Based on our findings here and previously [5, 22], we suggest that HSV-1 is recognized by intracellular sensors present in the cytoplasm or the nucleus of human macrophages. Moreover, we hypothesize that virus fusion events result in expression of specific mediators of innate immunity. Our data support that IFI16 plays a role during early HSV-1 infection of human primary macrophages. HSV-1 DNA released during early infection may be recognized by IFI16, resulting in expression of CCL3 (Figure 3C); while, CXCL10 expression is mediated via an IFI16-independent pathway (Figure 3D). Furthermore, we found that IFI16 is upregulated by type I IFN and by the nuclear factor kappa B (NF-κB)-activating TLR7/8 agonist R848 (Figure 2A and C); yet, it is only slightly upregulated by HSV-1 during very early infection (Figure 2B).
UV-inactivated HSV-1 elicited a greater CCL3 and CXCL10 response compared to infection with replication competent virus (Figure 1A and B); therefore, the present data support previous findings suggesting that viral replication products inhibit IFN and chemokine expression in human macrophages as well as other cells . For example, the viral protein ICP27 appears to harbor immune-modulatory capabilities, as demonstrated by more potent stimulation of CXCL10 and other cytokines in macrophages infected with a HSV-1 ICP27-deletion mutant compared with infection with wild type virus . Evidence of immune-evasive activity has also been shown for several other HSV proteins, including ICP0, ICP34.5, Us11, vhs, and US3, which were reviewed previously [27, 28]. IFN-inducible IFI16 was more effectively induced in macrophages infected with UV-inactivated HSV-1 compared with replication competent HSV, further supporting the notion that viral replication products counteract IFN responses.
The first cytoplasmic DNA receptor for HSV DNA was identified in 2007. The DNA-dependent activator of IFN regulatory factors (DAI) has been shown to mediate recognition of HSV DNA in a murine fibroblast cell line . Since the initial publications relating to DAI, accumulating evidence has suggested additional DNA receptors exist. Within the past two years, major breakthroughs have been achieved, and an increasing number of DNA receptors have been associated with innate sensing of HSV, including the pyrin and HIN domain-containing protein (PYHIN) family member IFI16, which recognizes HSV-1-derived DNA in murine macrophages and Hela cells .
CCL3 mRNA accumulation was found to be dependent on IFI16 (Figure 3C). However, IFI16 knockdown did not consistently reduce CCL3 protein levels after 20 h of infection (data not shown). Reasons for this discrepancy between the mRNA and protein data include the following: i) CCL3 protein secretion may be transcriptionally regulated by factors other than IFI16 during HSV-1 infection in monocyte-derived macrophages or ii) CCL3 induction at the later time point is partly due to other virally induced factors, such as induced TNF-α , which we have previously found to be secreted 5 h after infection . Therefore, studies evaluating CCL3 mRNA and protein accumulation at later stages during infection may be complicated by secretion of other mediators and by the accumulation of viral gene-products that selectively interfere with innate immune responses. However, the observed early IFI16-dependent CCL3 mRNA accumulation precedes the HSV-1-induced secretion of cytokines . Therefore, we believe the CCL3 mRNA accumulation is mediated by direct sensing of HSV-1 virions.
Very recent data suggest a multifunctional role for IFI16 during HSV-1, HSV-2, and CMV infection. Studies from Gariano et al. demonstrated that siRNA knockdown of IFI16 enhanced viral replication , indicating IFI16 is a restriction factor for herpes virus replication. In theory, IFI16 could mediate innate recognition of HSV-1, while restricting replication at the same time. Thus, knock-down of IFI16 could enhance replication-dependent innate responses. Our data showed a two- to three-fold increased induction of CXCL10 in response to HSV-1-infection when macrophages were subjected to IFI16 knock-down (Figure 3D). Reduced IFI16 protein levels might result in decreased competition for viral DNA binding; thus, other DNA receptors able to stimulate CXCL10 expression have greater access to HSV-1 DNA not sequestered by IFI16. Another explanation could be redundant systems leading to CXCL10 mRNA accumulation: one dependent on virus replication (possibly being sensitive to IFI16 inhibition) and one via a pre-replication step. Further studies are necessary to determine whether IFI16 regulates mRNA accumulation and/or transcription or IFI16 acts as an innate receptor, and whether IFI16 restricts HSV-1 replication in human primary cells and/or sequesters viral DNA.
