Yajie Hu ¹ Yunguang Hu ² Anguo Yin ² Yaming Lv ¹ Jiang Li ² Jingyuan Fan ¹ Baojiang Qian ¹ Jie Song ^2* Yunhui Zhang ¹

• ¹ The First People’s Hospital of Yunnan Province, Kunming, Yunnan Province, China
• ² Institute of Medical Biology, Chinese Academy of Medical Sciences and Peking Union Medical College, Kunming, Yunnan Province, China

The mechanisms underlying pathological changes in the central nervous system (CNS) following Coxsackievirus A16 (CV-A16) infection have not yet been elucidated. IFN-γ-inducible protein-10 (IP-10) is often used as a predictive factor to monitor early virus infection. It has also been reported that IP-10 plays a pivotal role in neuroinflammation. In this study, we aimed to explore the role of IP-10 in the neuropathogenesis of CV-A16 infection. We observed that the level of IP-10, as well as the TLR3-TRIF-TRAF3-TBK1-NF-κB and RIG-I/MDA5-MAVS-TRAFS-TBK1-NF-κB pathways, which are the upstream of IP-10, were significantly elevated during the course of CV-A16 infection. This increase was accompanied by an increase in a series of inflammatory cytokines at different time-points during CV-A16 infection. To determine whether IP-10 influences BBB integrity, we examined junctional complexes. Our results revealed that the expression levels of Claudin5, Occludin, ZO-1 and VE-Cadherin were notably decreased in CV-A16-infected HUVECs, but these indicators were restored in CV-A16-infected HUVECs with Eldelumab treatment. Nevertheless, IP-10 is only a chemokine that primarily traffics CXCR3-positive immune cells to inflammatory sites or promotes the production of inflammatory cytokines. Therefore, the interactions between IP-10 and inflammatory cytokines were evaluated. Our data revealed that IP-10 mediated the production of TNF-α, which was also observed to change the junctional complexes. Moreover, in a suckling mouse model, IP-10 and TNF-α treatments exacerbated clinical symptoms, mortality and pathological changes in the brain of CV-A16-infected mice, but Anti-IP-10 and Anti-TNF-α treatments alleviated these changes. Our data also revealed that IP-10 may be detected early in CV-A16 infection, whereas TNF-α was detected late in CV-A16 infection, and the production of TNF-α was also found to be positively correlated with IP-10. In addition, IP-10 and TNF-α were observed to reduce junctional complexes and enhance virus entry into the CNS. Taken together, this study provides the first evidence that CV-A16 activates the IP-10/TNF-α regulatory axis to cause BBB damage and accelerate the formation of neuroinflammation in infected hosts, which not only provides a new understanding of the neuropathogenesis caused by CV-A16, but also offers a promising target for the development of CV-A16 antiviral drugs.