We measured increased promoter activity of the human TAP1 gene and detected enhanced expression of TAP1 protein in HTNV-infected A549 cells. Similarly, paramyxoviruses have been shown to enhance TAP1 expression [30]. Thus, hantaviruses may augment transport of peptides ABT-888 ic50 into the ER similar to flaviviruses [31, 32]. Type I IFN was not absolutely required for HTNV-induced HLA-I expression. First, HTNV only moderately increased the number of IFN-β transcripts in A549

cells in line with recent studies [26, 33]. Second, Vero E6 cells, which lack type I IFN genes [25], also upregulate MHC-I upon HTNV infection. Third, although HTNV-infected A549 cells produced type III IFN (IFN-λ1 and IFN-λ2) transcripts confirming

a previous report [26], exogenously added type IFN-λ1 did not significantly increase MHC-I expression in Vero E6 cells. In addition, transfection of RNA derived from HTNV-infected cells triggered MHC-I upregulation, although Peptide 17 mw type III IFN could not be detected in the supernatant. Finally, IFN-λ1 was not detectable in HTNV stocks prepared from Vero E6 cells [34]. This points to an IFN-independent mechanism contributing to HTNV-associated MHC-I upregulation. On the other hand, we have previously observed that upregulation of HLA-I on human endothelial cells infected with hantavirus can be blocked in part by antibodies directed against type I IFN [35]. Taken together, our results suggest that both direct and indirect (IFN-driven) hantaviral mechanisms are required for efficient HLA-I upregulation. Activation of NF-κB could increase MHC-I transcription independently of IFN during hantavirus infection as reported for flaviviruses [36, 37]. In accordance, HTNV RNA has recently been described Fossariinae to trigger NF-κB promoter activity through RIG-I stimulation [21]. On the other hand, the HTNV N protein has been demonstrated to interfere with NF-κB activation [38]. Thus, hantavirus-triggered PRRs may facilitate the assembly of a MHC-I-specific enhanceosome that binds to promoter sequences different from the NF-κB binding site as shown for NLRC5 [39, 40]. Compared to DCs stimulated with TNF-α, HTNV-infected

DCs show increased macropinocytosis and receptor-mediated endocytosis [23], a prerequisite of cross-presentation. Indeed, we observed in this study that HTNV confers upon DCs the capacity to efficiently cross-present pp65, a HCMV-encoded model antigen. It is likely that HTNV-infected DCs also cross-present HTNV-derived antigens. In contrast, cross-presenting uninfected DCs that are activated indirectly by proinflammatory cytokines may induce tolerance rather than immunity [41]. It has been shown that HTNV-infected DCs do not undergo cell death [23]. Thus, lung DCs infected with HTNV after inhalation of virion containing aerosols could migrate to the draining lymph nodes and cross-prime powerful antiviral cytotoxic T cells.