It should be noted that induction of CXCL10 may also be mediated by fusion or membrane-membrane interactions between the virus and cell. Collins et al. showed that HSV-1 replication is not required for induction of the IFN-stimulated genes CXCl10 and ISG56 , which is supported by our data showing replication-independent CXCL10 induction (Figure 1B). Moreover, a very recent study demonstrated that fusion events may indeed trigger innate responses, including expression of CXCL10 . Accordingly, we found that membrane-membrane interactions induced CXCL10 in human primary macrophages and human peripheral blood mononuclear cells (Figure 4A and B). We hypothesize that viral fusion events could account for the observed replication- and IFI16-independent induction of CXCL10. Furthermore, we found that CCL3 is not induced by lipofectamin2000, suggesting that CCL3 mRNA accumulation is mediated via post-entry steps involving IFI16.
A number of other potential DNA sensors that recognize HSV-1 have been identified since this study was completed. The helicases DDX41, DDX60, DHX9, and DHX36 were identified as regulators of HSV-induced type I IFN and cytokine responses [33–35]. Zhang Z. et al. found that DDX41 mediates IFN-β, TNF-α, and IL-6 responses by recognizing viral DNA in the cytoplasm following HSV-1 infection in murine DCs . Miyashita et al. found that DDX60 binds dsRNA and dsDNA in the cytoplasm and mediates induction of IFN-β and CXCL10 during HSV-1 infection in Hela cells; in addition, they found DDX60 promotes RIG-I and MDA-5 signaling . Therefore, it remains to be determined whether DDX60 senses the HSV-1 genome or rather enhances the RLR-mediated signaling after HSV infection. Finally, Kim T. et al. found that DHX9 mediates TNF-α expression and DHX36 mediates IFN-α production in a human plasmacytoid cell line. In addition to the helicases, the DNA binding protein Ku70 was recently found to mediate type III IFN responses in HEK293 cells after DNA was introduced into the cells or cells were infected with HSV-2 .
Future investigations are necessary to further elucidate the role of IFI16 and determine the following: i) whether IFI16 functions as a primary innate receptor for HSV-1 DNA in human primary macrophages; ii) whether the aforementioned helicases DDX41, DDX60, DHX9 and DHX36, and Ku70 are sensors of viral DNA in human primary cells, including human primary macrophages; iii) which cellular responses these proteins regulate in primary cells; iv) whether HSV infection is sensed by DDX40, DDX41, DHX9, or DHX36 in human primary cells, including human primary macrophages; v) whether the specificity of DNA-induced innate responses is based on involvement of specific signaling adaptors and/or recognition of specific DNA sequences or lengths of DNA; vi) whether one of the newly identified DNA receptors mediates expression of CXCL10; vii) whether lipid-lipid interactions induce CXCL10/ISG expression in human primary macrophages during early HSV-1 infection or infection with other enveloped viruses; and viii) whether IFI16 is a restriction factor of HSV-1 infection in human primary cells.
Further details on the innate DNA receptors may provide important knowledge for the development of novel DNA vaccine adjuvants, specifically targeting receptors of interest. As proposed by A. Iwasaki, timed and targeted cytokine and chemokine responses may facilitate an influx of cells necessary for efficient vaccine responses in mucosal areas . CXCL9 and CXCL10 play an important role in regulating NK cell and T lymphocyte influx to the site of infection and for control of vaginal HSV infection ; therefore, modulated production of specific chemokines may be imperative for future vaccine adjuvant candidates . CCL3 could be important in terms of HSV clearance, as mice lacking the CCL3 receptor CCR5 are highly susceptible to vaginal HSV-2 infection . In the context of vaccines, it is of interest that intracellular delivery of DNA forming a complex with lipids has yielded promising results in guinea pigs, decreasing viral replication and the severity of infection [38–40]. Based on the current knowledge, lipid-delivered DNA may provide two independent innate signals, one via lipid-interaction with the target cell membrane and one from the DNA targeting a cytoplasmic or nuclear DNA receptor.
Further investigation is necessary to determine the role of IFI16 and other DNA receptors in the development of novel DNA vaccine adjuvants targeting the cytoplasm. Furthermore, it remains to be determined whether DNA adjuvants will improve HSV vaccines in humans. However, a detailed understanding of cytoplasmic/nuclear DNA-host receptor interactions and the subsequent innate responses generated may provide rationale for use of delivering DNA to the cytoplasm or nucleus of cells as a vaccine adjuvant or direct antiviral therapy.