIEC mice. The IKKα-deficient epithelium produced increased amounts of TSLP, and treatment with neutralizing antibodies against TLSP largely restored IL-22 production and effective immunity to C. rodentium. Interestingly, recombinant TSLP inhibited IL-22 production by RORγt⁺ ILCs when added in splenocyte cultures but not in purified ILC cultures, suggesting that the suppressive effect of TSLP on IL-22 production in ILCs is indirect.

The study by Giacomin et al. adds novel insight to our understanding of the epithelial–immune cell dialog that controls intestinal immunity, but it also raises new questions. Does epithelial IKKα act via noncanonical NF-κB signaling, as suggested by previous studies implicating epithelial lymphotoxin β receptor and stromal p52 in controlling C. rodentium infection, or do other, NF-κB–independent, IKKα functions regulate intestinal immunity? Additional studies in mice with epithelial cell specific knockout of specific NF-κB subunits will be required to address this question. The target cell type that responds to TSLP and suppresses IL-22 production by ILCs also remains to be identified. Dendritic cells (DCs) are a likely candidate, as TSLP affects DC polarization and cytokine production, which could in turn affect IL-22 expression by ILCs. Moreover, TSLP neutralization only partly restored immunity to C. rodentium in IKKα^ΔIEC mice, suggesting that additional, as yet unknown, mechanisms also contribute. From a clinical perspective, this study raises the interesting possibility that TSLP neutralization might have beneficial effects in enteropathogenic bacterial infections, although the diverse biological functions of TSLP may make it difficult to exploit this newly identified mechanism therapeutically.