The first report that described cells with plasma cell morphology in the T cell areas of human reactive lymph nodes was published in 1958 (Lennert and Remmele 1958). These cells were named T-associated plasma cells. Only in 1999, after much debate and several controversial manuscripts, Siegal et al. (1999) reported that the plasmacytoid DCs indeed represented the previously characterized IFN-producing cells (Fitzgerald-Bocarsly 1993; Svensson et al. 1996). In the intervening years, the morphology and functions of pDCs have been fully characterized, together with their intracellular signaling cascades (Barchet et al. 2005; Liu 2005). Following viral infections, human and mouse pDCs are capable of producing up to 10 pg/cell of type I IFNs, making them 10- to 100fold more efficient than other cell types, including mDCs (Fitzgerald-Bocarsly et al. 1988; Siegal et al. 2001). Moreover, unlike mDCs, pDCs do not express
TLR2, TLR3, TLR4, or TLR5, and therefore they do not respond to the ligands of these TLRs. Remarkably, the TLRs expressed by pDCs are restricted to those that enable recognition of DNA and RNA viruses. In fact, human and murine
Fig. 1 TRIF-dependent pathways regulating TLR3- and TLR4-mediated activation of IRF3/7 and NF-kB. The adapter molecule Mal/TIRAP contains a phosphati-dylinositol 4,5-bisphosphate (PIP2) binding domain, which is important in mediating the recruitment of MyD88 to TLR4. MyD88 associates with the downstream serine/threonine kinases IRAK-1 and -4. A dimeric E2 (or ubiquitin conjugating enzyme) consisting of Ubc13 and Uev1A polyubiquitinates target proteins, including TRAF6. K63-polyubiquintated TRAF6 mediates activation of TAK1-associated proteins TAB2 and TAB3, which interact with K63-ubiquitin chains. The IKK complex is then activated, leading to NF-kB activation. TLR3 signaling to this pathway bypasses MyD88 and IRAKs and possibly TRAF6. Instead TLR3 uses RIP1, which may also be ubiquitinated by TRAF6. Both TLR3 and TLR4-mediated activation of IRF3/7 and the induction of IFN-P take place in a MyD88-independent manner and require TRIF and the IKK-related kinases, IKKe and TBK1. The adapter TRAM (TRIF-related adaptor molecule) is tethered to the plasma membrane via N-terminal myristoylation, which is required to recruit TRIF to the TLR4 cytoplasmic domain. IRF7 is also activated by the IKKe/TBK1 pathway, although it is unclear if transcriptional regulation via IFN-P is required or if this is direct. The TRIF-dependent pathways are negatively regulated by SARM
pDCs express only TLR7 and TLR9 (Bauer et al. 2001; Boonstra et al. 2003; Iwasaki and Medzhitov 2004; Jarrossay et al. 2001; Kadowaki et al. 2001; Krug et al. 2001) and can promptly produce large amounts of type I IFNs in response to either imidazoquinoline compounds (Ito et al. 2002), ssRNA-ODNs, ssRNA viruses (Heil et al. 2004), or CpG-ODNs and DNA viruses (Kadowaki et al. 2001; Krug et al. 2001).
TLR7 is closely related to TLR9 phylogenetically and as such these two receptors have several features in common (Wagner 2004). The signaling pathways activated by these TLRs are completely dependent on MyD88, and there is no evidence that other TIR-domain-containing adapters are involved (Hemmi et al. 2003). In contrast to what was observed in TLR3- and TLR4-activated signaling to IFN genes, TRIF is completely dispensable for type I IFN gene induction in the TLR7 and TLR9 pathways (Hemmi et al. 2000, 2002, 2000). Because the induction of type I IFNs is crucially dependent on the activation of IRFs, this raised the intriguing question of how these TLRs could activate IRFs without the help of TRIF. Compared to mDCs, pDCs express constitutively very high levels of IRF7 (Coccia et al. 2004; Izaguirre et al. 2003). Most cell types, including mDC, require upregulation of IRF7 in response to type I IFN feedback signaling, in order to secrete IFN-a subtypes. In contrast, pDCs are capable of rapidly secreting IFN-a even in the absence of the IFN autocrine loop due to this high basal expression of IRF7 (Barchet et al. 2002). Come clarity to this issue was provided by the observation that the engagement of TLR7 and TLR9 did not lead to the activation of IRF3, but instead activated the related factors IRF7 (Honda et al. 2004; Kawai et al. 2004) and IRF5 (Schoenemeyer et al. 2005). In a key paper from Honda et al., IRF7 has been named the master regulator of type I IFN-dependent immune response (Honda et al. 2005). Using splenic pDCs purified from IRF7 knockout mice, the authors demonstrated that the induction of IFN-a and IFN-P upon HSV-1 and VSV infection, which activate TLR9 (Krug et al. 2004) and TLR7 (Lund et al. 2004), respectively, is completely dependent on IRF7, whereas no difference was observed in IRF3-deficient pDCs. Type I IFN induction was also completely IRF7-dependent when the cells were stimulated with the TLR9 ligand, CpG-ODNs (Honda et al. 2005). Thus in the pDCs, IRF7 and not IRF3 is the key mediator of IFN-a and IFN-P gene expression.
Major advances in understanding how type I IFN production is triggered in the TLR7 and TLR9-activated pathways have been made with the discovery that IRF7 interacts directly with MyD88 to form a complex in the cytoplasm (Honda et al. 2004; Kawai et al. 2004). Moreover, this complex involves the IRAK1/4 kinases and TRAF6 (Honda et al. 2004; Kawai et al. 2004). Data from Kawai et al. has suggested that in addition to being phosphorylated, IRF7 is also ubiquitinated and that the ubiquitin ligase activity of TRAF6 is important for this event (Kawai et al. 2004). Although IRF7 activation can occur via phosphorylation through the action of the IKKe and/or TBK1 kinases as part of the secondary feedback loop (Caillaud et al. 2005; Sharma et al. 2003), it is unclear at present if either of these kinases participate in TLR7/9 signaling to IRF7 in pDCs. What is clear is that the IRAK kinases participate in the phosphorylation of IRF7 in pDCs (Uematsu et al. 2005). IRAKI interacts with and phosphorylates IRF7 in vitro and the kinase activity of IRAK1 is necessary for the activation of IRF7. TLR7 and TLR9 ligands are severely impaired in their ability to activate IRF7 and induce IFN-a in IRAKI- and IRAK4-deficient pDCs. A very recent study has also identified a role for IKKa in IRF7 activation in TLR7/9 signaling (Hoshino et al. 2006). Hoshino et al. demonstrated that TLR7/9-induced IFN-a production was severely impaired in IKKa-deficient pDCs and a kinase-deficient IKKa blocked the ability of MyD88 to activate the IFN-a promoter in synergy with IRF7 in overexpression experiments. All of these findings highlight the importance of IRF7 in TLR7 and TLR9 signaling and are summarized in Fig. 2.
Was this article helpful?