Alerting dendritic cells to pathogens: The importance of Toll-like receptor signaling of stromal cells

Eradication of intracellular pathogens relies on differentiation of naïve CD4⁺ T cells into T helper (Th) 1 cells that produce IFN-γ, which promotes macrophage activation and production of complement-fixing and opsonizing antibodies (1-3). CD4⁺ T cells differentiate into Th1 cells by following instructions provided by dendritic cells (DCs) that have perceived pathogen incursion through specific receptors for microbial components, the Toll-like receptors (TLRs) (4). In this issue of PNAS, Sato and Iwasaki (5) demonstrate that TLR-mediated vigilance of DCs alone is not sufficient to trigger Th1 responses against intracellular pathogens. In fact, they show that epithelial and stromal cells, which constitute the mucosal barrier primarily attacked and infected by pathogens, also generate TLR-mediated signals that contribute to initiation of Th1 responses (5).

DCs are bone marrow-derived cells that seed peripheral tissues and function as sentinels of pathogenic invasion (6). Upon detecting the presence of infectious agents, DCs capture microbial antigens and deliver them to draining lymph nodes (6). There, DCs present microbial antigens to rare antigen-specific naïve CD4⁺ T cells, triggering their proliferation and expression of the functional receptor for IL-12 (IL-12R) (1, 3, 6) (Fig. 1). Moreover, DCs produce IL-12, which is the major cytokine controlling Th1 differentiation (7), as well as additional factors synergizing with IL-12, including IL-18 and two IL-12-related cytokines, IL-23 and IL-27 (7) (Fig. 1). Studies over the last few years have demonstrated that the ability of DCs to stimulate Th1 responses after exposure to infectious agents is mediated by TLRs (4). TLRs comprise a family of cell surface receptors that recognize pathogen-associated molecular patterns (PAMPs), including lipopoly-saccharide (LPS) and hypomethylated CpG-rich DNA, as well as double-stranded and single-stranded RNA (8) (Fig. 1). Upon recognition of microbial products, TLRs and their associated adapters, MyD88 and TRIF, recruit intracellular signaling mediators that activate transcription factors, such as NF-κB. As a result of transcriptional activation, DCs express chemokine receptors that drive their migration into lymph nodes, up-regulate MHC, costimulatory, and adhesion molecules that activate T cells, and secrete IL-12 and other cytokines that induce Th1 differentiation. Thus, TLRs link recognition of pathogens with multiple coordinated functional changes in DCs, which are collectively known as DC maturation and which allow DCs to reach CD4⁺ T cells and dictate their Th1 differentiation.

Fig. 1.

Two-step amplification cascade in TLR-mediated detection of microbial components. Detection of LPS, CpG-rich DNA and RNA by stromal TLRs induces the release of various mediators that directly engage TLRs in immature DCs (iDC) or facilitate the ability of TLRs to encounter microbial components. TLR-activated mature DCs (mDC) prime Th1 responses in lymph nodes.

Epithelial and stromal cells contribute to initiation of T helper 1 responses.

In their present study, Sato and Iwasaki (5) analyzed the impact of TLR signaling on Th1 responses against herpes simplex virus (HSV)-2, a virus that infects the genital mucosae of humans and mice (5). Impaired TLR signaling due to absence of the adapter MyD88 caused a defective Th1 response against HSV-2 in MyD88-deficient mice. Because TLR9 is known to mediate HSV-2 recognition by DCs (9), it seemed plausible that the defect in these mice was due to the inability of DCs to effectively recognize HSV-2. Consistent with this hypothesis, when the hematopoietic cells (including DCs) of normal mice were replaced with those from MyD88-deficient mice by bone marrow transplantation, the mice were no longer able to mount a Th1 response against HSV-2. Remarkably, however, MyD88-deficient mice reconstituted with normal hematopoietic cells also had impaired Th1 responses against HSV-2. Therefore, TLR-mediated recognition of HSV-2 by nonhematopoietic cells is as important as DC recognition for instructing Th1 responses.

At this stage it is unclear how TLR signals from nonhematopoietic cells and DCs are integrated into a strong Th1 directive. Among nonhematopoietic cells that could generate TLR-mediated signals, epithelial cells lining genital mucosae are the most likely candidates because they are the primary targets of HSV-2 infection and express a broad repertoire of TLRs, including TLR9. Engagement of TLR on epithelial cells has been extensively shown to trigger secretion of proin flammatory chemokines and cytokines, including IL-8, MIP-1α, MIP-1β, RANTES, MCP-1, Gro-α and IL-1β, which may attract and activate DC precursors at the site of infection (Fig. 1). However, Sato and Iwasaki (5) did not find strong evidence for increased recruitment of DCs. Thus, alternative scenarios may be evoked. Microbial infections of barrier epithelial cells induce secretion of large quantities of defensins (10, 11). These encompass a family of small antimicrobial peptides that directly contribute to the killing of microbes (12). A recent study demonstrated that β-defensin 2 can induce DC maturation by engaging TLR4 (13). Therefore, it is possible that β-defensin 2 or other members of this large family of antimicrobial peptides activate mucosal DCs by engaging TLRs (Fig. 1). TLR-activated epithelial cells also secrete multiple soluble proteins, collectively known as pentraxins, which act as carriers of microbial products and apoptotic cells (14, 15). These soluble proteins may facilitate efficient delivery of PAMPs or dying infected cells to DCs for their recognition through TLR (Fig. 1).

Whatever the mechanisms, the insightful study by Sato and Iwasaki (5) demonstrates that pathogenic signals alert the adaptive immune system through a two-step amplification cascade. This mechanism may be particularly advantageous when the local concentration of PAMPs existing at the site of infection is hardly sufficient to directly activate DCs. Interestingly, it has been shown that DCs are also activated by mediators generated by damaged or inflamed tissues, such as tumor necrosis factor (16) and uric acid (17). Thus, it is clear that signals generated during mucosal innate responses have a profound impact in shaping adaptive immunity. Recognizing the importance of TLR-mediated signals in epithelial/stromal cells and understanding their integration by DCs may provide new avenues for the development of vaccination regimens aimed at activating Th1 responses against intracellular pathogens.