Intestinal commensal bacteria communicate continuously with host epithelial and innate and adaptive immune cells under both steady-state and stressful conditions (Hooper and Macpherson, 2010
; Lee and Mazmanian, 2010
; Ley et al., 2008
). These interactions have abundant beneficial effects on host metabolism and immune function. However, these interactions require host factors for the maintenance of immune homeostasis. The oral administration of DSS, which is directly toxic to enterocytes, has been a crucial tool for the identification of the cell signaling pathways involved in the regulation of host-commensal bacteria interactions (Saleh and Trinchieri, 2011
). Among these factors, the TLR and IL-1 family receptor signaling adaptor protein MyD88 plays a central role in resistance to experimental DSS-induced colitis. Myd88−/−
mice, which cannot signal through IL-1 family receptors or most TLRs, are acutely susceptible to DSS-induced colonic injury (Rakoff-Nahoum et al., 2004
). It appears that both TLR-dependent and IL-18-dependent activation of MyD88 are essential for host resistance to DSS-induced colonic damage (Dupaul-Chicoine et al., 2010
; Rakoff-Nahoum et al., 2004
; Salcedo et al., 2010
; Saleh and Trinchieri, 2011
; Zaki et al., 2010
Our results in mice simultaneously lacking multiple TLRs that are involved in bacterial recognition suggested that TLR9- and caspase-1-dependent activation of MyD88 plays a major role in host resistance to DSS-induced colonic damage. We also determined that MyD88 signaling in B cells is required to control commensal bacteria following intestinal damage. Our results highlighted the physiological importance of TLR signaling in B cells for IgM- and complement-mediated host protection following DSS-induced colonic damage.
It is well known that the lack of MyD88 in all cell types results in rapid mortality following DSS treatment in mice (Abraham and Medzhitov, 2011
; Saleh and Elson, 2011
). Based on the results of previous studies, this outcome can be attributed to the deregulated epithelial cell homeostasis that is a result of the impaired recognition of commensal bacteria by the cells. The intestinal epithelium provides not only a physical barrier against the excessive entry of luminal microbiota but also executes the MyD88-dependent microbe sensing programs that are required for the restriction of intestinal bacteria via the production of antimicrobial peptides and epithelial cell proliferation (Hooper and Macpherson, 2010
). Surprisingly, the lack of MyD88 in epithelial cells had no effect on the survival of mice following DSS treatment. Furthermore, colonic damage was comparable between epithelial cell-specific Myd88
-deficient mice (Villin-Cre × Myd88flox/flox
) and their WT counterparts (Myd88flox/flox
mice without Cre expression or Villin-Cre × Myd88flox/wt
We observed that the uncontrolled dissemination of commensal bacteria was responsible for the mortality of Myd88−/− mice during prolonged DSS-induced colonic damage. The depletion of commensal bacteria by antibiotic treatment during DSS-induced colonic damage completely rescued Myd88−/− mice from the mortality that typically occurs in DSS-treated mice. Furthermore, reconstitution of the intestinal microbiota in DSS-treated Myd88−/− mice reverted their resistance and resulted in the mortality that is associated with the dissemination of commensal bacteria throughout the peripheral tissues. These results formally demonstrated that MyD88 is indispensible in restricting intestinal bacteria from invasion and dissemination when the integrity of the epithelial cell layer is compromised. In our attempts to identify the cell-specific requirements for MyD88 signaling, we observed that whereas TLR signaling in DCs and macrophages contributed to host protection, B cell-intrinsic MyD88 played the most critical role in orchestrating host protection following DSS-induced colonic injury. Although MyD88 activation has multiple cell-specific downstream effects, our results suggested that B cell-intrinsic MyD88 signaling is required to coordinate the IgM- and complement-mediated control of intestinal bacteria following DSS-induced colonic damage.
Because the fraction of intestinal bacteria that was coated with IgM and complement C3 fragments was substantially decreased by the deletion of MyD88 in B cells, it is likely that the key protective function of MyD88 in B cells is to promote the production of IgM antibodies that subsequently bind to commensal organisms. This function is essential for preventing bacteria from gaining access to the subepithelial tissue following DSS-induced breaches of the epithelial barrier.
Our work also revealed an important role for IgA in conferring protection against DSS-induced colitis. However, IgA responses were only partially regulated by MyD88, and it is likely that TLR, IL-1R, and IL-18R signaling in cell types other than B cells is involved in the regulation of commensal bacteria-specific IgA responses. In addition, because mutations in the antigen-binding variable region of IgA are crucial for the control of commensal bacteria, further studies will be needed to investigate whether B cell intrinsic or extrinsic MyD88 signaling is involved in somatic hypermutation (Wei et al., 2011
). Our experiments cannot rule out the possibility that the lack of B cell-intrinsic MyD88 reduces protection against colitis by impairing the induction of somatic hypermutations.
While the mechanism we identified does not exclude additional functions for TLR signaling in B cells and other innate and adaptive immune cells, our work does emphasize the importance of B cell-intrinsic MyD88 in the regulation of intestinal homeostasis via the restriction of commensal bacteria from dissemination into the peripheral tissues. Our results suggest that impaired MyD88-dependent recognition of commensal bacteria causes normal microbiota to function as pathogenic microorganisms and induces mortality in the mammalian host when the barrier function of the epithelial layer is compromised.