In this study, we demonstrated that, in human bronchial epithelial NCI-H292 cells, IFN-γ treatment dramatically enhanced IL-6 production stimulated by poly(I:C), a double-stranded RNA analog that accumulates in virus-infected cells. When NCI-H292 cells were stimulated by IFN-γ or poly(I:C) alone, only a slight increase in IL-6 gene expression was observed; however, when cells were pretreated with IFN-γ, poly(I:C) significantly enhanced IL-6 production depending on the IFN-γ-preincubation time. These reactions were observed not only in NCI H292 cells, but also in other human bronchial epithelial A549 cells and primary cultured human normal bronchial epithelial cells HNBE, suggesting that this reaction occurs commonly in bronchial epithelial cells. Previous studies reported that bronchial epithelial cells produce various inflammatory cytokines, including IL-6, in response to poly(I:C) [11, 22–24]. However, to the best of our knowledge, this is the first study to demonstrate that IFN-γ priming markedly enhances poly(I:C)-induced IL-6 production. Alveolar macrophages and infiltrating immune cells are thought to be the major source of IL-6 [25], but our finding that poly(I:C)-induced IL-6 production in bronchial epithelial cells is enhanced by IFN-γ priming may be important for understanding the mechanism of cytokine overproduction associated with the exacerbation of viral infections, such as SARS-Cov-2 and influenza.
Poly(I:C) is a synthetic analog of viral dsRNA and a well-characterized ligand of TLR3 [12]. The results of the present study suggest that the marked enhancement of poly(I:C)-induced IL-6 production by IFN-γ is accompanied by the upregulation of TLR3. IFN-γ increased TLR3 mRNA expression in NCI-H292 cells, which increased the protein level of TLR3. Furthermore, knockdown of TLR3 with siRNA effectively abrogated the enhancement by IFN-γ of poly(I:C)-induced IL-6 production. Interestingly, upregulation of TLR3 expression in NCI-H292 cells was not observed with type I IFN, TNF-α, IL-1β, or the TLR4 ligand LPS, which is consistent with finding that the enhancement of IL-6 production was not significant with such cytokines. In addition to TLR3, virus-derived dsRNAs are recognized by various innate immune sensors including retinoic acid-inducible gene I (RIG-I), melanoma differentiation-associated gene 5 (MDA-5), and dsRNA-dependent protein kinase (PKR) [26]. Matsukura et al. reported that the poly(I:C)-induced inflammatory cytokine production in bronchial epithelial cells was inhibited by siRNA targeting TLR3, but not by RIG-I, MDA-5, or PKR [27]. The role of RNA sensors other than TLR3 in the response to the combination of poly(I:C) and IFN-γ needs to be clarified in future studies.
IFN-γ is a key endogenous activator of macrophages and potentiates the actions of many TLRs and inflammatory cytokines [19]. The IFN-γ receptor is composed of the ligand-binding subunit IFNGR1 and signaling subunit IFNGR2. Binding of IFN-γ to IFNGR1 forms a functional complex with IFNGR2 to activate receptor-associated tyrosine kinases JAK1 and JAK2, which in turn phosphorylate STAT1 to form homodimers, translocate to the nucleus, and activate the transcription of target genes [28]. Enhancement of poly(I:C)-induced IL-6 production by IFN-γ was strongly suppressed by the JAK inhibitor tofacitinib, suggesting the importance of STAT1-mediated regulation. IFN-α also phosphorylates STAT1, but forms a heterotrimer with phosphorylated STAT2 and IRF9 [29]. This difference may account for the differential effects of IFN-α and IFN-γ on TLR3 induction.
Expression of the IL-6 gene is under the control of many different transcription factors such as NF-κB, SP (specificity protein) 1, AP-1 (activator-protein-1), CREB (cyclic AMP-responsive element-binding protein), and C/EBP (CCAAT-enhancer-binding protein [30]. The results of this study show that poly(I:C) activated the NF-kB pathway, but alone had a weak effect on IL-6 mRNA expression. However, stimulation of IFN-γ receptor activate JAK-STAT signaling pathway and phosphorylated STAT1 homodimer is translocated to nuclei to associate with IFN-γ activation site (GAS) element on DNA to regulate gene expression [31]. Although the human IL-6 promoter does not contain GAS regulatory elements, STATs recruit histone acetyltransferases and chromatin remodeling enzymes [28]. These enzymes are known to induce chromatin remodeling and indirectly enhance gene expression by transcription factors such as NF-κB [18]. The present results demonstrate that the priming effect of IFN-γ on poly(I:C)-induced IL-6 production may occur via a similar mechanism. In fact, ChIP assay showed that IFN-γ stimulates H3K27ac and H3K4me3 at the IL-6 gene locus in NCI H292 cells. Unlike IFN-γ, poly(I:C) induced no such histone modifications by itself, and only slightly increased IFN-γ–induced histone modifications, which was much less than the synergistic effect observed in enhancing IL-6 production. In addition, tofacitinib completely inhibited IFN-γ-induced histone methylation and acetylation at the IL-6 gene locus. These results suggest that IFN-γ induces chromatin remodeling of the IL6 gene via the JAK-STAT1 signaling pathway and promotes IL-6 gene expression induced by the poly(I:C)-TLR3-NF-κB pathway.
Finally, we demonstrated that the enhanced poly(I:C)-induced IL-6 production by IFN-γ observed in human bronchial epithelial cells could be reproduced in a mouse model of poly(I:C)-induced acute pneumonia. Similar to the results obtained with cell lines in vitro, IL-6 gene expression in lung tissue and IL-6 levels in BALF in a poly(I:C)-induced acute pneumonia were significantly enhanced by IFN-γ pretreatment. Although the involvement of cells other than bronchial epithelial cells cannot be excluded from the results obtained in this mouse model, bronchial epithelial cells may also be involved in the accumulation of IL-6 in the lungs. To address this issue, further investigations using bronchial epithelium-specific TLR3 knockout mice are required.
In this study, we focused on the expression of IL-6, an inflammatory cytokine; however, increased production of other inflammatory chemokines such as IP-10 and CXCL-11 was also observed (data not shown). The augmentation of poly(I:C)-induced IL-6 production by IFN-γ shown in this study may serve as a model for excess inflammatory cytokine production in the lungs of severe COVID19 patients. In fact, IFN-γ is reported to be an independent risk factor associated with mortality in patients with moderate and severe COVID-19 infection (PMID:32979474). Considering these reports and our findings, the prophylactic administration of JAK inhibitors, such as tofacitinib, may be a useful strategy to prevent the exacerbation of various viral infections, including COVID-19.
Taken together, our current findings indicate that IFN-γ upregulate the expression of IL-6 in response to synthetic dsRNAs poly(I:C) via TLR3-NF-κB signaling in bronchial epithelial cells. Although further research using viruses and clinical studies in patients are needed, the present results may expand our understanding of the mechanisms underlying bronchial viral infection and subsequent bronchial inflammation.