Linking Viral DNA to Endosomal Innate Immune Recognition

Activation of the adaptive immune system has long been known to depend on two signals: Ag specificity alone is not enough, but is cross-checked for the presence of a second, independent signal confirming the irregularity of the first Ag-specific signal. Prominent examples are B cell responses that are controlled by Th cells specific for the same Ag, or T cell responses that are controlled through the presence of activated APCs, typically caused by infections. In this regard, infections are recognized by distinct signals, such as pathogen-associated structural patterns (1), including Ag repetitiveness, which is mostly recognized by B cells and the humoral innate immune system, as well as by pathogen-associated molecular patterns, recognized by cell surface, endosomal, and intracellular receptors (2). Most intuitive ligands for these receptors, such as LPS, only exist in bacterial outer membranes and are recognized, in this case, by TLR4 on the APC surface. Other ligands, such as dsRNA produced by viral replication, are sensed by a magnitude of intracellular and endosomal receptors (3, 4). Less intuitive are structures such as ssRNA or dsDNA, which are abundant in virtually all cell types, and hence do not appear to be ideal pathogen-associated molecular patterns because they are not exclusively pathogen associated. However, in this case, the immune system harnesses another fundamental aspect of cellular biology, namely subcellular location. Indeed, under normal circumstances, and in the absence of infection, neither RNA nor DNA is found in endosomes, which explains why TLR7/8 (recognizing RNA) and TLR9 (recognizing DNA) are all localized in the endosome (5). In the case of TLR9, recognition of microbe-specific DNA is enhanced by the presence of unmethylated CG-rich motifs, which are suppressed in eukaryotic DNA. Recognition of microbial DNA was initially described for the tuberculosis vaccine bacillus Calmette–Guérin, which is rich in palindromic sequences (6), and subsequently extended to somewhat species-specific DNA oligomers, in particular when they were stabilized by thioester bonds (CpGs) (7). Soon thereafter, TRAF6 and MyD88 were identified as key signaling mediators in TLR pathways for such endosomal DNA (2). Although TLR9 was clearly involved in recognition of bacterial DNA, much less was known about the role of viral DNA.

In the present Pillars of Immunology article (8), this question was beautifully elucidated. A relatively recently discovered cell type, at that time, came into the spotlight—the plasmacytoid dendritic cell. The paper has a stunningly clear and interesting storyline demonstrating that plasmacytoid dendritic cells recognize DNA of HSV-2 in a TLR9- and MyD88-dependent fashion, and they rapidly produce vast amounts of IFN-α, a strongly antiviral cytokine (9). For TLR9 activation to occur, endosomal uptake and acidification were required. Interestingly, and perhaps underestimated at the time, packaging of viral DNA into virions made TLR9 activation much more efficient. This demonstrated that free-floating DNA does not have the same impact as DNA in a viral particle, and it indicated that endosomal targeting as well as intraendosomal handling may impact TLR9 activation. Endosomal trafficking has indeed been confirmed later to be important for the type of response induced by TLR9 stimulation (10).

The Pillars of Immunology article was not only of fundamental elegance for basic immunology, but it led to the finding that activation of TLR9 by CpGs strongly enhanced specific T cell responses. This discovery promoted the development of vaccines and immunomodulators based on viral particles loaded with CpG (11–13). Indeed, ligands for TLR9 are one of the most sought out class of new adjuvants. For example, a CpG oligonucleotide is used in a marketed vaccine against hepatitis B to enhance protective Ab responses, overcoming the well-known problem of relatively high nonresponder rates to hepatitis B vaccination without CpG (14). TLR9 ligands are also used for the development of vaccines and immune stimulators for cancer patients (15, 16). CpGs may well help fighting cancer because they promote type 1 immunity, including cytotoxic T cell and NK cell activation, as well as inhibition of regulatory T cells and immunosuppressive myeloid cells (17, 18). For these reasons, TLR9 stimulation by DNA is increasingly exploited for therapeutic purposes. Thus, this Pillars of Immunology article uncovered recognition of viral DNA, in particular when packaged in viral particles, by TLR9 in endosomes and paved the way for the development of novel, CpG-based prophylactic and therapeutic vaccines.